diff options
Diffstat (limited to 'arch')
288 files changed, 15680 insertions, 2398 deletions
diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index 3d74801..56a4df9 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -70,10 +70,6 @@ config GENERIC_ISA_DMA bool default y -config GENERIC_IOMAP - bool - default n - source "init/Kconfig" source "kernel/Kconfig.freezer" @@ -319,6 +315,7 @@ config ISA_DMA_API config PCI bool depends on !ALPHA_JENSEN + select GENERIC_PCI_IOMAP default y help Find out whether you have a PCI motherboard. PCI is the name of a diff --git a/arch/alpha/kernel/pci-noop.c b/arch/alpha/kernel/pci-noop.c index 246100e..04eea48 100644 --- a/arch/alpha/kernel/pci-noop.c +++ b/arch/alpha/kernel/pci-noop.c @@ -185,15 +185,3 @@ struct dma_map_ops alpha_noop_ops = { struct dma_map_ops *dma_ops = &alpha_noop_ops; EXPORT_SYMBOL(dma_ops); - -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) -{ - return NULL; -} - -void pci_iounmap(struct pci_dev *dev, void __iomem * addr) -{ -} - -EXPORT_SYMBOL(pci_iomap); -EXPORT_SYMBOL(pci_iounmap); diff --git a/arch/alpha/kernel/pci.c b/arch/alpha/kernel/pci.c index c9ab94e..f3cae27 100644 --- a/arch/alpha/kernel/pci.c +++ b/arch/alpha/kernel/pci.c @@ -508,30 +508,7 @@ sys_pciconfig_iobase(long which, unsigned long bus, unsigned long dfn) return -EOPNOTSUPP; } -/* Create an __iomem token from a PCI BAR. Copied from lib/iomap.c with - no changes, since we don't want the other things in that object file. */ - -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) -{ - resource_size_t start = pci_resource_start(dev, bar); - resource_size_t len = pci_resource_len(dev, bar); - unsigned long flags = pci_resource_flags(dev, bar); - - if (!len || !start) - return NULL; - if (maxlen && len > maxlen) - len = maxlen; - if (flags & IORESOURCE_IO) - return ioport_map(start, len); - if (flags & IORESOURCE_MEM) { - /* Not checking IORESOURCE_CACHEABLE because alpha does - not distinguish between ioremap and ioremap_nocache. */ - return ioremap(start, len); - } - return NULL; -} - -/* Destroy that token. Not copied from lib/iomap.c. */ +/* Destroy an __iomem token. Not copied from lib/iomap.c. */ void pci_iounmap(struct pci_dev *dev, void __iomem * addr) { @@ -539,7 +516,6 @@ void pci_iounmap(struct pci_dev *dev, void __iomem * addr) iounmap(addr); } -EXPORT_SYMBOL(pci_iomap); EXPORT_SYMBOL(pci_iounmap); /* FIXME: Some boxes have multiple ISA bridges! */ diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 9d66dfc..24626b0 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -16,6 +16,7 @@ config ARM select HAVE_FTRACE_MCOUNT_RECORD if (!XIP_KERNEL) select HAVE_DYNAMIC_FTRACE if (!XIP_KERNEL) select HAVE_FUNCTION_GRAPH_TRACER if (!THUMB2_KERNEL) + select ARCH_BINFMT_ELF_RANDOMIZE_PIE select HAVE_GENERIC_DMA_COHERENT select HAVE_KERNEL_GZIP select HAVE_KERNEL_LZO @@ -30,6 +31,7 @@ config ARM select HAVE_SPARSE_IRQ select GENERIC_IRQ_SHOW select CPU_PM if (SUSPEND || CPU_IDLE) + select GENERIC_PCI_IOMAP help The ARM series is a line of low-power-consumption RISC chip designs licensed by ARM Ltd and targeted at embedded applications and diff --git a/arch/arm/include/asm/io.h b/arch/arm/include/asm/io.h index 065d100..9275828 100644 --- a/arch/arm/include/asm/io.h +++ b/arch/arm/include/asm/io.h @@ -27,6 +27,7 @@ #include <asm/byteorder.h> #include <asm/memory.h> #include <asm/system.h> +#include <asm-generic/pci_iomap.h> /* * ISA I/O bus memory addresses are 1:1 with the physical address. @@ -306,7 +307,6 @@ extern void ioport_unmap(void __iomem *addr); struct pci_dev; -extern void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen); extern void pci_iounmap(struct pci_dev *dev, void __iomem *addr); /* diff --git a/arch/arm/mach-davinci/board-da850-evm.c b/arch/arm/mach-davinci/board-da850-evm.c index f8a682f..6b22b54 100644 --- a/arch/arm/mach-davinci/board-da850-evm.c +++ b/arch/arm/mach-davinci/board-da850-evm.c @@ -127,7 +127,7 @@ static void da850_evm_m25p80_notify_add(struct mtd_info *mtd) size_t retlen; if (!strcmp(mtd->name, "MAC-Address")) { - mtd->read(mtd, 0, ETH_ALEN, &retlen, mac_addr); + mtd_read(mtd, 0, ETH_ALEN, &retlen, mac_addr); if (retlen == ETH_ALEN) pr_info("Read MAC addr from SPI Flash: %pM\n", mac_addr); diff --git a/arch/arm/mach-omap2/devices.c b/arch/arm/mach-omap2/devices.c index 35d5dff..46dfd1a 100644 --- a/arch/arm/mach-omap2/devices.c +++ b/arch/arm/mach-omap2/devices.c @@ -28,6 +28,7 @@ #include <plat/board.h> #include <plat/mcbsp.h> #include <plat/mmc.h> +#include <plat/iommu.h> #include <plat/dma.h> #include <plat/omap_hwmod.h> #include <plat/omap_device.h> @@ -211,9 +212,15 @@ static struct platform_device omap3isp_device = { .resource = omap3isp_resources, }; +static struct omap_iommu_arch_data omap3_isp_iommu = { + .name = "isp", +}; + int omap3_init_camera(struct isp_platform_data *pdata) { omap3isp_device.dev.platform_data = pdata; + omap3isp_device.dev.archdata.iommu = &omap3_isp_iommu; + return platform_device_register(&omap3isp_device); } diff --git a/arch/arm/mach-tegra/include/mach/kbc.h b/arch/arm/mach-tegra/include/mach/kbc.h index 4f3572a..20bb054 100644 --- a/arch/arm/mach-tegra/include/mach/kbc.h +++ b/arch/arm/mach-tegra/include/mach/kbc.h @@ -53,6 +53,7 @@ struct tegra_kbc_platform_data { struct tegra_kbc_pin_cfg pin_cfg[KBC_MAX_GPIO]; const struct matrix_keymap_data *keymap_data; + u32 wakeup_key; bool wakeup; bool use_fn_map; bool use_ghost_filter; diff --git a/arch/arm/mach-u300/Kconfig b/arch/arm/mach-u300/Kconfig index 1cbcd4f..54d8f34 100644 --- a/arch/arm/mach-u300/Kconfig +++ b/arch/arm/mach-u300/Kconfig @@ -7,8 +7,8 @@ comment "ST-Ericsson Mobile Platform Products" config MACH_U300 bool "U300" select PINCTRL - select PINMUX_U300 - select GPIO_U300 + select PINCTRL_U300 + select PINCTRL_COH901 comment "ST-Ericsson U300/U330/U335/U365 Feature Selections" diff --git a/arch/arm/mach-u300/core.c b/arch/arm/mach-u300/core.c index 6979307..b4c6926 100644 --- a/arch/arm/mach-u300/core.c +++ b/arch/arm/mach-u300/core.c @@ -1605,15 +1605,15 @@ static struct platform_device pinmux_device = { }; /* Pinmux settings */ -static struct pinmux_map u300_pinmux_map[] = { +static struct pinmux_map __initdata u300_pinmux_map[] = { /* anonymous maps for chip power and EMIFs */ - PINMUX_MAP_PRIMARY_SYS_HOG("POWER", "power"), - PINMUX_MAP_PRIMARY_SYS_HOG("EMIF0", "emif0"), - PINMUX_MAP_PRIMARY_SYS_HOG("EMIF1", "emif1"), + PINMUX_MAP_SYS_HOG("POWER", "pinmux-u300", "power"), + PINMUX_MAP_SYS_HOG("EMIF0", "pinmux-u300", "emif0"), + PINMUX_MAP_SYS_HOG("EMIF1", "pinmux-u300", "emif1"), /* per-device maps for MMC/SD, SPI and UART */ - PINMUX_MAP_PRIMARY("MMCSD", "mmc0", "mmci"), - PINMUX_MAP_PRIMARY("SPI", "spi0", "pl022"), - PINMUX_MAP_PRIMARY("UART0", "uart0", "uart0"), + PINMUX_MAP("MMCSD", "pinmux-u300", "mmc0", "mmci"), + PINMUX_MAP("SPI", "pinmux-u300", "spi0", "pl022"), + PINMUX_MAP("UART0", "pinmux-u300", "uart0", "uart0"), }; struct u300_mux_hog { diff --git a/arch/arm/mach-u300/include/mach/gpio-u300.h b/arch/arm/mach-u300/include/mach/gpio-u300.h index 0c2b202..bf4c793 100644 --- a/arch/arm/mach-u300/include/mach/gpio-u300.h +++ b/arch/arm/mach-u300/include/mach/gpio-u300.h @@ -9,121 +9,6 @@ #ifndef __MACH_U300_GPIO_U300_H #define __MACH_U300_GPIO_U300_H -/* - * Individual pin assignments for the B26/S26. Notice that the - * actual usage of these pins depends on the PAD MUX settings, that - * is why the same number can potentially appear several times. - * In the reference design each pin is only used for one purpose. - * These were determined by inspecting the B26/S26 schematic: - * 2/1911-ROA 128 1603 - */ -#ifdef CONFIG_MACH_U300_BS2X -#define U300_GPIO_PIN_UART_RX 0 -#define U300_GPIO_PIN_UART_TX 1 -#define U300_GPIO_PIN_GPIO02 2 /* Unrouted */ -#define U300_GPIO_PIN_GPIO03 3 /* Unrouted */ -#define U300_GPIO_PIN_CAM_SLEEP 4 -#define U300_GPIO_PIN_CAM_REG_EN 5 -#define U300_GPIO_PIN_GPIO06 6 /* Unrouted */ -#define U300_GPIO_PIN_GPIO07 7 /* Unrouted */ - -#define U300_GPIO_PIN_GPIO08 8 /* Service point SP2321 */ -#define U300_GPIO_PIN_GPIO09 9 /* Service point SP2322 */ -#define U300_GPIO_PIN_PHFSENSE 10 /* Headphone jack sensing */ -#define U300_GPIO_PIN_MMC_CLKRET 11 /* Clock return from MMC/SD card */ -#define U300_GPIO_PIN_MMC_CD 12 /* MMC Card insertion detection */ -#define U300_GPIO_PIN_FLIPSENSE 13 /* Mechanical flip sensing */ -#define U300_GPIO_PIN_GPIO14 14 /* DSP JTAG Port RTCK */ -#define U300_GPIO_PIN_GPIO15 15 /* Unrouted */ - -#define U300_GPIO_PIN_GPIO16 16 /* Unrouted */ -#define U300_GPIO_PIN_GPIO17 17 /* Unrouted */ -#define U300_GPIO_PIN_GPIO18 18 /* Unrouted */ -#define U300_GPIO_PIN_GPIO19 19 /* Unrouted */ -#define U300_GPIO_PIN_GPIO20 20 /* Unrouted */ -#define U300_GPIO_PIN_GPIO21 21 /* Unrouted */ -#define U300_GPIO_PIN_GPIO22 22 /* Unrouted */ -#define U300_GPIO_PIN_GPIO23 23 /* Unrouted */ -#endif - -/* - * Individual pin assignments for the B330/S330 and B365/S365. - * Notice that the actual usage of these pins depends on the - * PAD MUX settings, that is why the same number can potentially - * appear several times. In the reference design each pin is only - * used for one purpose. These were determined by inspecting the - * S365 schematic. - */ -#if defined(CONFIG_MACH_U300_BS330) || defined(CONFIG_MACH_U300_BS365) || \ - defined(CONFIG_MACH_U300_BS335) -#define U300_GPIO_PIN_UART_RX 0 -#define U300_GPIO_PIN_UART_TX 1 -#define U300_GPIO_PIN_UART_CTS 2 -#define U300_GPIO_PIN_UART_RTS 3 -#define U300_GPIO_PIN_CAM_MAIN_STANDBY 4 /* Camera MAIN standby */ -#define U300_GPIO_PIN_GPIO05 5 /* Unrouted */ -#define U300_GPIO_PIN_MS_CD 6 /* Memory Stick Card insertion */ -#define U300_GPIO_PIN_GPIO07 7 /* Test point TP2430 */ - -#define U300_GPIO_PIN_GPIO08 8 /* Test point TP2437 */ -#define U300_GPIO_PIN_GPIO09 9 /* Test point TP2431 */ -#define U300_GPIO_PIN_GPIO10 10 /* Test point TP2432 */ -#define U300_GPIO_PIN_MMC_CLKRET 11 /* Clock return from MMC/SD card */ -#define U300_GPIO_PIN_MMC_CD 12 /* MMC Card insertion detection */ -#define U300_GPIO_PIN_CAM_SUB_STANDBY 13 /* Camera SUB standby */ -#define U300_GPIO_PIN_GPIO14 14 /* Test point TP2436 */ -#define U300_GPIO_PIN_GPIO15 15 /* Unrouted */ - -#define U300_GPIO_PIN_GPIO16 16 /* Test point TP2438 */ -#define U300_GPIO_PIN_PHFSENSE 17 /* Headphone jack sensing */ -#define U300_GPIO_PIN_GPIO18 18 /* Test point TP2439 */ -#define U300_GPIO_PIN_GPIO19 19 /* Routed somewhere */ -#define U300_GPIO_PIN_GPIO20 20 /* Unrouted */ -#define U300_GPIO_PIN_GPIO21 21 /* Unrouted */ -#define U300_GPIO_PIN_GPIO22 22 /* Unrouted */ -#define U300_GPIO_PIN_GPIO23 23 /* Unrouted */ - -#define U300_GPIO_PIN_GPIO24 24 /* Unrouted */ -#define U300_GPIO_PIN_GPIO25 25 /* Unrouted */ -#define U300_GPIO_PIN_GPIO26 26 /* Unrouted */ -#define U300_GPIO_PIN_GPIO27 27 /* Unrouted */ -#define U300_GPIO_PIN_GPIO28 28 /* Unrouted */ -#define U300_GPIO_PIN_GPIO29 29 /* Unrouted */ -#define U300_GPIO_PIN_GPIO30 30 /* Unrouted */ -#define U300_GPIO_PIN_GPIO31 31 /* Unrouted */ - -#define U300_GPIO_PIN_GPIO32 32 /* Unrouted */ -#define U300_GPIO_PIN_GPIO33 33 /* Unrouted */ -#define U300_GPIO_PIN_GPIO34 34 /* Unrouted */ -#define U300_GPIO_PIN_GPIO35 35 /* Unrouted */ -#define U300_GPIO_PIN_GPIO36 36 /* Unrouted */ -#define U300_GPIO_PIN_GPIO37 37 /* Unrouted */ -#define U300_GPIO_PIN_GPIO38 38 /* Unrouted */ -#define U300_GPIO_PIN_GPIO39 39 /* Unrouted */ - -#ifdef CONFIG_MACH_U300_BS335 - -#define U300_GPIO_PIN_GPIO40 40 /* Unrouted */ -#define U300_GPIO_PIN_GPIO41 41 /* Unrouted */ -#define U300_GPIO_PIN_GPIO42 42 /* Unrouted */ -#define U300_GPIO_PIN_GPIO43 43 /* Unrouted */ -#define U300_GPIO_PIN_GPIO44 44 /* Unrouted */ -#define U300_GPIO_PIN_GPIO45 45 /* Unrouted */ -#define U300_GPIO_PIN_GPIO46 46 /* Unrouted */ -#define U300_GPIO_PIN_GPIO47 47 /* Unrouted */ - -#define U300_GPIO_PIN_GPIO48 48 /* Unrouted */ -#define U300_GPIO_PIN_GPIO49 49 /* Unrouted */ -#define U300_GPIO_PIN_GPIO50 50 /* Unrouted */ -#define U300_GPIO_PIN_GPIO51 51 /* Unrouted */ -#define U300_GPIO_PIN_GPIO52 52 /* Unrouted */ -#define U300_GPIO_PIN_GPIO53 53 /* Unrouted */ -#define U300_GPIO_PIN_GPIO54 54 /* Unrouted */ -#define U300_GPIO_PIN_GPIO55 55 /* Unrouted */ -#endif - -#endif - /** * enum u300_gpio_variant - the type of U300 GPIO employed */ diff --git a/arch/arm/mach-u300/include/mach/irqs.h b/arch/arm/mach-u300/include/mach/irqs.h index db3fbfa..ee78a26 100644 --- a/arch/arm/mach-u300/include/mach/irqs.h +++ b/arch/arm/mach-u300/include/mach/irqs.h @@ -110,7 +110,7 @@ #endif /* Maximum 8*7 GPIO lines */ -#ifdef CONFIG_GPIO_U300 +#ifdef CONFIG_PINCTRL_COH901 #define IRQ_U300_GPIO_BASE (U300_VIC_IRQS_END) #define IRQ_U300_GPIO_END (IRQ_U300_GPIO_BASE + 56) #else diff --git a/arch/arm/mach-u300/mmc.c b/arch/arm/mach-u300/mmc.c index 4d482aa..05abd6a 100644 --- a/arch/arm/mach-u300/mmc.c +++ b/arch/arm/mach-u300/mmc.c @@ -18,8 +18,8 @@ #include <linux/slab.h> #include <mach/coh901318.h> #include <mach/dma_channels.h> -#include <mach/gpio-u300.h> +#include "u300-gpio.h" #include "mmc.h" static struct mmci_platform_data mmc0_plat_data = { diff --git a/arch/arm/mach-u300/u300-gpio.h b/arch/arm/mach-u300/u300-gpio.h new file mode 100644 index 0000000..847dc25 --- /dev/null +++ b/arch/arm/mach-u300/u300-gpio.h @@ -0,0 +1,114 @@ +/* + * Individual pin assignments for the B26/S26. Notice that the + * actual usage of these pins depends on the PAD MUX settings, that + * is why the same number can potentially appear several times. + * In the reference design each pin is only used for one purpose. + * These were determined by inspecting the B26/S26 schematic: + * 2/1911-ROA 128 1603 + */ +#ifdef CONFIG_MACH_U300_BS2X +#define U300_GPIO_PIN_UART_RX 0 +#define U300_GPIO_PIN_UART_TX 1 +#define U300_GPIO_PIN_GPIO02 2 /* Unrouted */ +#define U300_GPIO_PIN_GPIO03 3 /* Unrouted */ +#define U300_GPIO_PIN_CAM_SLEEP 4 +#define U300_GPIO_PIN_CAM_REG_EN 5 +#define U300_GPIO_PIN_GPIO06 6 /* Unrouted */ +#define U300_GPIO_PIN_GPIO07 7 /* Unrouted */ + +#define U300_GPIO_PIN_GPIO08 8 /* Service point SP2321 */ +#define U300_GPIO_PIN_GPIO09 9 /* Service point SP2322 */ +#define U300_GPIO_PIN_PHFSENSE 10 /* Headphone jack sensing */ +#define U300_GPIO_PIN_MMC_CLKRET 11 /* Clock return from MMC/SD card */ +#define U300_GPIO_PIN_MMC_CD 12 /* MMC Card insertion detection */ +#define U300_GPIO_PIN_FLIPSENSE 13 /* Mechanical flip sensing */ +#define U300_GPIO_PIN_GPIO14 14 /* DSP JTAG Port RTCK */ +#define U300_GPIO_PIN_GPIO15 15 /* Unrouted */ + +#define U300_GPIO_PIN_GPIO16 16 /* Unrouted */ +#define U300_GPIO_PIN_GPIO17 17 /* Unrouted */ +#define U300_GPIO_PIN_GPIO18 18 /* Unrouted */ +#define U300_GPIO_PIN_GPIO19 19 /* Unrouted */ +#define U300_GPIO_PIN_GPIO20 20 /* Unrouted */ +#define U300_GPIO_PIN_GPIO21 21 /* Unrouted */ +#define U300_GPIO_PIN_GPIO22 22 /* Unrouted */ +#define U300_GPIO_PIN_GPIO23 23 /* Unrouted */ +#endif + +/* + * Individual pin assignments for the B330/S330 and B365/S365. + * Notice that the actual usage of these pins depends on the + * PAD MUX settings, that is why the same number can potentially + * appear several times. In the reference design each pin is only + * used for one purpose. These were determined by inspecting the + * S365 schematic. + */ +#if defined(CONFIG_MACH_U300_BS330) || defined(CONFIG_MACH_U300_BS365) || \ + defined(CONFIG_MACH_U300_BS335) +#define U300_GPIO_PIN_UART_RX 0 +#define U300_GPIO_PIN_UART_TX 1 +#define U300_GPIO_PIN_UART_CTS 2 +#define U300_GPIO_PIN_UART_RTS 3 +#define U300_GPIO_PIN_CAM_MAIN_STANDBY 4 /* Camera MAIN standby */ +#define U300_GPIO_PIN_GPIO05 5 /* Unrouted */ +#define U300_GPIO_PIN_MS_CD 6 /* Memory Stick Card insertion */ +#define U300_GPIO_PIN_GPIO07 7 /* Test point TP2430 */ + +#define U300_GPIO_PIN_GPIO08 8 /* Test point TP2437 */ +#define U300_GPIO_PIN_GPIO09 9 /* Test point TP2431 */ +#define U300_GPIO_PIN_GPIO10 10 /* Test point TP2432 */ +#define U300_GPIO_PIN_MMC_CLKRET 11 /* Clock return from MMC/SD card */ +#define U300_GPIO_PIN_MMC_CD 12 /* MMC Card insertion detection */ +#define U300_GPIO_PIN_CAM_SUB_STANDBY 13 /* Camera SUB standby */ +#define U300_GPIO_PIN_GPIO14 14 /* Test point TP2436 */ +#define U300_GPIO_PIN_GPIO15 15 /* Unrouted */ + +#define U300_GPIO_PIN_GPIO16 16 /* Test point TP2438 */ +#define U300_GPIO_PIN_PHFSENSE 17 /* Headphone jack sensing */ +#define U300_GPIO_PIN_GPIO18 18 /* Test point TP2439 */ +#define U300_GPIO_PIN_GPIO19 19 /* Routed somewhere */ +#define U300_GPIO_PIN_GPIO20 20 /* Unrouted */ +#define U300_GPIO_PIN_GPIO21 21 /* Unrouted */ +#define U300_GPIO_PIN_GPIO22 22 /* Unrouted */ +#define U300_GPIO_PIN_GPIO23 23 /* Unrouted */ + +#define U300_GPIO_PIN_GPIO24 24 /* Unrouted */ +#define U300_GPIO_PIN_GPIO25 25 /* Unrouted */ +#define U300_GPIO_PIN_GPIO26 26 /* Unrouted */ +#define U300_GPIO_PIN_GPIO27 27 /* Unrouted */ +#define U300_GPIO_PIN_GPIO28 28 /* Unrouted */ +#define U300_GPIO_PIN_GPIO29 29 /* Unrouted */ +#define U300_GPIO_PIN_GPIO30 30 /* Unrouted */ +#define U300_GPIO_PIN_GPIO31 31 /* Unrouted */ + +#define U300_GPIO_PIN_GPIO32 32 /* Unrouted */ +#define U300_GPIO_PIN_GPIO33 33 /* Unrouted */ +#define U300_GPIO_PIN_GPIO34 34 /* Unrouted */ +#define U300_GPIO_PIN_GPIO35 35 /* Unrouted */ +#define U300_GPIO_PIN_GPIO36 36 /* Unrouted */ +#define U300_GPIO_PIN_GPIO37 37 /* Unrouted */ +#define U300_GPIO_PIN_GPIO38 38 /* Unrouted */ +#define U300_GPIO_PIN_GPIO39 39 /* Unrouted */ + +#ifdef CONFIG_MACH_U300_BS335 + +#define U300_GPIO_PIN_GPIO40 40 /* Unrouted */ +#define U300_GPIO_PIN_GPIO41 41 /* Unrouted */ +#define U300_GPIO_PIN_GPIO42 42 /* Unrouted */ +#define U300_GPIO_PIN_GPIO43 43 /* Unrouted */ +#define U300_GPIO_PIN_GPIO44 44 /* Unrouted */ +#define U300_GPIO_PIN_GPIO45 45 /* Unrouted */ +#define U300_GPIO_PIN_GPIO46 46 /* Unrouted */ +#define U300_GPIO_PIN_GPIO47 47 /* Unrouted */ + +#define U300_GPIO_PIN_GPIO48 48 /* Unrouted */ +#define U300_GPIO_PIN_GPIO49 49 /* Unrouted */ +#define U300_GPIO_PIN_GPIO50 50 /* Unrouted */ +#define U300_GPIO_PIN_GPIO51 51 /* Unrouted */ +#define U300_GPIO_PIN_GPIO52 52 /* Unrouted */ +#define U300_GPIO_PIN_GPIO53 53 /* Unrouted */ +#define U300_GPIO_PIN_GPIO54 54 /* Unrouted */ +#define U300_GPIO_PIN_GPIO55 55 /* Unrouted */ +#endif + +#endif diff --git a/arch/arm/mm/iomap.c b/arch/arm/mm/iomap.c index 430df1a..e62956e 100644 --- a/arch/arm/mm/iomap.c +++ b/arch/arm/mm/iomap.c @@ -35,27 +35,6 @@ EXPORT_SYMBOL(pcibios_min_mem); unsigned int pci_flags = PCI_REASSIGN_ALL_RSRC; EXPORT_SYMBOL(pci_flags); -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) -{ - resource_size_t start = pci_resource_start(dev, bar); - resource_size_t len = pci_resource_len(dev, bar); - unsigned long flags = pci_resource_flags(dev, bar); - - if (!len || !start) - return NULL; - if (maxlen && len > maxlen) - len = maxlen; - if (flags & IORESOURCE_IO) - return ioport_map(start, len); - if (flags & IORESOURCE_MEM) { - if (flags & IORESOURCE_CACHEABLE) - return ioremap(start, len); - return ioremap_nocache(start, len); - } - return NULL; -} -EXPORT_SYMBOL(pci_iomap); - void pci_iounmap(struct pci_dev *dev, void __iomem *addr) { if ((unsigned long)addr >= VMALLOC_START && diff --git a/arch/arm/plat-omap/include/plat/iommu.h b/arch/arm/plat-omap/include/plat/iommu.h index a1d79ee..88be3e6 100644 --- a/arch/arm/plat-omap/include/plat/iommu.h +++ b/arch/arm/plat-omap/include/plat/iommu.h @@ -111,6 +111,32 @@ struct iommu_platform_data { u32 da_end; }; +/** + * struct iommu_arch_data - omap iommu private data + * @name: name of the iommu device + * @iommu_dev: handle of the iommu device + * + * This is an omap iommu private data object, which binds an iommu user + * to its iommu device. This object should be placed at the iommu user's + * dev_archdata so generic IOMMU API can be used without having to + * utilize omap-specific plumbing anymore. + */ +struct omap_iommu_arch_data { + const char *name; + struct omap_iommu *iommu_dev; +}; + +/** + * dev_to_omap_iommu() - retrieves an omap iommu object from a user device + * @dev: iommu client device + */ +static inline struct omap_iommu *dev_to_omap_iommu(struct device *dev) +{ + struct omap_iommu_arch_data *arch_data = dev->archdata.iommu; + + return arch_data->iommu_dev; +} + /* IOMMU errors */ #define OMAP_IOMMU_ERR_TLB_MISS (1 << 0) #define OMAP_IOMMU_ERR_TRANS_FAULT (1 << 1) @@ -163,8 +189,8 @@ extern int omap_iommu_set_isr(const char *name, void *priv), void *isr_priv); -extern void omap_iommu_save_ctx(struct omap_iommu *obj); -extern void omap_iommu_restore_ctx(struct omap_iommu *obj); +extern void omap_iommu_save_ctx(struct device *dev); +extern void omap_iommu_restore_ctx(struct device *dev); extern int omap_install_iommu_arch(const struct iommu_functions *ops); extern void omap_uninstall_iommu_arch(const struct iommu_functions *ops); @@ -176,6 +202,5 @@ extern ssize_t omap_iommu_dump_ctx(struct omap_iommu *obj, char *buf, ssize_t len); extern size_t omap_dump_tlb_entries(struct omap_iommu *obj, char *buf, ssize_t len); -struct device *omap_find_iommu_device(const char *name); #endif /* __MACH_IOMMU_H */ diff --git a/arch/arm/plat-omap/include/plat/iovmm.h b/arch/arm/plat-omap/include/plat/iovmm.h index 6af1a91..498e57c 100644 --- a/arch/arm/plat-omap/include/plat/iovmm.h +++ b/arch/arm/plat-omap/include/plat/iovmm.h @@ -72,18 +72,18 @@ struct iovm_struct { #define IOVMF_DA_FIXED (1 << (4 + IOVMF_SW_SHIFT)) -extern struct iovm_struct *omap_find_iovm_area(struct omap_iommu *obj, u32 da); +extern struct iovm_struct *omap_find_iovm_area(struct device *dev, u32 da); extern u32 -omap_iommu_vmap(struct iommu_domain *domain, struct omap_iommu *obj, u32 da, +omap_iommu_vmap(struct iommu_domain *domain, struct device *dev, u32 da, const struct sg_table *sgt, u32 flags); extern struct sg_table *omap_iommu_vunmap(struct iommu_domain *domain, - struct omap_iommu *obj, u32 da); + struct device *dev, u32 da); extern u32 -omap_iommu_vmalloc(struct iommu_domain *domain, struct omap_iommu *obj, +omap_iommu_vmalloc(struct iommu_domain *domain, struct device *dev, u32 da, size_t bytes, u32 flags); extern void -omap_iommu_vfree(struct iommu_domain *domain, struct omap_iommu *obj, +omap_iommu_vfree(struct iommu_domain *domain, struct device *dev, const u32 da); -extern void *omap_da_to_va(struct omap_iommu *obj, u32 da); +extern void *omap_da_to_va(struct device *dev, u32 da); #endif /* __IOMMU_MMAP_H */ diff --git a/arch/arm/plat-samsung/include/plat/keypad.h b/arch/arm/plat-samsung/include/plat/keypad.h index b59a648..c81ace3 100644 --- a/arch/arm/plat-samsung/include/plat/keypad.h +++ b/arch/arm/plat-samsung/include/plat/keypad.h @@ -13,32 +13,7 @@ #ifndef __PLAT_SAMSUNG_KEYPAD_H #define __PLAT_SAMSUNG_KEYPAD_H -#include <linux/input/matrix_keypad.h> - -#define SAMSUNG_MAX_ROWS 8 -#define SAMSUNG_MAX_COLS 8 - -/** - * struct samsung_keypad_platdata - Platform device data for Samsung Keypad. - * @keymap_data: pointer to &matrix_keymap_data. - * @rows: number of keypad row supported. - * @cols: number of keypad col supported. - * @no_autorepeat: disable key autorepeat. - * @wakeup: controls whether the device should be set up as wakeup source. - * @cfg_gpio: configure the GPIO. - * - * Initialisation data specific to either the machine or the platform - * for the device driver to use or call-back when configuring gpio. - */ -struct samsung_keypad_platdata { - const struct matrix_keymap_data *keymap_data; - unsigned int rows; - unsigned int cols; - bool no_autorepeat; - bool wakeup; - - void (*cfg_gpio)(unsigned int rows, unsigned int cols); -}; +#include <linux/input/samsung-keypad.h> /** * samsung_keypad_set_platdata - Set platform data for Samsung Keypad device. diff --git a/arch/blackfin/configs/BF518F-EZBRD_defconfig b/arch/blackfin/configs/BF518F-EZBRD_defconfig index 5edcb58..0b7039c 100644 --- a/arch/blackfin/configs/BF518F-EZBRD_defconfig +++ b/arch/blackfin/configs/BF518F-EZBRD_defconfig @@ -80,7 +80,7 @@ CONFIG_I2C_CHARDEV=y CONFIG_I2C_BLACKFIN_TWI=y CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=100 CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y # CONFIG_HWMON is not set diff --git a/arch/blackfin/configs/BF526-EZBRD_defconfig b/arch/blackfin/configs/BF526-EZBRD_defconfig index 2e54957..5553205 100644 --- a/arch/blackfin/configs/BF526-EZBRD_defconfig +++ b/arch/blackfin/configs/BF526-EZBRD_defconfig @@ -97,7 +97,7 @@ CONFIG_I2C_CHARDEV=m CONFIG_I2C_BLACKFIN_TWI=y CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=100 CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y CONFIG_WATCHDOG=y diff --git a/arch/blackfin/configs/BF527-AD7160-EVAL_defconfig b/arch/blackfin/configs/BF527-AD7160-EVAL_defconfig index ad0881b..d95658f 100644 --- a/arch/blackfin/configs/BF527-AD7160-EVAL_defconfig +++ b/arch/blackfin/configs/BF527-AD7160-EVAL_defconfig @@ -68,7 +68,7 @@ CONFIG_I2C_ALGOBIT=y CONFIG_I2C_BLACKFIN_TWI=y CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=400 CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y # CONFIG_HWMON is not set diff --git a/arch/blackfin/configs/BF527-EZKIT-V2_defconfig b/arch/blackfin/configs/BF527-EZKIT-V2_defconfig index 8465b3e..498f64a 100644 --- a/arch/blackfin/configs/BF527-EZKIT-V2_defconfig +++ b/arch/blackfin/configs/BF527-EZKIT-V2_defconfig @@ -105,7 +105,7 @@ CONFIG_I2C_CHARDEV=m CONFIG_I2C_BLACKFIN_TWI=y CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=100 CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y # CONFIG_HWMON is not set diff --git a/arch/blackfin/configs/BF527-EZKIT_defconfig b/arch/blackfin/configs/BF527-EZKIT_defconfig index 5e7321b..72e0317 100644 --- a/arch/blackfin/configs/BF527-EZKIT_defconfig +++ b/arch/blackfin/configs/BF527-EZKIT_defconfig @@ -99,7 +99,7 @@ CONFIG_I2C_CHARDEV=m CONFIG_I2C_BLACKFIN_TWI=y CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=100 CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y # CONFIG_HWMON is not set diff --git a/arch/blackfin/configs/BF533-EZKIT_defconfig b/arch/blackfin/configs/BF533-EZKIT_defconfig index a7eb54b..2f075e0 100644 --- a/arch/blackfin/configs/BF533-EZKIT_defconfig +++ b/arch/blackfin/configs/BF533-EZKIT_defconfig @@ -81,7 +81,7 @@ CONFIG_SERIAL_BFIN_CONSOLE=y # CONFIG_LEGACY_PTYS is not set # CONFIG_HW_RANDOM is not set CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y # CONFIG_HWMON is not set diff --git a/arch/blackfin/configs/BF533-STAMP_defconfig b/arch/blackfin/configs/BF533-STAMP_defconfig index b90d379..ab38a82 100644 --- a/arch/blackfin/configs/BF533-STAMP_defconfig +++ b/arch/blackfin/configs/BF533-STAMP_defconfig @@ -84,7 +84,7 @@ CONFIG_I2C=m CONFIG_I2C_CHARDEV=m CONFIG_I2C_GPIO=m CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y # CONFIG_HWMON is not set diff --git a/arch/blackfin/configs/BF537-STAMP_defconfig b/arch/blackfin/configs/BF537-STAMP_defconfig index 0053625..5c802d6 100644 --- a/arch/blackfin/configs/BF537-STAMP_defconfig +++ b/arch/blackfin/configs/BF537-STAMP_defconfig @@ -94,7 +94,7 @@ CONFIG_I2C_CHARDEV=m CONFIG_I2C_BLACKFIN_TWI=m CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=100 CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y # CONFIG_HWMON is not set diff --git a/arch/blackfin/configs/BF538-EZKIT_defconfig b/arch/blackfin/configs/BF538-EZKIT_defconfig index 580bf42..972aa62 100644 --- a/arch/blackfin/configs/BF538-EZKIT_defconfig +++ b/arch/blackfin/configs/BF538-EZKIT_defconfig @@ -101,7 +101,7 @@ CONFIG_I2C=m CONFIG_I2C_BLACKFIN_TWI=m CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=100 CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y # CONFIG_HWMON is not set diff --git a/arch/blackfin/configs/BF548-EZKIT_defconfig b/arch/blackfin/configs/BF548-EZKIT_defconfig index 0e6d841..7a1e3bf 100644 --- a/arch/blackfin/configs/BF548-EZKIT_defconfig +++ b/arch/blackfin/configs/BF548-EZKIT_defconfig @@ -113,7 +113,7 @@ CONFIG_I2C_CHARDEV=y CONFIG_I2C_BLACKFIN_TWI=y CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=100 CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y # CONFIG_HWMON is not set diff --git a/arch/blackfin/configs/BF561-ACVILON_defconfig b/arch/blackfin/configs/BF561-ACVILON_defconfig index 77a27e3..0fdc4ec 100644 --- a/arch/blackfin/configs/BF561-ACVILON_defconfig +++ b/arch/blackfin/configs/BF561-ACVILON_defconfig @@ -85,7 +85,7 @@ CONFIG_I2C=y CONFIG_I2C_CHARDEV=y CONFIG_I2C_PCA_PLATFORM=y CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_SPI_SPIDEV=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y diff --git a/arch/blackfin/configs/BF561-EZKIT-SMP_defconfig b/arch/blackfin/configs/BF561-EZKIT-SMP_defconfig index f5ed34e..78adbbf 100644 --- a/arch/blackfin/configs/BF561-EZKIT-SMP_defconfig +++ b/arch/blackfin/configs/BF561-EZKIT-SMP_defconfig @@ -84,7 +84,7 @@ CONFIG_SERIAL_BFIN_CONSOLE=y # CONFIG_LEGACY_PTYS is not set # CONFIG_HW_RANDOM is not set CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y # CONFIG_HWMON is not set diff --git a/arch/blackfin/configs/BF561-EZKIT_defconfig b/arch/blackfin/configs/BF561-EZKIT_defconfig index d7ff2ae..d3cd0f5 100644 --- a/arch/blackfin/configs/BF561-EZKIT_defconfig +++ b/arch/blackfin/configs/BF561-EZKIT_defconfig @@ -86,7 +86,7 @@ CONFIG_SERIAL_BFIN_CONSOLE=y # CONFIG_LEGACY_PTYS is not set # CONFIG_HW_RANDOM is not set CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y # CONFIG_HWMON is not set diff --git a/arch/blackfin/configs/BlackStamp_defconfig b/arch/blackfin/configs/BlackStamp_defconfig index 8501431..7b982d0 100644 --- a/arch/blackfin/configs/BlackStamp_defconfig +++ b/arch/blackfin/configs/BlackStamp_defconfig @@ -80,7 +80,7 @@ CONFIG_I2C=m CONFIG_I2C_CHARDEV=m CONFIG_I2C_GPIO=m CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_SPI_SPIDEV=m # CONFIG_HWMON is not set CONFIG_WATCHDOG=y diff --git a/arch/blackfin/configs/CM-BF527_defconfig b/arch/blackfin/configs/CM-BF527_defconfig index dbf750c..c280a50 100644 --- a/arch/blackfin/configs/CM-BF527_defconfig +++ b/arch/blackfin/configs/CM-BF527_defconfig @@ -88,7 +88,7 @@ CONFIG_I2C_CHARDEV=m CONFIG_I2C_BLACKFIN_TWI=m CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=100 CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y CONFIG_WATCHDOG=y diff --git a/arch/blackfin/configs/CM-BF533_defconfig b/arch/blackfin/configs/CM-BF533_defconfig index 07ffbda..c940a1e 100644 --- a/arch/blackfin/configs/CM-BF533_defconfig +++ b/arch/blackfin/configs/CM-BF533_defconfig @@ -57,7 +57,7 @@ CONFIG_SERIAL_BFIN_CONSOLE=y # CONFIG_LEGACY_PTYS is not set # CONFIG_HW_RANDOM is not set CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y # CONFIG_HWMON is not set # CONFIG_USB_SUPPORT is not set CONFIG_MMC=y diff --git a/arch/blackfin/configs/CM-BF537E_defconfig b/arch/blackfin/configs/CM-BF537E_defconfig index 707cbf8..2e47df7 100644 --- a/arch/blackfin/configs/CM-BF537E_defconfig +++ b/arch/blackfin/configs/CM-BF537E_defconfig @@ -78,7 +78,7 @@ CONFIG_SERIAL_BFIN_UART1=y # CONFIG_LEGACY_PTYS is not set # CONFIG_HW_RANDOM is not set CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y CONFIG_USB_GADGET=m diff --git a/arch/blackfin/configs/CM-BF537U_defconfig b/arch/blackfin/configs/CM-BF537U_defconfig index 4596935..6da629f 100644 --- a/arch/blackfin/configs/CM-BF537U_defconfig +++ b/arch/blackfin/configs/CM-BF537U_defconfig @@ -72,7 +72,7 @@ CONFIG_SERIAL_BFIN_UART1=y # CONFIG_LEGACY_PTYS is not set # CONFIG_HW_RANDOM is not set CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y CONFIG_USB_GADGET=y diff --git a/arch/blackfin/configs/CM-BF548_defconfig b/arch/blackfin/configs/CM-BF548_defconfig index 9f1d084..349922b 100644 --- a/arch/blackfin/configs/CM-BF548_defconfig +++ b/arch/blackfin/configs/CM-BF548_defconfig @@ -89,7 +89,7 @@ CONFIG_I2C=y CONFIG_I2C_CHARDEV=y CONFIG_I2C_BLACKFIN_TWI=y CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y # CONFIG_HWMON is not set CONFIG_WATCHDOG=y CONFIG_BFIN_WDT=y diff --git a/arch/blackfin/configs/CM-BF561_defconfig b/arch/blackfin/configs/CM-BF561_defconfig index 6c7b215..0456dea 100644 --- a/arch/blackfin/configs/CM-BF561_defconfig +++ b/arch/blackfin/configs/CM-BF561_defconfig @@ -78,7 +78,7 @@ CONFIG_SERIAL_BFIN_CONSOLE=y # CONFIG_LEGACY_PTYS is not set # CONFIG_HW_RANDOM is not set CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y CONFIG_USB_GADGET=m diff --git a/arch/blackfin/configs/DNP5370_defconfig b/arch/blackfin/configs/DNP5370_defconfig index b192acf..89162d0 100644 --- a/arch/blackfin/configs/DNP5370_defconfig +++ b/arch/blackfin/configs/DNP5370_defconfig @@ -78,7 +78,7 @@ CONFIG_I2C=y CONFIG_I2C_CHARDEV=y CONFIG_I2C_BLACKFIN_TWI=y CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_SPI_SPIDEV=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y diff --git a/arch/blackfin/configs/H8606_defconfig b/arch/blackfin/configs/H8606_defconfig index 06e9f49..a26436b 100644 --- a/arch/blackfin/configs/H8606_defconfig +++ b/arch/blackfin/configs/H8606_defconfig @@ -68,7 +68,7 @@ CONFIG_SERIAL_BFIN_CONSOLE=y # CONFIG_LEGACY_PTYS is not set # CONFIG_HW_RANDOM is not set CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_SPI_SPIDEV=y CONFIG_WATCHDOG=y CONFIG_SOUND=m diff --git a/arch/blackfin/configs/IP0X_defconfig b/arch/blackfin/configs/IP0X_defconfig index 5e797cf..6479915 100644 --- a/arch/blackfin/configs/IP0X_defconfig +++ b/arch/blackfin/configs/IP0X_defconfig @@ -70,7 +70,7 @@ CONFIG_SERIAL_BFIN_CONSOLE=y # CONFIG_LEGACY_PTYS is not set CONFIG_HW_RANDOM=y CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y # CONFIG_HWMON is not set CONFIG_WATCHDOG=y CONFIG_USB=y diff --git a/arch/blackfin/configs/PNAV-10_defconfig b/arch/blackfin/configs/PNAV-10_defconfig index a566a2f..8fd9b44 100644 --- a/arch/blackfin/configs/PNAV-10_defconfig +++ b/arch/blackfin/configs/PNAV-10_defconfig @@ -84,7 +84,7 @@ CONFIG_I2C_CHARDEV=y CONFIG_I2C_BLACKFIN_TWI=y CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=100 CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_FB=y CONFIG_FIRMWARE_EDID=y CONFIG_BACKLIGHT_LCD_SUPPORT=y diff --git a/arch/blackfin/configs/SRV1_defconfig b/arch/blackfin/configs/SRV1_defconfig index 12e66cd..0520c16 100644 --- a/arch/blackfin/configs/SRV1_defconfig +++ b/arch/blackfin/configs/SRV1_defconfig @@ -71,7 +71,7 @@ CONFIG_I2C_CHARDEV=y CONFIG_I2C_BLACKFIN_TWI=y CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=100 CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_HWMON=m CONFIG_WATCHDOG=y CONFIG_BFIN_WDT=y diff --git a/arch/blackfin/configs/TCM-BF518_defconfig b/arch/blackfin/configs/TCM-BF518_defconfig index d496ae9..e4ed865 100644 --- a/arch/blackfin/configs/TCM-BF518_defconfig +++ b/arch/blackfin/configs/TCM-BF518_defconfig @@ -92,7 +92,7 @@ CONFIG_I2C_CHARDEV=y CONFIG_I2C_BLACKFIN_TWI=y CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=100 CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y CONFIG_GPIOLIB=y CONFIG_GPIO_SYSFS=y # CONFIG_HWMON is not set diff --git a/arch/blackfin/configs/TCM-BF537_defconfig b/arch/blackfin/configs/TCM-BF537_defconfig index 65f6421..c1f45f1 100644 --- a/arch/blackfin/configs/TCM-BF537_defconfig +++ b/arch/blackfin/configs/TCM-BF537_defconfig @@ -70,7 +70,7 @@ CONFIG_SERIAL_BFIN_UART1=y # CONFIG_LEGACY_PTYS is not set # CONFIG_HW_RANDOM is not set CONFIG_SPI=y -CONFIG_SPI_BFIN=y +CONFIG_SPI_BFIN5XX=y # CONFIG_HWMON is not set CONFIG_WATCHDOG=y CONFIG_BFIN_WDT=y diff --git a/arch/blackfin/include/asm/bfin_serial.h b/arch/blackfin/include/asm/bfin_serial.h index ecacdf3..68bcc3d 100644 --- a/arch/blackfin/include/asm/bfin_serial.h +++ b/arch/blackfin/include/asm/bfin_serial.h @@ -51,9 +51,6 @@ struct bfin_serial_port { #elif ANOMALY_05000363 unsigned int anomaly_threshold; #endif -#ifdef CONFIG_SERIAL_BFIN_HARD_CTSRTS - int scts; -#endif #if defined(CONFIG_SERIAL_BFIN_CTSRTS) || \ defined(CONFIG_SERIAL_BFIN_HARD_CTSRTS) int cts_pin; diff --git a/arch/blackfin/include/asm/cpu.h b/arch/blackfin/include/asm/cpu.h index 0504378..e349631 100644 --- a/arch/blackfin/include/asm/cpu.h +++ b/arch/blackfin/include/asm/cpu.h @@ -14,6 +14,9 @@ struct blackfin_cpudata { struct cpu cpu; unsigned int imemctl; unsigned int dmemctl; +#ifdef CONFIG_SMP + struct task_struct *idle; +#endif }; DECLARE_PER_CPU(struct blackfin_cpudata, cpu_data); diff --git a/arch/blackfin/include/asm/smp.h b/arch/blackfin/include/asm/smp.h index af6c0aa..dc3d144 100644 --- a/arch/blackfin/include/asm/smp.h +++ b/arch/blackfin/include/asm/smp.h @@ -37,7 +37,7 @@ extern unsigned long dcache_invld_count[NR_CPUS]; #endif void smp_icache_flush_range_others(unsigned long start, - unsigned long end); + unsigned long end); #ifdef CONFIG_HOTPLUG_CPU void coreb_die(void); void cpu_die(void); @@ -46,4 +46,7 @@ int __cpu_disable(void); int __cpu_die(unsigned int cpu); #endif +void smp_timer_broadcast(const struct cpumask *mask); + + #endif /* !__ASM_BLACKFIN_SMP_H */ diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c index dfa2525..d6102c8 100644 --- a/arch/blackfin/kernel/setup.c +++ b/arch/blackfin/kernel/setup.c @@ -828,10 +828,18 @@ static inline int __init get_mem_size(void) u32 ddrctl = bfin_read_EBIU_DDRCTL1(); int ret = 0; switch (ddrctl & 0xc0000) { - case DEVSZ_64: ret = 64 / 8; - case DEVSZ_128: ret = 128 / 8; - case DEVSZ_256: ret = 256 / 8; - case DEVSZ_512: ret = 512 / 8; + case DEVSZ_64: + ret = 64 / 8; + break; + case DEVSZ_128: + ret = 128 / 8; + break; + case DEVSZ_256: + ret = 256 / 8; + break; + case DEVSZ_512: + ret = 512 / 8; + break; } switch (ddrctl & 0x30000) { case DEVWD_4: ret *= 2; diff --git a/arch/blackfin/kernel/time-ts.c b/arch/blackfin/kernel/time-ts.c index 1bcf3a3..d98f2d6 100644 --- a/arch/blackfin/kernel/time-ts.c +++ b/arch/blackfin/kernel/time-ts.c @@ -219,7 +219,7 @@ static void __init bfin_gptmr0_clockevent_init(struct clock_event_device *evt) #if defined(CONFIG_TICKSOURCE_CORETMR) /* per-cpu local core timer */ -static DEFINE_PER_CPU(struct clock_event_device, coretmr_events); +DEFINE_PER_CPU(struct clock_event_device, coretmr_events); static int bfin_coretmr_set_next_event(unsigned long cycles, struct clock_event_device *evt) @@ -281,6 +281,7 @@ void bfin_coretmr_init(void) #ifdef CONFIG_CORE_TIMER_IRQ_L1 __attribute__((l1_text)) #endif + irqreturn_t bfin_coretmr_interrupt(int irq, void *dev_id) { int cpu = smp_processor_id(); @@ -306,6 +307,11 @@ void bfin_coretmr_clockevent_init(void) unsigned int cpu = smp_processor_id(); struct clock_event_device *evt = &per_cpu(coretmr_events, cpu); +#ifdef CONFIG_SMP + evt->broadcast = smp_timer_broadcast; +#endif + + evt->name = "bfin_core_timer"; evt->rating = 350; evt->irq = -1; diff --git a/arch/blackfin/mach-bf518/boards/ezbrd.c b/arch/blackfin/mach-bf518/boards/ezbrd.c index d1c0c0c..a2d96d3 100644 --- a/arch/blackfin/mach-bf518/boards/ezbrd.c +++ b/arch/blackfin/mach-bf518/boards/ezbrd.c @@ -61,7 +61,7 @@ static struct physmap_flash_data ezbrd_flash_data = { static struct resource ezbrd_flash_resource = { .start = 0x20000000, -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) .end = 0x202fffff, #else .end = 0x203fffff, @@ -122,6 +122,8 @@ static struct bfin_mii_bus_platform_data bfin_mii_bus_data = { #if defined(CONFIG_NET_DSA_KSZ8893M) || defined(CONFIG_NET_DSA_KSZ8893M_MODULE) .phy_mask = 0xfff7, /* Only probe the port phy connect to the on chip MAC */ #endif + .vlan1_mask = 1, + .vlan2_mask = 2, }; static struct platform_device bfin_mii_bus = { @@ -292,7 +294,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { }; /* SPI controller data */ -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI (0) */ static struct bfin5xx_spi_master bfin_spi0_info = { .num_chipselect = 6, @@ -715,7 +717,7 @@ static struct platform_device *stamp_devices[] __initdata = { #endif #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, &bfin_spi1_device, #endif @@ -777,7 +779,7 @@ static int __init ezbrd_init(void) spi_register_board_info(bfin_spi_board_info, ARRAY_SIZE(bfin_spi_board_info)); /* setup BF518-EZBRD GPIO pin PG11 to AMS2, PG15 to AMS3. */ peripheral_request(P_AMS2, "ParaFlash"); -#if !defined(CONFIG_SPI_BFIN) && !defined(CONFIG_SPI_BFIN_MODULE) +#if !defined(CONFIG_SPI_BFIN5XX) && !defined(CONFIG_SPI_BFIN5XX_MODULE) peripheral_request(P_AMS3, "ParaFlash"); #endif return 0; diff --git a/arch/blackfin/mach-bf518/boards/tcm-bf518.c b/arch/blackfin/mach-bf518/boards/tcm-bf518.c index 5470bf8..f271310 100644 --- a/arch/blackfin/mach-bf518/boards/tcm-bf518.c +++ b/arch/blackfin/mach-bf518/boards/tcm-bf518.c @@ -228,7 +228,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { }; /* SPI controller data */ -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI (0) */ static struct bfin5xx_spi_master bfin_spi0_info = { .num_chipselect = 6, @@ -635,7 +635,7 @@ static struct platform_device *tcm_devices[] __initdata = { &bfin_mac_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, &bfin_spi1_device, #endif diff --git a/arch/blackfin/mach-bf527/boards/ad7160eval.c b/arch/blackfin/mach-bf527/boards/ad7160eval.c index 5bc6938..c8d5d2b 100644 --- a/arch/blackfin/mach-bf527/boards/ad7160eval.c +++ b/arch/blackfin/mach-bf527/boards/ad7160eval.c @@ -334,7 +334,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { #endif }; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI controller data */ static struct bfin5xx_spi_master bfin_spi0_info = { .num_chipselect = MAX_CTRL_CS + MAX_BLACKFIN_GPIOS, @@ -744,7 +744,7 @@ static struct platform_device *stamp_devices[] __initdata = { &bfin_mac_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif diff --git a/arch/blackfin/mach-bf527/boards/cm_bf527.c b/arch/blackfin/mach-bf527/boards/cm_bf527.c index cd28969..7330607 100644 --- a/arch/blackfin/mach-bf527/boards/cm_bf527.c +++ b/arch/blackfin/mach-bf527/boards/cm_bf527.c @@ -444,7 +444,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { #endif }; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI controller data */ static struct bfin5xx_spi_master bfin_spi0_info = { .num_chipselect = 8, @@ -893,7 +893,7 @@ static struct platform_device *cmbf527_devices[] __initdata = { &net2272_bfin_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif diff --git a/arch/blackfin/mach-bf527/boards/ezbrd.c b/arch/blackfin/mach-bf527/boards/ezbrd.c index 9f792ea..db3ecfc 100644 --- a/arch/blackfin/mach-bf527/boards/ezbrd.c +++ b/arch/blackfin/mach-bf527/boards/ezbrd.c @@ -371,7 +371,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { #endif }; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI controller data */ static struct bfin5xx_spi_master bfin_spi0_info = { .num_chipselect = 8, @@ -776,7 +776,7 @@ static struct platform_device *stamp_devices[] __initdata = { &bfin_mac_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif diff --git a/arch/blackfin/mach-bf527/boards/ezkit.c b/arch/blackfin/mach-bf527/boards/ezkit.c index 3ecafff..dfdd8e6 100644 --- a/arch/blackfin/mach-bf527/boards/ezkit.c +++ b/arch/blackfin/mach-bf527/boards/ezkit.c @@ -664,7 +664,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { #endif }; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI controller data */ static struct bfin5xx_spi_master bfin_spi0_info = { .num_chipselect = 8, @@ -1189,7 +1189,7 @@ static struct platform_device *stamp_devices[] __initdata = { &net2272_bfin_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif diff --git a/arch/blackfin/mach-bf527/boards/tll6527m.c b/arch/blackfin/mach-bf527/boards/tll6527m.c index 3a92c43..360e97f 100644 --- a/arch/blackfin/mach-bf527/boards/tll6527m.c +++ b/arch/blackfin/mach-bf527/boards/tll6527m.c @@ -448,7 +448,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { #endif }; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI controller data */ static struct bfin5xx_spi_master bfin_spi0_info = { .num_chipselect = EXP_GPIO_SPISEL_BASE + 8 + MAX_CTRL_CS, @@ -831,7 +831,7 @@ static struct platform_device *tll6527m_devices[] __initdata = { &bfin_mac_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif diff --git a/arch/blackfin/mach-bf533/boards/H8606.c b/arch/blackfin/mach-bf533/boards/H8606.c index 47cadd3..6cb7b3e 100644 --- a/arch/blackfin/mach-bf533/boards/H8606.c +++ b/arch/blackfin/mach-bf533/boards/H8606.c @@ -125,7 +125,7 @@ static struct platform_device net2272_bfin_device = { }; #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* all SPI peripherals info goes here */ #if defined(CONFIG_MTD_M25P80) || defined(CONFIG_MTD_M25P80_MODULE) @@ -398,7 +398,7 @@ static struct platform_device *h8606_devices[] __initdata = { &net2272_bfin_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif @@ -428,7 +428,7 @@ static int __init H8606_init(void) printk(KERN_INFO "HV Sistemas H8606 board support by http://www.hvsistemas.com\n"); printk(KERN_INFO "%s(): registering device resources\n", __func__); platform_add_devices(h8606_devices, ARRAY_SIZE(h8606_devices)); -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) spi_register_board_info(bfin_spi_board_info, ARRAY_SIZE(bfin_spi_board_info)); #endif return 0; diff --git a/arch/blackfin/mach-bf533/boards/blackstamp.c b/arch/blackfin/mach-bf533/boards/blackstamp.c index 18817d5..de44a37 100644 --- a/arch/blackfin/mach-bf533/boards/blackstamp.c +++ b/arch/blackfin/mach-bf533/boards/blackstamp.c @@ -146,7 +146,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { #endif }; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI (0) */ static struct resource bfin_spi0_resource[] = { [0] = { @@ -422,7 +422,7 @@ static struct platform_device *stamp_devices[] __initdata = { #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif diff --git a/arch/blackfin/mach-bf533/boards/cm_bf533.c b/arch/blackfin/mach-bf533/boards/cm_bf533.c index 2c8f30e..fe47e04 100644 --- a/arch/blackfin/mach-bf533/boards/cm_bf533.c +++ b/arch/blackfin/mach-bf533/boards/cm_bf533.c @@ -29,7 +29,7 @@ */ const char bfin_board_name[] = "Bluetechnix CM BF533"; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* all SPI peripherals info goes here */ #if defined(CONFIG_MTD_M25P80) || defined(CONFIG_MTD_M25P80_MODULE) static struct mtd_partition bfin_spi_flash_partitions[] = { @@ -536,7 +536,7 @@ static struct platform_device *cm_bf533_devices[] __initdata = { &net2272_bfin_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif @@ -549,7 +549,7 @@ static int __init cm_bf533_init(void) { printk(KERN_INFO "%s(): registering device resources\n", __func__); platform_add_devices(cm_bf533_devices, ARRAY_SIZE(cm_bf533_devices)); -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) spi_register_board_info(bfin_spi_board_info, ARRAY_SIZE(bfin_spi_board_info)); #endif return 0; diff --git a/arch/blackfin/mach-bf533/boards/ezkit.c b/arch/blackfin/mach-bf533/boards/ezkit.c index 144556e..07811c2 100644 --- a/arch/blackfin/mach-bf533/boards/ezkit.c +++ b/arch/blackfin/mach-bf533/boards/ezkit.c @@ -245,7 +245,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { #endif }; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI (0) */ static struct resource bfin_spi0_resource[] = { [0] = { @@ -484,7 +484,7 @@ static struct platform_device *ezkit_devices[] __initdata = { &smc91x_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif diff --git a/arch/blackfin/mach-bf533/boards/ip0x.c b/arch/blackfin/mach-bf533/boards/ip0x.c index b597d4e..e303dae 100644 --- a/arch/blackfin/mach-bf533/boards/ip0x.c +++ b/arch/blackfin/mach-bf533/boards/ip0x.c @@ -104,7 +104,7 @@ static struct platform_device dm9000_device2 = { #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* all SPI peripherals info goes here */ #if defined(CONFIG_MMC_SPI) || defined(CONFIG_MMC_SPI_MODULE) @@ -270,7 +270,7 @@ static struct platform_device *ip0x_devices[] __initdata = { #endif #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &spi_bfin_master_device, #endif diff --git a/arch/blackfin/mach-bf533/boards/stamp.c b/arch/blackfin/mach-bf533/boards/stamp.c index 2afd02e..ce88a71 100644 --- a/arch/blackfin/mach-bf533/boards/stamp.c +++ b/arch/blackfin/mach-bf533/boards/stamp.c @@ -219,9 +219,10 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { }, #endif -#if defined(CONFIG_SND_BF5XX_SOC_AD183X) || defined(CONFIG_SND_BF5XX_SOC_AD183X_MODULE) +#if defined(CONFIG_SND_BF5XX_SOC_AD1836) || \ + defined(CONFIG_SND_BF5XX_SOC_AD1836_MODULE) { - .modalias = "ad183x", + .modalias = "ad1836", .max_speed_hz = 3125000, /* max spi clock (SCK) speed in HZ */ .bus_num = 0, .chip_select = 4, @@ -251,7 +252,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { #endif }; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI (0) */ static struct resource bfin_spi0_resource[] = { [0] = { @@ -471,7 +472,7 @@ static struct i2c_gpio_platform_data i2c_gpio_data = { .scl_pin = GPIO_PF3, .sda_is_open_drain = 0, .scl_is_open_drain = 0, - .udelay = 40, + .udelay = 10, }; static struct platform_device i2c_gpio_device = { @@ -540,27 +541,150 @@ static struct platform_device bfin_dpmc = { }, }; +#if defined(CONFIG_SND_BF5XX_I2S) || defined(CONFIG_SND_BF5XX_I2S_MODULE) || \ + defined(CONFIG_SND_BF5XX_TDM) || defined(CONFIG_SND_BF5XX_TDM_MODULE) \ + || defined(CONFIG_SND_BF5XX_AC97) || \ + defined(CONFIG_SND_BF5XX_AC97_MODULE) + +#include <asm/bfin_sport.h> + +#define SPORT_REQ(x) \ + [x] = {P_SPORT##x##_TFS, P_SPORT##x##_DTPRI, P_SPORT##x##_TSCLK, \ + P_SPORT##x##_RFS, P_SPORT##x##_DRPRI, P_SPORT##x##_RSCLK, 0} + +static const u16 bfin_snd_pin[][7] = { + SPORT_REQ(0), + SPORT_REQ(1), +}; + +static struct bfin_snd_platform_data bfin_snd_data[] = { + { + .pin_req = &bfin_snd_pin[0][0], + }, + { + .pin_req = &bfin_snd_pin[1][0], + }, +}; + +#define BFIN_SND_RES(x) \ + [x] = { \ + { \ + .start = SPORT##x##_TCR1, \ + .end = SPORT##x##_TCR1, \ + .flags = IORESOURCE_MEM \ + }, \ + { \ + .start = CH_SPORT##x##_RX, \ + .end = CH_SPORT##x##_RX, \ + .flags = IORESOURCE_DMA, \ + }, \ + { \ + .start = CH_SPORT##x##_TX, \ + .end = CH_SPORT##x##_TX, \ + .flags = IORESOURCE_DMA, \ + }, \ + { \ + .start = IRQ_SPORT##x##_ERROR, \ + .end = IRQ_SPORT##x##_ERROR, \ + .flags = IORESOURCE_IRQ, \ + } \ + } + +static struct resource bfin_snd_resources[][4] = { + BFIN_SND_RES(0), + BFIN_SND_RES(1), +}; +#endif + #if defined(CONFIG_SND_BF5XX_I2S) || defined(CONFIG_SND_BF5XX_I2S_MODULE) +static struct platform_device bfin_i2s_pcm = { + .name = "bfin-i2s-pcm-audio", + .id = -1, +}; +#endif + +#if defined(CONFIG_SND_BF5XX_TDM) || defined(CONFIG_SND_BF5XX_TDM_MODULE) +static struct platform_device bfin_tdm_pcm = { + .name = "bfin-tdm-pcm-audio", + .id = -1, +}; +#endif + +#if defined(CONFIG_SND_BF5XX_AC97) || defined(CONFIG_SND_BF5XX_AC97_MODULE) +static struct platform_device bfin_ac97_pcm = { + .name = "bfin-ac97-pcm-audio", + .id = -1, +}; +#endif + +#if defined(CONFIG_SND_BF5XX_SOC_AD73311) || \ + defined(CONFIG_SND_BF5XX_SOC_AD73311_MODULE) +static const unsigned ad73311_gpio[] = { + GPIO_PF4, +}; + +static struct platform_device bfin_ad73311_machine = { + .name = "bfin-snd-ad73311", + .id = 1, + .dev = { + .platform_data = (void *)ad73311_gpio, + }, +}; +#endif + +#if defined(CONFIG_SND_SOC_AD73311) || defined(CONFIG_SND_SOC_AD73311_MODULE) +static struct platform_device bfin_ad73311_codec_device = { + .name = "ad73311", + .id = -1, +}; +#endif + +#if defined(CONFIG_SND_SOC_AD74111) || defined(CONFIG_SND_SOC_AD74111_MODULE) +static struct platform_device bfin_ad74111_codec_device = { + .name = "ad74111", + .id = -1, +}; +#endif + +#if defined(CONFIG_SND_BF5XX_SOC_I2S) || \ + defined(CONFIG_SND_BF5XX_SOC_I2S_MODULE) static struct platform_device bfin_i2s = { .name = "bfin-i2s", .id = CONFIG_SND_BF5XX_SPORT_NUM, - /* TODO: add platform data here */ + .num_resources = + ARRAY_SIZE(bfin_snd_resources[CONFIG_SND_BF5XX_SPORT_NUM]), + .resource = bfin_snd_resources[CONFIG_SND_BF5XX_SPORT_NUM], + .dev = { + .platform_data = &bfin_snd_data[CONFIG_SND_BF5XX_SPORT_NUM], + }, }; #endif -#if defined(CONFIG_SND_BF5XX_TDM) || defined(CONFIG_SND_BF5XX_TDM_MODULE) +#if defined(CONFIG_SND_BF5XX_SOC_TDM) || \ + defined(CONFIG_SND_BF5XX_SOC_TDM_MODULE) static struct platform_device bfin_tdm = { .name = "bfin-tdm", .id = CONFIG_SND_BF5XX_SPORT_NUM, - /* TODO: add platform data here */ + .num_resources = + ARRAY_SIZE(bfin_snd_resources[CONFIG_SND_BF5XX_SPORT_NUM]), + .resource = bfin_snd_resources[CONFIG_SND_BF5XX_SPORT_NUM], + .dev = { + .platform_data = &bfin_snd_data[CONFIG_SND_BF5XX_SPORT_NUM], + }, }; #endif -#if defined(CONFIG_SND_BF5XX_AC97) || defined(CONFIG_SND_BF5XX_AC97_MODULE) +#if defined(CONFIG_SND_BF5XX_SOC_AC97) || \ + defined(CONFIG_SND_BF5XX_SOC_AC97_MODULE) static struct platform_device bfin_ac97 = { .name = "bfin-ac97", .id = CONFIG_SND_BF5XX_SPORT_NUM, - /* TODO: add platform data here */ + .num_resources = + ARRAY_SIZE(bfin_snd_resources[CONFIG_SND_BF5XX_SPORT_NUM]), + .resource = bfin_snd_resources[CONFIG_SND_BF5XX_SPORT_NUM], + .dev = { + .platform_data = &bfin_snd_data[CONFIG_SND_BF5XX_SPORT_NUM], + }, }; #endif @@ -580,7 +704,7 @@ static struct platform_device *stamp_devices[] __initdata = { &net2272_bfin_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif @@ -596,7 +720,8 @@ static struct platform_device *stamp_devices[] __initdata = { #endif #endif -#if defined(CONFIG_SERIAL_BFIN_SPORT) || defined(CONFIG_SERIAL_BFIN_SPORT_MODULE) +#if defined(CONFIG_SERIAL_BFIN_SPORT) || \ + defined(CONFIG_SERIAL_BFIN_SPORT_MODULE) #ifdef CONFIG_SERIAL_BFIN_SPORT0_UART &bfin_sport0_uart_device, #endif @@ -618,14 +743,42 @@ static struct platform_device *stamp_devices[] __initdata = { #endif #if defined(CONFIG_SND_BF5XX_I2S) || defined(CONFIG_SND_BF5XX_I2S_MODULE) - &bfin_i2s, + &bfin_i2s_pcm, #endif #if defined(CONFIG_SND_BF5XX_TDM) || defined(CONFIG_SND_BF5XX_TDM_MODULE) - &bfin_tdm, + &bfin_tdm_pcm, #endif #if defined(CONFIG_SND_BF5XX_AC97) || defined(CONFIG_SND_BF5XX_AC97_MODULE) + &bfin_ac97_pcm, +#endif + +#if defined(CONFIG_SND_BF5XX_SOC_AD73311) || \ + defined(CONFIG_SND_BF5XX_SOC_AD73311_MODULE) + &bfin_ad73311_machine, +#endif + +#if defined(CONFIG_SND_SOC_AD73311) || defined(CONFIG_SND_SOC_AD73311_MODULE) + &bfin_ad73311_codec_device, +#endif + +#if defined(CONFIG_SND_SOC_AD74111) || defined(CONFIG_SND_SOC_AD74111_MODULE) + &bfin_ad74111_codec_device, +#endif + +#if defined(CONFIG_SND_BF5XX_SOC_I2S) || \ + defined(CONFIG_SND_BF5XX_SOC_I2S_MODULE) + &bfin_i2s, +#endif + +#if defined(CONFIG_SND_BF5XX_SOC_TDM) || \ + defined(CONFIG_SND_BF5XX_SOC_TDM_MODULE) + &bfin_tdm, +#endif + +#if defined(CONFIG_SND_BF5XX_SOC_AC97) || \ + defined(CONFIG_SND_BF5XX_SOC_AC97_MODULE) &bfin_ac97, #endif }; diff --git a/arch/blackfin/mach-bf537/boards/cm_bf537e.c b/arch/blackfin/mach-bf537/boards/cm_bf537e.c index 604a430..0d4a2f6 100644 --- a/arch/blackfin/mach-bf537/boards/cm_bf537e.c +++ b/arch/blackfin/mach-bf537/boards/cm_bf537e.c @@ -31,7 +31,7 @@ */ const char bfin_board_name[] = "Bluetechnix CM BF537E"; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* all SPI peripherals info goes here */ #if defined(CONFIG_MTD_M25P80) || defined(CONFIG_MTD_M25P80_MODULE) @@ -735,7 +735,7 @@ static struct platform_device *cm_bf537e_devices[] __initdata = { &net2272_bfin_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif @@ -770,7 +770,7 @@ static int __init cm_bf537e_init(void) { printk(KERN_INFO "%s(): registering device resources\n", __func__); platform_add_devices(cm_bf537e_devices, ARRAY_SIZE(cm_bf537e_devices)); -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) spi_register_board_info(bfin_spi_board_info, ARRAY_SIZE(bfin_spi_board_info)); #endif diff --git a/arch/blackfin/mach-bf537/boards/cm_bf537u.c b/arch/blackfin/mach-bf537/boards/cm_bf537u.c index d916b46..f553698 100644 --- a/arch/blackfin/mach-bf537/boards/cm_bf537u.c +++ b/arch/blackfin/mach-bf537/boards/cm_bf537u.c @@ -32,7 +32,7 @@ */ const char bfin_board_name[] = "Bluetechnix CM BF537U"; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* all SPI peripherals info goes here */ #if defined(CONFIG_MTD_M25P80) || defined(CONFIG_MTD_M25P80_MODULE) @@ -700,7 +700,7 @@ static struct platform_device *cm_bf537u_devices[] __initdata = { &net2272_bfin_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif @@ -747,7 +747,7 @@ static int __init cm_bf537u_init(void) { printk(KERN_INFO "%s(): registering device resources\n", __func__); platform_add_devices(cm_bf537u_devices, ARRAY_SIZE(cm_bf537u_devices)); -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) spi_register_board_info(bfin_spi_board_info, ARRAY_SIZE(bfin_spi_board_info)); #endif diff --git a/arch/blackfin/mach-bf537/boards/dnp5370.c b/arch/blackfin/mach-bf537/boards/dnp5370.c index 5f30722..11dadeb 100644 --- a/arch/blackfin/mach-bf537/boards/dnp5370.c +++ b/arch/blackfin/mach-bf537/boards/dnp5370.c @@ -125,7 +125,7 @@ static struct platform_device asmb_flash_device = { }; #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) #if defined(CONFIG_MMC_SPI) || defined(CONFIG_MMC_SPI_MODULE) @@ -370,7 +370,7 @@ static struct platform_device *dnp5370_devices[] __initdata = { &bfin_mac_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &spi_bfin_master_device, #endif diff --git a/arch/blackfin/mach-bf537/boards/minotaur.c b/arch/blackfin/mach-bf537/boards/minotaur.c index 3901dd0..d2d7128 100644 --- a/arch/blackfin/mach-bf537/boards/minotaur.c +++ b/arch/blackfin/mach-bf537/boards/minotaur.c @@ -121,7 +121,7 @@ static struct platform_device net2272_bfin_device = { }; #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* all SPI peripherals info goes here */ #if defined(CONFIG_MTD_M25P80) \ @@ -496,7 +496,7 @@ static struct platform_device *minotaur_devices[] __initdata = { &net2272_bfin_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif @@ -537,7 +537,7 @@ static int __init minotaur_init(void) { printk(KERN_INFO "%s(): registering device resources\n", __func__); platform_add_devices(minotaur_devices, ARRAY_SIZE(minotaur_devices)); -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) spi_register_board_info(bfin_spi_board_info, ARRAY_SIZE(bfin_spi_board_info)); #endif diff --git a/arch/blackfin/mach-bf537/boards/pnav10.c b/arch/blackfin/mach-bf537/boards/pnav10.c index aebd31c..6fd8470 100644 --- a/arch/blackfin/mach-bf537/boards/pnav10.c +++ b/arch/blackfin/mach-bf537/boards/pnav10.c @@ -154,7 +154,7 @@ static struct platform_device net2272_bfin_device = { }; #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* all SPI peripherals info goes here */ #if defined(CONFIG_MTD_M25P80) \ @@ -477,7 +477,7 @@ static struct platform_device *stamp_devices[] __initdata = { &net2272_bfin_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif @@ -508,7 +508,7 @@ static int __init pnav_init(void) { printk(KERN_INFO "%s(): registering device resources\n", __func__); platform_add_devices(stamp_devices, ARRAY_SIZE(stamp_devices)); -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) spi_register_board_info(bfin_spi_board_info, ARRAY_SIZE(bfin_spi_board_info)); #endif diff --git a/arch/blackfin/mach-bf537/boards/stamp.c b/arch/blackfin/mach-bf537/boards/stamp.c index 7fbb0bb..2221173 100644 --- a/arch/blackfin/mach-bf537/boards/stamp.c +++ b/arch/blackfin/mach-bf537/boards/stamp.c @@ -1420,7 +1420,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { #endif }; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI controller data */ static struct bfin5xx_spi_master bfin_spi0_info = { .num_chipselect = MAX_CTRL_CS + MAX_BLACKFIN_GPIOS, @@ -1462,7 +1462,7 @@ static struct platform_device bfin_spi0_device = { /* SPORT SPI controller data */ static struct bfin5xx_spi_master bfin_sport_spi0_info = { - .num_chipselect = 1, /* master only supports one device */ + .num_chipselect = MAX_BLACKFIN_GPIOS, .enable_dma = 0, /* master don't support DMA */ .pin_req = {P_SPORT0_DTPRI, P_SPORT0_TSCLK, P_SPORT0_DRPRI, P_SPORT0_RSCLK, P_SPORT0_TFS, P_SPORT0_RFS, 0}, @@ -1492,7 +1492,7 @@ static struct platform_device bfin_sport_spi0_device = { }; static struct bfin5xx_spi_master bfin_sport_spi1_info = { - .num_chipselect = 1, /* master only supports one device */ + .num_chipselect = MAX_BLACKFIN_GPIOS, .enable_dma = 0, /* master don't support DMA */ .pin_req = {P_SPORT1_DTPRI, P_SPORT1_TSCLK, P_SPORT1_DRPRI, P_SPORT1_RSCLK, P_SPORT1_TFS, P_SPORT1_RFS, 0}, @@ -1558,6 +1558,71 @@ static struct platform_device bfin_lq035q1_device = { }; #endif +#if defined(CONFIG_VIDEO_BLACKFIN_CAPTURE) \ + || defined(CONFIG_VIDEO_BLACKFIN_CAPTURE_MODULE) +#include <linux/videodev2.h> +#include <media/blackfin/bfin_capture.h> +#include <media/blackfin/ppi.h> + +static const unsigned short ppi_req[] = { + P_PPI0_D0, P_PPI0_D1, P_PPI0_D2, P_PPI0_D3, + P_PPI0_D4, P_PPI0_D5, P_PPI0_D6, P_PPI0_D7, + P_PPI0_CLK, P_PPI0_FS1, P_PPI0_FS2, + 0, +}; + +static const struct ppi_info ppi_info = { + .type = PPI_TYPE_PPI, + .dma_ch = CH_PPI, + .irq_err = IRQ_PPI_ERROR, + .base = (void __iomem *)PPI_CONTROL, + .pin_req = ppi_req, +}; + +#if defined(CONFIG_VIDEO_VS6624) \ + || defined(CONFIG_VIDEO_VS6624_MODULE) +static struct v4l2_input vs6624_inputs[] = { + { + .index = 0, + .name = "Camera", + .type = V4L2_INPUT_TYPE_CAMERA, + .std = V4L2_STD_UNKNOWN, + }, +}; + +static struct bcap_route vs6624_routes[] = { + { + .input = 0, + .output = 0, + }, +}; + +static const unsigned vs6624_ce_pin = GPIO_PF10; + +static struct bfin_capture_config bfin_capture_data = { + .card_name = "BF537", + .inputs = vs6624_inputs, + .num_inputs = ARRAY_SIZE(vs6624_inputs), + .routes = vs6624_routes, + .i2c_adapter_id = 0, + .board_info = { + .type = "vs6624", + .addr = 0x10, + .platform_data = (void *)&vs6624_ce_pin, + }, + .ppi_info = &ppi_info, + .ppi_control = (PACK_EN | DLEN_8 | XFR_TYPE | 0x0020), +}; +#endif + +static struct platform_device bfin_capture_device = { + .name = "bfin_capture", + .dev = { + .platform_data = &bfin_capture_data, + }, +}; +#endif + #if defined(CONFIG_SERIAL_BFIN) || defined(CONFIG_SERIAL_BFIN_MODULE) #ifdef CONFIG_SERIAL_BFIN_UART0 static struct resource bfin_uart0_resources[] = { @@ -2716,7 +2781,7 @@ static struct platform_device *stamp_devices[] __initdata = { &net2272_bfin_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif @@ -2733,6 +2798,11 @@ static struct platform_device *stamp_devices[] __initdata = { &bfin_lq035q1_device, #endif +#if defined(CONFIG_VIDEO_BLACKFIN_CAPTURE) \ + || defined(CONFIG_VIDEO_BLACKFIN_CAPTURE_MODULE) + &bfin_capture_device, +#endif + #if defined(CONFIG_SERIAL_BFIN) || defined(CONFIG_SERIAL_BFIN_MODULE) #ifdef CONFIG_SERIAL_BFIN_UART0 &bfin_uart0_device, diff --git a/arch/blackfin/mach-bf537/boards/tcm_bf537.c b/arch/blackfin/mach-bf537/boards/tcm_bf537.c index 6917ce2..9885176 100644 --- a/arch/blackfin/mach-bf537/boards/tcm_bf537.c +++ b/arch/blackfin/mach-bf537/boards/tcm_bf537.c @@ -32,7 +32,7 @@ */ const char bfin_board_name[] = "Bluetechnix TCM BF537"; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* all SPI peripherals info goes here */ #if defined(CONFIG_MTD_M25P80) || defined(CONFIG_MTD_M25P80_MODULE) @@ -702,7 +702,7 @@ static struct platform_device *cm_bf537_devices[] __initdata = { &net2272_bfin_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif @@ -737,7 +737,7 @@ static int __init tcm_bf537_init(void) { printk(KERN_INFO "%s(): registering device resources\n", __func__); platform_add_devices(cm_bf537_devices, ARRAY_SIZE(cm_bf537_devices)); -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) spi_register_board_info(bfin_spi_board_info, ARRAY_SIZE(bfin_spi_board_info)); #endif diff --git a/arch/blackfin/mach-bf538/boards/ezkit.c b/arch/blackfin/mach-bf538/boards/ezkit.c index 8356eb5..1633a6f 100644 --- a/arch/blackfin/mach-bf538/boards/ezkit.c +++ b/arch/blackfin/mach-bf538/boards/ezkit.c @@ -490,7 +490,7 @@ static struct platform_device smc91x_device = { }; #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* all SPI peripherals info goes here */ #if defined(CONFIG_MTD_M25P80) \ || defined(CONFIG_MTD_M25P80_MODULE) @@ -874,7 +874,7 @@ static struct platform_device *cm_bf538_devices[] __initdata = { #endif #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bf538_spi_master0, &bf538_spi_master1, &bf538_spi_master2, @@ -938,7 +938,7 @@ static int __init ezkit_init(void) printk(KERN_INFO "%s(): registering device resources\n", __func__); platform_add_devices(cm_bf538_devices, ARRAY_SIZE(cm_bf538_devices)); -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) spi_register_board_info(bf538_spi_board_info, ARRAY_SIZE(bf538_spi_board_info)); #endif diff --git a/arch/blackfin/mach-bf548/boards/cm_bf548.c b/arch/blackfin/mach-bf548/boards/cm_bf548.c index 0350eac..68af594 100644 --- a/arch/blackfin/mach-bf548/boards/cm_bf548.c +++ b/arch/blackfin/mach-bf548/boards/cm_bf548.c @@ -854,7 +854,7 @@ static struct platform_device para_flash_device = { }; #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* all SPI peripherals info goes here */ #if defined(CONFIG_MTD_M25P80) \ || defined(CONFIG_MTD_M25P80_MODULE) @@ -1175,7 +1175,7 @@ static struct platform_device *cm_bf548_devices[] __initdata = { &bf54x_sdh_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bf54x_spi_master0, &bf54x_spi_master1, #endif @@ -1210,7 +1210,7 @@ static int __init cm_bf548_init(void) printk(KERN_INFO "%s(): registering device resources\n", __func__); platform_add_devices(cm_bf548_devices, ARRAY_SIZE(cm_bf548_devices)); -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) spi_register_board_info(bf54x_spi_board_info, ARRAY_SIZE(bf54x_spi_board_info)); #endif diff --git a/arch/blackfin/mach-bf548/boards/ezkit.c b/arch/blackfin/mach-bf548/boards/ezkit.c index bb868ac..3ea45f8 100644 --- a/arch/blackfin/mach-bf548/boards/ezkit.c +++ b/arch/blackfin/mach-bf548/boards/ezkit.c @@ -1110,7 +1110,7 @@ static struct spi_board_info bfin_spi_board_info[] __initdata = { }, #endif }; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI (0) */ static struct resource bfin_spi0_resource[] = { [0] = { @@ -1183,6 +1183,71 @@ static struct platform_device bf54x_spi_master1 = { }; #endif /* spi master and devices */ +#if defined(CONFIG_VIDEO_BLACKFIN_CAPTURE) \ + || defined(CONFIG_VIDEO_BLACKFIN_CAPTURE_MODULE) +#include <linux/videodev2.h> +#include <media/blackfin/bfin_capture.h> +#include <media/blackfin/ppi.h> + +static const unsigned short ppi_req[] = { + P_PPI1_D0, P_PPI1_D1, P_PPI1_D2, P_PPI1_D3, + P_PPI1_D4, P_PPI1_D5, P_PPI1_D6, P_PPI1_D7, + P_PPI1_CLK, P_PPI1_FS1, P_PPI1_FS2, + 0, +}; + +static const struct ppi_info ppi_info = { + .type = PPI_TYPE_EPPI, + .dma_ch = CH_EPPI1, + .irq_err = IRQ_EPPI1_ERROR, + .base = (void __iomem *)EPPI1_STATUS, + .pin_req = ppi_req, +}; + +#if defined(CONFIG_VIDEO_VS6624) \ + || defined(CONFIG_VIDEO_VS6624_MODULE) +static struct v4l2_input vs6624_inputs[] = { + { + .index = 0, + .name = "Camera", + .type = V4L2_INPUT_TYPE_CAMERA, + .std = V4L2_STD_UNKNOWN, + }, +}; + +static struct bcap_route vs6624_routes[] = { + { + .input = 0, + .output = 0, + }, +}; + +static const unsigned vs6624_ce_pin = GPIO_PG6; + +static struct bfin_capture_config bfin_capture_data = { + .card_name = "BF548", + .inputs = vs6624_inputs, + .num_inputs = ARRAY_SIZE(vs6624_inputs), + .routes = vs6624_routes, + .i2c_adapter_id = 0, + .board_info = { + .type = "vs6624", + .addr = 0x10, + .platform_data = (void *)&vs6624_ce_pin, + }, + .ppi_info = &ppi_info, + .ppi_control = (POLC | PACKEN | DLEN_8 | XFR_TYPE | 0x20), +}; +#endif + +static struct platform_device bfin_capture_device = { + .name = "bfin_capture", + .dev = { + .platform_data = &bfin_capture_data, + }, +}; +#endif + #if defined(CONFIG_I2C_BLACKFIN_TWI) || defined(CONFIG_I2C_BLACKFIN_TWI_MODULE) static struct resource bfin_twi0_resource[] = { [0] = { @@ -1502,10 +1567,14 @@ static struct platform_device *ezkit_devices[] __initdata = { &bf54x_sdh_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bf54x_spi_master0, &bf54x_spi_master1, #endif +#if defined(CONFIG_VIDEO_BLACKFIN_CAPTURE) \ + || defined(CONFIG_VIDEO_BLACKFIN_CAPTURE_MODULE) + &bfin_capture_device, +#endif #if defined(CONFIG_KEYBOARD_BFIN) || defined(CONFIG_KEYBOARD_BFIN_MODULE) &bf54x_kpad_device, diff --git a/arch/blackfin/mach-bf561/boards/acvilon.c b/arch/blackfin/mach-bf561/boards/acvilon.c index b1b7339..f6ffd6f 100644 --- a/arch/blackfin/mach-bf561/boards/acvilon.c +++ b/arch/blackfin/mach-bf561/boards/acvilon.c @@ -372,7 +372,7 @@ static struct bfin5xx_spi_chip data_flash_chip_info = { }; #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI (0) */ static struct resource bfin_spi0_resource[] = { [0] = { @@ -475,7 +475,7 @@ static struct platform_device bfin_dpmc = { static struct platform_device *acvilon_devices[] __initdata = { &bfin_dpmc, -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif diff --git a/arch/blackfin/mach-bf561/boards/cm_bf561.c b/arch/blackfin/mach-bf561/boards/cm_bf561.c index c017cf0..d81450f 100644 --- a/arch/blackfin/mach-bf561/boards/cm_bf561.c +++ b/arch/blackfin/mach-bf561/boards/cm_bf561.c @@ -29,7 +29,7 @@ */ const char bfin_board_name[] = "Bluetechnix CM BF561"; -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* all SPI peripherals info goes here */ #if defined(CONFIG_MTD_M25P80) || defined(CONFIG_MTD_M25P80_MODULE) @@ -488,7 +488,7 @@ static struct platform_device *cm_bf561_devices[] __initdata = { &net2272_bfin_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif @@ -523,7 +523,7 @@ static int __init cm_bf561_init(void) { printk(KERN_INFO "%s(): registering device resources\n", __func__); platform_add_devices(cm_bf561_devices, ARRAY_SIZE(cm_bf561_devices)); -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) spi_register_board_info(bfin_spi_board_info, ARRAY_SIZE(bfin_spi_board_info)); #endif diff --git a/arch/blackfin/mach-bf561/boards/ezkit.c b/arch/blackfin/mach-bf561/boards/ezkit.c index 27f22ed..8389788 100644 --- a/arch/blackfin/mach-bf561/boards/ezkit.c +++ b/arch/blackfin/mach-bf561/boards/ezkit.c @@ -291,7 +291,7 @@ static struct platform_device ezkit_flash_device = { }; #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) /* SPI (0) */ static struct resource bfin_spi0_resource[] = { [0] = { @@ -383,7 +383,7 @@ static struct i2c_gpio_platform_data i2c_gpio_data = { .scl_pin = GPIO_PF0, .sda_is_open_drain = 0, .scl_is_open_drain = 0, - .udelay = 40, + .udelay = 10, }; static struct platform_device i2c_gpio_device = { @@ -422,6 +422,96 @@ static struct platform_device bfin_dpmc = { }, }; +#if defined(CONFIG_VIDEO_BLACKFIN_CAPTURE) \ + || defined(CONFIG_VIDEO_BLACKFIN_CAPTURE_MODULE) +#include <linux/videodev2.h> +#include <media/blackfin/bfin_capture.h> +#include <media/blackfin/ppi.h> + +static const unsigned short ppi_req[] = { + P_PPI0_D0, P_PPI0_D1, P_PPI0_D2, P_PPI0_D3, + P_PPI0_D4, P_PPI0_D5, P_PPI0_D6, P_PPI0_D7, + P_PPI0_CLK, P_PPI0_FS1, P_PPI0_FS2, + 0, +}; + +static const struct ppi_info ppi_info = { + .type = PPI_TYPE_PPI, + .dma_ch = CH_PPI0, + .irq_err = IRQ_PPI1_ERROR, + .base = (void __iomem *)PPI0_CONTROL, + .pin_req = ppi_req, +}; + +#if defined(CONFIG_VIDEO_ADV7183) \ + || defined(CONFIG_VIDEO_ADV7183_MODULE) +#include <media/adv7183.h> +static struct v4l2_input adv7183_inputs[] = { + { + .index = 0, + .name = "Composite", + .type = V4L2_INPUT_TYPE_CAMERA, + .std = V4L2_STD_ALL, + }, + { + .index = 1, + .name = "S-Video", + .type = V4L2_INPUT_TYPE_CAMERA, + .std = V4L2_STD_ALL, + }, + { + .index = 2, + .name = "Component", + .type = V4L2_INPUT_TYPE_CAMERA, + .std = V4L2_STD_ALL, + }, +}; + +static struct bcap_route adv7183_routes[] = { + { + .input = ADV7183_COMPOSITE4, + .output = ADV7183_8BIT_OUT, + }, + { + .input = ADV7183_SVIDEO0, + .output = ADV7183_8BIT_OUT, + }, + { + .input = ADV7183_COMPONENT0, + .output = ADV7183_8BIT_OUT, + }, +}; + + +static const unsigned adv7183_gpio[] = { + GPIO_PF13, /* reset pin */ + GPIO_PF2, /* output enable pin */ +}; + +static struct bfin_capture_config bfin_capture_data = { + .card_name = "BF561", + .inputs = adv7183_inputs, + .num_inputs = ARRAY_SIZE(adv7183_inputs), + .routes = adv7183_routes, + .i2c_adapter_id = 0, + .board_info = { + .type = "adv7183", + .addr = 0x20, + .platform_data = (void *)adv7183_gpio, + }, + .ppi_info = &ppi_info, + .ppi_control = (PACK_EN | DLEN_8 | DMA32 | FLD_SEL), +}; +#endif + +static struct platform_device bfin_capture_device = { + .name = "bfin_capture", + .dev = { + .platform_data = &bfin_capture_data, + }, +}; +#endif + #if defined(CONFIG_SND_BF5XX_I2S) || defined(CONFIG_SND_BF5XX_I2S_MODULE) static struct platform_device bfin_i2s = { .name = "bfin-i2s", @@ -462,7 +552,7 @@ static struct platform_device *ezkit_devices[] __initdata = { &bfin_isp1760_device, #endif -#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE) +#if defined(CONFIG_SPI_BFIN5XX) || defined(CONFIG_SPI_BFIN5XX_MODULE) &bfin_spi0_device, #endif @@ -494,6 +584,11 @@ static struct platform_device *ezkit_devices[] __initdata = { &ezkit_flash_device, #endif +#if defined(CONFIG_VIDEO_BLACKFIN_CAPTURE) \ + || defined(CONFIG_VIDEO_BLACKFIN_CAPTURE_MODULE) + &bfin_capture_device, +#endif + #if defined(CONFIG_SND_BF5XX_I2S) || defined(CONFIG_SND_BF5XX_I2S_MODULE) &bfin_i2s, #endif diff --git a/arch/blackfin/mach-bf561/include/mach/pll.h b/arch/blackfin/mach-bf561/include/mach/pll.h index 7977db2..00bdace 100644 --- a/arch/blackfin/mach-bf561/include/mach/pll.h +++ b/arch/blackfin/mach-bf561/include/mach/pll.h @@ -16,6 +16,7 @@ #include <mach/irq.h> #define SUPPLE_0_WAKEUP ((IRQ_SUPPLE_0 - (IRQ_CORETMR + 1)) % 32) +#define SUPPLE_1_WAKEUP ((IRQ_SUPPLE_1 - (IRQ_CORETMR + 1)) % 32) static inline void bfin_iwr_restore(unsigned long iwr0, unsigned long iwr1, unsigned long iwr2) @@ -42,7 +43,8 @@ bfin_iwr_save(unsigned long niwr0, unsigned long niwr1, unsigned long niwr2, static inline void bfin_iwr_set_sup0(unsigned long *iwr0, unsigned long *iwr1, unsigned long *iwr2) { - bfin_iwr_save(0, IWR_ENABLE(SUPPLE_0_WAKEUP), 0, iwr0, iwr1, iwr2); + bfin_iwr_save(0, IWR_ENABLE(SUPPLE_0_WAKEUP) | + IWR_ENABLE(SUPPLE_1_WAKEUP), 0, iwr0, iwr1, iwr2); } #endif diff --git a/arch/blackfin/mach-bf561/smp.c b/arch/blackfin/mach-bf561/smp.c index db22401..ab1c617 100644 --- a/arch/blackfin/mach-bf561/smp.c +++ b/arch/blackfin/mach-bf561/smp.c @@ -84,7 +84,7 @@ int __cpuinit platform_boot_secondary(unsigned int cpu, struct task_struct *idle if ((bfin_read_SYSCR() & COREB_SRAM_INIT) == 0) { /* CoreB already running, sending ipi to wakeup it */ - platform_send_ipi_cpu(cpu, IRQ_SUPPLE_0); + smp_send_reschedule(cpu); } else { /* Kick CoreB, which should start execution from CORE_SRAM_BASE. */ bfin_write_SYSCR(bfin_read_SYSCR() & ~COREB_SRAM_INIT); @@ -114,7 +114,8 @@ void __init platform_request_ipi(int irq, void *handler) int ret; const char *name = (irq == IRQ_SUPPLE_0) ? supple0 : supple1; - ret = request_irq(irq, handler, IRQF_PERCPU, name, handler); + ret = request_irq(irq, handler, IRQF_PERCPU | IRQF_NO_SUSPEND | + IRQF_FORCE_RESUME, name, handler); if (ret) panic("Cannot request %s for IPI service", name); } diff --git a/arch/blackfin/mach-common/smp.c b/arch/blackfin/mach-common/smp.c index 0784a52..ac8f8a4 100644 --- a/arch/blackfin/mach-common/smp.c +++ b/arch/blackfin/mach-common/smp.c @@ -14,6 +14,7 @@ #include <linux/sched.h> #include <linux/interrupt.h> #include <linux/cache.h> +#include <linux/clockchips.h> #include <linux/profile.h> #include <linux/errno.h> #include <linux/mm.h> @@ -47,9 +48,10 @@ unsigned long blackfin_iflush_l1_entry[NR_CPUS]; struct blackfin_initial_pda __cpuinitdata initial_pda_coreb; -#define BFIN_IPI_RESCHEDULE 0 -#define BFIN_IPI_CALL_FUNC 1 -#define BFIN_IPI_CPU_STOP 2 +#define BFIN_IPI_TIMER 0 +#define BFIN_IPI_RESCHEDULE 1 +#define BFIN_IPI_CALL_FUNC 2 +#define BFIN_IPI_CPU_STOP 3 struct blackfin_flush_data { unsigned long start; @@ -160,6 +162,14 @@ static irqreturn_t ipi_handler_int0(int irq, void *dev_instance) return IRQ_HANDLED; } +DECLARE_PER_CPU(struct clock_event_device, coretmr_events); +void ipi_timer(void) +{ + int cpu = smp_processor_id(); + struct clock_event_device *evt = &per_cpu(coretmr_events, cpu); + evt->event_handler(evt); +} + static irqreturn_t ipi_handler_int1(int irq, void *dev_instance) { struct ipi_message *msg; @@ -176,18 +186,17 @@ static irqreturn_t ipi_handler_int1(int irq, void *dev_instance) while (msg_queue->count) { msg = &msg_queue->ipi_message[msg_queue->head]; switch (msg->type) { + case BFIN_IPI_TIMER: + ipi_timer(); + break; case BFIN_IPI_RESCHEDULE: scheduler_ipi(); break; case BFIN_IPI_CALL_FUNC: - spin_unlock_irqrestore(&msg_queue->lock, flags); ipi_call_function(cpu, msg); - spin_lock_irqsave(&msg_queue->lock, flags); break; case BFIN_IPI_CPU_STOP: - spin_unlock_irqrestore(&msg_queue->lock, flags); ipi_cpu_stop(cpu); - spin_lock_irqsave(&msg_queue->lock, flags); break; default: printk(KERN_CRIT "CPU%u: Unknown IPI message 0x%lx\n", @@ -297,8 +306,6 @@ void smp_send_reschedule(int cpu) { cpumask_t callmap; /* simply trigger an ipi */ - if (cpu_is_offline(cpu)) - return; cpumask_clear(&callmap); cpumask_set_cpu(cpu, &callmap); @@ -308,6 +315,16 @@ void smp_send_reschedule(int cpu) return; } +void smp_send_msg(const struct cpumask *mask, unsigned long type) +{ + smp_send_message(*mask, type, NULL, NULL, 0); +} + +void smp_timer_broadcast(const struct cpumask *mask) +{ + smp_send_msg(mask, BFIN_IPI_TIMER); +} + void smp_send_stop(void) { cpumask_t callmap; @@ -326,17 +343,24 @@ void smp_send_stop(void) int __cpuinit __cpu_up(unsigned int cpu) { int ret; - static struct task_struct *idle; + struct blackfin_cpudata *ci = &per_cpu(cpu_data, cpu); + struct task_struct *idle = ci->idle; - if (idle) + if (idle) { free_task(idle); - - idle = fork_idle(cpu); - if (IS_ERR(idle)) { - printk(KERN_ERR "CPU%u: fork() failed\n", cpu); - return PTR_ERR(idle); + idle = NULL; } + if (!idle) { + idle = fork_idle(cpu); + if (IS_ERR(idle)) { + printk(KERN_ERR "CPU%u: fork() failed\n", cpu); + return PTR_ERR(idle); + } + ci->idle = idle; + } else { + init_idle(idle, cpu); + } secondary_stack = task_stack_page(idle) + THREAD_SIZE; ret = platform_boot_secondary(cpu, idle); @@ -411,6 +435,7 @@ void __cpuinit secondary_start_kernel(void) bfin_setup_caches(cpu); + notify_cpu_starting(cpu); /* * Calibrate loops per jiffy value. * IRQs need to be enabled here - D-cache can be invalidated @@ -453,8 +478,10 @@ void smp_icache_flush_range_others(unsigned long start, unsigned long end) smp_flush_data.start = start; smp_flush_data.end = end; - if (smp_call_function(&ipi_flush_icache, &smp_flush_data, 0)) + preempt_disable(); + if (smp_call_function(&ipi_flush_icache, &smp_flush_data, 1)) printk(KERN_WARNING "SMP: failed to run I-cache flush request on other CPUs\n"); + preempt_enable(); } EXPORT_SYMBOL_GPL(smp_icache_flush_range_others); diff --git a/arch/c6x/Kconfig b/arch/c6x/Kconfig new file mode 100644 index 0000000..26e67f0 --- /dev/null +++ b/arch/c6x/Kconfig @@ -0,0 +1,174 @@ +# +# For a description of the syntax of this configuration file, +# see Documentation/kbuild/kconfig-language.txt. +# + +config TMS320C6X + def_bool y + select CLKDEV_LOOKUP + select GENERIC_IRQ_SHOW + select HAVE_ARCH_TRACEHOOK + select HAVE_DMA_API_DEBUG + select HAVE_GENERIC_HARDIRQS + select HAVE_MEMBLOCK + select HAVE_SPARSE_IRQ + select OF + select OF_EARLY_FLATTREE + +config MMU + def_bool n + +config ZONE_DMA + def_bool y + +config FPU + def_bool n + +config HIGHMEM + def_bool n + +config NUMA + def_bool n + +config RWSEM_GENERIC_SPINLOCK + def_bool y + +config RWSEM_XCHGADD_ALGORITHM + def_bool n + +config GENERIC_CALIBRATE_DELAY + def_bool y + +config GENERIC_HWEIGHT + def_bool y + +config GENERIC_CLOCKEVENTS + def_bool y + +config GENERIC_CLOCKEVENTS_BROADCAST + bool + +config GENERIC_BUG + def_bool y + +config COMMON_CLKDEV + def_bool y + +config C6X_BIG_KERNEL + bool "Build a big kernel" + help + The C6X function call instruction has a limited range of +/- 2MiB. + This is sufficient for most kernels, but some kernel configurations + with lots of compiled-in functionality may require a larger range + for function calls. Use this option to have the compiler generate + function calls with 32-bit range. This will make the kernel both + larger and slower. + + If unsure, say N. + +source "init/Kconfig" + +# Use the generic interrupt handling code in kernel/irq/ + +source "kernel/Kconfig.freezer" + +config CMDLINE_BOOL + bool "Default bootloader kernel arguments" + +config CMDLINE + string "Kernel command line" + depends on CMDLINE_BOOL + default "console=ttyS0,57600" + help + On some architectures there is currently no way for the boot loader + to pass arguments to the kernel. For these architectures, you should + supply some command-line options at build time by entering them + here. + +config CMDLINE_FORCE + bool "Force default kernel command string" + depends on CMDLINE_BOOL + default n + help + Set this to have arguments from the default kernel command string + override those passed by the boot loader. + +config CPU_BIG_ENDIAN + bool "Build big-endian kernel" + default n + help + Say Y if you plan on running a kernel in big-endian mode. + Note that your board must be properly built and your board + port must properly enable any big-endian related features + of your chipset/board/processor. + +config FORCE_MAX_ZONEORDER + int "Maximum zone order" + default "13" + help + The kernel memory allocator divides physically contiguous memory + blocks into "zones", where each zone is a power of two number of + pages. This option selects the largest power of two that the kernel + keeps in the memory allocator. If you need to allocate very large + blocks of physically contiguous memory, then you may need to + increase this value. + + This config option is actually maximum order plus one. For example, + a value of 11 means that the largest free memory block is 2^10 pages. + +menu "Processor type and features" + +source "arch/c6x/platforms/Kconfig" + +config TMS320C6X_CACHES_ON + bool "L2 cache support" + default y + +config KERNEL_RAM_BASE_ADDRESS + hex "Virtual address of memory base" + default 0xe0000000 if SOC_TMS320C6455 + default 0xe0000000 if SOC_TMS320C6457 + default 0xe0000000 if SOC_TMS320C6472 + default 0x80000000 + +source "mm/Kconfig" + +source "kernel/Kconfig.preempt" + +source "kernel/Kconfig.hz" +source "kernel/time/Kconfig" + +endmenu + +menu "Executable file formats" + +source "fs/Kconfig.binfmt" + +endmenu + +source "net/Kconfig" + +source "drivers/Kconfig" + +source "fs/Kconfig" + +source "security/Kconfig" + +source "crypto/Kconfig" + +source "lib/Kconfig" + +menu "Kernel hacking" + +source "lib/Kconfig.debug" + +config ACCESS_CHECK + bool "Check the user pointer address" + default y + help + Usually the pointer transfer from user space is checked to see if its + address is in the kernel space. + + Say N here to disable that check to improve the performance. + +endmenu diff --git a/arch/c6x/Makefile b/arch/c6x/Makefile new file mode 100644 index 0000000..1d08dd0 --- /dev/null +++ b/arch/c6x/Makefile @@ -0,0 +1,60 @@ +# +# linux/arch/c6x/Makefile +# +# This file is subject to the terms and conditions of the GNU General Public +# License. See the file "COPYING" in the main directory of this archive +# for more details. +# + +cflags-y += -mno-dsbt -msdata=none + +cflags-$(CONFIG_C6X_BIG_KERNEL) += -mlong-calls + +CFLAGS_MODULE += -mlong-calls -mno-dsbt -msdata=none + +CHECKFLAGS += + +KBUILD_CFLAGS += $(cflags-y) +KBUILD_AFLAGS += $(cflags-y) + +ifdef CONFIG_CPU_BIG_ENDIAN +KBUILD_CFLAGS += -mbig-endian +KBUILD_AFLAGS += -mbig-endian +LINKFLAGS += -mbig-endian +KBUILD_LDFLAGS += -mbig-endian +LDFLAGS += -EB +endif + +head-y := arch/c6x/kernel/head.o +core-y += arch/c6x/kernel/ arch/c6x/mm/ arch/c6x/platforms/ +libs-y += arch/c6x/lib/ + +# Default to vmlinux.bin, override when needed +all: vmlinux.bin + +boot := arch/$(ARCH)/boot + +# Are we making a dtbImage.<boardname> target? If so, crack out the boardname +DTB:=$(subst dtbImage.,,$(filter dtbImage.%, $(MAKECMDGOALS))) +export DTB + +ifneq ($(DTB),) +core-y += $(boot)/ +endif + +# With make 3.82 we cannot mix normal and wildcard targets + +vmlinux.bin: vmlinux + $(Q)$(MAKE) $(build)=$(boot) $(patsubst %,$(boot)/%,$@) + +dtbImage.%: vmlinux + $(Q)$(MAKE) $(build)=$(boot) $(patsubst %,$(boot)/%,$@) + +archclean: + $(Q)$(MAKE) $(clean)=$(boot) + +define archhelp + @echo ' vmlinux.bin - Binary kernel image (arch/$(ARCH)/boot/vmlinux.bin)' + @echo ' dtbImage.<dt> - ELF image with $(arch)/boot/dts/<dt>.dts linked in' + @echo ' - stripped elf with fdt blob' +endef diff --git a/arch/c6x/boot/Makefile b/arch/c6x/boot/Makefile new file mode 100644 index 0000000..ecca820 --- /dev/null +++ b/arch/c6x/boot/Makefile @@ -0,0 +1,30 @@ +# +# Makefile for bootable kernel images +# + +OBJCOPYFLAGS_vmlinux.bin := -O binary +$(obj)/vmlinux.bin: vmlinux FORCE + $(call if_changed,objcopy) + +DTC_FLAGS ?= -p 1024 + +ifneq ($(DTB),) +obj-y += linked_dtb.o +endif + +$(obj)/%.dtb: $(src)/dts/%.dts FORCE + $(call cmd,dtc) + +quiet_cmd_cp = CP $< $@$2 + cmd_cp = cat $< >$@$2 || (rm -f $@ && echo false) + +# Generate builtin.dtb from $(DTB).dtb +$(obj)/builtin.dtb: $(obj)/$(DTB).dtb + $(call if_changed,cp) + +$(obj)/linked_dtb.o: $(obj)/builtin.dtb + +$(obj)/dtbImage.%: vmlinux + $(call if_changed,objcopy) + +clean-files := $(obj)/*.dtb diff --git a/arch/c6x/boot/dts/dsk6455.dts b/arch/c6x/boot/dts/dsk6455.dts new file mode 100644 index 0000000..2b71f80 --- /dev/null +++ b/arch/c6x/boot/dts/dsk6455.dts @@ -0,0 +1,62 @@ +/* + * arch/c6x/boot/dts/dsk6455.dts + * + * DSK6455 Evaluation Platform For TMS320C6455 + * Copyright (C) 2011 Texas Instruments Incorporated + * + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + */ + +/dts-v1/; + +/include/ "tms320c6455.dtsi" + +/ { + model = "Spectrum Digital DSK6455"; + compatible = "spectrum-digital,dsk6455"; + + chosen { + bootargs = "root=/dev/nfs ip=dhcp rw"; + }; + + memory { + device_type = "memory"; + reg = <0xE0000000 0x08000000>; + }; + + soc { + megamod_pic: interrupt-controller@1800000 { + interrupts = < 12 13 14 15 >; + }; + + emifa@70000000 { + flash@3,0 { + #address-cells = <1>; + #size-cells = <1>; + compatible = "cfi-flash"; + reg = <0x3 0x0 0x400000>; + bank-width = <1>; + device-width = <1>; + partition@0 { + reg = <0x0 0x400000>; + label = "NOR"; + }; + }; + }; + + timer1: timer@2980000 { + interrupt-parent = <&megamod_pic>; + interrupts = < 69 >; + }; + + clock-controller@029a0000 { + clock-frequency = <50000000>; + }; + }; +}; diff --git a/arch/c6x/boot/dts/evmc6457.dts b/arch/c6x/boot/dts/evmc6457.dts new file mode 100644 index 0000000..0301eb9a --- /dev/null +++ b/arch/c6x/boot/dts/evmc6457.dts @@ -0,0 +1,48 @@ +/* + * arch/c6x/boot/dts/evmc6457.dts + * + * EVMC6457 Evaluation Platform For TMS320C6457 + * + * Copyright (C) 2011 Texas Instruments Incorporated + * + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + */ + +/dts-v1/; + +/include/ "tms320c6457.dtsi" + +/ { + model = "eInfochips EVMC6457"; + compatible = "einfochips,evmc6457"; + + chosen { + bootargs = "console=hvc root=/dev/nfs ip=dhcp rw"; + }; + + memory { + device_type = "memory"; + reg = <0xE0000000 0x10000000>; + }; + + soc { + megamod_pic: interrupt-controller@1800000 { + interrupts = < 12 13 14 15 >; + }; + + timer0: timer@2940000 { + interrupt-parent = <&megamod_pic>; + interrupts = < 67 >; + }; + + clock-controller@29a0000 { + clock-frequency = <60000000>; + }; + }; +}; diff --git a/arch/c6x/boot/dts/evmc6472.dts b/arch/c6x/boot/dts/evmc6472.dts new file mode 100644 index 0000000..3e207b4 --- /dev/null +++ b/arch/c6x/boot/dts/evmc6472.dts @@ -0,0 +1,73 @@ +/* + * arch/c6x/boot/dts/evmc6472.dts + * + * EVMC6472 Evaluation Platform For TMS320C6472 + * + * Copyright (C) 2011 Texas Instruments Incorporated + * + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + */ + +/dts-v1/; + +/include/ "tms320c6472.dtsi" + +/ { + model = "eInfochips EVMC6472"; + compatible = "einfochips,evmc6472"; + + chosen { + bootargs = "console=hvc root=/dev/nfs ip=dhcp rw"; + }; + + memory { + device_type = "memory"; + reg = <0xE0000000 0x10000000>; + }; + + soc { + megamod_pic: interrupt-controller@1800000 { + interrupts = < 12 13 14 15 >; + }; + + timer0: timer@25e0000 { + interrupt-parent = <&megamod_pic>; + interrupts = < 16 >; + }; + + timer1: timer@25f0000 { + interrupt-parent = <&megamod_pic>; + interrupts = < 16 >; + }; + + timer2: timer@2600000 { + interrupt-parent = <&megamod_pic>; + interrupts = < 16 >; + }; + + timer3: timer@2610000 { + interrupt-parent = <&megamod_pic>; + interrupts = < 16 >; + }; + + timer4: timer@2620000 { + interrupt-parent = <&megamod_pic>; + interrupts = < 16 >; + }; + + timer5: timer@2630000 { + interrupt-parent = <&megamod_pic>; + interrupts = < 16 >; + }; + + clock-controller@29a0000 { + clock-frequency = <25000000>; + }; + }; +}; diff --git a/arch/c6x/boot/dts/evmc6474.dts b/arch/c6x/boot/dts/evmc6474.dts new file mode 100644 index 0000000..4dc2912 --- /dev/null +++ b/arch/c6x/boot/dts/evmc6474.dts @@ -0,0 +1,58 @@ +/* + * arch/c6x/boot/dts/evmc6474.dts + * + * EVMC6474 Evaluation Platform For TMS320C6474 + * + * Copyright (C) 2011 Texas Instruments Incorporated + * + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + */ + +/dts-v1/; + +/include/ "tms320c6474.dtsi" + +/ { + model = "Spectrum Digital EVMC6474"; + compatible = "spectrum-digital,evmc6474"; + + chosen { + bootargs = "console=hvc root=/dev/nfs ip=dhcp rw"; + }; + + memory { + device_type = "memory"; + reg = <0x80000000 0x08000000>; + }; + + soc { + megamod_pic: interrupt-controller@1800000 { + interrupts = < 12 13 14 15 >; + }; + + timer3: timer@2940000 { + interrupt-parent = <&megamod_pic>; + interrupts = < 39 >; + }; + + timer4: timer@2950000 { + interrupt-parent = <&megamod_pic>; + interrupts = < 41 >; + }; + + timer5: timer@2960000 { + interrupt-parent = <&megamod_pic>; + interrupts = < 43 >; + }; + + clock-controller@29a0000 { + clock-frequency = <50000000>; + }; + }; +}; diff --git a/arch/c6x/boot/dts/tms320c6455.dtsi b/arch/c6x/boot/dts/tms320c6455.dtsi new file mode 100644 index 0000000..a804ec1 --- /dev/null +++ b/arch/c6x/boot/dts/tms320c6455.dtsi @@ -0,0 +1,96 @@ + +/ { + #address-cells = <1>; + #size-cells = <1>; + + cpus { + #address-cells = <1>; + #size-cells = <0>; + + cpu@0 { + device_type = "cpu"; + model = "ti,c64x+"; + reg = <0>; + }; + }; + + soc { + compatible = "simple-bus"; + model = "tms320c6455"; + #address-cells = <1>; + #size-cells = <1>; + ranges; + + core_pic: interrupt-controller { + interrupt-controller; + #interrupt-cells = <1>; + compatible = "ti,c64x+core-pic"; + }; + + /* + * Megamodule interrupt controller + */ + megamod_pic: interrupt-controller@1800000 { + compatible = "ti,c64x+megamod-pic"; + interrupt-controller; + #interrupt-cells = <1>; + reg = <0x1800000 0x1000>; + interrupt-parent = <&core_pic>; + }; + + cache-controller@1840000 { + compatible = "ti,c64x+cache"; + reg = <0x01840000 0x8400>; + }; + + emifa@70000000 { + compatible = "ti,c64x+emifa", "simple-bus"; + #address-cells = <2>; + #size-cells = <1>; + reg = <0x70000000 0x100>; + ranges = <0x2 0x0 0xa0000000 0x00000008 + 0x3 0x0 0xb0000000 0x00400000 + 0x4 0x0 0xc0000000 0x10000000 + 0x5 0x0 0xD0000000 0x10000000>; + + ti,dscr-dev-enable = <13>; + ti,emifa-burst-priority = <255>; + ti,emifa-ce-config = <0x00240120 + 0x00240120 + 0x00240122 + 0x00240122>; + }; + + timer1: timer@2980000 { + compatible = "ti,c64x+timer64"; + reg = <0x2980000 0x40>; + ti,dscr-dev-enable = <4>; + }; + + clock-controller@029a0000 { + compatible = "ti,c6455-pll", "ti,c64x+pll"; + reg = <0x029a0000 0x200>; + ti,c64x+pll-bypass-delay = <1440>; + ti,c64x+pll-reset-delay = <15360>; + ti,c64x+pll-lock-delay = <24000>; + }; + + device-state-config-regs@2a80000 { + compatible = "ti,c64x+dscr"; + reg = <0x02a80000 0x41000>; + + ti,dscr-devstat = <0>; + ti,dscr-silicon-rev = <8 28 0xf>; + ti,dscr-rmii-resets = <0 0x40020 0x00040000>; + + ti,dscr-locked-regs = <0x40008 0x40004 0x0f0a0b00>; + ti,dscr-devstate-ctl-regs = + <0 12 0x40008 1 0 0 2 + 12 1 0x40008 3 0 30 2 + 13 2 0x4002c 1 0xffffffff 0 1>; + ti,dscr-devstate-stat-regs = + <0 10 0x40014 1 0 0 3 + 10 2 0x40018 1 0 0 3>; + }; + }; +}; diff --git a/arch/c6x/boot/dts/tms320c6457.dtsi b/arch/c6x/boot/dts/tms320c6457.dtsi new file mode 100644 index 0000000..35f4070 --- /dev/null +++ b/arch/c6x/boot/dts/tms320c6457.dtsi @@ -0,0 +1,68 @@ + +/ { + #address-cells = <1>; + #size-cells = <1>; + + cpus { + #address-cells = <1>; + #size-cells = <0>; + + cpu@0 { + device_type = "cpu"; + model = "ti,c64x+"; + reg = <0>; + }; + }; + + soc { + compatible = "simple-bus"; + model = "tms320c6457"; + #address-cells = <1>; + #size-cells = <1>; + ranges; + + core_pic: interrupt-controller { + interrupt-controller; + #interrupt-cells = <1>; + compatible = "ti,c64x+core-pic"; + }; + + megamod_pic: interrupt-controller@1800000 { + compatible = "ti,c64x+megamod-pic"; + interrupt-controller; + #interrupt-cells = <1>; + interrupt-parent = <&core_pic>; + reg = <0x1800000 0x1000>; + }; + + cache-controller@1840000 { + compatible = "ti,c64x+cache"; + reg = <0x01840000 0x8400>; + }; + + device-state-controller@2880800 { + compatible = "ti,c64x+dscr"; + reg = <0x02880800 0x400>; + + ti,dscr-devstat = <0x20>; + ti,dscr-silicon-rev = <0x18 28 0xf>; + ti,dscr-mac-fuse-regs = <0x114 3 4 5 6 + 0x118 0 0 1 2>; + ti,dscr-kick-regs = <0x38 0x83E70B13 + 0x3c 0x95A4F1E0>; + }; + + timer0: timer@2940000 { + compatible = "ti,c64x+timer64"; + reg = <0x2940000 0x40>; + }; + + clock-controller@29a0000 { + compatible = "ti,c6457-pll", "ti,c64x+pll"; + reg = <0x029a0000 0x200>; + ti,c64x+pll-bypass-delay = <300>; + ti,c64x+pll-reset-delay = <24000>; + ti,c64x+pll-lock-delay = <50000>; + }; + }; +}; diff --git a/arch/c6x/boot/dts/tms320c6472.dtsi b/arch/c6x/boot/dts/tms320c6472.dtsi new file mode 100644 index 0000000..b488aae --- /dev/null +++ b/arch/c6x/boot/dts/tms320c6472.dtsi @@ -0,0 +1,134 @@ + +/ { + #address-cells = <1>; + #size-cells = <1>; + + cpus { + #address-cells = <1>; + #size-cells = <0>; + + cpu@0 { + device_type = "cpu"; + reg = <0>; + model = "ti,c64x+"; + }; + cpu@1 { + device_type = "cpu"; + reg = <1>; + model = "ti,c64x+"; + }; + cpu@2 { + device_type = "cpu"; + reg = <2>; + model = "ti,c64x+"; + }; + cpu@3 { + device_type = "cpu"; + reg = <3>; + model = "ti,c64x+"; + }; + cpu@4 { + device_type = "cpu"; + reg = <4>; + model = "ti,c64x+"; + }; + cpu@5 { + device_type = "cpu"; + reg = <5>; + model = "ti,c64x+"; + }; + }; + + soc { + compatible = "simple-bus"; + model = "tms320c6472"; + #address-cells = <1>; + #size-cells = <1>; + ranges; + + core_pic: interrupt-controller { + compatible = "ti,c64x+core-pic"; + interrupt-controller; + #interrupt-cells = <1>; + }; + + megamod_pic: interrupt-controller@1800000 { + compatible = "ti,c64x+megamod-pic"; + interrupt-controller; + #interrupt-cells = <1>; + reg = <0x1800000 0x1000>; + interrupt-parent = <&core_pic>; + }; + + cache-controller@1840000 { + compatible = "ti,c64x+cache"; + reg = <0x01840000 0x8400>; + }; + + timer0: timer@25e0000 { + compatible = "ti,c64x+timer64"; + ti,core-mask = < 0x01 >; + reg = <0x25e0000 0x40>; + }; + + timer1: timer@25f0000 { + compatible = "ti,c64x+timer64"; + ti,core-mask = < 0x02 >; + reg = <0x25f0000 0x40>; + }; + + timer2: timer@2600000 { + compatible = "ti,c64x+timer64"; + ti,core-mask = < 0x04 >; + reg = <0x2600000 0x40>; + }; + + timer3: timer@2610000 { + compatible = "ti,c64x+timer64"; + ti,core-mask = < 0x08 >; + reg = <0x2610000 0x40>; + }; + + timer4: timer@2620000 { + compatible = "ti,c64x+timer64"; + ti,core-mask = < 0x10 >; + reg = <0x2620000 0x40>; + }; + + timer5: timer@2630000 { + compatible = "ti,c64x+timer64"; + ti,core-mask = < 0x20 >; + reg = <0x2630000 0x40>; + }; + + clock-controller@29a0000 { + compatible = "ti,c6472-pll", "ti,c64x+pll"; + reg = <0x029a0000 0x200>; + ti,c64x+pll-bypass-delay = <200>; + ti,c64x+pll-reset-delay = <12000>; + ti,c64x+pll-lock-delay = <80000>; + }; + + device-state-controller@2a80000 { + compatible = "ti,c64x+dscr"; + reg = <0x02a80000 0x1000>; + + ti,dscr-devstat = <0>; + ti,dscr-silicon-rev = <0x70c 16 0xff>; + + ti,dscr-mac-fuse-regs = <0x700 1 2 3 4 + 0x704 5 6 0 0>; + + ti,dscr-rmii-resets = <0x208 1 + 0x20c 1>; + + ti,dscr-locked-regs = <0x200 0x204 0x0a1e183a + 0x40c 0x420 0xbea7 + 0x41c 0x420 0xbea7>; + + ti,dscr-privperm = <0x41c 0xaaaaaaaa>; + + ti,dscr-devstate-ctl-regs = <0 13 0x200 1 0 0 1>; + }; + }; +}; diff --git a/arch/c6x/boot/dts/tms320c6474.dtsi b/arch/c6x/boot/dts/tms320c6474.dtsi new file mode 100644 index 0000000..cc601bf --- /dev/null +++ b/arch/c6x/boot/dts/tms320c6474.dtsi @@ -0,0 +1,89 @@ + +/ { + #address-cells = <1>; + #size-cells = <1>; + + cpus { + #address-cells = <1>; + #size-cells = <0>; + + cpu@0 { + device_type = "cpu"; + reg = <0>; + model = "ti,c64x+"; + }; + cpu@1 { + device_type = "cpu"; + reg = <1>; + model = "ti,c64x+"; + }; + cpu@2 { + device_type = "cpu"; + reg = <2>; + model = "ti,c64x+"; + }; + }; + + soc { + compatible = "simple-bus"; + model = "tms320c6474"; + #address-cells = <1>; + #size-cells = <1>; + ranges; + + core_pic: interrupt-controller { + interrupt-controller; + #interrupt-cells = <1>; + compatible = "ti,c64x+core-pic"; + }; + + megamod_pic: interrupt-controller@1800000 { + compatible = "ti,c64x+megamod-pic"; + interrupt-controller; + #interrupt-cells = <1>; + reg = <0x1800000 0x1000>; + interrupt-parent = <&core_pic>; + }; + + cache-controller@1840000 { + compatible = "ti,c64x+cache"; + reg = <0x01840000 0x8400>; + }; + + timer3: timer@2940000 { + compatible = "ti,c64x+timer64"; + ti,core-mask = < 0x04 >; + reg = <0x2940000 0x40>; + }; + + timer4: timer@2950000 { + compatible = "ti,c64x+timer64"; + ti,core-mask = < 0x02 >; + reg = <0x2950000 0x40>; + }; + + timer5: timer@2960000 { + compatible = "ti,c64x+timer64"; + ti,core-mask = < 0x01 >; + reg = <0x2960000 0x40>; + }; + + device-state-controller@2880800 { + compatible = "ti,c64x+dscr"; + reg = <0x02880800 0x400>; + + ti,dscr-devstat = <0x004>; + ti,dscr-silicon-rev = <0x014 28 0xf>; + ti,dscr-mac-fuse-regs = <0x34 3 4 5 6 + 0x38 0 0 1 2>; + }; + + clock-controller@29a0000 { + compatible = "ti,c6474-pll", "ti,c64x+pll"; + reg = <0x029a0000 0x200>; + ti,c64x+pll-bypass-delay = <120>; + ti,c64x+pll-reset-delay = <30000>; + ti,c64x+pll-lock-delay = <60000>; + }; + }; +}; diff --git a/arch/c6x/boot/linked_dtb.S b/arch/c6x/boot/linked_dtb.S new file mode 100644 index 0000000..57a4454 --- /dev/null +++ b/arch/c6x/boot/linked_dtb.S @@ -0,0 +1,2 @@ +.section __fdt_blob,"a" +.incbin "arch/c6x/boot/builtin.dtb" diff --git a/arch/c6x/configs/dsk6455_defconfig b/arch/c6x/configs/dsk6455_defconfig new file mode 100644 index 0000000..4663487 --- /dev/null +++ b/arch/c6x/configs/dsk6455_defconfig @@ -0,0 +1,44 @@ +CONFIG_SOC_TMS320C6455=y +CONFIG_EXPERIMENTAL=y +# CONFIG_LOCALVERSION_AUTO is not set +CONFIG_SYSVIPC=y +CONFIG_SPARSE_IRQ=y +CONFIG_LOG_BUF_SHIFT=14 +CONFIG_NAMESPACES=y +# CONFIG_UTS_NS is not set +# CONFIG_USER_NS is not set +# CONFIG_PID_NS is not set +CONFIG_BLK_DEV_INITRD=y +CONFIG_CC_OPTIMIZE_FOR_SIZE=y +CONFIG_EXPERT=y +# CONFIG_FUTEX is not set +# CONFIG_SLUB_DEBUG is not set +CONFIG_MODULES=y +CONFIG_MODULE_FORCE_LOAD=y +CONFIG_MODULE_UNLOAD=y +CONFIG_MODULE_FORCE_UNLOAD=y +CONFIG_CMDLINE_BOOL=y +CONFIG_CMDLINE="" +CONFIG_NO_HZ=y +CONFIG_HIGH_RES_TIMERS=y +CONFIG_BLK_DEV_LOOP=y +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_COUNT=2 +CONFIG_BLK_DEV_RAM_SIZE=17000 +CONFIG_MISC_DEVICES=y +# CONFIG_INPUT is not set +# CONFIG_SERIO is not set +# CONFIG_VT is not set +# CONFIG_HW_RANDOM is not set +# CONFIG_HWMON is not set +# CONFIG_USB_SUPPORT is not set +# CONFIG_IOMMU_SUPPORT is not set +# CONFIG_MISC_FILESYSTEMS is not set +CONFIG_CRC16=y +# CONFIG_ENABLE_MUST_CHECK is not set +# CONFIG_SCHED_DEBUG is not set +# CONFIG_DEBUG_BUGVERBOSE is not set +CONFIG_MTD=y +CONFIG_MTD_CFI=y +CONFIG_MTD_CFI_AMDSTD=y +CONFIG_MTD_PHYSMAP_OF=y diff --git a/arch/c6x/configs/evmc6457_defconfig b/arch/c6x/configs/evmc6457_defconfig new file mode 100644 index 0000000..bba40e1 --- /dev/null +++ b/arch/c6x/configs/evmc6457_defconfig @@ -0,0 +1,41 @@ +CONFIG_SOC_TMS320C6457=y +CONFIG_EXPERIMENTAL=y +# CONFIG_LOCALVERSION_AUTO is not set +CONFIG_SYSVIPC=y +CONFIG_SPARSE_IRQ=y +CONFIG_LOG_BUF_SHIFT=14 +CONFIG_NAMESPACES=y +# CONFIG_UTS_NS is not set +# CONFIG_USER_NS is not set +# CONFIG_PID_NS is not set +CONFIG_BLK_DEV_INITRD=y +CONFIG_CC_OPTIMIZE_FOR_SIZE=y +CONFIG_EXPERT=y +# CONFIG_FUTEX is not set +# CONFIG_SLUB_DEBUG is not set +CONFIG_MODULES=y +CONFIG_MODULE_FORCE_LOAD=y +CONFIG_MODULE_UNLOAD=y +CONFIG_MODULE_FORCE_UNLOAD=y +CONFIG_CMDLINE_BOOL=y +CONFIG_CMDLINE="" +CONFIG_BOARD_EVM6457=y +CONFIG_NO_HZ=y +CONFIG_HIGH_RES_TIMERS=y +CONFIG_BLK_DEV_LOOP=y +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_COUNT=2 +CONFIG_BLK_DEV_RAM_SIZE=17000 +CONFIG_MISC_DEVICES=y +# CONFIG_INPUT is not set +# CONFIG_SERIO is not set +# CONFIG_VT is not set +# CONFIG_HW_RANDOM is not set +# CONFIG_HWMON is not set +# CONFIG_USB_SUPPORT is not set +# CONFIG_IOMMU_SUPPORT is not set +# CONFIG_MISC_FILESYSTEMS is not set +CONFIG_CRC16=y +# CONFIG_ENABLE_MUST_CHECK is not set +# CONFIG_SCHED_DEBUG is not set +# CONFIG_DEBUG_BUGVERBOSE is not set diff --git a/arch/c6x/configs/evmc6472_defconfig b/arch/c6x/configs/evmc6472_defconfig new file mode 100644 index 0000000..8c46155 --- /dev/null +++ b/arch/c6x/configs/evmc6472_defconfig @@ -0,0 +1,42 @@ +CONFIG_SOC_TMS320C6472=y +CONFIG_EXPERIMENTAL=y +# CONFIG_LOCALVERSION_AUTO is not set +CONFIG_SYSVIPC=y +CONFIG_SPARSE_IRQ=y +CONFIG_LOG_BUF_SHIFT=14 +CONFIG_NAMESPACES=y +# CONFIG_UTS_NS is not set +# CONFIG_USER_NS is not set +# CONFIG_PID_NS is not set +CONFIG_BLK_DEV_INITRD=y +CONFIG_CC_OPTIMIZE_FOR_SIZE=y +CONFIG_EXPERT=y +# CONFIG_FUTEX is not set +# CONFIG_SLUB_DEBUG is not set +CONFIG_MODULES=y +CONFIG_MODULE_FORCE_LOAD=y +CONFIG_MODULE_UNLOAD=y +CONFIG_MODULE_FORCE_UNLOAD=y +CONFIG_CMDLINE_BOOL=y +CONFIG_CMDLINE="" +# CONFIG_CMDLINE_FORCE is not set +CONFIG_BOARD_EVM6472=y +CONFIG_NO_HZ=y +CONFIG_HIGH_RES_TIMERS=y +CONFIG_BLK_DEV_LOOP=y +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_COUNT=2 +CONFIG_BLK_DEV_RAM_SIZE=17000 +CONFIG_MISC_DEVICES=y +# CONFIG_INPUT is not set +# CONFIG_SERIO is not set +# CONFIG_VT is not set +# CONFIG_HW_RANDOM is not set +# CONFIG_HWMON is not set +# CONFIG_USB_SUPPORT is not set +# CONFIG_IOMMU_SUPPORT is not set +# CONFIG_MISC_FILESYSTEMS is not set +CONFIG_CRC16=y +# CONFIG_ENABLE_MUST_CHECK is not set +# CONFIG_SCHED_DEBUG is not set +# CONFIG_DEBUG_BUGVERBOSE is not set diff --git a/arch/c6x/configs/evmc6474_defconfig b/arch/c6x/configs/evmc6474_defconfig new file mode 100644 index 0000000..15533f6 --- /dev/null +++ b/arch/c6x/configs/evmc6474_defconfig @@ -0,0 +1,42 @@ +CONFIG_SOC_TMS320C6474=y +CONFIG_EXPERIMENTAL=y +# CONFIG_LOCALVERSION_AUTO is not set +CONFIG_SYSVIPC=y +CONFIG_SPARSE_IRQ=y +CONFIG_LOG_BUF_SHIFT=14 +CONFIG_NAMESPACES=y +# CONFIG_UTS_NS is not set +# CONFIG_USER_NS is not set +# CONFIG_PID_NS is not set +CONFIG_BLK_DEV_INITRD=y +CONFIG_CC_OPTIMIZE_FOR_SIZE=y +CONFIG_EXPERT=y +# CONFIG_FUTEX is not set +# CONFIG_SLUB_DEBUG is not set +CONFIG_MODULES=y +CONFIG_MODULE_FORCE_LOAD=y +CONFIG_MODULE_UNLOAD=y +CONFIG_MODULE_FORCE_UNLOAD=y +CONFIG_CMDLINE_BOOL=y +CONFIG_CMDLINE="" +# CONFIG_CMDLINE_FORCE is not set +CONFIG_BOARD_EVM6474=y +CONFIG_NO_HZ=y +CONFIG_HIGH_RES_TIMERS=y +CONFIG_BLK_DEV_LOOP=y +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_COUNT=2 +CONFIG_BLK_DEV_RAM_SIZE=17000 +CONFIG_MISC_DEVICES=y +# CONFIG_INPUT is not set +# CONFIG_SERIO is not set +# CONFIG_VT is not set +# CONFIG_HW_RANDOM is not set +# CONFIG_HWMON is not set +# CONFIG_USB_SUPPORT is not set +# CONFIG_IOMMU_SUPPORT is not set +# CONFIG_MISC_FILESYSTEMS is not set +CONFIG_CRC16=y +# CONFIG_ENABLE_MUST_CHECK is not set +# CONFIG_SCHED_DEBUG is not set +# CONFIG_DEBUG_BUGVERBOSE is not set diff --git a/arch/c6x/include/asm/Kbuild b/arch/c6x/include/asm/Kbuild new file mode 100644 index 0000000..13dcf78 --- /dev/null +++ b/arch/c6x/include/asm/Kbuild @@ -0,0 +1,54 @@ +include include/asm-generic/Kbuild.asm + +generic-y += atomic.h +generic-y += auxvec.h +generic-y += bitsperlong.h +generic-y += bug.h +generic-y += bugs.h +generic-y += cputime.h +generic-y += current.h +generic-y += device.h +generic-y += div64.h +generic-y += dma.h +generic-y += emergency-restart.h +generic-y += errno.h +generic-y += fb.h +generic-y += fcntl.h +generic-y += futex.h +generic-y += hw_irq.h +generic-y += io.h +generic-y += ioctl.h +generic-y += ioctls.h +generic-y += ipcbuf.h +generic-y += irq_regs.h +generic-y += kdebug.h +generic-y += kmap_types.h +generic-y += local.h +generic-y += mman.h +generic-y += mmu_context.h +generic-y += msgbuf.h +generic-y += param.h +generic-y += pci.h +generic-y += percpu.h +generic-y += pgalloc.h +generic-y += poll.h +generic-y += posix_types.h +generic-y += resource.h +generic-y += scatterlist.h +generic-y += segment.h +generic-y += sembuf.h +generic-y += shmbuf.h +generic-y += shmparam.h +generic-y += siginfo.h +generic-y += socket.h +generic-y += sockios.h +generic-y += stat.h +generic-y += statfs.h +generic-y += termbits.h +generic-y += termios.h +generic-y += tlbflush.h +generic-y += topology.h +generic-y += types.h +generic-y += ucontext.h +generic-y += user.h +generic-y += vga.h diff --git a/arch/c6x/include/asm/asm-offsets.h b/arch/c6x/include/asm/asm-offsets.h new file mode 100644 index 0000000..d370ee3 --- /dev/null +++ b/arch/c6x/include/asm/asm-offsets.h @@ -0,0 +1 @@ +#include <generated/asm-offsets.h> diff --git a/arch/c6x/include/asm/bitops.h b/arch/c6x/include/asm/bitops.h new file mode 100644 index 0000000..39ab7e8 --- /dev/null +++ b/arch/c6x/include/asm/bitops.h @@ -0,0 +1,105 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_BITOPS_H +#define _ASM_C6X_BITOPS_H + +#ifdef __KERNEL__ + +#include <linux/bitops.h> + +#include <asm/system.h> +#include <asm/byteorder.h> + +/* + * clear_bit() doesn't provide any barrier for the compiler. + */ +#define smp_mb__before_clear_bit() barrier() +#define smp_mb__after_clear_bit() barrier() + +/* + * We are lucky, DSP is perfect for bitops: do it in 3 cycles + */ + +/** + * __ffs - find first bit in word. + * @word: The word to search + * + * Undefined if no bit exists, so code should check against 0 first. + * Note __ffs(0) = undef, __ffs(1) = 0, __ffs(0x80000000) = 31. + * + */ +static inline unsigned long __ffs(unsigned long x) +{ + asm (" bitr .M1 %0,%0\n" + " nop\n" + " lmbd .L1 1,%0,%0\n" + : "+a"(x)); + + return x; +} + +/* + * ffz - find first zero in word. + * @word: The word to search + * + * Undefined if no zero exists, so code should check against ~0UL first. + */ +#define ffz(x) __ffs(~(x)) + +/** + * fls - find last (most-significant) bit set + * @x: the word to search + * + * This is defined the same way as ffs. + * Note fls(0) = 0, fls(1) = 1, fls(0x80000000) = 32. + */ +static inline int fls(int x) +{ + if (!x) + return 0; + + asm (" lmbd .L1 1,%0,%0\n" : "+a"(x)); + + return 32 - x; +} + +/** + * ffs - find first bit set + * @x: the word to search + * + * This is defined the same way as + * the libc and compiler builtin ffs routines, therefore + * differs in spirit from the above ffz (man ffs). + * Note ffs(0) = 0, ffs(1) = 1, ffs(0x80000000) = 32. + */ +static inline int ffs(int x) +{ + if (!x) + return 0; + + return __ffs(x) + 1; +} + +#include <asm-generic/bitops/__fls.h> +#include <asm-generic/bitops/fls64.h> +#include <asm-generic/bitops/find.h> + +#include <asm-generic/bitops/sched.h> +#include <asm-generic/bitops/hweight.h> +#include <asm-generic/bitops/lock.h> + +#include <asm-generic/bitops/atomic.h> +#include <asm-generic/bitops/non-atomic.h> +#include <asm-generic/bitops/le.h> +#include <asm-generic/bitops/ext2-atomic.h> + +#endif /* __KERNEL__ */ +#endif /* _ASM_C6X_BITOPS_H */ diff --git a/arch/c6x/include/asm/byteorder.h b/arch/c6x/include/asm/byteorder.h new file mode 100644 index 0000000..166038d --- /dev/null +++ b/arch/c6x/include/asm/byteorder.h @@ -0,0 +1,12 @@ +#ifndef _ASM_C6X_BYTEORDER_H +#define _ASM_C6X_BYTEORDER_H + +#include <asm/types.h> + +#ifdef _BIG_ENDIAN +#include <linux/byteorder/big_endian.h> +#else /* _BIG_ENDIAN */ +#include <linux/byteorder/little_endian.h> +#endif /* _BIG_ENDIAN */ + +#endif /* _ASM_BYTEORDER_H */ diff --git a/arch/c6x/include/asm/cache.h b/arch/c6x/include/asm/cache.h new file mode 100644 index 0000000..6d521d9 --- /dev/null +++ b/arch/c6x/include/asm/cache.h @@ -0,0 +1,90 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2005, 2006, 2009, 2010 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_CACHE_H +#define _ASM_C6X_CACHE_H + +#include <linux/irqflags.h> + +/* + * Cache line size + */ +#define L1D_CACHE_BYTES 64 +#define L1P_CACHE_BYTES 32 +#define L2_CACHE_BYTES 128 + +/* + * L2 used as cache + */ +#define L2MODE_SIZE L2MODE_256K_CACHE + +/* + * For practical reasons the L1_CACHE_BYTES defines should not be smaller than + * the L2 line size + */ +#define L1_CACHE_BYTES L2_CACHE_BYTES + +#define L2_CACHE_ALIGN_LOW(x) \ + (((x) & ~(L2_CACHE_BYTES - 1))) +#define L2_CACHE_ALIGN_UP(x) \ + (((x) + (L2_CACHE_BYTES - 1)) & ~(L2_CACHE_BYTES - 1)) +#define L2_CACHE_ALIGN_CNT(x) \ + (((x) + (sizeof(int) - 1)) & ~(sizeof(int) - 1)) + +#define ARCH_DMA_MINALIGN L1_CACHE_BYTES +#define ARCH_SLAB_MINALIGN L1_CACHE_BYTES + +/* + * This is the granularity of hardware cacheability control. + */ +#define CACHEABILITY_ALIGN 0x01000000 + +/* + * Align a physical address to MAR regions + */ +#define CACHE_REGION_START(v) \ + (((u32) (v)) & ~(CACHEABILITY_ALIGN - 1)) +#define CACHE_REGION_END(v) \ + (((u32) (v) + (CACHEABILITY_ALIGN - 1)) & ~(CACHEABILITY_ALIGN - 1)) + +extern void __init c6x_cache_init(void); + +extern void enable_caching(unsigned long start, unsigned long end); +extern void disable_caching(unsigned long start, unsigned long end); + +extern void L1_cache_off(void); +extern void L1_cache_on(void); + +extern void L1P_cache_global_invalidate(void); +extern void L1D_cache_global_invalidate(void); +extern void L1D_cache_global_writeback(void); +extern void L1D_cache_global_writeback_invalidate(void); +extern void L2_cache_set_mode(unsigned int mode); +extern void L2_cache_global_writeback_invalidate(void); +extern void L2_cache_global_writeback(void); + +extern void L1P_cache_block_invalidate(unsigned int start, unsigned int end); +extern void L1D_cache_block_invalidate(unsigned int start, unsigned int end); +extern void L1D_cache_block_writeback_invalidate(unsigned int start, + unsigned int end); +extern void L1D_cache_block_writeback(unsigned int start, unsigned int end); +extern void L2_cache_block_invalidate(unsigned int start, unsigned int end); +extern void L2_cache_block_writeback(unsigned int start, unsigned int end); +extern void L2_cache_block_writeback_invalidate(unsigned int start, + unsigned int end); +extern void L2_cache_block_invalidate_nowait(unsigned int start, + unsigned int end); +extern void L2_cache_block_writeback_nowait(unsigned int start, + unsigned int end); + +extern void L2_cache_block_writeback_invalidate_nowait(unsigned int start, + unsigned int end); + +#endif /* _ASM_C6X_CACHE_H */ diff --git a/arch/c6x/include/asm/cacheflush.h b/arch/c6x/include/asm/cacheflush.h new file mode 100644 index 0000000..df5db90 --- /dev/null +++ b/arch/c6x/include/asm/cacheflush.h @@ -0,0 +1,65 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_CACHEFLUSH_H +#define _ASM_C6X_CACHEFLUSH_H + +#include <linux/spinlock.h> + +#include <asm/setup.h> +#include <asm/cache.h> +#include <asm/mman.h> +#include <asm/page.h> +#include <asm/string.h> + +/* + * virtually-indexed cache management (our cache is physically indexed) + */ +#define flush_cache_all() do {} while (0) +#define flush_cache_mm(mm) do {} while (0) +#define flush_cache_dup_mm(mm) do {} while (0) +#define flush_cache_range(mm, start, end) do {} while (0) +#define flush_cache_page(vma, vmaddr, pfn) do {} while (0) +#define flush_cache_vmap(start, end) do {} while (0) +#define flush_cache_vunmap(start, end) do {} while (0) +#define ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE 0 +#define flush_dcache_page(page) do {} while (0) +#define flush_dcache_mmap_lock(mapping) do {} while (0) +#define flush_dcache_mmap_unlock(mapping) do {} while (0) + +/* + * physically-indexed cache management + */ +#define flush_icache_range(s, e) \ +do { \ + L1D_cache_block_writeback((s), (e)); \ + L1P_cache_block_invalidate((s), (e)); \ +} while (0) + +#define flush_icache_page(vma, page) \ +do { \ + if ((vma)->vm_flags & PROT_EXEC) \ + L1D_cache_block_writeback_invalidate(page_address(page), \ + (unsigned long) page_address(page) + PAGE_SIZE)); \ + L1P_cache_block_invalidate(page_address(page), \ + (unsigned long) page_address(page) + PAGE_SIZE)); \ +} while (0) + + +#define copy_to_user_page(vma, page, vaddr, dst, src, len) \ +do { \ + memcpy(dst, src, len); \ + flush_icache_range((unsigned) (dst), (unsigned) (dst) + (len)); \ +} while (0) + +#define copy_from_user_page(vma, page, vaddr, dst, src, len) \ + memcpy(dst, src, len) + +#endif /* _ASM_C6X_CACHEFLUSH_H */ diff --git a/arch/c6x/include/asm/checksum.h b/arch/c6x/include/asm/checksum.h new file mode 100644 index 0000000..7246816 --- /dev/null +++ b/arch/c6x/include/asm/checksum.h @@ -0,0 +1,34 @@ +/* + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_CHECKSUM_H +#define _ASM_C6X_CHECKSUM_H + +static inline __wsum +csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len, + unsigned short proto, __wsum sum) +{ + unsigned long long tmp; + + asm ("add .d1 %1,%5,%1\n" + "|| addu .l1 %3,%4,%0\n" + "addu .l1 %2,%0,%0\n" +#ifndef CONFIG_CPU_BIG_ENDIAN + "|| shl .s1 %1,8,%1\n" +#endif + "addu .l1 %1,%0,%0\n" + "add .l1 %P0,%p0,%2\n" + : "=&a"(tmp), "+a"(len), "+a"(sum) + : "a" (saddr), "a" (daddr), "a" (proto)); + return sum; +} +#define csum_tcpudp_nofold csum_tcpudp_nofold + +#include <asm-generic/checksum.h> + +#endif /* _ASM_C6X_CHECKSUM_H */ diff --git a/arch/c6x/include/asm/clkdev.h b/arch/c6x/include/asm/clkdev.h new file mode 100644 index 0000000..76a070b --- /dev/null +++ b/arch/c6x/include/asm/clkdev.h @@ -0,0 +1,22 @@ +#ifndef _ASM_CLKDEV_H +#define _ASM_CLKDEV_H + +#include <linux/slab.h> + +struct clk; + +static inline int __clk_get(struct clk *clk) +{ + return 1; +} + +static inline void __clk_put(struct clk *clk) +{ +} + +static inline struct clk_lookup_alloc *__clkdev_alloc(size_t size) +{ + return kzalloc(size, GFP_KERNEL); +} + +#endif /* _ASM_CLKDEV_H */ diff --git a/arch/c6x/include/asm/clock.h b/arch/c6x/include/asm/clock.h new file mode 100644 index 0000000..bcf42b2 --- /dev/null +++ b/arch/c6x/include/asm/clock.h @@ -0,0 +1,148 @@ +/* + * TI C64X clock definitions + * + * Copyright (C) 2010, 2011 Texas Instruments. + * Contributed by: Mark Salter <msalter@redhat.com> + * + * Copied heavily from arm/mach-davinci/clock.h, so: + * + * Copyright (C) 2006-2007 Texas Instruments. + * Copyright (C) 2008-2009 Deep Root Systems, LLC + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef _ASM_C6X_CLOCK_H +#define _ASM_C6X_CLOCK_H + +#ifndef __ASSEMBLER__ + +#include <linux/list.h> + +/* PLL/Reset register offsets */ +#define PLLCTL 0x100 +#define PLLM 0x110 +#define PLLPRE 0x114 +#define PLLDIV1 0x118 +#define PLLDIV2 0x11c +#define PLLDIV3 0x120 +#define PLLPOST 0x128 +#define PLLCMD 0x138 +#define PLLSTAT 0x13c +#define PLLALNCTL 0x140 +#define PLLDCHANGE 0x144 +#define PLLCKEN 0x148 +#define PLLCKSTAT 0x14c +#define PLLSYSTAT 0x150 +#define PLLDIV4 0x160 +#define PLLDIV5 0x164 +#define PLLDIV6 0x168 +#define PLLDIV7 0x16c +#define PLLDIV8 0x170 +#define PLLDIV9 0x174 +#define PLLDIV10 0x178 +#define PLLDIV11 0x17c +#define PLLDIV12 0x180 +#define PLLDIV13 0x184 +#define PLLDIV14 0x188 +#define PLLDIV15 0x18c +#define PLLDIV16 0x190 + +/* PLLM register bits */ +#define PLLM_PLLM_MASK 0xff +#define PLLM_VAL(x) ((x) - 1) + +/* PREDIV register bits */ +#define PLLPREDIV_EN BIT(15) +#define PLLPREDIV_VAL(x) ((x) - 1) + +/* PLLCTL register bits */ +#define PLLCTL_PLLEN BIT(0) +#define PLLCTL_PLLPWRDN BIT(1) +#define PLLCTL_PLLRST BIT(3) +#define PLLCTL_PLLDIS BIT(4) +#define PLLCTL_PLLENSRC BIT(5) +#define PLLCTL_CLKMODE BIT(8) + +/* PLLCMD register bits */ +#define PLLCMD_GOSTAT BIT(0) + +/* PLLSTAT register bits */ +#define PLLSTAT_GOSTAT BIT(0) + +/* PLLDIV register bits */ +#define PLLDIV_EN BIT(15) +#define PLLDIV_RATIO_MASK 0x1f +#define PLLDIV_RATIO(x) ((x) - 1) + +struct pll_data; + +struct clk { + struct list_head node; + struct module *owner; + const char *name; + unsigned long rate; + int usecount; + u32 flags; + struct clk *parent; + struct list_head children; /* list of children */ + struct list_head childnode; /* parent's child list node */ + struct pll_data *pll_data; + u32 div; + unsigned long (*recalc) (struct clk *); + int (*set_rate) (struct clk *clk, unsigned long rate); + int (*round_rate) (struct clk *clk, unsigned long rate); +}; + +/* Clock flags: SoC-specific flags start at BIT(16) */ +#define ALWAYS_ENABLED BIT(1) +#define CLK_PLL BIT(2) /* PLL-derived clock */ +#define PRE_PLL BIT(3) /* source is before PLL mult/div */ +#define FIXED_DIV_PLL BIT(4) /* fixed divisor from PLL */ +#define FIXED_RATE_PLL BIT(5) /* fixed ouput rate PLL */ + +#define MAX_PLL_SYSCLKS 16 + +struct pll_data { + void __iomem *base; + u32 num; + u32 flags; + u32 input_rate; + u32 bypass_delay; /* in loops */ + u32 reset_delay; /* in loops */ + u32 lock_delay; /* in loops */ + struct clk sysclks[MAX_PLL_SYSCLKS + 1]; +}; + +/* pll_data flag bit */ +#define PLL_HAS_PRE BIT(0) +#define PLL_HAS_MUL BIT(1) +#define PLL_HAS_POST BIT(2) + +#define CLK(dev, con, ck) \ + { \ + .dev_id = dev, \ + .con_id = con, \ + .clk = ck, \ + } \ + +extern void c6x_clks_init(struct clk_lookup *clocks); +extern int clk_register(struct clk *clk); +extern void clk_unregister(struct clk *clk); +extern void c64x_setup_clocks(void); + +extern struct pll_data c6x_soc_pll1; + +extern struct clk clkin1; +extern struct clk c6x_core_clk; +extern struct clk c6x_i2c_clk; +extern struct clk c6x_watchdog_clk; +extern struct clk c6x_mcbsp1_clk; +extern struct clk c6x_mcbsp2_clk; +extern struct clk c6x_mdio_clk; + +#endif + +#endif /* _ASM_C6X_CLOCK_H */ diff --git a/arch/c6x/include/asm/delay.h b/arch/c6x/include/asm/delay.h new file mode 100644 index 0000000..f314c2e --- /dev/null +++ b/arch/c6x/include/asm/delay.h @@ -0,0 +1,67 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_DELAY_H +#define _ASM_C6X_DELAY_H + +#include <linux/kernel.h> + +extern unsigned int ticks_per_ns_scaled; + +static inline void __delay(unsigned long loops) +{ + uint32_t tmp; + + /* 6 cycles per loop */ + asm volatile (" mv .s1 %0,%1\n" + "0: [%1] b .s1 0b\n" + " add .l1 -6,%0,%0\n" + " cmplt .l1 1,%0,%1\n" + " nop 3\n" + : "+a"(loops), "=A"(tmp)); +} + +static inline void _c6x_tickdelay(unsigned int x) +{ + uint32_t cnt, endcnt; + + asm volatile (" mvc .s2 TSCL,%0\n" + " add .s2x %0,%1,%2\n" + " || mvk .l2 1,B0\n" + "0: [B0] b .s2 0b\n" + " mvc .s2 TSCL,%0\n" + " sub .s2 %0,%2,%0\n" + " cmpgt .l2 0,%0,B0\n" + " nop 2\n" + : "=b"(cnt), "+a"(x), "=b"(endcnt) : : "B0"); +} + +/* use scaled math to avoid slow division */ +#define C6X_NDELAY_SCALE 10 + +static inline void _ndelay(unsigned int n) +{ + _c6x_tickdelay((ticks_per_ns_scaled * n) >> C6X_NDELAY_SCALE); +} + +static inline void _udelay(unsigned int n) +{ + while (n >= 10) { + _ndelay(10000); + n -= 10; + } + while (n-- > 0) + _ndelay(1000); +} + +#define udelay(x) _udelay((unsigned int)(x)) +#define ndelay(x) _ndelay((unsigned int)(x)) + +#endif /* _ASM_C6X_DELAY_H */ diff --git a/arch/c6x/include/asm/dma-mapping.h b/arch/c6x/include/asm/dma-mapping.h new file mode 100644 index 0000000..03579fd --- /dev/null +++ b/arch/c6x/include/asm/dma-mapping.h @@ -0,0 +1,91 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot <aurelien.jacquiot@ti.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ +#ifndef _ASM_C6X_DMA_MAPPING_H +#define _ASM_C6X_DMA_MAPPING_H + +#include <linux/dma-debug.h> +#include <asm-generic/dma-coherent.h> + +#define dma_supported(d, m) 1 + +static inline int dma_set_mask(struct device *dev, u64 dma_mask) +{ + if (!dev->dma_mask || !dma_supported(dev, dma_mask)) + return -EIO; + + *dev->dma_mask = dma_mask; + + return 0; +} + +/* + * DMA errors are defined by all-bits-set in the DMA address. + */ +static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) +{ + return dma_addr == ~0; +} + +extern dma_addr_t dma_map_single(struct device *dev, void *cpu_addr, + size_t size, enum dma_data_direction dir); + +extern void dma_unmap_single(struct device *dev, dma_addr_t handle, + size_t size, enum dma_data_direction dir); + +extern int dma_map_sg(struct device *dev, struct scatterlist *sglist, + int nents, enum dma_data_direction direction); + +extern void dma_unmap_sg(struct device *dev, struct scatterlist *sglist, + int nents, enum dma_data_direction direction); + +static inline dma_addr_t dma_map_page(struct device *dev, struct page *page, + unsigned long offset, size_t size, + enum dma_data_direction dir) +{ + dma_addr_t handle; + + handle = dma_map_single(dev, page_address(page) + offset, size, dir); + + debug_dma_map_page(dev, page, offset, size, dir, handle, false); + + return handle; +} + +static inline void dma_unmap_page(struct device *dev, dma_addr_t handle, + size_t size, enum dma_data_direction dir) +{ + dma_unmap_single(dev, handle, size, dir); + + debug_dma_unmap_page(dev, handle, size, dir, false); +} + +extern void dma_sync_single_for_cpu(struct device *dev, dma_addr_t handle, + size_t size, enum dma_data_direction dir); + +extern void dma_sync_single_for_device(struct device *dev, dma_addr_t handle, + size_t size, + enum dma_data_direction dir); + +extern void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir); + +extern void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir); + +extern void coherent_mem_init(u32 start, u32 size); +extern void *dma_alloc_coherent(struct device *, size_t, dma_addr_t *, gfp_t); +extern void dma_free_coherent(struct device *, size_t, void *, dma_addr_t); + +#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent((d), (s), (h), (f)) +#define dma_free_noncoherent(d, s, v, h) dma_free_coherent((d), (s), (v), (h)) + +#endif /* _ASM_C6X_DMA_MAPPING_H */ diff --git a/arch/c6x/include/asm/dscr.h b/arch/c6x/include/asm/dscr.h new file mode 100644 index 0000000..561ba83 --- /dev/null +++ b/arch/c6x/include/asm/dscr.h @@ -0,0 +1,34 @@ +/* + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ +#ifndef _ASM_C6X_DSCR_H +#define _ASM_C6X_DSCR_H + +enum dscr_devstate_t { + DSCR_DEVSTATE_ENABLED, + DSCR_DEVSTATE_DISABLED, +}; + +/* + * Set the device state of the device with the given ID. + * + * Individual drivers should use this to enable or disable the + * hardware device. The devid used to identify the device being + * controlled should be a property in the device's tree node. + */ +extern void dscr_set_devstate(int devid, enum dscr_devstate_t state); + +/* + * Assert or de-assert an RMII reset. + */ +extern void dscr_rmii_reset(int id, int assert); + +extern void dscr_probe(void); + +#endif /* _ASM_C6X_DSCR_H */ diff --git a/arch/c6x/include/asm/elf.h b/arch/c6x/include/asm/elf.h new file mode 100644 index 0000000..d57865b --- /dev/null +++ b/arch/c6x/include/asm/elf.h @@ -0,0 +1,113 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_ELF_H +#define _ASM_C6X_ELF_H + +/* + * ELF register definitions.. + */ +#include <asm/ptrace.h> + +typedef unsigned long elf_greg_t; +typedef unsigned long elf_fpreg_t; + +#define ELF_NGREG 58 +#define ELF_NFPREG 1 + +typedef elf_greg_t elf_gregset_t[ELF_NGREG]; +typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG]; + +/* + * This is used to ensure we don't load something for the wrong architecture. + */ +#define elf_check_arch(x) ((x)->e_machine == EM_TI_C6000) + +#define elf_check_const_displacement(x) (1) + +/* + * These are used to set parameters in the core dumps. + */ +#ifdef __LITTLE_ENDIAN__ +#define ELF_DATA ELFDATA2LSB +#else +#define ELF_DATA ELFDATA2MSB +#endif + +#define ELF_CLASS ELFCLASS32 +#define ELF_ARCH EM_TI_C6000 + +/* Nothing for now. Need to setup DP... */ +#define ELF_PLAT_INIT(_r) + +#define USE_ELF_CORE_DUMP +#define ELF_EXEC_PAGESIZE 4096 + +#define ELF_CORE_COPY_REGS(_dest, _regs) \ + memcpy((char *) &_dest, (char *) _regs, \ + sizeof(struct pt_regs)); + +/* This yields a mask that user programs can use to figure out what + instruction set this cpu supports. */ + +#define ELF_HWCAP (0) + +/* This yields a string that ld.so will use to load implementation + specific libraries for optimization. This is more specific in + intent than poking at uname or /proc/cpuinfo. */ + +#define ELF_PLATFORM (NULL) + +#define SET_PERSONALITY(ex) set_personality(PER_LINUX) + +/* C6X specific section types */ +#define SHT_C6000_UNWIND 0x70000001 +#define SHT_C6000_PREEMPTMAP 0x70000002 +#define SHT_C6000_ATTRIBUTES 0x70000003 + +/* C6X specific DT_ tags */ +#define DT_C6000_DSBT_BASE 0x70000000 +#define DT_C6000_DSBT_SIZE 0x70000001 +#define DT_C6000_PREEMPTMAP 0x70000002 +#define DT_C6000_DSBT_INDEX 0x70000003 + +/* C6X specific relocs */ +#define R_C6000_NONE 0 +#define R_C6000_ABS32 1 +#define R_C6000_ABS16 2 +#define R_C6000_ABS8 3 +#define R_C6000_PCR_S21 4 +#define R_C6000_PCR_S12 5 +#define R_C6000_PCR_S10 6 +#define R_C6000_PCR_S7 7 +#define R_C6000_ABS_S16 8 +#define R_C6000_ABS_L16 9 +#define R_C6000_ABS_H16 10 +#define R_C6000_SBR_U15_B 11 +#define R_C6000_SBR_U15_H 12 +#define R_C6000_SBR_U15_W 13 +#define R_C6000_SBR_S16 14 +#define R_C6000_SBR_L16_B 15 +#define R_C6000_SBR_L16_H 16 +#define R_C6000_SBR_L16_W 17 +#define R_C6000_SBR_H16_B 18 +#define R_C6000_SBR_H16_H 19 +#define R_C6000_SBR_H16_W 20 +#define R_C6000_SBR_GOT_U15_W 21 +#define R_C6000_SBR_GOT_L16_W 22 +#define R_C6000_SBR_GOT_H16_W 23 +#define R_C6000_DSBT_INDEX 24 +#define R_C6000_PREL31 25 +#define R_C6000_COPY 26 +#define R_C6000_ALIGN 253 +#define R_C6000_FPHEAD 254 +#define R_C6000_NOCMP 255 + +#endif /*_ASM_C6X_ELF_H */ diff --git a/arch/c6x/include/asm/ftrace.h b/arch/c6x/include/asm/ftrace.h new file mode 100644 index 0000000..3701958 --- /dev/null +++ b/arch/c6x/include/asm/ftrace.h @@ -0,0 +1,6 @@ +#ifndef _ASM_C6X_FTRACE_H +#define _ASM_C6X_FTRACE_H + +/* empty */ + +#endif /* _ASM_C6X_FTRACE_H */ diff --git a/arch/c6x/include/asm/hardirq.h b/arch/c6x/include/asm/hardirq.h new file mode 100644 index 0000000..9621954 --- /dev/null +++ b/arch/c6x/include/asm/hardirq.h @@ -0,0 +1,20 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef _ASM_C6X_HARDIRQ_H +#define _ASM_C6X_HARDIRQ_H + +extern void ack_bad_irq(int irq); +#define ack_bad_irq ack_bad_irq + +#include <asm-generic/hardirq.h> + +#endif /* _ASM_C6X_HARDIRQ_H */ diff --git a/arch/c6x/include/asm/irq.h b/arch/c6x/include/asm/irq.h new file mode 100644 index 0000000..a6ae3c9 --- /dev/null +++ b/arch/c6x/include/asm/irq.h @@ -0,0 +1,302 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2006, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * Large parts taken directly from powerpc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_IRQ_H +#define _ASM_C6X_IRQ_H + +#include <linux/threads.h> +#include <linux/list.h> +#include <linux/radix-tree.h> +#include <asm/percpu.h> + +#define irq_canonicalize(irq) (irq) + +/* + * The C64X+ core has 16 IRQ vectors. One each is used by Reset and NMI. Two + * are reserved. The remaining 12 vectors are used to route SoC interrupts. + * These interrupt vectors are prioritized with IRQ 4 having the highest + * priority and IRQ 15 having the lowest. + * + * The C64x+ megamodule provides a PIC which combines SoC IRQ sources into a + * single core IRQ vector. There are four combined sources, each of which + * feed into one of the 12 general interrupt vectors. The remaining 8 vectors + * can each route a single SoC interrupt directly. + */ +#define NR_PRIORITY_IRQS 16 + +#define NR_IRQS_LEGACY NR_PRIORITY_IRQS + +/* Total number of virq in the platform */ +#define NR_IRQS 256 + +/* This number is used when no interrupt has been assigned */ +#define NO_IRQ 0 + +/* This type is the placeholder for a hardware interrupt number. It has to + * be big enough to enclose whatever representation is used by a given + * platform. + */ +typedef unsigned long irq_hw_number_t; + +/* Interrupt controller "host" data structure. This could be defined as a + * irq domain controller. That is, it handles the mapping between hardware + * and virtual interrupt numbers for a given interrupt domain. The host + * structure is generally created by the PIC code for a given PIC instance + * (though a host can cover more than one PIC if they have a flat number + * model). It's the host callbacks that are responsible for setting the + * irq_chip on a given irq_desc after it's been mapped. + * + * The host code and data structures are fairly agnostic to the fact that + * we use an open firmware device-tree. We do have references to struct + * device_node in two places: in irq_find_host() to find the host matching + * a given interrupt controller node, and of course as an argument to its + * counterpart host->ops->match() callback. However, those are treated as + * generic pointers by the core and the fact that it's actually a device-node + * pointer is purely a convention between callers and implementation. This + * code could thus be used on other architectures by replacing those two + * by some sort of arch-specific void * "token" used to identify interrupt + * controllers. + */ +struct irq_host; +struct radix_tree_root; +struct device_node; + +/* Functions below are provided by the host and called whenever a new mapping + * is created or an old mapping is disposed. The host can then proceed to + * whatever internal data structures management is required. It also needs + * to setup the irq_desc when returning from map(). + */ +struct irq_host_ops { + /* Match an interrupt controller device node to a host, returns + * 1 on a match + */ + int (*match)(struct irq_host *h, struct device_node *node); + + /* Create or update a mapping between a virtual irq number and a hw + * irq number. This is called only once for a given mapping. + */ + int (*map)(struct irq_host *h, unsigned int virq, irq_hw_number_t hw); + + /* Dispose of such a mapping */ + void (*unmap)(struct irq_host *h, unsigned int virq); + + /* Translate device-tree interrupt specifier from raw format coming + * from the firmware to a irq_hw_number_t (interrupt line number) and + * type (sense) that can be passed to set_irq_type(). In the absence + * of this callback, irq_create_of_mapping() and irq_of_parse_and_map() + * will return the hw number in the first cell and IRQ_TYPE_NONE for + * the type (which amount to keeping whatever default value the + * interrupt controller has for that line) + */ + int (*xlate)(struct irq_host *h, struct device_node *ctrler, + const u32 *intspec, unsigned int intsize, + irq_hw_number_t *out_hwirq, unsigned int *out_type); +}; + +struct irq_host { + struct list_head link; + + /* type of reverse mapping technique */ + unsigned int revmap_type; +#define IRQ_HOST_MAP_PRIORITY 0 /* core priority irqs, get irqs 1..15 */ +#define IRQ_HOST_MAP_NOMAP 1 /* no fast reverse mapping */ +#define IRQ_HOST_MAP_LINEAR 2 /* linear map of interrupts */ +#define IRQ_HOST_MAP_TREE 3 /* radix tree */ + union { + struct { + unsigned int size; + unsigned int *revmap; + } linear; + struct radix_tree_root tree; + } revmap_data; + struct irq_host_ops *ops; + void *host_data; + irq_hw_number_t inval_irq; + + /* Optional device node pointer */ + struct device_node *of_node; +}; + +struct irq_data; +extern irq_hw_number_t irqd_to_hwirq(struct irq_data *d); +extern irq_hw_number_t virq_to_hw(unsigned int virq); +extern bool virq_is_host(unsigned int virq, struct irq_host *host); + +/** + * irq_alloc_host - Allocate a new irq_host data structure + * @of_node: optional device-tree node of the interrupt controller + * @revmap_type: type of reverse mapping to use + * @revmap_arg: for IRQ_HOST_MAP_LINEAR linear only: size of the map + * @ops: map/unmap host callbacks + * @inval_irq: provide a hw number in that host space that is always invalid + * + * Allocates and initialize and irq_host structure. Note that in the case of + * IRQ_HOST_MAP_LEGACY, the map() callback will be called before this returns + * for all legacy interrupts except 0 (which is always the invalid irq for + * a legacy controller). For a IRQ_HOST_MAP_LINEAR, the map is allocated by + * this call as well. For a IRQ_HOST_MAP_TREE, the radix tree will be allocated + * later during boot automatically (the reverse mapping will use the slow path + * until that happens). + */ +extern struct irq_host *irq_alloc_host(struct device_node *of_node, + unsigned int revmap_type, + unsigned int revmap_arg, + struct irq_host_ops *ops, + irq_hw_number_t inval_irq); + + +/** + * irq_find_host - Locates a host for a given device node + * @node: device-tree node of the interrupt controller + */ +extern struct irq_host *irq_find_host(struct device_node *node); + + +/** + * irq_set_default_host - Set a "default" host + * @host: default host pointer + * + * For convenience, it's possible to set a "default" host that will be used + * whenever NULL is passed to irq_create_mapping(). It makes life easier for + * platforms that want to manipulate a few hard coded interrupt numbers that + * aren't properly represented in the device-tree. + */ +extern void irq_set_default_host(struct irq_host *host); + + +/** + * irq_set_virq_count - Set the maximum number of virt irqs + * @count: number of linux virtual irqs, capped with NR_IRQS + * + * This is mainly for use by platforms like iSeries who want to program + * the virtual irq number in the controller to avoid the reverse mapping + */ +extern void irq_set_virq_count(unsigned int count); + + +/** + * irq_create_mapping - Map a hardware interrupt into linux virq space + * @host: host owning this hardware interrupt or NULL for default host + * @hwirq: hardware irq number in that host space + * + * Only one mapping per hardware interrupt is permitted. Returns a linux + * virq number. + * If the sense/trigger is to be specified, set_irq_type() should be called + * on the number returned from that call. + */ +extern unsigned int irq_create_mapping(struct irq_host *host, + irq_hw_number_t hwirq); + + +/** + * irq_dispose_mapping - Unmap an interrupt + * @virq: linux virq number of the interrupt to unmap + */ +extern void irq_dispose_mapping(unsigned int virq); + +/** + * irq_find_mapping - Find a linux virq from an hw irq number. + * @host: host owning this hardware interrupt + * @hwirq: hardware irq number in that host space + * + * This is a slow path, for use by generic code. It's expected that an + * irq controller implementation directly calls the appropriate low level + * mapping function. + */ +extern unsigned int irq_find_mapping(struct irq_host *host, + irq_hw_number_t hwirq); + +/** + * irq_create_direct_mapping - Allocate a virq for direct mapping + * @host: host to allocate the virq for or NULL for default host + * + * This routine is used for irq controllers which can choose the hardware + * interrupt numbers they generate. In such a case it's simplest to use + * the linux virq as the hardware interrupt number. + */ +extern unsigned int irq_create_direct_mapping(struct irq_host *host); + +/** + * irq_radix_revmap_insert - Insert a hw irq to linux virq number mapping. + * @host: host owning this hardware interrupt + * @virq: linux irq number + * @hwirq: hardware irq number in that host space + * + * This is for use by irq controllers that use a radix tree reverse + * mapping for fast lookup. + */ +extern void irq_radix_revmap_insert(struct irq_host *host, unsigned int virq, + irq_hw_number_t hwirq); + +/** + * irq_radix_revmap_lookup - Find a linux virq from a hw irq number. + * @host: host owning this hardware interrupt + * @hwirq: hardware irq number in that host space + * + * This is a fast path, for use by irq controller code that uses radix tree + * revmaps + */ +extern unsigned int irq_radix_revmap_lookup(struct irq_host *host, + irq_hw_number_t hwirq); + +/** + * irq_linear_revmap - Find a linux virq from a hw irq number. + * @host: host owning this hardware interrupt + * @hwirq: hardware irq number in that host space + * + * This is a fast path, for use by irq controller code that uses linear + * revmaps. It does fallback to the slow path if the revmap doesn't exist + * yet and will create the revmap entry with appropriate locking + */ + +extern unsigned int irq_linear_revmap(struct irq_host *host, + irq_hw_number_t hwirq); + + + +/** + * irq_alloc_virt - Allocate virtual irq numbers + * @host: host owning these new virtual irqs + * @count: number of consecutive numbers to allocate + * @hint: pass a hint number, the allocator will try to use a 1:1 mapping + * + * This is a low level function that is used internally by irq_create_mapping() + * and that can be used by some irq controllers implementations for things + * like allocating ranges of numbers for MSIs. The revmaps are left untouched. + */ +extern unsigned int irq_alloc_virt(struct irq_host *host, + unsigned int count, + unsigned int hint); + +/** + * irq_free_virt - Free virtual irq numbers + * @virq: virtual irq number of the first interrupt to free + * @count: number of interrupts to free + * + * This function is the opposite of irq_alloc_virt. It will not clear reverse + * maps, this should be done previously by unmap'ing the interrupt. In fact, + * all interrupts covered by the range being freed should have been unmapped + * prior to calling this. + */ +extern void irq_free_virt(unsigned int virq, unsigned int count); + +extern void __init init_pic_c64xplus(void); + +extern void init_IRQ(void); + +struct pt_regs; + +extern asmlinkage void c6x_do_IRQ(unsigned int prio, struct pt_regs *regs); + +extern unsigned long irq_err_count; + +#endif /* _ASM_C6X_IRQ_H */ diff --git a/arch/c6x/include/asm/irqflags.h b/arch/c6x/include/asm/irqflags.h new file mode 100644 index 0000000..cf78e09 --- /dev/null +++ b/arch/c6x/include/asm/irqflags.h @@ -0,0 +1,72 @@ +/* + * C6X IRQ flag handling + * + * Copyright (C) 2010 Texas Instruments Incorporated + * Written by Mark Salter (msalter@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#ifndef _ASM_IRQFLAGS_H +#define _ASM_IRQFLAGS_H + +#ifndef __ASSEMBLY__ + +/* read interrupt enabled status */ +static inline unsigned long arch_local_save_flags(void) +{ + unsigned long flags; + + asm volatile (" mvc .s2 CSR,%0\n" : "=b"(flags)); + return flags; +} + +/* set interrupt enabled status */ +static inline void arch_local_irq_restore(unsigned long flags) +{ + asm volatile (" mvc .s2 %0,CSR\n" : : "b"(flags)); +} + +/* unconditionally enable interrupts */ +static inline void arch_local_irq_enable(void) +{ + unsigned long flags = arch_local_save_flags(); + flags |= 1; + arch_local_irq_restore(flags); +} + +/* unconditionally disable interrupts */ +static inline void arch_local_irq_disable(void) +{ + unsigned long flags = arch_local_save_flags(); + flags &= ~1; + arch_local_irq_restore(flags); +} + +/* get status and disable interrupts */ +static inline unsigned long arch_local_irq_save(void) +{ + unsigned long flags; + + flags = arch_local_save_flags(); + arch_local_irq_restore(flags & ~1); + return flags; +} + +/* test flags */ +static inline int arch_irqs_disabled_flags(unsigned long flags) +{ + return (flags & 1) == 0; +} + +/* test hardware interrupt enable bit */ +static inline int arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +#endif /* __ASSEMBLY__ */ +#endif /* __ASM_IRQFLAGS_H */ diff --git a/arch/c6x/include/asm/linkage.h b/arch/c6x/include/asm/linkage.h new file mode 100644 index 0000000..376925c --- /dev/null +++ b/arch/c6x/include/asm/linkage.h @@ -0,0 +1,30 @@ +#ifndef _ASM_C6X_LINKAGE_H +#define _ASM_C6X_LINKAGE_H + +#ifdef __ASSEMBLER__ + +#define __ALIGN .align 2 +#define __ALIGN_STR ".align 2" + +#ifndef __DSBT__ +#define ENTRY(name) \ + .global name @ \ + __ALIGN @ \ +name: +#else +#define ENTRY(name) \ + .global name @ \ + .hidden name @ \ + __ALIGN @ \ +name: +#endif + +#define ENDPROC(name) \ + .type name, @function @ \ + .size name, . - name + +#endif + +#include <asm-generic/linkage.h> + +#endif /* _ASM_C6X_LINKAGE_H */ diff --git a/arch/c6x/include/asm/megamod-pic.h b/arch/c6x/include/asm/megamod-pic.h new file mode 100644 index 0000000..eca0a86 --- /dev/null +++ b/arch/c6x/include/asm/megamod-pic.h @@ -0,0 +1,9 @@ +#ifndef _C6X_MEGAMOD_PIC_H +#define _C6X_MEGAMOD_PIC_H + +#ifdef __KERNEL__ + +extern void __init megamod_pic_init(void); + +#endif /* __KERNEL__ */ +#endif /* _C6X_MEGAMOD_PIC_H */ diff --git a/arch/c6x/include/asm/mmu.h b/arch/c6x/include/asm/mmu.h new file mode 100644 index 0000000..41592bf --- /dev/null +++ b/arch/c6x/include/asm/mmu.h @@ -0,0 +1,18 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_MMU_H +#define _ASM_C6X_MMU_H + +typedef struct { + unsigned long end_brk; +} mm_context_t; + +#endif /* _ASM_C6X_MMU_H */ diff --git a/arch/c6x/include/asm/module.h b/arch/c6x/include/asm/module.h new file mode 100644 index 0000000..a453f97 --- /dev/null +++ b/arch/c6x/include/asm/module.h @@ -0,0 +1,33 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * Updated for 2.6.34 by: Mark Salter (msalter@redhat.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_MODULE_H +#define _ASM_C6X_MODULE_H + +#define Elf_Shdr Elf32_Shdr +#define Elf_Sym Elf32_Sym +#define Elf_Ehdr Elf32_Ehdr +#define Elf_Addr Elf32_Addr +#define Elf_Word Elf32_Word + +/* + * This file contains the C6x architecture specific module code. + */ +struct mod_arch_specific { +}; + +struct loaded_sections { + unsigned int new_vaddr; + unsigned int loaded; +}; + +#endif /* _ASM_C6X_MODULE_H */ diff --git a/arch/c6x/include/asm/mutex.h b/arch/c6x/include/asm/mutex.h new file mode 100644 index 0000000..7a7248e --- /dev/null +++ b/arch/c6x/include/asm/mutex.h @@ -0,0 +1,6 @@ +#ifndef _ASM_C6X_MUTEX_H +#define _ASM_C6X_MUTEX_H + +#include <asm-generic/mutex-null.h> + +#endif /* _ASM_C6X_MUTEX_H */ diff --git a/arch/c6x/include/asm/page.h b/arch/c6x/include/asm/page.h new file mode 100644 index 0000000..d18e2b0 --- /dev/null +++ b/arch/c6x/include/asm/page.h @@ -0,0 +1,11 @@ +#ifndef _ASM_C6X_PAGE_H +#define _ASM_C6X_PAGE_H + +#define VM_DATA_DEFAULT_FLAGS \ + (VM_READ | VM_WRITE | \ + ((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0) | \ + VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) + +#include <asm-generic/page.h> + +#endif /* _ASM_C6X_PAGE_H */ diff --git a/arch/c6x/include/asm/pgtable.h b/arch/c6x/include/asm/pgtable.h new file mode 100644 index 0000000..68c8af4 --- /dev/null +++ b/arch/c6x/include/asm/pgtable.h @@ -0,0 +1,81 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_PGTABLE_H +#define _ASM_C6X_PGTABLE_H + +#include <asm-generic/4level-fixup.h> + +#include <asm/setup.h> +#include <asm/page.h> + +/* + * All 32bit addresses are effectively valid for vmalloc... + * Sort of meaningless for non-VM targets. + */ +#define VMALLOC_START 0 +#define VMALLOC_END 0xffffffff + +#define pgd_present(pgd) (1) +#define pgd_none(pgd) (0) +#define pgd_bad(pgd) (0) +#define pgd_clear(pgdp) +#define kern_addr_valid(addr) (1) + +#define pmd_offset(a, b) ((void *)0) +#define pmd_none(x) (!pmd_val(x)) +#define pmd_present(x) (pmd_val(x)) +#define pmd_clear(xp) do { set_pmd(xp, __pmd(0)); } while (0) +#define pmd_bad(x) (pmd_val(x) & ~PAGE_MASK) + +#define PAGE_NONE __pgprot(0) /* these mean nothing to NO_MM */ +#define PAGE_SHARED __pgprot(0) /* these mean nothing to NO_MM */ +#define PAGE_COPY __pgprot(0) /* these mean nothing to NO_MM */ +#define PAGE_READONLY __pgprot(0) /* these mean nothing to NO_MM */ +#define PAGE_KERNEL __pgprot(0) /* these mean nothing to NO_MM */ +#define pgprot_noncached(prot) (prot) + +extern void paging_init(void); + +#define __swp_type(x) (0) +#define __swp_offset(x) (0) +#define __swp_entry(typ, off) ((swp_entry_t) { ((typ) | ((off) << 7)) }) +#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) +#define __swp_entry_to_pte(x) ((pte_t) { (x).val }) + +static inline int pte_file(pte_t pte) +{ + return 0; +} + +#define set_pte(pteptr, pteval) (*(pteptr) = pteval) +#define set_pte_at(mm, addr, ptep, pteval) set_pte(ptep, pteval) + +/* + * ZERO_PAGE is a global shared page that is always zero: used + * for zero-mapped memory areas etc.. + */ +#define ZERO_PAGE(vaddr) virt_to_page(empty_zero_page) +extern unsigned long empty_zero_page; + +#define swapper_pg_dir ((pgd_t *) 0) + +/* + * No page table caches to initialise + */ +#define pgtable_cache_init() do { } while (0) +#define io_remap_pfn_range remap_pfn_range + +#define io_remap_page_range(vma, vaddr, paddr, size, prot) \ + remap_pfn_range(vma, vaddr, (paddr) >> PAGE_SHIFT, size, prot) + +#include <asm-generic/pgtable.h> + +#endif /* _ASM_C6X_PGTABLE_H */ diff --git a/arch/c6x/include/asm/processor.h b/arch/c6x/include/asm/processor.h new file mode 100644 index 0000000..8154c4e --- /dev/null +++ b/arch/c6x/include/asm/processor.h @@ -0,0 +1,132 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * Updated for 2.6.34: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_PROCESSOR_H +#define _ASM_C6X_PROCESSOR_H + +#include <asm/ptrace.h> +#include <asm/page.h> +#include <asm/current.h> + +/* + * Default implementation of macro that returns current + * instruction pointer ("program counter"). + */ +#define current_text_addr() \ +({ \ + void *__pc; \ + asm("mvc .S2 pce1,%0\n" : "=b"(__pc)); \ + __pc; \ +}) + +/* + * User space process size. This is mostly meaningless for NOMMU + * but some C6X processors may have RAM addresses up to 0xFFFFFFFF. + * Since calls like mmap() can return an address or an error, we + * have to allow room for error returns when code does something + * like: + * + * addr = do_mmap(...) + * if ((unsigned long)addr >= TASK_SIZE) + * ... its an error code, not an address ... + * + * Here, we allow for 4096 error codes which means we really can't + * use the last 4K page on systems with RAM extending all the way + * to the end of the 32-bit address space. + */ +#define TASK_SIZE 0xFFFFF000 + +/* + * This decides where the kernel will search for a free chunk of vm + * space during mmap's. We won't be using it + */ +#define TASK_UNMAPPED_BASE 0 + +struct thread_struct { + unsigned long long b15_14; + unsigned long long a15_14; + unsigned long long b13_12; + unsigned long long a13_12; + unsigned long long b11_10; + unsigned long long a11_10; + unsigned long long ricl_icl; + unsigned long usp; /* user stack pointer */ + unsigned long pc; /* kernel pc */ + unsigned long wchan; +}; + +#define INIT_THREAD \ +{ \ + .usp = 0, \ + .wchan = 0, \ +} + +#define INIT_MMAP { \ + &init_mm, 0, 0, NULL, PAGE_SHARED, VM_READ | VM_WRITE | VM_EXEC, 1, \ + NULL, NULL } + +#define task_pt_regs(task) \ + ((struct pt_regs *)(THREAD_START_SP + task_stack_page(task)) - 1) + +#define alloc_kernel_stack() __get_free_page(GFP_KERNEL) +#define free_kernel_stack(page) free_page((page)) + + +/* Forward declaration, a strange C thing */ +struct task_struct; + +extern void start_thread(struct pt_regs *regs, unsigned int pc, + unsigned long usp); + +/* Free all resources held by a thread. */ +static inline void release_thread(struct task_struct *dead_task) +{ +} + +/* Prepare to copy thread state - unlazy all lazy status */ +#define prepare_to_copy(tsk) do { } while (0) + +extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags); + +#define copy_segments(tsk, mm) do { } while (0) +#define release_segments(mm) do { } while (0) + +/* + * saved PC of a blocked thread. + */ +#define thread_saved_pc(tsk) (task_pt_regs(tsk)->pc) + +/* + * saved kernel SP and DP of a blocked thread. + */ +#ifdef _BIG_ENDIAN +#define thread_saved_ksp(tsk) \ + (*(unsigned long *)&(tsk)->thread.b15_14) +#define thread_saved_dp(tsk) \ + (*(((unsigned long *)&(tsk)->thread.b15_14) + 1)) +#else +#define thread_saved_ksp(tsk) \ + (*(((unsigned long *)&(tsk)->thread.b15_14) + 1)) +#define thread_saved_dp(tsk) \ + (*(unsigned long *)&(tsk)->thread.b15_14) +#endif + +extern unsigned long get_wchan(struct task_struct *p); + +#define KSTK_EIP(tsk) (task_pt_regs(task)->pc) +#define KSTK_ESP(tsk) (task_pt_regs(task)->sp) + +#define cpu_relax() do { } while (0) + +extern const struct seq_operations cpuinfo_op; + +#endif /* ASM_C6X_PROCESSOR_H */ diff --git a/arch/c6x/include/asm/procinfo.h b/arch/c6x/include/asm/procinfo.h new file mode 100644 index 0000000..c139d1e --- /dev/null +++ b/arch/c6x/include/asm/procinfo.h @@ -0,0 +1,28 @@ +/* + * Copyright (C) 2010 Texas Instruments Incorporated + * Author: Mark Salter (msalter@redhat.com) + * + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_PROCINFO_H +#define _ASM_C6X_PROCINFO_H + +#ifdef __KERNEL__ + +struct proc_info_list { + unsigned int cpu_val; + unsigned int cpu_mask; + const char *arch_name; + const char *elf_name; + unsigned int elf_hwcap; +}; + +#else /* __KERNEL__ */ +#include <asm/elf.h> +#warning "Please include asm/elf.h instead" +#endif /* __KERNEL__ */ + +#endif /* _ASM_C6X_PROCINFO_H */ diff --git a/arch/c6x/include/asm/prom.h b/arch/c6x/include/asm/prom.h new file mode 100644 index 0000000..b4ec95f --- /dev/null +++ b/arch/c6x/include/asm/prom.h @@ -0,0 +1 @@ +/* dummy prom.h; here to make linux/of.h's #includes happy */ diff --git a/arch/c6x/include/asm/ptrace.h b/arch/c6x/include/asm/ptrace.h new file mode 100644 index 0000000..21e8d79 --- /dev/null +++ b/arch/c6x/include/asm/ptrace.h @@ -0,0 +1,174 @@ +/* + * Copyright (C) 2004, 2006, 2009, 2010 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * Updated for 2.6.34: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_PTRACE_H +#define _ASM_C6X_PTRACE_H + +#define BKPT_OPCODE 0x56454314 /* illegal opcode */ + +#ifdef _BIG_ENDIAN +#define PT_LO(odd, even) odd +#define PT_HI(odd, even) even +#else +#define PT_LO(odd, even) even +#define PT_HI(odd, even) odd +#endif + +#define PT_A4_ORG PT_LO(1, 0) +#define PT_TSR PT_HI(1, 0) +#define PT_ILC PT_LO(3, 2) +#define PT_RILC PT_HI(3, 2) +#define PT_CSR PT_LO(5, 4) +#define PT_PC PT_HI(5, 4) +#define PT_B16 PT_LO(7, 6) +#define PT_B17 PT_HI(7, 6) +#define PT_B18 PT_LO(9, 8) +#define PT_B19 PT_HI(9, 8) +#define PT_B20 PT_LO(11, 10) +#define PT_B21 PT_HI(11, 10) +#define PT_B22 PT_LO(13, 12) +#define PT_B23 PT_HI(13, 12) +#define PT_B24 PT_LO(15, 14) +#define PT_B25 PT_HI(15, 14) +#define PT_B26 PT_LO(17, 16) +#define PT_B27 PT_HI(17, 16) +#define PT_B28 PT_LO(19, 18) +#define PT_B29 PT_HI(19, 18) +#define PT_B30 PT_LO(21, 20) +#define PT_B31 PT_HI(21, 20) +#define PT_B0 PT_LO(23, 22) +#define PT_B1 PT_HI(23, 22) +#define PT_B2 PT_LO(25, 24) +#define PT_B3 PT_HI(25, 24) +#define PT_B4 PT_LO(27, 26) +#define PT_B5 PT_HI(27, 26) +#define PT_B6 PT_LO(29, 28) +#define PT_B7 PT_HI(29, 28) +#define PT_B8 PT_LO(31, 30) +#define PT_B9 PT_HI(31, 30) +#define PT_B10 PT_LO(33, 32) +#define PT_B11 PT_HI(33, 32) +#define PT_B12 PT_LO(35, 34) +#define PT_B13 PT_HI(35, 34) +#define PT_A16 PT_LO(37, 36) +#define PT_A17 PT_HI(37, 36) +#define PT_A18 PT_LO(39, 38) +#define PT_A19 PT_HI(39, 38) +#define PT_A20 PT_LO(41, 40) +#define PT_A21 PT_HI(41, 40) +#define PT_A22 PT_LO(43, 42) +#define PT_A23 PT_HI(43, 42) +#define PT_A24 PT_LO(45, 44) +#define PT_A25 PT_HI(45, 44) +#define PT_A26 PT_LO(47, 46) +#define PT_A27 PT_HI(47, 46) +#define PT_A28 PT_LO(49, 48) +#define PT_A29 PT_HI(49, 48) +#define PT_A30 PT_LO(51, 50) +#define PT_A31 PT_HI(51, 50) +#define PT_A0 PT_LO(53, 52) +#define PT_A1 PT_HI(53, 52) +#define PT_A2 PT_LO(55, 54) +#define PT_A3 PT_HI(55, 54) +#define PT_A4 PT_LO(57, 56) +#define PT_A5 PT_HI(57, 56) +#define PT_A6 PT_LO(59, 58) +#define PT_A7 PT_HI(59, 58) +#define PT_A8 PT_LO(61, 60) +#define PT_A9 PT_HI(61, 60) +#define PT_A10 PT_LO(63, 62) +#define PT_A11 PT_HI(63, 62) +#define PT_A12 PT_LO(65, 64) +#define PT_A13 PT_HI(65, 64) +#define PT_A14 PT_LO(67, 66) +#define PT_A15 PT_HI(67, 66) +#define PT_B14 PT_LO(69, 68) +#define PT_B15 PT_HI(69, 68) + +#define NR_PTREGS 70 + +#define PT_DP PT_B14 /* Data Segment Pointer (B14) */ +#define PT_SP PT_B15 /* Stack Pointer (B15) */ + +#ifndef __ASSEMBLY__ + +#ifdef _BIG_ENDIAN +#define REG_PAIR(odd, even) unsigned long odd; unsigned long even +#else +#define REG_PAIR(odd, even) unsigned long even; unsigned long odd +#endif + +/* + * this struct defines the way the registers are stored on the + * stack during a system call. fields defined with REG_PAIR + * are saved and restored using double-word memory operations + * which means the word ordering of the pair depends on endianess. + */ +struct pt_regs { + REG_PAIR(tsr, orig_a4); + REG_PAIR(rilc, ilc); + REG_PAIR(pc, csr); + + REG_PAIR(b17, b16); + REG_PAIR(b19, b18); + REG_PAIR(b21, b20); + REG_PAIR(b23, b22); + REG_PAIR(b25, b24); + REG_PAIR(b27, b26); + REG_PAIR(b29, b28); + REG_PAIR(b31, b30); + + REG_PAIR(b1, b0); + REG_PAIR(b3, b2); + REG_PAIR(b5, b4); + REG_PAIR(b7, b6); + REG_PAIR(b9, b8); + REG_PAIR(b11, b10); + REG_PAIR(b13, b12); + + REG_PAIR(a17, a16); + REG_PAIR(a19, a18); + REG_PAIR(a21, a20); + REG_PAIR(a23, a22); + REG_PAIR(a25, a24); + REG_PAIR(a27, a26); + REG_PAIR(a29, a28); + REG_PAIR(a31, a30); + + REG_PAIR(a1, a0); + REG_PAIR(a3, a2); + REG_PAIR(a5, a4); + REG_PAIR(a7, a6); + REG_PAIR(a9, a8); + REG_PAIR(a11, a10); + REG_PAIR(a13, a12); + + REG_PAIR(a15, a14); + REG_PAIR(sp, dp); +}; + +#ifdef __KERNEL__ + +#include <linux/linkage.h> + +#define user_mode(regs) ((((regs)->tsr) & 0x40) != 0) + +#define instruction_pointer(regs) ((regs)->pc) +#define profile_pc(regs) instruction_pointer(regs) +#define user_stack_pointer(regs) ((regs)->sp) + +extern void show_regs(struct pt_regs *); + +extern asmlinkage unsigned long syscall_trace_entry(struct pt_regs *regs); +extern asmlinkage void syscall_trace_exit(struct pt_regs *regs); + +#endif /* __KERNEL__ */ +#endif /* __ASSEMBLY__ */ +#endif /* _ASM_C6X_PTRACE_H */ diff --git a/arch/c6x/include/asm/sections.h b/arch/c6x/include/asm/sections.h new file mode 100644 index 0000000..f703989 --- /dev/null +++ b/arch/c6x/include/asm/sections.h @@ -0,0 +1,12 @@ +#ifndef _ASM_C6X_SECTIONS_H +#define _ASM_C6X_SECTIONS_H + +#include <asm-generic/sections.h> + +extern char _vectors_start[]; +extern char _vectors_end[]; + +extern char _data_lma[]; +extern char _fdt_start[], _fdt_end[]; + +#endif /* _ASM_C6X_SECTIONS_H */ diff --git a/arch/c6x/include/asm/setup.h b/arch/c6x/include/asm/setup.h new file mode 100644 index 0000000..1808f27 --- /dev/null +++ b/arch/c6x/include/asm/setup.h @@ -0,0 +1,32 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_SETUP_H +#define _ASM_C6X_SETUP_H + +#define COMMAND_LINE_SIZE 1024 + +#ifndef __ASSEMBLY__ +extern char c6x_command_line[COMMAND_LINE_SIZE]; + +extern int c6x_add_memory(phys_addr_t start, unsigned long size); + +extern unsigned long ram_start; +extern unsigned long ram_end; + +extern int c6x_num_cores; +extern unsigned int c6x_silicon_rev; +extern unsigned int c6x_devstat; +extern unsigned char c6x_fuse_mac[6]; + +extern void machine_init(unsigned long dt_ptr); + +#endif /* !__ASSEMBLY__ */ +#endif /* _ASM_C6X_SETUP_H */ diff --git a/arch/c6x/include/asm/sigcontext.h b/arch/c6x/include/asm/sigcontext.h new file mode 100644 index 0000000..eb702f3 --- /dev/null +++ b/arch/c6x/include/asm/sigcontext.h @@ -0,0 +1,80 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_SIGCONTEXT_H +#define _ASM_C6X_SIGCONTEXT_H + + +struct sigcontext { + unsigned long sc_mask; /* old sigmask */ + unsigned long sc_sp; /* old user stack pointer */ + + unsigned long sc_a4; + unsigned long sc_b4; + unsigned long sc_a6; + unsigned long sc_b6; + unsigned long sc_a8; + unsigned long sc_b8; + + unsigned long sc_a0; + unsigned long sc_a1; + unsigned long sc_a2; + unsigned long sc_a3; + unsigned long sc_a5; + unsigned long sc_a7; + unsigned long sc_a9; + + unsigned long sc_b0; + unsigned long sc_b1; + unsigned long sc_b2; + unsigned long sc_b3; + unsigned long sc_b5; + unsigned long sc_b7; + unsigned long sc_b9; + + unsigned long sc_a16; + unsigned long sc_a17; + unsigned long sc_a18; + unsigned long sc_a19; + unsigned long sc_a20; + unsigned long sc_a21; + unsigned long sc_a22; + unsigned long sc_a23; + unsigned long sc_a24; + unsigned long sc_a25; + unsigned long sc_a26; + unsigned long sc_a27; + unsigned long sc_a28; + unsigned long sc_a29; + unsigned long sc_a30; + unsigned long sc_a31; + + unsigned long sc_b16; + unsigned long sc_b17; + unsigned long sc_b18; + unsigned long sc_b19; + unsigned long sc_b20; + unsigned long sc_b21; + unsigned long sc_b22; + unsigned long sc_b23; + unsigned long sc_b24; + unsigned long sc_b25; + unsigned long sc_b26; + unsigned long sc_b27; + unsigned long sc_b28; + unsigned long sc_b29; + unsigned long sc_b30; + unsigned long sc_b31; + + unsigned long sc_csr; + unsigned long sc_pc; +}; + +#endif /* _ASM_C6X_SIGCONTEXT_H */ diff --git a/arch/c6x/include/asm/signal.h b/arch/c6x/include/asm/signal.h new file mode 100644 index 0000000..f1cd870 --- /dev/null +++ b/arch/c6x/include/asm/signal.h @@ -0,0 +1,17 @@ +#ifndef _ASM_C6X_SIGNAL_H +#define _ASM_C6X_SIGNAL_H + +#include <asm-generic/signal.h> + +#ifndef __ASSEMBLY__ +#include <linux/linkage.h> + +struct pt_regs; + +extern asmlinkage int do_rt_sigreturn(struct pt_regs *regs); +extern asmlinkage void do_notify_resume(struct pt_regs *regs, + u32 thread_info_flags, + int syscall); +#endif + +#endif /* _ASM_C6X_SIGNAL_H */ diff --git a/arch/c6x/include/asm/soc.h b/arch/c6x/include/asm/soc.h new file mode 100644 index 0000000..43f5015 --- /dev/null +++ b/arch/c6x/include/asm/soc.h @@ -0,0 +1,35 @@ +/* + * Miscellaneous SoC-specific hooks. + * + * Copyright (C) 2011 Texas Instruments Incorporated + * + * Author: Mark Salter <msalter@redhat.com> + * + * This file is licensed under the terms of the GNU General Public License + * version 2. This program is licensed "as is" without any warranty of any + * kind, whether express or implied. + */ +#ifndef _ASM_C6X_SOC_H +#define _ASM_C6X_SOC_H + +struct soc_ops { + /* Return active exception event or -1 if none */ + int (*get_exception)(void); + + /* Assert an event */ + void (*assert_event)(unsigned int evt); +}; + +extern struct soc_ops soc_ops; + +extern int soc_get_exception(void); +extern void soc_assert_event(unsigned int event); +extern int soc_mac_addr(unsigned int index, u8 *addr); + +/* + * for mmio on SoC devices. regs are always same byte order as cpu. + */ +#define soc_readl(addr) __raw_readl(addr) +#define soc_writel(b, addr) __raw_writel((b), (addr)) + +#endif /* _ASM_C6X_SOC_H */ diff --git a/arch/c6x/include/asm/string.h b/arch/c6x/include/asm/string.h new file mode 100644 index 0000000..b21517c --- /dev/null +++ b/arch/c6x/include/asm/string.h @@ -0,0 +1,21 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_STRING_H +#define _ASM_C6X_STRING_H + +#include <asm/page.h> +#include <linux/linkage.h> + +asmlinkage extern void *memcpy(void *to, const void *from, size_t n); + +#define __HAVE_ARCH_MEMCPY + +#endif /* _ASM_C6X_STRING_H */ diff --git a/arch/c6x/include/asm/swab.h b/arch/c6x/include/asm/swab.h new file mode 100644 index 0000000..fd4bb05 --- /dev/null +++ b/arch/c6x/include/asm/swab.h @@ -0,0 +1,54 @@ +/* + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_SWAB_H +#define _ASM_C6X_SWAB_H + +static inline __attribute_const__ __u16 __c6x_swab16(__u16 val) +{ + asm("swap4 .l1 %0,%0\n" : "+a"(val)); + return val; +} + +static inline __attribute_const__ __u32 __c6x_swab32(__u32 val) +{ + asm("swap4 .l1 %0,%0\n" + "swap2 .l1 %0,%0\n" + : "+a"(val)); + return val; +} + +static inline __attribute_const__ __u64 __c6x_swab64(__u64 val) +{ + asm(" swap2 .s1 %p0,%P0\n" + "|| swap2 .l1 %P0,%p0\n" + " swap4 .l1 %p0,%p0\n" + " swap4 .l1 %P0,%P0\n" + : "+a"(val)); + return val; +} + +static inline __attribute_const__ __u32 __c6x_swahw32(__u32 val) +{ + asm("swap2 .l1 %0,%0\n" : "+a"(val)); + return val; +} + +static inline __attribute_const__ __u32 __c6x_swahb32(__u32 val) +{ + asm("swap4 .l1 %0,%0\n" : "+a"(val)); + return val; +} + +#define __arch_swab16 __c6x_swab16 +#define __arch_swab32 __c6x_swab32 +#define __arch_swab64 __c6x_swab64 +#define __arch_swahw32 __c6x_swahw32 +#define __arch_swahb32 __c6x_swahb32 + +#endif /* _ASM_C6X_SWAB_H */ diff --git a/arch/c6x/include/asm/syscall.h b/arch/c6x/include/asm/syscall.h new file mode 100644 index 0000000..ae2be31 --- /dev/null +++ b/arch/c6x/include/asm/syscall.h @@ -0,0 +1,123 @@ +/* + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __ASM_C6X_SYSCALL_H +#define __ASM_C6X_SYSCALL_H + +#include <linux/err.h> +#include <linux/sched.h> + +static inline int syscall_get_nr(struct task_struct *task, + struct pt_regs *regs) +{ + return regs->b0; +} + +static inline void syscall_rollback(struct task_struct *task, + struct pt_regs *regs) +{ + /* do nothing */ +} + +static inline long syscall_get_error(struct task_struct *task, + struct pt_regs *regs) +{ + return IS_ERR_VALUE(regs->a4) ? regs->a4 : 0; +} + +static inline long syscall_get_return_value(struct task_struct *task, + struct pt_regs *regs) +{ + return regs->a4; +} + +static inline void syscall_set_return_value(struct task_struct *task, + struct pt_regs *regs, + int error, long val) +{ + regs->a4 = error ?: val; +} + +static inline void syscall_get_arguments(struct task_struct *task, + struct pt_regs *regs, unsigned int i, + unsigned int n, unsigned long *args) +{ + switch (i) { + case 0: + if (!n--) + break; + *args++ = regs->a4; + case 1: + if (!n--) + break; + *args++ = regs->b4; + case 2: + if (!n--) + break; + *args++ = regs->a6; + case 3: + if (!n--) + break; + *args++ = regs->b6; + case 4: + if (!n--) + break; + *args++ = regs->a8; + case 5: + if (!n--) + break; + *args++ = regs->b8; + case 6: + if (!n--) + break; + default: + BUG(); + } +} + +static inline void syscall_set_arguments(struct task_struct *task, + struct pt_regs *regs, + unsigned int i, unsigned int n, + const unsigned long *args) +{ + switch (i) { + case 0: + if (!n--) + break; + regs->a4 = *args++; + case 1: + if (!n--) + break; + regs->b4 = *args++; + case 2: + if (!n--) + break; + regs->a6 = *args++; + case 3: + if (!n--) + break; + regs->b6 = *args++; + case 4: + if (!n--) + break; + regs->a8 = *args++; + case 5: + if (!n--) + break; + regs->a9 = *args++; + case 6: + if (!n) + break; + default: + BUG(); + } +} + +#endif /* __ASM_C6X_SYSCALLS_H */ diff --git a/arch/c6x/include/asm/syscalls.h b/arch/c6x/include/asm/syscalls.h new file mode 100644 index 0000000..aed53da --- /dev/null +++ b/arch/c6x/include/asm/syscalls.h @@ -0,0 +1,55 @@ +/* + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation, version 2. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or + * NON INFRINGEMENT. See the GNU General Public License for + * more details. + */ + +#ifndef __ASM_C6X_SYSCALLS_H +#define __ASM_C6X_SYSCALLS_H + +#include <linux/compiler.h> +#include <linux/linkage.h> +#include <linux/types.h> + +/* The array of function pointers for syscalls. */ +extern void *sys_call_table[]; + +/* The following are trampolines in entry.S to handle 64-bit arguments */ +extern long sys_pread_c6x(unsigned int fd, char __user *buf, + size_t count, off_t pos_low, off_t pos_high); +extern long sys_pwrite_c6x(unsigned int fd, const char __user *buf, + size_t count, off_t pos_low, off_t pos_high); +extern long sys_truncate64_c6x(const char __user *path, + off_t length_low, off_t length_high); +extern long sys_ftruncate64_c6x(unsigned int fd, + off_t length_low, off_t length_high); +extern long sys_fadvise64_c6x(int fd, u32 offset_lo, u32 offset_hi, + u32 len, int advice); +extern long sys_fadvise64_64_c6x(int fd, u32 offset_lo, u32 offset_hi, + u32 len_lo, u32 len_hi, int advice); +extern long sys_fallocate_c6x(int fd, int mode, + u32 offset_lo, u32 offset_hi, + u32 len_lo, u32 len_hi); +extern int sys_cache_sync(unsigned long s, unsigned long e); + +struct pt_regs; + +extern asmlinkage long sys_c6x_clone(struct pt_regs *regs); +extern asmlinkage long sys_c6x_execve(const char __user *name, + const char __user *const __user *argv, + const char __user *const __user *envp, + struct pt_regs *regs); + + +#include <asm-generic/syscalls.h> + +#endif /* __ASM_C6X_SYSCALLS_H */ diff --git a/arch/c6x/include/asm/system.h b/arch/c6x/include/asm/system.h new file mode 100644 index 0000000..e076dc0 --- /dev/null +++ b/arch/c6x/include/asm/system.h @@ -0,0 +1,168 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_SYSTEM_H +#define _ASM_C6X_SYSTEM_H + +#include <linux/linkage.h> +#include <linux/irqflags.h> + +#define prepare_to_switch() do { } while (0) + +struct task_struct; +struct thread_struct; +asmlinkage void *__switch_to(struct thread_struct *prev, + struct thread_struct *next, + struct task_struct *tsk); + +#define switch_to(prev, next, last) \ + do { \ + current->thread.wchan = (u_long) __builtin_return_address(0); \ + (last) = __switch_to(&(prev)->thread, \ + &(next)->thread, (prev)); \ + mb(); \ + current->thread.wchan = 0; \ + } while (0) + +/* Reset the board */ +#define HARD_RESET_NOW() + +#define get_creg(reg) \ + ({ unsigned int __x; \ + asm volatile ("mvc .s2 " #reg ",%0\n" : "=b"(__x)); __x; }) + +#define set_creg(reg, v) \ + do { unsigned int __x = (unsigned int)(v); \ + asm volatile ("mvc .s2 %0," #reg "\n" : : "b"(__x)); \ + } while (0) + +#define or_creg(reg, n) \ + do { unsigned __x, __n = (unsigned)(n); \ + asm volatile ("mvc .s2 " #reg ",%0\n" \ + "or .l2 %1,%0,%0\n" \ + "mvc .s2 %0," #reg "\n" \ + "nop\n" \ + : "=&b"(__x) : "b"(__n)); \ + } while (0) + +#define and_creg(reg, n) \ + do { unsigned __x, __n = (unsigned)(n); \ + asm volatile ("mvc .s2 " #reg ",%0\n" \ + "and .l2 %1,%0,%0\n" \ + "mvc .s2 %0," #reg "\n" \ + "nop\n" \ + : "=&b"(__x) : "b"(__n)); \ + } while (0) + +#define get_coreid() (get_creg(DNUM) & 0xff) + +/* Set/get IST */ +#define set_ist(x) set_creg(ISTP, x) +#define get_ist() get_creg(ISTP) + +/* + * Exception management + */ +asmlinkage void enable_exception(void); +#define disable_exception() +#define get_except_type() get_creg(EFR) +#define ack_exception(type) set_creg(ECR, 1 << (type)) +#define get_iexcept() get_creg(IERR) +#define set_iexcept(mask) set_creg(IERR, (mask)) + +/* + * Misc. functions + */ +#define nop() asm("NOP\n"); +#define mb() barrier() +#define rmb() barrier() +#define wmb() barrier() +#define set_mb(var, value) do { var = value; mb(); } while (0) +#define set_wmb(var, value) do { var = value; wmb(); } while (0) + +#define smp_mb() barrier() +#define smp_rmb() barrier() +#define smp_wmb() barrier() +#define smp_read_barrier_depends() do { } while (0) + +#define xchg(ptr, x) \ + ((__typeof__(*(ptr)))__xchg((unsigned int)(x), (void *) (ptr), \ + sizeof(*(ptr)))) +#define tas(ptr) xchg((ptr), 1) + +unsigned int _lmbd(unsigned int, unsigned int); +unsigned int _bitr(unsigned int); + +struct __xchg_dummy { unsigned int a[100]; }; +#define __xg(x) ((volatile struct __xchg_dummy *)(x)) + +static inline unsigned int __xchg(unsigned int x, volatile void *ptr, int size) +{ + unsigned int tmp; + unsigned long flags; + + local_irq_save(flags); + + switch (size) { + case 1: + tmp = 0; + tmp = *((unsigned char *) ptr); + *((unsigned char *) ptr) = (unsigned char) x; + break; + case 2: + tmp = 0; + tmp = *((unsigned short *) ptr); + *((unsigned short *) ptr) = x; + break; + case 4: + tmp = 0; + tmp = *((unsigned int *) ptr); + *((unsigned int *) ptr) = x; + break; + } + local_irq_restore(flags); + return tmp; +} + +#include <asm-generic/cmpxchg-local.h> + +/* + * cmpxchg_local and cmpxchg64_local are atomic wrt current CPU. Always make + * them available. + */ +#define cmpxchg_local(ptr, o, n) \ + ((__typeof__(*(ptr)))__cmpxchg_local_generic((ptr), \ + (unsigned long)(o), \ + (unsigned long)(n), \ + sizeof(*(ptr)))) +#define cmpxchg64_local(ptr, o, n) __cmpxchg64_local_generic((ptr), (o), (n)) + +#include <asm-generic/cmpxchg.h> + +#define _extu(x, s, e) \ + ({ unsigned int __x; \ + asm volatile ("extu .S2 %3,%1,%2,%0\n" : \ + "=b"(__x) : "n"(s), "n"(e), "b"(x)); \ + __x; }) + + +extern unsigned int c6x_core_freq; + +struct pt_regs; + +extern void die(char *str, struct pt_regs *fp, int nr); +extern asmlinkage int process_exception(struct pt_regs *regs); +extern void time_init(void); +extern void free_initmem(void); + +extern void (*c6x_restart)(void); +extern void (*c6x_halt)(void); + +#endif /* _ASM_C6X_SYSTEM_H */ diff --git a/arch/c6x/include/asm/thread_info.h b/arch/c6x/include/asm/thread_info.h new file mode 100644 index 0000000..fd99148 --- /dev/null +++ b/arch/c6x/include/asm/thread_info.h @@ -0,0 +1,121 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * Updated for 2.6.3x: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_THREAD_INFO_H +#define _ASM_C6X_THREAD_INFO_H + +#ifdef __KERNEL__ + +#include <asm/page.h> + +#ifdef CONFIG_4KSTACKS +#define THREAD_SIZE 4096 +#define THREAD_SHIFT 12 +#define THREAD_ORDER 0 +#else +#define THREAD_SIZE 8192 +#define THREAD_SHIFT 13 +#define THREAD_ORDER 1 +#endif + +#define THREAD_START_SP (THREAD_SIZE - 8) + +#ifndef __ASSEMBLY__ + +typedef struct { + unsigned long seg; +} mm_segment_t; + +/* + * low level task data. + */ +struct thread_info { + struct task_struct *task; /* main task structure */ + struct exec_domain *exec_domain; /* execution domain */ + unsigned long flags; /* low level flags */ + int cpu; /* cpu we're on */ + int preempt_count; /* 0 = preemptable, <0 = BUG */ + mm_segment_t addr_limit; /* thread address space */ + struct restart_block restart_block; +}; + +/* + * macros/functions for gaining access to the thread information structure + * + * preempt_count needs to be 1 initially, until the scheduler is functional. + */ +#define INIT_THREAD_INFO(tsk) \ +{ \ + .task = &tsk, \ + .exec_domain = &default_exec_domain, \ + .flags = 0, \ + .cpu = 0, \ + .preempt_count = INIT_PREEMPT_COUNT, \ + .addr_limit = KERNEL_DS, \ + .restart_block = { \ + .fn = do_no_restart_syscall, \ + }, \ +} + +#define init_thread_info (init_thread_union.thread_info) +#define init_stack (init_thread_union.stack) + +/* get the thread information struct of current task */ +static inline __attribute__((const)) +struct thread_info *current_thread_info(void) +{ + struct thread_info *ti; + asm volatile (" clr .s2 B15,0,%1,%0\n" + : "=b" (ti) + : "Iu5" (THREAD_SHIFT - 1)); + return ti; +} + +#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR + +/* thread information allocation */ +#ifdef CONFIG_DEBUG_STACK_USAGE +#define THREAD_FLAGS (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO) +#else +#define THREAD_FLAGS (GFP_KERNEL | __GFP_NOTRACK) +#endif + +#define alloc_thread_info_node(tsk, node) \ + ((struct thread_info *)__get_free_pages(THREAD_FLAGS, THREAD_ORDER)) + +#define free_thread_info(ti) free_pages((unsigned long) (ti), THREAD_ORDER) +#define get_thread_info(ti) get_task_struct((ti)->task) +#define put_thread_info(ti) put_task_struct((ti)->task) +#endif /* __ASSEMBLY__ */ + +#define PREEMPT_ACTIVE 0x10000000 + +/* + * thread information flag bit numbers + * - pending work-to-be-done flags are in LSW + * - other flags in MSW + */ +#define TIF_SYSCALL_TRACE 0 /* syscall trace active */ +#define TIF_NOTIFY_RESUME 1 /* resumption notification requested */ +#define TIF_SIGPENDING 2 /* signal pending */ +#define TIF_NEED_RESCHED 3 /* rescheduling necessary */ +#define TIF_RESTORE_SIGMASK 4 /* restore signal mask in do_signal() */ + +#define TIF_POLLING_NRFLAG 16 /* true if polling TIF_NEED_RESCHED */ +#define TIF_MEMDIE 17 /* OOM killer killed process */ + +#define TIF_WORK_MASK 0x00007FFE /* work on irq/exception return */ +#define TIF_ALLWORK_MASK 0x00007FFF /* work on any return to u-space */ + +#endif /* __KERNEL__ */ + +#endif /* _ASM_C6X_THREAD_INFO_H */ diff --git a/arch/c6x/include/asm/timer64.h b/arch/c6x/include/asm/timer64.h new file mode 100644 index 0000000..bbe27bb --- /dev/null +++ b/arch/c6x/include/asm/timer64.h @@ -0,0 +1,6 @@ +#ifndef _C6X_TIMER64_H +#define _C6X_TIMER64_H + +extern void __init timer64_init(void); + +#endif /* _C6X_TIMER64_H */ diff --git a/arch/c6x/include/asm/timex.h b/arch/c6x/include/asm/timex.h new file mode 100644 index 0000000..508c3ec --- /dev/null +++ b/arch/c6x/include/asm/timex.h @@ -0,0 +1,33 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * Modified for 2.6.34: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_TIMEX_H +#define _ASM_C6X_TIMEX_H + +#define CLOCK_TICK_RATE ((1000 * 1000000UL) / 6) + +/* 64-bit timestamp */ +typedef unsigned long long cycles_t; + +static inline cycles_t get_cycles(void) +{ + unsigned l, h; + + asm volatile (" dint\n" + " mvc .s2 TSCL,%0\n" + " mvc .s2 TSCH,%1\n" + " rint\n" + : "=b"(l), "=b"(h)); + return ((cycles_t)h << 32) | l; +} + +#endif /* _ASM_C6X_TIMEX_H */ diff --git a/arch/c6x/include/asm/tlb.h b/arch/c6x/include/asm/tlb.h new file mode 100644 index 0000000..8709e5e --- /dev/null +++ b/arch/c6x/include/asm/tlb.h @@ -0,0 +1,8 @@ +#ifndef _ASM_C6X_TLB_H +#define _ASM_C6X_TLB_H + +#define tlb_flush(tlb) flush_tlb_mm((tlb)->mm) + +#include <asm-generic/tlb.h> + +#endif /* _ASM_C6X_TLB_H */ diff --git a/arch/c6x/include/asm/traps.h b/arch/c6x/include/asm/traps.h new file mode 100644 index 0000000..62124d7 --- /dev/null +++ b/arch/c6x/include/asm/traps.h @@ -0,0 +1,36 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_TRAPS_H +#define _ASM_C6X_TRAPS_H + +#define EXCEPT_TYPE_NXF 31 /* NMI */ +#define EXCEPT_TYPE_EXC 30 /* external exception */ +#define EXCEPT_TYPE_IXF 1 /* internal exception */ +#define EXCEPT_TYPE_SXF 0 /* software exception */ + +#define EXCEPT_CAUSE_LBX (1 << 7) /* loop buffer exception */ +#define EXCEPT_CAUSE_PRX (1 << 6) /* privilege exception */ +#define EXCEPT_CAUSE_RAX (1 << 5) /* resource access exception */ +#define EXCEPT_CAUSE_RCX (1 << 4) /* resource conflict exception */ +#define EXCEPT_CAUSE_OPX (1 << 3) /* opcode exception */ +#define EXCEPT_CAUSE_EPX (1 << 2) /* execute packet exception */ +#define EXCEPT_CAUSE_FPX (1 << 1) /* fetch packet exception */ +#define EXCEPT_CAUSE_IFX (1 << 0) /* instruction fetch exception */ + +struct exception_info { + char *kernel_str; + int signo; + int code; +}; + +extern int (*c6x_nmi_handler)(struct pt_regs *regs); + +#endif /* _ASM_C6X_TRAPS_H */ diff --git a/arch/c6x/include/asm/uaccess.h b/arch/c6x/include/asm/uaccess.h new file mode 100644 index 0000000..453dd26 --- /dev/null +++ b/arch/c6x/include/asm/uaccess.h @@ -0,0 +1,107 @@ +/* + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_UACCESS_H +#define _ASM_C6X_UACCESS_H + +#include <linux/types.h> +#include <linux/compiler.h> +#include <linux/string.h> + +#ifdef CONFIG_ACCESS_CHECK +#define __access_ok _access_ok +#endif + +/* + * __copy_from_user/copy_to_user are based on ones in asm-generic/uaccess.h + * + * C6X supports unaligned 32 and 64 bit loads and stores. + */ +static inline __must_check long __copy_from_user(void *to, + const void __user *from, unsigned long n) +{ + u32 tmp32; + u64 tmp64; + + if (__builtin_constant_p(n)) { + switch (n) { + case 1: + *(u8 *)to = *(u8 __force *)from; + return 0; + case 4: + asm volatile ("ldnw .d1t1 *%2,%0\n" + "nop 4\n" + "stnw .d1t1 %0,*%1\n" + : "=&a"(tmp32) + : "A"(to), "a"(from) + : "memory"); + return 0; + case 8: + asm volatile ("ldndw .d1t1 *%2,%0\n" + "nop 4\n" + "stndw .d1t1 %0,*%1\n" + : "=&a"(tmp64) + : "a"(to), "a"(from) + : "memory"); + return 0; + default: + break; + } + } + + memcpy(to, (const void __force *)from, n); + return 0; +} + +static inline __must_check long __copy_to_user(void __user *to, + const void *from, unsigned long n) +{ + u32 tmp32; + u64 tmp64; + + if (__builtin_constant_p(n)) { + switch (n) { + case 1: + *(u8 __force *)to = *(u8 *)from; + return 0; + case 4: + asm volatile ("ldnw .d1t1 *%2,%0\n" + "nop 4\n" + "stnw .d1t1 %0,*%1\n" + : "=&a"(tmp32) + : "a"(to), "a"(from) + : "memory"); + return 0; + case 8: + asm volatile ("ldndw .d1t1 *%2,%0\n" + "nop 4\n" + "stndw .d1t1 %0,*%1\n" + : "=&a"(tmp64) + : "a"(to), "a"(from) + : "memory"); + return 0; + default: + break; + } + } + + memcpy((void __force *)to, from, n); + return 0; +} + +#define __copy_to_user __copy_to_user +#define __copy_from_user __copy_from_user + +extern int _access_ok(unsigned long addr, unsigned long size); +#ifdef CONFIG_ACCESS_CHECK +#define __access_ok _access_ok +#endif + +#include <asm-generic/uaccess.h> + +#endif /* _ASM_C6X_UACCESS_H */ diff --git a/arch/c6x/include/asm/unaligned.h b/arch/c6x/include/asm/unaligned.h new file mode 100644 index 0000000..b976cb7 --- /dev/null +++ b/arch/c6x/include/asm/unaligned.h @@ -0,0 +1,170 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * Rewritten for 2.6.3x: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef _ASM_C6X_UNALIGNED_H +#define _ASM_C6X_UNALIGNED_H + +#include <linux/swab.h> + +/* + * The C64x+ can do unaligned word and dword accesses in hardware + * using special load/store instructions. + */ + +static inline u16 get_unaligned_le16(const void *p) +{ + const u8 *_p = p; + return _p[0] | _p[1] << 8; +} + +static inline u16 get_unaligned_be16(const void *p) +{ + const u8 *_p = p; + return _p[0] << 8 | _p[1]; +} + +static inline void put_unaligned_le16(u16 val, void *p) +{ + u8 *_p = p; + _p[0] = val; + _p[1] = val >> 8; +} + +static inline void put_unaligned_be16(u16 val, void *p) +{ + u8 *_p = p; + _p[0] = val >> 8; + _p[1] = val; +} + +static inline u32 get_unaligned32(const void *p) +{ + u32 val = (u32) p; + asm (" ldnw .d1t1 *%0,%0\n" + " nop 4\n" + : "+a"(val)); + return val; +} + +static inline void put_unaligned32(u32 val, void *p) +{ + asm volatile (" stnw .d2t1 %0,*%1\n" + : : "a"(val), "b"(p) : "memory"); +} + +static inline u64 get_unaligned64(const void *p) +{ + u64 val; + asm volatile (" ldndw .d1t1 *%1,%0\n" + " nop 4\n" + : "=a"(val) : "a"(p)); + return val; +} + +static inline void put_unaligned64(u64 val, const void *p) +{ + asm volatile (" stndw .d2t1 %0,*%1\n" + : : "a"(val), "b"(p) : "memory"); +} + +#ifdef CONFIG_CPU_BIG_ENDIAN + +#define get_unaligned_le32(p) __swab32(get_unaligned32(p)) +#define get_unaligned_le64(p) __swab64(get_unaligned64(p)) +#define get_unaligned_be32(p) get_unaligned32(p) +#define get_unaligned_be64(p) get_unaligned64(p) +#define put_unaligned_le32(v, p) put_unaligned32(__swab32(v), (p)) +#define put_unaligned_le64(v, p) put_unaligned64(__swab64(v), (p)) +#define put_unaligned_be32(v, p) put_unaligned32((v), (p)) +#define put_unaligned_be64(v, p) put_unaligned64((v), (p)) +#define get_unaligned __get_unaligned_be +#define put_unaligned __put_unaligned_be + +#else + +#define get_unaligned_le32(p) get_unaligned32(p) +#define get_unaligned_le64(p) get_unaligned64(p) +#define get_unaligned_be32(p) __swab32(get_unaligned32(p)) +#define get_unaligned_be64(p) __swab64(get_unaligned64(p)) +#define put_unaligned_le32(v, p) put_unaligned32((v), (p)) +#define put_unaligned_le64(v, p) put_unaligned64((v), (p)) +#define put_unaligned_be32(v, p) put_unaligned32(__swab32(v), (p)) +#define put_unaligned_be64(v, p) put_unaligned64(__swab64(v), (p)) +#define get_unaligned __get_unaligned_le +#define put_unaligned __put_unaligned_le + +#endif + +/* + * Cause a link-time error if we try an unaligned access other than + * 1,2,4 or 8 bytes long + */ +extern int __bad_unaligned_access_size(void); + +#define __get_unaligned_le(ptr) (typeof(*(ptr)))({ \ + sizeof(*(ptr)) == 1 ? *(ptr) : \ + (sizeof(*(ptr)) == 2 ? get_unaligned_le16((ptr)) : \ + (sizeof(*(ptr)) == 4 ? get_unaligned_le32((ptr)) : \ + (sizeof(*(ptr)) == 8 ? get_unaligned_le64((ptr)) : \ + __bad_unaligned_access_size()))); \ + }) + +#define __get_unaligned_be(ptr) (__force typeof(*(ptr)))({ \ + sizeof(*(ptr)) == 1 ? *(ptr) : \ + (sizeof(*(ptr)) == 2 ? get_unaligned_be16((ptr)) : \ + (sizeof(*(ptr)) == 4 ? get_unaligned_be32((ptr)) : \ + (sizeof(*(ptr)) == 8 ? get_unaligned_be64((ptr)) : \ + __bad_unaligned_access_size()))); \ + }) + +#define __put_unaligned_le(val, ptr) ({ \ + void *__gu_p = (ptr); \ + switch (sizeof(*(ptr))) { \ + case 1: \ + *(u8 *)__gu_p = (__force u8)(val); \ + break; \ + case 2: \ + put_unaligned_le16((__force u16)(val), __gu_p); \ + break; \ + case 4: \ + put_unaligned_le32((__force u32)(val), __gu_p); \ + break; \ + case 8: \ + put_unaligned_le64((__force u64)(val), __gu_p); \ + break; \ + default: \ + __bad_unaligned_access_size(); \ + break; \ + } \ + (void)0; }) + +#define __put_unaligned_be(val, ptr) ({ \ + void *__gu_p = (ptr); \ + switch (sizeof(*(ptr))) { \ + case 1: \ + *(u8 *)__gu_p = (__force u8)(val); \ + break; \ + case 2: \ + put_unaligned_be16((__force u16)(val), __gu_p); \ + break; \ + case 4: \ + put_unaligned_be32((__force u32)(val), __gu_p); \ + break; \ + case 8: \ + put_unaligned_be64((__force u64)(val), __gu_p); \ + break; \ + default: \ + __bad_unaligned_access_size(); \ + break; \ + } \ + (void)0; }) + +#endif /* _ASM_C6X_UNALIGNED_H */ diff --git a/arch/c6x/include/asm/unistd.h b/arch/c6x/include/asm/unistd.h new file mode 100644 index 0000000..6d54ea4 --- /dev/null +++ b/arch/c6x/include/asm/unistd.h @@ -0,0 +1,26 @@ +/* + * Copyright (C) 2011 Texas Instruments Incorporated + * + * Based on arch/tile version. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation, version 2. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or + * NON INFRINGEMENT. See the GNU General Public License for + * more details. + */ +#if !defined(_ASM_C6X_UNISTD_H) || defined(__SYSCALL) +#define _ASM_C6X_UNISTD_H + +/* Use the standard ABI for syscalls. */ +#include <asm-generic/unistd.h> + +/* C6X-specific syscalls. */ +#define __NR_cache_sync (__NR_arch_specific_syscall + 0) +__SYSCALL(__NR_cache_sync, sys_cache_sync) + +#endif /* _ASM_C6X_UNISTD_H */ diff --git a/arch/c6x/kernel/Makefile b/arch/c6x/kernel/Makefile new file mode 100644 index 0000000..580a515 --- /dev/null +++ b/arch/c6x/kernel/Makefile @@ -0,0 +1,12 @@ +# +# Makefile for arch/c6x/kernel/ +# + +extra-y := head.o vmlinux.lds + +obj-y := process.o traps.o irq.o signal.o ptrace.o +obj-y += setup.o sys_c6x.o time.o devicetree.o +obj-y += switch_to.o entry.o vectors.o c6x_ksyms.o +obj-y += soc.o dma.o + +obj-$(CONFIG_MODULES) += module.o diff --git a/arch/c6x/kernel/asm-offsets.c b/arch/c6x/kernel/asm-offsets.c new file mode 100644 index 0000000..759ad6d --- /dev/null +++ b/arch/c6x/kernel/asm-offsets.c @@ -0,0 +1,123 @@ +/* + * Generate definitions needed by assembly language modules. + * This code generates raw asm output which is post-processed + * to extract and format the required data. + */ + +#include <linux/sched.h> +#include <linux/thread_info.h> +#include <asm/procinfo.h> +#include <linux/kbuild.h> +#include <linux/unistd.h> + +void foo(void) +{ + OFFSET(REGS_A16, pt_regs, a16); + OFFSET(REGS_A17, pt_regs, a17); + OFFSET(REGS_A18, pt_regs, a18); + OFFSET(REGS_A19, pt_regs, a19); + OFFSET(REGS_A20, pt_regs, a20); + OFFSET(REGS_A21, pt_regs, a21); + OFFSET(REGS_A22, pt_regs, a22); + OFFSET(REGS_A23, pt_regs, a23); + OFFSET(REGS_A24, pt_regs, a24); + OFFSET(REGS_A25, pt_regs, a25); + OFFSET(REGS_A26, pt_regs, a26); + OFFSET(REGS_A27, pt_regs, a27); + OFFSET(REGS_A28, pt_regs, a28); + OFFSET(REGS_A29, pt_regs, a29); + OFFSET(REGS_A30, pt_regs, a30); + OFFSET(REGS_A31, pt_regs, a31); + + OFFSET(REGS_B16, pt_regs, b16); + OFFSET(REGS_B17, pt_regs, b17); + OFFSET(REGS_B18, pt_regs, b18); + OFFSET(REGS_B19, pt_regs, b19); + OFFSET(REGS_B20, pt_regs, b20); + OFFSET(REGS_B21, pt_regs, b21); + OFFSET(REGS_B22, pt_regs, b22); + OFFSET(REGS_B23, pt_regs, b23); + OFFSET(REGS_B24, pt_regs, b24); + OFFSET(REGS_B25, pt_regs, b25); + OFFSET(REGS_B26, pt_regs, b26); + OFFSET(REGS_B27, pt_regs, b27); + OFFSET(REGS_B28, pt_regs, b28); + OFFSET(REGS_B29, pt_regs, b29); + OFFSET(REGS_B30, pt_regs, b30); + OFFSET(REGS_B31, pt_regs, b31); + + OFFSET(REGS_A0, pt_regs, a0); + OFFSET(REGS_A1, pt_regs, a1); + OFFSET(REGS_A2, pt_regs, a2); + OFFSET(REGS_A3, pt_regs, a3); + OFFSET(REGS_A4, pt_regs, a4); + OFFSET(REGS_A5, pt_regs, a5); + OFFSET(REGS_A6, pt_regs, a6); + OFFSET(REGS_A7, pt_regs, a7); + OFFSET(REGS_A8, pt_regs, a8); + OFFSET(REGS_A9, pt_regs, a9); + OFFSET(REGS_A10, pt_regs, a10); + OFFSET(REGS_A11, pt_regs, a11); + OFFSET(REGS_A12, pt_regs, a12); + OFFSET(REGS_A13, pt_regs, a13); + OFFSET(REGS_A14, pt_regs, a14); + OFFSET(REGS_A15, pt_regs, a15); + + OFFSET(REGS_B0, pt_regs, b0); + OFFSET(REGS_B1, pt_regs, b1); + OFFSET(REGS_B2, pt_regs, b2); + OFFSET(REGS_B3, pt_regs, b3); + OFFSET(REGS_B4, pt_regs, b4); + OFFSET(REGS_B5, pt_regs, b5); + OFFSET(REGS_B6, pt_regs, b6); + OFFSET(REGS_B7, pt_regs, b7); + OFFSET(REGS_B8, pt_regs, b8); + OFFSET(REGS_B9, pt_regs, b9); + OFFSET(REGS_B10, pt_regs, b10); + OFFSET(REGS_B11, pt_regs, b11); + OFFSET(REGS_B12, pt_regs, b12); + OFFSET(REGS_B13, pt_regs, b13); + OFFSET(REGS_DP, pt_regs, dp); + OFFSET(REGS_SP, pt_regs, sp); + + OFFSET(REGS_TSR, pt_regs, tsr); + OFFSET(REGS_ORIG_A4, pt_regs, orig_a4); + + DEFINE(REGS__END, sizeof(struct pt_regs)); + BLANK(); + + OFFSET(THREAD_PC, thread_struct, pc); + OFFSET(THREAD_B15_14, thread_struct, b15_14); + OFFSET(THREAD_A15_14, thread_struct, a15_14); + OFFSET(THREAD_B13_12, thread_struct, b13_12); + OFFSET(THREAD_A13_12, thread_struct, a13_12); + OFFSET(THREAD_B11_10, thread_struct, b11_10); + OFFSET(THREAD_A11_10, thread_struct, a11_10); + OFFSET(THREAD_RICL_ICL, thread_struct, ricl_icl); + BLANK(); + + OFFSET(TASK_STATE, task_struct, state); + BLANK(); + + OFFSET(THREAD_INFO_FLAGS, thread_info, flags); + OFFSET(THREAD_INFO_PREEMPT_COUNT, thread_info, preempt_count); + BLANK(); + + /* These would be unneccessary if we ran asm files + * through the preprocessor. + */ + DEFINE(KTHREAD_SIZE, THREAD_SIZE); + DEFINE(KTHREAD_SHIFT, THREAD_SHIFT); + DEFINE(KTHREAD_START_SP, THREAD_START_SP); + DEFINE(ENOSYS_, ENOSYS); + DEFINE(NR_SYSCALLS_, __NR_syscalls); + + DEFINE(_TIF_SYSCALL_TRACE, (1<<TIF_SYSCALL_TRACE)); + DEFINE(_TIF_NOTIFY_RESUME, (1<<TIF_NOTIFY_RESUME)); + DEFINE(_TIF_SIGPENDING, (1<<TIF_SIGPENDING)); + DEFINE(_TIF_NEED_RESCHED, (1<<TIF_NEED_RESCHED)); + DEFINE(_TIF_POLLING_NRFLAG, (1<<TIF_POLLING_NRFLAG)); + + DEFINE(_TIF_ALLWORK_MASK, TIF_ALLWORK_MASK); + DEFINE(_TIF_WORK_MASK, TIF_WORK_MASK); +} diff --git a/arch/c6x/kernel/c6x_ksyms.c b/arch/c6x/kernel/c6x_ksyms.c new file mode 100644 index 0000000..0ba3e0b --- /dev/null +++ b/arch/c6x/kernel/c6x_ksyms.c @@ -0,0 +1,66 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ +#include <linux/module.h> +#include <asm/checksum.h> +#include <linux/io.h> + +/* + * libgcc functions - used internally by the compiler... + */ +extern int __c6xabi_divi(int dividend, int divisor); +EXPORT_SYMBOL(__c6xabi_divi); + +extern unsigned __c6xabi_divu(unsigned dividend, unsigned divisor); +EXPORT_SYMBOL(__c6xabi_divu); + +extern int __c6xabi_remi(int dividend, int divisor); +EXPORT_SYMBOL(__c6xabi_remi); + +extern unsigned __c6xabi_remu(unsigned dividend, unsigned divisor); +EXPORT_SYMBOL(__c6xabi_remu); + +extern int __c6xabi_divremi(int dividend, int divisor); +EXPORT_SYMBOL(__c6xabi_divremi); + +extern unsigned __c6xabi_divremu(unsigned dividend, unsigned divisor); +EXPORT_SYMBOL(__c6xabi_divremu); + +extern unsigned long long __c6xabi_mpyll(unsigned long long src1, + unsigned long long src2); +EXPORT_SYMBOL(__c6xabi_mpyll); + +extern long long __c6xabi_negll(long long src); +EXPORT_SYMBOL(__c6xabi_negll); + +extern unsigned long long __c6xabi_llshl(unsigned long long src1, uint src2); +EXPORT_SYMBOL(__c6xabi_llshl); + +extern long long __c6xabi_llshr(long long src1, uint src2); +EXPORT_SYMBOL(__c6xabi_llshr); + +extern unsigned long long __c6xabi_llshru(unsigned long long src1, uint src2); +EXPORT_SYMBOL(__c6xabi_llshru); + +extern void __c6xabi_strasgi(int *dst, const int *src, unsigned cnt); +EXPORT_SYMBOL(__c6xabi_strasgi); + +extern void __c6xabi_push_rts(void); +EXPORT_SYMBOL(__c6xabi_push_rts); + +extern void __c6xabi_pop_rts(void); +EXPORT_SYMBOL(__c6xabi_pop_rts); + +extern void __c6xabi_strasgi_64plus(int *dst, const int *src, unsigned cnt); +EXPORT_SYMBOL(__c6xabi_strasgi_64plus); + +/* lib functions */ +EXPORT_SYMBOL(memcpy); diff --git a/arch/c6x/kernel/devicetree.c b/arch/c6x/kernel/devicetree.c new file mode 100644 index 0000000..bdb56f0 --- /dev/null +++ b/arch/c6x/kernel/devicetree.c @@ -0,0 +1,53 @@ +/* + * Architecture specific OF callbacks. + * + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ +#include <linux/init.h> +#include <linux/of.h> +#include <linux/of_fdt.h> +#include <linux/initrd.h> +#include <linux/memblock.h> + +void __init early_init_devtree(void *params) +{ + /* Setup flat device-tree pointer */ + initial_boot_params = params; + + /* Retrieve various informations from the /chosen node of the + * device-tree, including the platform type, initrd location and + * size and more ... + */ + of_scan_flat_dt(early_init_dt_scan_chosen, c6x_command_line); + + /* Scan memory nodes and rebuild MEMBLOCKs */ + of_scan_flat_dt(early_init_dt_scan_root, NULL); + of_scan_flat_dt(early_init_dt_scan_memory, NULL); +} + + +#ifdef CONFIG_BLK_DEV_INITRD +void __init early_init_dt_setup_initrd_arch(unsigned long start, + unsigned long end) +{ + initrd_start = (unsigned long)__va(start); + initrd_end = (unsigned long)__va(end); + initrd_below_start_ok = 1; +} +#endif + +void __init early_init_dt_add_memory_arch(u64 base, u64 size) +{ + c6x_add_memory(base, size); +} + +void * __init early_init_dt_alloc_memory_arch(u64 size, u64 align) +{ + return __va(memblock_alloc(size, align)); +} diff --git a/arch/c6x/kernel/dma.c b/arch/c6x/kernel/dma.c new file mode 100644 index 0000000..ab7b12d --- /dev/null +++ b/arch/c6x/kernel/dma.c @@ -0,0 +1,153 @@ +/* + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/module.h> +#include <linux/dma-mapping.h> +#include <linux/mm.h> +#include <linux/mm_types.h> +#include <linux/scatterlist.h> + +#include <asm/cacheflush.h> + +static void c6x_dma_sync(dma_addr_t handle, size_t size, + enum dma_data_direction dir) +{ + unsigned long paddr = handle; + + BUG_ON(!valid_dma_direction(dir)); + + switch (dir) { + case DMA_FROM_DEVICE: + L2_cache_block_invalidate(paddr, paddr + size); + break; + case DMA_TO_DEVICE: + L2_cache_block_writeback(paddr, paddr + size); + break; + case DMA_BIDIRECTIONAL: + L2_cache_block_writeback_invalidate(paddr, paddr + size); + break; + default: + break; + } +} + +dma_addr_t dma_map_single(struct device *dev, void *ptr, size_t size, + enum dma_data_direction dir) +{ + dma_addr_t addr = virt_to_phys(ptr); + + c6x_dma_sync(addr, size, dir); + + debug_dma_map_page(dev, virt_to_page(ptr), + (unsigned long)ptr & ~PAGE_MASK, size, + dir, addr, true); + return addr; +} +EXPORT_SYMBOL(dma_map_single); + + +void dma_unmap_single(struct device *dev, dma_addr_t handle, + size_t size, enum dma_data_direction dir) +{ + c6x_dma_sync(handle, size, dir); + + debug_dma_unmap_page(dev, handle, size, dir, true); +} +EXPORT_SYMBOL(dma_unmap_single); + + +int dma_map_sg(struct device *dev, struct scatterlist *sglist, + int nents, enum dma_data_direction dir) +{ + struct scatterlist *sg; + int i; + + for_each_sg(sglist, sg, nents, i) + sg->dma_address = dma_map_single(dev, sg_virt(sg), sg->length, + dir); + + debug_dma_map_sg(dev, sglist, nents, nents, dir); + + return nents; +} +EXPORT_SYMBOL(dma_map_sg); + + +void dma_unmap_sg(struct device *dev, struct scatterlist *sglist, + int nents, enum dma_data_direction dir) +{ + struct scatterlist *sg; + int i; + + for_each_sg(sglist, sg, nents, i) + dma_unmap_single(dev, sg_dma_address(sg), sg->length, dir); + + debug_dma_unmap_sg(dev, sglist, nents, dir); +} +EXPORT_SYMBOL(dma_unmap_sg); + +void dma_sync_single_for_cpu(struct device *dev, dma_addr_t handle, + size_t size, enum dma_data_direction dir) +{ + c6x_dma_sync(handle, size, dir); + + debug_dma_sync_single_for_cpu(dev, handle, size, dir); +} +EXPORT_SYMBOL(dma_sync_single_for_cpu); + + +void dma_sync_single_for_device(struct device *dev, dma_addr_t handle, + size_t size, enum dma_data_direction dir) +{ + c6x_dma_sync(handle, size, dir); + + debug_dma_sync_single_for_device(dev, handle, size, dir); +} +EXPORT_SYMBOL(dma_sync_single_for_device); + + +void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sglist, + int nents, enum dma_data_direction dir) +{ + struct scatterlist *sg; + int i; + + for_each_sg(sglist, sg, nents, i) + dma_sync_single_for_cpu(dev, sg_dma_address(sg), + sg->length, dir); + + debug_dma_sync_sg_for_cpu(dev, sglist, nents, dir); +} +EXPORT_SYMBOL(dma_sync_sg_for_cpu); + + +void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sglist, + int nents, enum dma_data_direction dir) +{ + struct scatterlist *sg; + int i; + + for_each_sg(sglist, sg, nents, i) + dma_sync_single_for_device(dev, sg_dma_address(sg), + sg->length, dir); + + debug_dma_sync_sg_for_device(dev, sglist, nents, dir); +} +EXPORT_SYMBOL(dma_sync_sg_for_device); + + +/* Number of entries preallocated for DMA-API debugging */ +#define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16) + +static int __init dma_init(void) +{ + dma_debug_init(PREALLOC_DMA_DEBUG_ENTRIES); + + return 0; +} +fs_initcall(dma_init); diff --git a/arch/c6x/kernel/entry.S b/arch/c6x/kernel/entry.S new file mode 100644 index 0000000..3e977cc --- /dev/null +++ b/arch/c6x/kernel/entry.S @@ -0,0 +1,803 @@ +; +; Port on Texas Instruments TMS320C6x architecture +; +; Copyright (C) 2004-2011 Texas Instruments Incorporated +; Author: Aurelien Jacquiot (aurelien.jacquiot@virtuallogix.com) +; Updated for 2.6.34: Mark Salter <msalter@redhat.com> +; +; This program is free software; you can redistribute it and/or modify +; it under the terms of the GNU General Public License version 2 as +; published by the Free Software Foundation. +; + +#include <linux/sys.h> +#include <linux/linkage.h> +#include <asm/thread_info.h> +#include <asm/asm-offsets.h> +#include <asm/unistd.h> +#include <asm/errno.h> + +; Registers naming +#define DP B14 +#define SP B15 + +#ifndef CONFIG_PREEMPT +#define resume_kernel restore_all +#endif + + .altmacro + + .macro MASK_INT reg + MVC .S2 CSR,reg + CLR .S2 reg,0,0,reg + MVC .S2 reg,CSR + .endm + + .macro UNMASK_INT reg + MVC .S2 CSR,reg + SET .S2 reg,0,0,reg + MVC .S2 reg,CSR + .endm + + .macro GET_THREAD_INFO reg + SHR .S1X SP,THREAD_SHIFT,reg + SHL .S1 reg,THREAD_SHIFT,reg + .endm + + ;; + ;; This defines the normal kernel pt_regs layout. + ;; + .macro SAVE_ALL __rp __tsr + STW .D2T2 B0,*SP--[2] ; save original B0 + MVKL .S2 current_ksp,B0 + MVKH .S2 current_ksp,B0 + LDW .D2T2 *B0,B1 ; KSP + + NOP 3 + STW .D2T2 B1,*+SP[1] ; save original B1 + XOR .D2 SP,B1,B0 ; (SP ^ KSP) + LDW .D2T2 *+SP[1],B1 ; restore B0/B1 + LDW .D2T2 *++SP[2],B0 + SHR .S2 B0,THREAD_SHIFT,B0 ; 0 if already using kstack + [B0] STDW .D2T2 SP:DP,*--B1[1] ; user: save user sp/dp kstack + [B0] MV .S2 B1,SP ; and switch to kstack +||[!B0] STDW .D2T2 SP:DP,*--SP[1] ; kernel: save on current stack + + SUBAW .D2 SP,2,SP + + ADD .D1X SP,-8,A15 + || STDW .D2T1 A15:A14,*SP--[16] ; save A15:A14 + + STDW .D2T2 B13:B12,*SP--[1] + || STDW .D1T1 A13:A12,*A15--[1] + || MVC .S2 __rp,B13 + + STDW .D2T2 B11:B10,*SP--[1] + || STDW .D1T1 A11:A10,*A15--[1] + || MVC .S2 CSR,B12 + + STDW .D2T2 B9:B8,*SP--[1] + || STDW .D1T1 A9:A8,*A15--[1] + || MVC .S2 RILC,B11 + STDW .D2T2 B7:B6,*SP--[1] + || STDW .D1T1 A7:A6,*A15--[1] + || MVC .S2 ILC,B10 + + STDW .D2T2 B5:B4,*SP--[1] + || STDW .D1T1 A5:A4,*A15--[1] + + STDW .D2T2 B3:B2,*SP--[1] + || STDW .D1T1 A3:A2,*A15--[1] + || MVC .S2 __tsr,B5 + + STDW .D2T2 B1:B0,*SP--[1] + || STDW .D1T1 A1:A0,*A15--[1] + || MV .S1X B5,A5 + + STDW .D2T2 B31:B30,*SP--[1] + || STDW .D1T1 A31:A30,*A15--[1] + STDW .D2T2 B29:B28,*SP--[1] + || STDW .D1T1 A29:A28,*A15--[1] + STDW .D2T2 B27:B26,*SP--[1] + || STDW .D1T1 A27:A26,*A15--[1] + STDW .D2T2 B25:B24,*SP--[1] + || STDW .D1T1 A25:A24,*A15--[1] + STDW .D2T2 B23:B22,*SP--[1] + || STDW .D1T1 A23:A22,*A15--[1] + STDW .D2T2 B21:B20,*SP--[1] + || STDW .D1T1 A21:A20,*A15--[1] + STDW .D2T2 B19:B18,*SP--[1] + || STDW .D1T1 A19:A18,*A15--[1] + STDW .D2T2 B17:B16,*SP--[1] + || STDW .D1T1 A17:A16,*A15--[1] + + STDW .D2T2 B13:B12,*SP--[1] ; save PC and CSR + + STDW .D2T2 B11:B10,*SP--[1] ; save RILC and ILC + STDW .D2T1 A5:A4,*SP--[1] ; save TSR and orig A4 + + ;; We left an unused word on the stack just above pt_regs. + ;; It is used to save whether or not this frame is due to + ;; a syscall. It is cleared here, but the syscall handler + ;; sets it to a non-zero value. + MVK .L2 0,B1 + STW .D2T2 B1,*+SP(REGS__END+8) ; clear syscall flag + .endm + + .macro RESTORE_ALL __rp __tsr + LDDW .D2T2 *++SP[1],B9:B8 ; get TSR (B9) + LDDW .D2T2 *++SP[1],B11:B10 ; get RILC (B11) and ILC (B10) + LDDW .D2T2 *++SP[1],B13:B12 ; get PC (B13) and CSR (B12) + + ADDAW .D1X SP,30,A15 + + LDDW .D1T1 *++A15[1],A17:A16 + || LDDW .D2T2 *++SP[1],B17:B16 + LDDW .D1T1 *++A15[1],A19:A18 + || LDDW .D2T2 *++SP[1],B19:B18 + LDDW .D1T1 *++A15[1],A21:A20 + || LDDW .D2T2 *++SP[1],B21:B20 + LDDW .D1T1 *++A15[1],A23:A22 + || LDDW .D2T2 *++SP[1],B23:B22 + LDDW .D1T1 *++A15[1],A25:A24 + || LDDW .D2T2 *++SP[1],B25:B24 + LDDW .D1T1 *++A15[1],A27:A26 + || LDDW .D2T2 *++SP[1],B27:B26 + LDDW .D1T1 *++A15[1],A29:A28 + || LDDW .D2T2 *++SP[1],B29:B28 + LDDW .D1T1 *++A15[1],A31:A30 + || LDDW .D2T2 *++SP[1],B31:B30 + + LDDW .D1T1 *++A15[1],A1:A0 + || LDDW .D2T2 *++SP[1],B1:B0 + + LDDW .D1T1 *++A15[1],A3:A2 + || LDDW .D2T2 *++SP[1],B3:B2 + || MVC .S2 B9,__tsr + LDDW .D1T1 *++A15[1],A5:A4 + || LDDW .D2T2 *++SP[1],B5:B4 + || MVC .S2 B11,RILC + LDDW .D1T1 *++A15[1],A7:A6 + || LDDW .D2T2 *++SP[1],B7:B6 + || MVC .S2 B10,ILC + + LDDW .D1T1 *++A15[1],A9:A8 + || LDDW .D2T2 *++SP[1],B9:B8 + || MVC .S2 B13,__rp + + LDDW .D1T1 *++A15[1],A11:A10 + || LDDW .D2T2 *++SP[1],B11:B10 + || MVC .S2 B12,CSR + + LDDW .D1T1 *++A15[1],A13:A12 + || LDDW .D2T2 *++SP[1],B13:B12 + + MV .D2X A15,SP + || MVKL .S1 current_ksp,A15 + MVKH .S1 current_ksp,A15 + || ADDAW .D1X SP,6,A14 + STW .D1T1 A14,*A15 ; save kernel stack pointer + + LDDW .D2T1 *++SP[1],A15:A14 + + B .S2 __rp ; return from interruption + LDDW .D2T2 *+SP[1],SP:DP + NOP 4 + .endm + + .section .text + + ;; + ;; Jump to schedule() then return to ret_from_exception + ;; +_reschedule: +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S1 schedule,A0 + MVKH .S1 schedule,A0 + B .S2X A0 +#else + B .S1 schedule +#endif + ADDKPC .S2 ret_from_exception,B3,4 + + ;; + ;; Called before syscall handler when process is being debugged + ;; +tracesys_on: +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S1 syscall_trace_entry,A0 + MVKH .S1 syscall_trace_entry,A0 + B .S2X A0 +#else + B .S1 syscall_trace_entry +#endif + ADDKPC .S2 ret_from_syscall_trace,B3,3 + ADD .S1X 8,SP,A4 + +ret_from_syscall_trace: + ;; tracing returns (possibly new) syscall number + MV .D2X A4,B0 + || MVK .S2 __NR_syscalls,B1 + CMPLTU .L2 B0,B1,B1 + + [!B1] BNOP .S2 ret_from_syscall_function,5 + || MVK .S1 -ENOSYS,A4 + + ;; reload syscall args from (possibly modified) stack frame + ;; and get syscall handler addr from sys_call_table: + LDW .D2T2 *+SP(REGS_B4+8),B4 + || MVKL .S2 sys_call_table,B1 + LDW .D2T1 *+SP(REGS_A6+8),A6 + || MVKH .S2 sys_call_table,B1 + LDW .D2T2 *+B1[B0],B0 + || MVKL .S2 ret_from_syscall_function,B3 + LDW .D2T2 *+SP(REGS_B6+8),B6 + || MVKH .S2 ret_from_syscall_function,B3 + LDW .D2T1 *+SP(REGS_A8+8),A8 + LDW .D2T2 *+SP(REGS_B8+8),B8 + NOP + ; B0 = sys_call_table[__NR_*] + BNOP .S2 B0,5 ; branch to syscall handler + || LDW .D2T1 *+SP(REGS_ORIG_A4+8),A4 + +syscall_exit_work: + AND .D1 _TIF_SYSCALL_TRACE,A2,A0 + [!A0] BNOP .S1 work_pending,5 + [A0] B .S2 syscall_trace_exit + ADDKPC .S2 resume_userspace,B3,1 + MVC .S2 CSR,B1 + SET .S2 B1,0,0,B1 + MVC .S2 B1,CSR ; enable ints + +work_pending: + AND .D1 _TIF_NEED_RESCHED,A2,A0 + [!A0] BNOP .S1 work_notifysig,5 + +work_resched: +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S1 schedule,A1 + MVKH .S1 schedule,A1 + B .S2X A1 +#else + B .S2 schedule +#endif + ADDKPC .S2 work_rescheduled,B3,4 +work_rescheduled: + ;; make sure we don't miss an interrupt setting need_resched or + ;; sigpending between sampling and the rti + MASK_INT B2 + GET_THREAD_INFO A12 + LDW .D1T1 *+A12(THREAD_INFO_FLAGS),A2 + MVK .S1 _TIF_WORK_MASK,A1 + MVK .S1 _TIF_NEED_RESCHED,A3 + NOP 2 + AND .D1 A1,A2,A0 + || AND .S1 A3,A2,A1 + [!A0] BNOP .S1 restore_all,5 + [A1] BNOP .S1 work_resched,5 + +work_notifysig: + B .S2 do_notify_resume + LDW .D2T1 *+SP(REGS__END+8),A6 ; syscall flag + ADDKPC .S2 resume_userspace,B3,1 + ADD .S1X 8,SP,A4 ; pt_regs pointer is first arg + MV .D2X A2,B4 ; thread_info flags is second arg + + ;; + ;; On C64x+, the return way from exception and interrupt + ;; is a little bit different + ;; +ENTRY(ret_from_exception) +#ifdef CONFIG_PREEMPT + MASK_INT B2 +#endif + +ENTRY(ret_from_interrupt) + ;; + ;; Check if we are comming from user mode. + ;; + LDW .D2T2 *+SP(REGS_TSR+8),B0 + MVK .S2 0x40,B1 + NOP 3 + AND .D2 B0,B1,B0 + [!B0] BNOP .S2 resume_kernel,5 + +resume_userspace: + ;; make sure we don't miss an interrupt setting need_resched or + ;; sigpending between sampling and the rti + MASK_INT B2 + GET_THREAD_INFO A12 + LDW .D1T1 *+A12(THREAD_INFO_FLAGS),A2 + MVK .S1 _TIF_WORK_MASK,A1 + MVK .S1 _TIF_NEED_RESCHED,A3 + NOP 2 + AND .D1 A1,A2,A0 + [A0] BNOP .S1 work_pending,5 + BNOP .S1 restore_all,5 + + ;; + ;; System call handling + ;; B0 = syscall number (in sys_call_table) + ;; A4,B4,A6,B6,A8,B8 = arguments of the syscall function + ;; A4 is the return value register + ;; +system_call_saved: + MVK .L2 1,B2 + STW .D2T2 B2,*+SP(REGS__END+8) ; set syscall flag + MVC .S2 B2,ECR ; ack the software exception + + UNMASK_INT B2 ; re-enable global IT + +system_call_saved_noack: + ;; Check system call number + MVK .S2 __NR_syscalls,B1 +#ifdef CONFIG_C6X_BIG_KERNEL + || MVKL .S1 sys_ni_syscall,A0 +#endif + CMPLTU .L2 B0,B1,B1 +#ifdef CONFIG_C6X_BIG_KERNEL + || MVKH .S1 sys_ni_syscall,A0 +#endif + + ;; Check for ptrace + GET_THREAD_INFO A12 + +#ifdef CONFIG_C6X_BIG_KERNEL + [!B1] B .S2X A0 +#else + [!B1] B .S2 sys_ni_syscall +#endif + [!B1] ADDKPC .S2 ret_from_syscall_function,B3,4 + + ;; Get syscall handler addr from sys_call_table + ;; call tracesys_on or call syscall handler + LDW .D1T1 *+A12(THREAD_INFO_FLAGS),A2 + || MVKL .S2 sys_call_table,B1 + MVKH .S2 sys_call_table,B1 + LDW .D2T2 *+B1[B0],B0 + NOP 2 + ; A2 = thread_info flags + AND .D1 _TIF_SYSCALL_TRACE,A2,A2 + [A2] BNOP .S1 tracesys_on,5 + ;; B0 = _sys_call_table[__NR_*] + B .S2 B0 + ADDKPC .S2 ret_from_syscall_function,B3,4 + +ret_from_syscall_function: + STW .D2T1 A4,*+SP(REGS_A4+8) ; save return value in A4 + ; original A4 is in orig_A4 +syscall_exit: + ;; make sure we don't miss an interrupt setting need_resched or + ;; sigpending between sampling and the rti + MASK_INT B2 + LDW .D1T1 *+A12(THREAD_INFO_FLAGS),A2 + MVK .S1 _TIF_ALLWORK_MASK,A1 + NOP 3 + AND .D1 A1,A2,A2 ; check for work to do + [A2] BNOP .S1 syscall_exit_work,5 + +restore_all: + RESTORE_ALL NRP,NTSR + + ;; + ;; After a fork we jump here directly from resume, + ;; so that A4 contains the previous task structure. + ;; +ENTRY(ret_from_fork) +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S1 schedule_tail,A0 + MVKH .S1 schedule_tail,A0 + B .S2X A0 +#else + B .S2 schedule_tail +#endif + ADDKPC .S2 ret_from_fork_2,B3,4 +ret_from_fork_2: + ;; return 0 in A4 for child process + GET_THREAD_INFO A12 + BNOP .S2 syscall_exit,3 + MVK .L2 0,B0 + STW .D2T2 B0,*+SP(REGS_A4+8) +ENDPROC(ret_from_fork) + + ;; + ;; These are the interrupt handlers, responsible for calling __do_IRQ() + ;; int6 is used for syscalls (see _system_call entry) + ;; + .macro SAVE_ALL_INT + SAVE_ALL IRP,ITSR + .endm + + .macro CALL_INT int +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S1 c6x_do_IRQ,A0 + MVKH .S1 c6x_do_IRQ,A0 + BNOP .S2X A0,1 + MVK .S1 int,A4 + ADDAW .D2 SP,2,B4 + MVKL .S2 ret_from_interrupt,B3 + MVKH .S2 ret_from_interrupt,B3 +#else + CALLP .S2 c6x_do_IRQ,B3 + || MVK .S1 int,A4 + || ADDAW .D2 SP,2,B4 + B .S1 ret_from_interrupt + NOP 5 +#endif + .endm + +ENTRY(_int4_handler) + SAVE_ALL_INT + CALL_INT 4 +ENDPROC(_int4_handler) + +ENTRY(_int5_handler) + SAVE_ALL_INT + CALL_INT 5 +ENDPROC(_int5_handler) + +ENTRY(_int6_handler) + SAVE_ALL_INT + CALL_INT 6 +ENDPROC(_int6_handler) + +ENTRY(_int7_handler) + SAVE_ALL_INT + CALL_INT 7 +ENDPROC(_int7_handler) + +ENTRY(_int8_handler) + SAVE_ALL_INT + CALL_INT 8 +ENDPROC(_int8_handler) + +ENTRY(_int9_handler) + SAVE_ALL_INT + CALL_INT 9 +ENDPROC(_int9_handler) + +ENTRY(_int10_handler) + SAVE_ALL_INT + CALL_INT 10 +ENDPROC(_int10_handler) + +ENTRY(_int11_handler) + SAVE_ALL_INT + CALL_INT 11 +ENDPROC(_int11_handler) + +ENTRY(_int12_handler) + SAVE_ALL_INT + CALL_INT 12 +ENDPROC(_int12_handler) + +ENTRY(_int13_handler) + SAVE_ALL_INT + CALL_INT 13 +ENDPROC(_int13_handler) + +ENTRY(_int14_handler) + SAVE_ALL_INT + CALL_INT 14 +ENDPROC(_int14_handler) + +ENTRY(_int15_handler) + SAVE_ALL_INT + CALL_INT 15 +ENDPROC(_int15_handler) + + ;; + ;; Handler for uninitialized and spurious interrupts + ;; +ENTRY(_bad_interrupt) + B .S2 IRP + NOP 5 +ENDPROC(_bad_interrupt) + + ;; + ;; Entry for NMI/exceptions/syscall + ;; +ENTRY(_nmi_handler) + SAVE_ALL NRP,NTSR + + MVC .S2 EFR,B2 + CMPEQ .L2 1,B2,B2 + || MVC .S2 TSR,B1 + CLR .S2 B1,10,10,B1 + MVC .S2 B1,TSR +#ifdef CONFIG_C6X_BIG_KERNEL + [!B2] MVKL .S1 process_exception,A0 + [!B2] MVKH .S1 process_exception,A0 + [!B2] B .S2X A0 +#else + [!B2] B .S2 process_exception +#endif + [B2] B .S2 system_call_saved + [!B2] ADDAW .D2 SP,2,B1 + [!B2] MV .D1X B1,A4 + ADDKPC .S2 ret_from_trap,B3,2 + +ret_from_trap: + MV .D2X A4,B0 + [!B0] BNOP .S2 ret_from_exception,5 + +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S2 system_call_saved_noack,B3 + MVKH .S2 system_call_saved_noack,B3 +#endif + LDW .D2T2 *+SP(REGS_B0+8),B0 + LDW .D2T1 *+SP(REGS_A4+8),A4 + LDW .D2T2 *+SP(REGS_B4+8),B4 + LDW .D2T1 *+SP(REGS_A6+8),A6 + LDW .D2T2 *+SP(REGS_B6+8),B6 + LDW .D2T1 *+SP(REGS_A8+8),A8 +#ifdef CONFIG_C6X_BIG_KERNEL + || B .S2 B3 +#else + || B .S2 system_call_saved_noack +#endif + LDW .D2T2 *+SP(REGS_B8+8),B8 + NOP 4 +ENDPROC(_nmi_handler) + + ;; + ;; Jump to schedule() then return to ret_from_isr + ;; +#ifdef CONFIG_PREEMPT +resume_kernel: + GET_THREAD_INFO A12 + LDW .D1T1 *+A12(THREAD_INFO_PREEMPT_COUNT),A1 + NOP 4 + [A1] BNOP .S2 restore_all,5 + +preempt_schedule: + GET_THREAD_INFO A2 + LDW .D1T1 *+A2(THREAD_INFO_FLAGS),A1 +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S2 preempt_schedule_irq,B0 + MVKH .S2 preempt_schedule_irq,B0 + NOP 2 +#else + NOP 4 +#endif + AND .D1 _TIF_NEED_RESCHED,A1,A1 + [!A1] BNOP .S2 restore_all,5 +#ifdef CONFIG_C6X_BIG_KERNEL + B .S2 B0 +#else + B .S2 preempt_schedule_irq +#endif + ADDKPC .S2 preempt_schedule,B3,4 +#endif /* CONFIG_PREEMPT */ + +ENTRY(enable_exception) + DINT + MVC .S2 TSR,B0 + MVC .S2 B3,NRP + MVK .L2 0xc,B1 + OR .D2 B0,B1,B0 + MVC .S2 B0,TSR ; Set GEE and XEN in TSR + B .S2 NRP + NOP 5 +ENDPROC(enable_exception) + +ENTRY(sys_sigaltstack) +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S1 do_sigaltstack,A0 ; branch to do_sigaltstack + MVKH .S1 do_sigaltstack,A0 + B .S2X A0 +#else + B .S2 do_sigaltstack +#endif + LDW .D2T1 *+SP(REGS_SP+8),A6 + NOP 4 +ENDPROC(sys_sigaltstack) + + ;; kernel_execve +ENTRY(kernel_execve) + MVK .S2 __NR_execve,B0 + SWE + BNOP .S2 B3,5 +ENDPROC(kernel_execve) + + ;; + ;; Special system calls + ;; return address is in B3 + ;; +ENTRY(sys_clone) + ADD .D1X SP,8,A4 +#ifdef CONFIG_C6X_BIG_KERNEL + || MVKL .S1 sys_c6x_clone,A0 + MVKH .S1 sys_c6x_clone,A0 + BNOP .S2X A0,5 +#else + || B .S2 sys_c6x_clone + NOP 5 +#endif +ENDPROC(sys_clone) + +ENTRY(sys_rt_sigreturn) + ADD .D1X SP,8,A4 +#ifdef CONFIG_C6X_BIG_KERNEL + || MVKL .S1 do_rt_sigreturn,A0 + MVKH .S1 do_rt_sigreturn,A0 + BNOP .S2X A0,5 +#else + || B .S2 do_rt_sigreturn + NOP 5 +#endif +ENDPROC(sys_rt_sigreturn) + +ENTRY(sys_execve) + ADDAW .D2 SP,2,B6 ; put regs addr in 4th parameter + ; & adjust regs stack addr + LDW .D2T2 *+SP(REGS_B4+8),B4 + + ;; c6x_execve(char *name, char **argv, + ;; char **envp, struct pt_regs *regs) +#ifdef CONFIG_C6X_BIG_KERNEL + || MVKL .S1 sys_c6x_execve,A0 + MVKH .S1 sys_c6x_execve,A0 + B .S2X A0 +#else + || B .S2 sys_c6x_execve +#endif + STW .D2T2 B3,*SP--[2] + ADDKPC .S2 ret_from_c6x_execve,B3,3 + +ret_from_c6x_execve: + LDW .D2T2 *++SP[2],B3 + NOP 4 + BNOP .S2 B3,5 +ENDPROC(sys_execve) + +ENTRY(sys_pread_c6x) + MV .D2X A8,B7 +#ifdef CONFIG_C6X_BIG_KERNEL + || MVKL .S1 sys_pread64,A0 + MVKH .S1 sys_pread64,A0 + BNOP .S2X A0,5 +#else + || B .S2 sys_pread64 + NOP 5 +#endif +ENDPROC(sys_pread_c6x) + +ENTRY(sys_pwrite_c6x) + MV .D2X A8,B7 +#ifdef CONFIG_C6X_BIG_KERNEL + || MVKL .S1 sys_pwrite64,A0 + MVKH .S1 sys_pwrite64,A0 + BNOP .S2X A0,5 +#else + || B .S2 sys_pwrite64 + NOP 5 +#endif +ENDPROC(sys_pwrite_c6x) + +;; On Entry +;; A4 - path +;; B4 - offset_lo (LE), offset_hi (BE) +;; A6 - offset_lo (BE), offset_hi (LE) +ENTRY(sys_truncate64_c6x) +#ifdef CONFIG_CPU_BIG_ENDIAN + MV .S2 B4,B5 + MV .D2X A6,B4 +#else + MV .D2X A6,B5 +#endif +#ifdef CONFIG_C6X_BIG_KERNEL + || MVKL .S1 sys_truncate64,A0 + MVKH .S1 sys_truncate64,A0 + BNOP .S2X A0,5 +#else + || B .S2 sys_truncate64 + NOP 5 +#endif +ENDPROC(sys_truncate64_c6x) + +;; On Entry +;; A4 - fd +;; B4 - offset_lo (LE), offset_hi (BE) +;; A6 - offset_lo (BE), offset_hi (LE) +ENTRY(sys_ftruncate64_c6x) +#ifdef CONFIG_CPU_BIG_ENDIAN + MV .S2 B4,B5 + MV .D2X A6,B4 +#else + MV .D2X A6,B5 +#endif +#ifdef CONFIG_C6X_BIG_KERNEL + || MVKL .S1 sys_ftruncate64,A0 + MVKH .S1 sys_ftruncate64,A0 + BNOP .S2X A0,5 +#else + || B .S2 sys_ftruncate64 + NOP 5 +#endif +ENDPROC(sys_ftruncate64_c6x) + +#ifdef __ARCH_WANT_SYSCALL_OFF_T +;; On Entry +;; A4 - fd +;; B4 - offset_lo (LE), offset_hi (BE) +;; A6 - offset_lo (BE), offset_hi (LE) +;; B6 - len +;; A8 - advice +ENTRY(sys_fadvise64_c6x) +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S1 sys_fadvise64,A0 + MVKH .S1 sys_fadvise64,A0 + BNOP .S2X A0,2 +#else + B .S2 sys_fadvise64 + NOP 2 +#endif +#ifdef CONFIG_CPU_BIG_ENDIAN + MV .L2 B4,B5 + || MV .D2X A6,B4 +#else + MV .D2X A6,B5 +#endif + MV .D1X B6,A6 + MV .D2X A8,B6 +#endif +ENDPROC(sys_fadvise64_c6x) + +;; On Entry +;; A4 - fd +;; B4 - offset_lo (LE), offset_hi (BE) +;; A6 - offset_lo (BE), offset_hi (LE) +;; B6 - len_lo (LE), len_hi (BE) +;; A8 - len_lo (BE), len_hi (LE) +;; B8 - advice +ENTRY(sys_fadvise64_64_c6x) +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S1 sys_fadvise64_64,A0 + MVKH .S1 sys_fadvise64_64,A0 + BNOP .S2X A0,2 +#else + B .S2 sys_fadvise64_64 + NOP 2 +#endif +#ifdef CONFIG_CPU_BIG_ENDIAN + MV .L2 B4,B5 + || MV .D2X A6,B4 + MV .L1 A8,A6 + || MV .D1X B6,A7 +#else + MV .D2X A6,B5 + MV .L1 A8,A7 + || MV .D1X B6,A6 +#endif + MV .L2 B8,B6 +ENDPROC(sys_fadvise64_64_c6x) + +;; On Entry +;; A4 - fd +;; B4 - mode +;; A6 - offset_hi +;; B6 - offset_lo +;; A8 - len_hi +;; B8 - len_lo +ENTRY(sys_fallocate_c6x) +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S1 sys_fallocate,A0 + MVKH .S1 sys_fallocate,A0 + BNOP .S2X A0,1 +#else + B .S2 sys_fallocate + NOP +#endif + MV .D1 A6,A7 + MV .D1X B6,A6 + MV .D2X A8,B7 + MV .D2 B8,B6 +ENDPROC(sys_fallocate_c6x) + + ;; put this in .neardata for faster access when using DSBT mode + .section .neardata,"aw",@progbits + .global current_ksp + .hidden current_ksp +current_ksp: + .word init_thread_union + THREAD_START_SP diff --git a/arch/c6x/kernel/head.S b/arch/c6x/kernel/head.S new file mode 100644 index 0000000..133eab6 --- /dev/null +++ b/arch/c6x/kernel/head.S @@ -0,0 +1,84 @@ +; +; Port on Texas Instruments TMS320C6x architecture +; +; Copyright (C) 2004, 2009, 2010, 2011 Texas Instruments Incorporated +; Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) +; +; This program is free software; you can redistribute it and/or modify +; it under the terms of the GNU General Public License version 2 as +; published by the Free Software Foundation. +; +#include <linux/linkage.h> +#include <linux/of_fdt.h> +#include <asm/asm-offsets.h> + + __HEAD +ENTRY(_c_int00) + ;; Save magic and pointer + MV .S1 A4,A10 + MV .S2 B4,B10 + MVKL .S2 __bss_start,B5 + MVKH .S2 __bss_start,B5 + MVKL .S2 __bss_stop,B6 + MVKH .S2 __bss_stop,B6 + SUB .L2 B6,B5,B6 ; bss size + + ;; Set the stack pointer + MVKL .S2 current_ksp,B0 + MVKH .S2 current_ksp,B0 + LDW .D2T2 *B0,B15 + + ;; clear bss + SHR .S2 B6,3,B0 ; number of dwords to clear + ZERO .L2 B13 + ZERO .L2 B12 +bss_loop: + BDEC .S2 bss_loop,B0 + NOP 3 + CMPLT .L2 B0,0,B1 + [!B1] STDW .D2T2 B13:B12,*B5++[1] + + NOP 4 + AND .D2 ~7,B15,B15 + + ;; Clear GIE and PGIE + MVC .S2 CSR,B2 + CLR .S2 B2,0,1,B2 + MVC .S2 B2,CSR + MVC .S2 TSR,B2 + CLR .S2 B2,0,1,B2 + MVC .S2 B2,TSR + MVC .S2 ITSR,B2 + CLR .S2 B2,0,1,B2 + MVC .S2 B2,ITSR + MVC .S2 NTSR,B2 + CLR .S2 B2,0,1,B2 + MVC .S2 B2,NTSR + + ;; pass DTB pointer to machine_init (or zero if none) + MVKL .S1 OF_DT_HEADER,A0 + MVKH .S1 OF_DT_HEADER,A0 + CMPEQ .L1 A10,A0,A0 + [A0] MV .S1X B10,A4 + [!A0] MVK .S1 0,A4 + +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S1 machine_init,A0 + MVKH .S1 machine_init,A0 + B .S2X A0 + ADDKPC .S2 0f,B3,4 +0: +#else + CALLP .S2 machine_init,B3 +#endif + + ;; Jump to Linux init +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S1 start_kernel,A0 + MVKH .S1 start_kernel,A0 + B .S2X A0 +#else + B .S2 start_kernel +#endif + NOP 5 +L1: BNOP .S2 L1,5 diff --git a/arch/c6x/kernel/irq.c b/arch/c6x/kernel/irq.c new file mode 100644 index 0000000..0929e4b --- /dev/null +++ b/arch/c6x/kernel/irq.c @@ -0,0 +1,728 @@ +/* + * Copyright (C) 2011 Texas Instruments Incorporated + * + * This borrows heavily from powerpc version, which is: + * + * Derived from arch/i386/kernel/irq.c + * Copyright (C) 1992 Linus Torvalds + * Adapted from arch/i386 by Gary Thomas + * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) + * Updated and modified by Cort Dougan <cort@fsmlabs.com> + * Copyright (C) 1996-2001 Cort Dougan + * Adapted for Power Macintosh by Paul Mackerras + * Copyright (C) 1996 Paul Mackerras (paulus@cs.anu.edu.au) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#include <linux/slab.h> +#include <linux/seq_file.h> +#include <linux/radix-tree.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/of_irq.h> +#include <linux/interrupt.h> +#include <linux/kernel_stat.h> + +#include <asm/megamod-pic.h> + +unsigned long irq_err_count; + +static DEFINE_RAW_SPINLOCK(core_irq_lock); + +static void mask_core_irq(struct irq_data *data) +{ + unsigned int prio = data->irq; + + BUG_ON(prio < 4 || prio >= NR_PRIORITY_IRQS); + + raw_spin_lock(&core_irq_lock); + and_creg(IER, ~(1 << prio)); + raw_spin_unlock(&core_irq_lock); +} + +static void unmask_core_irq(struct irq_data *data) +{ + unsigned int prio = data->irq; + + raw_spin_lock(&core_irq_lock); + or_creg(IER, 1 << prio); + raw_spin_unlock(&core_irq_lock); +} + +static struct irq_chip core_chip = { + .name = "core", + .irq_mask = mask_core_irq, + .irq_unmask = unmask_core_irq, +}; + +asmlinkage void c6x_do_IRQ(unsigned int prio, struct pt_regs *regs) +{ + struct pt_regs *old_regs = set_irq_regs(regs); + + irq_enter(); + + BUG_ON(prio < 4 || prio >= NR_PRIORITY_IRQS); + + generic_handle_irq(prio); + + irq_exit(); + + set_irq_regs(old_regs); +} + +static struct irq_host *core_host; + +static int core_host_map(struct irq_host *h, unsigned int virq, + irq_hw_number_t hw) +{ + if (hw < 4 || hw >= NR_PRIORITY_IRQS) + return -EINVAL; + + irq_set_status_flags(virq, IRQ_LEVEL); + irq_set_chip_and_handler(virq, &core_chip, handle_level_irq); + return 0; +} + +static struct irq_host_ops core_host_ops = { + .map = core_host_map, +}; + +void __init init_IRQ(void) +{ + struct device_node *np; + + /* Mask all priority IRQs */ + and_creg(IER, ~0xfff0); + + np = of_find_compatible_node(NULL, NULL, "ti,c64x+core-pic"); + if (np != NULL) { + /* create the core host */ + core_host = irq_alloc_host(np, IRQ_HOST_MAP_PRIORITY, 0, + &core_host_ops, 0); + if (core_host) + irq_set_default_host(core_host); + of_node_put(np); + } + + printk(KERN_INFO "Core interrupt controller initialized\n"); + + /* now we're ready for other SoC controllers */ + megamod_pic_init(); + + /* Clear all general IRQ flags */ + set_creg(ICR, 0xfff0); +} + +void ack_bad_irq(int irq) +{ + printk(KERN_ERR "IRQ: spurious interrupt %d\n", irq); + irq_err_count++; +} + +int arch_show_interrupts(struct seq_file *p, int prec) +{ + seq_printf(p, "%*s: %10lu\n", prec, "Err", irq_err_count); + return 0; +} + +/* + * IRQ controller and virtual interrupts + */ + +/* The main irq map itself is an array of NR_IRQ entries containing the + * associate host and irq number. An entry with a host of NULL is free. + * An entry can be allocated if it's free, the allocator always then sets + * hwirq first to the host's invalid irq number and then fills ops. + */ +struct irq_map_entry { + irq_hw_number_t hwirq; + struct irq_host *host; +}; + +static LIST_HEAD(irq_hosts); +static DEFINE_RAW_SPINLOCK(irq_big_lock); +static DEFINE_MUTEX(revmap_trees_mutex); +static struct irq_map_entry irq_map[NR_IRQS]; +static unsigned int irq_virq_count = NR_IRQS; +static struct irq_host *irq_default_host; + +irq_hw_number_t irqd_to_hwirq(struct irq_data *d) +{ + return irq_map[d->irq].hwirq; +} +EXPORT_SYMBOL_GPL(irqd_to_hwirq); + +irq_hw_number_t virq_to_hw(unsigned int virq) +{ + return irq_map[virq].hwirq; +} +EXPORT_SYMBOL_GPL(virq_to_hw); + +bool virq_is_host(unsigned int virq, struct irq_host *host) +{ + return irq_map[virq].host == host; +} +EXPORT_SYMBOL_GPL(virq_is_host); + +static int default_irq_host_match(struct irq_host *h, struct device_node *np) +{ + return h->of_node != NULL && h->of_node == np; +} + +struct irq_host *irq_alloc_host(struct device_node *of_node, + unsigned int revmap_type, + unsigned int revmap_arg, + struct irq_host_ops *ops, + irq_hw_number_t inval_irq) +{ + struct irq_host *host; + unsigned int size = sizeof(struct irq_host); + unsigned int i; + unsigned int *rmap; + unsigned long flags; + + /* Allocate structure and revmap table if using linear mapping */ + if (revmap_type == IRQ_HOST_MAP_LINEAR) + size += revmap_arg * sizeof(unsigned int); + host = kzalloc(size, GFP_KERNEL); + if (host == NULL) + return NULL; + + /* Fill structure */ + host->revmap_type = revmap_type; + host->inval_irq = inval_irq; + host->ops = ops; + host->of_node = of_node_get(of_node); + + if (host->ops->match == NULL) + host->ops->match = default_irq_host_match; + + raw_spin_lock_irqsave(&irq_big_lock, flags); + + /* Check for the priority controller. */ + if (revmap_type == IRQ_HOST_MAP_PRIORITY) { + if (irq_map[0].host != NULL) { + raw_spin_unlock_irqrestore(&irq_big_lock, flags); + of_node_put(host->of_node); + kfree(host); + return NULL; + } + irq_map[0].host = host; + } + + list_add(&host->link, &irq_hosts); + raw_spin_unlock_irqrestore(&irq_big_lock, flags); + + /* Additional setups per revmap type */ + switch (revmap_type) { + case IRQ_HOST_MAP_PRIORITY: + /* 0 is always the invalid number for priority */ + host->inval_irq = 0; + /* setup us as the host for all priority interrupts */ + for (i = 1; i < NR_PRIORITY_IRQS; i++) { + irq_map[i].hwirq = i; + smp_wmb(); + irq_map[i].host = host; + smp_wmb(); + + ops->map(host, i, i); + } + break; + case IRQ_HOST_MAP_LINEAR: + rmap = (unsigned int *)(host + 1); + for (i = 0; i < revmap_arg; i++) + rmap[i] = NO_IRQ; + host->revmap_data.linear.size = revmap_arg; + smp_wmb(); + host->revmap_data.linear.revmap = rmap; + break; + case IRQ_HOST_MAP_TREE: + INIT_RADIX_TREE(&host->revmap_data.tree, GFP_KERNEL); + break; + default: + break; + } + + pr_debug("irq: Allocated host of type %d @0x%p\n", revmap_type, host); + + return host; +} + +struct irq_host *irq_find_host(struct device_node *node) +{ + struct irq_host *h, *found = NULL; + unsigned long flags; + + /* We might want to match the legacy controller last since + * it might potentially be set to match all interrupts in + * the absence of a device node. This isn't a problem so far + * yet though... + */ + raw_spin_lock_irqsave(&irq_big_lock, flags); + list_for_each_entry(h, &irq_hosts, link) + if (h->ops->match(h, node)) { + found = h; + break; + } + raw_spin_unlock_irqrestore(&irq_big_lock, flags); + return found; +} +EXPORT_SYMBOL_GPL(irq_find_host); + +void irq_set_default_host(struct irq_host *host) +{ + pr_debug("irq: Default host set to @0x%p\n", host); + + irq_default_host = host; +} + +void irq_set_virq_count(unsigned int count) +{ + pr_debug("irq: Trying to set virq count to %d\n", count); + + BUG_ON(count < NR_PRIORITY_IRQS); + if (count < NR_IRQS) + irq_virq_count = count; +} + +static int irq_setup_virq(struct irq_host *host, unsigned int virq, + irq_hw_number_t hwirq) +{ + int res; + + res = irq_alloc_desc_at(virq, 0); + if (res != virq) { + pr_debug("irq: -> allocating desc failed\n"); + goto error; + } + + /* map it */ + smp_wmb(); + irq_map[virq].hwirq = hwirq; + smp_mb(); + + if (host->ops->map(host, virq, hwirq)) { + pr_debug("irq: -> mapping failed, freeing\n"); + goto errdesc; + } + + irq_clear_status_flags(virq, IRQ_NOREQUEST); + + return 0; + +errdesc: + irq_free_descs(virq, 1); +error: + irq_free_virt(virq, 1); + return -1; +} + +unsigned int irq_create_direct_mapping(struct irq_host *host) +{ + unsigned int virq; + + if (host == NULL) + host = irq_default_host; + + BUG_ON(host == NULL); + WARN_ON(host->revmap_type != IRQ_HOST_MAP_NOMAP); + + virq = irq_alloc_virt(host, 1, 0); + if (virq == NO_IRQ) { + pr_debug("irq: create_direct virq allocation failed\n"); + return NO_IRQ; + } + + pr_debug("irq: create_direct obtained virq %d\n", virq); + + if (irq_setup_virq(host, virq, virq)) + return NO_IRQ; + + return virq; +} + +unsigned int irq_create_mapping(struct irq_host *host, + irq_hw_number_t hwirq) +{ + unsigned int virq, hint; + + pr_debug("irq: irq_create_mapping(0x%p, 0x%lx)\n", host, hwirq); + + /* Look for default host if nececssary */ + if (host == NULL) + host = irq_default_host; + if (host == NULL) { + printk(KERN_WARNING "irq_create_mapping called for" + " NULL host, hwirq=%lx\n", hwirq); + WARN_ON(1); + return NO_IRQ; + } + pr_debug("irq: -> using host @%p\n", host); + + /* Check if mapping already exists */ + virq = irq_find_mapping(host, hwirq); + if (virq != NO_IRQ) { + pr_debug("irq: -> existing mapping on virq %d\n", virq); + return virq; + } + + /* Allocate a virtual interrupt number */ + hint = hwirq % irq_virq_count; + virq = irq_alloc_virt(host, 1, hint); + if (virq == NO_IRQ) { + pr_debug("irq: -> virq allocation failed\n"); + return NO_IRQ; + } + + if (irq_setup_virq(host, virq, hwirq)) + return NO_IRQ; + + pr_debug("irq: irq %lu on host %s mapped to virtual irq %u\n", + hwirq, host->of_node ? host->of_node->full_name : "null", virq); + + return virq; +} +EXPORT_SYMBOL_GPL(irq_create_mapping); + +unsigned int irq_create_of_mapping(struct device_node *controller, + const u32 *intspec, unsigned int intsize) +{ + struct irq_host *host; + irq_hw_number_t hwirq; + unsigned int type = IRQ_TYPE_NONE; + unsigned int virq; + + if (controller == NULL) + host = irq_default_host; + else + host = irq_find_host(controller); + if (host == NULL) { + printk(KERN_WARNING "irq: no irq host found for %s !\n", + controller->full_name); + return NO_IRQ; + } + + /* If host has no translation, then we assume interrupt line */ + if (host->ops->xlate == NULL) + hwirq = intspec[0]; + else { + if (host->ops->xlate(host, controller, intspec, intsize, + &hwirq, &type)) + return NO_IRQ; + } + + /* Create mapping */ + virq = irq_create_mapping(host, hwirq); + if (virq == NO_IRQ) + return virq; + + /* Set type if specified and different than the current one */ + if (type != IRQ_TYPE_NONE && + type != (irqd_get_trigger_type(irq_get_irq_data(virq)))) + irq_set_irq_type(virq, type); + return virq; +} +EXPORT_SYMBOL_GPL(irq_create_of_mapping); + +void irq_dispose_mapping(unsigned int virq) +{ + struct irq_host *host; + irq_hw_number_t hwirq; + + if (virq == NO_IRQ) + return; + + /* Never unmap priority interrupts */ + if (virq < NR_PRIORITY_IRQS) + return; + + host = irq_map[virq].host; + if (WARN_ON(host == NULL)) + return; + + irq_set_status_flags(virq, IRQ_NOREQUEST); + + /* remove chip and handler */ + irq_set_chip_and_handler(virq, NULL, NULL); + + /* Make sure it's completed */ + synchronize_irq(virq); + + /* Tell the PIC about it */ + if (host->ops->unmap) + host->ops->unmap(host, virq); + smp_mb(); + + /* Clear reverse map */ + hwirq = irq_map[virq].hwirq; + switch (host->revmap_type) { + case IRQ_HOST_MAP_LINEAR: + if (hwirq < host->revmap_data.linear.size) + host->revmap_data.linear.revmap[hwirq] = NO_IRQ; + break; + case IRQ_HOST_MAP_TREE: + mutex_lock(&revmap_trees_mutex); + radix_tree_delete(&host->revmap_data.tree, hwirq); + mutex_unlock(&revmap_trees_mutex); + break; + } + + /* Destroy map */ + smp_mb(); + irq_map[virq].hwirq = host->inval_irq; + + irq_free_descs(virq, 1); + /* Free it */ + irq_free_virt(virq, 1); +} +EXPORT_SYMBOL_GPL(irq_dispose_mapping); + +unsigned int irq_find_mapping(struct irq_host *host, + irq_hw_number_t hwirq) +{ + unsigned int i; + unsigned int hint = hwirq % irq_virq_count; + + /* Look for default host if nececssary */ + if (host == NULL) + host = irq_default_host; + if (host == NULL) + return NO_IRQ; + + /* Slow path does a linear search of the map */ + i = hint; + do { + if (irq_map[i].host == host && + irq_map[i].hwirq == hwirq) + return i; + i++; + if (i >= irq_virq_count) + i = 4; + } while (i != hint); + return NO_IRQ; +} +EXPORT_SYMBOL_GPL(irq_find_mapping); + +unsigned int irq_radix_revmap_lookup(struct irq_host *host, + irq_hw_number_t hwirq) +{ + struct irq_map_entry *ptr; + unsigned int virq; + + if (WARN_ON_ONCE(host->revmap_type != IRQ_HOST_MAP_TREE)) + return irq_find_mapping(host, hwirq); + + /* + * The ptr returned references the static global irq_map. + * but freeing an irq can delete nodes along the path to + * do the lookup via call_rcu. + */ + rcu_read_lock(); + ptr = radix_tree_lookup(&host->revmap_data.tree, hwirq); + rcu_read_unlock(); + + /* + * If found in radix tree, then fine. + * Else fallback to linear lookup - this should not happen in practice + * as it means that we failed to insert the node in the radix tree. + */ + if (ptr) + virq = ptr - irq_map; + else + virq = irq_find_mapping(host, hwirq); + + return virq; +} + +void irq_radix_revmap_insert(struct irq_host *host, unsigned int virq, + irq_hw_number_t hwirq) +{ + if (WARN_ON(host->revmap_type != IRQ_HOST_MAP_TREE)) + return; + + if (virq != NO_IRQ) { + mutex_lock(&revmap_trees_mutex); + radix_tree_insert(&host->revmap_data.tree, hwirq, + &irq_map[virq]); + mutex_unlock(&revmap_trees_mutex); + } +} + +unsigned int irq_linear_revmap(struct irq_host *host, + irq_hw_number_t hwirq) +{ + unsigned int *revmap; + + if (WARN_ON_ONCE(host->revmap_type != IRQ_HOST_MAP_LINEAR)) + return irq_find_mapping(host, hwirq); + + /* Check revmap bounds */ + if (unlikely(hwirq >= host->revmap_data.linear.size)) + return irq_find_mapping(host, hwirq); + + /* Check if revmap was allocated */ + revmap = host->revmap_data.linear.revmap; + if (unlikely(revmap == NULL)) + return irq_find_mapping(host, hwirq); + + /* Fill up revmap with slow path if no mapping found */ + if (unlikely(revmap[hwirq] == NO_IRQ)) + revmap[hwirq] = irq_find_mapping(host, hwirq); + + return revmap[hwirq]; +} + +unsigned int irq_alloc_virt(struct irq_host *host, + unsigned int count, + unsigned int hint) +{ + unsigned long flags; + unsigned int i, j, found = NO_IRQ; + + if (count == 0 || count > (irq_virq_count - NR_PRIORITY_IRQS)) + return NO_IRQ; + + raw_spin_lock_irqsave(&irq_big_lock, flags); + + /* Use hint for 1 interrupt if any */ + if (count == 1 && hint >= NR_PRIORITY_IRQS && + hint < irq_virq_count && irq_map[hint].host == NULL) { + found = hint; + goto hint_found; + } + + /* Look for count consecutive numbers in the allocatable + * (non-legacy) space + */ + for (i = NR_PRIORITY_IRQS, j = 0; i < irq_virq_count; i++) { + if (irq_map[i].host != NULL) + j = 0; + else + j++; + + if (j == count) { + found = i - count + 1; + break; + } + } + if (found == NO_IRQ) { + raw_spin_unlock_irqrestore(&irq_big_lock, flags); + return NO_IRQ; + } + hint_found: + for (i = found; i < (found + count); i++) { + irq_map[i].hwirq = host->inval_irq; + smp_wmb(); + irq_map[i].host = host; + } + raw_spin_unlock_irqrestore(&irq_big_lock, flags); + return found; +} + +void irq_free_virt(unsigned int virq, unsigned int count) +{ + unsigned long flags; + unsigned int i; + + WARN_ON(virq < NR_PRIORITY_IRQS); + WARN_ON(count == 0 || (virq + count) > irq_virq_count); + + if (virq < NR_PRIORITY_IRQS) { + if (virq + count < NR_PRIORITY_IRQS) + return; + count -= NR_PRIORITY_IRQS - virq; + virq = NR_PRIORITY_IRQS; + } + + if (count > irq_virq_count || virq > irq_virq_count - count) { + if (virq > irq_virq_count) + return; + count = irq_virq_count - virq; + } + + raw_spin_lock_irqsave(&irq_big_lock, flags); + for (i = virq; i < (virq + count); i++) { + struct irq_host *host; + + host = irq_map[i].host; + irq_map[i].hwirq = host->inval_irq; + smp_wmb(); + irq_map[i].host = NULL; + } + raw_spin_unlock_irqrestore(&irq_big_lock, flags); +} + +#ifdef CONFIG_VIRQ_DEBUG +static int virq_debug_show(struct seq_file *m, void *private) +{ + unsigned long flags; + struct irq_desc *desc; + const char *p; + static const char none[] = "none"; + void *data; + int i; + + seq_printf(m, "%-5s %-7s %-15s %-18s %s\n", "virq", "hwirq", + "chip name", "chip data", "host name"); + + for (i = 1; i < nr_irqs; i++) { + desc = irq_to_desc(i); + if (!desc) + continue; + + raw_spin_lock_irqsave(&desc->lock, flags); + + if (desc->action && desc->action->handler) { + struct irq_chip *chip; + + seq_printf(m, "%5d ", i); + seq_printf(m, "0x%05lx ", irq_map[i].hwirq); + + chip = irq_desc_get_chip(desc); + if (chip && chip->name) + p = chip->name; + else + p = none; + seq_printf(m, "%-15s ", p); + + data = irq_desc_get_chip_data(desc); + seq_printf(m, "0x%16p ", data); + + if (irq_map[i].host && irq_map[i].host->of_node) + p = irq_map[i].host->of_node->full_name; + else + p = none; + seq_printf(m, "%s\n", p); + } + + raw_spin_unlock_irqrestore(&desc->lock, flags); + } + + return 0; +} + +static int virq_debug_open(struct inode *inode, struct file *file) +{ + return single_open(file, virq_debug_show, inode->i_private); +} + +static const struct file_operations virq_debug_fops = { + .open = virq_debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static int __init irq_debugfs_init(void) +{ + if (debugfs_create_file("virq_mapping", S_IRUGO, powerpc_debugfs_root, + NULL, &virq_debug_fops) == NULL) + return -ENOMEM; + + return 0; +} +device_initcall(irq_debugfs_init); +#endif /* CONFIG_VIRQ_DEBUG */ diff --git a/arch/c6x/kernel/module.c b/arch/c6x/kernel/module.c new file mode 100644 index 0000000..5fc03f1 --- /dev/null +++ b/arch/c6x/kernel/module.c @@ -0,0 +1,123 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2005, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Thomas Charleux (thomas.charleux@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ +#include <linux/moduleloader.h> +#include <linux/elf.h> +#include <linux/vmalloc.h> +#include <linux/kernel.h> + +static inline int fixup_pcr(u32 *ip, Elf32_Addr dest, u32 maskbits, int shift) +{ + u32 opcode; + long ep = (long)ip & ~31; + long delta = ((long)dest - ep) >> 2; + long mask = (1 << maskbits) - 1; + + if ((delta >> (maskbits - 1)) == 0 || + (delta >> (maskbits - 1)) == -1) { + opcode = *ip; + opcode &= ~(mask << shift); + opcode |= ((delta & mask) << shift); + *ip = opcode; + + pr_debug("REL PCR_S%d[%p] dest[%p] opcode[%08x]\n", + maskbits, ip, (void *)dest, opcode); + + return 0; + } + pr_err("PCR_S%d reloc %p -> %p out of range!\n", + maskbits, ip, (void *)dest); + + return -1; +} + +/* + * apply a RELA relocation + */ +int apply_relocate_add(Elf32_Shdr *sechdrs, + const char *strtab, + unsigned int symindex, + unsigned int relsec, + struct module *me) +{ + Elf32_Rela *rel = (void *) sechdrs[relsec].sh_addr; + Elf_Sym *sym; + u32 *location, opcode; + unsigned int i; + Elf32_Addr v; + Elf_Addr offset = 0; + + pr_debug("Applying relocate section %u to %u with offset 0x%x\n", + relsec, sechdrs[relsec].sh_info, offset); + + for (i = 0; i < sechdrs[relsec].sh_size / sizeof(*rel); i++) { + /* This is where to make the change */ + location = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr + + rel[i].r_offset - offset; + + /* This is the symbol it is referring to. Note that all + undefined symbols have been resolved. */ + sym = (Elf_Sym *)sechdrs[symindex].sh_addr + + ELF32_R_SYM(rel[i].r_info); + + /* this is the adjustment to be made */ + v = sym->st_value + rel[i].r_addend; + + switch (ELF32_R_TYPE(rel[i].r_info)) { + case R_C6000_ABS32: + pr_debug("RELA ABS32: [%p] = 0x%x\n", location, v); + *location = v; + break; + case R_C6000_ABS16: + pr_debug("RELA ABS16: [%p] = 0x%x\n", location, v); + *(u16 *)location = v; + break; + case R_C6000_ABS8: + pr_debug("RELA ABS8: [%p] = 0x%x\n", location, v); + *(u8 *)location = v; + break; + case R_C6000_ABS_L16: + opcode = *location; + opcode &= ~0x7fff80; + opcode |= ((v & 0xffff) << 7); + pr_debug("RELA ABS_L16[%p] v[0x%x] opcode[0x%x]\n", + location, v, opcode); + *location = opcode; + break; + case R_C6000_ABS_H16: + opcode = *location; + opcode &= ~0x7fff80; + opcode |= ((v >> 9) & 0x7fff80); + pr_debug("RELA ABS_H16[%p] v[0x%x] opcode[0x%x]\n", + location, v, opcode); + *location = opcode; + break; + case R_C6000_PCR_S21: + if (fixup_pcr(location, v, 21, 7)) + return -ENOEXEC; + break; + case R_C6000_PCR_S12: + if (fixup_pcr(location, v, 12, 16)) + return -ENOEXEC; + break; + case R_C6000_PCR_S10: + if (fixup_pcr(location, v, 10, 13)) + return -ENOEXEC; + break; + default: + pr_err("module %s: Unknown RELA relocation: %u\n", + me->name, ELF32_R_TYPE(rel[i].r_info)); + return -ENOEXEC; + } + } + + return 0; +} diff --git a/arch/c6x/kernel/process.c b/arch/c6x/kernel/process.c new file mode 100644 index 0000000..7ca8c41 --- /dev/null +++ b/arch/c6x/kernel/process.c @@ -0,0 +1,265 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2006, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ +#include <linux/module.h> +#include <linux/unistd.h> +#include <linux/ptrace.h> +#include <linux/init_task.h> +#include <linux/tick.h> +#include <linux/mqueue.h> +#include <linux/syscalls.h> +#include <linux/reboot.h> + +#include <asm/syscalls.h> + +/* hooks for board specific support */ +void (*c6x_restart)(void); +void (*c6x_halt)(void); + +extern asmlinkage void ret_from_fork(void); + +static struct signal_struct init_signals = INIT_SIGNALS(init_signals); +static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand); + +/* + * Initial thread structure. + */ +union thread_union init_thread_union __init_task_data = { + INIT_THREAD_INFO(init_task) +}; + +/* + * Initial task structure. + */ +struct task_struct init_task = INIT_TASK(init_task); +EXPORT_SYMBOL(init_task); + +/* + * power off function, if any + */ +void (*pm_power_off)(void); +EXPORT_SYMBOL(pm_power_off); + +static void c6x_idle(void) +{ + unsigned long tmp; + + /* + * Put local_irq_enable and idle in same execute packet + * to make them atomic and avoid race to idle with + * interrupts enabled. + */ + asm volatile (" mvc .s2 CSR,%0\n" + " or .d2 1,%0,%0\n" + " mvc .s2 %0,CSR\n" + "|| idle\n" + : "=b"(tmp)); +} + +/* + * The idle loop for C64x + */ +void cpu_idle(void) +{ + /* endless idle loop with no priority at all */ + while (1) { + tick_nohz_idle_enter(); + rcu_idle_enter(); + while (1) { + local_irq_disable(); + if (need_resched()) { + local_irq_enable(); + break; + } + c6x_idle(); /* enables local irqs */ + } + rcu_idle_exit(); + tick_nohz_idle_exit(); + + preempt_enable_no_resched(); + schedule(); + preempt_disable(); + } +} + +static void halt_loop(void) +{ + printk(KERN_EMERG "System Halted, OK to turn off power\n"); + local_irq_disable(); + while (1) + asm volatile("idle\n"); +} + +void machine_restart(char *__unused) +{ + if (c6x_restart) + c6x_restart(); + halt_loop(); +} + +void machine_halt(void) +{ + if (c6x_halt) + c6x_halt(); + halt_loop(); +} + +void machine_power_off(void) +{ + if (pm_power_off) + pm_power_off(); + halt_loop(); +} + +static void kernel_thread_helper(int dummy, void *arg, int (*fn)(void *)) +{ + do_exit(fn(arg)); +} + +/* + * Create a kernel thread + */ +int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) +{ + struct pt_regs regs; + + /* + * copy_thread sets a4 to zero (child return from fork) + * so we can't just set things up to directly return to + * fn. + */ + memset(®s, 0, sizeof(regs)); + regs.b4 = (unsigned long) arg; + regs.a6 = (unsigned long) fn; + regs.pc = (unsigned long) kernel_thread_helper; + local_save_flags(regs.csr); + regs.csr |= 1; + regs.tsr = 5; /* Set GEE and GIE in TSR */ + + /* Ok, create the new process.. */ + return do_fork(flags | CLONE_VM | CLONE_UNTRACED, -1, ®s, + 0, NULL, NULL); +} +EXPORT_SYMBOL(kernel_thread); + +void flush_thread(void) +{ +} + +void exit_thread(void) +{ +} + +SYSCALL_DEFINE1(c6x_clone, struct pt_regs *, regs) +{ + unsigned long clone_flags; + unsigned long newsp; + + /* syscall puts clone_flags in A4 and usp in B4 */ + clone_flags = regs->orig_a4; + if (regs->b4) + newsp = regs->b4; + else + newsp = regs->sp; + + return do_fork(clone_flags, newsp, regs, 0, (int __user *)regs->a6, + (int __user *)regs->b6); +} + +/* + * Do necessary setup to start up a newly executed thread. + */ +void start_thread(struct pt_regs *regs, unsigned int pc, unsigned long usp) +{ + /* + * The binfmt loader will setup a "full" stack, but the C6X + * operates an "empty" stack. So we adjust the usp so that + * argc doesn't get destroyed if an interrupt is taken before + * it is read from the stack. + * + * NB: Library startup code needs to match this. + */ + usp -= 8; + + set_fs(USER_DS); + regs->pc = pc; + regs->sp = usp; + regs->tsr |= 0x40; /* set user mode */ + current->thread.usp = usp; +} + +/* + * Copy a new thread context in its stack. + */ +int copy_thread(unsigned long clone_flags, unsigned long usp, + unsigned long ustk_size, + struct task_struct *p, struct pt_regs *regs) +{ + struct pt_regs *childregs; + + childregs = task_pt_regs(p); + + *childregs = *regs; + childregs->a4 = 0; + + if (usp == -1) + /* case of __kernel_thread: we return to supervisor space */ + childregs->sp = (unsigned long)(childregs + 1); + else + /* Otherwise use the given stack */ + childregs->sp = usp; + + /* Set usp/ksp */ + p->thread.usp = childregs->sp; + /* switch_to uses stack to save/restore 14 callee-saved regs */ + thread_saved_ksp(p) = (unsigned long)childregs - 8; + p->thread.pc = (unsigned int) ret_from_fork; + p->thread.wchan = (unsigned long) ret_from_fork; +#ifdef __DSBT__ + { + unsigned long dp; + + asm volatile ("mv .S2 b14,%0\n" : "=b"(dp)); + + thread_saved_dp(p) = dp; + if (usp == -1) + childregs->dp = dp; + } +#endif + return 0; +} + +/* + * c6x_execve() executes a new program. + */ +SYSCALL_DEFINE4(c6x_execve, const char __user *, name, + const char __user *const __user *, argv, + const char __user *const __user *, envp, + struct pt_regs *, regs) +{ + int error; + char *filename; + + filename = getname(name); + error = PTR_ERR(filename); + if (IS_ERR(filename)) + goto out; + + error = do_execve(filename, argv, envp, regs); + putname(filename); +out: + return error; +} + +unsigned long get_wchan(struct task_struct *p) +{ + return p->thread.wchan; +} diff --git a/arch/c6x/kernel/ptrace.c b/arch/c6x/kernel/ptrace.c new file mode 100644 index 0000000..3c494e8 --- /dev/null +++ b/arch/c6x/kernel/ptrace.c @@ -0,0 +1,187 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2006, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * Updated for 2.6.34: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/ptrace.h> +#include <linux/tracehook.h> +#include <linux/regset.h> +#include <linux/elf.h> + +#include <asm/cacheflush.h> + +#define PT_REG_SIZE (sizeof(struct pt_regs)) + +/* + * Called by kernel/ptrace.c when detaching. + */ +void ptrace_disable(struct task_struct *child) +{ + /* nothing to do */ +} + +/* + * Get a register number from live pt_regs for the specified task. + */ +static inline long get_reg(struct task_struct *task, int regno) +{ + long *addr = (long *)task_pt_regs(task); + + if (regno == PT_TSR || regno == PT_CSR) + return 0; + + return addr[regno]; +} + +/* + * Write contents of register REGNO in task TASK. + */ +static inline int put_reg(struct task_struct *task, + int regno, + unsigned long data) +{ + unsigned long *addr = (unsigned long *)task_pt_regs(task); + + if (regno != PT_TSR && regno != PT_CSR) + addr[regno] = data; + + return 0; +} + +/* regset get/set implementations */ + +static int gpr_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + struct pt_regs *regs = task_pt_regs(target); + + return user_regset_copyout(&pos, &count, &kbuf, &ubuf, + regs, + 0, sizeof(*regs)); +} + +static int gpr_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + int ret; + struct pt_regs *regs = task_pt_regs(target); + + /* Don't copyin TSR or CSR */ + ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf, + ®s, + 0, PT_TSR * sizeof(long)); + if (ret) + return ret; + + ret = user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf, + PT_TSR * sizeof(long), + (PT_TSR + 1) * sizeof(long)); + if (ret) + return ret; + + ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf, + ®s, + (PT_TSR + 1) * sizeof(long), + PT_CSR * sizeof(long)); + if (ret) + return ret; + + ret = user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf, + PT_CSR * sizeof(long), + (PT_CSR + 1) * sizeof(long)); + if (ret) + return ret; + + ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf, + ®s, + (PT_CSR + 1) * sizeof(long), -1); + return ret; +} + +enum c6x_regset { + REGSET_GPR, +}; + +static const struct user_regset c6x_regsets[] = { + [REGSET_GPR] = { + .core_note_type = NT_PRSTATUS, + .n = ELF_NGREG, + .size = sizeof(u32), + .align = sizeof(u32), + .get = gpr_get, + .set = gpr_set + }, +}; + +static const struct user_regset_view user_c6x_native_view = { + .name = "tic6x", + .e_machine = EM_TI_C6000, + .regsets = c6x_regsets, + .n = ARRAY_SIZE(c6x_regsets), +}; + +const struct user_regset_view *task_user_regset_view(struct task_struct *task) +{ + return &user_c6x_native_view; +} + +/* + * Perform ptrace request + */ +long arch_ptrace(struct task_struct *child, long request, + unsigned long addr, unsigned long data) +{ + int ret = 0; + + switch (request) { + /* + * write the word at location addr. + */ + case PTRACE_POKETEXT: + ret = generic_ptrace_pokedata(child, addr, data); + if (ret == 0 && request == PTRACE_POKETEXT) + flush_icache_range(addr, addr + 4); + break; + default: + ret = ptrace_request(child, request, addr, data); + break; + } + + return ret; +} + +/* + * handle tracing of system call entry + * - return the revised system call number or ULONG_MAX to cause ENOSYS + */ +asmlinkage unsigned long syscall_trace_entry(struct pt_regs *regs) +{ + if (tracehook_report_syscall_entry(regs)) + /* tracing decided this syscall should not happen, so + * We'll return a bogus call number to get an ENOSYS + * error, but leave the original number in + * regs->orig_a4 + */ + return ULONG_MAX; + + return regs->b0; +} + +/* + * handle tracing of system call exit + */ +asmlinkage void syscall_trace_exit(struct pt_regs *regs) +{ + tracehook_report_syscall_exit(regs, 0); +} diff --git a/arch/c6x/kernel/setup.c b/arch/c6x/kernel/setup.c new file mode 100644 index 0000000..0c07921 --- /dev/null +++ b/arch/c6x/kernel/setup.c @@ -0,0 +1,510 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2006, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/dma-mapping.h> +#include <linux/memblock.h> +#include <linux/seq_file.h> +#include <linux/bootmem.h> +#include <linux/clkdev.h> +#include <linux/initrd.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/of_fdt.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/cache.h> +#include <linux/delay.h> +#include <linux/sched.h> +#include <linux/clk.h> +#include <linux/cpu.h> +#include <linux/fs.h> +#include <linux/of.h> + + +#include <asm/sections.h> +#include <asm/div64.h> +#include <asm/setup.h> +#include <asm/dscr.h> +#include <asm/clock.h> +#include <asm/soc.h> + +static const char *c6x_soc_name; + +int c6x_num_cores; +EXPORT_SYMBOL_GPL(c6x_num_cores); + +unsigned int c6x_silicon_rev; +EXPORT_SYMBOL_GPL(c6x_silicon_rev); + +/* + * Device status register. This holds information + * about device configuration needed by some drivers. + */ +unsigned int c6x_devstat; +EXPORT_SYMBOL_GPL(c6x_devstat); + +/* + * Some SoCs have fuse registers holding a unique MAC + * address. This is parsed out of the device tree with + * the resulting MAC being held here. + */ +unsigned char c6x_fuse_mac[6]; + +unsigned long memory_start; +unsigned long memory_end; + +unsigned long ram_start; +unsigned long ram_end; + +/* Uncached memory for DMA consistent use (memdma=) */ +static unsigned long dma_start __initdata; +static unsigned long dma_size __initdata; + +char c6x_command_line[COMMAND_LINE_SIZE]; + +#if defined(CONFIG_CMDLINE_BOOL) +static const char default_command_line[COMMAND_LINE_SIZE] __section(.cmdline) = + CONFIG_CMDLINE; +#endif + +struct cpuinfo_c6x { + const char *cpu_name; + const char *cpu_voltage; + const char *mmu; + const char *fpu; + char *cpu_rev; + unsigned int core_id; + char __cpu_rev[5]; +}; + +static DEFINE_PER_CPU(struct cpuinfo_c6x, cpu_data); + +unsigned int ticks_per_ns_scaled; +EXPORT_SYMBOL(ticks_per_ns_scaled); + +unsigned int c6x_core_freq; + +static void __init get_cpuinfo(void) +{ + unsigned cpu_id, rev_id, csr; + struct clk *coreclk = clk_get_sys(NULL, "core"); + unsigned long core_khz; + u64 tmp; + struct cpuinfo_c6x *p; + struct device_node *node, *np; + + p = &per_cpu(cpu_data, smp_processor_id()); + + if (!IS_ERR(coreclk)) + c6x_core_freq = clk_get_rate(coreclk); + else { + printk(KERN_WARNING + "Cannot find core clock frequency. Using 700MHz\n"); + c6x_core_freq = 700000000; + } + + core_khz = c6x_core_freq / 1000; + + tmp = (uint64_t)core_khz << C6X_NDELAY_SCALE; + do_div(tmp, 1000000); + ticks_per_ns_scaled = tmp; + + csr = get_creg(CSR); + cpu_id = csr >> 24; + rev_id = (csr >> 16) & 0xff; + + p->mmu = "none"; + p->fpu = "none"; + p->cpu_voltage = "unknown"; + + switch (cpu_id) { + case 0: + p->cpu_name = "C67x"; + p->fpu = "yes"; + break; + case 2: + p->cpu_name = "C62x"; + break; + case 8: + p->cpu_name = "C64x"; + break; + case 12: + p->cpu_name = "C64x"; + break; + case 16: + p->cpu_name = "C64x+"; + p->cpu_voltage = "1.2"; + break; + default: + p->cpu_name = "unknown"; + break; + } + + if (cpu_id < 16) { + switch (rev_id) { + case 0x1: + if (cpu_id > 8) { + p->cpu_rev = "DM640/DM641/DM642/DM643"; + p->cpu_voltage = "1.2 - 1.4"; + } else { + p->cpu_rev = "C6201"; + p->cpu_voltage = "2.5"; + } + break; + case 0x2: + p->cpu_rev = "C6201B/C6202/C6211"; + p->cpu_voltage = "1.8"; + break; + case 0x3: + p->cpu_rev = "C6202B/C6203/C6204/C6205"; + p->cpu_voltage = "1.5"; + break; + case 0x201: + p->cpu_rev = "C6701 revision 0 (early CPU)"; + p->cpu_voltage = "1.8"; + break; + case 0x202: + p->cpu_rev = "C6701/C6711/C6712"; + p->cpu_voltage = "1.8"; + break; + case 0x801: + p->cpu_rev = "C64x"; + p->cpu_voltage = "1.5"; + break; + default: + p->cpu_rev = "unknown"; + } + } else { + p->cpu_rev = p->__cpu_rev; + snprintf(p->__cpu_rev, sizeof(p->__cpu_rev), "0x%x", cpu_id); + } + + p->core_id = get_coreid(); + + node = of_find_node_by_name(NULL, "cpus"); + if (node) { + for_each_child_of_node(node, np) + if (!strcmp("cpu", np->name)) + ++c6x_num_cores; + of_node_put(node); + } + + node = of_find_node_by_name(NULL, "soc"); + if (node) { + if (of_property_read_string(node, "model", &c6x_soc_name)) + c6x_soc_name = "unknown"; + of_node_put(node); + } else + c6x_soc_name = "unknown"; + + printk(KERN_INFO "CPU%d: %s rev %s, %s volts, %uMHz\n", + p->core_id, p->cpu_name, p->cpu_rev, + p->cpu_voltage, c6x_core_freq / 1000000); +} + +/* + * Early parsing of the command line + */ +static u32 mem_size __initdata; + +/* "mem=" parsing. */ +static int __init early_mem(char *p) +{ + if (!p) + return -EINVAL; + + mem_size = memparse(p, &p); + /* don't remove all of memory when handling "mem={invalid}" */ + if (mem_size == 0) + return -EINVAL; + + return 0; +} +early_param("mem", early_mem); + +/* "memdma=<size>[@<address>]" parsing. */ +static int __init early_memdma(char *p) +{ + if (!p) + return -EINVAL; + + dma_size = memparse(p, &p); + if (*p == '@') + dma_start = memparse(p, &p); + + return 0; +} +early_param("memdma", early_memdma); + +int __init c6x_add_memory(phys_addr_t start, unsigned long size) +{ + static int ram_found __initdata; + + /* We only handle one bank (the one with PAGE_OFFSET) for now */ + if (ram_found) + return -EINVAL; + + if (start > PAGE_OFFSET || PAGE_OFFSET >= (start + size)) + return 0; + + ram_start = start; + ram_end = start + size; + + ram_found = 1; + return 0; +} + +/* + * Do early machine setup and device tree parsing. This is called very + * early on the boot process. + */ +notrace void __init machine_init(unsigned long dt_ptr) +{ + struct boot_param_header *dtb = __va(dt_ptr); + struct boot_param_header *fdt = (struct boot_param_header *)_fdt_start; + + /* interrupts must be masked */ + set_creg(IER, 2); + + /* + * Set the Interrupt Service Table (IST) to the beginning of the + * vector table. + */ + set_ist(_vectors_start); + + lockdep_init(); + + /* + * dtb is passed in from bootloader. + * fdt is linked in blob. + */ + if (dtb && dtb != fdt) + fdt = dtb; + + /* Do some early initialization based on the flat device tree */ + early_init_devtree(fdt); + + /* parse_early_param needs a boot_command_line */ + strlcpy(boot_command_line, c6x_command_line, COMMAND_LINE_SIZE); + parse_early_param(); +} + +void __init setup_arch(char **cmdline_p) +{ + int bootmap_size; + struct memblock_region *reg; + + printk(KERN_INFO "Initializing kernel\n"); + + /* Initialize command line */ + *cmdline_p = c6x_command_line; + + memory_end = ram_end; + memory_end &= ~(PAGE_SIZE - 1); + + if (mem_size && (PAGE_OFFSET + PAGE_ALIGN(mem_size)) < memory_end) + memory_end = PAGE_OFFSET + PAGE_ALIGN(mem_size); + + /* add block that this kernel can use */ + memblock_add(PAGE_OFFSET, memory_end - PAGE_OFFSET); + + /* reserve kernel text/data/bss */ + memblock_reserve(PAGE_OFFSET, + PAGE_ALIGN((unsigned long)&_end - PAGE_OFFSET)); + + if (dma_size) { + /* align to cacheability granularity */ + dma_size = CACHE_REGION_END(dma_size); + + if (!dma_start) + dma_start = memory_end - dma_size; + + /* align to cacheability granularity */ + dma_start = CACHE_REGION_START(dma_start); + + /* reserve DMA memory taken from kernel memory */ + if (memblock_is_region_memory(dma_start, dma_size)) + memblock_reserve(dma_start, dma_size); + } + + memory_start = PAGE_ALIGN((unsigned int) &_end); + + printk(KERN_INFO "Memory Start=%08lx, Memory End=%08lx\n", + memory_start, memory_end); + +#ifdef CONFIG_BLK_DEV_INITRD + /* + * Reserve initrd memory if in kernel memory. + */ + if (initrd_start < initrd_end) + if (memblock_is_region_memory(initrd_start, + initrd_end - initrd_start)) + memblock_reserve(initrd_start, + initrd_end - initrd_start); +#endif + + init_mm.start_code = (unsigned long) &_stext; + init_mm.end_code = (unsigned long) &_etext; + init_mm.end_data = memory_start; + init_mm.brk = memory_start; + + /* + * Give all the memory to the bootmap allocator, tell it to put the + * boot mem_map at the start of memory + */ + bootmap_size = init_bootmem_node(NODE_DATA(0), + memory_start >> PAGE_SHIFT, + PAGE_OFFSET >> PAGE_SHIFT, + memory_end >> PAGE_SHIFT); + memblock_reserve(memory_start, bootmap_size); + + unflatten_device_tree(); + + c6x_cache_init(); + + /* Set the whole external memory as non-cacheable */ + disable_caching(ram_start, ram_end - 1); + + /* Set caching of external RAM used by Linux */ + for_each_memblock(memory, reg) + enable_caching(CACHE_REGION_START(reg->base), + CACHE_REGION_START(reg->base + reg->size - 1)); + +#ifdef CONFIG_BLK_DEV_INITRD + /* + * Enable caching for initrd which falls outside kernel memory. + */ + if (initrd_start < initrd_end) { + if (!memblock_is_region_memory(initrd_start, + initrd_end - initrd_start)) + enable_caching(CACHE_REGION_START(initrd_start), + CACHE_REGION_START(initrd_end - 1)); + } +#endif + + /* + * Disable caching for dma coherent memory taken from kernel memory. + */ + if (dma_size && memblock_is_region_memory(dma_start, dma_size)) + disable_caching(dma_start, + CACHE_REGION_START(dma_start + dma_size - 1)); + + /* Initialize the coherent memory allocator */ + coherent_mem_init(dma_start, dma_size); + + /* + * Free all memory as a starting point. + */ + free_bootmem(PAGE_OFFSET, memory_end - PAGE_OFFSET); + + /* + * Then reserve memory which is already being used. + */ + for_each_memblock(reserved, reg) { + pr_debug("reserved - 0x%08x-0x%08x\n", + (u32) reg->base, (u32) reg->size); + reserve_bootmem(reg->base, reg->size, BOOTMEM_DEFAULT); + } + + max_low_pfn = PFN_DOWN(memory_end); + min_low_pfn = PFN_UP(memory_start); + max_mapnr = max_low_pfn - min_low_pfn; + + /* Get kmalloc into gear */ + paging_init(); + + /* + * Probe for Device State Configuration Registers. + * We have to do this early in case timer needs to be enabled + * through DSCR. + */ + dscr_probe(); + + /* We do this early for timer and core clock frequency */ + c64x_setup_clocks(); + + /* Get CPU info */ + get_cpuinfo(); + +#if defined(CONFIG_VT) && defined(CONFIG_DUMMY_CONSOLE) + conswitchp = &dummy_con; +#endif +} + +#define cpu_to_ptr(n) ((void *)((long)(n)+1)) +#define ptr_to_cpu(p) ((long)(p) - 1) + +static int show_cpuinfo(struct seq_file *m, void *v) +{ + int n = ptr_to_cpu(v); + struct cpuinfo_c6x *p = &per_cpu(cpu_data, n); + + if (n == 0) { + seq_printf(m, + "soc\t\t: %s\n" + "soc revision\t: 0x%x\n" + "soc cores\t: %d\n", + c6x_soc_name, c6x_silicon_rev, c6x_num_cores); + } + + seq_printf(m, + "\n" + "processor\t: %d\n" + "cpu\t\t: %s\n" + "core revision\t: %s\n" + "core voltage\t: %s\n" + "core id\t\t: %d\n" + "mmu\t\t: %s\n" + "fpu\t\t: %s\n" + "cpu MHz\t\t: %u\n" + "bogomips\t: %lu.%02lu\n\n", + n, + p->cpu_name, p->cpu_rev, p->cpu_voltage, + p->core_id, p->mmu, p->fpu, + (c6x_core_freq + 500000) / 1000000, + (loops_per_jiffy/(500000/HZ)), + (loops_per_jiffy/(5000/HZ))%100); + + return 0; +} + +static void *c_start(struct seq_file *m, loff_t *pos) +{ + return *pos < nr_cpu_ids ? cpu_to_ptr(*pos) : NULL; +} +static void *c_next(struct seq_file *m, void *v, loff_t *pos) +{ + ++*pos; + return NULL; +} +static void c_stop(struct seq_file *m, void *v) +{ +} + +const struct seq_operations cpuinfo_op = { + c_start, + c_stop, + c_next, + show_cpuinfo +}; + +static struct cpu cpu_devices[NR_CPUS]; + +static int __init topology_init(void) +{ + int i; + + for_each_present_cpu(i) + register_cpu(&cpu_devices[i], i); + + return 0; +} + +subsys_initcall(topology_init); diff --git a/arch/c6x/kernel/signal.c b/arch/c6x/kernel/signal.c new file mode 100644 index 0000000..304f675 --- /dev/null +++ b/arch/c6x/kernel/signal.c @@ -0,0 +1,377 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2006, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * Updated for 2.6.34: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/kernel.h> +#include <linux/uaccess.h> +#include <linux/syscalls.h> +#include <linux/tracehook.h> + +#include <asm/ucontext.h> +#include <asm/cacheflush.h> + + +#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) + +/* + * Do a signal return, undo the signal stack. + */ + +#define RETCODE_SIZE (9 << 2) /* 9 instructions = 36 bytes */ + +struct rt_sigframe { + struct siginfo __user *pinfo; + void __user *puc; + struct siginfo info; + struct ucontext uc; + unsigned long retcode[RETCODE_SIZE >> 2]; +}; + +static int restore_sigcontext(struct pt_regs *regs, + struct sigcontext __user *sc) +{ + int err = 0; + + /* The access_ok check was done by caller, so use __get_user here */ +#define COPY(x) (err |= __get_user(regs->x, &sc->sc_##x)) + + COPY(sp); COPY(a4); COPY(b4); COPY(a6); COPY(b6); COPY(a8); COPY(b8); + COPY(a0); COPY(a1); COPY(a2); COPY(a3); COPY(a5); COPY(a7); COPY(a9); + COPY(b0); COPY(b1); COPY(b2); COPY(b3); COPY(b5); COPY(b7); COPY(b9); + + COPY(a16); COPY(a17); COPY(a18); COPY(a19); + COPY(a20); COPY(a21); COPY(a22); COPY(a23); + COPY(a24); COPY(a25); COPY(a26); COPY(a27); + COPY(a28); COPY(a29); COPY(a30); COPY(a31); + COPY(b16); COPY(b17); COPY(b18); COPY(b19); + COPY(b20); COPY(b21); COPY(b22); COPY(b23); + COPY(b24); COPY(b25); COPY(b26); COPY(b27); + COPY(b28); COPY(b29); COPY(b30); COPY(b31); + + COPY(csr); COPY(pc); + +#undef COPY + + return err; +} + +asmlinkage int do_rt_sigreturn(struct pt_regs *regs) +{ + struct rt_sigframe __user *frame; + sigset_t set; + + /* + * Since we stacked the signal on a dword boundary, + * 'sp' should be dword aligned here. If it's + * not, then the user is trying to mess with us. + */ + if (regs->sp & 7) + goto badframe; + + frame = (struct rt_sigframe __user *) ((unsigned long) regs->sp + 8); + + if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) + goto badframe; + if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set))) + goto badframe; + + sigdelsetmask(&set, ~_BLOCKABLE); + spin_lock_irq(¤t->sighand->siglock); + current->blocked = set; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + if (restore_sigcontext(regs, &frame->uc.uc_mcontext)) + goto badframe; + + return regs->a4; + +badframe: + force_sig(SIGSEGV, current); + return 0; +} + +static int setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs, + unsigned long mask) +{ + int err = 0; + + err |= __put_user(mask, &sc->sc_mask); + + /* The access_ok check was done by caller, so use __put_user here */ +#define COPY(x) (err |= __put_user(regs->x, &sc->sc_##x)) + + COPY(sp); COPY(a4); COPY(b4); COPY(a6); COPY(b6); COPY(a8); COPY(b8); + COPY(a0); COPY(a1); COPY(a2); COPY(a3); COPY(a5); COPY(a7); COPY(a9); + COPY(b0); COPY(b1); COPY(b2); COPY(b3); COPY(b5); COPY(b7); COPY(b9); + + COPY(a16); COPY(a17); COPY(a18); COPY(a19); + COPY(a20); COPY(a21); COPY(a22); COPY(a23); + COPY(a24); COPY(a25); COPY(a26); COPY(a27); + COPY(a28); COPY(a29); COPY(a30); COPY(a31); + COPY(b16); COPY(b17); COPY(b18); COPY(b19); + COPY(b20); COPY(b21); COPY(b22); COPY(b23); + COPY(b24); COPY(b25); COPY(b26); COPY(b27); + COPY(b28); COPY(b29); COPY(b30); COPY(b31); + + COPY(csr); COPY(pc); + +#undef COPY + + return err; +} + +static inline void __user *get_sigframe(struct k_sigaction *ka, + struct pt_regs *regs, + unsigned long framesize) +{ + unsigned long sp = regs->sp; + + /* + * This is the X/Open sanctioned signal stack switching. + */ + if ((ka->sa.sa_flags & SA_ONSTACK) && sas_ss_flags(sp) == 0) + sp = current->sas_ss_sp + current->sas_ss_size; + + /* + * No matter what happens, 'sp' must be dword + * aligned. Otherwise, nasty things will happen + */ + return (void __user *)((sp - framesize) & ~7); +} + +static int setup_rt_frame(int signr, struct k_sigaction *ka, siginfo_t *info, + sigset_t *set, struct pt_regs *regs) +{ + struct rt_sigframe __user *frame; + unsigned long __user *retcode; + int err = 0; + + frame = get_sigframe(ka, regs, sizeof(*frame)); + + if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) + goto segv_and_exit; + + err |= __put_user(&frame->info, &frame->pinfo); + err |= __put_user(&frame->uc, &frame->puc); + err |= copy_siginfo_to_user(&frame->info, info); + + /* Clear all the bits of the ucontext we don't use. */ + err |= __clear_user(&frame->uc, offsetof(struct ucontext, uc_mcontext)); + + err |= setup_sigcontext(&frame->uc.uc_mcontext, regs, set->sig[0]); + err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); + + /* Set up to return from userspace */ + retcode = (unsigned long __user *) &frame->retcode; + + /* The access_ok check was done above, so use __put_user here */ +#define COPY(x) (err |= __put_user(x, retcode++)) + + COPY(0x0000002AUL | (__NR_rt_sigreturn << 7)); + /* MVK __NR_rt_sigreturn,B0 */ + COPY(0x10000000UL); /* SWE */ + COPY(0x00006000UL); /* NOP 4 */ + COPY(0x00006000UL); /* NOP 4 */ + COPY(0x00006000UL); /* NOP 4 */ + COPY(0x00006000UL); /* NOP 4 */ + COPY(0x00006000UL); /* NOP 4 */ + COPY(0x00006000UL); /* NOP 4 */ + COPY(0x00006000UL); /* NOP 4 */ + +#undef COPY + + if (err) + goto segv_and_exit; + + flush_icache_range((unsigned long) &frame->retcode, + (unsigned long) &frame->retcode + RETCODE_SIZE); + + retcode = (unsigned long __user *) &frame->retcode; + + /* Change user context to branch to signal handler */ + regs->sp = (unsigned long) frame - 8; + regs->b3 = (unsigned long) retcode; + regs->pc = (unsigned long) ka->sa.sa_handler; + + /* Give the signal number to the handler */ + regs->a4 = signr; + + /* + * For realtime signals we must also set the second and third + * arguments for the signal handler. + * -- Peter Maydell <pmaydell@chiark.greenend.org.uk> 2000-12-06 + */ + regs->b4 = (unsigned long)&frame->info; + regs->a6 = (unsigned long)&frame->uc; + + return 0; + +segv_and_exit: + force_sigsegv(signr, current); + return -EFAULT; +} + +static inline void +handle_restart(struct pt_regs *regs, struct k_sigaction *ka, int has_handler) +{ + switch (regs->a4) { + case -ERESTARTNOHAND: + if (!has_handler) + goto do_restart; + regs->a4 = -EINTR; + break; + + case -ERESTARTSYS: + if (has_handler && !(ka->sa.sa_flags & SA_RESTART)) { + regs->a4 = -EINTR; + break; + } + /* fallthrough */ + case -ERESTARTNOINTR: +do_restart: + regs->a4 = regs->orig_a4; + regs->pc -= 4; + break; + } +} + +/* + * handle the actual delivery of a signal to userspace + */ +static int handle_signal(int sig, + siginfo_t *info, struct k_sigaction *ka, + sigset_t *oldset, struct pt_regs *regs, + int syscall) +{ + int ret; + + /* Are we from a system call? */ + if (syscall) { + /* If so, check system call restarting.. */ + switch (regs->a4) { + case -ERESTART_RESTARTBLOCK: + case -ERESTARTNOHAND: + regs->a4 = -EINTR; + break; + + case -ERESTARTSYS: + if (!(ka->sa.sa_flags & SA_RESTART)) { + regs->a4 = -EINTR; + break; + } + + /* fallthrough */ + case -ERESTARTNOINTR: + regs->a4 = regs->orig_a4; + regs->pc -= 4; + } + } + + /* Set up the stack frame */ + ret = setup_rt_frame(sig, ka, info, oldset, regs); + if (ret == 0) { + spin_lock_irq(¤t->sighand->siglock); + sigorsets(¤t->blocked, ¤t->blocked, + &ka->sa.sa_mask); + if (!(ka->sa.sa_flags & SA_NODEFER)) + sigaddset(¤t->blocked, sig); + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + } + + return ret; +} + +/* + * handle a potential signal + */ +static void do_signal(struct pt_regs *regs, int syscall) +{ + struct k_sigaction ka; + siginfo_t info; + sigset_t *oldset; + int signr; + + /* we want the common case to go fast, which is why we may in certain + * cases get here from kernel mode */ + if (!user_mode(regs)) + return; + + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + oldset = ¤t->saved_sigmask; + else + oldset = ¤t->blocked; + + signr = get_signal_to_deliver(&info, &ka, regs, NULL); + if (signr > 0) { + if (handle_signal(signr, &info, &ka, oldset, + regs, syscall) == 0) { + /* a signal was successfully delivered; the saved + * sigmask will have been stored in the signal frame, + * and will be restored by sigreturn, so we can simply + * clear the TIF_RESTORE_SIGMASK flag */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) + clear_thread_flag(TIF_RESTORE_SIGMASK); + + tracehook_signal_handler(signr, &info, &ka, regs, 0); + } + + return; + } + + /* did we come from a system call? */ + if (syscall) { + /* restart the system call - no handlers present */ + switch (regs->a4) { + case -ERESTARTNOHAND: + case -ERESTARTSYS: + case -ERESTARTNOINTR: + regs->a4 = regs->orig_a4; + regs->pc -= 4; + break; + + case -ERESTART_RESTARTBLOCK: + regs->a4 = regs->orig_a4; + regs->b0 = __NR_restart_syscall; + regs->pc -= 4; + break; + } + } + + /* if there's no signal to deliver, we just put the saved sigmask + * back */ + if (test_thread_flag(TIF_RESTORE_SIGMASK)) { + clear_thread_flag(TIF_RESTORE_SIGMASK); + sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); + } +} + +/* + * notification of userspace execution resumption + * - triggered by current->work.notify_resume + */ +asmlinkage void do_notify_resume(struct pt_regs *regs, u32 thread_info_flags, + int syscall) +{ + /* deal with pending signal delivery */ + if (thread_info_flags & ((1 << TIF_SIGPENDING) | + (1 << TIF_RESTORE_SIGMASK))) + do_signal(regs, syscall); + + if (thread_info_flags & (1 << TIF_NOTIFY_RESUME)) { + clear_thread_flag(TIF_NOTIFY_RESUME); + tracehook_notify_resume(regs); + if (current->replacement_session_keyring) + key_replace_session_keyring(); + } +} diff --git a/arch/c6x/kernel/soc.c b/arch/c6x/kernel/soc.c new file mode 100644 index 0000000..dd45bc3 --- /dev/null +++ b/arch/c6x/kernel/soc.c @@ -0,0 +1,91 @@ +/* + * Miscellaneous SoC-specific hooks. + * + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/module.h> +#include <linux/ctype.h> +#include <linux/etherdevice.h> +#include <asm/system.h> +#include <asm/setup.h> +#include <asm/soc.h> + +struct soc_ops soc_ops; + +int soc_get_exception(void) +{ + if (!soc_ops.get_exception) + return -1; + return soc_ops.get_exception(); +} + +void soc_assert_event(unsigned int evt) +{ + if (soc_ops.assert_event) + soc_ops.assert_event(evt); +} + +static u8 cmdline_mac[6]; + +static int __init get_mac_addr_from_cmdline(char *str) +{ + int count, i, val; + + for (count = 0; count < 6 && *str; count++, str += 3) { + if (!isxdigit(str[0]) || !isxdigit(str[1])) + return 0; + if (str[2] != ((count < 5) ? ':' : '\0')) + return 0; + + for (i = 0, val = 0; i < 2; i++) { + val = val << 4; + val |= isdigit(str[i]) ? + str[i] - '0' : toupper(str[i]) - 'A' + 10; + } + cmdline_mac[count] = val; + } + return 1; +} +__setup("emac_addr=", get_mac_addr_from_cmdline); + +/* + * Setup the MAC address for SoC ethernet devices. + * + * Before calling this function, the ethernet driver will have + * initialized the addr with local-mac-address from the device + * tree (if found). Allow command line to override, but not + * the fused address. + */ +int soc_mac_addr(unsigned int index, u8 *addr) +{ + int i, have_dt_mac = 0, have_cmdline_mac = 0, have_fuse_mac = 0; + + for (i = 0; i < 6; i++) { + if (cmdline_mac[i]) + have_cmdline_mac = 1; + if (c6x_fuse_mac[i]) + have_fuse_mac = 1; + if (addr[i]) + have_dt_mac = 1; + } + + /* cmdline overrides all */ + if (have_cmdline_mac) + memcpy(addr, cmdline_mac, 6); + else if (!have_dt_mac) { + if (have_fuse_mac) + memcpy(addr, c6x_fuse_mac, 6); + else + random_ether_addr(addr); + } + + /* adjust for specific EMAC device */ + addr[5] += index * c6x_num_cores; + return 1; +} +EXPORT_SYMBOL_GPL(soc_mac_addr); diff --git a/arch/c6x/kernel/switch_to.S b/arch/c6x/kernel/switch_to.S new file mode 100644 index 0000000..09177ed --- /dev/null +++ b/arch/c6x/kernel/switch_to.S @@ -0,0 +1,74 @@ +/* + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter (msalter@redhat.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/linkage.h> +#include <asm/asm-offsets.h> + +#define SP B15 + + /* + * void __switch_to(struct thread_info *prev, + * struct thread_info *next, + * struct task_struct *tsk) ; + */ +ENTRY(__switch_to) + LDDW .D2T2 *+B4(THREAD_B15_14),B7:B6 + || MV .L2X A4,B5 ; prev + || MV .L1X B4,A5 ; next + || MVC .S2 RILC,B1 + + STW .D2T2 B3,*+B5(THREAD_PC) + || STDW .D1T1 A13:A12,*+A4(THREAD_A13_12) + || MVC .S2 ILC,B0 + + LDW .D2T2 *+B4(THREAD_PC),B3 + || LDDW .D1T1 *+A5(THREAD_A13_12),A13:A12 + + STDW .D1T1 A11:A10,*+A4(THREAD_A11_10) + || STDW .D2T2 B1:B0,*+B5(THREAD_RICL_ICL) +#ifndef __DSBT__ + || MVKL .S2 current_ksp,B1 +#endif + + STDW .D2T2 B15:B14,*+B5(THREAD_B15_14) + || STDW .D1T1 A15:A14,*+A4(THREAD_A15_14) +#ifndef __DSBT__ + || MVKH .S2 current_ksp,B1 +#endif + + ;; Switch to next SP + MV .S2 B7,SP +#ifdef __DSBT__ + || STW .D2T2 B7,*+B14(current_ksp) +#else + || STW .D2T2 B7,*B1 + || MV .L2 B6,B14 +#endif + || LDDW .D1T1 *+A5(THREAD_RICL_ICL),A1:A0 + + STDW .D2T2 B11:B10,*+B5(THREAD_B11_10) + || LDDW .D1T1 *+A5(THREAD_A15_14),A15:A14 + + STDW .D2T2 B13:B12,*+B5(THREAD_B13_12) + || LDDW .D1T1 *+A5(THREAD_A11_10),A11:A10 + + B .S2 B3 ; return in next E1 + || LDDW .D2T2 *+B4(THREAD_B13_12),B13:B12 + + LDDW .D2T2 *+B4(THREAD_B11_10),B11:B10 + NOP + + MV .L2X A0,B0 + || MV .S1 A6,A4 + + MVC .S2 B0,ILC + || MV .L2X A1,B1 + + MVC .S2 B1,RILC +ENDPROC(__switch_to) diff --git a/arch/c6x/kernel/sys_c6x.c b/arch/c6x/kernel/sys_c6x.c new file mode 100644 index 0000000..3e9bdfb --- /dev/null +++ b/arch/c6x/kernel/sys_c6x.c @@ -0,0 +1,74 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/module.h> +#include <linux/syscalls.h> +#include <linux/uaccess.h> + +#include <asm/syscalls.h> + +#ifdef CONFIG_ACCESS_CHECK +int _access_ok(unsigned long addr, unsigned long size) +{ + if (!size) + return 1; + + if (!addr || addr > (0xffffffffUL - (size - 1))) + goto _bad_access; + + if (segment_eq(get_fs(), KERNEL_DS)) + return 1; + + if (memory_start <= addr && (addr + size - 1) < memory_end) + return 1; + +_bad_access: + pr_debug("Bad access attempt: pid[%d] addr[%08lx] size[0x%lx]\n", + current->pid, addr, size); + return 0; +} +EXPORT_SYMBOL(_access_ok); +#endif + +/* sys_cache_sync -- sync caches over given range */ +asmlinkage int sys_cache_sync(unsigned long s, unsigned long e) +{ + L1D_cache_block_writeback_invalidate(s, e); + L1P_cache_block_invalidate(s, e); + + return 0; +} + +/* Provide the actual syscall number to call mapping. */ +#undef __SYSCALL +#define __SYSCALL(nr, call) [nr] = (call), + +/* + * Use trampolines + */ +#define sys_pread64 sys_pread_c6x +#define sys_pwrite64 sys_pwrite_c6x +#define sys_truncate64 sys_truncate64_c6x +#define sys_ftruncate64 sys_ftruncate64_c6x +#define sys_fadvise64 sys_fadvise64_c6x +#define sys_fadvise64_64 sys_fadvise64_64_c6x +#define sys_fallocate sys_fallocate_c6x + +/* Use sys_mmap_pgoff directly */ +#define sys_mmap2 sys_mmap_pgoff + +/* + * Note that we can't include <linux/unistd.h> here since the header + * guard will defeat us; <asm/unistd.h> checks for __SYSCALL as well. + */ +void *sys_call_table[__NR_syscalls] = { + [0 ... __NR_syscalls-1] = sys_ni_syscall, +#include <asm/unistd.h> +}; diff --git a/arch/c6x/kernel/time.c b/arch/c6x/kernel/time.c new file mode 100644 index 0000000..4c9f136 --- /dev/null +++ b/arch/c6x/kernel/time.c @@ -0,0 +1,65 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/kernel.h> +#include <linux/clocksource.h> +#include <linux/errno.h> +#include <linux/sched.h> +#include <linux/param.h> +#include <linux/string.h> +#include <linux/mm.h> +#include <linux/interrupt.h> +#include <linux/timex.h> +#include <linux/profile.h> + +#include <asm/timer64.h> + +static u32 sched_clock_multiplier; +#define SCHED_CLOCK_SHIFT 16 + +static cycle_t tsc_read(struct clocksource *cs) +{ + return get_cycles(); +} + +static struct clocksource clocksource_tsc = { + .name = "timestamp", + .rating = 300, + .read = tsc_read, + .mask = CLOCKSOURCE_MASK(64), + .flags = CLOCK_SOURCE_IS_CONTINUOUS, +}; + +/* + * scheduler clock - returns current time in nanoseconds. + */ +u64 sched_clock(void) +{ + u64 tsc = get_cycles(); + + return (tsc * sched_clock_multiplier) >> SCHED_CLOCK_SHIFT; +} + +void time_init(void) +{ + u64 tmp = (u64)NSEC_PER_SEC << SCHED_CLOCK_SHIFT; + + do_div(tmp, c6x_core_freq); + sched_clock_multiplier = tmp; + + clocksource_register_hz(&clocksource_tsc, c6x_core_freq); + + /* write anything into TSCL to enable counting */ + set_creg(TSCL, 0); + + /* probe for timer64 event timer */ + timer64_init(); +} diff --git a/arch/c6x/kernel/traps.c b/arch/c6x/kernel/traps.c new file mode 100644 index 0000000..f50e3ed --- /dev/null +++ b/arch/c6x/kernel/traps.c @@ -0,0 +1,423 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2006, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/module.h> +#include <linux/ptrace.h> +#include <linux/kallsyms.h> +#include <linux/bug.h> + +#include <asm/soc.h> +#include <asm/traps.h> + +int (*c6x_nmi_handler)(struct pt_regs *regs); + +void __init trap_init(void) +{ + ack_exception(EXCEPT_TYPE_NXF); + ack_exception(EXCEPT_TYPE_EXC); + ack_exception(EXCEPT_TYPE_IXF); + ack_exception(EXCEPT_TYPE_SXF); + enable_exception(); +} + +void show_regs(struct pt_regs *regs) +{ + pr_err("\n"); + pr_err("PC: %08lx SP: %08lx\n", regs->pc, regs->sp); + pr_err("Status: %08lx ORIG_A4: %08lx\n", regs->csr, regs->orig_a4); + pr_err("A0: %08lx B0: %08lx\n", regs->a0, regs->b0); + pr_err("A1: %08lx B1: %08lx\n", regs->a1, regs->b1); + pr_err("A2: %08lx B2: %08lx\n", regs->a2, regs->b2); + pr_err("A3: %08lx B3: %08lx\n", regs->a3, regs->b3); + pr_err("A4: %08lx B4: %08lx\n", regs->a4, regs->b4); + pr_err("A5: %08lx B5: %08lx\n", regs->a5, regs->b5); + pr_err("A6: %08lx B6: %08lx\n", regs->a6, regs->b6); + pr_err("A7: %08lx B7: %08lx\n", regs->a7, regs->b7); + pr_err("A8: %08lx B8: %08lx\n", regs->a8, regs->b8); + pr_err("A9: %08lx B9: %08lx\n", regs->a9, regs->b9); + pr_err("A10: %08lx B10: %08lx\n", regs->a10, regs->b10); + pr_err("A11: %08lx B11: %08lx\n", regs->a11, regs->b11); + pr_err("A12: %08lx B12: %08lx\n", regs->a12, regs->b12); + pr_err("A13: %08lx B13: %08lx\n", regs->a13, regs->b13); + pr_err("A14: %08lx B14: %08lx\n", regs->a14, regs->dp); + pr_err("A15: %08lx B15: %08lx\n", regs->a15, regs->sp); + pr_err("A16: %08lx B16: %08lx\n", regs->a16, regs->b16); + pr_err("A17: %08lx B17: %08lx\n", regs->a17, regs->b17); + pr_err("A18: %08lx B18: %08lx\n", regs->a18, regs->b18); + pr_err("A19: %08lx B19: %08lx\n", regs->a19, regs->b19); + pr_err("A20: %08lx B20: %08lx\n", regs->a20, regs->b20); + pr_err("A21: %08lx B21: %08lx\n", regs->a21, regs->b21); + pr_err("A22: %08lx B22: %08lx\n", regs->a22, regs->b22); + pr_err("A23: %08lx B23: %08lx\n", regs->a23, regs->b23); + pr_err("A24: %08lx B24: %08lx\n", regs->a24, regs->b24); + pr_err("A25: %08lx B25: %08lx\n", regs->a25, regs->b25); + pr_err("A26: %08lx B26: %08lx\n", regs->a26, regs->b26); + pr_err("A27: %08lx B27: %08lx\n", regs->a27, regs->b27); + pr_err("A28: %08lx B28: %08lx\n", regs->a28, regs->b28); + pr_err("A29: %08lx B29: %08lx\n", regs->a29, regs->b29); + pr_err("A30: %08lx B30: %08lx\n", regs->a30, regs->b30); + pr_err("A31: %08lx B31: %08lx\n", regs->a31, regs->b31); +} + +void dump_stack(void) +{ + unsigned long stack; + + show_stack(current, &stack); +} +EXPORT_SYMBOL(dump_stack); + + +void die(char *str, struct pt_regs *fp, int nr) +{ + console_verbose(); + pr_err("%s: %08x\n", str, nr); + show_regs(fp); + + pr_err("Process %s (pid: %d, stackpage=%08lx)\n", + current->comm, current->pid, (PAGE_SIZE + + (unsigned long) current)); + + dump_stack(); + while (1) + ; +} + +static void die_if_kernel(char *str, struct pt_regs *fp, int nr) +{ + if (user_mode(fp)) + return; + + die(str, fp, nr); +} + + +/* Internal exceptions */ +static struct exception_info iexcept_table[10] = { + { "Oops - instruction fetch", SIGBUS, BUS_ADRERR }, + { "Oops - fetch packet", SIGBUS, BUS_ADRERR }, + { "Oops - execute packet", SIGILL, ILL_ILLOPC }, + { "Oops - undefined instruction", SIGILL, ILL_ILLOPC }, + { "Oops - resource conflict", SIGILL, ILL_ILLOPC }, + { "Oops - resource access", SIGILL, ILL_PRVREG }, + { "Oops - privilege", SIGILL, ILL_PRVOPC }, + { "Oops - loops buffer", SIGILL, ILL_ILLOPC }, + { "Oops - software exception", SIGILL, ILL_ILLTRP }, + { "Oops - unknown exception", SIGILL, ILL_ILLOPC } +}; + +/* External exceptions */ +static struct exception_info eexcept_table[128] = { + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - external exception", SIGBUS, BUS_ADRERR }, + { "Oops - CPU memory protection fault", SIGSEGV, SEGV_ACCERR }, + { "Oops - CPU memory protection fault in L1P", SIGSEGV, SEGV_ACCERR }, + { "Oops - DMA memory protection fault in L1P", SIGSEGV, SEGV_ACCERR }, + { "Oops - CPU memory protection fault in L1D", SIGSEGV, SEGV_ACCERR }, + { "Oops - DMA memory protection fault in L1D", SIGSEGV, SEGV_ACCERR }, + { "Oops - CPU memory protection fault in L2", SIGSEGV, SEGV_ACCERR }, + { "Oops - DMA memory protection fault in L2", SIGSEGV, SEGV_ACCERR }, + { "Oops - EMC CPU memory protection fault", SIGSEGV, SEGV_ACCERR }, + { "Oops - EMC bus error", SIGBUS, BUS_ADRERR } +}; + +static void do_trap(struct exception_info *except_info, struct pt_regs *regs) +{ + unsigned long addr = instruction_pointer(regs); + siginfo_t info; + + if (except_info->code != TRAP_BRKPT) + pr_err("TRAP: %s PC[0x%lx] signo[%d] code[%d]\n", + except_info->kernel_str, regs->pc, + except_info->signo, except_info->code); + + die_if_kernel(except_info->kernel_str, regs, addr); + + info.si_signo = except_info->signo; + info.si_errno = 0; + info.si_code = except_info->code; + info.si_addr = (void __user *)addr; + + force_sig_info(except_info->signo, &info, current); +} + +/* + * Process an internal exception (non maskable) + */ +static int process_iexcept(struct pt_regs *regs) +{ + unsigned int iexcept_report = get_iexcept(); + unsigned int iexcept_num; + + ack_exception(EXCEPT_TYPE_IXF); + + pr_err("IEXCEPT: PC[0x%lx]\n", regs->pc); + + while (iexcept_report) { + iexcept_num = __ffs(iexcept_report); + iexcept_report &= ~(1 << iexcept_num); + set_iexcept(iexcept_report); + if (*(unsigned int *)regs->pc == BKPT_OPCODE) { + /* This is a breakpoint */ + struct exception_info bkpt_exception = { + "Oops - undefined instruction", + SIGTRAP, TRAP_BRKPT + }; + do_trap(&bkpt_exception, regs); + iexcept_report &= ~(0xFF); + set_iexcept(iexcept_report); + continue; + } + + do_trap(&iexcept_table[iexcept_num], regs); + } + return 0; +} + +/* + * Process an external exception (maskable) + */ +static void process_eexcept(struct pt_regs *regs) +{ + int evt; + + pr_err("EEXCEPT: PC[0x%lx]\n", regs->pc); + + while ((evt = soc_get_exception()) >= 0) + do_trap(&eexcept_table[evt], regs); + + ack_exception(EXCEPT_TYPE_EXC); +} + +/* + * Main exception processing + */ +asmlinkage int process_exception(struct pt_regs *regs) +{ + unsigned int type; + unsigned int type_num; + unsigned int ie_num = 9; /* default is unknown exception */ + + while ((type = get_except_type()) != 0) { + type_num = fls(type) - 1; + + switch (type_num) { + case EXCEPT_TYPE_NXF: + ack_exception(EXCEPT_TYPE_NXF); + if (c6x_nmi_handler) + (c6x_nmi_handler)(regs); + else + pr_alert("NMI interrupt!\n"); + break; + + case EXCEPT_TYPE_IXF: + if (process_iexcept(regs)) + return 1; + break; + + case EXCEPT_TYPE_EXC: + process_eexcept(regs); + break; + + case EXCEPT_TYPE_SXF: + ie_num = 8; + default: + ack_exception(type_num); + do_trap(&iexcept_table[ie_num], regs); + break; + } + } + return 0; +} + +static int kstack_depth_to_print = 48; + +static void show_trace(unsigned long *stack, unsigned long *endstack) +{ + unsigned long addr; + int i; + + pr_debug("Call trace:"); + i = 0; + while (stack + 1 <= endstack) { + addr = *stack++; + /* + * If the address is either in the text segment of the + * kernel, or in the region which contains vmalloc'ed + * memory, it *may* be the address of a calling + * routine; if so, print it so that someone tracing + * down the cause of the crash will be able to figure + * out the call path that was taken. + */ + if (__kernel_text_address(addr)) { +#ifndef CONFIG_KALLSYMS + if (i % 5 == 0) + pr_debug("\n "); +#endif + pr_debug(" [<%08lx>]", addr); + print_symbol(" %s\n", addr); + i++; + } + } + pr_debug("\n"); +} + +void show_stack(struct task_struct *task, unsigned long *stack) +{ + unsigned long *p, *endstack; + int i; + + if (!stack) { + if (task && task != current) + /* We know this is a kernel stack, + so this is the start/end */ + stack = (unsigned long *)thread_saved_ksp(task); + else + stack = (unsigned long *)&stack; + } + endstack = (unsigned long *)(((unsigned long)stack + THREAD_SIZE - 1) + & -THREAD_SIZE); + + pr_debug("Stack from %08lx:", (unsigned long)stack); + for (i = 0, p = stack; i < kstack_depth_to_print; i++) { + if (p + 1 > endstack) + break; + if (i % 8 == 0) + pr_cont("\n "); + pr_cont(" %08lx", *p++); + } + pr_cont("\n"); + show_trace(stack, endstack); +} + +int is_valid_bugaddr(unsigned long addr) +{ + return __kernel_text_address(addr); +} diff --git a/arch/c6x/kernel/vectors.S b/arch/c6x/kernel/vectors.S new file mode 100644 index 0000000..c95c66f --- /dev/null +++ b/arch/c6x/kernel/vectors.S @@ -0,0 +1,81 @@ +; +; Port on Texas Instruments TMS320C6x architecture +; +; Copyright (C) 2004, 2006, 2009, 2010, 2011 Texas Instruments Incorporated +; Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) +; +; This program is free software; you can redistribute it and/or modify +; it under the terms of the GNU General Public License version 2 as +; published by the Free Software Foundation. +; +; This section handles all the interrupt vector routines. +; At RESET the processor sets up the DRAM timing parameters and +; branches to the label _c_int00 which handles initialization for the C code. +; + +#define ALIGNMENT 5 + + .macro IRQVEC name, handler + .align ALIGNMENT + .hidden \name + .global \name +\name: +#ifdef CONFIG_C6X_BIG_KERNEL + STW .D2T1 A0,*B15--[2] + || MVKL .S1 \handler,A0 + MVKH .S1 \handler,A0 + B .S2X A0 + LDW .D2T1 *++B15[2],A0 + NOP 4 + NOP + NOP + .endm +#else /* CONFIG_C6X_BIG_KERNEL */ + B .S2 \handler + NOP + NOP + NOP + NOP + NOP + NOP + NOP + .endm +#endif /* CONFIG_C6X_BIG_KERNEL */ + + .sect ".vectors","ax" + .align ALIGNMENT + .global RESET + .hidden RESET +RESET: +#ifdef CONFIG_C6X_BIG_KERNEL + MVKL .S1 _c_int00,A0 ; branch to _c_int00 + MVKH .S1 _c_int00,A0 + B .S2X A0 +#else + B .S2 _c_int00 + NOP + NOP +#endif + NOP + NOP + NOP + NOP + NOP + + + IRQVEC NMI,_nmi_handler ; NMI interrupt + IRQVEC AINT,_bad_interrupt ; reserved + IRQVEC MSGINT,_bad_interrupt ; reserved + + IRQVEC INT4,_int4_handler + IRQVEC INT5,_int5_handler + IRQVEC INT6,_int6_handler + IRQVEC INT7,_int7_handler + IRQVEC INT8,_int8_handler + IRQVEC INT9,_int9_handler + IRQVEC INT10,_int10_handler + IRQVEC INT11,_int11_handler + IRQVEC INT12,_int12_handler + IRQVEC INT13,_int13_handler + IRQVEC INT14,_int14_handler + IRQVEC INT15,_int15_handler diff --git a/arch/c6x/kernel/vmlinux.lds.S b/arch/c6x/kernel/vmlinux.lds.S new file mode 100644 index 0000000..1d81c4c --- /dev/null +++ b/arch/c6x/kernel/vmlinux.lds.S @@ -0,0 +1,162 @@ +/* + * ld script for the c6x kernel + * + * Copyright (C) 2010, 2011 Texas Instruments Incorporated + * Mark Salter <msalter@redhat.com> + */ +#include <asm-generic/vmlinux.lds.h> +#include <asm/thread_info.h> +#include <asm/page.h> + +ENTRY(_c_int00) + +#if defined(CONFIG_CPU_BIG_ENDIAN) +jiffies = jiffies_64 + 4; +#else +jiffies = jiffies_64; +#endif + +#define READONLY_SEGMENT_START \ + . = PAGE_OFFSET; +#define READWRITE_SEGMENT_START \ + . = ALIGN(128); \ + _data_lma = .; + +SECTIONS +{ + /* + * Start kernel read only segment + */ + READONLY_SEGMENT_START + + .vectors : + { + _vectors_start = .; + *(.vectors) + . = ALIGN(0x400); + _vectors_end = .; + } + + . = ALIGN(0x1000); + .cmdline : + { + *(.cmdline) + } + + /* + * This section contains data which may be shared with other + * cores. It needs to be a fixed offset from PAGE_OFFSET + * regardless of kernel configuration. + */ + .virtio_ipc_dev : + { + *(.virtio_ipc_dev) + } + + . = ALIGN(PAGE_SIZE); + .init : + { + _stext = .; + _sinittext = .; + HEAD_TEXT + INIT_TEXT + _einittext = .; + } + + __init_begin = _stext; + INIT_DATA_SECTION(16) + + PERCPU_SECTION(128) + + . = ALIGN(PAGE_SIZE); + __init_end = .; + + .text : + { + _text = .; + TEXT_TEXT + SCHED_TEXT + LOCK_TEXT + IRQENTRY_TEXT + KPROBES_TEXT + *(.fixup) + *(.gnu.warning) + } + + EXCEPTION_TABLE(16) + NOTES + + RO_DATA_SECTION(PAGE_SIZE) + .const : + { + *(.const .const.* .gnu.linkonce.r.*) + *(.switch) + } + + . = ALIGN (8) ; + __fdt_blob : AT(ADDR(__fdt_blob) - LOAD_OFFSET) + { + _fdt_start = . ; /* place for fdt blob */ + *(__fdt_blob) ; /* Any link-placed DTB */ + BYTE(0); /* section always has contents */ + . = _fdt_start + 0x4000; /* Pad up to 16kbyte */ + _fdt_end = . ; + } + + _etext = .; + + /* + * Start kernel read-write segment. + */ + READWRITE_SEGMENT_START + _sdata = .; + + .fardata : AT(ADDR(.fardata) - LOAD_OFFSET) + { + INIT_TASK_DATA(THREAD_SIZE) + NOSAVE_DATA + PAGE_ALIGNED_DATA(PAGE_SIZE) + CACHELINE_ALIGNED_DATA(128) + READ_MOSTLY_DATA(128) + DATA_DATA + CONSTRUCTORS + *(.data1) + *(.fardata .fardata.*) + *(.data.debug_bpt) + } + + .neardata ALIGN(8) : AT(ADDR(.neardata) - LOAD_OFFSET) + { + *(.neardata2 .neardata2.* .gnu.linkonce.s2.*) + *(.neardata .neardata.* .gnu.linkonce.s.*) + . = ALIGN(8); + } + + _edata = .; + + __bss_start = .; + SBSS(8) + BSS(8) + .far : + { + . = ALIGN(8); + *(.dynfar) + *(.far .far.* .gnu.linkonce.b.*) + . = ALIGN(8); + } + __bss_stop = .; + + _end = .; + + DWARF_DEBUG + + /DISCARD/ : + { + EXIT_TEXT + EXIT_DATA + EXIT_CALL + *(.discard) + *(.discard.*) + *(.interp) + } +} diff --git a/arch/c6x/lib/Makefile b/arch/c6x/lib/Makefile new file mode 100644 index 0000000..ffd3c65 --- /dev/null +++ b/arch/c6x/lib/Makefile @@ -0,0 +1,7 @@ +# +# Makefile for arch/c6x/lib/ +# + +lib-y := divu.o divi.o pop_rts.o push_rts.o remi.o remu.o strasgi.o llshru.o +lib-y += llshr.o llshl.o negll.o mpyll.o divremi.o divremu.o +lib-y += checksum.o csum_64plus.o memcpy_64plus.o strasgi_64plus.o diff --git a/arch/c6x/lib/checksum.c b/arch/c6x/lib/checksum.c new file mode 100644 index 0000000..67cc93b --- /dev/null +++ b/arch/c6x/lib/checksum.c @@ -0,0 +1,36 @@ +/* + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#include <linux/module.h> +#include <net/checksum.h> + +#include <asm/byteorder.h> + +/* + * copy from fs while checksumming, otherwise like csum_partial + */ +__wsum +csum_partial_copy_from_user(const void __user *src, void *dst, int len, + __wsum sum, int *csum_err) +{ + int missing; + + missing = __copy_from_user(dst, src, len); + if (missing) { + memset(dst + len - missing, 0, missing); + *csum_err = -EFAULT; + } else + *csum_err = 0; + + return csum_partial(dst, len, sum); +} +EXPORT_SYMBOL(csum_partial_copy_from_user); + +/* These are from csum_64plus.S */ +EXPORT_SYMBOL(csum_partial); +EXPORT_SYMBOL(csum_partial_copy); +EXPORT_SYMBOL(ip_compute_csum); +EXPORT_SYMBOL(ip_fast_csum); diff --git a/arch/c6x/lib/csum_64plus.S b/arch/c6x/lib/csum_64plus.S new file mode 100644 index 0000000..6d25896 --- /dev/null +++ b/arch/c6x/lib/csum_64plus.S @@ -0,0 +1,419 @@ +; +; linux/arch/c6x/lib/csum_64plus.s +; +; Port on Texas Instruments TMS320C6x architecture +; +; Copyright (C) 2006, 2009, 2010, 2011 Texas Instruments Incorporated +; Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) +; +; This program is free software; you can redistribute it and/or modify +; it under the terms of the GNU General Public License version 2 as +; published by the Free Software Foundation. +; +#include <linux/linkage.h> + +; +;unsigned int csum_partial_copy(const char *src, char * dst, +; int len, int sum) +; +; A4: src +; B4: dst +; A6: len +; B6: sum +; return csum in A4 +; + + .text +ENTRY(csum_partial_copy) + MVC .S2 ILC,B30 + + MV .D1X B6,A31 ; given csum + ZERO .D1 A9 ; csum (a side) +|| ZERO .D2 B9 ; csum (b side) +|| SHRU .S2X A6,2,B5 ; len / 4 + + ;; Check alignment and size + AND .S1 3,A4,A1 +|| AND .S2 3,B4,B0 + OR .L2X B0,A1,B0 ; non aligned condition +|| MVC .S2 B5,ILC +|| MVK .D2 1,B2 +|| MV .D1X B5,A1 ; words condition + [!A1] B .S1 L8 + [B0] BNOP .S1 L6,5 + + SPLOOP 1 + + ;; Main loop for aligned words + LDW .D1T1 *A4++,A7 + NOP 4 + MV .S2X A7,B7 +|| EXTU .S1 A7,0,16,A16 + STW .D2T2 B7,*B4++ +|| MPYU .M2 B7,B2,B8 +|| ADD .L1 A16,A9,A9 + NOP + SPKERNEL 8,0 +|| ADD .L2 B8,B9,B9 + + ZERO .D1 A1 +|| ADD .L1X A9,B9,A9 ; add csum from a and b sides + +L6: + [!A1] BNOP .S1 L8,5 + + ;; Main loop for non-aligned words + SPLOOP 2 + || MVK .L1 1,A2 + + LDNW .D1T1 *A4++,A7 + NOP 3 + + NOP + MV .S2X A7,B7 + || EXTU .S1 A7,0,16,A16 + || MPYU .M1 A7,A2,A8 + + ADD .L1 A16,A9,A9 + SPKERNEL 6,0 + || STNW .D2T2 B7,*B4++ + || ADD .L1 A8,A9,A9 + +L8: AND .S2X 2,A6,B5 + CMPGT .L2 B5,0,B0 + [!B0] BNOP .S1 L82,4 + + ;; Manage half-word + ZERO .L1 A7 +|| ZERO .D1 A8 + +#ifdef CONFIG_CPU_BIG_ENDIAN + + LDBU .D1T1 *A4++,A7 + LDBU .D1T1 *A4++,A8 + NOP 3 + SHL .S1 A7,8,A0 + ADD .S1 A8,A9,A9 + STB .D2T1 A7,*B4++ +|| ADD .S1 A0,A9,A9 + STB .D2T1 A8,*B4++ + +#else + + LDBU .D1T1 *A4++,A7 + LDBU .D1T1 *A4++,A8 + NOP 3 + ADD .S1 A7,A9,A9 + SHL .S1 A8,8,A0 + + STB .D2T1 A7,*B4++ +|| ADD .S1 A0,A9,A9 + STB .D2T1 A8,*B4++ + +#endif + + ;; Manage eventually the last byte +L82: AND .S2X 1,A6,B0 + [!B0] BNOP .S1 L9,5 + +|| ZERO .L1 A7 + +L83: LDBU .D1T1 *A4++,A7 + NOP 4 + + MV .L2X A7,B7 + +#ifdef CONFIG_CPU_BIG_ENDIAN + + STB .D2T2 B7,*B4++ +|| SHL .S1 A7,8,A7 + ADD .S1 A7,A9,A9 + +#else + + STB .D2T2 B7,*B4++ +|| ADD .S1 A7,A9,A9 + +#endif + + ;; Fold the csum +L9: SHRU .S2X A9,16,B0 + [!B0] BNOP .S1 L10,5 + +L91: SHRU .S2X A9,16,B4 +|| EXTU .S1 A9,16,16,A3 + ADD .D1X A3,B4,A9 + + SHRU .S1 A9,16,A0 + [A0] BNOP .S1 L91,5 + +L10: ADD .D1 A31,A9,A9 + MV .D1 A9,A4 + + BNOP .S2 B3,4 + MVC .S2 B30,ILC +ENDPROC(csum_partial_copy) + +; +;unsigned short +;ip_fast_csum(unsigned char *iph, unsigned int ihl) +;{ +; unsigned int checksum = 0; +; unsigned short *tosum = (unsigned short *) iph; +; int len; +; +; len = ihl*4; +; +; if (len <= 0) +; return 0; +; +; while(len) { +; len -= 2; +; checksum += *tosum++; +; } +; if (len & 1) +; checksum += *(unsigned char*) tosum; +; +; while(checksum >> 16) +; checksum = (checksum & 0xffff) + (checksum >> 16); +; +; return ~checksum; +;} +; +; A4: iph +; B4: ihl +; return checksum in A4 +; + .text + +ENTRY(ip_fast_csum) + ZERO .D1 A5 + || MVC .S2 ILC,B30 + SHL .S2 B4,2,B0 + CMPGT .L2 B0,0,B1 + [!B1] BNOP .S1 L15,4 + [!B1] ZERO .D1 A3 + + [!B0] B .S1 L12 + SHRU .S2 B0,1,B0 + MVC .S2 B0,ILC + NOP 3 + + SPLOOP 1 + LDHU .D1T1 *A4++,A3 + NOP 3 + NOP + SPKERNEL 5,0 + || ADD .L1 A3,A5,A5 + +L12: SHRU .S1 A5,16,A0 + [!A0] BNOP .S1 L14,5 + +L13: SHRU .S2X A5,16,B4 + EXTU .S1 A5,16,16,A3 + ADD .D1X A3,B4,A5 + SHRU .S1 A5,16,A0 + [A0] BNOP .S1 L13,5 + +L14: NOT .D1 A5,A3 + EXTU .S1 A3,16,16,A3 + +L15: BNOP .S2 B3,3 + MVC .S2 B30,ILC + MV .D1 A3,A4 +ENDPROC(ip_fast_csum) + +; +;unsigned short +;do_csum(unsigned char *buff, unsigned int len) +;{ +; int odd, count; +; unsigned int result = 0; +; +; if (len <= 0) +; goto out; +; odd = 1 & (unsigned long) buff; +; if (odd) { +;#ifdef __LITTLE_ENDIAN +; result += (*buff << 8); +;#else +; result = *buff; +;#endif +; len--; +; buff++; +; } +; count = len >> 1; /* nr of 16-bit words.. */ +; if (count) { +; if (2 & (unsigned long) buff) { +; result += *(unsigned short *) buff; +; count--; +; len -= 2; +; buff += 2; +; } +; count >>= 1; /* nr of 32-bit words.. */ +; if (count) { +; unsigned int carry = 0; +; do { +; unsigned int w = *(unsigned int *) buff; +; count--; +; buff += 4; +; result += carry; +; result += w; +; carry = (w > result); +; } while (count); +; result += carry; +; result = (result & 0xffff) + (result >> 16); +; } +; if (len & 2) { +; result += *(unsigned short *) buff; +; buff += 2; +; } +; } +; if (len & 1) +;#ifdef __LITTLE_ENDIAN +; result += *buff; +;#else +; result += (*buff << 8); +;#endif +; result = (result & 0xffff) + (result >> 16); +; /* add up carry.. */ +; result = (result & 0xffff) + (result >> 16); +; if (odd) +; result = ((result >> 8) & 0xff) | ((result & 0xff) << 8); +;out: +; return result; +;} +; +; A4: buff +; B4: len +; return checksum in A4 +; + +ENTRY(do_csum) + CMPGT .L2 B4,0,B0 + [!B0] BNOP .S1 L26,3 + EXTU .S1 A4,31,31,A0 + + MV .L1 A0,A3 +|| MV .S1X B3,A5 +|| MV .L2 B4,B3 +|| ZERO .D1 A1 + +#ifdef CONFIG_CPU_BIG_ENDIAN + [A0] SUB .L2 B3,1,B3 +|| [A0] LDBU .D1T1 *A4++,A1 +#else + [!A0] BNOP .S1 L21,5 +|| [A0] LDBU .D1T1 *A4++,A0 + SUB .L2 B3,1,B3 +|| SHL .S1 A0,8,A1 +L21: +#endif + SHR .S2 B3,1,B0 + [!B0] BNOP .S1 L24,3 + MVK .L1 2,A0 + AND .L1 A4,A0,A0 + + [!A0] BNOP .S1 L22,5 +|| [A0] LDHU .D1T1 *A4++,A0 + SUB .L2 B0,1,B0 +|| SUB .S2 B3,2,B3 +|| ADD .L1 A0,A1,A1 +L22: + SHR .S2 B0,1,B0 +|| ZERO .L1 A0 + + [!B0] BNOP .S1 L23,5 +|| [B0] MVC .S2 B0,ILC + + SPLOOP 3 + SPMASK L1 +|| MV .L1 A1,A2 +|| LDW .D1T1 *A4++,A1 + + NOP 4 + ADD .L1 A0,A1,A0 + ADD .L1 A2,A0,A2 + + SPKERNEL 1,2 +|| CMPGTU .L1 A1,A2,A0 + + ADD .L1 A0,A2,A6 + EXTU .S1 A6,16,16,A7 + SHRU .S2X A6,16,B0 + NOP 1 + ADD .L1X A7,B0,A1 +L23: + MVK .L2 2,B0 + AND .L2 B3,B0,B0 + [B0] LDHU .D1T1 *A4++,A0 + NOP 4 + [B0] ADD .L1 A0,A1,A1 +L24: + EXTU .S2 B3,31,31,B0 +#ifdef CONFIG_CPU_BIG_ENDIAN + [!B0] BNOP .S1 L25,4 +|| [B0] LDBU .D1T1 *A4,A0 + SHL .S1 A0,8,A0 + ADD .L1 A0,A1,A1 +L25: +#else + [B0] LDBU .D1T1 *A4,A0 + NOP 4 + [B0] ADD .L1 A0,A1,A1 +#endif + EXTU .S1 A1,16,16,A0 + SHRU .S2X A1,16,B0 + NOP 1 + ADD .L1X A0,B0,A0 + SHRU .S1 A0,16,A1 + ADD .L1 A0,A1,A0 + EXTU .S1 A0,16,16,A1 + EXTU .S1 A1,16,24,A2 + + EXTU .S1 A1,24,16,A0 +|| MV .L2X A3,B0 + + [B0] OR .L1 A0,A2,A1 +L26: + NOP 1 + BNOP .S2X A5,4 + MV .L1 A1,A4 +ENDPROC(do_csum) + +;__wsum csum_partial(const void *buff, int len, __wsum wsum) +;{ +; unsigned int sum = (__force unsigned int)wsum; +; unsigned int result = do_csum(buff, len); +; +; /* add in old sum, and carry.. */ +; result += sum; +; if (sum > result) +; result += 1; +; return (__force __wsum)result; +;} +; +ENTRY(csum_partial) + MV .L1X B3,A9 +|| CALLP .S2 do_csum,B3 +|| MV .S1 A6,A8 + BNOP .S2X A9,2 + ADD .L1 A8,A4,A1 + CMPGTU .L1 A8,A1,A0 + ADD .L1 A1,A0,A4 +ENDPROC(csum_partial) + +;unsigned short +;ip_compute_csum(unsigned char *buff, unsigned int len) +; +; A4: buff +; B4: len +; return checksum in A4 + +ENTRY(ip_compute_csum) + MV .L1X B3,A9 +|| CALLP .S2 do_csum,B3 + BNOP .S2X A9,3 + NOT .S1 A4,A4 + CLR .S1 A4,16,31,A4 +ENDPROC(ip_compute_csum) diff --git a/arch/c6x/lib/divi.S b/arch/c6x/lib/divi.S new file mode 100644 index 0000000..4bde924 --- /dev/null +++ b/arch/c6x/lib/divi.S @@ -0,0 +1,53 @@ +;; Copyright 2010 Free Software Foundation, Inc. +;; Contributed by Bernd Schmidt <bernds@codesourcery.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +#include <linux/linkage.h> + + ;; ABI considerations for the divide functions + ;; The following registers are call-used: + ;; __c6xabi_divi A0,A1,A2,A4,A6,B0,B1,B2,B4,B5 + ;; __c6xabi_divu A0,A1,A2,A4,A6,B0,B1,B2,B4 + ;; __c6xabi_remi A1,A2,A4,A5,A6,B0,B1,B2,B4 + ;; __c6xabi_remu A1,A4,A5,A7,B0,B1,B2,B4 + ;; + ;; In our implementation, divu and remu are leaf functions, + ;; while both divi and remi call into divu. + ;; A0 is not clobbered by any of the functions. + ;; divu does not clobber B2 either, which is taken advantage of + ;; in remi. + ;; divi uses B5 to hold the original return address during + ;; the call to divu. + ;; remi uses B2 and A5 to hold the input values during the + ;; call to divu. It stores B3 in on the stack. + + .text +ENTRY(__c6xabi_divi) + call .s2 __c6xabi_divu +|| mv .d2 B3, B5 +|| cmpgt .l1 0, A4, A1 +|| cmpgt .l2 0, B4, B1 + + [A1] neg .l1 A4, A4 +|| [B1] neg .l2 B4, B4 +|| xor .s1x A1, B1, A1 + [A1] addkpc .s2 _divu_ret, B3, 4 +_divu_ret: + neg .l1 A4, A4 +|| mv .l2 B3,B5 +|| ret .s2 B5 + nop 5 +ENDPROC(__c6xabi_divi) diff --git a/arch/c6x/lib/divremi.S b/arch/c6x/lib/divremi.S new file mode 100644 index 0000000..64bc5aa --- /dev/null +++ b/arch/c6x/lib/divremi.S @@ -0,0 +1,46 @@ +;; Copyright 2010 Free Software Foundation, Inc. +;; Contributed by Bernd Schmidt <bernds@codesourcery.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +#include <linux/linkage.h> + + .text +ENTRY(__c6xabi_divremi) + stw .d2t2 B3, *B15--[2] +|| cmpgt .l1 0, A4, A1 +|| cmpgt .l2 0, B4, B2 +|| mv .s1 A4, A5 +|| call .s2 __c6xabi_divu + + [A1] neg .l1 A4, A4 +|| [B2] neg .l2 B4, B4 +|| xor .s2x B2, A1, B0 +|| mv .d2 B4, B2 + + [B0] addkpc .s2 _divu_ret_1, B3, 1 + [!B0] addkpc .s2 _divu_ret_2, B3, 1 + nop 2 +_divu_ret_1: + neg .l1 A4, A4 +_divu_ret_2: + ldw .d2t2 *++B15[2], B3 + + mpy32 .m1x A4, B2, A6 + nop 3 + ret .s2 B3 + sub .l1 A5, A6, A5 + nop 4 +ENDPROC(__c6xabi_divremi) diff --git a/arch/c6x/lib/divremu.S b/arch/c6x/lib/divremu.S new file mode 100644 index 0000000..caa9f23 --- /dev/null +++ b/arch/c6x/lib/divremu.S @@ -0,0 +1,87 @@ +;; Copyright 2011 Free Software Foundation, Inc. +;; Contributed by Bernd Schmidt <bernds@codesourcery.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +#include <linux/linkage.h> + + .text +ENTRY(__c6xabi_divremu) + ;; We use a series of up to 31 subc instructions. First, we find + ;; out how many leading zero bits there are in the divisor. This + ;; gives us both a shift count for aligning (shifting) the divisor + ;; to the, and the number of times we have to execute subc. + + ;; At the end, we have both the remainder and most of the quotient + ;; in A4. The top bit of the quotient is computed first and is + ;; placed in A2. + + ;; Return immediately if the dividend is zero. Setting B4 to 1 + ;; is a trick to allow us to leave the following insns in the jump + ;; delay slot without affecting the result. + mv .s2x A4, B1 + + [b1] lmbd .l2 1, B4, B1 +||[!b1] b .s2 B3 ; RETURN A +||[!b1] mvk .d2 1, B4 + +||[!b1] zero .s1 A5 + mv .l1x B1, A6 +|| shl .s2 B4, B1, B4 + + ;; The loop performs a maximum of 28 steps, so we do the + ;; first 3 here. + cmpltu .l1x A4, B4, A2 + [!A2] sub .l1x A4, B4, A4 +|| shru .s2 B4, 1, B4 +|| xor .s1 1, A2, A2 + + shl .s1 A2, 31, A2 +|| [b1] subc .l1x A4,B4,A4 +|| [b1] add .s2 -1, B1, B1 + [b1] subc .l1x A4,B4,A4 +|| [b1] add .s2 -1, B1, B1 + + ;; RETURN A may happen here (note: must happen before the next branch) +__divremu0: + cmpgt .l2 B1, 7, B0 +|| [b1] subc .l1x A4,B4,A4 +|| [b1] add .s2 -1, B1, B1 + [b1] subc .l1x A4,B4,A4 +|| [b1] add .s2 -1, B1, B1 +|| [b0] b .s1 __divremu0 + [b1] subc .l1x A4,B4,A4 +|| [b1] add .s2 -1, B1, B1 + [b1] subc .l1x A4,B4,A4 +|| [b1] add .s2 -1, B1, B1 + [b1] subc .l1x A4,B4,A4 +|| [b1] add .s2 -1, B1, B1 + [b1] subc .l1x A4,B4,A4 +|| [b1] add .s2 -1, B1, B1 + [b1] subc .l1x A4,B4,A4 +|| [b1] add .s2 -1, B1, B1 + ;; loop backwards branch happens here + + ret .s2 B3 +|| mvk .s1 32, A1 + sub .l1 A1, A6, A6 +|| extu .s1 A4, A6, A5 + shl .s1 A4, A6, A4 + shru .s1 A4, 1, A4 +|| sub .l1 A6, 1, A6 + or .l1 A2, A4, A4 + shru .s1 A4, A6, A4 + nop +ENDPROC(__c6xabi_divremu) diff --git a/arch/c6x/lib/divu.S b/arch/c6x/lib/divu.S new file mode 100644 index 0000000..64af3c0 --- /dev/null +++ b/arch/c6x/lib/divu.S @@ -0,0 +1,98 @@ +;; Copyright 2010 Free Software Foundation, Inc. +;; Contributed by Bernd Schmidt <bernds@codesourcery.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +#include <linux/linkage.h> + + ;; ABI considerations for the divide functions + ;; The following registers are call-used: + ;; __c6xabi_divi A0,A1,A2,A4,A6,B0,B1,B2,B4,B5 + ;; __c6xabi_divu A0,A1,A2,A4,A6,B0,B1,B2,B4 + ;; __c6xabi_remi A1,A2,A4,A5,A6,B0,B1,B2,B4 + ;; __c6xabi_remu A1,A4,A5,A7,B0,B1,B2,B4 + ;; + ;; In our implementation, divu and remu are leaf functions, + ;; while both divi and remi call into divu. + ;; A0 is not clobbered by any of the functions. + ;; divu does not clobber B2 either, which is taken advantage of + ;; in remi. + ;; divi uses B5 to hold the original return address during + ;; the call to divu. + ;; remi uses B2 and A5 to hold the input values during the + ;; call to divu. It stores B3 in on the stack. + + .text +ENTRY(__c6xabi_divu) + ;; We use a series of up to 31 subc instructions. First, we find + ;; out how many leading zero bits there are in the divisor. This + ;; gives us both a shift count for aligning (shifting) the divisor + ;; to the, and the number of times we have to execute subc. + + ;; At the end, we have both the remainder and most of the quotient + ;; in A4. The top bit of the quotient is computed first and is + ;; placed in A2. + + ;; Return immediately if the dividend is zero. + mv .s2x A4, B1 + [B1] lmbd .l2 1, B4, B1 +|| [!B1] b .s2 B3 ; RETURN A +|| [!B1] mvk .d2 1, B4 + mv .l1x B1, A6 +|| shl .s2 B4, B1, B4 + + ;; The loop performs a maximum of 28 steps, so we do the + ;; first 3 here. + cmpltu .l1x A4, B4, A2 + [!A2] sub .l1x A4, B4, A4 +|| shru .s2 B4, 1, B4 +|| xor .s1 1, A2, A2 + + shl .s1 A2, 31, A2 +|| [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + + ;; RETURN A may happen here (note: must happen before the next branch) +_divu_loop: + cmpgt .l2 B1, 7, B0 +|| [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 +|| [B0] b .s1 _divu_loop + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + ;; loop backwards branch happens here + + ret .s2 B3 +|| mvk .s1 32, A1 + sub .l1 A1, A6, A6 + shl .s1 A4, A6, A4 + shru .s1 A4, 1, A4 +|| sub .l1 A6, 1, A6 + or .l1 A2, A4, A4 + shru .s1 A4, A6, A4 + nop +ENDPROC(__c6xabi_divu) diff --git a/arch/c6x/lib/llshl.S b/arch/c6x/lib/llshl.S new file mode 100644 index 0000000..7b105e2 --- /dev/null +++ b/arch/c6x/lib/llshl.S @@ -0,0 +1,37 @@ +;; Copyright (C) 2010 Texas Instruments Incorporated +;; Contributed by Mark Salter <msalter@redhat.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +;; uint64_t __c6xabi_llshl(uint64_t val, uint shift) + +#include <linux/linkage.h> + + .text +ENTRY(__c6xabi_llshl) + mv .l1x B4,A1 + [!A1] b .s2 B3 ; just return if zero shift + mvk .s1 32,A0 + sub .d1 A0,A1,A0 + cmplt .l1 0,A0,A2 + [A2] shru .s1 A4,A0,A0 + [!A2] neg .l1 A0,A5 +|| [A2] shl .s1 A5,A1,A5 + [!A2] shl .s1 A4,A5,A5 +|| [A2] or .d1 A5,A0,A5 +|| [!A2] mvk .l1 0,A4 + [A2] shl .s1 A4,A1,A4 + bnop .s2 B3,5 +ENDPROC(__c6xabi_llshl) diff --git a/arch/c6x/lib/llshr.S b/arch/c6x/lib/llshr.S new file mode 100644 index 0000000..fde1bec --- /dev/null +++ b/arch/c6x/lib/llshr.S @@ -0,0 +1,38 @@ +;; Copyright (C) 2010 Texas Instruments Incorporated +;; Contributed by Mark Salter <msalter@redhat.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +;; uint64_t __c6xabi_llshr(uint64_t val, uint shift) + +#include <linux/linkage.h> + + .text +ENTRY(__c6xabi_llshr) + mv .l1x B4,A1 + [!A1] b .s2 B3 ; return if zero shift count + mvk .s1 32,A0 + sub .d1 A0,A1,A0 + cmplt .l1 0,A0,A2 + [A2] shl .s1 A5,A0,A0 + nop + [!A2] neg .l1 A0,A4 +|| [A2] shru .s1 A4,A1,A4 + [!A2] shr .s1 A5,A4,A4 +|| [A2] or .d1 A4,A0,A4 + [!A2] shr .s1 A5,0x1f,A5 + [A2] shr .s1 A5,A1,A5 + bnop .s2 B3,5 +ENDPROC(__c6xabi_llshr) diff --git a/arch/c6x/lib/llshru.S b/arch/c6x/lib/llshru.S new file mode 100644 index 0000000..596ae3f --- /dev/null +++ b/arch/c6x/lib/llshru.S @@ -0,0 +1,38 @@ +;; Copyright (C) 2010 Texas Instruments Incorporated +;; Contributed by Mark Salter <msalter@redhat.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +;; uint64_t __c6xabi_llshru(uint64_t val, uint shift) + +#include <linux/linkage.h> + + .text +ENTRY(__c6xabi_llshru) + mv .l1x B4,A1 + [!A1] b .s2 B3 ; return if zero shift count + mvk .s1 32,A0 + sub .d1 A0,A1,A0 + cmplt .l1 0,A0,A2 + [A2] shl .s1 A5,A0,A0 + nop + [!A2] neg .l1 A0,A4 +|| [A2] shru .s1 A4,A1,A4 + [!A2] shru .s1 A5,A4,A4 +|| [A2] or .d1 A4,A0,A4 +|| [!A2] mvk .l1 0,A5 + [A2] shru .s1 A5,A1,A5 + bnop .s2 B3,5 +ENDPROC(__c6xabi_llshru) diff --git a/arch/c6x/lib/memcpy_64plus.S b/arch/c6x/lib/memcpy_64plus.S new file mode 100644 index 0000000..0bbc2cb --- /dev/null +++ b/arch/c6x/lib/memcpy_64plus.S @@ -0,0 +1,46 @@ +; Port on Texas Instruments TMS320C6x architecture +; +; Copyright (C) 2006, 2009, 2010 Texas Instruments Incorporated +; Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) +; +; This program is free software; you can redistribute it and/or modify +; it under the terms of the GNU General Public License version 2 as +; published by the Free Software Foundation. +; + +#include <linux/linkage.h> + + .text + +ENTRY(memcpy) + AND .L1 0x1,A6,A0 + || AND .S1 0x2,A6,A1 + || AND .L2X 0x4,A6,B0 + || MV .D1 A4,A3 + || MVC .S2 ILC,B2 + + [A0] LDB .D2T1 *B4++,A5 + [A1] LDB .D2T1 *B4++,A7 + [A1] LDB .D2T1 *B4++,A8 + [B0] LDNW .D2T1 *B4++,A9 + || SHRU .S2X A6,0x3,B1 + [!B1] BNOP .S2 B3,1 + + [A0] STB .D1T1 A5,*A3++ + ||[B1] MVC .S2 B1,ILC + [A1] STB .D1T1 A7,*A3++ + [A1] STB .D1T1 A8,*A3++ + [B0] STNW .D1T1 A9,*A3++ ; return when len < 8 + + SPLOOP 2 + + LDNDW .D2T1 *B4++,A9:A8 + NOP 3 + + NOP + SPKERNEL 0,0 + || STNDW .D1T1 A9:A8,*A3++ + + BNOP .S2 B3,4 + MVC .S2 B2,ILC +ENDPROC(memcpy) diff --git a/arch/c6x/lib/mpyll.S b/arch/c6x/lib/mpyll.S new file mode 100644 index 0000000..f103441 --- /dev/null +++ b/arch/c6x/lib/mpyll.S @@ -0,0 +1,49 @@ +;; Copyright (C) 2010 Texas Instruments Incorporated +;; Contributed by Mark Salter <msalter@redhat.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +#include <linux/linkage.h> + + ;; uint64_t __c6xabi_mpyll(uint64_t x, uint64_t y) + ;; + ;; 64x64 multiply + ;; First compute partial results using 32-bit parts of x and y: + ;; + ;; b63 b32 b31 b0 + ;; ----------------------------- + ;; | 1 | 0 | + ;; ----------------------------- + ;; + ;; P0 = X0*Y0 + ;; P1 = X0*Y1 + X1*Y0 + ;; P2 = X1*Y1 + ;; + ;; result = (P2 << 64) + (P1 << 32) + P0 + ;; + ;; Since the result is also 64-bit, we can skip the P2 term. + + .text +ENTRY(__c6xabi_mpyll) + mpy32u .m1x A4,B4,A1:A0 ; X0*Y0 + b .s2 B3 + || mpy32u .m2x B5,A4,B1:B0 ; X0*Y1 (don't need upper 32-bits) + || mpy32u .m1x A5,B4,A3:A2 ; X1*Y0 (don't need upper 32-bits) + nop + nop + mv .s1 A0,A4 + add .l1x A2,B0,A5 + add .s1 A1,A5,A5 +ENDPROC(__c6xabi_mpyll) diff --git a/arch/c6x/lib/negll.S b/arch/c6x/lib/negll.S new file mode 100644 index 0000000..82f4bce --- /dev/null +++ b/arch/c6x/lib/negll.S @@ -0,0 +1,31 @@ +;; Copyright (C) 2010 Texas Instruments Incorporated +;; Contributed by Mark Salter <msalter@redhat.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +;; int64_t __c6xabi_negll(int64_t val) + +#include <linux/linkage.h> + + .text +ENTRY(__c6xabi_negll) + b .s2 B3 + mvk .l1 0,A0 + subu .l1 A0,A4,A3:A2 + sub .l1 A0,A5,A0 +|| ext .s1 A3,24,24,A5 + add .l1 A5,A0,A5 + mv .s1 A2,A4 +ENDPROC(__c6xabi_negll) diff --git a/arch/c6x/lib/pop_rts.S b/arch/c6x/lib/pop_rts.S new file mode 100644 index 0000000..d7d96c7 --- /dev/null +++ b/arch/c6x/lib/pop_rts.S @@ -0,0 +1,32 @@ +;; Copyright 2010 Free Software Foundation, Inc. +;; Contributed by Bernd Schmidt <bernds@codesourcery.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +#include <linux/linkage.h> + + .text + +ENTRY(__c6xabi_pop_rts) + lddw .d2t2 *++B15, B3:B2 + lddw .d2t1 *++B15, A11:A10 + lddw .d2t2 *++B15, B11:B10 + lddw .d2t1 *++B15, A13:A12 + lddw .d2t2 *++B15, B13:B12 + lddw .d2t1 *++B15, A15:A14 +|| b .s2 B3 + ldw .d2t2 *++B15[2], B14 + nop 4 +ENDPROC(__c6xabi_pop_rts) diff --git a/arch/c6x/lib/push_rts.S b/arch/c6x/lib/push_rts.S new file mode 100644 index 0000000..f6e3db3 --- /dev/null +++ b/arch/c6x/lib/push_rts.S @@ -0,0 +1,31 @@ +;; Copyright 2010 Free Software Foundation, Inc. +;; Contributed by Bernd Schmidt <bernds@codesourcery.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +#include <linux/linkage.h> + + .text + +ENTRY(__c6xabi_push_rts) + stw .d2t2 B14, *B15--[2] + stdw .d2t1 A15:A14, *B15-- +|| b .s2x A3 + stdw .d2t2 B13:B12, *B15-- + stdw .d2t1 A13:A12, *B15-- + stdw .d2t2 B11:B10, *B15-- + stdw .d2t1 A11:A10, *B15-- + stdw .d2t2 B3:B2, *B15-- +ENDPROC(__c6xabi_push_rts) diff --git a/arch/c6x/lib/remi.S b/arch/c6x/lib/remi.S new file mode 100644 index 0000000..6f2ca18 --- /dev/null +++ b/arch/c6x/lib/remi.S @@ -0,0 +1,64 @@ +;; Copyright 2010 Free Software Foundation, Inc. +;; Contributed by Bernd Schmidt <bernds@codesourcery.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +#include <linux/linkage.h> + + ;; ABI considerations for the divide functions + ;; The following registers are call-used: + ;; __c6xabi_divi A0,A1,A2,A4,A6,B0,B1,B2,B4,B5 + ;; __c6xabi_divu A0,A1,A2,A4,A6,B0,B1,B2,B4 + ;; __c6xabi_remi A1,A2,A4,A5,A6,B0,B1,B2,B4 + ;; __c6xabi_remu A1,A4,A5,A7,B0,B1,B2,B4 + ;; + ;; In our implementation, divu and remu are leaf functions, + ;; while both divi and remi call into divu. + ;; A0 is not clobbered by any of the functions. + ;; divu does not clobber B2 either, which is taken advantage of + ;; in remi. + ;; divi uses B5 to hold the original return address during + ;; the call to divu. + ;; remi uses B2 and A5 to hold the input values during the + ;; call to divu. It stores B3 in on the stack. + + .text + +ENTRY(__c6xabi_remi) + stw .d2t2 B3, *B15--[2] +|| cmpgt .l1 0, A4, A1 +|| cmpgt .l2 0, B4, B2 +|| mv .s1 A4, A5 +|| call .s2 __c6xabi_divu + + [A1] neg .l1 A4, A4 +|| [B2] neg .l2 B4, B4 +|| xor .s2x B2, A1, B0 +|| mv .d2 B4, B2 + + [B0] addkpc .s2 _divu_ret_1, B3, 1 + [!B0] addkpc .s2 _divu_ret_2, B3, 1 + nop 2 +_divu_ret_1: + neg .l1 A4, A4 +_divu_ret_2: + ldw .d2t2 *++B15[2], B3 + + mpy32 .m1x A4, B2, A6 + nop 3 + ret .s2 B3 + sub .l1 A5, A6, A4 + nop 4 +ENDPROC(__c6xabi_remi) diff --git a/arch/c6x/lib/remu.S b/arch/c6x/lib/remu.S new file mode 100644 index 0000000..3fae719 --- /dev/null +++ b/arch/c6x/lib/remu.S @@ -0,0 +1,82 @@ +;; Copyright 2010 Free Software Foundation, Inc. +;; Contributed by Bernd Schmidt <bernds@codesourcery.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +#include <linux/linkage.h> + + ;; ABI considerations for the divide functions + ;; The following registers are call-used: + ;; __c6xabi_divi A0,A1,A2,A4,A6,B0,B1,B2,B4,B5 + ;; __c6xabi_divu A0,A1,A2,A4,A6,B0,B1,B2,B4 + ;; __c6xabi_remi A1,A2,A4,A5,A6,B0,B1,B2,B4 + ;; __c6xabi_remu A1,A4,A5,A7,B0,B1,B2,B4 + ;; + ;; In our implementation, divu and remu are leaf functions, + ;; while both divi and remi call into divu. + ;; A0 is not clobbered by any of the functions. + ;; divu does not clobber B2 either, which is taken advantage of + ;; in remi. + ;; divi uses B5 to hold the original return address during + ;; the call to divu. + ;; remi uses B2 and A5 to hold the input values during the + ;; call to divu. It stores B3 in on the stack. + + + .text + +ENTRY(__c6xabi_remu) + ;; The ABI seems designed to prevent these functions calling each other, + ;; so we duplicate most of the divsi3 code here. + mv .s2x A4, B1 + lmbd .l2 1, B4, B1 +|| [!B1] b .s2 B3 ; RETURN A +|| [!B1] mvk .d2 1, B4 + + mv .l1x B1, A7 +|| shl .s2 B4, B1, B4 + + cmpltu .l1x A4, B4, A1 + [!A1] sub .l1x A4, B4, A4 + shru .s2 B4, 1, B4 + +_remu_loop: + cmpgt .l2 B1, 7, B0 +|| [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + ;; RETURN A may happen here (note: must happen before the next branch) + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 +|| [B0] b .s1 _remu_loop + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + ;; loop backwards branch happens here + + ret .s2 B3 + [B1] subc .l1x A4,B4,A4 +|| [B1] add .s2 -1, B1, B1 + [B1] subc .l1x A4,B4,A4 + + extu .s1 A4, A7, A4 + nop 2 +ENDPROC(__c6xabi_remu) diff --git a/arch/c6x/lib/strasgi.S b/arch/c6x/lib/strasgi.S new file mode 100644 index 0000000..de274076 --- /dev/null +++ b/arch/c6x/lib/strasgi.S @@ -0,0 +1,89 @@ +;; Copyright 2010 Free Software Foundation, Inc. +;; Contributed by Bernd Schmidt <bernds@codesourcery.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +#include <linux/linkage.h> + + .text + +ENTRY(__c6xabi_strasgi) + ;; This is essentially memcpy, with alignment known to be at least + ;; 4, and the size a multiple of 4 greater than or equal to 28. + ldw .d2t1 *B4++, A0 +|| mvk .s2 16, B1 + ldw .d2t1 *B4++, A1 +|| mvk .s2 20, B2 +|| sub .d1 A6, 24, A6 + ldw .d2t1 *B4++, A5 + ldw .d2t1 *B4++, A7 +|| mv .l2x A6, B7 + ldw .d2t1 *B4++, A8 + ldw .d2t1 *B4++, A9 +|| mv .s2x A0, B5 +|| cmpltu .l2 B2, B7, B0 + +_strasgi_loop: + stw .d1t2 B5, *A4++ +|| [B0] ldw .d2t1 *B4++, A0 +|| mv .s2x A1, B5 +|| mv .l2 B7, B6 + + [B0] sub .d2 B6, 24, B7 +|| [B0] b .s2 _strasgi_loop +|| cmpltu .l2 B1, B6, B0 + + [B0] ldw .d2t1 *B4++, A1 +|| stw .d1t2 B5, *A4++ +|| mv .s2x A5, B5 +|| cmpltu .l2 12, B6, B0 + + [B0] ldw .d2t1 *B4++, A5 +|| stw .d1t2 B5, *A4++ +|| mv .s2x A7, B5 +|| cmpltu .l2 8, B6, B0 + + [B0] ldw .d2t1 *B4++, A7 +|| stw .d1t2 B5, *A4++ +|| mv .s2x A8, B5 +|| cmpltu .l2 4, B6, B0 + + [B0] ldw .d2t1 *B4++, A8 +|| stw .d1t2 B5, *A4++ +|| mv .s2x A9, B5 +|| cmpltu .l2 0, B6, B0 + + [B0] ldw .d2t1 *B4++, A9 +|| stw .d1t2 B5, *A4++ +|| mv .s2x A0, B5 +|| cmpltu .l2 B2, B7, B0 + + ;; loop back branch happens here + + cmpltu .l2 B1, B6, B0 +|| ret .s2 b3 + + [B0] stw .d1t1 A1, *A4++ +|| cmpltu .l2 12, B6, B0 + [B0] stw .d1t1 A5, *A4++ +|| cmpltu .l2 8, B6, B0 + [B0] stw .d1t1 A7, *A4++ +|| cmpltu .l2 4, B6, B0 + [B0] stw .d1t1 A8, *A4++ +|| cmpltu .l2 0, B6, B0 + [B0] stw .d1t1 A9, *A4++ + + ;; return happens here +ENDPROC(__c6xabi_strasgi) diff --git a/arch/c6x/lib/strasgi_64plus.S b/arch/c6x/lib/strasgi_64plus.S new file mode 100644 index 0000000..c9fd159 --- /dev/null +++ b/arch/c6x/lib/strasgi_64plus.S @@ -0,0 +1,39 @@ +;; Copyright 2010 Free Software Foundation, Inc. +;; Contributed by Bernd Schmidt <bernds@codesourcery.com>. +;; +;; This program is free software; you can redistribute it and/or modify +;; it under the terms of the GNU General Public License as published by +;; the Free Software Foundation; either version 2 of the License, or +;; (at your option) any later version. +;; +;; This program is distributed in the hope that it will be useful, +;; but WITHOUT ANY WARRANTY; without even the implied warranty of +;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +;; GNU General Public License for more details. +;; +;; You should have received a copy of the GNU General Public License +;; along with this program; if not, write to the Free Software +;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + +#include <linux/linkage.h> + + .text + +ENTRY(__c6xabi_strasgi_64plus) + shru .s2x a6, 2, b31 +|| mv .s1 a4, a30 +|| mv .d2 b4, b30 + + add .s2 -4, b31, b31 + + sploopd 1 +|| mvc .s2 b31, ilc + ldw .d2t2 *b30++, b31 + nop 4 + mv .s1x b31,a31 + spkernel 6, 0 +|| stw .d1t1 a31, *a30++ + + ret .s2 b3 + nop 5 +ENDPROC(__c6xabi_strasgi_64plus) diff --git a/arch/c6x/mm/Makefile b/arch/c6x/mm/Makefile new file mode 100644 index 0000000..136a975 --- /dev/null +++ b/arch/c6x/mm/Makefile @@ -0,0 +1,5 @@ +# +# Makefile for the linux c6x-specific parts of the memory manager. +# + +obj-y := init.o dma-coherent.o diff --git a/arch/c6x/mm/dma-coherent.c b/arch/c6x/mm/dma-coherent.c new file mode 100644 index 0000000..4187e51 --- /dev/null +++ b/arch/c6x/mm/dma-coherent.c @@ -0,0 +1,143 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot <aurelien.jacquiot@ti.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * DMA uncached mapping support. + * + * Using code pulled from ARM + * Copyright (C) 2000-2004 Russell King + * + */ +#include <linux/slab.h> +#include <linux/bitmap.h> +#include <linux/bitops.h> +#include <linux/module.h> +#include <linux/interrupt.h> +#include <linux/dma-mapping.h> +#include <linux/memblock.h> + +#include <asm/page.h> + +/* + * DMA coherent memory management, can be redefined using the memdma= + * kernel command line + */ + +/* none by default */ +static phys_addr_t dma_base; +static u32 dma_size; +static u32 dma_pages; + +static unsigned long *dma_bitmap; + +/* bitmap lock */ +static DEFINE_SPINLOCK(dma_lock); + +/* + * Return a DMA coherent and contiguous memory chunk from the DMA memory + */ +static inline u32 __alloc_dma_pages(int order) +{ + unsigned long flags; + u32 pos; + + spin_lock_irqsave(&dma_lock, flags); + pos = bitmap_find_free_region(dma_bitmap, dma_pages, order); + spin_unlock_irqrestore(&dma_lock, flags); + + return dma_base + (pos << PAGE_SHIFT); +} + +static void __free_dma_pages(u32 addr, int order) +{ + unsigned long flags; + u32 pos = (addr - dma_base) >> PAGE_SHIFT; + + if (addr < dma_base || (pos + (1 << order)) >= dma_pages) { + printk(KERN_ERR "%s: freeing outside range.\n", __func__); + BUG(); + } + + spin_lock_irqsave(&dma_lock, flags); + bitmap_release_region(dma_bitmap, pos, order); + spin_unlock_irqrestore(&dma_lock, flags); +} + +/* + * Allocate DMA coherent memory space and return both the kernel + * virtual and DMA address for that space. + */ +void *dma_alloc_coherent(struct device *dev, size_t size, + dma_addr_t *handle, gfp_t gfp) +{ + u32 paddr; + int order; + + if (!dma_size || !size) + return NULL; + + order = get_count_order(((size - 1) >> PAGE_SHIFT) + 1); + + paddr = __alloc_dma_pages(order); + + if (handle) + *handle = paddr; + + if (!paddr) + return NULL; + + return phys_to_virt(paddr); +} +EXPORT_SYMBOL(dma_alloc_coherent); + +/* + * Free DMA coherent memory as defined by the above mapping. + */ +void dma_free_coherent(struct device *dev, size_t size, void *vaddr, + dma_addr_t dma_handle) +{ + int order; + + if (!dma_size || !size) + return; + + order = get_count_order(((size - 1) >> PAGE_SHIFT) + 1); + + __free_dma_pages(virt_to_phys(vaddr), order); +} +EXPORT_SYMBOL(dma_free_coherent); + +/* + * Initialise the coherent DMA memory allocator using the given uncached region. + */ +void __init coherent_mem_init(phys_addr_t start, u32 size) +{ + phys_addr_t bitmap_phys; + + if (!size) + return; + + printk(KERN_INFO + "Coherent memory (DMA) region start=0x%x size=0x%x\n", + start, size); + + dma_base = start; + dma_size = size; + + /* allocate bitmap */ + dma_pages = dma_size >> PAGE_SHIFT; + if (dma_size & (PAGE_SIZE - 1)) + ++dma_pages; + + bitmap_phys = memblock_alloc(BITS_TO_LONGS(dma_pages) * sizeof(long), + sizeof(long)); + + dma_bitmap = phys_to_virt(bitmap_phys); + memset(dma_bitmap, 0, dma_pages * PAGE_SIZE); +} diff --git a/arch/c6x/mm/init.c b/arch/c6x/mm/init.c new file mode 100644 index 0000000..89395f0 --- /dev/null +++ b/arch/c6x/mm/init.c @@ -0,0 +1,113 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2004, 2009, 2010, 2011 Texas Instruments Incorporated + * Author: Aurelien Jacquiot (aurelien.jacquiot@jaluna.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/mm.h> +#include <linux/swap.h> +#include <linux/module.h> +#include <linux/bootmem.h> +#ifdef CONFIG_BLK_DEV_RAM +#include <linux/blkdev.h> +#endif +#include <linux/initrd.h> + +#include <asm/sections.h> + +/* + * ZERO_PAGE is a special page that is used for zero-initialized + * data and COW. + */ +unsigned long empty_zero_page; +EXPORT_SYMBOL(empty_zero_page); + +/* + * paging_init() continues the virtual memory environment setup which + * was begun by the code in arch/head.S. + * The parameters are pointers to where to stick the starting and ending + * addresses of available kernel virtual memory. + */ +void __init paging_init(void) +{ + struct pglist_data *pgdat = NODE_DATA(0); + unsigned long zones_size[MAX_NR_ZONES] = {0, }; + + empty_zero_page = (unsigned long) alloc_bootmem_pages(PAGE_SIZE); + memset((void *)empty_zero_page, 0, PAGE_SIZE); + + /* + * Set up user data space + */ + set_fs(KERNEL_DS); + + /* + * Define zones + */ + zones_size[ZONE_NORMAL] = (memory_end - PAGE_OFFSET) >> PAGE_SHIFT; + pgdat->node_zones[ZONE_NORMAL].zone_start_pfn = + __pa(PAGE_OFFSET) >> PAGE_SHIFT; + + free_area_init(zones_size); +} + +void __init mem_init(void) +{ + int codek, datak; + unsigned long tmp; + unsigned long len = memory_end - memory_start; + + high_memory = (void *)(memory_end & PAGE_MASK); + + /* this will put all memory onto the freelists */ + totalram_pages = free_all_bootmem(); + + codek = (_etext - _stext) >> 10; + datak = (_end - _sdata) >> 10; + + tmp = nr_free_pages() << PAGE_SHIFT; + printk(KERN_INFO "Memory: %luk/%luk RAM (%dk kernel code, %dk data)\n", + tmp >> 10, len >> 10, codek, datak); +} + +#ifdef CONFIG_BLK_DEV_INITRD +void __init free_initrd_mem(unsigned long start, unsigned long end) +{ + int pages = 0; + for (; start < end; start += PAGE_SIZE) { + ClearPageReserved(virt_to_page(start)); + init_page_count(virt_to_page(start)); + free_page(start); + totalram_pages++; + pages++; + } + printk(KERN_INFO "Freeing initrd memory: %luk freed\n", + (pages * PAGE_SIZE) >> 10); +} +#endif + +void __init free_initmem(void) +{ + unsigned long addr; + + /* + * The following code should be cool even if these sections + * are not page aligned. + */ + addr = PAGE_ALIGN((unsigned long)(__init_begin)); + + /* next to check that the page we free is not a partial page */ + for (; addr + PAGE_SIZE < (unsigned long)(__init_end); + addr += PAGE_SIZE) { + ClearPageReserved(virt_to_page(addr)); + init_page_count(virt_to_page(addr)); + free_page(addr); + totalram_pages++; + } + printk(KERN_INFO "Freeing unused kernel memory: %dK freed\n", + (int) ((addr - PAGE_ALIGN((long) &__init_begin)) >> 10)); +} diff --git a/arch/c6x/platforms/Kconfig b/arch/c6x/platforms/Kconfig new file mode 100644 index 0000000..401ee67 --- /dev/null +++ b/arch/c6x/platforms/Kconfig @@ -0,0 +1,16 @@ + +config SOC_TMS320C6455 + bool "TMS320C6455" + default n + +config SOC_TMS320C6457 + bool "TMS320C6457" + default n + +config SOC_TMS320C6472 + bool "TMS320C6472" + default n + +config SOC_TMS320C6474 + bool "TMS320C6474" + default n diff --git a/arch/c6x/platforms/Makefile b/arch/c6x/platforms/Makefile new file mode 100644 index 0000000..9a95b9b --- /dev/null +++ b/arch/c6x/platforms/Makefile @@ -0,0 +1,12 @@ +# +# Makefile for arch/c6x/platforms +# +# Copyright 2010, 2011 Texas Instruments Incorporated +# + +obj-y = platform.o cache.o megamod-pic.o pll.o plldata.o timer64.o +obj-y += dscr.o + +# SoC objects +obj-$(CONFIG_SOC_TMS320C6455) += emif.o +obj-$(CONFIG_SOC_TMS320C6457) += emif.o diff --git a/arch/c6x/platforms/cache.c b/arch/c6x/platforms/cache.c new file mode 100644 index 0000000..86318a1 --- /dev/null +++ b/arch/c6x/platforms/cache.c @@ -0,0 +1,445 @@ +/* + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/of.h> +#include <linux/of_address.h> +#include <linux/io.h> + +#include <asm/cache.h> +#include <asm/soc.h> + +/* + * Internal Memory Control Registers for caches + */ +#define IMCR_CCFG 0x0000 +#define IMCR_L1PCFG 0x0020 +#define IMCR_L1PCC 0x0024 +#define IMCR_L1DCFG 0x0040 +#define IMCR_L1DCC 0x0044 +#define IMCR_L2ALLOC0 0x2000 +#define IMCR_L2ALLOC1 0x2004 +#define IMCR_L2ALLOC2 0x2008 +#define IMCR_L2ALLOC3 0x200c +#define IMCR_L2WBAR 0x4000 +#define IMCR_L2WWC 0x4004 +#define IMCR_L2WIBAR 0x4010 +#define IMCR_L2WIWC 0x4014 +#define IMCR_L2IBAR 0x4018 +#define IMCR_L2IWC 0x401c +#define IMCR_L1PIBAR 0x4020 +#define IMCR_L1PIWC 0x4024 +#define IMCR_L1DWIBAR 0x4030 +#define IMCR_L1DWIWC 0x4034 +#define IMCR_L1DWBAR 0x4040 +#define IMCR_L1DWWC 0x4044 +#define IMCR_L1DIBAR 0x4048 +#define IMCR_L1DIWC 0x404c +#define IMCR_L2WB 0x5000 +#define IMCR_L2WBINV 0x5004 +#define IMCR_L2INV 0x5008 +#define IMCR_L1PINV 0x5028 +#define IMCR_L1DWB 0x5040 +#define IMCR_L1DWBINV 0x5044 +#define IMCR_L1DINV 0x5048 +#define IMCR_MAR_BASE 0x8000 +#define IMCR_MAR96_111 0x8180 +#define IMCR_MAR128_191 0x8200 +#define IMCR_MAR224_239 0x8380 +#define IMCR_L2MPFAR 0xa000 +#define IMCR_L2MPFSR 0xa004 +#define IMCR_L2MPFCR 0xa008 +#define IMCR_L2MPLK0 0xa100 +#define IMCR_L2MPLK1 0xa104 +#define IMCR_L2MPLK2 0xa108 +#define IMCR_L2MPLK3 0xa10c +#define IMCR_L2MPLKCMD 0xa110 +#define IMCR_L2MPLKSTAT 0xa114 +#define IMCR_L2MPPA_BASE 0xa200 +#define IMCR_L1PMPFAR 0xa400 +#define IMCR_L1PMPFSR 0xa404 +#define IMCR_L1PMPFCR 0xa408 +#define IMCR_L1PMPLK0 0xa500 +#define IMCR_L1PMPLK1 0xa504 +#define IMCR_L1PMPLK2 0xa508 +#define IMCR_L1PMPLK3 0xa50c +#define IMCR_L1PMPLKCMD 0xa510 +#define IMCR_L1PMPLKSTAT 0xa514 +#define IMCR_L1PMPPA_BASE 0xa600 +#define IMCR_L1DMPFAR 0xac00 +#define IMCR_L1DMPFSR 0xac04 +#define IMCR_L1DMPFCR 0xac08 +#define IMCR_L1DMPLK0 0xad00 +#define IMCR_L1DMPLK1 0xad04 +#define IMCR_L1DMPLK2 0xad08 +#define IMCR_L1DMPLK3 0xad0c +#define IMCR_L1DMPLKCMD 0xad10 +#define IMCR_L1DMPLKSTAT 0xad14 +#define IMCR_L1DMPPA_BASE 0xae00 +#define IMCR_L2PDWAKE0 0xc040 +#define IMCR_L2PDWAKE1 0xc044 +#define IMCR_L2PDSLEEP0 0xc050 +#define IMCR_L2PDSLEEP1 0xc054 +#define IMCR_L2PDSTAT0 0xc060 +#define IMCR_L2PDSTAT1 0xc064 + +/* + * CCFG register values and bits + */ +#define L2MODE_0K_CACHE 0x0 +#define L2MODE_32K_CACHE 0x1 +#define L2MODE_64K_CACHE 0x2 +#define L2MODE_128K_CACHE 0x3 +#define L2MODE_256K_CACHE 0x7 + +#define L2PRIO_URGENT 0x0 +#define L2PRIO_HIGH 0x1 +#define L2PRIO_MEDIUM 0x2 +#define L2PRIO_LOW 0x3 + +#define CCFG_ID 0x100 /* Invalidate L1P bit */ +#define CCFG_IP 0x200 /* Invalidate L1D bit */ + +static void __iomem *cache_base; + +/* + * L1 & L2 caches generic functions + */ +#define imcr_get(reg) soc_readl(cache_base + (reg)) +#define imcr_set(reg, value) \ +do { \ + soc_writel((value), cache_base + (reg)); \ + soc_readl(cache_base + (reg)); \ +} while (0) + +static void cache_block_operation_wait(unsigned int wc_reg) +{ + /* Wait for completion */ + while (imcr_get(wc_reg)) + cpu_relax(); +} + +static DEFINE_SPINLOCK(cache_lock); + +/* + * Generic function to perform a block cache operation as + * invalidate or writeback/invalidate + */ +static void cache_block_operation(unsigned int *start, + unsigned int *end, + unsigned int bar_reg, + unsigned int wc_reg) +{ + unsigned long flags; + unsigned int wcnt = + (L2_CACHE_ALIGN_CNT((unsigned int) end) + - L2_CACHE_ALIGN_LOW((unsigned int) start)) >> 2; + unsigned int wc = 0; + + for (; wcnt; wcnt -= wc, start += wc) { +loop: + spin_lock_irqsave(&cache_lock, flags); + + /* + * If another cache operation is occuring + */ + if (unlikely(imcr_get(wc_reg))) { + spin_unlock_irqrestore(&cache_lock, flags); + + /* Wait for previous operation completion */ + cache_block_operation_wait(wc_reg); + + /* Try again */ + goto loop; + } + + imcr_set(bar_reg, L2_CACHE_ALIGN_LOW((unsigned int) start)); + + if (wcnt > 0xffff) + wc = 0xffff; + else + wc = wcnt; + + /* Set word count value in the WC register */ + imcr_set(wc_reg, wc & 0xffff); + + spin_unlock_irqrestore(&cache_lock, flags); + + /* Wait for completion */ + cache_block_operation_wait(wc_reg); + } +} + +static void cache_block_operation_nowait(unsigned int *start, + unsigned int *end, + unsigned int bar_reg, + unsigned int wc_reg) +{ + unsigned long flags; + unsigned int wcnt = + (L2_CACHE_ALIGN_CNT((unsigned int) end) + - L2_CACHE_ALIGN_LOW((unsigned int) start)) >> 2; + unsigned int wc = 0; + + for (; wcnt; wcnt -= wc, start += wc) { + + spin_lock_irqsave(&cache_lock, flags); + + imcr_set(bar_reg, L2_CACHE_ALIGN_LOW((unsigned int) start)); + + if (wcnt > 0xffff) + wc = 0xffff; + else + wc = wcnt; + + /* Set word count value in the WC register */ + imcr_set(wc_reg, wc & 0xffff); + + spin_unlock_irqrestore(&cache_lock, flags); + + /* Don't wait for completion on last cache operation */ + if (wcnt > 0xffff) + cache_block_operation_wait(wc_reg); + } +} + +/* + * L1 caches management + */ + +/* + * Disable L1 caches + */ +void L1_cache_off(void) +{ + unsigned int dummy; + + imcr_set(IMCR_L1PCFG, 0); + dummy = imcr_get(IMCR_L1PCFG); + + imcr_set(IMCR_L1DCFG, 0); + dummy = imcr_get(IMCR_L1DCFG); +} + +/* + * Enable L1 caches + */ +void L1_cache_on(void) +{ + unsigned int dummy; + + imcr_set(IMCR_L1PCFG, 7); + dummy = imcr_get(IMCR_L1PCFG); + + imcr_set(IMCR_L1DCFG, 7); + dummy = imcr_get(IMCR_L1DCFG); +} + +/* + * L1P global-invalidate all + */ +void L1P_cache_global_invalidate(void) +{ + unsigned int set = 1; + imcr_set(IMCR_L1PINV, set); + while (imcr_get(IMCR_L1PINV) & 1) + cpu_relax(); +} + +/* + * L1D global-invalidate all + * + * Warning: this operation causes all updated data in L1D to + * be discarded rather than written back to the lower levels of + * memory + */ +void L1D_cache_global_invalidate(void) +{ + unsigned int set = 1; + imcr_set(IMCR_L1DINV, set); + while (imcr_get(IMCR_L1DINV) & 1) + cpu_relax(); +} + +void L1D_cache_global_writeback(void) +{ + unsigned int set = 1; + imcr_set(IMCR_L1DWB, set); + while (imcr_get(IMCR_L1DWB) & 1) + cpu_relax(); +} + +void L1D_cache_global_writeback_invalidate(void) +{ + unsigned int set = 1; + imcr_set(IMCR_L1DWBINV, set); + while (imcr_get(IMCR_L1DWBINV) & 1) + cpu_relax(); +} + +/* + * L2 caches management + */ + +/* + * Set L2 operation mode + */ +void L2_cache_set_mode(unsigned int mode) +{ + unsigned int ccfg = imcr_get(IMCR_CCFG); + + /* Clear and set the L2MODE bits in CCFG */ + ccfg &= ~7; + ccfg |= (mode & 7); + imcr_set(IMCR_CCFG, ccfg); + ccfg = imcr_get(IMCR_CCFG); +} + +/* + * L2 global-writeback and global-invalidate all + */ +void L2_cache_global_writeback_invalidate(void) +{ + imcr_set(IMCR_L2WBINV, 1); + while (imcr_get(IMCR_L2WBINV)) + cpu_relax(); +} + +/* + * L2 global-writeback all + */ +void L2_cache_global_writeback(void) +{ + imcr_set(IMCR_L2WB, 1); + while (imcr_get(IMCR_L2WB)) + cpu_relax(); +} + +/* + * Cacheability controls + */ +void enable_caching(unsigned long start, unsigned long end) +{ + unsigned int mar = IMCR_MAR_BASE + ((start >> 24) << 2); + unsigned int mar_e = IMCR_MAR_BASE + ((end >> 24) << 2); + + for (; mar <= mar_e; mar += 4) + imcr_set(mar, imcr_get(mar) | 1); +} + +void disable_caching(unsigned long start, unsigned long end) +{ + unsigned int mar = IMCR_MAR_BASE + ((start >> 24) << 2); + unsigned int mar_e = IMCR_MAR_BASE + ((end >> 24) << 2); + + for (; mar <= mar_e; mar += 4) + imcr_set(mar, imcr_get(mar) & ~1); +} + + +/* + * L1 block operations + */ +void L1P_cache_block_invalidate(unsigned int start, unsigned int end) +{ + cache_block_operation((unsigned int *) start, + (unsigned int *) end, + IMCR_L1PIBAR, IMCR_L1PIWC); +} + +void L1D_cache_block_invalidate(unsigned int start, unsigned int end) +{ + cache_block_operation((unsigned int *) start, + (unsigned int *) end, + IMCR_L1DIBAR, IMCR_L1DIWC); +} + +void L1D_cache_block_writeback_invalidate(unsigned int start, unsigned int end) +{ + cache_block_operation((unsigned int *) start, + (unsigned int *) end, + IMCR_L1DWIBAR, IMCR_L1DWIWC); +} + +void L1D_cache_block_writeback(unsigned int start, unsigned int end) +{ + cache_block_operation((unsigned int *) start, + (unsigned int *) end, + IMCR_L1DWBAR, IMCR_L1DWWC); +} + +/* + * L2 block operations + */ +void L2_cache_block_invalidate(unsigned int start, unsigned int end) +{ + cache_block_operation((unsigned int *) start, + (unsigned int *) end, + IMCR_L2IBAR, IMCR_L2IWC); +} + +void L2_cache_block_writeback(unsigned int start, unsigned int end) +{ + cache_block_operation((unsigned int *) start, + (unsigned int *) end, + IMCR_L2WBAR, IMCR_L2WWC); +} + +void L2_cache_block_writeback_invalidate(unsigned int start, unsigned int end) +{ + cache_block_operation((unsigned int *) start, + (unsigned int *) end, + IMCR_L2WIBAR, IMCR_L2WIWC); +} + +void L2_cache_block_invalidate_nowait(unsigned int start, unsigned int end) +{ + cache_block_operation_nowait((unsigned int *) start, + (unsigned int *) end, + IMCR_L2IBAR, IMCR_L2IWC); +} + +void L2_cache_block_writeback_nowait(unsigned int start, unsigned int end) +{ + cache_block_operation_nowait((unsigned int *) start, + (unsigned int *) end, + IMCR_L2WBAR, IMCR_L2WWC); +} + +void L2_cache_block_writeback_invalidate_nowait(unsigned int start, + unsigned int end) +{ + cache_block_operation_nowait((unsigned int *) start, + (unsigned int *) end, + IMCR_L2WIBAR, IMCR_L2WIWC); +} + + +/* + * L1 and L2 caches configuration + */ +void __init c6x_cache_init(void) +{ + struct device_node *node; + + node = of_find_compatible_node(NULL, NULL, "ti,c64x+cache"); + if (!node) + return; + + cache_base = of_iomap(node, 0); + + of_node_put(node); + + if (!cache_base) + return; + + /* Set L2 caches on the the whole L2 SRAM memory */ + L2_cache_set_mode(L2MODE_SIZE); + + /* Enable L1 */ + L1_cache_on(); +} diff --git a/arch/c6x/platforms/dscr.c b/arch/c6x/platforms/dscr.c new file mode 100644 index 0000000..f848a65 --- /dev/null +++ b/arch/c6x/platforms/dscr.c @@ -0,0 +1,598 @@ +/* + * Device State Control Registers driver + * + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +/* + * The Device State Control Registers (DSCR) provide SoC level control over + * a number of peripherals. Details vary considerably among the various SoC + * parts. In general, the DSCR block will provide one or more configuration + * registers often protected by a lock register. One or more key values must + * be written to a lock register in order to unlock the configuration register. + * The configuration register may be used to enable (and disable in some + * cases) SoC pin drivers, peripheral clock sources (internal or pin), etc. + * In some cases, a configuration register is write once or the individual + * bits are write once. That is, you may be able to enable a device, but + * will not be able to disable it. + * + * In addition to device configuration, the DSCR block may provide registers + * which are used to reset SoC peripherals, provide device ID information, + * provide MAC addresses, and other miscellaneous functions. + */ + +#include <linux/of.h> +#include <linux/of_address.h> +#include <linux/of_platform.h> +#include <linux/module.h> +#include <linux/io.h> +#include <linux/delay.h> +#include <asm/soc.h> +#include <asm/dscr.h> + +#define MAX_DEVSTATE_IDS 32 +#define MAX_DEVCTL_REGS 8 +#define MAX_DEVSTAT_REGS 8 +#define MAX_LOCKED_REGS 4 +#define MAX_SOC_EMACS 2 + +struct rmii_reset_reg { + u32 reg; + u32 mask; +}; + +/* + * Some registerd may be locked. In order to write to these + * registers, the key value must first be written to the lockreg. + */ +struct locked_reg { + u32 reg; /* offset from base */ + u32 lockreg; /* offset from base */ + u32 key; /* unlock key */ +}; + +/* + * This describes a contiguous area of like control bits used to enable/disable + * SoC devices. Each controllable device is given an ID which is used by the + * individual device drivers to control the device state. These IDs start at + * zero and are assigned sequentially to the control bitfield ranges described + * by this structure. + */ +struct devstate_ctl_reg { + u32 reg; /* register holding the control bits */ + u8 start_id; /* start id of this range */ + u8 num_ids; /* number of devices in this range */ + u8 enable_only; /* bits are write-once to enable only */ + u8 enable; /* value used to enable device */ + u8 disable; /* value used to disable device */ + u8 shift; /* starting (rightmost) bit in range */ + u8 nbits; /* number of bits per device */ +}; + + +/* + * This describes a region of status bits indicating the state of + * various devices. This is used internally to wait for status + * change completion when enabling/disabling a device. Status is + * optional and not all device controls will have a corresponding + * status. + */ +struct devstate_stat_reg { + u32 reg; /* register holding the status bits */ + u8 start_id; /* start id of this range */ + u8 num_ids; /* number of devices in this range */ + u8 enable; /* value indicating enabled state */ + u8 disable; /* value indicating disabled state */ + u8 shift; /* starting (rightmost) bit in range */ + u8 nbits; /* number of bits per device */ +}; + +struct devstate_info { + struct devstate_ctl_reg *ctl; + struct devstate_stat_reg *stat; +}; + +/* These are callbacks to SOC-specific code. */ +struct dscr_ops { + void (*init)(struct device_node *node); +}; + +struct dscr_regs { + spinlock_t lock; + void __iomem *base; + u32 kick_reg[2]; + u32 kick_key[2]; + struct locked_reg locked[MAX_LOCKED_REGS]; + struct devstate_info devstate_info[MAX_DEVSTATE_IDS]; + struct rmii_reset_reg rmii_resets[MAX_SOC_EMACS]; + struct devstate_ctl_reg devctl[MAX_DEVCTL_REGS]; + struct devstate_stat_reg devstat[MAX_DEVSTAT_REGS]; +}; + +static struct dscr_regs dscr; + +static struct locked_reg *find_locked_reg(u32 reg) +{ + int i; + + for (i = 0; i < MAX_LOCKED_REGS; i++) + if (dscr.locked[i].key && reg == dscr.locked[i].reg) + return &dscr.locked[i]; + return NULL; +} + +/* + * Write to a register with one lock + */ +static void dscr_write_locked1(u32 reg, u32 val, + u32 lock, u32 key) +{ + void __iomem *reg_addr = dscr.base + reg; + void __iomem *lock_addr = dscr.base + lock; + + /* + * For some registers, the lock is relocked after a short number + * of cycles. We have to put the lock write and register write in + * the same fetch packet to meet this timing. The .align ensures + * the two stw instructions are in the same fetch packet. + */ + asm volatile ("b .s2 0f\n" + "nop 5\n" + " .align 5\n" + "0:\n" + "stw .D1T2 %3,*%2\n" + "stw .D1T2 %1,*%0\n" + : + : "a"(reg_addr), "b"(val), "a"(lock_addr), "b"(key) + ); + + /* in case the hw doesn't reset the lock */ + soc_writel(0, lock_addr); +} + +/* + * Write to a register protected by two lock registers + */ +static void dscr_write_locked2(u32 reg, u32 val, + u32 lock0, u32 key0, + u32 lock1, u32 key1) +{ + soc_writel(key0, dscr.base + lock0); + soc_writel(key1, dscr.base + lock1); + soc_writel(val, dscr.base + reg); + soc_writel(0, dscr.base + lock0); + soc_writel(0, dscr.base + lock1); +} + +static void dscr_write(u32 reg, u32 val) +{ + struct locked_reg *lock; + + lock = find_locked_reg(reg); + if (lock) + dscr_write_locked1(reg, val, lock->lockreg, lock->key); + else if (dscr.kick_key[0]) + dscr_write_locked2(reg, val, dscr.kick_reg[0], dscr.kick_key[0], + dscr.kick_reg[1], dscr.kick_key[1]); + else + soc_writel(val, dscr.base + reg); +} + + +/* + * Drivers can use this interface to enable/disable SoC IP blocks. + */ +void dscr_set_devstate(int id, enum dscr_devstate_t state) +{ + struct devstate_ctl_reg *ctl; + struct devstate_stat_reg *stat; + struct devstate_info *info; + u32 ctl_val, val; + int ctl_shift, ctl_mask; + unsigned long flags; + + if (!dscr.base) + return; + + if (id < 0 || id >= MAX_DEVSTATE_IDS) + return; + + info = &dscr.devstate_info[id]; + ctl = info->ctl; + stat = info->stat; + + if (ctl == NULL) + return; + + ctl_shift = ctl->shift + ctl->nbits * (id - ctl->start_id); + ctl_mask = ((1 << ctl->nbits) - 1) << ctl_shift; + + switch (state) { + case DSCR_DEVSTATE_ENABLED: + ctl_val = ctl->enable << ctl_shift; + break; + case DSCR_DEVSTATE_DISABLED: + if (ctl->enable_only) + return; + ctl_val = ctl->disable << ctl_shift; + break; + default: + return; + } + + spin_lock_irqsave(&dscr.lock, flags); + + val = soc_readl(dscr.base + ctl->reg); + val &= ~ctl_mask; + val |= ctl_val; + + dscr_write(ctl->reg, val); + + spin_unlock_irqrestore(&dscr.lock, flags); + + if (!stat) + return; + + ctl_shift = stat->shift + stat->nbits * (id - stat->start_id); + + if (state == DSCR_DEVSTATE_ENABLED) + ctl_val = stat->enable; + else + ctl_val = stat->disable; + + do { + val = soc_readl(dscr.base + stat->reg); + val >>= ctl_shift; + val &= ((1 << stat->nbits) - 1); + } while (val != ctl_val); +} +EXPORT_SYMBOL(dscr_set_devstate); + +/* + * Drivers can use this to reset RMII module. + */ +void dscr_rmii_reset(int id, int assert) +{ + struct rmii_reset_reg *r; + unsigned long flags; + u32 val; + + if (id < 0 || id >= MAX_SOC_EMACS) + return; + + r = &dscr.rmii_resets[id]; + if (r->mask == 0) + return; + + spin_lock_irqsave(&dscr.lock, flags); + + val = soc_readl(dscr.base + r->reg); + if (assert) + dscr_write(r->reg, val | r->mask); + else + dscr_write(r->reg, val & ~(r->mask)); + + spin_unlock_irqrestore(&dscr.lock, flags); +} +EXPORT_SYMBOL(dscr_rmii_reset); + +static void __init dscr_parse_devstat(struct device_node *node, + void __iomem *base) +{ + u32 val; + int err; + + err = of_property_read_u32_array(node, "ti,dscr-devstat", &val, 1); + if (!err) + c6x_devstat = soc_readl(base + val); + printk(KERN_INFO "DEVSTAT: %08x\n", c6x_devstat); +} + +static void __init dscr_parse_silicon_rev(struct device_node *node, + void __iomem *base) +{ + u32 vals[3]; + int err; + + err = of_property_read_u32_array(node, "ti,dscr-silicon-rev", vals, 3); + if (!err) { + c6x_silicon_rev = soc_readl(base + vals[0]); + c6x_silicon_rev >>= vals[1]; + c6x_silicon_rev &= vals[2]; + } +} + +/* + * Some SoCs will have a pair of fuse registers which hold + * an ethernet MAC address. The "ti,dscr-mac-fuse-regs" + * property is a mapping from fuse register bytes to MAC + * address bytes. The expected format is: + * + * ti,dscr-mac-fuse-regs = <reg0 b3 b2 b1 b0 + * reg1 b3 b2 b1 b0> + * + * reg0 and reg1 are the offsets of the two fuse registers. + * b3-b0 positionally represent bytes within the fuse register. + * b3 is the most significant byte and b0 is the least. + * Allowable values for b3-b0 are: + * + * 0 = fuse register byte not used in MAC address + * 1-6 = index+1 into c6x_fuse_mac[] + */ +static void __init dscr_parse_mac_fuse(struct device_node *node, + void __iomem *base) +{ + u32 vals[10], fuse; + int f, i, j, err; + + err = of_property_read_u32_array(node, "ti,dscr-mac-fuse-regs", + vals, 10); + if (err) + return; + + for (f = 0; f < 2; f++) { + fuse = soc_readl(base + vals[f * 5]); + for (j = (f * 5) + 1, i = 24; i >= 0; i -= 8, j++) + if (vals[j] && vals[j] <= 6) + c6x_fuse_mac[vals[j] - 1] = fuse >> i; + } +} + +static void __init dscr_parse_rmii_resets(struct device_node *node, + void __iomem *base) +{ + const __be32 *p; + int i, size; + + /* look for RMII reset registers */ + p = of_get_property(node, "ti,dscr-rmii-resets", &size); + if (p) { + /* parse all the reg/mask pairs we can handle */ + size /= (sizeof(*p) * 2); + if (size > MAX_SOC_EMACS) + size = MAX_SOC_EMACS; + + for (i = 0; i < size; i++) { + dscr.rmii_resets[i].reg = be32_to_cpup(p++); + dscr.rmii_resets[i].mask = be32_to_cpup(p++); + } + } +} + + +static void __init dscr_parse_privperm(struct device_node *node, + void __iomem *base) +{ + u32 vals[2]; + int err; + + err = of_property_read_u32_array(node, "ti,dscr-privperm", vals, 2); + if (err) + return; + dscr_write(vals[0], vals[1]); +} + +/* + * SoCs may have "locked" DSCR registers which can only be written + * to only after writing a key value to a lock registers. These + * regisers can be described with the "ti,dscr-locked-regs" property. + * This property provides a list of register descriptions with each + * description consisting of three values. + * + * ti,dscr-locked-regs = <reg0 lockreg0 key0 + * ... + * regN lockregN keyN>; + * + * reg is the offset of the locked register + * lockreg is the offset of the lock register + * key is the unlock key written to lockreg + * + */ +static void __init dscr_parse_locked_regs(struct device_node *node, + void __iomem *base) +{ + struct locked_reg *r; + const __be32 *p; + int i, size; + + p = of_get_property(node, "ti,dscr-locked-regs", &size); + if (p) { + /* parse all the register descriptions we can handle */ + size /= (sizeof(*p) * 3); + if (size > MAX_LOCKED_REGS) + size = MAX_LOCKED_REGS; + + for (i = 0; i < size; i++) { + r = &dscr.locked[i]; + + r->reg = be32_to_cpup(p++); + r->lockreg = be32_to_cpup(p++); + r->key = be32_to_cpup(p++); + } + } +} + +/* + * SoCs may have DSCR registers which are only write enabled after + * writing specific key values to two registers. The two key registers + * and the key values can be parsed from a "ti,dscr-kick-regs" + * propety with the following layout: + * + * ti,dscr-kick-regs = <kickreg0 key0 kickreg1 key1> + * + * kickreg is the offset of the "kick" register + * key is the value which unlocks writing for protected regs + */ +static void __init dscr_parse_kick_regs(struct device_node *node, + void __iomem *base) +{ + u32 vals[4]; + int err; + + err = of_property_read_u32_array(node, "ti,dscr-kick-regs", vals, 4); + if (!err) { + dscr.kick_reg[0] = vals[0]; + dscr.kick_key[0] = vals[1]; + dscr.kick_reg[1] = vals[2]; + dscr.kick_key[1] = vals[3]; + } +} + + +/* + * SoCs may provide controls to enable/disable individual IP blocks. These + * controls in the DSCR usually control pin drivers but also may control + * clocking and or resets. The device tree is used to describe the bitfields + * in registers used to control device state. The number of bits and their + * values may vary even within the same register. + * + * The layout of these bitfields is described by the ti,dscr-devstate-ctl-regs + * property. This property is a list where each element describes a contiguous + * range of control fields with like properties. Each element of the list + * consists of 7 cells with the following values: + * + * start_id num_ids reg enable disable start_bit nbits + * + * start_id is device id for the first device control in the range + * num_ids is the number of device controls in the range + * reg is the offset of the register holding the control bits + * enable is the value to enable a device + * disable is the value to disable a device (0xffffffff if cannot disable) + * start_bit is the bit number of the first bit in the range + * nbits is the number of bits per device control + */ +static void __init dscr_parse_devstate_ctl_regs(struct device_node *node, + void __iomem *base) +{ + struct devstate_ctl_reg *r; + const __be32 *p; + int i, j, size; + + p = of_get_property(node, "ti,dscr-devstate-ctl-regs", &size); + if (p) { + /* parse all the ranges we can handle */ + size /= (sizeof(*p) * 7); + if (size > MAX_DEVCTL_REGS) + size = MAX_DEVCTL_REGS; + + for (i = 0; i < size; i++) { + r = &dscr.devctl[i]; + + r->start_id = be32_to_cpup(p++); + r->num_ids = be32_to_cpup(p++); + r->reg = be32_to_cpup(p++); + r->enable = be32_to_cpup(p++); + r->disable = be32_to_cpup(p++); + if (r->disable == 0xffffffff) + r->enable_only = 1; + r->shift = be32_to_cpup(p++); + r->nbits = be32_to_cpup(p++); + + for (j = r->start_id; + j < (r->start_id + r->num_ids); + j++) + dscr.devstate_info[j].ctl = r; + } + } +} + +/* + * SoCs may provide status registers indicating the state (enabled/disabled) of + * devices on the SoC. The device tree is used to describe the bitfields in + * registers used to provide device status. The number of bits and their + * values used to provide status may vary even within the same register. + * + * The layout of these bitfields is described by the ti,dscr-devstate-stat-regs + * property. This property is a list where each element describes a contiguous + * range of status fields with like properties. Each element of the list + * consists of 7 cells with the following values: + * + * start_id num_ids reg enable disable start_bit nbits + * + * start_id is device id for the first device status in the range + * num_ids is the number of devices covered by the range + * reg is the offset of the register holding the status bits + * enable is the value indicating device is enabled + * disable is the value indicating device is disabled + * start_bit is the bit number of the first bit in the range + * nbits is the number of bits per device status + */ +static void __init dscr_parse_devstate_stat_regs(struct device_node *node, + void __iomem *base) +{ + struct devstate_stat_reg *r; + const __be32 *p; + int i, j, size; + + p = of_get_property(node, "ti,dscr-devstate-stat-regs", &size); + if (p) { + /* parse all the ranges we can handle */ + size /= (sizeof(*p) * 7); + if (size > MAX_DEVSTAT_REGS) + size = MAX_DEVSTAT_REGS; + + for (i = 0; i < size; i++) { + r = &dscr.devstat[i]; + + r->start_id = be32_to_cpup(p++); + r->num_ids = be32_to_cpup(p++); + r->reg = be32_to_cpup(p++); + r->enable = be32_to_cpup(p++); + r->disable = be32_to_cpup(p++); + r->shift = be32_to_cpup(p++); + r->nbits = be32_to_cpup(p++); + + for (j = r->start_id; + j < (r->start_id + r->num_ids); + j++) + dscr.devstate_info[j].stat = r; + } + } +} + +static struct of_device_id dscr_ids[] __initdata = { + { .compatible = "ti,c64x+dscr" }, + {} +}; + +/* + * Probe for DSCR area. + * + * This has to be done early on in case timer or interrupt controller + * needs something. e.g. On C6455 SoC, timer must be enabled through + * DSCR before it is functional. + */ +void __init dscr_probe(void) +{ + struct device_node *node; + void __iomem *base; + + spin_lock_init(&dscr.lock); + + node = of_find_matching_node(NULL, dscr_ids); + if (!node) + return; + + base = of_iomap(node, 0); + if (!base) { + of_node_put(node); + return; + } + + dscr.base = base; + + dscr_parse_devstat(node, base); + dscr_parse_silicon_rev(node, base); + dscr_parse_mac_fuse(node, base); + dscr_parse_rmii_resets(node, base); + dscr_parse_locked_regs(node, base); + dscr_parse_kick_regs(node, base); + dscr_parse_devstate_ctl_regs(node, base); + dscr_parse_devstate_stat_regs(node, base); + dscr_parse_privperm(node, base); +} diff --git a/arch/c6x/platforms/emif.c b/arch/c6x/platforms/emif.c new file mode 100644 index 0000000..8b564de --- /dev/null +++ b/arch/c6x/platforms/emif.c @@ -0,0 +1,87 @@ +/* + * External Memory Interface + * + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/of.h> +#include <linux/of_address.h> +#include <linux/io.h> +#include <asm/soc.h> +#include <asm/dscr.h> + +#define NUM_EMIFA_CHIP_ENABLES 4 + +struct emifa_regs { + u32 midr; + u32 stat; + u32 reserved1[6]; + u32 bprio; + u32 reserved2[23]; + u32 cecfg[NUM_EMIFA_CHIP_ENABLES]; + u32 reserved3[4]; + u32 awcc; + u32 reserved4[7]; + u32 intraw; + u32 intmsk; + u32 intmskset; + u32 intmskclr; +}; + +static struct of_device_id emifa_match[] __initdata = { + { .compatible = "ti,c64x+emifa" }, + {} +}; + +/* + * Parse device tree for existence of an EMIF (External Memory Interface) + * and initialize it if found. + */ +static int __init c6x_emifa_init(void) +{ + struct emifa_regs __iomem *regs; + struct device_node *node; + const __be32 *p; + u32 val; + int i, len, err; + + node = of_find_matching_node(NULL, emifa_match); + if (!node) + return 0; + + regs = of_iomap(node, 0); + if (!regs) + return 0; + + /* look for a dscr-based enable for emifa pin buffers */ + err = of_property_read_u32_array(node, "ti,dscr-dev-enable", &val, 1); + if (!err) + dscr_set_devstate(val, DSCR_DEVSTATE_ENABLED); + + /* set up the chip enables */ + p = of_get_property(node, "ti,emifa-ce-config", &len); + if (p) { + len /= sizeof(u32); + if (len > NUM_EMIFA_CHIP_ENABLES) + len = NUM_EMIFA_CHIP_ENABLES; + for (i = 0; i <= len; i++) + soc_writel(be32_to_cpup(&p[i]), ®s->cecfg[i]); + } + + err = of_property_read_u32_array(node, "ti,emifa-burst-priority", &val, 1); + if (!err) + soc_writel(val, ®s->bprio); + + err = of_property_read_u32_array(node, "ti,emifa-async-wait-control", &val, 1); + if (!err) + soc_writel(val, ®s->awcc); + + iounmap(regs); + of_node_put(node); + return 0; +} +pure_initcall(c6x_emifa_init); diff --git a/arch/c6x/platforms/megamod-pic.c b/arch/c6x/platforms/megamod-pic.c new file mode 100644 index 0000000..7c37a94 --- /dev/null +++ b/arch/c6x/platforms/megamod-pic.c @@ -0,0 +1,349 @@ +/* + * Support for C64x+ Megamodule Interrupt Controller + * + * Copyright (C) 2010, 2011 Texas Instruments Incorporated + * Contributed by: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/module.h> +#include <linux/interrupt.h> +#include <linux/io.h> +#include <linux/of.h> +#include <linux/of_irq.h> +#include <linux/of_address.h> +#include <linux/slab.h> +#include <asm/soc.h> +#include <asm/megamod-pic.h> + +#define NR_COMBINERS 4 +#define NR_MUX_OUTPUTS 12 + +#define IRQ_UNMAPPED 0xffff + +/* + * Megamodule Interrupt Controller register layout + */ +struct megamod_regs { + u32 evtflag[8]; + u32 evtset[8]; + u32 evtclr[8]; + u32 reserved0[8]; + u32 evtmask[8]; + u32 mevtflag[8]; + u32 expmask[8]; + u32 mexpflag[8]; + u32 intmux_unused; + u32 intmux[7]; + u32 reserved1[8]; + u32 aegmux[2]; + u32 reserved2[14]; + u32 intxstat; + u32 intxclr; + u32 intdmask; + u32 reserved3[13]; + u32 evtasrt; +}; + +struct megamod_pic { + struct irq_host *irqhost; + struct megamod_regs __iomem *regs; + raw_spinlock_t lock; + + /* hw mux mapping */ + unsigned int output_to_irq[NR_MUX_OUTPUTS]; +}; + +static struct megamod_pic *mm_pic; + +struct megamod_cascade_data { + struct megamod_pic *pic; + int index; +}; + +static struct megamod_cascade_data cascade_data[NR_COMBINERS]; + +static void mask_megamod(struct irq_data *data) +{ + struct megamod_pic *pic = irq_data_get_irq_chip_data(data); + irq_hw_number_t src = irqd_to_hwirq(data); + u32 __iomem *evtmask = &pic->regs->evtmask[src / 32]; + + raw_spin_lock(&pic->lock); + soc_writel(soc_readl(evtmask) | (1 << (src & 31)), evtmask); + raw_spin_unlock(&pic->lock); +} + +static void unmask_megamod(struct irq_data *data) +{ + struct megamod_pic *pic = irq_data_get_irq_chip_data(data); + irq_hw_number_t src = irqd_to_hwirq(data); + u32 __iomem *evtmask = &pic->regs->evtmask[src / 32]; + + raw_spin_lock(&pic->lock); + soc_writel(soc_readl(evtmask) & ~(1 << (src & 31)), evtmask); + raw_spin_unlock(&pic->lock); +} + +static struct irq_chip megamod_chip = { + .name = "megamod", + .irq_mask = mask_megamod, + .irq_unmask = unmask_megamod, +}; + +static void megamod_irq_cascade(unsigned int irq, struct irq_desc *desc) +{ + struct megamod_cascade_data *cascade; + struct megamod_pic *pic; + u32 events; + int n, idx; + + cascade = irq_desc_get_handler_data(desc); + + pic = cascade->pic; + idx = cascade->index; + + while ((events = soc_readl(&pic->regs->mevtflag[idx])) != 0) { + n = __ffs(events); + + irq = irq_linear_revmap(pic->irqhost, idx * 32 + n); + + soc_writel(1 << n, &pic->regs->evtclr[idx]); + + generic_handle_irq(irq); + } +} + +static int megamod_map(struct irq_host *h, unsigned int virq, + irq_hw_number_t hw) +{ + struct megamod_pic *pic = h->host_data; + int i; + + /* We shouldn't see a hwirq which is muxed to core controller */ + for (i = 0; i < NR_MUX_OUTPUTS; i++) + if (pic->output_to_irq[i] == hw) + return -1; + + irq_set_chip_data(virq, pic); + irq_set_chip_and_handler(virq, &megamod_chip, handle_level_irq); + + /* Set default irq type */ + irq_set_irq_type(virq, IRQ_TYPE_NONE); + + return 0; +} + +static int megamod_xlate(struct irq_host *h, struct device_node *ct, + const u32 *intspec, unsigned int intsize, + irq_hw_number_t *out_hwirq, unsigned int *out_type) + +{ + /* megamod intspecs must have 1 cell */ + BUG_ON(intsize != 1); + *out_hwirq = intspec[0]; + *out_type = IRQ_TYPE_NONE; + return 0; +} + +static struct irq_host_ops megamod_host_ops = { + .map = megamod_map, + .xlate = megamod_xlate, +}; + +static void __init set_megamod_mux(struct megamod_pic *pic, int src, int output) +{ + int index, offset; + u32 val; + + if (src < 0 || src >= (NR_COMBINERS * 32)) { + pic->output_to_irq[output] = IRQ_UNMAPPED; + return; + } + + /* four mappings per mux register */ + index = output / 4; + offset = (output & 3) * 8; + + val = soc_readl(&pic->regs->intmux[index]); + val &= ~(0xff << offset); + val |= src << offset; + soc_writel(val, &pic->regs->intmux[index]); +} + +/* + * Parse the MUX mapping, if one exists. + * + * The MUX map is an array of up to 12 cells; one for each usable core priority + * interrupt. The value of a given cell is the megamodule interrupt source + * which is to me MUXed to the output corresponding to the cell position + * withing the array. The first cell in the array corresponds to priority + * 4 and the last (12th) cell corresponds to priority 15. The allowed + * values are 4 - ((NR_COMBINERS * 32) - 1). Note that the combined interrupt + * sources (0 - 3) are not allowed to be mapped through this property. They + * are handled through the "interrupts" property. This allows us to use a + * value of zero as a "do not map" placeholder. + */ +static void __init parse_priority_map(struct megamod_pic *pic, + int *mapping, int size) +{ + struct device_node *np = pic->irqhost->of_node; + const __be32 *map; + int i, maplen; + u32 val; + + map = of_get_property(np, "ti,c64x+megamod-pic-mux", &maplen); + if (map) { + maplen /= 4; + if (maplen > size) + maplen = size; + + for (i = 0; i < maplen; i++) { + val = be32_to_cpup(map); + if (val && val >= 4) + mapping[i] = val; + ++map; + } + } +} + +static struct megamod_pic * __init init_megamod_pic(struct device_node *np) +{ + struct megamod_pic *pic; + int i, irq; + int mapping[NR_MUX_OUTPUTS]; + + pr_info("Initializing C64x+ Megamodule PIC\n"); + + pic = kzalloc(sizeof(struct megamod_pic), GFP_KERNEL); + if (!pic) { + pr_err("%s: Could not alloc PIC structure.\n", np->full_name); + return NULL; + } + + pic->irqhost = irq_alloc_host(np, IRQ_HOST_MAP_LINEAR, + NR_COMBINERS * 32, &megamod_host_ops, + IRQ_UNMAPPED); + if (!pic->irqhost) { + pr_err("%s: Could not alloc host.\n", np->full_name); + goto error_free; + } + + pic->irqhost->host_data = pic; + + raw_spin_lock_init(&pic->lock); + + pic->regs = of_iomap(np, 0); + if (!pic->regs) { + pr_err("%s: Could not map registers.\n", np->full_name); + goto error_free; + } + + /* Initialize MUX map */ + for (i = 0; i < ARRAY_SIZE(mapping); i++) + mapping[i] = IRQ_UNMAPPED; + + parse_priority_map(pic, mapping, ARRAY_SIZE(mapping)); + + /* + * We can have up to 12 interrupts cascading to the core controller. + * These cascades can be from the combined interrupt sources or for + * individual interrupt sources. The "interrupts" property only + * deals with the cascaded combined interrupts. The individual + * interrupts muxed to the core controller use the core controller + * as their interrupt parent. + */ + for (i = 0; i < NR_COMBINERS; i++) { + + irq = irq_of_parse_and_map(np, i); + if (irq == NO_IRQ) + continue; + + /* + * We count on the core priority interrupts (4 - 15) being + * direct mapped. Check that device tree provided something + * in that range. + */ + if (irq < 4 || irq >= NR_PRIORITY_IRQS) { + pr_err("%s: combiner-%d virq %d out of range!\n", + np->full_name, i, irq); + continue; + } + + /* record the mapping */ + mapping[irq - 4] = i; + + pr_debug("%s: combiner-%d cascading to virq %d\n", + np->full_name, i, irq); + + cascade_data[i].pic = pic; + cascade_data[i].index = i; + + /* mask and clear all events in combiner */ + soc_writel(~0, &pic->regs->evtmask[i]); + soc_writel(~0, &pic->regs->evtclr[i]); + + irq_set_handler_data(irq, &cascade_data[i]); + irq_set_chained_handler(irq, megamod_irq_cascade); + } + + /* Finally, set up the MUX registers */ + for (i = 0; i < NR_MUX_OUTPUTS; i++) { + if (mapping[i] != IRQ_UNMAPPED) { + pr_debug("%s: setting mux %d to priority %d\n", + np->full_name, mapping[i], i + 4); + set_megamod_mux(pic, mapping[i], i); + } + } + + return pic; + +error_free: + kfree(pic); + + return NULL; +} + +/* + * Return next active event after ACK'ing it. + * Return -1 if no events active. + */ +static int get_exception(void) +{ + int i, bit; + u32 mask; + + for (i = 0; i < NR_COMBINERS; i++) { + mask = soc_readl(&mm_pic->regs->mexpflag[i]); + if (mask) { + bit = __ffs(mask); + soc_writel(1 << bit, &mm_pic->regs->evtclr[i]); + return (i * 32) + bit; + } + } + return -1; +} + +static void assert_event(unsigned int val) +{ + soc_writel(val, &mm_pic->regs->evtasrt); +} + +void __init megamod_pic_init(void) +{ + struct device_node *np; + + np = of_find_compatible_node(NULL, NULL, "ti,c64x+megamod-pic"); + if (!np) + return; + + mm_pic = init_megamod_pic(np); + of_node_put(np); + + soc_ops.get_exception = get_exception; + soc_ops.assert_event = assert_event; + + return; +} diff --git a/arch/c6x/platforms/platform.c b/arch/c6x/platforms/platform.c new file mode 100644 index 0000000..26c1a35 --- /dev/null +++ b/arch/c6x/platforms/platform.c @@ -0,0 +1,17 @@ +/* + * Copyright 2011 Texas Instruments Incorporated + * + * This file is licensed under the terms of the GNU General Public License + * version 2. This program is licensed "as is" without any warranty of any + * kind, whether express or implied. + */ + +#include <linux/init.h> +#include <linux/of_platform.h> + +static int __init c6x_device_probe(void) +{ + of_platform_populate(NULL, of_default_bus_match_table, NULL, NULL); + return 0; +} +core_initcall(c6x_device_probe); diff --git a/arch/c6x/platforms/pll.c b/arch/c6x/platforms/pll.c new file mode 100644 index 0000000..3aa898f --- /dev/null +++ b/arch/c6x/platforms/pll.c @@ -0,0 +1,444 @@ +/* + * Clock and PLL control for C64x+ devices + * + * Copyright (C) 2010, 2011 Texas Instruments. + * Contributed by: Mark Salter <msalter@redhat.com> + * + * Copied heavily from arm/mach-davinci/clock.c, so: + * + * Copyright (C) 2006-2007 Texas Instruments. + * Copyright (C) 2008-2009 Deep Root Systems, LLC + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/module.h> +#include <linux/clkdev.h> +#include <linux/clk.h> +#include <linux/io.h> +#include <linux/err.h> + +#include <asm/clock.h> +#include <asm/soc.h> + +static LIST_HEAD(clocks); +static DEFINE_MUTEX(clocks_mutex); +static DEFINE_SPINLOCK(clockfw_lock); + +static void __clk_enable(struct clk *clk) +{ + if (clk->parent) + __clk_enable(clk->parent); + clk->usecount++; +} + +static void __clk_disable(struct clk *clk) +{ + if (WARN_ON(clk->usecount == 0)) + return; + --clk->usecount; + + if (clk->parent) + __clk_disable(clk->parent); +} + +int clk_enable(struct clk *clk) +{ + unsigned long flags; + + if (clk == NULL || IS_ERR(clk)) + return -EINVAL; + + spin_lock_irqsave(&clockfw_lock, flags); + __clk_enable(clk); + spin_unlock_irqrestore(&clockfw_lock, flags); + + return 0; +} +EXPORT_SYMBOL(clk_enable); + +void clk_disable(struct clk *clk) +{ + unsigned long flags; + + if (clk == NULL || IS_ERR(clk)) + return; + + spin_lock_irqsave(&clockfw_lock, flags); + __clk_disable(clk); + spin_unlock_irqrestore(&clockfw_lock, flags); +} +EXPORT_SYMBOL(clk_disable); + +unsigned long clk_get_rate(struct clk *clk) +{ + if (clk == NULL || IS_ERR(clk)) + return -EINVAL; + + return clk->rate; +} +EXPORT_SYMBOL(clk_get_rate); + +long clk_round_rate(struct clk *clk, unsigned long rate) +{ + if (clk == NULL || IS_ERR(clk)) + return -EINVAL; + + if (clk->round_rate) + return clk->round_rate(clk, rate); + + return clk->rate; +} +EXPORT_SYMBOL(clk_round_rate); + +/* Propagate rate to children */ +static void propagate_rate(struct clk *root) +{ + struct clk *clk; + + list_for_each_entry(clk, &root->children, childnode) { + if (clk->recalc) + clk->rate = clk->recalc(clk); + propagate_rate(clk); + } +} + +int clk_set_rate(struct clk *clk, unsigned long rate) +{ + unsigned long flags; + int ret = -EINVAL; + + if (clk == NULL || IS_ERR(clk)) + return ret; + + if (clk->set_rate) + ret = clk->set_rate(clk, rate); + + spin_lock_irqsave(&clockfw_lock, flags); + if (ret == 0) { + if (clk->recalc) + clk->rate = clk->recalc(clk); + propagate_rate(clk); + } + spin_unlock_irqrestore(&clockfw_lock, flags); + + return ret; +} +EXPORT_SYMBOL(clk_set_rate); + +int clk_set_parent(struct clk *clk, struct clk *parent) +{ + unsigned long flags; + + if (clk == NULL || IS_ERR(clk)) + return -EINVAL; + + /* Cannot change parent on enabled clock */ + if (WARN_ON(clk->usecount)) + return -EINVAL; + + mutex_lock(&clocks_mutex); + clk->parent = parent; + list_del_init(&clk->childnode); + list_add(&clk->childnode, &clk->parent->children); + mutex_unlock(&clocks_mutex); + + spin_lock_irqsave(&clockfw_lock, flags); + if (clk->recalc) + clk->rate = clk->recalc(clk); + propagate_rate(clk); + spin_unlock_irqrestore(&clockfw_lock, flags); + + return 0; +} +EXPORT_SYMBOL(clk_set_parent); + +int clk_register(struct clk *clk) +{ + if (clk == NULL || IS_ERR(clk)) + return -EINVAL; + + if (WARN(clk->parent && !clk->parent->rate, + "CLK: %s parent %s has no rate!\n", + clk->name, clk->parent->name)) + return -EINVAL; + + mutex_lock(&clocks_mutex); + list_add_tail(&clk->node, &clocks); + if (clk->parent) + list_add_tail(&clk->childnode, &clk->parent->children); + mutex_unlock(&clocks_mutex); + + /* If rate is already set, use it */ + if (clk->rate) + return 0; + + /* Else, see if there is a way to calculate it */ + if (clk->recalc) + clk->rate = clk->recalc(clk); + + /* Otherwise, default to parent rate */ + else if (clk->parent) + clk->rate = clk->parent->rate; + + return 0; +} +EXPORT_SYMBOL(clk_register); + +void clk_unregister(struct clk *clk) +{ + if (clk == NULL || IS_ERR(clk)) + return; + + mutex_lock(&clocks_mutex); + list_del(&clk->node); + list_del(&clk->childnode); + mutex_unlock(&clocks_mutex); +} +EXPORT_SYMBOL(clk_unregister); + + +static u32 pll_read(struct pll_data *pll, int reg) +{ + return soc_readl(pll->base + reg); +} + +static unsigned long clk_sysclk_recalc(struct clk *clk) +{ + u32 v, plldiv = 0; + struct pll_data *pll; + unsigned long rate = clk->rate; + + if (WARN_ON(!clk->parent)) + return rate; + + rate = clk->parent->rate; + + /* the parent must be a PLL */ + if (WARN_ON(!clk->parent->pll_data)) + return rate; + + pll = clk->parent->pll_data; + + /* If pre-PLL, source clock is before the multiplier and divider(s) */ + if (clk->flags & PRE_PLL) + rate = pll->input_rate; + + if (!clk->div) { + pr_debug("%s: (no divider) rate = %lu KHz\n", + clk->name, rate / 1000); + return rate; + } + + if (clk->flags & FIXED_DIV_PLL) { + rate /= clk->div; + pr_debug("%s: (fixed divide by %d) rate = %lu KHz\n", + clk->name, clk->div, rate / 1000); + return rate; + } + + v = pll_read(pll, clk->div); + if (v & PLLDIV_EN) + plldiv = (v & PLLDIV_RATIO_MASK) + 1; + + if (plldiv == 0) + plldiv = 1; + + rate /= plldiv; + + pr_debug("%s: (divide by %d) rate = %lu KHz\n", + clk->name, plldiv, rate / 1000); + + return rate; +} + +static unsigned long clk_leafclk_recalc(struct clk *clk) +{ + if (WARN_ON(!clk->parent)) + return clk->rate; + + pr_debug("%s: (parent %s) rate = %lu KHz\n", + clk->name, clk->parent->name, clk->parent->rate / 1000); + + return clk->parent->rate; +} + +static unsigned long clk_pllclk_recalc(struct clk *clk) +{ + u32 ctrl, mult = 0, prediv = 0, postdiv = 0; + u8 bypass; + struct pll_data *pll = clk->pll_data; + unsigned long rate = clk->rate; + + if (clk->flags & FIXED_RATE_PLL) + return rate; + + ctrl = pll_read(pll, PLLCTL); + rate = pll->input_rate = clk->parent->rate; + + if (ctrl & PLLCTL_PLLEN) + bypass = 0; + else + bypass = 1; + + if (pll->flags & PLL_HAS_MUL) { + mult = pll_read(pll, PLLM); + mult = (mult & PLLM_PLLM_MASK) + 1; + } + if (pll->flags & PLL_HAS_PRE) { + prediv = pll_read(pll, PLLPRE); + if (prediv & PLLDIV_EN) + prediv = (prediv & PLLDIV_RATIO_MASK) + 1; + else + prediv = 0; + } + if (pll->flags & PLL_HAS_POST) { + postdiv = pll_read(pll, PLLPOST); + if (postdiv & PLLDIV_EN) + postdiv = (postdiv & PLLDIV_RATIO_MASK) + 1; + else + postdiv = 1; + } + + if (!bypass) { + if (prediv) + rate /= prediv; + if (mult) + rate *= mult; + if (postdiv) + rate /= postdiv; + + pr_debug("PLL%d: input = %luMHz, pre[%d] mul[%d] post[%d] " + "--> %luMHz output.\n", + pll->num, clk->parent->rate / 1000000, + prediv, mult, postdiv, rate / 1000000); + } else + pr_debug("PLL%d: input = %luMHz, bypass mode.\n", + pll->num, clk->parent->rate / 1000000); + + return rate; +} + + +static void __init __init_clk(struct clk *clk) +{ + INIT_LIST_HEAD(&clk->node); + INIT_LIST_HEAD(&clk->children); + INIT_LIST_HEAD(&clk->childnode); + + if (!clk->recalc) { + + /* Check if clock is a PLL */ + if (clk->pll_data) + clk->recalc = clk_pllclk_recalc; + + /* Else, if it is a PLL-derived clock */ + else if (clk->flags & CLK_PLL) + clk->recalc = clk_sysclk_recalc; + + /* Otherwise, it is a leaf clock (PSC clock) */ + else if (clk->parent) + clk->recalc = clk_leafclk_recalc; + } +} + +void __init c6x_clks_init(struct clk_lookup *clocks) +{ + struct clk_lookup *c; + struct clk *clk; + size_t num_clocks = 0; + + for (c = clocks; c->clk; c++) { + clk = c->clk; + + __init_clk(clk); + clk_register(clk); + num_clocks++; + + /* Turn on clocks that Linux doesn't otherwise manage */ + if (clk->flags & ALWAYS_ENABLED) + clk_enable(clk); + } + + clkdev_add_table(clocks, num_clocks); +} + +#ifdef CONFIG_DEBUG_FS + +#include <linux/debugfs.h> +#include <linux/seq_file.h> + +#define CLKNAME_MAX 10 /* longest clock name */ +#define NEST_DELTA 2 +#define NEST_MAX 4 + +static void +dump_clock(struct seq_file *s, unsigned nest, struct clk *parent) +{ + char *state; + char buf[CLKNAME_MAX + NEST_DELTA * NEST_MAX]; + struct clk *clk; + unsigned i; + + if (parent->flags & CLK_PLL) + state = "pll"; + else + state = ""; + + /* <nest spaces> name <pad to end> */ + memset(buf, ' ', sizeof(buf) - 1); + buf[sizeof(buf) - 1] = 0; + i = strlen(parent->name); + memcpy(buf + nest, parent->name, + min(i, (unsigned)(sizeof(buf) - 1 - nest))); + + seq_printf(s, "%s users=%2d %-3s %9ld Hz\n", + buf, parent->usecount, state, clk_get_rate(parent)); + /* REVISIT show device associations too */ + + /* cost is now small, but not linear... */ + list_for_each_entry(clk, &parent->children, childnode) { + dump_clock(s, nest + NEST_DELTA, clk); + } +} + +static int c6x_ck_show(struct seq_file *m, void *v) +{ + struct clk *clk; + + /* + * Show clock tree; We trust nonzero usecounts equate to PSC enables... + */ + mutex_lock(&clocks_mutex); + list_for_each_entry(clk, &clocks, node) + if (!clk->parent) + dump_clock(m, 0, clk); + mutex_unlock(&clocks_mutex); + + return 0; +} + +static int c6x_ck_open(struct inode *inode, struct file *file) +{ + return single_open(file, c6x_ck_show, NULL); +} + +static const struct file_operations c6x_ck_operations = { + .open = c6x_ck_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static int __init c6x_clk_debugfs_init(void) +{ + debugfs_create_file("c6x_clocks", S_IFREG | S_IRUGO, NULL, NULL, + &c6x_ck_operations); + + return 0; +} +device_initcall(c6x_clk_debugfs_init); +#endif /* CONFIG_DEBUG_FS */ diff --git a/arch/c6x/platforms/plldata.c b/arch/c6x/platforms/plldata.c new file mode 100644 index 0000000..2cfd6f4 --- /dev/null +++ b/arch/c6x/platforms/plldata.c @@ -0,0 +1,404 @@ +/* + * Port on Texas Instruments TMS320C6x architecture + * + * Copyright (C) 2011 Texas Instruments Incorporated + * Author: Mark Salter <msalter@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/kernel.h> +#include <linux/delay.h> +#include <linux/errno.h> +#include <linux/string.h> +#include <linux/ioport.h> +#include <linux/clkdev.h> +#include <linux/of.h> +#include <linux/of_address.h> + +#include <asm/clock.h> +#include <asm/setup.h> +#include <asm/irq.h> + +/* + * Common SoC clock support. + */ + +/* Default input for PLL1 */ +struct clk clkin1 = { + .name = "clkin1", + .node = LIST_HEAD_INIT(clkin1.node), + .children = LIST_HEAD_INIT(clkin1.children), + .childnode = LIST_HEAD_INIT(clkin1.childnode), +}; + +struct pll_data c6x_soc_pll1 = { + .num = 1, + .sysclks = { + { + .name = "pll1", + .parent = &clkin1, + .pll_data = &c6x_soc_pll1, + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk1", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk2", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk3", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk4", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk5", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk6", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk7", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk8", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk9", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk10", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk11", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk12", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk13", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk14", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk15", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + { + .name = "pll1_sysclk16", + .parent = &c6x_soc_pll1.sysclks[0], + .flags = CLK_PLL, + }, + }, +}; + +/* CPU core clock */ +struct clk c6x_core_clk = { + .name = "core", +}; + +/* miscellaneous IO clocks */ +struct clk c6x_i2c_clk = { + .name = "i2c", +}; + +struct clk c6x_watchdog_clk = { + .name = "watchdog", +}; + +struct clk c6x_mcbsp1_clk = { + .name = "mcbsp1", +}; + +struct clk c6x_mcbsp2_clk = { + .name = "mcbsp2", +}; + +struct clk c6x_mdio_clk = { + .name = "mdio", +}; + + +#ifdef CONFIG_SOC_TMS320C6455 +static struct clk_lookup c6455_clks[] = { + CLK(NULL, "pll1", &c6x_soc_pll1.sysclks[0]), + CLK(NULL, "pll1_sysclk2", &c6x_soc_pll1.sysclks[2]), + CLK(NULL, "pll1_sysclk3", &c6x_soc_pll1.sysclks[3]), + CLK(NULL, "pll1_sysclk4", &c6x_soc_pll1.sysclks[4]), + CLK(NULL, "pll1_sysclk5", &c6x_soc_pll1.sysclks[5]), + CLK(NULL, "core", &c6x_core_clk), + CLK("i2c_davinci.1", NULL, &c6x_i2c_clk), + CLK("watchdog", NULL, &c6x_watchdog_clk), + CLK("2c81800.mdio", NULL, &c6x_mdio_clk), + CLK("", NULL, NULL) +}; + + +static void __init c6455_setup_clocks(struct device_node *node) +{ + struct pll_data *pll = &c6x_soc_pll1; + struct clk *sysclks = pll->sysclks; + + pll->flags = PLL_HAS_PRE | PLL_HAS_MUL; + + sysclks[2].flags |= FIXED_DIV_PLL; + sysclks[2].div = 3; + sysclks[3].flags |= FIXED_DIV_PLL; + sysclks[3].div = 6; + sysclks[4].div = PLLDIV4; + sysclks[5].div = PLLDIV5; + + c6x_core_clk.parent = &sysclks[0]; + c6x_i2c_clk.parent = &sysclks[3]; + c6x_watchdog_clk.parent = &sysclks[3]; + c6x_mdio_clk.parent = &sysclks[3]; + + c6x_clks_init(c6455_clks); +} +#endif /* CONFIG_SOC_TMS320C6455 */ + +#ifdef CONFIG_SOC_TMS320C6457 +static struct clk_lookup c6457_clks[] = { + CLK(NULL, "pll1", &c6x_soc_pll1.sysclks[0]), + CLK(NULL, "pll1_sysclk1", &c6x_soc_pll1.sysclks[1]), + CLK(NULL, "pll1_sysclk2", &c6x_soc_pll1.sysclks[2]), + CLK(NULL, "pll1_sysclk3", &c6x_soc_pll1.sysclks[3]), + CLK(NULL, "pll1_sysclk4", &c6x_soc_pll1.sysclks[4]), + CLK(NULL, "pll1_sysclk5", &c6x_soc_pll1.sysclks[5]), + CLK(NULL, "core", &c6x_core_clk), + CLK("i2c_davinci.1", NULL, &c6x_i2c_clk), + CLK("watchdog", NULL, &c6x_watchdog_clk), + CLK("2c81800.mdio", NULL, &c6x_mdio_clk), + CLK("", NULL, NULL) +}; + +static void __init c6457_setup_clocks(struct device_node *node) +{ + struct pll_data *pll = &c6x_soc_pll1; + struct clk *sysclks = pll->sysclks; + + pll->flags = PLL_HAS_MUL | PLL_HAS_POST; + + sysclks[1].flags |= FIXED_DIV_PLL; + sysclks[1].div = 1; + sysclks[2].flags |= FIXED_DIV_PLL; + sysclks[2].div = 3; + sysclks[3].flags |= FIXED_DIV_PLL; + sysclks[3].div = 6; + sysclks[4].div = PLLDIV4; + sysclks[5].div = PLLDIV5; + + c6x_core_clk.parent = &sysclks[1]; + c6x_i2c_clk.parent = &sysclks[3]; + c6x_watchdog_clk.parent = &sysclks[5]; + c6x_mdio_clk.parent = &sysclks[5]; + + c6x_clks_init(c6457_clks); +} +#endif /* CONFIG_SOC_TMS320C6455 */ + +#ifdef CONFIG_SOC_TMS320C6472 +static struct clk_lookup c6472_clks[] = { + CLK(NULL, "pll1", &c6x_soc_pll1.sysclks[0]), + CLK(NULL, "pll1_sysclk1", &c6x_soc_pll1.sysclks[1]), + CLK(NULL, "pll1_sysclk2", &c6x_soc_pll1.sysclks[2]), + CLK(NULL, "pll1_sysclk3", &c6x_soc_pll1.sysclks[3]), + CLK(NULL, "pll1_sysclk4", &c6x_soc_pll1.sysclks[4]), + CLK(NULL, "pll1_sysclk5", &c6x_soc_pll1.sysclks[5]), + CLK(NULL, "pll1_sysclk6", &c6x_soc_pll1.sysclks[6]), + CLK(NULL, "pll1_sysclk7", &c6x_soc_pll1.sysclks[7]), + CLK(NULL, "pll1_sysclk8", &c6x_soc_pll1.sysclks[8]), + CLK(NULL, "pll1_sysclk9", &c6x_soc_pll1.sysclks[9]), + CLK(NULL, "pll1_sysclk10", &c6x_soc_pll1.sysclks[10]), + CLK(NULL, "core", &c6x_core_clk), + CLK("i2c_davinci.1", NULL, &c6x_i2c_clk), + CLK("watchdog", NULL, &c6x_watchdog_clk), + CLK("2c81800.mdio", NULL, &c6x_mdio_clk), + CLK("", NULL, NULL) +}; + +/* assumptions used for delay loop calculations */ +#define MIN_CLKIN1_KHz 15625 +#define MAX_CORE_KHz 700000 +#define MIN_PLLOUT_KHz MIN_CLKIN1_KHz + +static void __init c6472_setup_clocks(struct device_node *node) +{ + struct pll_data *pll = &c6x_soc_pll1; + struct clk *sysclks = pll->sysclks; + int i; + + pll->flags = PLL_HAS_MUL; + + for (i = 1; i <= 6; i++) { + sysclks[i].flags |= FIXED_DIV_PLL; + sysclks[i].div = 1; + } + + sysclks[7].flags |= FIXED_DIV_PLL; + sysclks[7].div = 3; + sysclks[8].flags |= FIXED_DIV_PLL; + sysclks[8].div = 6; + sysclks[9].flags |= FIXED_DIV_PLL; + sysclks[9].div = 2; + sysclks[10].div = PLLDIV10; + + c6x_core_clk.parent = &sysclks[get_coreid() + 1]; + c6x_i2c_clk.parent = &sysclks[8]; + c6x_watchdog_clk.parent = &sysclks[8]; + c6x_mdio_clk.parent = &sysclks[5]; + + c6x_clks_init(c6472_clks); +} +#endif /* CONFIG_SOC_TMS320C6472 */ + + +#ifdef CONFIG_SOC_TMS320C6474 +static struct clk_lookup c6474_clks[] = { + CLK(NULL, "pll1", &c6x_soc_pll1.sysclks[0]), + CLK(NULL, "pll1_sysclk7", &c6x_soc_pll1.sysclks[7]), + CLK(NULL, "pll1_sysclk9", &c6x_soc_pll1.sysclks[9]), + CLK(NULL, "pll1_sysclk10", &c6x_soc_pll1.sysclks[10]), + CLK(NULL, "pll1_sysclk11", &c6x_soc_pll1.sysclks[11]), + CLK(NULL, "pll1_sysclk12", &c6x_soc_pll1.sysclks[12]), + CLK(NULL, "pll1_sysclk13", &c6x_soc_pll1.sysclks[13]), + CLK(NULL, "core", &c6x_core_clk), + CLK("i2c_davinci.1", NULL, &c6x_i2c_clk), + CLK("mcbsp.1", NULL, &c6x_mcbsp1_clk), + CLK("mcbsp.2", NULL, &c6x_mcbsp2_clk), + CLK("watchdog", NULL, &c6x_watchdog_clk), + CLK("2c81800.mdio", NULL, &c6x_mdio_clk), + CLK("", NULL, NULL) +}; + +static void __init c6474_setup_clocks(struct device_node *node) +{ + struct pll_data *pll = &c6x_soc_pll1; + struct clk *sysclks = pll->sysclks; + + pll->flags = PLL_HAS_MUL; + + sysclks[7].flags |= FIXED_DIV_PLL; + sysclks[7].div = 1; + sysclks[9].flags |= FIXED_DIV_PLL; + sysclks[9].div = 3; + sysclks[10].flags |= FIXED_DIV_PLL; + sysclks[10].div = 6; + + sysclks[11].div = PLLDIV11; + + sysclks[12].flags |= FIXED_DIV_PLL; + sysclks[12].div = 2; + + sysclks[13].div = PLLDIV13; + + c6x_core_clk.parent = &sysclks[7]; + c6x_i2c_clk.parent = &sysclks[10]; + c6x_watchdog_clk.parent = &sysclks[10]; + c6x_mcbsp1_clk.parent = &sysclks[10]; + c6x_mcbsp2_clk.parent = &sysclks[10]; + + c6x_clks_init(c6474_clks); +} +#endif /* CONFIG_SOC_TMS320C6474 */ + +static struct of_device_id c6x_clkc_match[] __initdata = { +#ifdef CONFIG_SOC_TMS320C6455 + { .compatible = "ti,c6455-pll", .data = c6455_setup_clocks }, +#endif +#ifdef CONFIG_SOC_TMS320C6457 + { .compatible = "ti,c6457-pll", .data = c6457_setup_clocks }, +#endif +#ifdef CONFIG_SOC_TMS320C6472 + { .compatible = "ti,c6472-pll", .data = c6472_setup_clocks }, +#endif +#ifdef CONFIG_SOC_TMS320C6474 + { .compatible = "ti,c6474-pll", .data = c6474_setup_clocks }, +#endif + { .compatible = "ti,c64x+pll" }, + {} +}; + +void __init c64x_setup_clocks(void) +{ + void (*__setup_clocks)(struct device_node *np); + struct pll_data *pll = &c6x_soc_pll1; + struct device_node *node; + const struct of_device_id *id; + int err; + u32 val; + + node = of_find_matching_node(NULL, c6x_clkc_match); + if (!node) + return; + + pll->base = of_iomap(node, 0); + if (!pll->base) + goto out; + + err = of_property_read_u32(node, "clock-frequency", &val); + if (err || val == 0) { + pr_err("%s: no clock-frequency found! Using %dMHz\n", + node->full_name, (int)val / 1000000); + val = 25000000; + } + clkin1.rate = val; + + err = of_property_read_u32(node, "ti,c64x+pll-bypass-delay", &val); + if (err) + val = 5000; + pll->bypass_delay = val; + + err = of_property_read_u32(node, "ti,c64x+pll-reset-delay", &val); + if (err) + val = 30000; + pll->reset_delay = val; + + err = of_property_read_u32(node, "ti,c64x+pll-lock-delay", &val); + if (err) + val = 30000; + pll->lock_delay = val; + + /* id->data is a pointer to SoC-specific setup */ + id = of_match_node(c6x_clkc_match, node); + if (id && id->data) { + __setup_clocks = id->data; + __setup_clocks(node); + } + +out: + of_node_put(node); +} diff --git a/arch/c6x/platforms/timer64.c b/arch/c6x/platforms/timer64.c new file mode 100644 index 0000000..03c03c2 --- /dev/null +++ b/arch/c6x/platforms/timer64.c @@ -0,0 +1,244 @@ +/* + * Copyright (C) 2010, 2011 Texas Instruments Incorporated + * Contributed by: Mark Salter (msalter@redhat.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/clockchips.h> +#include <linux/interrupt.h> +#include <linux/io.h> +#include <linux/of.h> +#include <linux/of_irq.h> +#include <linux/of_address.h> +#include <asm/soc.h> +#include <asm/dscr.h> +#include <asm/timer64.h> + +struct timer_regs { + u32 reserved0; + u32 emumgt; + u32 reserved1; + u32 reserved2; + u32 cntlo; + u32 cnthi; + u32 prdlo; + u32 prdhi; + u32 tcr; + u32 tgcr; + u32 wdtcr; +}; + +static struct timer_regs __iomem *timer; + +#define TCR_TSTATLO 0x001 +#define TCR_INVOUTPLO 0x002 +#define TCR_INVINPLO 0x004 +#define TCR_CPLO 0x008 +#define TCR_ENAMODELO_ONCE 0x040 +#define TCR_ENAMODELO_CONT 0x080 +#define TCR_ENAMODELO_MASK 0x0c0 +#define TCR_PWIDLO_MASK 0x030 +#define TCR_CLKSRCLO 0x100 +#define TCR_TIENLO 0x200 +#define TCR_TSTATHI (0x001 << 16) +#define TCR_INVOUTPHI (0x002 << 16) +#define TCR_CPHI (0x008 << 16) +#define TCR_PWIDHI_MASK (0x030 << 16) +#define TCR_ENAMODEHI_ONCE (0x040 << 16) +#define TCR_ENAMODEHI_CONT (0x080 << 16) +#define TCR_ENAMODEHI_MASK (0x0c0 << 16) + +#define TGCR_TIMLORS 0x001 +#define TGCR_TIMHIRS 0x002 +#define TGCR_TIMMODE_UD32 0x004 +#define TGCR_TIMMODE_WDT64 0x008 +#define TGCR_TIMMODE_CD32 0x00c +#define TGCR_TIMMODE_MASK 0x00c +#define TGCR_PSCHI_MASK (0x00f << 8) +#define TGCR_TDDRHI_MASK (0x00f << 12) + +/* + * Timer clocks are divided down from the CPU clock + * The divisor is in the EMUMGTCLKSPD register + */ +#define TIMER_DIVISOR \ + ((soc_readl(&timer->emumgt) & (0xf << 16)) >> 16) + +#define TIMER64_RATE (c6x_core_freq / TIMER_DIVISOR) + +#define TIMER64_MODE_DISABLED 0 +#define TIMER64_MODE_ONE_SHOT TCR_ENAMODELO_ONCE +#define TIMER64_MODE_PERIODIC TCR_ENAMODELO_CONT + +static int timer64_mode; +static int timer64_devstate_id = -1; + +static void timer64_config(unsigned long period) +{ + u32 tcr = soc_readl(&timer->tcr) & ~TCR_ENAMODELO_MASK; + + soc_writel(tcr, &timer->tcr); + soc_writel(period - 1, &timer->prdlo); + soc_writel(0, &timer->cntlo); + tcr |= timer64_mode; + soc_writel(tcr, &timer->tcr); +} + +static void timer64_enable(void) +{ + u32 val; + + if (timer64_devstate_id >= 0) + dscr_set_devstate(timer64_devstate_id, DSCR_DEVSTATE_ENABLED); + + /* disable timer, reset count */ + soc_writel(soc_readl(&timer->tcr) & ~TCR_ENAMODELO_MASK, &timer->tcr); + soc_writel(0, &timer->prdlo); + + /* use internal clock and 1 cycle pulse width */ + val = soc_readl(&timer->tcr); + soc_writel(val & ~(TCR_CLKSRCLO | TCR_PWIDLO_MASK), &timer->tcr); + + /* dual 32-bit unchained mode */ + val = soc_readl(&timer->tgcr) & ~TGCR_TIMMODE_MASK; + soc_writel(val, &timer->tgcr); + soc_writel(val | (TGCR_TIMLORS | TGCR_TIMMODE_UD32), &timer->tgcr); +} + +static void timer64_disable(void) +{ + /* disable timer, reset count */ + soc_writel(soc_readl(&timer->tcr) & ~TCR_ENAMODELO_MASK, &timer->tcr); + soc_writel(0, &timer->prdlo); + + if (timer64_devstate_id >= 0) + dscr_set_devstate(timer64_devstate_id, DSCR_DEVSTATE_DISABLED); +} + +static int next_event(unsigned long delta, + struct clock_event_device *evt) +{ + timer64_config(delta); + return 0; +} + +static void set_clock_mode(enum clock_event_mode mode, + struct clock_event_device *evt) +{ + switch (mode) { + case CLOCK_EVT_MODE_PERIODIC: + timer64_enable(); + timer64_mode = TIMER64_MODE_PERIODIC; + timer64_config(TIMER64_RATE / HZ); + break; + case CLOCK_EVT_MODE_ONESHOT: + timer64_enable(); + timer64_mode = TIMER64_MODE_ONE_SHOT; + break; + case CLOCK_EVT_MODE_UNUSED: + case CLOCK_EVT_MODE_SHUTDOWN: + timer64_mode = TIMER64_MODE_DISABLED; + timer64_disable(); + break; + case CLOCK_EVT_MODE_RESUME: + break; + } +} + +static struct clock_event_device t64_clockevent_device = { + .name = "TIMER64_EVT32_TIMER", + .features = CLOCK_EVT_FEAT_ONESHOT | CLOCK_EVT_FEAT_PERIODIC, + .rating = 200, + .set_mode = set_clock_mode, + .set_next_event = next_event, +}; + +static irqreturn_t timer_interrupt(int irq, void *dev_id) +{ + struct clock_event_device *cd = &t64_clockevent_device; + + cd->event_handler(cd); + + return IRQ_HANDLED; +} + +static struct irqaction timer_iact = { + .name = "timer", + .flags = IRQF_TIMER, + .handler = timer_interrupt, + .dev_id = &t64_clockevent_device, +}; + +void __init timer64_init(void) +{ + struct clock_event_device *cd = &t64_clockevent_device; + struct device_node *np, *first = NULL; + u32 val; + int err, found = 0; + + for_each_compatible_node(np, NULL, "ti,c64x+timer64") { + err = of_property_read_u32(np, "ti,core-mask", &val); + if (!err) { + if (val & (1 << get_coreid())) { + found = 1; + break; + } + } else if (!first) + first = np; + } + if (!found) { + /* try first one with no core-mask */ + if (first) + np = of_node_get(first); + else { + pr_debug("Cannot find ti,c64x+timer64 timer.\n"); + return; + } + } + + timer = of_iomap(np, 0); + if (!timer) { + pr_debug("%s: Cannot map timer registers.\n", np->full_name); + goto out; + } + pr_debug("%s: Timer registers=%p.\n", np->full_name, timer); + + cd->irq = irq_of_parse_and_map(np, 0); + if (cd->irq == NO_IRQ) { + pr_debug("%s: Cannot find interrupt.\n", np->full_name); + iounmap(timer); + goto out; + } + + /* If there is a device state control, save the ID. */ + err = of_property_read_u32(np, "ti,dscr-dev-enable", &val); + if (!err) { + timer64_devstate_id = val; + + /* + * It is necessary to enable the timer block here because + * the TIMER_DIVISOR macro needs to read a timer register + * to get the divisor. + */ + dscr_set_devstate(timer64_devstate_id, DSCR_DEVSTATE_ENABLED); + } + + pr_debug("%s: Timer irq=%d.\n", np->full_name, cd->irq); + + clockevents_calc_mult_shift(cd, c6x_core_freq / TIMER_DIVISOR, 5); + + cd->max_delta_ns = clockevent_delta2ns(0x7fffffff, cd); + cd->min_delta_ns = clockevent_delta2ns(250, cd); + + cd->cpumask = cpumask_of(smp_processor_id()); + + clockevents_register_device(cd); + setup_irq(cd->irq, &timer_iact); + +out: + of_node_put(np); + return; +} diff --git a/arch/cris/Kconfig b/arch/cris/Kconfig index 408b055..b3abfb0 100644 --- a/arch/cris/Kconfig +++ b/arch/cris/Kconfig @@ -19,10 +19,6 @@ config GENERIC_CMOS_UPDATE config ARCH_USES_GETTIMEOFFSET def_bool n -config GENERIC_IOMAP - bool - default y - config ARCH_HAS_ILOG2_U32 bool default n @@ -52,6 +48,7 @@ config CRIS select HAVE_IDE select HAVE_GENERIC_HARDIRQS select GENERIC_IRQ_SHOW + select GENERIC_IOMAP config HZ int diff --git a/arch/cris/arch-v32/drivers/axisflashmap.c b/arch/cris/arch-v32/drivers/axisflashmap.c index a2bde37..b34438e 100644 --- a/arch/cris/arch-v32/drivers/axisflashmap.c +++ b/arch/cris/arch-v32/drivers/axisflashmap.c @@ -404,8 +404,7 @@ static int __init init_axis_flash(void) */ int blockstat; do { - blockstat = main_mtd->block_isbad(main_mtd, - ptable_sector); + blockstat = mtd_block_isbad(main_mtd, ptable_sector); if (blockstat < 0) ptable_sector = 0; /* read error */ else if (blockstat) @@ -413,8 +412,8 @@ static int __init init_axis_flash(void) } while (blockstat && ptable_sector); #endif if (ptable_sector) { - main_mtd->read(main_mtd, ptable_sector, PAGESIZE, - &len, page); + mtd_read(main_mtd, ptable_sector, PAGESIZE, &len, + page); ptable_head = &((struct partitiontable *) page)->head; } diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig index c5e69ab..bbbf7927 100644 --- a/arch/frv/Kconfig +++ b/arch/frv/Kconfig @@ -317,6 +317,7 @@ config PCI bool "Use PCI" depends on MB93090_MB00 default y + select GENERIC_PCI_IOMAP help Some FR-V systems (such as the MB93090-MB00 VDK) have PCI onboard. If you have one of these boards and you wish to use the PCI diff --git a/arch/frv/include/asm/io.h b/arch/frv/include/asm/io.h index ca7475e..8cb50a2 100644 --- a/arch/frv/include/asm/io.h +++ b/arch/frv/include/asm/io.h @@ -21,6 +21,7 @@ #include <asm/virtconvert.h> #include <asm/string.h> #include <asm/mb-regs.h> +#include <asm-generic/pci_iomap.h> #include <linux/delay.h> /* @@ -370,7 +371,6 @@ static inline void iowrite32_rep(void __iomem *p, const void *src, unsigned long /* Create a virtual mapping cookie for a PCI BAR (memory or IO) */ struct pci_dev; -extern void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max); static inline void pci_iounmap(struct pci_dev *dev, void __iomem *p) { } diff --git a/arch/frv/mb93090-mb00/Makefile b/arch/frv/mb93090-mb00/Makefile index b73b542..21f1df1 100644 --- a/arch/frv/mb93090-mb00/Makefile +++ b/arch/frv/mb93090-mb00/Makefile @@ -3,7 +3,7 @@ # ifeq "$(CONFIG_PCI)" "y" -obj-y := pci-frv.o pci-irq.o pci-vdk.o pci-iomap.o +obj-y := pci-frv.o pci-irq.o pci-vdk.o ifeq "$(CONFIG_MMU)" "y" obj-y += pci-dma.o diff --git a/arch/frv/mb93090-mb00/pci-iomap.c b/arch/frv/mb93090-mb00/pci-iomap.c deleted file mode 100644 index 35f6df2..0000000 --- a/arch/frv/mb93090-mb00/pci-iomap.c +++ /dev/null @@ -1,29 +0,0 @@ -/* pci-iomap.c: description - * - * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <linux/pci.h> -#include <linux/module.h> - -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) -{ - resource_size_t start = pci_resource_start(dev, bar); - resource_size_t len = pci_resource_len(dev, bar); - unsigned long flags = pci_resource_flags(dev, bar); - - if (!len || !start) - return NULL; - - if ((flags & IORESOURCE_IO) || (flags & IORESOURCE_MEM)) - return (void __iomem *) start; - - return NULL; -} - -EXPORT_SYMBOL(pci_iomap); diff --git a/arch/hexagon/Kconfig b/arch/hexagon/Kconfig index 02513c2..9059e39 100644 --- a/arch/hexagon/Kconfig +++ b/arch/hexagon/Kconfig @@ -26,6 +26,7 @@ config HEXAGON select HAVE_ARCH_KGDB select HAVE_ARCH_TRACEHOOK select NO_IOPORT + select GENERIC_IOMAP # mostly generic routines, with some accelerated ones ---help--- Qualcomm Hexagon is a processor architecture designed for high @@ -73,9 +74,6 @@ config GENERIC_CSUM config GENERIC_IRQ_PROBE def_bool y -config GENERIC_IOMAP - def_bool y - #config ZONE_DMA # bool # default y diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 3b7a7c4..bd72669 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -32,6 +32,7 @@ config IA64 select GENERIC_IRQ_SHOW select ARCH_WANT_OPTIONAL_GPIOLIB select ARCH_HAVE_NMI_SAFE_CMPXCHG + select GENERIC_IOMAP default y help The Itanium Processor Family is Intel's 64-bit successor to @@ -105,10 +106,6 @@ config EFI bool default y -config GENERIC_IOMAP - bool - default y - config ARCH_CLOCKSOURCE_DATA def_bool y diff --git a/arch/ia64/include/asm/iommu.h b/arch/ia64/include/asm/iommu.h index 105c93b..b6a809f 100644 --- a/arch/ia64/include/asm/iommu.h +++ b/arch/ia64/include/asm/iommu.h @@ -11,10 +11,12 @@ extern void no_iommu_init(void); extern int force_iommu, no_iommu; extern int iommu_pass_through; extern int iommu_detected; +extern int iommu_group_mf; #else #define iommu_pass_through (0) #define no_iommu (1) #define iommu_detected (0) +#define iommu_group_mf (0) #endif extern void iommu_dma_init(void); extern void machvec_init(const char *name); diff --git a/arch/ia64/include/asm/xen/interface.h b/arch/ia64/include/asm/xen/interface.h index 1d2427d..fbb5198 100644 --- a/arch/ia64/include/asm/xen/interface.h +++ b/arch/ia64/include/asm/xen/interface.h @@ -71,7 +71,7 @@ __DEFINE_GUEST_HANDLE(uchar, unsigned char); __DEFINE_GUEST_HANDLE(uint, unsigned int); __DEFINE_GUEST_HANDLE(ulong, unsigned long); -__DEFINE_GUEST_HANDLE(u64, unsigned long); + DEFINE_GUEST_HANDLE(char); DEFINE_GUEST_HANDLE(int); DEFINE_GUEST_HANDLE(long); diff --git a/arch/ia64/kernel/pci-dma.c b/arch/ia64/kernel/pci-dma.c index c16162c..eb11757 100644 --- a/arch/ia64/kernel/pci-dma.c +++ b/arch/ia64/kernel/pci-dma.c @@ -33,6 +33,7 @@ int force_iommu __read_mostly; #endif int iommu_pass_through; +int iommu_group_mf; /* Dummy device used for NULL arguments (normally ISA). Better would be probably a smaller DMA mask, but this is bug-to-bug compatible diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index 43f4c92..4050520 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -774,13 +774,13 @@ struct kvm *kvm_arch_alloc_vm(void) return kvm; } -struct kvm_io_range { +struct kvm_ia64_io_range { unsigned long start; unsigned long size; unsigned long type; }; -static const struct kvm_io_range io_ranges[] = { +static const struct kvm_ia64_io_range io_ranges[] = { {VGA_IO_START, VGA_IO_SIZE, GPFN_FRAME_BUFFER}, {MMIO_START, MMIO_SIZE, GPFN_LOW_MMIO}, {LEGACY_IO_START, LEGACY_IO_SIZE, GPFN_LEGACY_IO}, @@ -1366,14 +1366,12 @@ static void kvm_release_vm_pages(struct kvm *kvm) { struct kvm_memslots *slots; struct kvm_memory_slot *memslot; - int i, j; + int j; unsigned long base_gfn; slots = kvm_memslots(kvm); - for (i = 0; i < slots->nmemslots; i++) { - memslot = &slots->memslots[i]; + kvm_for_each_memslot(memslot, slots) { base_gfn = memslot->base_gfn; - for (j = 0; j < memslot->npages; j++) { if (memslot->rmap[j]) put_page((struct page *)memslot->rmap[j]); @@ -1820,7 +1818,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, if (log->slot >= KVM_MEMORY_SLOTS) goto out; - memslot = &kvm->memslots->memslots[log->slot]; + memslot = id_to_memslot(kvm->memslots, log->slot); r = -ENOENT; if (!memslot->dirty_bitmap) goto out; diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig index 81fdaa7..99c3636 100644 --- a/arch/m68k/Kconfig +++ b/arch/m68k/Kconfig @@ -37,9 +37,6 @@ config GENERIC_CALIBRATE_DELAY bool default y -config GENERIC_IOMAP - def_bool MMU - config GENERIC_CSUM bool @@ -81,6 +78,7 @@ source "kernel/Kconfig.freezer" config MMU bool "MMU-based Paged Memory Management Support" default y + select GENERIC_IOMAP help Select if you want MMU-based virtualised addressing space support by paged memory management. If unsure, say 'Y'. diff --git a/arch/microblaze/Kconfig b/arch/microblaze/Kconfig index e446bab..f0eead7 100644 --- a/arch/microblaze/Kconfig +++ b/arch/microblaze/Kconfig @@ -17,6 +17,7 @@ config MICROBLAZE select HAVE_GENERIC_HARDIRQS select GENERIC_IRQ_PROBE select GENERIC_IRQ_SHOW + select GENERIC_PCI_IOMAP config SWAP def_bool n diff --git a/arch/microblaze/include/asm/irq.h b/arch/microblaze/include/asm/irq.h index cc54187..a175132 100644 --- a/arch/microblaze/include/asm/irq.h +++ b/arch/microblaze/include/asm/irq.h @@ -9,7 +9,14 @@ #ifndef _ASM_MICROBLAZE_IRQ_H #define _ASM_MICROBLAZE_IRQ_H -#define NR_IRQS 32 + +/* + * Linux IRQ# is currently offset by one to map to the hardware + * irq number. So hardware IRQ0 maps to Linux irq 1. + */ +#define NO_IRQ_OFFSET 1 +#define IRQ_OFFSET NO_IRQ_OFFSET +#define NR_IRQS (32 + IRQ_OFFSET) #include <asm-generic/irq.h> /* This type is the placeholder for a hardware interrupt number. It has to @@ -20,8 +27,6 @@ typedef unsigned long irq_hw_number_t; extern unsigned int nr_irq; -#define NO_IRQ (-1) - struct pt_regs; extern void do_IRQ(struct pt_regs *regs); diff --git a/arch/microblaze/include/asm/page.h b/arch/microblaze/include/asm/page.h index ed9d0f6..a25e6b5 100644 --- a/arch/microblaze/include/asm/page.h +++ b/arch/microblaze/include/asm/page.h @@ -174,15 +174,8 @@ extern int page_is_ram(unsigned long pfn); #define virt_addr_valid(vaddr) (pfn_valid(virt_to_pfn(vaddr))) - -# ifndef CONFIG_MMU -# define __pa(vaddr) ((unsigned long) (vaddr)) -# define __va(paddr) ((void *) (paddr)) -# else /* CONFIG_MMU */ -# define __pa(x) __virt_to_phys((unsigned long)(x)) -# define __va(x) ((void *)__phys_to_virt((unsigned long)(x))) -# endif /* CONFIG_MMU */ - +# define __pa(x) __virt_to_phys((unsigned long)(x)) +# define __va(x) ((void *)__phys_to_virt((unsigned long)(x))) /* Convert between virtual and physical address for MMU. */ /* Handle MicroBlaze processor with virtual memory. */ diff --git a/arch/microblaze/include/asm/setup.h b/arch/microblaze/include/asm/setup.h index 904e5ef..6c72ed7 100644 --- a/arch/microblaze/include/asm/setup.h +++ b/arch/microblaze/include/asm/setup.h @@ -26,12 +26,6 @@ int setup_early_printk(char *opt); void remap_early_printk(void); void disable_early_printk(void); -#if defined(CONFIG_EARLY_PRINTK) -#define eprintk early_printk -#else -#define eprintk printk -#endif - void heartbeat(void); void setup_heartbeat(void); diff --git a/arch/microblaze/include/asm/unistd.h b/arch/microblaze/include/asm/unistd.h index 7d7092b..d20ffbc 100644 --- a/arch/microblaze/include/asm/unistd.h +++ b/arch/microblaze/include/asm/unistd.h @@ -391,8 +391,11 @@ #define __NR_clock_adjtime 373 #define __NR_syncfs 374 #define __NR_setns 375 +#define __NR_sendmmsg 376 +#define __NR_process_vm_readv 377 +#define __NR_process_vm_writev 378 -#define __NR_syscalls 376 +#define __NR_syscalls 379 #ifdef __KERNEL__ #ifndef __ASSEMBLY__ diff --git a/arch/microblaze/kernel/early_printk.c b/arch/microblaze/kernel/early_printk.c index d26d92d..8356e47 100644 --- a/arch/microblaze/kernel/early_printk.c +++ b/arch/microblaze/kernel/early_printk.c @@ -50,9 +50,9 @@ static void early_printk_uartlite_write(struct console *unused, const char *s, unsigned n) { while (*s && n-- > 0) { - early_printk_uartlite_putc(*s); if (*s == '\n') early_printk_uartlite_putc('\r'); + early_printk_uartlite_putc(*s); s++; } } @@ -94,9 +94,9 @@ static void early_printk_uart16550_write(struct console *unused, const char *s, unsigned n) { while (*s && n-- > 0) { - early_printk_uart16550_putc(*s); if (*s == '\n') early_printk_uart16550_putc('\r'); + early_printk_uart16550_putc(*s); s++; } } diff --git a/arch/microblaze/kernel/entry.S b/arch/microblaze/kernel/entry.S index ca15bc5..66e34a3 100644 --- a/arch/microblaze/kernel/entry.S +++ b/arch/microblaze/kernel/entry.S @@ -468,7 +468,7 @@ C_ENTRY(sys_fork_wrapper): addi r5, r0, SIGCHLD /* Arg 0: flags */ lwi r6, r1, PT_R1 /* Arg 1: child SP (use parent's) */ addik r7, r1, 0 /* Arg 2: parent context */ - add r8. r0, r0 /* Arg 3: (unused) */ + add r8, r0, r0 /* Arg 3: (unused) */ add r9, r0, r0; /* Arg 4: (unused) */ brid do_fork /* Do real work (tail-call) */ add r10, r0, r0; /* Arg 5: (unused) */ diff --git a/arch/microblaze/kernel/intc.c b/arch/microblaze/kernel/intc.c index eb41441..44b177e 100644 --- a/arch/microblaze/kernel/intc.c +++ b/arch/microblaze/kernel/intc.c @@ -42,8 +42,9 @@ unsigned int nr_irq; static void intc_enable_or_unmask(struct irq_data *d) { - unsigned long mask = 1 << d->irq; - pr_debug("enable_or_unmask: %d\n", d->irq); + unsigned long mask = 1 << d->hwirq; + + pr_debug("enable_or_unmask: %ld\n", d->hwirq); out_be32(INTC_BASE + SIE, mask); /* ack level irqs because they can't be acked during @@ -56,20 +57,21 @@ static void intc_enable_or_unmask(struct irq_data *d) static void intc_disable_or_mask(struct irq_data *d) { - pr_debug("disable: %d\n", d->irq); - out_be32(INTC_BASE + CIE, 1 << d->irq); + pr_debug("disable: %ld\n", d->hwirq); + out_be32(INTC_BASE + CIE, 1 << d->hwirq); } static void intc_ack(struct irq_data *d) { - pr_debug("ack: %d\n", d->irq); - out_be32(INTC_BASE + IAR, 1 << d->irq); + pr_debug("ack: %ld\n", d->hwirq); + out_be32(INTC_BASE + IAR, 1 << d->hwirq); } static void intc_mask_ack(struct irq_data *d) { - unsigned long mask = 1 << d->irq; - pr_debug("disable_and_ack: %d\n", d->irq); + unsigned long mask = 1 << d->hwirq; + + pr_debug("disable_and_ack: %ld\n", d->hwirq); out_be32(INTC_BASE + CIE, mask); out_be32(INTC_BASE + IAR, mask); } @@ -91,7 +93,7 @@ unsigned int get_irq(struct pt_regs *regs) * order to handle multiple interrupt controllers. It currently * is hardcoded to check for interrupts only on the first INTC. */ - irq = in_be32(INTC_BASE + IVR); + irq = in_be32(INTC_BASE + IVR) + NO_IRQ_OFFSET; pr_debug("get_irq: %d\n", irq); return irq; @@ -99,7 +101,7 @@ unsigned int get_irq(struct pt_regs *regs) void __init init_IRQ(void) { - u32 i, j, intr_type; + u32 i, intr_mask; struct device_node *intc = NULL; #ifdef CONFIG_SELFMOD_INTC unsigned int intc_baseaddr = 0; @@ -113,35 +115,24 @@ void __init init_IRQ(void) 0 }; #endif - const char * const intc_list[] = { - "xlnx,xps-intc-1.00.a", - NULL - }; - - for (j = 0; intc_list[j] != NULL; j++) { - intc = of_find_compatible_node(NULL, NULL, intc_list[j]); - if (intc) - break; - } + intc = of_find_compatible_node(NULL, NULL, "xlnx,xps-intc-1.00.a"); BUG_ON(!intc); - intc_baseaddr = be32_to_cpup(of_get_property(intc, - "reg", NULL)); + intc_baseaddr = be32_to_cpup(of_get_property(intc, "reg", NULL)); intc_baseaddr = (unsigned long) ioremap(intc_baseaddr, PAGE_SIZE); nr_irq = be32_to_cpup(of_get_property(intc, "xlnx,num-intr-inputs", NULL)); - intr_type = - be32_to_cpup(of_get_property(intc, - "xlnx,kind-of-intr", NULL)); - if (intr_type > (u32)((1ULL << nr_irq) - 1)) + intr_mask = + be32_to_cpup(of_get_property(intc, "xlnx,kind-of-intr", NULL)); + if (intr_mask > (u32)((1ULL << nr_irq) - 1)) printk(KERN_INFO " ERROR: Mismatch in kind-of-intr param\n"); #ifdef CONFIG_SELFMOD_INTC selfmod_function((int *) arr_func, intc_baseaddr); #endif - printk(KERN_INFO "%s #0 at 0x%08x, num_irq=%d, edge=0x%x\n", - intc_list[j], intc_baseaddr, nr_irq, intr_type); + printk(KERN_INFO "XPS intc #0 at 0x%08x, num_irq=%d, edge=0x%x\n", + intc_baseaddr, nr_irq, intr_mask); /* * Disable all external interrupts until they are @@ -155,8 +146,8 @@ void __init init_IRQ(void) /* Turn on the Master Enable. */ out_be32(intc_baseaddr + MER, MER_HIE | MER_ME); - for (i = 0; i < nr_irq; ++i) { - if (intr_type & (0x00000001 << i)) { + for (i = IRQ_OFFSET; i < (nr_irq + IRQ_OFFSET); ++i) { + if (intr_mask & (0x00000001 << (i - IRQ_OFFSET))) { irq_set_chip_and_handler_name(i, &intc_dev, handle_edge_irq, "edge"); irq_clear_status_flags(i, IRQ_LEVEL); @@ -165,5 +156,6 @@ void __init init_IRQ(void) handle_level_irq, "level"); irq_set_status_flags(i, IRQ_LEVEL); } + irq_get_irq_data(i)->hwirq = i - IRQ_OFFSET; } } diff --git a/arch/microblaze/kernel/irq.c b/arch/microblaze/kernel/irq.c index e5d63a8..bbebcae 100644 --- a/arch/microblaze/kernel/irq.c +++ b/arch/microblaze/kernel/irq.c @@ -33,11 +33,12 @@ void __irq_entry do_IRQ(struct pt_regs *regs) irq_enter(); irq = get_irq(regs); next_irq: - BUG_ON(irq == -1U); - generic_handle_irq(irq); + BUG_ON(!irq); + /* Substract 1 because of get_irq */ + generic_handle_irq(irq + IRQ_OFFSET - NO_IRQ_OFFSET); irq = get_irq(regs); - if (irq != -1U) { + if (irq) { pr_debug("next irq: %d\n", irq); ++concurrent_irq; goto next_irq; @@ -52,13 +53,13 @@ next_irq: intc without any cascades or any connection that's why mapping is 1:1 */ unsigned int irq_create_mapping(struct irq_host *host, irq_hw_number_t hwirq) { - return hwirq; + return hwirq + IRQ_OFFSET; } EXPORT_SYMBOL_GPL(irq_create_mapping); unsigned int irq_create_of_mapping(struct device_node *controller, const u32 *intspec, unsigned int intsize) { - return intspec[0]; + return intspec[0] + IRQ_OFFSET; } EXPORT_SYMBOL_GPL(irq_create_of_mapping); diff --git a/arch/microblaze/kernel/module.c b/arch/microblaze/kernel/module.c index 142426f..f39257a 100644 --- a/arch/microblaze/kernel/module.c +++ b/arch/microblaze/kernel/module.c @@ -100,7 +100,7 @@ int apply_relocate_add(Elf32_Shdr *sechdrs, const char *strtab, break; case R_MICROBLAZE_64_NONE: - pr_debug("R_MICROBLAZE_NONE\n"); + pr_debug("R_MICROBLAZE_64_NONE\n"); break; case R_MICROBLAZE_NONE: diff --git a/arch/microblaze/kernel/setup.c b/arch/microblaze/kernel/setup.c index 0e654a1..604cd9d 100644 --- a/arch/microblaze/kernel/setup.c +++ b/arch/microblaze/kernel/setup.c @@ -145,32 +145,32 @@ void __init machine_early_init(const char *cmdline, unsigned int ram, setup_early_printk(NULL); #endif - eprintk("Ramdisk addr 0x%08x, ", ram); + printk("Ramdisk addr 0x%08x, ", ram); if (fdt) - eprintk("FDT at 0x%08x\n", fdt); + printk("FDT at 0x%08x\n", fdt); else - eprintk("Compiled-in FDT at 0x%08x\n", + printk("Compiled-in FDT at 0x%08x\n", (unsigned int)_fdt_start); #ifdef CONFIG_MTD_UCLINUX - eprintk("Found romfs @ 0x%08x (0x%08x)\n", + printk("Found romfs @ 0x%08x (0x%08x)\n", romfs_base, romfs_size); - eprintk("#### klimit %p ####\n", old_klimit); + printk("#### klimit %p ####\n", old_klimit); BUG_ON(romfs_size < 0); /* What else can we do? */ - eprintk("Moved 0x%08x bytes from 0x%08x to 0x%08x\n", + printk("Moved 0x%08x bytes from 0x%08x to 0x%08x\n", romfs_size, romfs_base, (unsigned)&_ebss); - eprintk("New klimit: 0x%08x\n", (unsigned)klimit); + printk("New klimit: 0x%08x\n", (unsigned)klimit); #endif #if CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR if (msr) - eprintk("!!!Your kernel has setup MSR instruction but " + printk("!!!Your kernel has setup MSR instruction but " "CPU don't have it %x\n", msr); #else if (!msr) - eprintk("!!!Your kernel not setup MSR instruction but " + printk("!!!Your kernel not setup MSR instruction but " "CPU have it %x\n", msr); #endif diff --git a/arch/microblaze/kernel/syscall_table.S b/arch/microblaze/kernel/syscall_table.S index 8789daa..6a2b294 100644 --- a/arch/microblaze/kernel/syscall_table.S +++ b/arch/microblaze/kernel/syscall_table.S @@ -380,3 +380,6 @@ ENTRY(sys_call_table) .long sys_clock_adjtime .long sys_syncfs .long sys_setns /* 375 */ + .long sys_sendmmsg + .long sys_process_vm_readv + .long sys_process_vm_writev diff --git a/arch/microblaze/kernel/timer.c b/arch/microblaze/kernel/timer.c index af74b11..3cb0bf6 100644 --- a/arch/microblaze/kernel/timer.c +++ b/arch/microblaze/kernel/timer.c @@ -243,7 +243,7 @@ static int timer_initialized; void __init time_init(void) { - u32 irq, i = 0; + u32 irq; u32 timer_num = 1; struct device_node *timer = NULL; const void *prop; @@ -258,33 +258,24 @@ void __init time_init(void) 0 }; #endif - const char * const timer_list[] = { - "xlnx,xps-timer-1.00.a", - NULL - }; - - for (i = 0; timer_list[i] != NULL; i++) { - timer = of_find_compatible_node(NULL, NULL, timer_list[i]); - if (timer) - break; - } + timer = of_find_compatible_node(NULL, NULL, "xlnx,xps-timer-1.00.a"); BUG_ON(!timer); timer_baseaddr = be32_to_cpup(of_get_property(timer, "reg", NULL)); timer_baseaddr = (unsigned long) ioremap(timer_baseaddr, PAGE_SIZE); - irq = be32_to_cpup(of_get_property(timer, "interrupts", NULL)); + irq = irq_of_parse_and_map(timer, 0); timer_num = be32_to_cpup(of_get_property(timer, "xlnx,one-timer-only", NULL)); if (timer_num) { - eprintk(KERN_EMERG "Please enable two timers in HW\n"); + printk(KERN_EMERG "Please enable two timers in HW\n"); BUG(); } #ifdef CONFIG_SELFMOD_TIMER selfmod_function((int *) arr_func, timer_baseaddr); #endif - printk(KERN_INFO "%s #0 at 0x%08x, irq=%d\n", - timer_list[i], timer_baseaddr, irq); + printk(KERN_INFO "XPS timer #0 at 0x%08x, irq=%d\n", + timer_baseaddr, irq); /* If there is clock-frequency property than use it */ prop = of_get_property(timer, "clock-frequency", NULL); diff --git a/arch/microblaze/lib/Makefile b/arch/microblaze/lib/Makefile index c13067b..844960e 100644 --- a/arch/microblaze/lib/Makefile +++ b/arch/microblaze/lib/Makefile @@ -20,6 +20,7 @@ lib-y += uaccess_old.o lib-y += ashldi3.o lib-y += ashrdi3.o +lib-y += cmpdi2.o lib-y += divsi3.o lib-y += lshrdi3.o lib-y += modsi3.o diff --git a/arch/microblaze/lib/cmpdi2.c b/arch/microblaze/lib/cmpdi2.c new file mode 100644 index 0000000..a708400 --- /dev/null +++ b/arch/microblaze/lib/cmpdi2.c @@ -0,0 +1,26 @@ +#include <linux/module.h> + +#include "libgcc.h" + +word_type __cmpdi2(long long a, long long b) +{ + const DWunion au = { + .ll = a + }; + const DWunion bu = { + .ll = b + }; + + if (au.s.high < bu.s.high) + return 0; + else if (au.s.high > bu.s.high) + return 2; + + if ((unsigned int) au.s.low < (unsigned int) bu.s.low) + return 0; + else if ((unsigned int) au.s.low > (unsigned int) bu.s.low) + return 2; + + return 1; +} +EXPORT_SYMBOL(__cmpdi2); diff --git a/arch/microblaze/pci/iomap.c b/arch/microblaze/pci/iomap.c index 57acda8..b07abba 100644 --- a/arch/microblaze/pci/iomap.c +++ b/arch/microblaze/pci/iomap.c @@ -10,25 +10,6 @@ #include <asm/io.h> #include <asm/pci-bridge.h> -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max) -{ - resource_size_t start = pci_resource_start(dev, bar); - resource_size_t len = pci_resource_len(dev, bar); - unsigned long flags = pci_resource_flags(dev, bar); - - if (!len) - return NULL; - if (max && len > max) - len = max; - if (flags & IORESOURCE_IO) - return ioport_map(start, len); - if (flags & IORESOURCE_MEM) - return ioremap(start, len); - /* What? */ - return NULL; -} -EXPORT_SYMBOL(pci_iomap); - void pci_iounmap(struct pci_dev *dev, void __iomem *addr) { if (isa_vaddr_is_ioport(addr)) diff --git a/arch/microblaze/pci/pci-common.c b/arch/microblaze/pci/pci-common.c index db841c7..0d71b2e 100644 --- a/arch/microblaze/pci/pci-common.c +++ b/arch/microblaze/pci/pci-common.c @@ -242,7 +242,7 @@ int pci_read_irq_line(struct pci_dev *pci_dev) line, pin); virq = irq_create_mapping(NULL, line); - if (virq != NO_IRQ) + if (virq) irq_set_irq_type(virq, IRQ_TYPE_LEVEL_LOW); } else { pr_debug(" Got one, spec %d cells (0x%08x 0x%08x...) on %s\n", @@ -253,7 +253,7 @@ int pci_read_irq_line(struct pci_dev *pci_dev) virq = irq_create_of_mapping(oirq.controller, oirq.specifier, oirq.size); } - if (virq == NO_IRQ) { + if (!virq) { pr_debug(" Failed to map !\n"); return -1; } diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index a7636d3..29d9218 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -16,6 +16,7 @@ config MIPS select HAVE_FUNCTION_GRAPH_TRACER select HAVE_KPROBES select HAVE_KRETPROBES + select ARCH_BINFMT_ELF_RANDOMIZE_PIE select RTC_LIB if !MACH_LOONGSON select GENERIC_ATOMIC64 if !64BIT select HAVE_DMA_ATTRS @@ -2316,6 +2317,7 @@ config PCI bool "Support for PCI controller" depends on HW_HAS_PCI select PCI_DOMAINS + select GENERIC_PCI_IOMAP help Find out whether you have a PCI motherboard. PCI is the name of a bus system, i.e. the way the CPU talks to the other stuff inside diff --git a/arch/mips/bcm63xx/boards/board_bcm963xx.c b/arch/mips/bcm63xx/boards/board_bcm963xx.c index 40b223b..c223854 100644 --- a/arch/mips/bcm63xx/boards/board_bcm963xx.c +++ b/arch/mips/bcm63xx/boards/board_bcm963xx.c @@ -834,10 +834,13 @@ static struct mtd_partition mtd_partitions[] = { } }; +static const char *bcm63xx_part_types[] = { "bcm63xxpart", NULL }; + static struct physmap_flash_data flash_data = { .width = 2, .nr_parts = ARRAY_SIZE(mtd_partitions), .parts = mtd_partitions, + .part_probe_types = bcm63xx_part_types, }; static struct resource mtd_resources[] = { diff --git a/arch/mips/include/asm/mach-bcm63xx/bcm963xx_tag.h b/arch/mips/include/asm/mach-bcm63xx/bcm963xx_tag.h index ed72e6a..1e6b587 100644 --- a/arch/mips/include/asm/mach-bcm63xx/bcm963xx_tag.h +++ b/arch/mips/include/asm/mach-bcm63xx/bcm963xx_tag.h @@ -16,7 +16,6 @@ #define TAGINFO1_LEN 30 /* Length of vendor information field1 in tag */ #define FLASHLAYOUTVER_LEN 4 /* Length of Flash Layout Version String tag */ #define TAGINFO2_LEN 16 /* Length of vendor information field2 in tag */ -#define CRC_LEN 4 /* Length of CRC in bytes */ #define ALTTAGINFO_LEN 54 /* Alternate length for vendor information; Pirelli */ #define NUM_PIRELLI 2 @@ -77,19 +76,19 @@ struct bcm_tag { /* 192-195: Version flash layout */ char flash_layout_ver[FLASHLAYOUTVER_LEN]; /* 196-199: kernel+rootfs CRC32 */ - char fskernel_crc[CRC_LEN]; + __u32 fskernel_crc; /* 200-215: Unused except on Alice Gate where is is information */ char information2[TAGINFO2_LEN]; /* 216-219: CRC32 of image less imagetag (kernel for Alice Gate) */ - char image_crc[CRC_LEN]; + __u32 image_crc; /* 220-223: CRC32 of rootfs partition */ - char rootfs_crc[CRC_LEN]; + __u32 rootfs_crc; /* 224-227: CRC32 of kernel partition */ - char kernel_crc[CRC_LEN]; + __u32 kernel_crc; /* 228-235: Unused at present */ char reserved1[8]; /* 236-239: CRC32 of header excluding last 20 bytes */ - char header_crc[CRC_LEN]; + __u32 header_crc; /* 240-255: Unused at present */ char reserved2[16]; }; diff --git a/arch/mips/lib/iomap-pci.c b/arch/mips/lib/iomap-pci.c index 2ab899c..2635b1a 100644 --- a/arch/mips/lib/iomap-pci.c +++ b/arch/mips/lib/iomap-pci.c @@ -40,32 +40,6 @@ static void __iomem *ioport_map_pci(struct pci_dev *dev, return (void __iomem *) (ctrl->io_map_base + port); } -/* - * Create a virtual mapping cookie for a PCI BAR (memory or IO) - */ -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) -{ - resource_size_t start = pci_resource_start(dev, bar); - resource_size_t len = pci_resource_len(dev, bar); - unsigned long flags = pci_resource_flags(dev, bar); - - if (!len || !start) - return NULL; - if (maxlen && len > maxlen) - len = maxlen; - if (flags & IORESOURCE_IO) - return ioport_map_pci(dev, start, len); - if (flags & IORESOURCE_MEM) { - if (flags & IORESOURCE_CACHEABLE) - return ioremap(start, len); - return ioremap_nocache(start, len); - } - /* What? */ - return NULL; -} - -EXPORT_SYMBOL(pci_iomap); - void pci_iounmap(struct pci_dev *dev, void __iomem * addr) { iounmap(addr); diff --git a/arch/mn10300/Kconfig b/arch/mn10300/Kconfig index 438db84..8f1c40d 100644 --- a/arch/mn10300/Kconfig +++ b/arch/mn10300/Kconfig @@ -252,6 +252,7 @@ config PCI bool "Use PCI" depends on MN10300_UNIT_ASB2305 default y + select GENERIC_PCI_IOMAP help Some systems (such as the ASB2305) have PCI onboard. If you have one of these boards and you wish to use the PCI facilities, say Y here. diff --git a/arch/mn10300/include/asm/io.h b/arch/mn10300/include/asm/io.h index 787255d..139df8c 100644 --- a/arch/mn10300/include/asm/io.h +++ b/arch/mn10300/include/asm/io.h @@ -229,7 +229,6 @@ static inline void outsl(unsigned long addr, const void *buffer, int count) /* Create a virtual mapping cookie for a PCI BAR (memory or IO) */ struct pci_dev; -extern void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max); static inline void pci_iounmap(struct pci_dev *dev, void __iomem *p) { } @@ -251,15 +250,15 @@ static inline void *phys_to_virt(unsigned long address) /* * Change "struct page" to physical address. */ -static inline void *__ioremap(unsigned long offset, unsigned long size, - unsigned long flags) +static inline void __iomem *__ioremap(unsigned long offset, unsigned long size, + unsigned long flags) { - return (void *) offset; + return (void __iomem *) offset; } -static inline void *ioremap(unsigned long offset, unsigned long size) +static inline void __iomem *ioremap(unsigned long offset, unsigned long size) { - return (void *) offset; + return (void __iomem *) offset; } /* @@ -267,14 +266,14 @@ static inline void *ioremap(unsigned long offset, unsigned long size) * area. it's useful if some control registers are in such an area and write * combining or read caching is not desirable: */ -static inline void *ioremap_nocache(unsigned long offset, unsigned long size) +static inline void __iomem *ioremap_nocache(unsigned long offset, unsigned long size) { - return (void *) (offset | 0x20000000); + return (void __iomem *) (offset | 0x20000000); } #define ioremap_wc ioremap_nocache -static inline void iounmap(void *addr) +static inline void iounmap(void __iomem *addr) { } diff --git a/arch/mn10300/unit-asb2305/Makefile b/arch/mn10300/unit-asb2305/Makefile index 0551022..cbc5aba 100644 --- a/arch/mn10300/unit-asb2305/Makefile +++ b/arch/mn10300/unit-asb2305/Makefile @@ -5,4 +5,4 @@ ############################################################################### obj-y := unit-init.o leds.o -obj-$(CONFIG_PCI) += pci.o pci-asb2305.o pci-irq.o pci-iomap.o +obj-$(CONFIG_PCI) += pci.o pci-asb2305.o pci-irq.o diff --git a/arch/mn10300/unit-asb2305/pci-iomap.c b/arch/mn10300/unit-asb2305/pci-iomap.c deleted file mode 100644 index c1a8d8f..0000000 --- a/arch/mn10300/unit-asb2305/pci-iomap.c +++ /dev/null @@ -1,31 +0,0 @@ -/* ASB2305 PCI I/O mapping handler - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public Licence - * as published by the Free Software Foundation; either version - * 2 of the Licence, or (at your option) any later version. - */ -#include <linux/pci.h> -#include <linux/module.h> - -/* - * Create a virtual mapping cookie for a PCI BAR (memory or IO) - */ -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) -{ - resource_size_t start = pci_resource_start(dev, bar); - resource_size_t len = pci_resource_len(dev, bar); - unsigned long flags = pci_resource_flags(dev, bar); - - if (!len || !start) - return NULL; - - if ((flags & IORESOURCE_IO) || (flags & IORESOURCE_MEM)) - return (void __iomem *) start; - - return NULL; -} -EXPORT_SYMBOL(pci_iomap); diff --git a/arch/openrisc/Kconfig b/arch/openrisc/Kconfig index e518a5a..081a54f 100644 --- a/arch/openrisc/Kconfig +++ b/arch/openrisc/Kconfig @@ -38,9 +38,6 @@ config RWSEM_XCHGADD_ALGORITHM config GENERIC_HWEIGHT def_bool y -config GENERIC_IOMAP - def_bool y - config NO_IOPORT def_bool y diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig index fdfd8be..242a1b7 100644 --- a/arch/parisc/Kconfig +++ b/arch/parisc/Kconfig @@ -14,6 +14,7 @@ config PARISC select GENERIC_ATOMIC64 if !64BIT select HAVE_GENERIC_HARDIRQS select GENERIC_IRQ_PROBE + select GENERIC_PCI_IOMAP select IRQ_PER_CPU select ARCH_HAVE_NMI_SAFE_CMPXCHG diff --git a/arch/parisc/lib/iomap.c b/arch/parisc/lib/iomap.c index 8f470c9..fb8e10a 100644 --- a/arch/parisc/lib/iomap.c +++ b/arch/parisc/lib/iomap.c @@ -436,28 +436,6 @@ void ioport_unmap(void __iomem *addr) } } -/* Create a virtual mapping cookie for a PCI BAR (memory or IO) */ -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) -{ - resource_size_t start = pci_resource_start(dev, bar); - resource_size_t len = pci_resource_len(dev, bar); - unsigned long flags = pci_resource_flags(dev, bar); - - if (!len || !start) - return NULL; - if (maxlen && len > maxlen) - len = maxlen; - if (flags & IORESOURCE_IO) - return ioport_map(start, len); - if (flags & IORESOURCE_MEM) { - if (flags & IORESOURCE_CACHEABLE) - return ioremap(start, len); - return ioremap_nocache(start, len); - } - /* What? */ - return NULL; -} - void pci_iounmap(struct pci_dev *dev, void __iomem * addr) { if (!INDIRECT_ADDR(addr)) { @@ -483,5 +461,4 @@ EXPORT_SYMBOL(iowrite16_rep); EXPORT_SYMBOL(iowrite32_rep); EXPORT_SYMBOL(ioport_map); EXPORT_SYMBOL(ioport_unmap); -EXPORT_SYMBOL(pci_iomap); EXPORT_SYMBOL(pci_iounmap); diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 692ac75..1919634 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -718,6 +718,7 @@ config PCI default PCI_PERMEDIA if !4xx && !CPM2 && !8xx default PCI_QSPAN if !4xx && !CPM2 && 8xx select ARCH_SUPPORTS_MSI + select GENERIC_PCI_IOMAP help Find out whether your system includes a PCI bus. PCI is the name of a bus system, i.e. the way the CPU talks to the other stuff inside diff --git a/arch/powerpc/include/asm/kvm.h b/arch/powerpc/include/asm/kvm.h index 0ad432b..f7727d9 100644 --- a/arch/powerpc/include/asm/kvm.h +++ b/arch/powerpc/include/asm/kvm.h @@ -170,8 +170,8 @@ struct kvm_sregs { } ppc64; struct { __u32 sr[16]; - __u64 ibat[8]; - __u64 dbat[8]; + __u64 ibat[8]; + __u64 dbat[8]; } ppc32; } s; struct { diff --git a/arch/powerpc/kernel/iomap.c b/arch/powerpc/kernel/iomap.c index 2627918..97a3715 100644 --- a/arch/powerpc/kernel/iomap.c +++ b/arch/powerpc/kernel/iomap.c @@ -119,24 +119,6 @@ EXPORT_SYMBOL(ioport_map); EXPORT_SYMBOL(ioport_unmap); #ifdef CONFIG_PCI -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max) -{ - resource_size_t start = pci_resource_start(dev, bar); - resource_size_t len = pci_resource_len(dev, bar); - unsigned long flags = pci_resource_flags(dev, bar); - - if (!len) - return NULL; - if (max && len > max) - len = max; - if (flags & IORESOURCE_IO) - return ioport_map(start, len); - if (flags & IORESOURCE_MEM) - return ioremap(start, len); - /* What? */ - return NULL; -} - void pci_iounmap(struct pci_dev *dev, void __iomem *addr) { if (isa_vaddr_is_ioport(addr)) @@ -146,6 +128,5 @@ void pci_iounmap(struct pci_dev *dev, void __iomem *addr) iounmap(addr); } -EXPORT_SYMBOL(pci_iomap); EXPORT_SYMBOL(pci_iounmap); #endif /* CONFIG_PCI */ diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c index a459479..e41ac6f 100644 --- a/arch/powerpc/kvm/book3s.c +++ b/arch/powerpc/kvm/book3s.c @@ -498,7 +498,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, /* If nothing is dirty, don't bother messing with page tables. */ if (is_dirty) { - memslot = &kvm->memslots->memslots[log->slot]; + memslot = id_to_memslot(kvm->memslots, log->slot); ga = memslot->base_gfn << PAGE_SHIFT; ga_end = ga + (memslot->npages << PAGE_SHIFT); diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c index 286f13d..a795a13 100644 --- a/arch/powerpc/kvm/book3s_hv_builtin.c +++ b/arch/powerpc/kvm/book3s_hv_builtin.c @@ -86,7 +86,7 @@ static inline int lpcr_rmls(unsigned long rma_size) * to allocate contiguous physical memory for the real memory * areas for guests. */ -void kvm_rma_init(void) +void __init kvm_rma_init(void) { unsigned long i; unsigned long j, npages; diff --git a/arch/powerpc/platforms/Kconfig b/arch/powerpc/platforms/Kconfig index 31e1ade..0cfb46d 100644 --- a/arch/powerpc/platforms/Kconfig +++ b/arch/powerpc/platforms/Kconfig @@ -175,9 +175,6 @@ config PPC_INDIRECT_MMIO config PPC_IO_WORKAROUNDS bool -config GENERIC_IOMAP - bool - source "drivers/cpufreq/Kconfig" menu "CPU Frequency drivers" diff --git a/arch/score/Kconfig b/arch/score/Kconfig index 8b0c946..3df65d3 100644 --- a/arch/score/Kconfig +++ b/arch/score/Kconfig @@ -4,6 +4,7 @@ config SCORE def_bool y select HAVE_GENERIC_HARDIRQS select GENERIC_IRQ_SHOW + select GENERIC_IOMAP select HAVE_MEMBLOCK select HAVE_MEMBLOCK_NODE_MAP select ARCH_DISCARD_MEMBLOCK @@ -36,9 +37,6 @@ endmenu config CPU_SCORE7 bool -config GENERIC_IOMAP - def_bool y - config NO_DMA bool default y diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index 47a2f1c..3c8db65 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -85,9 +85,6 @@ config GENERIC_GPIO config GENERIC_CALIBRATE_DELAY bool -config GENERIC_IOMAP - bool - config GENERIC_CLOCKEVENTS def_bool y @@ -861,6 +858,7 @@ config PCI bool "PCI support" depends on SYS_SUPPORTS_PCI select PCI_DOMAINS + select GENERIC_PCI_IOMAP help Find out whether you have a PCI motherboard. PCI is the name of a bus system, i.e. the way the CPU talks to the other stuff inside diff --git a/arch/sh/drivers/pci/pci.c b/arch/sh/drivers/pci/pci.c index c2691af..11aaf2f 100644 --- a/arch/sh/drivers/pci/pci.c +++ b/arch/sh/drivers/pci/pci.c @@ -393,29 +393,6 @@ static void __iomem *ioport_map_pci(struct pci_dev *dev, return (void __iomem *)(chan->io_map_base + port); } -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) -{ - resource_size_t start = pci_resource_start(dev, bar); - resource_size_t len = pci_resource_len(dev, bar); - unsigned long flags = pci_resource_flags(dev, bar); - - if (unlikely(!len || !start)) - return NULL; - if (maxlen && len > maxlen) - len = maxlen; - - if (flags & IORESOURCE_IO) - return ioport_map_pci(dev, start, len); - if (flags & IORESOURCE_MEM) { - if (flags & IORESOURCE_CACHEABLE) - return ioremap(start, len); - return ioremap_nocache(start, len); - } - - return NULL; -} -EXPORT_SYMBOL(pci_iomap); - void pci_iounmap(struct pci_dev *dev, void __iomem *addr) { iounmap(addr); diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index 868ea08..9665799 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -28,6 +28,7 @@ config SPARC select HAVE_GENERIC_HARDIRQS select GENERIC_IRQ_SHOW select USE_GENERIC_SMP_HELPERS if SMP + select GENERIC_PCI_IOMAP config SPARC32 def_bool !64BIT diff --git a/arch/sparc/include/asm/io_32.h b/arch/sparc/include/asm/io_32.h index c2ced21..2006e5d 100644 --- a/arch/sparc/include/asm/io_32.h +++ b/arch/sparc/include/asm/io_32.h @@ -7,6 +7,7 @@ #include <asm/page.h> /* IO address mapping routines need this */ #include <asm/system.h> +#include <asm-generic/pci_iomap.h> #define page_to_phys(page) (page_to_pfn(page) << PAGE_SHIFT) @@ -324,7 +325,6 @@ extern void ioport_unmap(void __iomem *); /* Create a virtual mapping cookie for a PCI BAR (memory or IO) */ struct pci_dev; -extern void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max); extern void pci_iounmap(struct pci_dev *dev, void __iomem *); /* diff --git a/arch/sparc/include/asm/io_64.h b/arch/sparc/include/asm/io_64.h index 9c89654..9481e5a 100644 --- a/arch/sparc/include/asm/io_64.h +++ b/arch/sparc/include/asm/io_64.h @@ -8,6 +8,7 @@ #include <asm/page.h> /* IO address mapping routines need this */ #include <asm/system.h> #include <asm/asi.h> +#include <asm-generic/pci_iomap.h> /* PC crapola... */ #define __SLOW_DOWN_IO do { } while (0) @@ -514,7 +515,6 @@ extern void ioport_unmap(void __iomem *); /* Create a virtual mapping cookie for a PCI BAR (memory or IO) */ struct pci_dev; -extern void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max); extern void pci_iounmap(struct pci_dev *dev, void __iomem *); static inline int sbus_can_dma_64bit(void) diff --git a/arch/sparc/include/asm/signal.h b/arch/sparc/include/asm/signal.h index e49b828..aa42fe3 100644 --- a/arch/sparc/include/asm/signal.h +++ b/arch/sparc/include/asm/signal.h @@ -143,10 +143,11 @@ struct sigstack { #define SA_ONSTACK _SV_SSTACK #define SA_RESTART _SV_INTR #define SA_ONESHOT _SV_RESET -#define SA_NOMASK 0x20u +#define SA_NODEFER 0x20u #define SA_NOCLDWAIT 0x100u #define SA_SIGINFO 0x200u +#define SA_NOMASK SA_NODEFER #define SIG_BLOCK 0x01 /* for blocking signals */ #define SIG_UNBLOCK 0x02 /* for unblocking signals */ diff --git a/arch/sparc/lib/iomap.c b/arch/sparc/lib/iomap.c index 9ef37e1..c4d42a5 100644 --- a/arch/sparc/lib/iomap.c +++ b/arch/sparc/lib/iomap.c @@ -18,31 +18,8 @@ void ioport_unmap(void __iomem *addr) EXPORT_SYMBOL(ioport_map); EXPORT_SYMBOL(ioport_unmap); -/* Create a virtual mapping cookie for a PCI BAR (memory or IO) */ -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) -{ - resource_size_t start = pci_resource_start(dev, bar); - resource_size_t len = pci_resource_len(dev, bar); - unsigned long flags = pci_resource_flags(dev, bar); - - if (!len || !start) - return NULL; - if (maxlen && len > maxlen) - len = maxlen; - if (flags & IORESOURCE_IO) - return ioport_map(start, len); - if (flags & IORESOURCE_MEM) { - if (flags & IORESOURCE_CACHEABLE) - return ioremap(start, len); - return ioremap_nocache(start, len); - } - /* What? */ - return NULL; -} - void pci_iounmap(struct pci_dev *dev, void __iomem * addr) { /* nothing to do */ } -EXPORT_SYMBOL(pci_iomap); EXPORT_SYMBOL(pci_iounmap); diff --git a/arch/tile/Kconfig b/arch/tile/Kconfig index 70a0de4..11270ca 100644 --- a/arch/tile/Kconfig +++ b/arch/tile/Kconfig @@ -321,6 +321,7 @@ config PCI bool "PCI support" default y select PCI_DOMAINS + select GENERIC_PCI_IOMAP ---help--- Enable PCI root complex support, so PCIe endpoint devices can be attached to the Tile chip. Many, but not all, PCI devices diff --git a/arch/tile/include/asm/io.h b/arch/tile/include/asm/io.h index c9ea165..d2152de 100644 --- a/arch/tile/include/asm/io.h +++ b/arch/tile/include/asm/io.h @@ -204,7 +204,8 @@ static inline long ioport_panic(void) static inline void __iomem *ioport_map(unsigned long port, unsigned int len) { - return (void __iomem *) ioport_panic(); + pr_info("ioport_map: mapping IO resources is unsupported on tile.\n"); + return NULL; } static inline void ioport_unmap(void __iomem *addr) diff --git a/arch/tile/include/asm/pci.h b/arch/tile/include/asm/pci.h index 7f03cef..1d25fea 100644 --- a/arch/tile/include/asm/pci.h +++ b/arch/tile/include/asm/pci.h @@ -16,6 +16,7 @@ #define _ASM_TILE_PCI_H #include <linux/pci.h> +#include <asm-generic/pci_iomap.h> /* * Structure of a PCI controller (host bridge) @@ -49,7 +50,6 @@ struct pci_controller { int __devinit tile_pci_init(void); int __devinit pcibios_init(void); -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max); static inline void pci_iounmap(struct pci_dev *dev, void __iomem *addr) {} void __devinit pcibios_fixup_bus(struct pci_bus *bus); diff --git a/arch/tile/kernel/pci.c b/arch/tile/kernel/pci.c index 9d610d3..2556793 100644 --- a/arch/tile/kernel/pci.c +++ b/arch/tile/kernel/pci.c @@ -466,27 +466,6 @@ int pcibios_enable_device(struct pci_dev *dev, int mask) return 0; } -void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max) -{ - unsigned long start = pci_resource_start(dev, bar); - unsigned long len = pci_resource_len(dev, bar); - unsigned long flags = pci_resource_flags(dev, bar); - - if (!len) - return NULL; - if (max && len > max) - len = max; - - if (!(flags & IORESOURCE_MEM)) { - pr_info("PCI: Trying to map invalid resource %#lx\n", flags); - start = 0; - } - - return (void __iomem *)start; -} -EXPORT_SYMBOL(pci_iomap); - - /**************************************************************** * * Tile PCI config space read/write routines diff --git a/arch/unicore32/Kconfig b/arch/unicore32/Kconfig index 942ed61..eeb8054 100644 --- a/arch/unicore32/Kconfig +++ b/arch/unicore32/Kconfig @@ -12,6 +12,7 @@ config UNICORE32 select GENERIC_IRQ_PROBE select GENERIC_IRQ_SHOW select ARCH_WANT_FRAME_POINTERS + select GENERIC_IOMAP help UniCore-32 is 32-bit Instruction Set Architecture, including a series of low-power-consumption RISC chip @@ -30,9 +31,6 @@ config GENERIC_CLOCKEVENTS config GENERIC_CSUM def_bool y -config GENERIC_IOMAP - def_bool y - config NO_IOPORT bool diff --git a/arch/unicore32/include/asm/io.h b/arch/unicore32/include/asm/io.h index 1a5c5a5..adddf6d 100644 --- a/arch/unicore32/include/asm/io.h +++ b/arch/unicore32/include/asm/io.h @@ -37,15 +37,9 @@ extern void __uc32_iounmap(volatile void __iomem *addr); */ #define ioremap(cookie, size) __uc32_ioremap(cookie, size) #define ioremap_cached(cookie, size) __uc32_ioremap_cached(cookie, size) +#define ioremap_nocache(cookie, size) __uc32_ioremap(cookie, size) #define iounmap(cookie) __uc32_iounmap(cookie) -/* - * Convert a physical pointer to a virtual kernel pointer for /dev/mem - * access - */ -#undef xlate_dev_mem_ptr -#define xlate_dev_mem_ptr(p) __va(p) - #define HAVE_ARCH_PIO_SIZE #define PIO_OFFSET (unsigned int)(PCI_IOBASE) #define PIO_MASK (unsigned int)(IO_SPACE_LIMIT) diff --git a/arch/unicore32/kernel/puv3-nb0916.c b/arch/unicore32/kernel/puv3-nb0916.c index 37b12a0..181108b 100644 --- a/arch/unicore32/kernel/puv3-nb0916.c +++ b/arch/unicore32/kernel/puv3-nb0916.c @@ -123,7 +123,7 @@ int __init mach_nb0916_init(void) if (request_irq(gpio_to_irq(GPI_LCD_CASE_OFF), &nb0916_lcdcaseoff_handler, - IRQF_DISABLED | IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING, + IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING, "NB0916 lcd case off", NULL) < 0) { printk(KERN_DEBUG "LCD-Case-OFF IRQ %d not available\n", @@ -131,7 +131,7 @@ int __init mach_nb0916_init(void) } if (request_irq(gpio_to_irq(GPI_OTP_INT), &nb0916_overheat_handler, - IRQF_DISABLED | IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING, + IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING, "NB0916 overheating protection", NULL) < 0) { printk(KERN_DEBUG "Overheating Protection IRQ %d not available\n", diff --git a/arch/unicore32/kernel/setup.c b/arch/unicore32/kernel/setup.c index 673d7a8..87adbf5 100644 --- a/arch/unicore32/kernel/setup.c +++ b/arch/unicore32/kernel/setup.c @@ -65,7 +65,7 @@ static char default_command_line[COMMAND_LINE_SIZE] __initdata = CONFIG_CMDLINE; */ static struct resource mem_res[] = { { - .name = "Kernel text", + .name = "Kernel code", .start = 0, .end = 0, .flags = IORESOURCE_MEM diff --git a/arch/unicore32/kernel/signal.c b/arch/unicore32/kernel/signal.c index b163fca..911b549 100644 --- a/arch/unicore32/kernel/signal.c +++ b/arch/unicore32/kernel/signal.c @@ -63,10 +63,7 @@ static int restore_sigframe(struct pt_regs *regs, struct sigframe __user *sf) err = __copy_from_user(&set, &sf->uc.uc_sigmask, sizeof(set)); if (err == 0) { sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); + set_current_blocked(&set); } err |= __get_user(regs->UCreg_00, &sf->uc.uc_mcontext.regs.UCreg_00); @@ -321,6 +318,7 @@ static int handle_signal(unsigned long sig, struct k_sigaction *ka, { struct thread_info *thread = current_thread_info(); struct task_struct *tsk = current; + sigset_t blocked; int usig = sig; int ret; @@ -372,13 +370,10 @@ static int handle_signal(unsigned long sig, struct k_sigaction *ka, /* * Block the signal if we were successful. */ - spin_lock_irq(&tsk->sighand->siglock); - sigorsets(&tsk->blocked, &tsk->blocked, - &ka->sa.sa_mask); + sigorsets(&blocked, &tsk->blocked, &ka->sa.sa_mask); if (!(ka->sa.sa_flags & SA_NODEFER)) - sigaddset(&tsk->blocked, sig); - recalc_sigpending(); - spin_unlock_irq(&tsk->sighand->siglock); + sigaddset(&blocked, sig); + set_current_blocked(&blocked); return 0; } diff --git a/arch/unicore32/kernel/time.c b/arch/unicore32/kernel/time.c index 080710c..d3824b2 100644 --- a/arch/unicore32/kernel/time.c +++ b/arch/unicore32/kernel/time.c @@ -86,7 +86,7 @@ static struct clocksource cksrc_puv3_oscr = { static struct irqaction puv3_timer_irq = { .name = "ost0", - .flags = IRQF_DISABLED | IRQF_TIMER | IRQF_IRQPOLL, + .flags = IRQF_TIMER | IRQF_IRQPOLL, .handler = puv3_ost0_interrupt, .dev_id = &ckevt_puv3_osmr0, }; diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 1d2a69d..1a31254 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -62,6 +62,7 @@ config X86 select ANON_INODES select HAVE_ARCH_KMEMCHECK select HAVE_USER_RETURN_NOTIFIER + select ARCH_BINFMT_ELF_RANDOMIZE_PIE select HAVE_ARCH_JUMP_LABEL select HAVE_TEXT_POKE_SMP select HAVE_GENERIC_HARDIRQS @@ -77,6 +78,7 @@ config X86 select HAVE_BPF_JIT if (X86_64 && NET) select CLKEVT_I8253 select ARCH_HAVE_NMI_SAFE_CMPXCHG + select GENERIC_IOMAP config INSTRUCTION_DECODER def_bool (KPROBES || PERF_EVENTS) @@ -142,9 +144,6 @@ config NEED_SG_DMA_LENGTH config GENERIC_ISA_DMA def_bool ISA_DMA_API -config GENERIC_IOMAP - def_bool y - config GENERIC_BUG def_bool y depends on BUG diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index f3444f7..17c5d4b 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -197,7 +197,10 @@ /* Intel-defined CPU features, CPUID level 0x00000007:0 (ebx), word 9 */ #define X86_FEATURE_FSGSBASE (9*32+ 0) /* {RD/WR}{FS/GS}BASE instructions*/ +#define X86_FEATURE_BMI1 (9*32+ 3) /* 1st group bit manipulation extensions */ +#define X86_FEATURE_AVX2 (9*32+ 5) /* AVX2 instructions */ #define X86_FEATURE_SMEP (9*32+ 7) /* Supervisor Mode Execution Protection */ +#define X86_FEATURE_BMI2 (9*32+ 8) /* 2nd group bit manipulation extensions */ #define X86_FEATURE_ERMS (9*32+ 9) /* Enhanced REP MOVSB/STOSB */ #if defined(__KERNEL__) && !defined(__ASSEMBLY__) diff --git a/arch/x86/include/asm/iommu.h b/arch/x86/include/asm/iommu.h index 345c99c..dffc38e 100644 --- a/arch/x86/include/asm/iommu.h +++ b/arch/x86/include/asm/iommu.h @@ -5,6 +5,7 @@ extern struct dma_map_ops nommu_dma_ops; extern int force_iommu, no_iommu; extern int iommu_detected; extern int iommu_pass_through; +extern int iommu_group_mf; /* 10 seconds */ #define DMAR_OPERATION_TIMEOUT ((cycles_t) tsc_khz*10*1000) diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index a026507..ab4092e 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -181,6 +181,7 @@ struct x86_emulate_ops { int (*set_dr)(struct x86_emulate_ctxt *ctxt, int dr, ulong value); int (*set_msr)(struct x86_emulate_ctxt *ctxt, u32 msr_index, u64 data); int (*get_msr)(struct x86_emulate_ctxt *ctxt, u32 msr_index, u64 *pdata); + int (*read_pmc)(struct x86_emulate_ctxt *ctxt, u32 pmc, u64 *pdata); void (*halt)(struct x86_emulate_ctxt *ctxt); void (*wbinvd)(struct x86_emulate_ctxt *ctxt); int (*fix_hypercall)(struct x86_emulate_ctxt *ctxt); @@ -364,6 +365,7 @@ enum x86_intercept { #endif int x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len); +bool x86_page_table_writing_insn(struct x86_emulate_ctxt *ctxt); #define EMULATION_FAILED -1 #define EMULATION_OK 0 #define EMULATION_RESTART 1 diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index b4973f4..52d6640 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -16,10 +16,12 @@ #include <linux/mmu_notifier.h> #include <linux/tracepoint.h> #include <linux/cpumask.h> +#include <linux/irq_work.h> #include <linux/kvm.h> #include <linux/kvm_para.h> #include <linux/kvm_types.h> +#include <linux/perf_event.h> #include <asm/pvclock-abi.h> #include <asm/desc.h> @@ -31,6 +33,8 @@ #define KVM_MEMORY_SLOTS 32 /* memory slots that does not exposed to userspace */ #define KVM_PRIVATE_MEM_SLOTS 4 +#define KVM_MEM_SLOTS_NUM (KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS) + #define KVM_MMIO_SIZE 16 #define KVM_PIO_PAGE_OFFSET 1 @@ -228,7 +232,7 @@ struct kvm_mmu_page { * One bit set per slot which has memory * in this shadow page. */ - DECLARE_BITMAP(slot_bitmap, KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS); + DECLARE_BITMAP(slot_bitmap, KVM_MEM_SLOTS_NUM); bool unsync; int root_count; /* Currently serving as active root */ unsigned int unsync_children; @@ -239,14 +243,9 @@ struct kvm_mmu_page { int clear_spte_count; #endif - struct rcu_head rcu; -}; + int write_flooding_count; -struct kvm_pv_mmu_op_buffer { - void *ptr; - unsigned len; - unsigned processed; - char buf[512] __aligned(sizeof(long)); + struct rcu_head rcu; }; struct kvm_pio_request { @@ -294,6 +293,37 @@ struct kvm_mmu { u64 pdptrs[4]; /* pae */ }; +enum pmc_type { + KVM_PMC_GP = 0, + KVM_PMC_FIXED, +}; + +struct kvm_pmc { + enum pmc_type type; + u8 idx; + u64 counter; + u64 eventsel; + struct perf_event *perf_event; + struct kvm_vcpu *vcpu; +}; + +struct kvm_pmu { + unsigned nr_arch_gp_counters; + unsigned nr_arch_fixed_counters; + unsigned available_event_types; + u64 fixed_ctr_ctrl; + u64 global_ctrl; + u64 global_status; + u64 global_ovf_ctrl; + u64 counter_bitmask[2]; + u64 global_ctrl_mask; + u8 version; + struct kvm_pmc gp_counters[X86_PMC_MAX_GENERIC]; + struct kvm_pmc fixed_counters[X86_PMC_MAX_FIXED]; + struct irq_work irq_work; + u64 reprogram_pmi; +}; + struct kvm_vcpu_arch { /* * rip and regs accesses must go through @@ -345,19 +375,10 @@ struct kvm_vcpu_arch { */ struct kvm_mmu *walk_mmu; - /* only needed in kvm_pv_mmu_op() path, but it's hot so - * put it here to avoid allocation */ - struct kvm_pv_mmu_op_buffer mmu_op_buffer; - struct kvm_mmu_memory_cache mmu_pte_list_desc_cache; struct kvm_mmu_memory_cache mmu_page_cache; struct kvm_mmu_memory_cache mmu_page_header_cache; - gfn_t last_pt_write_gfn; - int last_pt_write_count; - u64 *last_pte_updated; - gfn_t last_pte_gfn; - struct fpu guest_fpu; u64 xcr0; @@ -436,6 +457,8 @@ struct kvm_vcpu_arch { unsigned access; gfn_t mmio_gfn; + struct kvm_pmu pmu; + /* used for guest single stepping over the given code position */ unsigned long singlestep_rip; @@ -444,6 +467,9 @@ struct kvm_vcpu_arch { cpumask_var_t wbinvd_dirty_mask; + unsigned long last_retry_eip; + unsigned long last_retry_addr; + struct { bool halted; gfn_t gfns[roundup_pow_of_two(ASYNC_PF_PER_VCPU)]; @@ -459,7 +485,6 @@ struct kvm_arch { unsigned int n_requested_mmu_pages; unsigned int n_max_mmu_pages; unsigned int indirect_shadow_pages; - atomic_t invlpg_counter; struct hlist_head mmu_page_hash[KVM_NUM_MMU_PAGES]; /* * Hash table of struct kvm_mmu_page. @@ -660,6 +685,8 @@ void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask, int kvm_mmu_reset_context(struct kvm_vcpu *vcpu); void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot); +int kvm_mmu_rmap_write_protect(struct kvm *kvm, u64 gfn, + struct kvm_memory_slot *slot); void kvm_mmu_zap_all(struct kvm *kvm); unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm); void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages); @@ -668,8 +695,6 @@ int load_pdptrs(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, unsigned long cr3); int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa, const void *val, int bytes); -int kvm_pv_mmu_op(struct kvm_vcpu *vcpu, unsigned long bytes, - gpa_t addr, unsigned long *ret); u8 kvm_get_guest_memory_type(struct kvm_vcpu *vcpu, gfn_t gfn); extern bool tdp_enabled; @@ -692,6 +717,7 @@ enum emulation_result { #define EMULTYPE_NO_DECODE (1 << 0) #define EMULTYPE_TRAP_UD (1 << 1) #define EMULTYPE_SKIP (1 << 2) +#define EMULTYPE_RETRY (1 << 3) int x86_emulate_instruction(struct kvm_vcpu *vcpu, unsigned long cr2, int emulation_type, void *insn, int insn_len); @@ -734,6 +760,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data); unsigned long kvm_get_rflags(struct kvm_vcpu *vcpu); void kvm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags); +bool kvm_rdpmc(struct kvm_vcpu *vcpu); void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr); void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code); @@ -754,13 +781,14 @@ int fx_init(struct kvm_vcpu *vcpu); void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu); void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, - const u8 *new, int bytes, - bool guest_initiated); + const u8 *new, int bytes); +int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn); int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva); void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu); int kvm_mmu_load(struct kvm_vcpu *vcpu); void kvm_mmu_unload(struct kvm_vcpu *vcpu); void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu); +gpa_t translate_nested_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access); gpa_t kvm_mmu_gva_to_gpa_read(struct kvm_vcpu *vcpu, gva_t gva, struct x86_exception *exception); gpa_t kvm_mmu_gva_to_gpa_fetch(struct kvm_vcpu *vcpu, gva_t gva, @@ -782,6 +810,11 @@ void kvm_disable_tdp(void); int complete_pio(struct kvm_vcpu *vcpu); bool kvm_check_iopl(struct kvm_vcpu *vcpu); +static inline gpa_t translate_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access) +{ + return gpa; +} + static inline struct kvm_mmu_page *page_header(hpa_t shadow_page) { struct page *page = pfn_to_page(shadow_page >> PAGE_SHIFT); @@ -894,4 +927,17 @@ extern bool kvm_find_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn); void kvm_complete_insn_gp(struct kvm_vcpu *vcpu, int err); +int kvm_is_in_guest(void); + +void kvm_pmu_init(struct kvm_vcpu *vcpu); +void kvm_pmu_destroy(struct kvm_vcpu *vcpu); +void kvm_pmu_reset(struct kvm_vcpu *vcpu); +void kvm_pmu_cpuid_update(struct kvm_vcpu *vcpu); +bool kvm_pmu_msr(struct kvm_vcpu *vcpu, u32 msr); +int kvm_pmu_get_msr(struct kvm_vcpu *vcpu, u32 msr, u64 *data); +int kvm_pmu_set_msr(struct kvm_vcpu *vcpu, u32 msr, u64 data); +int kvm_pmu_read_pmc(struct kvm_vcpu *vcpu, unsigned pmc, u64 *data); +void kvm_handle_pmu_event(struct kvm_vcpu *vcpu); +void kvm_deliver_pmi(struct kvm_vcpu *vcpu); + #endif /* _ASM_X86_KVM_HOST_H */ diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c index a9c2116..f0c6fd6 100644 --- a/arch/x86/kernel/kvm.c +++ b/arch/x86/kernel/kvm.c @@ -39,8 +39,6 @@ #include <asm/desc.h> #include <asm/tlbflush.h> -#define MMU_QUEUE_SIZE 1024 - static int kvmapf = 1; static int parse_no_kvmapf(char *arg) @@ -60,21 +58,10 @@ static int parse_no_stealacc(char *arg) early_param("no-steal-acc", parse_no_stealacc); -struct kvm_para_state { - u8 mmu_queue[MMU_QUEUE_SIZE]; - int mmu_queue_len; -}; - -static DEFINE_PER_CPU(struct kvm_para_state, para_state); static DEFINE_PER_CPU(struct kvm_vcpu_pv_apf_data, apf_reason) __aligned(64); static DEFINE_PER_CPU(struct kvm_steal_time, steal_time) __aligned(64); static int has_steal_clock = 0; -static struct kvm_para_state *kvm_para_state(void) -{ - return &per_cpu(para_state, raw_smp_processor_id()); -} - /* * No need for any "IO delay" on KVM */ @@ -271,151 +258,6 @@ do_async_page_fault(struct pt_regs *regs, unsigned long error_code) } } -static void kvm_mmu_op(void *buffer, unsigned len) -{ - int r; - unsigned long a1, a2; - - do { - a1 = __pa(buffer); - a2 = 0; /* on i386 __pa() always returns <4G */ - r = kvm_hypercall3(KVM_HC_MMU_OP, len, a1, a2); - buffer += r; - len -= r; - } while (len); -} - -static void mmu_queue_flush(struct kvm_para_state *state) -{ - if (state->mmu_queue_len) { - kvm_mmu_op(state->mmu_queue, state->mmu_queue_len); - state->mmu_queue_len = 0; - } -} - -static void kvm_deferred_mmu_op(void *buffer, int len) -{ - struct kvm_para_state *state = kvm_para_state(); - - if (paravirt_get_lazy_mode() != PARAVIRT_LAZY_MMU) { - kvm_mmu_op(buffer, len); - return; - } - if (state->mmu_queue_len + len > sizeof state->mmu_queue) - mmu_queue_flush(state); - memcpy(state->mmu_queue + state->mmu_queue_len, buffer, len); - state->mmu_queue_len += len; -} - -static void kvm_mmu_write(void *dest, u64 val) -{ - __u64 pte_phys; - struct kvm_mmu_op_write_pte wpte; - -#ifdef CONFIG_HIGHPTE - struct page *page; - unsigned long dst = (unsigned long) dest; - - page = kmap_atomic_to_page(dest); - pte_phys = page_to_pfn(page); - pte_phys <<= PAGE_SHIFT; - pte_phys += (dst & ~(PAGE_MASK)); -#else - pte_phys = (unsigned long)__pa(dest); -#endif - wpte.header.op = KVM_MMU_OP_WRITE_PTE; - wpte.pte_val = val; - wpte.pte_phys = pte_phys; - - kvm_deferred_mmu_op(&wpte, sizeof wpte); -} - -/* - * We only need to hook operations that are MMU writes. We hook these so that - * we can use lazy MMU mode to batch these operations. We could probably - * improve the performance of the host code if we used some of the information - * here to simplify processing of batched writes. - */ -static void kvm_set_pte(pte_t *ptep, pte_t pte) -{ - kvm_mmu_write(ptep, pte_val(pte)); -} - -static void kvm_set_pte_at(struct mm_struct *mm, unsigned long addr, - pte_t *ptep, pte_t pte) -{ - kvm_mmu_write(ptep, pte_val(pte)); -} - -static void kvm_set_pmd(pmd_t *pmdp, pmd_t pmd) -{ - kvm_mmu_write(pmdp, pmd_val(pmd)); -} - -#if PAGETABLE_LEVELS >= 3 -#ifdef CONFIG_X86_PAE -static void kvm_set_pte_atomic(pte_t *ptep, pte_t pte) -{ - kvm_mmu_write(ptep, pte_val(pte)); -} - -static void kvm_pte_clear(struct mm_struct *mm, - unsigned long addr, pte_t *ptep) -{ - kvm_mmu_write(ptep, 0); -} - -static void kvm_pmd_clear(pmd_t *pmdp) -{ - kvm_mmu_write(pmdp, 0); -} -#endif - -static void kvm_set_pud(pud_t *pudp, pud_t pud) -{ - kvm_mmu_write(pudp, pud_val(pud)); -} - -#if PAGETABLE_LEVELS == 4 -static void kvm_set_pgd(pgd_t *pgdp, pgd_t pgd) -{ - kvm_mmu_write(pgdp, pgd_val(pgd)); -} -#endif -#endif /* PAGETABLE_LEVELS >= 3 */ - -static void kvm_flush_tlb(void) -{ - struct kvm_mmu_op_flush_tlb ftlb = { - .header.op = KVM_MMU_OP_FLUSH_TLB, - }; - - kvm_deferred_mmu_op(&ftlb, sizeof ftlb); -} - -static void kvm_release_pt(unsigned long pfn) -{ - struct kvm_mmu_op_release_pt rpt = { - .header.op = KVM_MMU_OP_RELEASE_PT, - .pt_phys = (u64)pfn << PAGE_SHIFT, - }; - - kvm_mmu_op(&rpt, sizeof rpt); -} - -static void kvm_enter_lazy_mmu(void) -{ - paravirt_enter_lazy_mmu(); -} - -static void kvm_leave_lazy_mmu(void) -{ - struct kvm_para_state *state = kvm_para_state(); - - mmu_queue_flush(state); - paravirt_leave_lazy_mmu(); -} - static void __init paravirt_ops_setup(void) { pv_info.name = "KVM"; @@ -424,29 +266,6 @@ static void __init paravirt_ops_setup(void) if (kvm_para_has_feature(KVM_FEATURE_NOP_IO_DELAY)) pv_cpu_ops.io_delay = kvm_io_delay; - if (kvm_para_has_feature(KVM_FEATURE_MMU_OP)) { - pv_mmu_ops.set_pte = kvm_set_pte; - pv_mmu_ops.set_pte_at = kvm_set_pte_at; - pv_mmu_ops.set_pmd = kvm_set_pmd; -#if PAGETABLE_LEVELS >= 3 -#ifdef CONFIG_X86_PAE - pv_mmu_ops.set_pte_atomic = kvm_set_pte_atomic; - pv_mmu_ops.pte_clear = kvm_pte_clear; - pv_mmu_ops.pmd_clear = kvm_pmd_clear; -#endif - pv_mmu_ops.set_pud = kvm_set_pud; -#if PAGETABLE_LEVELS == 4 - pv_mmu_ops.set_pgd = kvm_set_pgd; -#endif -#endif - pv_mmu_ops.flush_tlb_user = kvm_flush_tlb; - pv_mmu_ops.release_pte = kvm_release_pt; - pv_mmu_ops.release_pmd = kvm_release_pt; - pv_mmu_ops.release_pud = kvm_release_pt; - - pv_mmu_ops.lazy_mode.enter = kvm_enter_lazy_mmu; - pv_mmu_ops.lazy_mode.leave = kvm_leave_lazy_mmu; - } #ifdef CONFIG_X86_IO_APIC no_timer_check = 1; #endif diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 80dc793..1c4d769 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -45,6 +45,15 @@ int iommu_detected __read_mostly = 0; */ int iommu_pass_through __read_mostly; +/* + * Group multi-function PCI devices into a single device-group for the + * iommu_device_group interface. This tells the iommu driver to pretend + * it cannot distinguish between functions of a device, exposing only one + * group for the device. Useful for disallowing use of individual PCI + * functions from userspace drivers. + */ +int iommu_group_mf __read_mostly; + extern struct iommu_table_entry __iommu_table[], __iommu_table_end[]; /* Dummy device used for NULL arguments (normally ISA). */ @@ -169,6 +178,8 @@ static __init int iommu_setup(char *p) #endif if (!strncmp(p, "pt", 2)) iommu_pass_through = 1; + if (!strncmp(p, "group_mf", 8)) + iommu_group_mf = 1; gart_parse_options(p); diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c index 54ddaeb2..46a01bdc 100644 --- a/arch/x86/kernel/signal.c +++ b/arch/x86/kernel/signal.c @@ -682,7 +682,6 @@ static int handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, struct pt_regs *regs) { - sigset_t blocked; int ret; /* Are we from a system call? */ @@ -733,10 +732,7 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, */ regs->flags &= ~X86_EFLAGS_TF; - sigorsets(&blocked, ¤t->blocked, &ka->sa.sa_mask); - if (!(ka->sa.sa_flags & SA_NODEFER)) - sigaddset(&blocked, sig); - set_current_blocked(&blocked); + block_sigmask(ka, sig); tracehook_signal_handler(sig, info, ka, regs, test_thread_flag(TIF_SINGLESTEP)); diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig index ff5790d..1a7fe86 100644 --- a/arch/x86/kvm/Kconfig +++ b/arch/x86/kvm/Kconfig @@ -35,6 +35,7 @@ config KVM select KVM_MMIO select TASKSTATS select TASK_DELAY_ACCT + select PERF_EVENTS ---help--- Support hosting fully virtualized guest machines using hardware virtualization extensions. You will need a fairly recent @@ -52,6 +53,8 @@ config KVM config KVM_INTEL tristate "KVM for Intel processors support" depends on KVM + # for perf_guest_get_msrs(): + depends on CPU_SUP_INTEL ---help--- Provides support for KVM on Intel processors equipped with the VT extensions. diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile index f15501f43..4f579e8 100644 --- a/arch/x86/kvm/Makefile +++ b/arch/x86/kvm/Makefile @@ -12,7 +12,7 @@ kvm-$(CONFIG_IOMMU_API) += $(addprefix ../../../virt/kvm/, iommu.o) kvm-$(CONFIG_KVM_ASYNC_PF) += $(addprefix ../../../virt/kvm/, async_pf.o) kvm-y += x86.o mmu.o emulate.o i8259.o irq.o lapic.o \ - i8254.o timer.o + i8254.o timer.o cpuid.o pmu.o kvm-intel-y += vmx.o kvm-amd-y += svm.o diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c new file mode 100644 index 0000000..89b02bf --- /dev/null +++ b/arch/x86/kvm/cpuid.c @@ -0,0 +1,670 @@ +/* + * Kernel-based Virtual Machine driver for Linux + * cpuid support routines + * + * derived from arch/x86/kvm/x86.c + * + * Copyright 2011 Red Hat, Inc. and/or its affiliates. + * Copyright IBM Corporation, 2008 + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + */ + +#include <linux/kvm_host.h> +#include <linux/module.h> +#include <linux/vmalloc.h> +#include <linux/uaccess.h> +#include <asm/user.h> +#include <asm/xsave.h> +#include "cpuid.h" +#include "lapic.h" +#include "mmu.h" +#include "trace.h" + +void kvm_update_cpuid(struct kvm_vcpu *vcpu) +{ + struct kvm_cpuid_entry2 *best; + struct kvm_lapic *apic = vcpu->arch.apic; + + best = kvm_find_cpuid_entry(vcpu, 1, 0); + if (!best) + return; + + /* Update OSXSAVE bit */ + if (cpu_has_xsave && best->function == 0x1) { + best->ecx &= ~(bit(X86_FEATURE_OSXSAVE)); + if (kvm_read_cr4_bits(vcpu, X86_CR4_OSXSAVE)) + best->ecx |= bit(X86_FEATURE_OSXSAVE); + } + + if (apic) { + if (best->ecx & bit(X86_FEATURE_TSC_DEADLINE_TIMER)) + apic->lapic_timer.timer_mode_mask = 3 << 17; + else + apic->lapic_timer.timer_mode_mask = 1 << 17; + } + + kvm_pmu_cpuid_update(vcpu); +} + +static int is_efer_nx(void) +{ + unsigned long long efer = 0; + + rdmsrl_safe(MSR_EFER, &efer); + return efer & EFER_NX; +} + +static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu) +{ + int i; + struct kvm_cpuid_entry2 *e, *entry; + + entry = NULL; + for (i = 0; i < vcpu->arch.cpuid_nent; ++i) { + e = &vcpu->arch.cpuid_entries[i]; + if (e->function == 0x80000001) { + entry = e; + break; + } + } + if (entry && (entry->edx & (1 << 20)) && !is_efer_nx()) { + entry->edx &= ~(1 << 20); + printk(KERN_INFO "kvm: guest NX capability removed\n"); + } +} + +/* when an old userspace process fills a new kernel module */ +int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu, + struct kvm_cpuid *cpuid, + struct kvm_cpuid_entry __user *entries) +{ + int r, i; + struct kvm_cpuid_entry *cpuid_entries; + + r = -E2BIG; + if (cpuid->nent > KVM_MAX_CPUID_ENTRIES) + goto out; + r = -ENOMEM; + cpuid_entries = vmalloc(sizeof(struct kvm_cpuid_entry) * cpuid->nent); + if (!cpuid_entries) + goto out; + r = -EFAULT; + if (copy_from_user(cpuid_entries, entries, + cpuid->nent * sizeof(struct kvm_cpuid_entry))) + goto out_free; + for (i = 0; i < cpuid->nent; i++) { + vcpu->arch.cpuid_entries[i].function = cpuid_entries[i].function; + vcpu->arch.cpuid_entries[i].eax = cpuid_entries[i].eax; + vcpu->arch.cpuid_entries[i].ebx = cpuid_entries[i].ebx; + vcpu->arch.cpuid_entries[i].ecx = cpuid_entries[i].ecx; + vcpu->arch.cpuid_entries[i].edx = cpuid_entries[i].edx; + vcpu->arch.cpuid_entries[i].index = 0; + vcpu->arch.cpuid_entries[i].flags = 0; + vcpu->arch.cpuid_entries[i].padding[0] = 0; + vcpu->arch.cpuid_entries[i].padding[1] = 0; + vcpu->arch.cpuid_entries[i].padding[2] = 0; + } + vcpu->arch.cpuid_nent = cpuid->nent; + cpuid_fix_nx_cap(vcpu); + r = 0; + kvm_apic_set_version(vcpu); + kvm_x86_ops->cpuid_update(vcpu); + kvm_update_cpuid(vcpu); + +out_free: + vfree(cpuid_entries); +out: + return r; +} + +int kvm_vcpu_ioctl_set_cpuid2(struct kvm_vcpu *vcpu, + struct kvm_cpuid2 *cpuid, + struct kvm_cpuid_entry2 __user *entries) +{ + int r; + + r = -E2BIG; + if (cpuid->nent > KVM_MAX_CPUID_ENTRIES) + goto out; + r = -EFAULT; + if (copy_from_user(&vcpu->arch.cpuid_entries, entries, + cpuid->nent * sizeof(struct kvm_cpuid_entry2))) + goto out; + vcpu->arch.cpuid_nent = cpuid->nent; + kvm_apic_set_version(vcpu); + kvm_x86_ops->cpuid_update(vcpu); + kvm_update_cpuid(vcpu); + return 0; + +out: + return r; +} + +int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu, + struct kvm_cpuid2 *cpuid, + struct kvm_cpuid_entry2 __user *entries) +{ + int r; + + r = -E2BIG; + if (cpuid->nent < vcpu->arch.cpuid_nent) + goto out; + r = -EFAULT; + if (copy_to_user(entries, &vcpu->arch.cpuid_entries, + vcpu->arch.cpuid_nent * sizeof(struct kvm_cpuid_entry2))) + goto out; + return 0; + +out: + cpuid->nent = vcpu->arch.cpuid_nent; + return r; +} + +static void cpuid_mask(u32 *word, int wordnum) +{ + *word &= boot_cpu_data.x86_capability[wordnum]; +} + +static void do_cpuid_1_ent(struct kvm_cpuid_entry2 *entry, u32 function, + u32 index) +{ + entry->function = function; + entry->index = index; + cpuid_count(entry->function, entry->index, + &entry->eax, &entry->ebx, &entry->ecx, &entry->edx); + entry->flags = 0; +} + +static bool supported_xcr0_bit(unsigned bit) +{ + u64 mask = ((u64)1 << bit); + + return mask & (XSTATE_FP | XSTATE_SSE | XSTATE_YMM) & host_xcr0; +} + +#define F(x) bit(X86_FEATURE_##x) + +static int do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, + u32 index, int *nent, int maxnent) +{ + int r; + unsigned f_nx = is_efer_nx() ? F(NX) : 0; +#ifdef CONFIG_X86_64 + unsigned f_gbpages = (kvm_x86_ops->get_lpage_level() == PT_PDPE_LEVEL) + ? F(GBPAGES) : 0; + unsigned f_lm = F(LM); +#else + unsigned f_gbpages = 0; + unsigned f_lm = 0; +#endif + unsigned f_rdtscp = kvm_x86_ops->rdtscp_supported() ? F(RDTSCP) : 0; + + /* cpuid 1.edx */ + const u32 kvm_supported_word0_x86_features = + F(FPU) | F(VME) | F(DE) | F(PSE) | + F(TSC) | F(MSR) | F(PAE) | F(MCE) | + F(CX8) | F(APIC) | 0 /* Reserved */ | F(SEP) | + F(MTRR) | F(PGE) | F(MCA) | F(CMOV) | + F(PAT) | F(PSE36) | 0 /* PSN */ | F(CLFLSH) | + 0 /* Reserved, DS, ACPI */ | F(MMX) | + F(FXSR) | F(XMM) | F(XMM2) | F(SELFSNOOP) | + 0 /* HTT, TM, Reserved, PBE */; + /* cpuid 0x80000001.edx */ + const u32 kvm_supported_word1_x86_features = + F(FPU) | F(VME) | F(DE) | F(PSE) | + F(TSC) | F(MSR) | F(PAE) | F(MCE) | + F(CX8) | F(APIC) | 0 /* Reserved */ | F(SYSCALL) | + F(MTRR) | F(PGE) | F(MCA) | F(CMOV) | + F(PAT) | F(PSE36) | 0 /* Reserved */ | + f_nx | 0 /* Reserved */ | F(MMXEXT) | F(MMX) | + F(FXSR) | F(FXSR_OPT) | f_gbpages | f_rdtscp | + 0 /* Reserved */ | f_lm | F(3DNOWEXT) | F(3DNOW); + /* cpuid 1.ecx */ + const u32 kvm_supported_word4_x86_features = + F(XMM3) | F(PCLMULQDQ) | 0 /* DTES64, MONITOR */ | + 0 /* DS-CPL, VMX, SMX, EST */ | + 0 /* TM2 */ | F(SSSE3) | 0 /* CNXT-ID */ | 0 /* Reserved */ | + F(FMA) | F(CX16) | 0 /* xTPR Update, PDCM */ | + 0 /* Reserved, DCA */ | F(XMM4_1) | + F(XMM4_2) | F(X2APIC) | F(MOVBE) | F(POPCNT) | + 0 /* Reserved*/ | F(AES) | F(XSAVE) | 0 /* OSXSAVE */ | F(AVX) | + F(F16C) | F(RDRAND); + /* cpuid 0x80000001.ecx */ + const u32 kvm_supported_word6_x86_features = + F(LAHF_LM) | F(CMP_LEGACY) | 0 /*SVM*/ | 0 /* ExtApicSpace */ | + F(CR8_LEGACY) | F(ABM) | F(SSE4A) | F(MISALIGNSSE) | + F(3DNOWPREFETCH) | 0 /* OSVW */ | 0 /* IBS */ | F(XOP) | + 0 /* SKINIT, WDT, LWP */ | F(FMA4) | F(TBM); + + /* cpuid 0xC0000001.edx */ + const u32 kvm_supported_word5_x86_features = + F(XSTORE) | F(XSTORE_EN) | F(XCRYPT) | F(XCRYPT_EN) | + F(ACE2) | F(ACE2_EN) | F(PHE) | F(PHE_EN) | + F(PMM) | F(PMM_EN); + + /* cpuid 7.0.ebx */ + const u32 kvm_supported_word9_x86_features = + F(FSGSBASE) | F(BMI1) | F(AVX2) | F(SMEP) | F(BMI2) | F(ERMS); + + /* all calls to cpuid_count() should be made on the same cpu */ + get_cpu(); + + r = -E2BIG; + + if (*nent >= maxnent) + goto out; + + do_cpuid_1_ent(entry, function, index); + ++*nent; + + switch (function) { + case 0: + entry->eax = min(entry->eax, (u32)0xd); + break; + case 1: + entry->edx &= kvm_supported_word0_x86_features; + cpuid_mask(&entry->edx, 0); + entry->ecx &= kvm_supported_word4_x86_features; + cpuid_mask(&entry->ecx, 4); + /* we support x2apic emulation even if host does not support + * it since we emulate x2apic in software */ + entry->ecx |= F(X2APIC); + break; + /* function 2 entries are STATEFUL. That is, repeated cpuid commands + * may return different values. This forces us to get_cpu() before + * issuing the first command, and also to emulate this annoying behavior + * in kvm_emulate_cpuid() using KVM_CPUID_FLAG_STATE_READ_NEXT */ + case 2: { + int t, times = entry->eax & 0xff; + + entry->flags |= KVM_CPUID_FLAG_STATEFUL_FUNC; + entry->flags |= KVM_CPUID_FLAG_STATE_READ_NEXT; + for (t = 1; t < times; ++t) { + if (*nent >= maxnent) + goto out; + + do_cpuid_1_ent(&entry[t], function, 0); + entry[t].flags |= KVM_CPUID_FLAG_STATEFUL_FUNC; + ++*nent; + } + break; + } + /* function 4 has additional index. */ + case 4: { + int i, cache_type; + + entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX; + /* read more entries until cache_type is zero */ + for (i = 1; ; ++i) { + if (*nent >= maxnent) + goto out; + + cache_type = entry[i - 1].eax & 0x1f; + if (!cache_type) + break; + do_cpuid_1_ent(&entry[i], function, i); + entry[i].flags |= + KVM_CPUID_FLAG_SIGNIFCANT_INDEX; + ++*nent; + } + break; + } + case 7: { + entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX; + /* Mask ebx against host capbability word 9 */ + if (index == 0) { + entry->ebx &= kvm_supported_word9_x86_features; + cpuid_mask(&entry->ebx, 9); + } else + entry->ebx = 0; + entry->eax = 0; + entry->ecx = 0; + entry->edx = 0; + break; + } + case 9: + break; + case 0xa: { /* Architectural Performance Monitoring */ + struct x86_pmu_capability cap; + union cpuid10_eax eax; + union cpuid10_edx edx; + + perf_get_x86_pmu_capability(&cap); + + /* + * Only support guest architectural pmu on a host + * with architectural pmu. + */ + if (!cap.version) + memset(&cap, 0, sizeof(cap)); + + eax.split.version_id = min(cap.version, 2); + eax.split.num_counters = cap.num_counters_gp; + eax.split.bit_width = cap.bit_width_gp; + eax.split.mask_length = cap.events_mask_len; + + edx.split.num_counters_fixed = cap.num_counters_fixed; + edx.split.bit_width_fixed = cap.bit_width_fixed; + edx.split.reserved = 0; + + entry->eax = eax.full; + entry->ebx = cap.events_mask; + entry->ecx = 0; + entry->edx = edx.full; + break; + } + /* function 0xb has additional index. */ + case 0xb: { + int i, level_type; + + entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX; + /* read more entries until level_type is zero */ + for (i = 1; ; ++i) { + if (*nent >= maxnent) + goto out; + + level_type = entry[i - 1].ecx & 0xff00; + if (!level_type) + break; + do_cpuid_1_ent(&entry[i], function, i); + entry[i].flags |= + KVM_CPUID_FLAG_SIGNIFCANT_INDEX; + ++*nent; + } + break; + } + case 0xd: { + int idx, i; + + entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX; + for (idx = 1, i = 1; idx < 64; ++idx) { + if (*nent >= maxnent) + goto out; + + do_cpuid_1_ent(&entry[i], function, idx); + if (entry[i].eax == 0 || !supported_xcr0_bit(idx)) + continue; + entry[i].flags |= + KVM_CPUID_FLAG_SIGNIFCANT_INDEX; + ++*nent; + ++i; + } + break; + } + case KVM_CPUID_SIGNATURE: { + char signature[12] = "KVMKVMKVM\0\0"; + u32 *sigptr = (u32 *)signature; + entry->eax = 0; + entry->ebx = sigptr[0]; + entry->ecx = sigptr[1]; + entry->edx = sigptr[2]; + break; + } + case KVM_CPUID_FEATURES: + entry->eax = (1 << KVM_FEATURE_CLOCKSOURCE) | + (1 << KVM_FEATURE_NOP_IO_DELAY) | + (1 << KVM_FEATURE_CLOCKSOURCE2) | + (1 << KVM_FEATURE_ASYNC_PF) | + (1 << KVM_FEATURE_CLOCKSOURCE_STABLE_BIT); + + if (sched_info_on()) + entry->eax |= (1 << KVM_FEATURE_STEAL_TIME); + + entry->ebx = 0; + entry->ecx = 0; + entry->edx = 0; + break; + case 0x80000000: + entry->eax = min(entry->eax, 0x8000001a); + break; + case 0x80000001: + entry->edx &= kvm_supported_word1_x86_features; + cpuid_mask(&entry->edx, 1); + entry->ecx &= kvm_supported_word6_x86_features; + cpuid_mask(&entry->ecx, 6); + break; + case 0x80000008: { + unsigned g_phys_as = (entry->eax >> 16) & 0xff; + unsigned virt_as = max((entry->eax >> 8) & 0xff, 48U); + unsigned phys_as = entry->eax & 0xff; + + if (!g_phys_as) + g_phys_as = phys_as; + entry->eax = g_phys_as | (virt_as << 8); + entry->ebx = entry->edx = 0; + break; + } + case 0x80000019: + entry->ecx = entry->edx = 0; + break; + case 0x8000001a: + break; + case 0x8000001d: + break; + /*Add support for Centaur's CPUID instruction*/ + case 0xC0000000: + /*Just support up to 0xC0000004 now*/ + entry->eax = min(entry->eax, 0xC0000004); + break; + case 0xC0000001: + entry->edx &= kvm_supported_word5_x86_features; + cpuid_mask(&entry->edx, 5); + break; + case 3: /* Processor serial number */ + case 5: /* MONITOR/MWAIT */ + case 6: /* Thermal management */ + case 0x80000007: /* Advanced power management */ + case 0xC0000002: + case 0xC0000003: + case 0xC0000004: + default: + entry->eax = entry->ebx = entry->ecx = entry->edx = 0; + break; + } + + kvm_x86_ops->set_supported_cpuid(function, entry); + + r = 0; + +out: + put_cpu(); + + return r; +} + +#undef F + +struct kvm_cpuid_param { + u32 func; + u32 idx; + bool has_leaf_count; + bool (*qualifier)(struct kvm_cpuid_param *param); +}; + +static bool is_centaur_cpu(struct kvm_cpuid_param *param) +{ + return boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR; +} + +int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, + struct kvm_cpuid_entry2 __user *entries) +{ + struct kvm_cpuid_entry2 *cpuid_entries; + int limit, nent = 0, r = -E2BIG, i; + u32 func; + static struct kvm_cpuid_param param[] = { + { .func = 0, .has_leaf_count = true }, + { .func = 0x80000000, .has_leaf_count = true }, + { .func = 0xC0000000, .qualifier = is_centaur_cpu, .has_leaf_count = true }, + { .func = KVM_CPUID_SIGNATURE }, + { .func = KVM_CPUID_FEATURES }, + }; + + if (cpuid->nent < 1) + goto out; + if (cpuid->nent > KVM_MAX_CPUID_ENTRIES) + cpuid->nent = KVM_MAX_CPUID_ENTRIES; + r = -ENOMEM; + cpuid_entries = vmalloc(sizeof(struct kvm_cpuid_entry2) * cpuid->nent); + if (!cpuid_entries) + goto out; + + r = 0; + for (i = 0; i < ARRAY_SIZE(param); i++) { + struct kvm_cpuid_param *ent = ¶m[i]; + + if (ent->qualifier && !ent->qualifier(ent)) + continue; + + r = do_cpuid_ent(&cpuid_entries[nent], ent->func, ent->idx, + &nent, cpuid->nent); + + if (r) + goto out_free; + + if (!ent->has_leaf_count) + continue; + + limit = cpuid_entries[nent - 1].eax; + for (func = ent->func + 1; func <= limit && nent < cpuid->nent && r == 0; ++func) + r = do_cpuid_ent(&cpuid_entries[nent], func, ent->idx, + &nent, cpuid->nent); + + if (r) + goto out_free; + } + + r = -EFAULT; + if (copy_to_user(entries, cpuid_entries, + nent * sizeof(struct kvm_cpuid_entry2))) + goto out_free; + cpuid->nent = nent; + r = 0; + +out_free: + vfree(cpuid_entries); +out: + return r; +} + +static int move_to_next_stateful_cpuid_entry(struct kvm_vcpu *vcpu, int i) +{ + struct kvm_cpuid_entry2 *e = &vcpu->arch.cpuid_entries[i]; + int j, nent = vcpu->arch.cpuid_nent; + + e->flags &= ~KVM_CPUID_FLAG_STATE_READ_NEXT; + /* when no next entry is found, the current entry[i] is reselected */ + for (j = i + 1; ; j = (j + 1) % nent) { + struct kvm_cpuid_entry2 *ej = &vcpu->arch.cpuid_entries[j]; + if (ej->function == e->function) { + ej->flags |= KVM_CPUID_FLAG_STATE_READ_NEXT; + return j; + } + } + return 0; /* silence gcc, even though control never reaches here */ +} + +/* find an entry with matching function, matching index (if needed), and that + * should be read next (if it's stateful) */ +static int is_matching_cpuid_entry(struct kvm_cpuid_entry2 *e, + u32 function, u32 index) +{ + if (e->function != function) + return 0; + if ((e->flags & KVM_CPUID_FLAG_SIGNIFCANT_INDEX) && e->index != index) + return 0; + if ((e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC) && + !(e->flags & KVM_CPUID_FLAG_STATE_READ_NEXT)) + return 0; + return 1; +} + +struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu, + u32 function, u32 index) +{ + int i; + struct kvm_cpuid_entry2 *best = NULL; + + for (i = 0; i < vcpu->arch.cpuid_nent; ++i) { + struct kvm_cpuid_entry2 *e; + + e = &vcpu->arch.cpuid_entries[i]; + if (is_matching_cpuid_entry(e, function, index)) { + if (e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC) + move_to_next_stateful_cpuid_entry(vcpu, i); + best = e; + break; + } + } + return best; +} +EXPORT_SYMBOL_GPL(kvm_find_cpuid_entry); + +int cpuid_maxphyaddr(struct kvm_vcpu *vcpu) +{ + struct kvm_cpuid_entry2 *best; + + best = kvm_find_cpuid_entry(vcpu, 0x80000000, 0); + if (!best || best->eax < 0x80000008) + goto not_found; + best = kvm_find_cpuid_entry(vcpu, 0x80000008, 0); + if (best) + return best->eax & 0xff; +not_found: + return 36; +} + +/* + * If no match is found, check whether we exceed the vCPU's limit + * and return the content of the highest valid _standard_ leaf instead. + * This is to satisfy the CPUID specification. + */ +static struct kvm_cpuid_entry2* check_cpuid_limit(struct kvm_vcpu *vcpu, + u32 function, u32 index) +{ + struct kvm_cpuid_entry2 *maxlevel; + + maxlevel = kvm_find_cpuid_entry(vcpu, function & 0x80000000, 0); + if (!maxlevel || maxlevel->eax >= function) + return NULL; + if (function & 0x80000000) { + maxlevel = kvm_find_cpuid_entry(vcpu, 0, 0); + if (!maxlevel) + return NULL; + } + return kvm_find_cpuid_entry(vcpu, maxlevel->eax, index); +} + +void kvm_emulate_cpuid(struct kvm_vcpu *vcpu) +{ + u32 function, index; + struct kvm_cpuid_entry2 *best; + + function = kvm_register_read(vcpu, VCPU_REGS_RAX); + index = kvm_register_read(vcpu, VCPU_REGS_RCX); + kvm_register_write(vcpu, VCPU_REGS_RAX, 0); + kvm_register_write(vcpu, VCPU_REGS_RBX, 0); + kvm_register_write(vcpu, VCPU_REGS_RCX, 0); + kvm_register_write(vcpu, VCPU_REGS_RDX, 0); + best = kvm_find_cpuid_entry(vcpu, function, index); + + if (!best) + best = check_cpuid_limit(vcpu, function, index); + + if (best) { + kvm_register_write(vcpu, VCPU_REGS_RAX, best->eax); + kvm_register_write(vcpu, VCPU_REGS_RBX, best->ebx); + kvm_register_write(vcpu, VCPU_REGS_RCX, best->ecx); + kvm_register_write(vcpu, VCPU_REGS_RDX, best->edx); + } + kvm_x86_ops->skip_emulated_instruction(vcpu); + trace_kvm_cpuid(function, + kvm_register_read(vcpu, VCPU_REGS_RAX), + kvm_register_read(vcpu, VCPU_REGS_RBX), + kvm_register_read(vcpu, VCPU_REGS_RCX), + kvm_register_read(vcpu, VCPU_REGS_RDX)); +} +EXPORT_SYMBOL_GPL(kvm_emulate_cpuid); diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h new file mode 100644 index 0000000..5b97e17 --- /dev/null +++ b/arch/x86/kvm/cpuid.h @@ -0,0 +1,46 @@ +#ifndef ARCH_X86_KVM_CPUID_H +#define ARCH_X86_KVM_CPUID_H + +#include "x86.h" + +void kvm_update_cpuid(struct kvm_vcpu *vcpu); +struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu, + u32 function, u32 index); +int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, + struct kvm_cpuid_entry2 __user *entries); +int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu, + struct kvm_cpuid *cpuid, + struct kvm_cpuid_entry __user *entries); +int kvm_vcpu_ioctl_set_cpuid2(struct kvm_vcpu *vcpu, + struct kvm_cpuid2 *cpuid, + struct kvm_cpuid_entry2 __user *entries); +int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu, + struct kvm_cpuid2 *cpuid, + struct kvm_cpuid_entry2 __user *entries); + + +static inline bool guest_cpuid_has_xsave(struct kvm_vcpu *vcpu) +{ + struct kvm_cpuid_entry2 *best; + + best = kvm_find_cpuid_entry(vcpu, 1, 0); + return best && (best->ecx & bit(X86_FEATURE_XSAVE)); +} + +static inline bool guest_cpuid_has_smep(struct kvm_vcpu *vcpu) +{ + struct kvm_cpuid_entry2 *best; + + best = kvm_find_cpuid_entry(vcpu, 7, 0); + return best && (best->ebx & bit(X86_FEATURE_SMEP)); +} + +static inline bool guest_cpuid_has_fsgsbase(struct kvm_vcpu *vcpu) +{ + struct kvm_cpuid_entry2 *best; + + best = kvm_find_cpuid_entry(vcpu, 7, 0); + return best && (best->ebx & bit(X86_FEATURE_FSGSBASE)); +} + +#endif diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index f1e3be18..05a562b 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -125,8 +125,9 @@ #define Lock (1<<26) /* lock prefix is allowed for the instruction */ #define Priv (1<<27) /* instruction generates #GP if current CPL != 0 */ #define No64 (1<<28) +#define PageTable (1 << 29) /* instruction used to write page table */ /* Source 2 operand type */ -#define Src2Shift (29) +#define Src2Shift (30) #define Src2None (OpNone << Src2Shift) #define Src2CL (OpCL << Src2Shift) #define Src2ImmByte (OpImmByte << Src2Shift) @@ -1674,11 +1675,6 @@ static int em_jmp_far(struct x86_emulate_ctxt *ctxt) return X86EMUL_CONTINUE; } -static int em_grp1a(struct x86_emulate_ctxt *ctxt) -{ - return emulate_pop(ctxt, &ctxt->dst.val, ctxt->dst.bytes); -} - static int em_grp2(struct x86_emulate_ctxt *ctxt) { switch (ctxt->modrm_reg) { @@ -1788,7 +1784,7 @@ static int em_grp45(struct x86_emulate_ctxt *ctxt) return rc; } -static int em_grp9(struct x86_emulate_ctxt *ctxt) +static int em_cmpxchg8b(struct x86_emulate_ctxt *ctxt) { u64 old = ctxt->dst.orig_val64; @@ -1831,6 +1827,24 @@ static int em_ret_far(struct x86_emulate_ctxt *ctxt) return rc; } +static int em_cmpxchg(struct x86_emulate_ctxt *ctxt) +{ + /* Save real source value, then compare EAX against destination. */ + ctxt->src.orig_val = ctxt->src.val; + ctxt->src.val = ctxt->regs[VCPU_REGS_RAX]; + emulate_2op_SrcV(ctxt, "cmp"); + + if (ctxt->eflags & EFLG_ZF) { + /* Success: write back to memory. */ + ctxt->dst.val = ctxt->src.orig_val; + } else { + /* Failure: write the value we saw to EAX. */ + ctxt->dst.type = OP_REG; + ctxt->dst.addr.reg = (unsigned long *)&ctxt->regs[VCPU_REGS_RAX]; + } + return X86EMUL_CONTINUE; +} + static int em_lseg(struct x86_emulate_ctxt *ctxt) { int seg = ctxt->src2.val; @@ -2481,6 +2495,15 @@ static int em_das(struct x86_emulate_ctxt *ctxt) return X86EMUL_CONTINUE; } +static int em_call(struct x86_emulate_ctxt *ctxt) +{ + long rel = ctxt->src.val; + + ctxt->src.val = (unsigned long)ctxt->_eip; + jmp_rel(ctxt, rel); + return em_push(ctxt); +} + static int em_call_far(struct x86_emulate_ctxt *ctxt) { u16 sel, old_cs; @@ -2622,12 +2645,75 @@ static int em_rdtsc(struct x86_emulate_ctxt *ctxt) return X86EMUL_CONTINUE; } +static int em_rdpmc(struct x86_emulate_ctxt *ctxt) +{ + u64 pmc; + + if (ctxt->ops->read_pmc(ctxt, ctxt->regs[VCPU_REGS_RCX], &pmc)) + return emulate_gp(ctxt, 0); + ctxt->regs[VCPU_REGS_RAX] = (u32)pmc; + ctxt->regs[VCPU_REGS_RDX] = pmc >> 32; + return X86EMUL_CONTINUE; +} + static int em_mov(struct x86_emulate_ctxt *ctxt) { ctxt->dst.val = ctxt->src.val; return X86EMUL_CONTINUE; } +static int em_cr_write(struct x86_emulate_ctxt *ctxt) +{ + if (ctxt->ops->set_cr(ctxt, ctxt->modrm_reg, ctxt->src.val)) + return emulate_gp(ctxt, 0); + + /* Disable writeback. */ + ctxt->dst.type = OP_NONE; + return X86EMUL_CONTINUE; +} + +static int em_dr_write(struct x86_emulate_ctxt *ctxt) +{ + unsigned long val; + + if (ctxt->mode == X86EMUL_MODE_PROT64) + val = ctxt->src.val & ~0ULL; + else + val = ctxt->src.val & ~0U; + + /* #UD condition is already handled. */ + if (ctxt->ops->set_dr(ctxt, ctxt->modrm_reg, val) < 0) + return emulate_gp(ctxt, 0); + + /* Disable writeback. */ + ctxt->dst.type = OP_NONE; + return X86EMUL_CONTINUE; +} + +static int em_wrmsr(struct x86_emulate_ctxt *ctxt) +{ + u64 msr_data; + + msr_data = (u32)ctxt->regs[VCPU_REGS_RAX] + | ((u64)ctxt->regs[VCPU_REGS_RDX] << 32); + if (ctxt->ops->set_msr(ctxt, ctxt->regs[VCPU_REGS_RCX], msr_data)) + return emulate_gp(ctxt, 0); + + return X86EMUL_CONTINUE; +} + +static int em_rdmsr(struct x86_emulate_ctxt *ctxt) +{ + u64 msr_data; + + if (ctxt->ops->get_msr(ctxt, ctxt->regs[VCPU_REGS_RCX], &msr_data)) + return emulate_gp(ctxt, 0); + + ctxt->regs[VCPU_REGS_RAX] = (u32)msr_data; + ctxt->regs[VCPU_REGS_RDX] = msr_data >> 32; + return X86EMUL_CONTINUE; +} + static int em_mov_rm_sreg(struct x86_emulate_ctxt *ctxt) { if (ctxt->modrm_reg > VCPU_SREG_GS) @@ -2775,6 +2861,24 @@ static int em_jcxz(struct x86_emulate_ctxt *ctxt) return X86EMUL_CONTINUE; } +static int em_in(struct x86_emulate_ctxt *ctxt) +{ + if (!pio_in_emulated(ctxt, ctxt->dst.bytes, ctxt->src.val, + &ctxt->dst.val)) + return X86EMUL_IO_NEEDED; + + return X86EMUL_CONTINUE; +} + +static int em_out(struct x86_emulate_ctxt *ctxt) +{ + ctxt->ops->pio_out_emulated(ctxt, ctxt->src.bytes, ctxt->dst.val, + &ctxt->src.val, 1); + /* Disable writeback. */ + ctxt->dst.type = OP_NONE; + return X86EMUL_CONTINUE; +} + static int em_cli(struct x86_emulate_ctxt *ctxt) { if (emulator_bad_iopl(ctxt)) @@ -2794,6 +2898,69 @@ static int em_sti(struct x86_emulate_ctxt *ctxt) return X86EMUL_CONTINUE; } +static int em_bt(struct x86_emulate_ctxt *ctxt) +{ + /* Disable writeback. */ + ctxt->dst.type = OP_NONE; + /* only subword offset */ + ctxt->src.val &= (ctxt->dst.bytes << 3) - 1; + + emulate_2op_SrcV_nobyte(ctxt, "bt"); + return X86EMUL_CONTINUE; +} + +static int em_bts(struct x86_emulate_ctxt *ctxt) +{ + emulate_2op_SrcV_nobyte(ctxt, "bts"); + return X86EMUL_CONTINUE; +} + +static int em_btr(struct x86_emulate_ctxt *ctxt) +{ + emulate_2op_SrcV_nobyte(ctxt, "btr"); + return X86EMUL_CONTINUE; +} + +static int em_btc(struct x86_emulate_ctxt *ctxt) +{ + emulate_2op_SrcV_nobyte(ctxt, "btc"); + return X86EMUL_CONTINUE; +} + +static int em_bsf(struct x86_emulate_ctxt *ctxt) +{ + u8 zf; + + __asm__ ("bsf %2, %0; setz %1" + : "=r"(ctxt->dst.val), "=q"(zf) + : "r"(ctxt->src.val)); + + ctxt->eflags &= ~X86_EFLAGS_ZF; + if (zf) { + ctxt->eflags |= X86_EFLAGS_ZF; + /* Disable writeback. */ + ctxt->dst.type = OP_NONE; + } + return X86EMUL_CONTINUE; +} + +static int em_bsr(struct x86_emulate_ctxt *ctxt) +{ + u8 zf; + + __asm__ ("bsr %2, %0; setz %1" + : "=r"(ctxt->dst.val), "=q"(zf) + : "r"(ctxt->src.val)); + + ctxt->eflags &= ~X86_EFLAGS_ZF; + if (zf) { + ctxt->eflags |= X86_EFLAGS_ZF; + /* Disable writeback. */ + ctxt->dst.type = OP_NONE; + } + return X86EMUL_CONTINUE; +} + static bool valid_cr(int nr) { switch (nr) { @@ -2867,9 +3034,6 @@ static int check_cr_write(struct x86_emulate_ctxt *ctxt) break; } case 4: { - u64 cr4; - - cr4 = ctxt->ops->get_cr(ctxt, 4); ctxt->ops->get_msr(ctxt, MSR_EFER, &efer); if ((efer & EFER_LMA) && !(new_val & X86_CR4_PAE)) @@ -3003,6 +3167,8 @@ static int check_perm_out(struct x86_emulate_ctxt *ctxt) #define D2bv(_f) D((_f) | ByteOp), D(_f) #define D2bvIP(_f, _i, _p) DIP((_f) | ByteOp, _i, _p), DIP(_f, _i, _p) #define I2bv(_f, _e) I((_f) | ByteOp, _e), I(_f, _e) +#define I2bvIP(_f, _e, _i, _p) \ + IIP((_f) | ByteOp, _e, _i, _p), IIP(_f, _e, _i, _p) #define I6ALU(_f, _e) I2bv((_f) | DstMem | SrcReg | ModRM, _e), \ I2bv(((_f) | DstReg | SrcMem | ModRM) & ~Lock, _e), \ @@ -3033,17 +3199,17 @@ static struct opcode group7_rm7[] = { static struct opcode group1[] = { I(Lock, em_add), - I(Lock, em_or), + I(Lock | PageTable, em_or), I(Lock, em_adc), I(Lock, em_sbb), - I(Lock, em_and), + I(Lock | PageTable, em_and), I(Lock, em_sub), I(Lock, em_xor), I(0, em_cmp), }; static struct opcode group1A[] = { - D(DstMem | SrcNone | ModRM | Mov | Stack), N, N, N, N, N, N, N, + I(DstMem | SrcNone | ModRM | Mov | Stack, em_pop), N, N, N, N, N, N, N, }; static struct opcode group3[] = { @@ -3058,16 +3224,19 @@ static struct opcode group3[] = { }; static struct opcode group4[] = { - D(ByteOp | DstMem | SrcNone | ModRM | Lock), D(ByteOp | DstMem | SrcNone | ModRM | Lock), + I(ByteOp | DstMem | SrcNone | ModRM | Lock, em_grp45), + I(ByteOp | DstMem | SrcNone | ModRM | Lock, em_grp45), N, N, N, N, N, N, }; static struct opcode group5[] = { - D(DstMem | SrcNone | ModRM | Lock), D(DstMem | SrcNone | ModRM | Lock), - D(SrcMem | ModRM | Stack), + I(DstMem | SrcNone | ModRM | Lock, em_grp45), + I(DstMem | SrcNone | ModRM | Lock, em_grp45), + I(SrcMem | ModRM | Stack, em_grp45), I(SrcMemFAddr | ModRM | ImplicitOps | Stack, em_call_far), - D(SrcMem | ModRM | Stack), D(SrcMemFAddr | ModRM | ImplicitOps), - D(SrcMem | ModRM | Stack), N, + I(SrcMem | ModRM | Stack, em_grp45), + I(SrcMemFAddr | ModRM | ImplicitOps, em_grp45), + I(SrcMem | ModRM | Stack, em_grp45), N, }; static struct opcode group6[] = { @@ -3096,18 +3265,21 @@ static struct group_dual group7 = { { static struct opcode group8[] = { N, N, N, N, - D(DstMem | SrcImmByte | ModRM), D(DstMem | SrcImmByte | ModRM | Lock), - D(DstMem | SrcImmByte | ModRM | Lock), D(DstMem | SrcImmByte | ModRM | Lock), + I(DstMem | SrcImmByte | ModRM, em_bt), + I(DstMem | SrcImmByte | ModRM | Lock | PageTable, em_bts), + I(DstMem | SrcImmByte | ModRM | Lock, em_btr), + I(DstMem | SrcImmByte | ModRM | Lock | PageTable, em_btc), }; static struct group_dual group9 = { { - N, D(DstMem64 | ModRM | Lock), N, N, N, N, N, N, + N, I(DstMem64 | ModRM | Lock | PageTable, em_cmpxchg8b), N, N, N, N, N, N, }, { N, N, N, N, N, N, N, N, } }; static struct opcode group11[] = { - I(DstMem | SrcImm | ModRM | Mov, em_mov), X7(D(Undefined)), + I(DstMem | SrcImm | ModRM | Mov | PageTable, em_mov), + X7(D(Undefined)), }; static struct gprefix pfx_0f_6f_0f_7f = { @@ -3120,7 +3292,7 @@ static struct opcode opcode_table[256] = { I(ImplicitOps | Stack | No64 | Src2ES, em_push_sreg), I(ImplicitOps | Stack | No64 | Src2ES, em_pop_sreg), /* 0x08 - 0x0F */ - I6ALU(Lock, em_or), + I6ALU(Lock | PageTable, em_or), I(ImplicitOps | Stack | No64 | Src2CS, em_push_sreg), N, /* 0x10 - 0x17 */ @@ -3132,7 +3304,7 @@ static struct opcode opcode_table[256] = { I(ImplicitOps | Stack | No64 | Src2DS, em_push_sreg), I(ImplicitOps | Stack | No64 | Src2DS, em_pop_sreg), /* 0x20 - 0x27 */ - I6ALU(Lock, em_and), N, N, + I6ALU(Lock | PageTable, em_and), N, N, /* 0x28 - 0x2F */ I6ALU(Lock, em_sub), N, I(ByteOp | DstAcc | No64, em_das), /* 0x30 - 0x37 */ @@ -3155,8 +3327,8 @@ static struct opcode opcode_table[256] = { I(DstReg | SrcMem | ModRM | Src2Imm, em_imul_3op), I(SrcImmByte | Mov | Stack, em_push), I(DstReg | SrcMem | ModRM | Src2ImmByte, em_imul_3op), - D2bvIP(DstDI | SrcDX | Mov | String, ins, check_perm_in), /* insb, insw/insd */ - D2bvIP(SrcSI | DstDX | String, outs, check_perm_out), /* outsb, outsw/outsd */ + I2bvIP(DstDI | SrcDX | Mov | String, em_in, ins, check_perm_in), /* insb, insw/insd */ + I2bvIP(SrcSI | DstDX | String, em_out, outs, check_perm_out), /* outsb, outsw/outsd */ /* 0x70 - 0x7F */ X16(D(SrcImmByte)), /* 0x80 - 0x87 */ @@ -3165,11 +3337,11 @@ static struct opcode opcode_table[256] = { G(ByteOp | DstMem | SrcImm | ModRM | No64 | Group, group1), G(DstMem | SrcImmByte | ModRM | Group, group1), I2bv(DstMem | SrcReg | ModRM, em_test), - I2bv(DstMem | SrcReg | ModRM | Lock, em_xchg), + I2bv(DstMem | SrcReg | ModRM | Lock | PageTable, em_xchg), /* 0x88 - 0x8F */ - I2bv(DstMem | SrcReg | ModRM | Mov, em_mov), + I2bv(DstMem | SrcReg | ModRM | Mov | PageTable, em_mov), I2bv(DstReg | SrcMem | ModRM | Mov, em_mov), - I(DstMem | SrcNone | ModRM | Mov, em_mov_rm_sreg), + I(DstMem | SrcNone | ModRM | Mov | PageTable, em_mov_rm_sreg), D(ModRM | SrcMem | NoAccess | DstReg), I(ImplicitOps | SrcMem16 | ModRM, em_mov_sreg_rm), G(0, group1A), @@ -3182,7 +3354,7 @@ static struct opcode opcode_table[256] = { II(ImplicitOps | Stack, em_popf, popf), N, N, /* 0xA0 - 0xA7 */ I2bv(DstAcc | SrcMem | Mov | MemAbs, em_mov), - I2bv(DstMem | SrcAcc | Mov | MemAbs, em_mov), + I2bv(DstMem | SrcAcc | Mov | MemAbs | PageTable, em_mov), I2bv(SrcSI | DstDI | Mov | String, em_mov), I2bv(SrcSI | DstDI | String, em_cmp), /* 0xA8 - 0xAF */ @@ -3213,13 +3385,13 @@ static struct opcode opcode_table[256] = { /* 0xE0 - 0xE7 */ X3(I(SrcImmByte, em_loop)), I(SrcImmByte, em_jcxz), - D2bvIP(SrcImmUByte | DstAcc, in, check_perm_in), - D2bvIP(SrcAcc | DstImmUByte, out, check_perm_out), + I2bvIP(SrcImmUByte | DstAcc, em_in, in, check_perm_in), + I2bvIP(SrcAcc | DstImmUByte, em_out, out, check_perm_out), /* 0xE8 - 0xEF */ - D(SrcImm | Stack), D(SrcImm | ImplicitOps), + I(SrcImm | Stack, em_call), D(SrcImm | ImplicitOps), I(SrcImmFAddr | No64, em_jmp_far), D(SrcImmByte | ImplicitOps), - D2bvIP(SrcDX | DstAcc, in, check_perm_in), - D2bvIP(SrcAcc | DstDX, out, check_perm_out), + I2bvIP(SrcDX | DstAcc, em_in, in, check_perm_in), + I2bvIP(SrcAcc | DstDX, em_out, out, check_perm_out), /* 0xF0 - 0xF7 */ N, DI(ImplicitOps, icebp), N, N, DI(ImplicitOps | Priv, hlt), D(ImplicitOps), @@ -3242,15 +3414,15 @@ static struct opcode twobyte_table[256] = { /* 0x20 - 0x2F */ DIP(ModRM | DstMem | Priv | Op3264, cr_read, check_cr_read), DIP(ModRM | DstMem | Priv | Op3264, dr_read, check_dr_read), - DIP(ModRM | SrcMem | Priv | Op3264, cr_write, check_cr_write), - DIP(ModRM | SrcMem | Priv | Op3264, dr_write, check_dr_write), + IIP(ModRM | SrcMem | Priv | Op3264, em_cr_write, cr_write, check_cr_write), + IIP(ModRM | SrcMem | Priv | Op3264, em_dr_write, dr_write, check_dr_write), N, N, N, N, N, N, N, N, N, N, N, N, /* 0x30 - 0x3F */ - DI(ImplicitOps | Priv, wrmsr), + II(ImplicitOps | Priv, em_wrmsr, wrmsr), IIP(ImplicitOps, em_rdtsc, rdtsc, check_rdtsc), - DI(ImplicitOps | Priv, rdmsr), - DIP(ImplicitOps | Priv, rdpmc, check_rdpmc), + II(ImplicitOps | Priv, em_rdmsr, rdmsr), + IIP(ImplicitOps, em_rdpmc, rdpmc, check_rdpmc), I(ImplicitOps | VendorSpecific, em_sysenter), I(ImplicitOps | Priv | VendorSpecific, em_sysexit), N, N, @@ -3275,26 +3447,28 @@ static struct opcode twobyte_table[256] = { X16(D(ByteOp | DstMem | SrcNone | ModRM| Mov)), /* 0xA0 - 0xA7 */ I(Stack | Src2FS, em_push_sreg), I(Stack | Src2FS, em_pop_sreg), - DI(ImplicitOps, cpuid), D(DstMem | SrcReg | ModRM | BitOp), + DI(ImplicitOps, cpuid), I(DstMem | SrcReg | ModRM | BitOp, em_bt), D(DstMem | SrcReg | Src2ImmByte | ModRM), D(DstMem | SrcReg | Src2CL | ModRM), N, N, /* 0xA8 - 0xAF */ I(Stack | Src2GS, em_push_sreg), I(Stack | Src2GS, em_pop_sreg), - DI(ImplicitOps, rsm), D(DstMem | SrcReg | ModRM | BitOp | Lock), + DI(ImplicitOps, rsm), + I(DstMem | SrcReg | ModRM | BitOp | Lock | PageTable, em_bts), D(DstMem | SrcReg | Src2ImmByte | ModRM), D(DstMem | SrcReg | Src2CL | ModRM), D(ModRM), I(DstReg | SrcMem | ModRM, em_imul), /* 0xB0 - 0xB7 */ - D2bv(DstMem | SrcReg | ModRM | Lock), + I2bv(DstMem | SrcReg | ModRM | Lock | PageTable, em_cmpxchg), I(DstReg | SrcMemFAddr | ModRM | Src2SS, em_lseg), - D(DstMem | SrcReg | ModRM | BitOp | Lock), + I(DstMem | SrcReg | ModRM | BitOp | Lock, em_btr), I(DstReg | SrcMemFAddr | ModRM | Src2FS, em_lseg), I(DstReg | SrcMemFAddr | ModRM | Src2GS, em_lseg), D(ByteOp | DstReg | SrcMem | ModRM | Mov), D(DstReg | SrcMem16 | ModRM | Mov), /* 0xB8 - 0xBF */ N, N, - G(BitOp, group8), D(DstMem | SrcReg | ModRM | BitOp | Lock), - D(DstReg | SrcMem | ModRM), D(DstReg | SrcMem | ModRM), + G(BitOp, group8), + I(DstMem | SrcReg | ModRM | BitOp | Lock | PageTable, em_btc), + I(DstReg | SrcMem | ModRM, em_bsf), I(DstReg | SrcMem | ModRM, em_bsr), D(ByteOp | DstReg | SrcMem | ModRM | Mov), D(DstReg | SrcMem16 | ModRM | Mov), /* 0xC0 - 0xCF */ D2bv(DstMem | SrcReg | ModRM | Lock), @@ -3320,6 +3494,7 @@ static struct opcode twobyte_table[256] = { #undef D2bv #undef D2bvIP #undef I2bv +#undef I2bvIP #undef I6ALU static unsigned imm_size(struct x86_emulate_ctxt *ctxt) @@ -3697,6 +3872,11 @@ done: return (rc != X86EMUL_CONTINUE) ? EMULATION_FAILED : EMULATION_OK; } +bool x86_page_table_writing_insn(struct x86_emulate_ctxt *ctxt) +{ + return ctxt->d & PageTable; +} + static bool string_insn_completed(struct x86_emulate_ctxt *ctxt) { /* The second termination condition only applies for REPE @@ -3720,7 +3900,6 @@ static bool string_insn_completed(struct x86_emulate_ctxt *ctxt) int x86_emulate_insn(struct x86_emulate_ctxt *ctxt) { struct x86_emulate_ops *ops = ctxt->ops; - u64 msr_data; int rc = X86EMUL_CONTINUE; int saved_dst_type = ctxt->dst.type; @@ -3854,15 +4033,6 @@ special_insn: goto cannot_emulate; ctxt->dst.val = (s32) ctxt->src.val; break; - case 0x6c: /* insb */ - case 0x6d: /* insw/insd */ - ctxt->src.val = ctxt->regs[VCPU_REGS_RDX]; - goto do_io_in; - case 0x6e: /* outsb */ - case 0x6f: /* outsw/outsd */ - ctxt->dst.val = ctxt->regs[VCPU_REGS_RDX]; - goto do_io_out; - break; case 0x70 ... 0x7f: /* jcc (short) */ if (test_cc(ctxt->b, ctxt->eflags)) jmp_rel(ctxt, ctxt->src.val); @@ -3870,9 +4040,6 @@ special_insn: case 0x8d: /* lea r16/r32, m */ ctxt->dst.val = ctxt->src.addr.mem.ea; break; - case 0x8f: /* pop (sole member of Grp1a) */ - rc = em_grp1a(ctxt); - break; case 0x90 ... 0x97: /* nop / xchg reg, rax */ if (ctxt->dst.addr.reg == &ctxt->regs[VCPU_REGS_RAX]) break; @@ -3905,38 +4072,11 @@ special_insn: ctxt->src.val = ctxt->regs[VCPU_REGS_RCX]; rc = em_grp2(ctxt); break; - case 0xe4: /* inb */ - case 0xe5: /* in */ - goto do_io_in; - case 0xe6: /* outb */ - case 0xe7: /* out */ - goto do_io_out; - case 0xe8: /* call (near) */ { - long int rel = ctxt->src.val; - ctxt->src.val = (unsigned long) ctxt->_eip; - jmp_rel(ctxt, rel); - rc = em_push(ctxt); - break; - } case 0xe9: /* jmp rel */ case 0xeb: /* jmp rel short */ jmp_rel(ctxt, ctxt->src.val); ctxt->dst.type = OP_NONE; /* Disable writeback. */ break; - case 0xec: /* in al,dx */ - case 0xed: /* in (e/r)ax,dx */ - do_io_in: - if (!pio_in_emulated(ctxt, ctxt->dst.bytes, ctxt->src.val, - &ctxt->dst.val)) - goto done; /* IO is needed */ - break; - case 0xee: /* out dx,al */ - case 0xef: /* out dx,(e/r)ax */ - do_io_out: - ops->pio_out_emulated(ctxt, ctxt->src.bytes, ctxt->dst.val, - &ctxt->src.val, 1); - ctxt->dst.type = OP_NONE; /* Disable writeback. */ - break; case 0xf4: /* hlt */ ctxt->ops->halt(ctxt); break; @@ -3956,12 +4096,6 @@ special_insn: case 0xfd: /* std */ ctxt->eflags |= EFLG_DF; break; - case 0xfe: /* Grp4 */ - rc = em_grp45(ctxt); - break; - case 0xff: /* Grp5 */ - rc = em_grp45(ctxt); - break; default: goto cannot_emulate; } @@ -4036,49 +4170,6 @@ twobyte_insn: case 0x21: /* mov from dr to reg */ ops->get_dr(ctxt, ctxt->modrm_reg, &ctxt->dst.val); break; - case 0x22: /* mov reg, cr */ - if (ops->set_cr(ctxt, ctxt->modrm_reg, ctxt->src.val)) { - emulate_gp(ctxt, 0); - rc = X86EMUL_PROPAGATE_FAULT; - goto done; - } - ctxt->dst.type = OP_NONE; - break; - case 0x23: /* mov from reg to dr */ - if (ops->set_dr(ctxt, ctxt->modrm_reg, ctxt->src.val & - ((ctxt->mode == X86EMUL_MODE_PROT64) ? - ~0ULL : ~0U)) < 0) { - /* #UD condition is already handled by the code above */ - emulate_gp(ctxt, 0); - rc = X86EMUL_PROPAGATE_FAULT; - goto done; - } - - ctxt->dst.type = OP_NONE; /* no writeback */ - break; - case 0x30: - /* wrmsr */ - msr_data = (u32)ctxt->regs[VCPU_REGS_RAX] - | ((u64)ctxt->regs[VCPU_REGS_RDX] << 32); - if (ops->set_msr(ctxt, ctxt->regs[VCPU_REGS_RCX], msr_data)) { - emulate_gp(ctxt, 0); - rc = X86EMUL_PROPAGATE_FAULT; - goto done; - } - rc = X86EMUL_CONTINUE; - break; - case 0x32: - /* rdmsr */ - if (ops->get_msr(ctxt, ctxt->regs[VCPU_REGS_RCX], &msr_data)) { - emulate_gp(ctxt, 0); - rc = X86EMUL_PROPAGATE_FAULT; - goto done; - } else { - ctxt->regs[VCPU_REGS_RAX] = (u32)msr_data; - ctxt->regs[VCPU_REGS_RDX] = msr_data >> 32; - } - rc = X86EMUL_CONTINUE; - break; case 0x40 ... 0x4f: /* cmov */ ctxt->dst.val = ctxt->dst.orig_val = ctxt->src.val; if (!test_cc(ctxt->b, ctxt->eflags)) @@ -4091,93 +4182,21 @@ twobyte_insn: case 0x90 ... 0x9f: /* setcc r/m8 */ ctxt->dst.val = test_cc(ctxt->b, ctxt->eflags); break; - case 0xa3: - bt: /* bt */ - ctxt->dst.type = OP_NONE; - /* only subword offset */ - ctxt->src.val &= (ctxt->dst.bytes << 3) - 1; - emulate_2op_SrcV_nobyte(ctxt, "bt"); - break; case 0xa4: /* shld imm8, r, r/m */ case 0xa5: /* shld cl, r, r/m */ emulate_2op_cl(ctxt, "shld"); break; - case 0xab: - bts: /* bts */ - emulate_2op_SrcV_nobyte(ctxt, "bts"); - break; case 0xac: /* shrd imm8, r, r/m */ case 0xad: /* shrd cl, r, r/m */ emulate_2op_cl(ctxt, "shrd"); break; case 0xae: /* clflush */ break; - case 0xb0 ... 0xb1: /* cmpxchg */ - /* - * Save real source value, then compare EAX against - * destination. - */ - ctxt->src.orig_val = ctxt->src.val; - ctxt->src.val = ctxt->regs[VCPU_REGS_RAX]; - emulate_2op_SrcV(ctxt, "cmp"); - if (ctxt->eflags & EFLG_ZF) { - /* Success: write back to memory. */ - ctxt->dst.val = ctxt->src.orig_val; - } else { - /* Failure: write the value we saw to EAX. */ - ctxt->dst.type = OP_REG; - ctxt->dst.addr.reg = (unsigned long *)&ctxt->regs[VCPU_REGS_RAX]; - } - break; - case 0xb3: - btr: /* btr */ - emulate_2op_SrcV_nobyte(ctxt, "btr"); - break; case 0xb6 ... 0xb7: /* movzx */ ctxt->dst.bytes = ctxt->op_bytes; ctxt->dst.val = (ctxt->d & ByteOp) ? (u8) ctxt->src.val : (u16) ctxt->src.val; break; - case 0xba: /* Grp8 */ - switch (ctxt->modrm_reg & 3) { - case 0: - goto bt; - case 1: - goto bts; - case 2: - goto btr; - case 3: - goto btc; - } - break; - case 0xbb: - btc: /* btc */ - emulate_2op_SrcV_nobyte(ctxt, "btc"); - break; - case 0xbc: { /* bsf */ - u8 zf; - __asm__ ("bsf %2, %0; setz %1" - : "=r"(ctxt->dst.val), "=q"(zf) - : "r"(ctxt->src.val)); - ctxt->eflags &= ~X86_EFLAGS_ZF; - if (zf) { - ctxt->eflags |= X86_EFLAGS_ZF; - ctxt->dst.type = OP_NONE; /* Disable writeback. */ - } - break; - } - case 0xbd: { /* bsr */ - u8 zf; - __asm__ ("bsr %2, %0; setz %1" - : "=r"(ctxt->dst.val), "=q"(zf) - : "r"(ctxt->src.val)); - ctxt->eflags &= ~X86_EFLAGS_ZF; - if (zf) { - ctxt->eflags |= X86_EFLAGS_ZF; - ctxt->dst.type = OP_NONE; /* Disable writeback. */ - } - break; - } case 0xbe ... 0xbf: /* movsx */ ctxt->dst.bytes = ctxt->op_bytes; ctxt->dst.val = (ctxt->d & ByteOp) ? (s8) ctxt->src.val : @@ -4194,9 +4213,6 @@ twobyte_insn: ctxt->dst.val = (ctxt->op_bytes == 4) ? (u32) ctxt->src.val : (u64) ctxt->src.val; break; - case 0xc7: /* Grp9 (cmpxchg8b) */ - rc = em_grp9(ctxt); - break; default: goto cannot_emulate; } diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c index 405f262..d68f99d 100644 --- a/arch/x86/kvm/i8254.c +++ b/arch/x86/kvm/i8254.c @@ -344,7 +344,7 @@ static void create_pit_timer(struct kvm *kvm, u32 val, int is_period) struct kvm_timer *pt = &ps->pit_timer; s64 interval; - if (!irqchip_in_kernel(kvm)) + if (!irqchip_in_kernel(kvm) || ps->flags & KVM_PIT_FLAGS_HPET_LEGACY) return; interval = muldiv64(val, NSEC_PER_SEC, KVM_PIT_FREQ); @@ -397,15 +397,11 @@ static void pit_load_count(struct kvm *kvm, int channel, u32 val) case 1: /* FIXME: enhance mode 4 precision */ case 4: - if (!(ps->flags & KVM_PIT_FLAGS_HPET_LEGACY)) { - create_pit_timer(kvm, val, 0); - } + create_pit_timer(kvm, val, 0); break; case 2: case 3: - if (!(ps->flags & KVM_PIT_FLAGS_HPET_LEGACY)){ - create_pit_timer(kvm, val, 1); - } + create_pit_timer(kvm, val, 1); break; default: destroy_pit_timer(kvm->arch.vpit); diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c index cac4746..b6a7353 100644 --- a/arch/x86/kvm/i8259.c +++ b/arch/x86/kvm/i8259.c @@ -262,9 +262,10 @@ int kvm_pic_read_irq(struct kvm *kvm) void kvm_pic_reset(struct kvm_kpic_state *s) { - int irq; - struct kvm_vcpu *vcpu0 = s->pics_state->kvm->bsp_vcpu; + int irq, i; + struct kvm_vcpu *vcpu; u8 irr = s->irr, isr = s->imr; + bool found = false; s->last_irr = 0; s->irr = 0; @@ -281,12 +282,19 @@ void kvm_pic_reset(struct kvm_kpic_state *s) s->special_fully_nested_mode = 0; s->init4 = 0; - for (irq = 0; irq < PIC_NUM_PINS/2; irq++) { - if (vcpu0 && kvm_apic_accept_pic_intr(vcpu0)) - if (irr & (1 << irq) || isr & (1 << irq)) { - pic_clear_isr(s, irq); - } - } + kvm_for_each_vcpu(i, vcpu, s->pics_state->kvm) + if (kvm_apic_accept_pic_intr(vcpu)) { + found = true; + break; + } + + + if (!found) + return; + + for (irq = 0; irq < PIC_NUM_PINS/2; irq++) + if (irr & (1 << irq) || isr & (1 << irq)) + pic_clear_isr(s, irq); } static void pic_ioport_write(void *opaque, u32 addr, u32 val) diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 54abb40..cfdc6e0 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -38,6 +38,7 @@ #include "irq.h" #include "trace.h" #include "x86.h" +#include "cpuid.h" #ifndef CONFIG_X86_64 #define mod_64(x, y) ((x) - (y) * div64_u64(x, y)) @@ -1120,7 +1121,7 @@ int apic_has_pending_timer(struct kvm_vcpu *vcpu) return 0; } -static int kvm_apic_local_deliver(struct kvm_lapic *apic, int lvt_type) +int kvm_apic_local_deliver(struct kvm_lapic *apic, int lvt_type) { u32 reg = apic_get_reg(apic, lvt_type); int vector, mode, trig_mode; diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h index 138e8cc..6f4ce25 100644 --- a/arch/x86/kvm/lapic.h +++ b/arch/x86/kvm/lapic.h @@ -34,6 +34,7 @@ void kvm_apic_set_version(struct kvm_vcpu *vcpu); int kvm_apic_match_physical_addr(struct kvm_lapic *apic, u16 dest); int kvm_apic_match_logical_addr(struct kvm_lapic *apic, u8 mda); int kvm_apic_set_irq(struct kvm_vcpu *vcpu, struct kvm_lapic_irq *irq); +int kvm_apic_local_deliver(struct kvm_lapic *apic, int lvt_type); u64 kvm_get_apic_base(struct kvm_vcpu *vcpu); void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data); diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index f1b36cf..2a2a9b4 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -59,15 +59,6 @@ enum { AUDIT_POST_SYNC }; -char *audit_point_name[] = { - "pre page fault", - "post page fault", - "pre pte write", - "post pte write", - "pre sync", - "post sync" -}; - #undef MMU_DEBUG #ifdef MMU_DEBUG @@ -87,9 +78,6 @@ static int dbg = 0; module_param(dbg, bool, 0644); #endif -static int oos_shadow = 1; -module_param(oos_shadow, bool, 0644); - #ifndef MMU_DEBUG #define ASSERT(x) do { } while (0) #else @@ -593,6 +581,11 @@ static int mmu_topup_memory_cache(struct kvm_mmu_memory_cache *cache, return 0; } +static int mmu_memory_cache_free_objects(struct kvm_mmu_memory_cache *cache) +{ + return cache->nobjs; +} + static void mmu_free_memory_cache(struct kvm_mmu_memory_cache *mc, struct kmem_cache *cache) { @@ -953,21 +946,35 @@ static void pte_list_walk(unsigned long *pte_list, pte_list_walk_fn fn) } } +static unsigned long *__gfn_to_rmap(struct kvm *kvm, gfn_t gfn, int level, + struct kvm_memory_slot *slot) +{ + struct kvm_lpage_info *linfo; + + if (likely(level == PT_PAGE_TABLE_LEVEL)) + return &slot->rmap[gfn - slot->base_gfn]; + + linfo = lpage_info_slot(gfn, slot, level); + return &linfo->rmap_pde; +} + /* * Take gfn and return the reverse mapping to it. */ static unsigned long *gfn_to_rmap(struct kvm *kvm, gfn_t gfn, int level) { struct kvm_memory_slot *slot; - struct kvm_lpage_info *linfo; slot = gfn_to_memslot(kvm, gfn); - if (likely(level == PT_PAGE_TABLE_LEVEL)) - return &slot->rmap[gfn - slot->base_gfn]; + return __gfn_to_rmap(kvm, gfn, level, slot); +} - linfo = lpage_info_slot(gfn, slot, level); +static bool rmap_can_add(struct kvm_vcpu *vcpu) +{ + struct kvm_mmu_memory_cache *cache; - return &linfo->rmap_pde; + cache = &vcpu->arch.mmu_pte_list_desc_cache; + return mmu_memory_cache_free_objects(cache); } static int rmap_add(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn) @@ -1004,17 +1011,16 @@ static void drop_spte(struct kvm *kvm, u64 *sptep) rmap_remove(kvm, sptep); } -static int rmap_write_protect(struct kvm *kvm, u64 gfn) +int kvm_mmu_rmap_write_protect(struct kvm *kvm, u64 gfn, + struct kvm_memory_slot *slot) { unsigned long *rmapp; u64 *spte; int i, write_protected = 0; - rmapp = gfn_to_rmap(kvm, gfn, PT_PAGE_TABLE_LEVEL); - + rmapp = __gfn_to_rmap(kvm, gfn, PT_PAGE_TABLE_LEVEL, slot); spte = rmap_next(kvm, rmapp, NULL); while (spte) { - BUG_ON(!spte); BUG_ON(!(*spte & PT_PRESENT_MASK)); rmap_printk("rmap_write_protect: spte %p %llx\n", spte, *spte); if (is_writable_pte(*spte)) { @@ -1027,12 +1033,11 @@ static int rmap_write_protect(struct kvm *kvm, u64 gfn) /* check for huge page mappings */ for (i = PT_DIRECTORY_LEVEL; i < PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES; ++i) { - rmapp = gfn_to_rmap(kvm, gfn, i); + rmapp = __gfn_to_rmap(kvm, gfn, i, slot); spte = rmap_next(kvm, rmapp, NULL); while (spte) { - BUG_ON(!spte); BUG_ON(!(*spte & PT_PRESENT_MASK)); - BUG_ON((*spte & (PT_PAGE_SIZE_MASK|PT_PRESENT_MASK)) != (PT_PAGE_SIZE_MASK|PT_PRESENT_MASK)); + BUG_ON(!is_large_pte(*spte)); pgprintk("rmap_write_protect(large): spte %p %llx %lld\n", spte, *spte, gfn); if (is_writable_pte(*spte)) { drop_spte(kvm, spte); @@ -1047,6 +1052,14 @@ static int rmap_write_protect(struct kvm *kvm, u64 gfn) return write_protected; } +static int rmap_write_protect(struct kvm *kvm, u64 gfn) +{ + struct kvm_memory_slot *slot; + + slot = gfn_to_memslot(kvm, gfn); + return kvm_mmu_rmap_write_protect(kvm, gfn, slot); +} + static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp, unsigned long data) { @@ -1103,15 +1116,15 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva, int (*handler)(struct kvm *kvm, unsigned long *rmapp, unsigned long data)) { - int i, j; + int j; int ret; int retval = 0; struct kvm_memslots *slots; + struct kvm_memory_slot *memslot; slots = kvm_memslots(kvm); - for (i = 0; i < slots->nmemslots; i++) { - struct kvm_memory_slot *memslot = &slots->memslots[i]; + kvm_for_each_memslot(memslot, slots) { unsigned long start = memslot->userspace_addr; unsigned long end; @@ -1324,7 +1337,7 @@ static struct kvm_mmu_page *kvm_mmu_alloc_page(struct kvm_vcpu *vcpu, PAGE_SIZE); set_page_private(virt_to_page(sp->spt), (unsigned long)sp); list_add(&sp->link, &vcpu->kvm->arch.active_mmu_pages); - bitmap_zero(sp->slot_bitmap, KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS); + bitmap_zero(sp->slot_bitmap, KVM_MEM_SLOTS_NUM); sp->parent_ptes = 0; mmu_page_add_parent_pte(vcpu, sp, parent_pte); kvm_mod_used_mmu_pages(vcpu->kvm, +1); @@ -1511,6 +1524,13 @@ static int kvm_sync_page_transient(struct kvm_vcpu *vcpu, return ret; } +#ifdef CONFIG_KVM_MMU_AUDIT +#include "mmu_audit.c" +#else +static void kvm_mmu_audit(struct kvm_vcpu *vcpu, int point) { } +static void mmu_audit_disable(void) { } +#endif + static int kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, struct list_head *invalid_list) { @@ -1640,6 +1660,18 @@ static void init_shadow_page_table(struct kvm_mmu_page *sp) sp->spt[i] = 0ull; } +static void __clear_sp_write_flooding_count(struct kvm_mmu_page *sp) +{ + sp->write_flooding_count = 0; +} + +static void clear_sp_write_flooding_count(u64 *spte) +{ + struct kvm_mmu_page *sp = page_header(__pa(spte)); + + __clear_sp_write_flooding_count(sp); +} + static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, gfn_t gfn, gva_t gaddr, @@ -1683,6 +1715,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, } else if (sp->unsync) kvm_mmu_mark_parents_unsync(sp); + __clear_sp_write_flooding_count(sp); trace_kvm_mmu_get_page(sp, false); return sp; } @@ -1796,7 +1829,7 @@ static void validate_direct_spte(struct kvm_vcpu *vcpu, u64 *sptep, } } -static void mmu_page_zap_pte(struct kvm *kvm, struct kvm_mmu_page *sp, +static bool mmu_page_zap_pte(struct kvm *kvm, struct kvm_mmu_page *sp, u64 *spte) { u64 pte; @@ -1804,17 +1837,21 @@ static void mmu_page_zap_pte(struct kvm *kvm, struct kvm_mmu_page *sp, pte = *spte; if (is_shadow_present_pte(pte)) { - if (is_last_spte(pte, sp->role.level)) + if (is_last_spte(pte, sp->role.level)) { drop_spte(kvm, spte); - else { + if (is_large_pte(pte)) + --kvm->stat.lpages; + } else { child = page_header(pte & PT64_BASE_ADDR_MASK); drop_parent_pte(child, spte); } - } else if (is_mmio_spte(pte)) + return true; + } + + if (is_mmio_spte(pte)) mmu_spte_clear_no_track(spte); - if (is_large_pte(pte)) - --kvm->stat.lpages; + return false; } static void kvm_mmu_page_unlink_children(struct kvm *kvm, @@ -1831,15 +1868,6 @@ static void kvm_mmu_put_page(struct kvm_mmu_page *sp, u64 *parent_pte) mmu_page_remove_parent_pte(sp, parent_pte); } -static void kvm_mmu_reset_last_pte_updated(struct kvm *kvm) -{ - int i; - struct kvm_vcpu *vcpu; - - kvm_for_each_vcpu(i, vcpu, kvm) - vcpu->arch.last_pte_updated = NULL; -} - static void kvm_mmu_unlink_parents(struct kvm *kvm, struct kvm_mmu_page *sp) { u64 *parent_pte; @@ -1899,7 +1927,6 @@ static int kvm_mmu_prepare_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp, } sp->role.invalid = 1; - kvm_mmu_reset_last_pte_updated(kvm); return ret; } @@ -1985,7 +2012,7 @@ void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int goal_nr_mmu_pages) kvm->arch.n_max_mmu_pages = goal_nr_mmu_pages; } -static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn) +int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn) { struct kvm_mmu_page *sp; struct hlist_node *node; @@ -1994,7 +2021,7 @@ static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn) pgprintk("%s: looking for gfn %llx\n", __func__, gfn); r = 0; - + spin_lock(&kvm->mmu_lock); for_each_gfn_indirect_valid_sp(kvm, sp, gfn, node) { pgprintk("%s: gfn %llx role %x\n", __func__, gfn, sp->role.word); @@ -2002,22 +2029,11 @@ static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn) kvm_mmu_prepare_zap_page(kvm, sp, &invalid_list); } kvm_mmu_commit_zap_page(kvm, &invalid_list); - return r; -} - -static void mmu_unshadow(struct kvm *kvm, gfn_t gfn) -{ - struct kvm_mmu_page *sp; - struct hlist_node *node; - LIST_HEAD(invalid_list); + spin_unlock(&kvm->mmu_lock); - for_each_gfn_indirect_valid_sp(kvm, sp, gfn, node) { - pgprintk("%s: zap %llx %x\n", - __func__, gfn, sp->role.word); - kvm_mmu_prepare_zap_page(kvm, sp, &invalid_list); - } - kvm_mmu_commit_zap_page(kvm, &invalid_list); + return r; } +EXPORT_SYMBOL_GPL(kvm_mmu_unprotect_page); static void page_header_update_slot(struct kvm *kvm, void *pte, gfn_t gfn) { @@ -2169,8 +2185,6 @@ static int mmu_need_write_protect(struct kvm_vcpu *vcpu, gfn_t gfn, return 1; if (!need_unsync && !s->unsync) { - if (!oos_shadow) - return 1; need_unsync = true; } } @@ -2191,11 +2205,6 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep, if (set_mmio_spte(sptep, gfn, pfn, pte_access)) return 0; - /* - * We don't set the accessed bit, since we sometimes want to see - * whether the guest actually used the pte (in order to detect - * demand paging). - */ spte = PT_PRESENT_MASK; if (!speculative) spte |= shadow_accessed_mask; @@ -2346,10 +2355,6 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep, } } kvm_release_pfn_clean(pfn); - if (speculative) { - vcpu->arch.last_pte_updated = sptep; - vcpu->arch.last_pte_gfn = gfn; - } } static void nonpaging_new_cr3(struct kvm_vcpu *vcpu) @@ -2840,12 +2845,12 @@ static void mmu_sync_roots(struct kvm_vcpu *vcpu) return; vcpu_clear_mmio_info(vcpu, ~0ul); - trace_kvm_mmu_audit(vcpu, AUDIT_PRE_SYNC); + kvm_mmu_audit(vcpu, AUDIT_PRE_SYNC); if (vcpu->arch.mmu.root_level == PT64_ROOT_LEVEL) { hpa_t root = vcpu->arch.mmu.root_hpa; sp = page_header(root); mmu_sync_children(vcpu, sp); - trace_kvm_mmu_audit(vcpu, AUDIT_POST_SYNC); + kvm_mmu_audit(vcpu, AUDIT_POST_SYNC); return; } for (i = 0; i < 4; ++i) { @@ -2857,7 +2862,7 @@ static void mmu_sync_roots(struct kvm_vcpu *vcpu) mmu_sync_children(vcpu, sp); } } - trace_kvm_mmu_audit(vcpu, AUDIT_POST_SYNC); + kvm_mmu_audit(vcpu, AUDIT_POST_SYNC); } void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu) @@ -3510,28 +3515,119 @@ static void mmu_pte_write_flush_tlb(struct kvm_vcpu *vcpu, bool zap_page, kvm_mmu_flush_tlb(vcpu); } -static bool last_updated_pte_accessed(struct kvm_vcpu *vcpu) +static u64 mmu_pte_write_fetch_gpte(struct kvm_vcpu *vcpu, gpa_t *gpa, + const u8 *new, int *bytes) { - u64 *spte = vcpu->arch.last_pte_updated; + u64 gentry; + int r; + + /* + * Assume that the pte write on a page table of the same type + * as the current vcpu paging mode since we update the sptes only + * when they have the same mode. + */ + if (is_pae(vcpu) && *bytes == 4) { + /* Handle a 32-bit guest writing two halves of a 64-bit gpte */ + *gpa &= ~(gpa_t)7; + *bytes = 8; + r = kvm_read_guest(vcpu->kvm, *gpa, &gentry, min(*bytes, 8)); + if (r) + gentry = 0; + new = (const u8 *)&gentry; + } - return !!(spte && (*spte & shadow_accessed_mask)); + switch (*bytes) { + case 4: + gentry = *(const u32 *)new; + break; + case 8: + gentry = *(const u64 *)new; + break; + default: + gentry = 0; + break; + } + + return gentry; } -static void kvm_mmu_access_page(struct kvm_vcpu *vcpu, gfn_t gfn) +/* + * If we're seeing too many writes to a page, it may no longer be a page table, + * or we may be forking, in which case it is better to unmap the page. + */ +static bool detect_write_flooding(struct kvm_mmu_page *sp, u64 *spte) { - u64 *spte = vcpu->arch.last_pte_updated; + /* + * Skip write-flooding detected for the sp whose level is 1, because + * it can become unsync, then the guest page is not write-protected. + */ + if (sp->role.level == 1) + return false; - if (spte - && vcpu->arch.last_pte_gfn == gfn - && shadow_accessed_mask - && !(*spte & shadow_accessed_mask) - && is_shadow_present_pte(*spte)) - set_bit(PT_ACCESSED_SHIFT, (unsigned long *)spte); + return ++sp->write_flooding_count >= 3; +} + +/* + * Misaligned accesses are too much trouble to fix up; also, they usually + * indicate a page is not used as a page table. + */ +static bool detect_write_misaligned(struct kvm_mmu_page *sp, gpa_t gpa, + int bytes) +{ + unsigned offset, pte_size, misaligned; + + pgprintk("misaligned: gpa %llx bytes %d role %x\n", + gpa, bytes, sp->role.word); + + offset = offset_in_page(gpa); + pte_size = sp->role.cr4_pae ? 8 : 4; + + /* + * Sometimes, the OS only writes the last one bytes to update status + * bits, for example, in linux, andb instruction is used in clear_bit(). + */ + if (!(offset & (pte_size - 1)) && bytes == 1) + return false; + + misaligned = (offset ^ (offset + bytes - 1)) & ~(pte_size - 1); + misaligned |= bytes < 4; + + return misaligned; +} + +static u64 *get_written_sptes(struct kvm_mmu_page *sp, gpa_t gpa, int *nspte) +{ + unsigned page_offset, quadrant; + u64 *spte; + int level; + + page_offset = offset_in_page(gpa); + level = sp->role.level; + *nspte = 1; + if (!sp->role.cr4_pae) { + page_offset <<= 1; /* 32->64 */ + /* + * A 32-bit pde maps 4MB while the shadow pdes map + * only 2MB. So we need to double the offset again + * and zap two pdes instead of one. + */ + if (level == PT32_ROOT_LEVEL) { + page_offset &= ~7; /* kill rounding error */ + page_offset <<= 1; + *nspte = 2; + } + quadrant = page_offset >> PAGE_SHIFT; + page_offset &= ~PAGE_MASK; + if (quadrant != sp->role.quadrant) + return NULL; + } + + spte = &sp->spt[page_offset / sizeof(*spte)]; + return spte; } void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, - const u8 *new, int bytes, - bool guest_initiated) + const u8 *new, int bytes) { gfn_t gfn = gpa >> PAGE_SHIFT; union kvm_mmu_page_role mask = { .word = 0 }; @@ -3539,8 +3635,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, struct hlist_node *node; LIST_HEAD(invalid_list); u64 entry, gentry, *spte; - unsigned pte_size, page_offset, misaligned, quadrant, offset; - int level, npte, invlpg_counter, r, flooded = 0; + int npte; bool remote_flush, local_flush, zap_page; /* @@ -3551,112 +3646,45 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, return; zap_page = remote_flush = local_flush = false; - offset = offset_in_page(gpa); pgprintk("%s: gpa %llx bytes %d\n", __func__, gpa, bytes); - invlpg_counter = atomic_read(&vcpu->kvm->arch.invlpg_counter); + gentry = mmu_pte_write_fetch_gpte(vcpu, &gpa, new, &bytes); /* - * Assume that the pte write on a page table of the same type - * as the current vcpu paging mode since we update the sptes only - * when they have the same mode. + * No need to care whether allocation memory is successful + * or not since pte prefetch is skiped if it does not have + * enough objects in the cache. */ - if ((is_pae(vcpu) && bytes == 4) || !new) { - /* Handle a 32-bit guest writing two halves of a 64-bit gpte */ - if (is_pae(vcpu)) { - gpa &= ~(gpa_t)7; - bytes = 8; - } - r = kvm_read_guest(vcpu->kvm, gpa, &gentry, min(bytes, 8)); - if (r) - gentry = 0; - new = (const u8 *)&gentry; - } - - switch (bytes) { - case 4: - gentry = *(const u32 *)new; - break; - case 8: - gentry = *(const u64 *)new; - break; - default: - gentry = 0; - break; - } + mmu_topup_memory_caches(vcpu); spin_lock(&vcpu->kvm->mmu_lock); - if (atomic_read(&vcpu->kvm->arch.invlpg_counter) != invlpg_counter) - gentry = 0; - kvm_mmu_free_some_pages(vcpu); ++vcpu->kvm->stat.mmu_pte_write; - trace_kvm_mmu_audit(vcpu, AUDIT_PRE_PTE_WRITE); - if (guest_initiated) { - kvm_mmu_access_page(vcpu, gfn); - if (gfn == vcpu->arch.last_pt_write_gfn - && !last_updated_pte_accessed(vcpu)) { - ++vcpu->arch.last_pt_write_count; - if (vcpu->arch.last_pt_write_count >= 3) - flooded = 1; - } else { - vcpu->arch.last_pt_write_gfn = gfn; - vcpu->arch.last_pt_write_count = 1; - vcpu->arch.last_pte_updated = NULL; - } - } + kvm_mmu_audit(vcpu, AUDIT_PRE_PTE_WRITE); mask.cr0_wp = mask.cr4_pae = mask.nxe = 1; for_each_gfn_indirect_valid_sp(vcpu->kvm, sp, gfn, node) { - pte_size = sp->role.cr4_pae ? 8 : 4; - misaligned = (offset ^ (offset + bytes - 1)) & ~(pte_size - 1); - misaligned |= bytes < 4; - if (misaligned || flooded) { - /* - * Misaligned accesses are too much trouble to fix - * up; also, they usually indicate a page is not used - * as a page table. - * - * If we're seeing too many writes to a page, - * it may no longer be a page table, or we may be - * forking, in which case it is better to unmap the - * page. - */ - pgprintk("misaligned: gpa %llx bytes %d role %x\n", - gpa, bytes, sp->role.word); + spte = get_written_sptes(sp, gpa, &npte); + + if (detect_write_misaligned(sp, gpa, bytes) || + detect_write_flooding(sp, spte)) { zap_page |= !!kvm_mmu_prepare_zap_page(vcpu->kvm, sp, &invalid_list); ++vcpu->kvm->stat.mmu_flooded; continue; } - page_offset = offset; - level = sp->role.level; - npte = 1; - if (!sp->role.cr4_pae) { - page_offset <<= 1; /* 32->64 */ - /* - * A 32-bit pde maps 4MB while the shadow pdes map - * only 2MB. So we need to double the offset again - * and zap two pdes instead of one. - */ - if (level == PT32_ROOT_LEVEL) { - page_offset &= ~7; /* kill rounding error */ - page_offset <<= 1; - npte = 2; - } - quadrant = page_offset >> PAGE_SHIFT; - page_offset &= ~PAGE_MASK; - if (quadrant != sp->role.quadrant) - continue; - } + + spte = get_written_sptes(sp, gpa, &npte); + if (!spte) + continue; + local_flush = true; - spte = &sp->spt[page_offset / sizeof(*spte)]; while (npte--) { entry = *spte; mmu_page_zap_pte(vcpu->kvm, sp, spte); if (gentry && !((sp->role.word ^ vcpu->arch.mmu.base_role.word) - & mask.word)) + & mask.word) && rmap_can_add(vcpu)) mmu_pte_write_new_pte(vcpu, sp, spte, &gentry); if (!remote_flush && need_remote_flush(entry, *spte)) remote_flush = true; @@ -3665,7 +3693,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, } mmu_pte_write_flush_tlb(vcpu, zap_page, remote_flush, local_flush); kvm_mmu_commit_zap_page(vcpu->kvm, &invalid_list); - trace_kvm_mmu_audit(vcpu, AUDIT_POST_PTE_WRITE); + kvm_mmu_audit(vcpu, AUDIT_POST_PTE_WRITE); spin_unlock(&vcpu->kvm->mmu_lock); } @@ -3679,9 +3707,8 @@ int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva) gpa = kvm_mmu_gva_to_gpa_read(vcpu, gva, NULL); - spin_lock(&vcpu->kvm->mmu_lock); r = kvm_mmu_unprotect_page(vcpu->kvm, gpa >> PAGE_SHIFT); - spin_unlock(&vcpu->kvm->mmu_lock); + return r; } EXPORT_SYMBOL_GPL(kvm_mmu_unprotect_page_virt); @@ -3702,10 +3729,18 @@ void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu) kvm_mmu_commit_zap_page(vcpu->kvm, &invalid_list); } +static bool is_mmio_page_fault(struct kvm_vcpu *vcpu, gva_t addr) +{ + if (vcpu->arch.mmu.direct_map || mmu_is_nested(vcpu)) + return vcpu_match_mmio_gpa(vcpu, addr); + + return vcpu_match_mmio_gva(vcpu, addr); +} + int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code, void *insn, int insn_len) { - int r; + int r, emulation_type = EMULTYPE_RETRY; enum emulation_result er; r = vcpu->arch.mmu.page_fault(vcpu, cr2, error_code, false); @@ -3717,11 +3752,10 @@ int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code, goto out; } - r = mmu_topup_memory_caches(vcpu); - if (r) - goto out; + if (is_mmio_page_fault(vcpu, cr2)) + emulation_type = 0; - er = x86_emulate_instruction(vcpu, cr2, 0, insn, insn_len); + er = x86_emulate_instruction(vcpu, cr2, emulation_type, insn, insn_len); switch (er) { case EMULATE_DONE: @@ -3792,7 +3826,11 @@ static int alloc_mmu_pages(struct kvm_vcpu *vcpu) int kvm_mmu_create(struct kvm_vcpu *vcpu) { ASSERT(vcpu); - ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa)); + + vcpu->arch.walk_mmu = &vcpu->arch.mmu; + vcpu->arch.mmu.root_hpa = INVALID_PAGE; + vcpu->arch.mmu.translate_gpa = translate_gpa; + vcpu->arch.nested_mmu.translate_gpa = translate_nested_gpa; return alloc_mmu_pages(vcpu); } @@ -3852,14 +3890,14 @@ restart: spin_unlock(&kvm->mmu_lock); } -static int kvm_mmu_remove_some_alloc_mmu_pages(struct kvm *kvm, - struct list_head *invalid_list) +static void kvm_mmu_remove_some_alloc_mmu_pages(struct kvm *kvm, + struct list_head *invalid_list) { struct kvm_mmu_page *page; page = container_of(kvm->arch.active_mmu_pages.prev, struct kvm_mmu_page, link); - return kvm_mmu_prepare_zap_page(kvm, page, invalid_list); + kvm_mmu_prepare_zap_page(kvm, page, invalid_list); } static int mmu_shrink(struct shrinker *shrink, struct shrink_control *sc) @@ -3874,15 +3912,15 @@ static int mmu_shrink(struct shrinker *shrink, struct shrink_control *sc) raw_spin_lock(&kvm_lock); list_for_each_entry(kvm, &vm_list, vm_list) { - int idx, freed_pages; + int idx; LIST_HEAD(invalid_list); idx = srcu_read_lock(&kvm->srcu); spin_lock(&kvm->mmu_lock); if (!kvm_freed && nr_to_scan > 0 && kvm->arch.n_used_mmu_pages > 0) { - freed_pages = kvm_mmu_remove_some_alloc_mmu_pages(kvm, - &invalid_list); + kvm_mmu_remove_some_alloc_mmu_pages(kvm, + &invalid_list); kvm_freed = kvm; } nr_to_scan--; @@ -3944,15 +3982,15 @@ nomem: */ unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm) { - int i; unsigned int nr_mmu_pages; unsigned int nr_pages = 0; struct kvm_memslots *slots; + struct kvm_memory_slot *memslot; slots = kvm_memslots(kvm); - for (i = 0; i < slots->nmemslots; i++) - nr_pages += slots->memslots[i].npages; + kvm_for_each_memslot(memslot, slots) + nr_pages += memslot->npages; nr_mmu_pages = nr_pages * KVM_PERMILLE_MMU_PAGES / 1000; nr_mmu_pages = max(nr_mmu_pages, @@ -3961,127 +3999,6 @@ unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm) return nr_mmu_pages; } -static void *pv_mmu_peek_buffer(struct kvm_pv_mmu_op_buffer *buffer, - unsigned len) -{ - if (len > buffer->len) - return NULL; - return buffer->ptr; -} - -static void *pv_mmu_read_buffer(struct kvm_pv_mmu_op_buffer *buffer, - unsigned len) -{ - void *ret; - - ret = pv_mmu_peek_buffer(buffer, len); - if (!ret) - return ret; - buffer->ptr += len; - buffer->len -= len; - buffer->processed += len; - return ret; -} - -static int kvm_pv_mmu_write(struct kvm_vcpu *vcpu, - gpa_t addr, gpa_t value) -{ - int bytes = 8; - int r; - - if (!is_long_mode(vcpu) && !is_pae(vcpu)) - bytes = 4; - - r = mmu_topup_memory_caches(vcpu); - if (r) - return r; - - if (!emulator_write_phys(vcpu, addr, &value, bytes)) - return -EFAULT; - - return 1; -} - -static int kvm_pv_mmu_flush_tlb(struct kvm_vcpu *vcpu) -{ - (void)kvm_set_cr3(vcpu, kvm_read_cr3(vcpu)); - return 1; -} - -static int kvm_pv_mmu_release_pt(struct kvm_vcpu *vcpu, gpa_t addr) -{ - spin_lock(&vcpu->kvm->mmu_lock); - mmu_unshadow(vcpu->kvm, addr >> PAGE_SHIFT); - spin_unlock(&vcpu->kvm->mmu_lock); - return 1; -} - -static int kvm_pv_mmu_op_one(struct kvm_vcpu *vcpu, - struct kvm_pv_mmu_op_buffer *buffer) -{ - struct kvm_mmu_op_header *header; - - header = pv_mmu_peek_buffer(buffer, sizeof *header); - if (!header) - return 0; - switch (header->op) { - case KVM_MMU_OP_WRITE_PTE: { - struct kvm_mmu_op_write_pte *wpte; - - wpte = pv_mmu_read_buffer(buffer, sizeof *wpte); - if (!wpte) - return 0; - return kvm_pv_mmu_write(vcpu, wpte->pte_phys, - wpte->pte_val); - } - case KVM_MMU_OP_FLUSH_TLB: { - struct kvm_mmu_op_flush_tlb *ftlb; - - ftlb = pv_mmu_read_buffer(buffer, sizeof *ftlb); - if (!ftlb) - return 0; - return kvm_pv_mmu_flush_tlb(vcpu); - } - case KVM_MMU_OP_RELEASE_PT: { - struct kvm_mmu_op_release_pt *rpt; - - rpt = pv_mmu_read_buffer(buffer, sizeof *rpt); - if (!rpt) - return 0; - return kvm_pv_mmu_release_pt(vcpu, rpt->pt_phys); - } - default: return 0; - } -} - -int kvm_pv_mmu_op(struct kvm_vcpu *vcpu, unsigned long bytes, - gpa_t addr, unsigned long *ret) -{ - int r; - struct kvm_pv_mmu_op_buffer *buffer = &vcpu->arch.mmu_op_buffer; - - buffer->ptr = buffer->buf; - buffer->len = min_t(unsigned long, bytes, sizeof buffer->buf); - buffer->processed = 0; - - r = kvm_read_guest(vcpu->kvm, addr, buffer->buf, buffer->len); - if (r) - goto out; - - while (buffer->len) { - r = kvm_pv_mmu_op_one(vcpu, buffer); - if (r < 0) - goto out; - if (r == 0) - break; - } - - r = 1; -out: - *ret = buffer->processed; - return r; -} - int kvm_mmu_get_spte_hierarchy(struct kvm_vcpu *vcpu, u64 addr, u64 sptes[4]) { struct kvm_shadow_walk_iterator iterator; @@ -4110,12 +4027,6 @@ void kvm_mmu_destroy(struct kvm_vcpu *vcpu) mmu_free_memory_caches(vcpu); } -#ifdef CONFIG_KVM_MMU_AUDIT -#include "mmu_audit.c" -#else -static void mmu_audit_disable(void) { } -#endif - void kvm_mmu_module_exit(void) { mmu_destroy_caches(); diff --git a/arch/x86/kvm/mmu_audit.c b/arch/x86/kvm/mmu_audit.c index 746ec25..fe15dcc 100644 --- a/arch/x86/kvm/mmu_audit.c +++ b/arch/x86/kvm/mmu_audit.c @@ -19,6 +19,15 @@ #include <linux/ratelimit.h> +char const *audit_point_name[] = { + "pre page fault", + "post page fault", + "pre pte write", + "post pte write", + "pre sync", + "post sync" +}; + #define audit_printk(kvm, fmt, args...) \ printk(KERN_ERR "audit: (%s) error: " \ fmt, audit_point_name[kvm->arch.audit_point], ##args) @@ -224,7 +233,10 @@ static void audit_vcpu_spte(struct kvm_vcpu *vcpu) mmu_spte_walk(vcpu, audit_spte); } -static void kvm_mmu_audit(void *ignore, struct kvm_vcpu *vcpu, int point) +static bool mmu_audit; +static struct jump_label_key mmu_audit_key; + +static void __kvm_mmu_audit(struct kvm_vcpu *vcpu, int point) { static DEFINE_RATELIMIT_STATE(ratelimit_state, 5 * HZ, 10); @@ -236,18 +248,18 @@ static void kvm_mmu_audit(void *ignore, struct kvm_vcpu *vcpu, int point) audit_vcpu_spte(vcpu); } -static bool mmu_audit; +static inline void kvm_mmu_audit(struct kvm_vcpu *vcpu, int point) +{ + if (static_branch((&mmu_audit_key))) + __kvm_mmu_audit(vcpu, point); +} static void mmu_audit_enable(void) { - int ret; - if (mmu_audit) return; - ret = register_trace_kvm_mmu_audit(kvm_mmu_audit, NULL); - WARN_ON(ret); - + jump_label_inc(&mmu_audit_key); mmu_audit = true; } @@ -256,8 +268,7 @@ static void mmu_audit_disable(void) if (!mmu_audit) return; - unregister_trace_kvm_mmu_audit(kvm_mmu_audit, NULL); - tracepoint_synchronize_unregister(); + jump_label_dec(&mmu_audit_key); mmu_audit = false; } diff --git a/arch/x86/kvm/mmutrace.h b/arch/x86/kvm/mmutrace.h index eed67f3..89fb0e8 100644 --- a/arch/x86/kvm/mmutrace.h +++ b/arch/x86/kvm/mmutrace.h @@ -243,25 +243,6 @@ TRACE_EVENT( TP_printk("addr:%llx gfn %llx access %x", __entry->addr, __entry->gfn, __entry->access) ); - -TRACE_EVENT( - kvm_mmu_audit, - TP_PROTO(struct kvm_vcpu *vcpu, int audit_point), - TP_ARGS(vcpu, audit_point), - - TP_STRUCT__entry( - __field(struct kvm_vcpu *, vcpu) - __field(int, audit_point) - ), - - TP_fast_assign( - __entry->vcpu = vcpu; - __entry->audit_point = audit_point; - ), - - TP_printk("vcpu:%d %s", __entry->vcpu->cpu, - audit_point_name[__entry->audit_point]) -); #endif /* _TRACE_KVMMMU_H */ #undef TRACE_INCLUDE_PATH diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index 9299410..1561028 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -497,6 +497,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr, shadow_walk_next(&it)) { gfn_t table_gfn; + clear_sp_write_flooding_count(it.sptep); drop_large_spte(vcpu, it.sptep); sp = NULL; @@ -522,6 +523,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr, shadow_walk_next(&it)) { gfn_t direct_gfn; + clear_sp_write_flooding_count(it.sptep); validate_direct_spte(vcpu, it.sptep, direct_access); drop_large_spte(vcpu, it.sptep); @@ -536,6 +538,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr, link_shadow_page(it.sptep, sp); } + clear_sp_write_flooding_count(it.sptep); mmu_set_spte(vcpu, it.sptep, access, gw->pte_access, user_fault, write_fault, emulate, it.level, gw->gfn, pfn, prefault, map_writable); @@ -599,11 +602,9 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code, */ if (!r) { pgprintk("%s: guest page fault\n", __func__); - if (!prefault) { + if (!prefault) inject_page_fault(vcpu, &walker.fault); - /* reset fork detector */ - vcpu->arch.last_pt_write_count = 0; - } + return 0; } @@ -631,7 +632,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code, if (mmu_notifier_retry(vcpu, mmu_seq)) goto out_unlock; - trace_kvm_mmu_audit(vcpu, AUDIT_PRE_PAGE_FAULT); + kvm_mmu_audit(vcpu, AUDIT_PRE_PAGE_FAULT); kvm_mmu_free_some_pages(vcpu); if (!force_pt_level) transparent_hugepage_adjust(vcpu, &walker.gfn, &pfn, &level); @@ -641,11 +642,8 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code, pgprintk("%s: shadow pte %p %llx emulate %d\n", __func__, sptep, *sptep, emulate); - if (!emulate) - vcpu->arch.last_pt_write_count = 0; /* reset fork detector */ - ++vcpu->stat.pf_fixed; - trace_kvm_mmu_audit(vcpu, AUDIT_POST_PAGE_FAULT); + kvm_mmu_audit(vcpu, AUDIT_POST_PAGE_FAULT); spin_unlock(&vcpu->kvm->mmu_lock); return emulate; @@ -656,65 +654,66 @@ out_unlock: return 0; } +static gpa_t FNAME(get_level1_sp_gpa)(struct kvm_mmu_page *sp) +{ + int offset = 0; + + WARN_ON(sp->role.level != 1); + + if (PTTYPE == 32) + offset = sp->role.quadrant << PT64_LEVEL_BITS; + + return gfn_to_gpa(sp->gfn) + offset * sizeof(pt_element_t); +} + static void FNAME(invlpg)(struct kvm_vcpu *vcpu, gva_t gva) { struct kvm_shadow_walk_iterator iterator; struct kvm_mmu_page *sp; - gpa_t pte_gpa = -1; int level; u64 *sptep; - int need_flush = 0; vcpu_clear_mmio_info(vcpu, gva); - spin_lock(&vcpu->kvm->mmu_lock); + /* + * No need to check return value here, rmap_can_add() can + * help us to skip pte prefetch later. + */ + mmu_topup_memory_caches(vcpu); + spin_lock(&vcpu->kvm->mmu_lock); for_each_shadow_entry(vcpu, gva, iterator) { level = iterator.level; sptep = iterator.sptep; sp = page_header(__pa(sptep)); if (is_last_spte(*sptep, level)) { - int offset, shift; + pt_element_t gpte; + gpa_t pte_gpa; if (!sp->unsync) break; - shift = PAGE_SHIFT - - (PT_LEVEL_BITS - PT64_LEVEL_BITS) * level; - offset = sp->role.quadrant << shift; - - pte_gpa = (sp->gfn << PAGE_SHIFT) + offset; + pte_gpa = FNAME(get_level1_sp_gpa)(sp); pte_gpa += (sptep - sp->spt) * sizeof(pt_element_t); - if (is_shadow_present_pte(*sptep)) { - if (is_large_pte(*sptep)) - --vcpu->kvm->stat.lpages; - drop_spte(vcpu->kvm, sptep); - need_flush = 1; - } else if (is_mmio_spte(*sptep)) - mmu_spte_clear_no_track(sptep); + if (mmu_page_zap_pte(vcpu->kvm, sp, sptep)) + kvm_flush_remote_tlbs(vcpu->kvm); - break; + if (!rmap_can_add(vcpu)) + break; + + if (kvm_read_guest_atomic(vcpu->kvm, pte_gpa, &gpte, + sizeof(pt_element_t))) + break; + + FNAME(update_pte)(vcpu, sp, sptep, &gpte); } if (!is_shadow_present_pte(*sptep) || !sp->unsync_children) break; } - - if (need_flush) - kvm_flush_remote_tlbs(vcpu->kvm); - - atomic_inc(&vcpu->kvm->arch.invlpg_counter); - spin_unlock(&vcpu->kvm->mmu_lock); - - if (pte_gpa == -1) - return; - - if (mmu_topup_memory_caches(vcpu)) - return; - kvm_mmu_pte_write(vcpu, pte_gpa, NULL, sizeof(pt_element_t), 0); } static gpa_t FNAME(gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t vaddr, u32 access, @@ -769,19 +768,14 @@ static gpa_t FNAME(gva_to_gpa_nested)(struct kvm_vcpu *vcpu, gva_t vaddr, */ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) { - int i, offset, nr_present; + int i, nr_present = 0; bool host_writable; gpa_t first_pte_gpa; - offset = nr_present = 0; - /* direct kvm_mmu_page can not be unsync. */ BUG_ON(sp->role.direct); - if (PTTYPE == 32) - offset = sp->role.quadrant << PT64_LEVEL_BITS; - - first_pte_gpa = gfn_to_gpa(sp->gfn) + offset * sizeof(pt_element_t); + first_pte_gpa = FNAME(get_level1_sp_gpa)(sp); for (i = 0; i < PT64_ENT_PER_PAGE; i++) { unsigned pte_access; diff --git a/arch/x86/kvm/pmu.c b/arch/x86/kvm/pmu.c new file mode 100644 index 0000000..7aad544 --- /dev/null +++ b/arch/x86/kvm/pmu.c @@ -0,0 +1,533 @@ +/* + * Kernel-based Virtual Machine -- Performane Monitoring Unit support + * + * Copyright 2011 Red Hat, Inc. and/or its affiliates. + * + * Authors: + * Avi Kivity <avi@redhat.com> + * Gleb Natapov <gleb@redhat.com> + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + */ + +#include <linux/types.h> +#include <linux/kvm_host.h> +#include <linux/perf_event.h> +#include "x86.h" +#include "cpuid.h" +#include "lapic.h" + +static struct kvm_arch_event_perf_mapping { + u8 eventsel; + u8 unit_mask; + unsigned event_type; + bool inexact; +} arch_events[] = { + /* Index must match CPUID 0x0A.EBX bit vector */ + [0] = { 0x3c, 0x00, PERF_COUNT_HW_CPU_CYCLES }, + [1] = { 0xc0, 0x00, PERF_COUNT_HW_INSTRUCTIONS }, + [2] = { 0x3c, 0x01, PERF_COUNT_HW_BUS_CYCLES }, + [3] = { 0x2e, 0x4f, PERF_COUNT_HW_CACHE_REFERENCES }, + [4] = { 0x2e, 0x41, PERF_COUNT_HW_CACHE_MISSES }, + [5] = { 0xc4, 0x00, PERF_COUNT_HW_BRANCH_INSTRUCTIONS }, + [6] = { 0xc5, 0x00, PERF_COUNT_HW_BRANCH_MISSES }, +}; + +/* mapping between fixed pmc index and arch_events array */ +int fixed_pmc_events[] = {1, 0, 2}; + +static bool pmc_is_gp(struct kvm_pmc *pmc) +{ + return pmc->type == KVM_PMC_GP; +} + +static inline u64 pmc_bitmask(struct kvm_pmc *pmc) +{ + struct kvm_pmu *pmu = &pmc->vcpu->arch.pmu; + + return pmu->counter_bitmask[pmc->type]; +} + +static inline bool pmc_enabled(struct kvm_pmc *pmc) +{ + struct kvm_pmu *pmu = &pmc->vcpu->arch.pmu; + return test_bit(pmc->idx, (unsigned long *)&pmu->global_ctrl); +} + +static inline struct kvm_pmc *get_gp_pmc(struct kvm_pmu *pmu, u32 msr, + u32 base) +{ + if (msr >= base && msr < base + pmu->nr_arch_gp_counters) + return &pmu->gp_counters[msr - base]; + return NULL; +} + +static inline struct kvm_pmc *get_fixed_pmc(struct kvm_pmu *pmu, u32 msr) +{ + int base = MSR_CORE_PERF_FIXED_CTR0; + if (msr >= base && msr < base + pmu->nr_arch_fixed_counters) + return &pmu->fixed_counters[msr - base]; + return NULL; +} + +static inline struct kvm_pmc *get_fixed_pmc_idx(struct kvm_pmu *pmu, int idx) +{ + return get_fixed_pmc(pmu, MSR_CORE_PERF_FIXED_CTR0 + idx); +} + +static struct kvm_pmc *global_idx_to_pmc(struct kvm_pmu *pmu, int idx) +{ + if (idx < X86_PMC_IDX_FIXED) + return get_gp_pmc(pmu, MSR_P6_EVNTSEL0 + idx, MSR_P6_EVNTSEL0); + else + return get_fixed_pmc_idx(pmu, idx - X86_PMC_IDX_FIXED); +} + +void kvm_deliver_pmi(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.apic) + kvm_apic_local_deliver(vcpu->arch.apic, APIC_LVTPC); +} + +static void trigger_pmi(struct irq_work *irq_work) +{ + struct kvm_pmu *pmu = container_of(irq_work, struct kvm_pmu, + irq_work); + struct kvm_vcpu *vcpu = container_of(pmu, struct kvm_vcpu, + arch.pmu); + + kvm_deliver_pmi(vcpu); +} + +static void kvm_perf_overflow(struct perf_event *perf_event, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + struct kvm_pmc *pmc = perf_event->overflow_handler_context; + struct kvm_pmu *pmu = &pmc->vcpu->arch.pmu; + __set_bit(pmc->idx, (unsigned long *)&pmu->global_status); +} + +static void kvm_perf_overflow_intr(struct perf_event *perf_event, + struct perf_sample_data *data, struct pt_regs *regs) +{ + struct kvm_pmc *pmc = perf_event->overflow_handler_context; + struct kvm_pmu *pmu = &pmc->vcpu->arch.pmu; + if (!test_and_set_bit(pmc->idx, (unsigned long *)&pmu->reprogram_pmi)) { + kvm_perf_overflow(perf_event, data, regs); + kvm_make_request(KVM_REQ_PMU, pmc->vcpu); + /* + * Inject PMI. If vcpu was in a guest mode during NMI PMI + * can be ejected on a guest mode re-entry. Otherwise we can't + * be sure that vcpu wasn't executing hlt instruction at the + * time of vmexit and is not going to re-enter guest mode until, + * woken up. So we should wake it, but this is impossible from + * NMI context. Do it from irq work instead. + */ + if (!kvm_is_in_guest()) + irq_work_queue(&pmc->vcpu->arch.pmu.irq_work); + else + kvm_make_request(KVM_REQ_PMI, pmc->vcpu); + } +} + +static u64 read_pmc(struct kvm_pmc *pmc) +{ + u64 counter, enabled, running; + + counter = pmc->counter; + + if (pmc->perf_event) + counter += perf_event_read_value(pmc->perf_event, + &enabled, &running); + + /* FIXME: Scaling needed? */ + + return counter & pmc_bitmask(pmc); +} + +static void stop_counter(struct kvm_pmc *pmc) +{ + if (pmc->perf_event) { + pmc->counter = read_pmc(pmc); + perf_event_release_kernel(pmc->perf_event); + pmc->perf_event = NULL; + } +} + +static void reprogram_counter(struct kvm_pmc *pmc, u32 type, + unsigned config, bool exclude_user, bool exclude_kernel, + bool intr) +{ + struct perf_event *event; + struct perf_event_attr attr = { + .type = type, + .size = sizeof(attr), + .pinned = true, + .exclude_idle = true, + .exclude_host = 1, + .exclude_user = exclude_user, + .exclude_kernel = exclude_kernel, + .config = config, + }; + + attr.sample_period = (-pmc->counter) & pmc_bitmask(pmc); + + event = perf_event_create_kernel_counter(&attr, -1, current, + intr ? kvm_perf_overflow_intr : + kvm_perf_overflow, pmc); + if (IS_ERR(event)) { + printk_once("kvm: pmu event creation failed %ld\n", + PTR_ERR(event)); + return; + } + + pmc->perf_event = event; + clear_bit(pmc->idx, (unsigned long*)&pmc->vcpu->arch.pmu.reprogram_pmi); +} + +static unsigned find_arch_event(struct kvm_pmu *pmu, u8 event_select, + u8 unit_mask) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(arch_events); i++) + if (arch_events[i].eventsel == event_select + && arch_events[i].unit_mask == unit_mask + && (pmu->available_event_types & (1 << i))) + break; + + if (i == ARRAY_SIZE(arch_events)) + return PERF_COUNT_HW_MAX; + + return arch_events[i].event_type; +} + +static void reprogram_gp_counter(struct kvm_pmc *pmc, u64 eventsel) +{ + unsigned config, type = PERF_TYPE_RAW; + u8 event_select, unit_mask; + + pmc->eventsel = eventsel; + + stop_counter(pmc); + + if (!(eventsel & ARCH_PERFMON_EVENTSEL_ENABLE) || !pmc_enabled(pmc)) + return; + + event_select = eventsel & ARCH_PERFMON_EVENTSEL_EVENT; + unit_mask = (eventsel & ARCH_PERFMON_EVENTSEL_UMASK) >> 8; + + if (!(event_select & (ARCH_PERFMON_EVENTSEL_EDGE | + ARCH_PERFMON_EVENTSEL_INV | + ARCH_PERFMON_EVENTSEL_CMASK))) { + config = find_arch_event(&pmc->vcpu->arch.pmu, event_select, + unit_mask); + if (config != PERF_COUNT_HW_MAX) + type = PERF_TYPE_HARDWARE; + } + + if (type == PERF_TYPE_RAW) + config = eventsel & X86_RAW_EVENT_MASK; + + reprogram_counter(pmc, type, config, + !(eventsel & ARCH_PERFMON_EVENTSEL_USR), + !(eventsel & ARCH_PERFMON_EVENTSEL_OS), + eventsel & ARCH_PERFMON_EVENTSEL_INT); +} + +static void reprogram_fixed_counter(struct kvm_pmc *pmc, u8 en_pmi, int idx) +{ + unsigned en = en_pmi & 0x3; + bool pmi = en_pmi & 0x8; + + stop_counter(pmc); + + if (!en || !pmc_enabled(pmc)) + return; + + reprogram_counter(pmc, PERF_TYPE_HARDWARE, + arch_events[fixed_pmc_events[idx]].event_type, + !(en & 0x2), /* exclude user */ + !(en & 0x1), /* exclude kernel */ + pmi); +} + +static inline u8 fixed_en_pmi(u64 ctrl, int idx) +{ + return (ctrl >> (idx * 4)) & 0xf; +} + +static void reprogram_fixed_counters(struct kvm_pmu *pmu, u64 data) +{ + int i; + + for (i = 0; i < pmu->nr_arch_fixed_counters; i++) { + u8 en_pmi = fixed_en_pmi(data, i); + struct kvm_pmc *pmc = get_fixed_pmc_idx(pmu, i); + + if (fixed_en_pmi(pmu->fixed_ctr_ctrl, i) == en_pmi) + continue; + + reprogram_fixed_counter(pmc, en_pmi, i); + } + + pmu->fixed_ctr_ctrl = data; +} + +static void reprogram_idx(struct kvm_pmu *pmu, int idx) +{ + struct kvm_pmc *pmc = global_idx_to_pmc(pmu, idx); + + if (!pmc) + return; + + if (pmc_is_gp(pmc)) + reprogram_gp_counter(pmc, pmc->eventsel); + else { + int fidx = idx - X86_PMC_IDX_FIXED; + reprogram_fixed_counter(pmc, + fixed_en_pmi(pmu->fixed_ctr_ctrl, fidx), fidx); + } +} + +static void global_ctrl_changed(struct kvm_pmu *pmu, u64 data) +{ + int bit; + u64 diff = pmu->global_ctrl ^ data; + + pmu->global_ctrl = data; + + for_each_set_bit(bit, (unsigned long *)&diff, X86_PMC_IDX_MAX) + reprogram_idx(pmu, bit); +} + +bool kvm_pmu_msr(struct kvm_vcpu *vcpu, u32 msr) +{ + struct kvm_pmu *pmu = &vcpu->arch.pmu; + int ret; + + switch (msr) { + case MSR_CORE_PERF_FIXED_CTR_CTRL: + case MSR_CORE_PERF_GLOBAL_STATUS: + case MSR_CORE_PERF_GLOBAL_CTRL: + case MSR_CORE_PERF_GLOBAL_OVF_CTRL: + ret = pmu->version > 1; + break; + default: + ret = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0) + || get_gp_pmc(pmu, msr, MSR_P6_EVNTSEL0) + || get_fixed_pmc(pmu, msr); + break; + } + return ret; +} + +int kvm_pmu_get_msr(struct kvm_vcpu *vcpu, u32 index, u64 *data) +{ + struct kvm_pmu *pmu = &vcpu->arch.pmu; + struct kvm_pmc *pmc; + + switch (index) { + case MSR_CORE_PERF_FIXED_CTR_CTRL: + *data = pmu->fixed_ctr_ctrl; + return 0; + case MSR_CORE_PERF_GLOBAL_STATUS: + *data = pmu->global_status; + return 0; + case MSR_CORE_PERF_GLOBAL_CTRL: + *data = pmu->global_ctrl; + return 0; + case MSR_CORE_PERF_GLOBAL_OVF_CTRL: + *data = pmu->global_ovf_ctrl; + return 0; + default: + if ((pmc = get_gp_pmc(pmu, index, MSR_IA32_PERFCTR0)) || + (pmc = get_fixed_pmc(pmu, index))) { + *data = read_pmc(pmc); + return 0; + } else if ((pmc = get_gp_pmc(pmu, index, MSR_P6_EVNTSEL0))) { + *data = pmc->eventsel; + return 0; + } + } + return 1; +} + +int kvm_pmu_set_msr(struct kvm_vcpu *vcpu, u32 index, u64 data) +{ + struct kvm_pmu *pmu = &vcpu->arch.pmu; + struct kvm_pmc *pmc; + + switch (index) { + case MSR_CORE_PERF_FIXED_CTR_CTRL: + if (pmu->fixed_ctr_ctrl == data) + return 0; + if (!(data & 0xfffffffffffff444)) { + reprogram_fixed_counters(pmu, data); + return 0; + } + break; + case MSR_CORE_PERF_GLOBAL_STATUS: + break; /* RO MSR */ + case MSR_CORE_PERF_GLOBAL_CTRL: + if (pmu->global_ctrl == data) + return 0; + if (!(data & pmu->global_ctrl_mask)) { + global_ctrl_changed(pmu, data); + return 0; + } + break; + case MSR_CORE_PERF_GLOBAL_OVF_CTRL: + if (!(data & (pmu->global_ctrl_mask & ~(3ull<<62)))) { + pmu->global_status &= ~data; + pmu->global_ovf_ctrl = data; + return 0; + } + break; + default: + if ((pmc = get_gp_pmc(pmu, index, MSR_IA32_PERFCTR0)) || + (pmc = get_fixed_pmc(pmu, index))) { + data = (s64)(s32)data; + pmc->counter += data - read_pmc(pmc); + return 0; + } else if ((pmc = get_gp_pmc(pmu, index, MSR_P6_EVNTSEL0))) { + if (data == pmc->eventsel) + return 0; + if (!(data & 0xffffffff00200000ull)) { + reprogram_gp_counter(pmc, data); + return 0; + } + } + } + return 1; +} + +int kvm_pmu_read_pmc(struct kvm_vcpu *vcpu, unsigned pmc, u64 *data) +{ + struct kvm_pmu *pmu = &vcpu->arch.pmu; + bool fast_mode = pmc & (1u << 31); + bool fixed = pmc & (1u << 30); + struct kvm_pmc *counters; + u64 ctr; + + pmc &= (3u << 30) - 1; + if (!fixed && pmc >= pmu->nr_arch_gp_counters) + return 1; + if (fixed && pmc >= pmu->nr_arch_fixed_counters) + return 1; + counters = fixed ? pmu->fixed_counters : pmu->gp_counters; + ctr = read_pmc(&counters[pmc]); + if (fast_mode) + ctr = (u32)ctr; + *data = ctr; + + return 0; +} + +void kvm_pmu_cpuid_update(struct kvm_vcpu *vcpu) +{ + struct kvm_pmu *pmu = &vcpu->arch.pmu; + struct kvm_cpuid_entry2 *entry; + unsigned bitmap_len; + + pmu->nr_arch_gp_counters = 0; + pmu->nr_arch_fixed_counters = 0; + pmu->counter_bitmask[KVM_PMC_GP] = 0; + pmu->counter_bitmask[KVM_PMC_FIXED] = 0; + pmu->version = 0; + + entry = kvm_find_cpuid_entry(vcpu, 0xa, 0); + if (!entry) + return; + + pmu->version = entry->eax & 0xff; + if (!pmu->version) + return; + + pmu->nr_arch_gp_counters = min((int)(entry->eax >> 8) & 0xff, + X86_PMC_MAX_GENERIC); + pmu->counter_bitmask[KVM_PMC_GP] = + ((u64)1 << ((entry->eax >> 16) & 0xff)) - 1; + bitmap_len = (entry->eax >> 24) & 0xff; + pmu->available_event_types = ~entry->ebx & ((1ull << bitmap_len) - 1); + + if (pmu->version == 1) { + pmu->global_ctrl = (1 << pmu->nr_arch_gp_counters) - 1; + return; + } + + pmu->nr_arch_fixed_counters = min((int)(entry->edx & 0x1f), + X86_PMC_MAX_FIXED); + pmu->counter_bitmask[KVM_PMC_FIXED] = + ((u64)1 << ((entry->edx >> 5) & 0xff)) - 1; + pmu->global_ctrl_mask = ~(((1 << pmu->nr_arch_gp_counters) - 1) + | (((1ull << pmu->nr_arch_fixed_counters) - 1) + << X86_PMC_IDX_FIXED)); +} + +void kvm_pmu_init(struct kvm_vcpu *vcpu) +{ + int i; + struct kvm_pmu *pmu = &vcpu->arch.pmu; + + memset(pmu, 0, sizeof(*pmu)); + for (i = 0; i < X86_PMC_MAX_GENERIC; i++) { + pmu->gp_counters[i].type = KVM_PMC_GP; + pmu->gp_counters[i].vcpu = vcpu; + pmu->gp_counters[i].idx = i; + } + for (i = 0; i < X86_PMC_MAX_FIXED; i++) { + pmu->fixed_counters[i].type = KVM_PMC_FIXED; + pmu->fixed_counters[i].vcpu = vcpu; + pmu->fixed_counters[i].idx = i + X86_PMC_IDX_FIXED; + } + init_irq_work(&pmu->irq_work, trigger_pmi); + kvm_pmu_cpuid_update(vcpu); +} + +void kvm_pmu_reset(struct kvm_vcpu *vcpu) +{ + struct kvm_pmu *pmu = &vcpu->arch.pmu; + int i; + + irq_work_sync(&pmu->irq_work); + for (i = 0; i < X86_PMC_MAX_GENERIC; i++) { + struct kvm_pmc *pmc = &pmu->gp_counters[i]; + stop_counter(pmc); + pmc->counter = pmc->eventsel = 0; + } + + for (i = 0; i < X86_PMC_MAX_FIXED; i++) + stop_counter(&pmu->fixed_counters[i]); + + pmu->fixed_ctr_ctrl = pmu->global_ctrl = pmu->global_status = + pmu->global_ovf_ctrl = 0; +} + +void kvm_pmu_destroy(struct kvm_vcpu *vcpu) +{ + kvm_pmu_reset(vcpu); +} + +void kvm_handle_pmu_event(struct kvm_vcpu *vcpu) +{ + struct kvm_pmu *pmu = &vcpu->arch.pmu; + u64 bitmask; + int bit; + + bitmask = pmu->reprogram_pmi; + + for_each_set_bit(bit, (unsigned long *)&bitmask, X86_PMC_IDX_MAX) { + struct kvm_pmc *pmc = global_idx_to_pmc(pmu, bit); + + if (unlikely(!pmc || !pmc->perf_event)) { + clear_bit(bit, (unsigned long *)&pmu->reprogram_pmi); + continue; + } + + reprogram_idx(pmu, bit); + } +} diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index e32243e..5fa553b 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1014,6 +1014,7 @@ static void init_vmcb(struct vcpu_svm *svm) set_intercept(svm, INTERCEPT_NMI); set_intercept(svm, INTERCEPT_SMI); set_intercept(svm, INTERCEPT_SELECTIVE_CR0); + set_intercept(svm, INTERCEPT_RDPMC); set_intercept(svm, INTERCEPT_CPUID); set_intercept(svm, INTERCEPT_INVD); set_intercept(svm, INTERCEPT_HLT); @@ -2770,6 +2771,19 @@ static int emulate_on_interception(struct vcpu_svm *svm) return emulate_instruction(&svm->vcpu, 0) == EMULATE_DONE; } +static int rdpmc_interception(struct vcpu_svm *svm) +{ + int err; + + if (!static_cpu_has(X86_FEATURE_NRIPS)) + return emulate_on_interception(svm); + + err = kvm_rdpmc(&svm->vcpu); + kvm_complete_insn_gp(&svm->vcpu, err); + + return 1; +} + bool check_selective_cr0_intercepted(struct vcpu_svm *svm, unsigned long val) { unsigned long cr0 = svm->vcpu.arch.cr0; @@ -3190,6 +3204,7 @@ static int (*svm_exit_handlers[])(struct vcpu_svm *svm) = { [SVM_EXIT_SMI] = nop_on_interception, [SVM_EXIT_INIT] = nop_on_interception, [SVM_EXIT_VINTR] = interrupt_window_interception, + [SVM_EXIT_RDPMC] = rdpmc_interception, [SVM_EXIT_CPUID] = cpuid_interception, [SVM_EXIT_IRET] = iret_interception, [SVM_EXIT_INVD] = emulate_on_interception, diff --git a/arch/x86/kvm/timer.c b/arch/x86/kvm/timer.c index ae432ea..6b85cc6 100644 --- a/arch/x86/kvm/timer.c +++ b/arch/x86/kvm/timer.c @@ -18,9 +18,10 @@ #include <linux/atomic.h> #include "kvm_timer.h" -static int __kvm_timer_fn(struct kvm_vcpu *vcpu, struct kvm_timer *ktimer) +enum hrtimer_restart kvm_timer_fn(struct hrtimer *data) { - int restart_timer = 0; + struct kvm_timer *ktimer = container_of(data, struct kvm_timer, timer); + struct kvm_vcpu *vcpu = ktimer->vcpu; wait_queue_head_t *q = &vcpu->wq; /* @@ -40,26 +41,7 @@ static int __kvm_timer_fn(struct kvm_vcpu *vcpu, struct kvm_timer *ktimer) if (ktimer->t_ops->is_periodic(ktimer)) { hrtimer_add_expires_ns(&ktimer->timer, ktimer->period); - restart_timer = 1; - } - - return restart_timer; -} - -enum hrtimer_restart kvm_timer_fn(struct hrtimer *data) -{ - int restart_timer; - struct kvm_vcpu *vcpu; - struct kvm_timer *ktimer = container_of(data, struct kvm_timer, timer); - - vcpu = ktimer->vcpu; - if (!vcpu) - return HRTIMER_NORESTART; - - restart_timer = __kvm_timer_fn(vcpu, ktimer); - if (restart_timer) return HRTIMER_RESTART; - else + } else return HRTIMER_NORESTART; } - diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 579a0b5..906a7e8 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -18,6 +18,7 @@ #include "irq.h" #include "mmu.h" +#include "cpuid.h" #include <linux/kvm_host.h> #include <linux/module.h> @@ -1747,7 +1748,6 @@ static void setup_msrs(struct vcpu_vmx *vmx) int save_nmsrs, index; unsigned long *msr_bitmap; - vmx_load_host_state(vmx); save_nmsrs = 0; #ifdef CONFIG_X86_64 if (is_long_mode(&vmx->vcpu)) { @@ -1956,6 +1956,7 @@ static __init void nested_vmx_setup_ctls_msrs(void) #endif CPU_BASED_MOV_DR_EXITING | CPU_BASED_UNCOND_IO_EXITING | CPU_BASED_USE_IO_BITMAPS | CPU_BASED_MONITOR_EXITING | + CPU_BASED_RDPMC_EXITING | CPU_BASED_ACTIVATE_SECONDARY_CONTROLS; /* * We can allow some features even when not supported by the @@ -2142,12 +2143,10 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata) return 1; /* Otherwise falls through */ default: - vmx_load_host_state(to_vmx(vcpu)); if (vmx_get_vmx_msr(vcpu, msr_index, pdata)) return 0; msr = find_msr_entry(to_vmx(vcpu), msr_index); if (msr) { - vmx_load_host_state(to_vmx(vcpu)); data = msr->data; break; } @@ -2171,7 +2170,6 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data) switch (msr_index) { case MSR_EFER: - vmx_load_host_state(vmx); ret = kvm_set_msr_common(vcpu, msr_index, data); break; #ifdef CONFIG_X86_64 @@ -2220,7 +2218,6 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data) break; msr = find_msr_entry(vmx, msr_index); if (msr) { - vmx_load_host_state(vmx); msr->data = data; break; } @@ -2414,7 +2411,8 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf) CPU_BASED_USE_TSC_OFFSETING | CPU_BASED_MWAIT_EXITING | CPU_BASED_MONITOR_EXITING | - CPU_BASED_INVLPG_EXITING; + CPU_BASED_INVLPG_EXITING | + CPU_BASED_RDPMC_EXITING; if (yield_on_hlt) min |= CPU_BASED_HLT_EXITING; @@ -2716,11 +2714,13 @@ static gva_t rmode_tss_base(struct kvm *kvm) { if (!kvm->arch.tss_addr) { struct kvm_memslots *slots; + struct kvm_memory_slot *slot; gfn_t base_gfn; slots = kvm_memslots(kvm); - base_gfn = slots->memslots[0].base_gfn + - kvm->memslots->memslots[0].npages - 3; + slot = id_to_memslot(slots, 0); + base_gfn = slot->base_gfn + slot->npages - 3; + return base_gfn << PAGE_SHIFT; } return kvm->arch.tss_addr; @@ -3945,12 +3945,15 @@ static bool nested_exit_on_intr(struct kvm_vcpu *vcpu) static void enable_irq_window(struct kvm_vcpu *vcpu) { u32 cpu_based_vm_exec_control; - if (is_guest_mode(vcpu) && nested_exit_on_intr(vcpu)) - /* We can get here when nested_run_pending caused - * vmx_interrupt_allowed() to return false. In this case, do - * nothing - the interrupt will be injected later. + if (is_guest_mode(vcpu) && nested_exit_on_intr(vcpu)) { + /* + * We get here if vmx_interrupt_allowed() said we can't + * inject to L1 now because L2 must run. Ask L2 to exit + * right after entry, so we can inject to L1 more promptly. */ + kvm_make_request(KVM_REQ_IMMEDIATE_EXIT, vcpu); return; + } cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL); cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_INTR_PENDING; @@ -4077,11 +4080,12 @@ static void vmx_set_nmi_mask(struct kvm_vcpu *vcpu, bool masked) static int vmx_interrupt_allowed(struct kvm_vcpu *vcpu) { if (is_guest_mode(vcpu) && nested_exit_on_intr(vcpu)) { - struct vmcs12 *vmcs12; - if (to_vmx(vcpu)->nested.nested_run_pending) + struct vmcs12 *vmcs12 = get_vmcs12(vcpu); + if (to_vmx(vcpu)->nested.nested_run_pending || + (vmcs12->idt_vectoring_info_field & + VECTORING_INFO_VALID_MASK)) return 0; nested_vmx_vmexit(vcpu); - vmcs12 = get_vmcs12(vcpu); vmcs12->vm_exit_reason = EXIT_REASON_EXTERNAL_INTERRUPT; vmcs12->vm_exit_intr_info = 0; /* fall through to normal code, but now in L1, not L2 */ @@ -4611,6 +4615,16 @@ static int handle_invlpg(struct kvm_vcpu *vcpu) return 1; } +static int handle_rdpmc(struct kvm_vcpu *vcpu) +{ + int err; + + err = kvm_rdpmc(vcpu); + kvm_complete_insn_gp(vcpu, err); + + return 1; +} + static int handle_wbinvd(struct kvm_vcpu *vcpu) { skip_emulated_instruction(vcpu); @@ -5561,6 +5575,7 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu) = { [EXIT_REASON_HLT] = handle_halt, [EXIT_REASON_INVD] = handle_invd, [EXIT_REASON_INVLPG] = handle_invlpg, + [EXIT_REASON_RDPMC] = handle_rdpmc, [EXIT_REASON_VMCALL] = handle_vmcall, [EXIT_REASON_VMCLEAR] = handle_vmclear, [EXIT_REASON_VMLAUNCH] = handle_vmlaunch, diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 4c938da..1171def 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -26,6 +26,7 @@ #include "tss.h" #include "kvm_cache_regs.h" #include "x86.h" +#include "cpuid.h" #include <linux/clocksource.h> #include <linux/interrupt.h> @@ -82,8 +83,6 @@ static u64 __read_mostly efer_reserved_bits = ~((u64)EFER_SCE); #define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU static void update_cr8_intercept(struct kvm_vcpu *vcpu); -static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, - struct kvm_cpuid_entry2 __user *entries); static void process_nmi(struct kvm_vcpu *vcpu); struct kvm_x86_ops *kvm_x86_ops; @@ -574,54 +573,6 @@ int kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr) } EXPORT_SYMBOL_GPL(kvm_set_xcr); -static bool guest_cpuid_has_xsave(struct kvm_vcpu *vcpu) -{ - struct kvm_cpuid_entry2 *best; - - best = kvm_find_cpuid_entry(vcpu, 1, 0); - return best && (best->ecx & bit(X86_FEATURE_XSAVE)); -} - -static bool guest_cpuid_has_smep(struct kvm_vcpu *vcpu) -{ - struct kvm_cpuid_entry2 *best; - - best = kvm_find_cpuid_entry(vcpu, 7, 0); - return best && (best->ebx & bit(X86_FEATURE_SMEP)); -} - -static bool guest_cpuid_has_fsgsbase(struct kvm_vcpu *vcpu) -{ - struct kvm_cpuid_entry2 *best; - - best = kvm_find_cpuid_entry(vcpu, 7, 0); - return best && (best->ebx & bit(X86_FEATURE_FSGSBASE)); -} - -static void update_cpuid(struct kvm_vcpu *vcpu) -{ - struct kvm_cpuid_entry2 *best; - struct kvm_lapic *apic = vcpu->arch.apic; - - best = kvm_find_cpuid_entry(vcpu, 1, 0); - if (!best) - return; - - /* Update OSXSAVE bit */ - if (cpu_has_xsave && best->function == 0x1) { - best->ecx &= ~(bit(X86_FEATURE_OSXSAVE)); - if (kvm_read_cr4_bits(vcpu, X86_CR4_OSXSAVE)) - best->ecx |= bit(X86_FEATURE_OSXSAVE); - } - - if (apic) { - if (best->ecx & bit(X86_FEATURE_TSC_DEADLINE_TIMER)) - apic->lapic_timer.timer_mode_mask = 3 << 17; - else - apic->lapic_timer.timer_mode_mask = 1 << 17; - } -} - int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) { unsigned long old_cr4 = kvm_read_cr4(vcpu); @@ -655,7 +606,7 @@ int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) kvm_mmu_reset_context(vcpu); if ((cr4 ^ old_cr4) & X86_CR4_OSXSAVE) - update_cpuid(vcpu); + kvm_update_cpuid(vcpu); return 0; } @@ -809,6 +760,21 @@ int kvm_get_dr(struct kvm_vcpu *vcpu, int dr, unsigned long *val) } EXPORT_SYMBOL_GPL(kvm_get_dr); +bool kvm_rdpmc(struct kvm_vcpu *vcpu) +{ + u32 ecx = kvm_register_read(vcpu, VCPU_REGS_RCX); + u64 data; + int err; + + err = kvm_pmu_read_pmc(vcpu, ecx, &data); + if (err) + return err; + kvm_register_write(vcpu, VCPU_REGS_RAX, (u32)data); + kvm_register_write(vcpu, VCPU_REGS_RDX, data >> 32); + return err; +} +EXPORT_SYMBOL_GPL(kvm_rdpmc); + /* * List of msr numbers which we expose to userspace through KVM_GET_MSRS * and KVM_SET_MSRS, and KVM_GET_MSR_INDEX_LIST. @@ -1358,12 +1324,11 @@ static int xen_hvm_config(struct kvm_vcpu *vcpu, u64 data) if (page_num >= blob_size) goto out; r = -ENOMEM; - page = kzalloc(PAGE_SIZE, GFP_KERNEL); - if (!page) + page = memdup_user(blob_addr + (page_num * PAGE_SIZE), PAGE_SIZE); + if (IS_ERR(page)) { + r = PTR_ERR(page); goto out; - r = -EFAULT; - if (copy_from_user(page, blob_addr + (page_num * PAGE_SIZE), PAGE_SIZE)) - goto out_free; + } if (kvm_write_guest(kvm, page_addr, page, PAGE_SIZE)) goto out_free; r = 0; @@ -1652,8 +1617,6 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) * which we perfectly emulate ;-). Any other value should be at least * reported, some guests depend on them. */ - case MSR_P6_EVNTSEL0: - case MSR_P6_EVNTSEL1: case MSR_K7_EVNTSEL0: case MSR_K7_EVNTSEL1: case MSR_K7_EVNTSEL2: @@ -1665,8 +1628,6 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) /* at least RHEL 4 unconditionally writes to the perfctr registers, * so we ignore writes to make it happy. */ - case MSR_P6_PERFCTR0: - case MSR_P6_PERFCTR1: case MSR_K7_PERFCTR0: case MSR_K7_PERFCTR1: case MSR_K7_PERFCTR2: @@ -1703,6 +1664,8 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) default: if (msr && (msr == vcpu->kvm->arch.xen_hvm_config.msr)) return xen_hvm_config(vcpu, data); + if (kvm_pmu_msr(vcpu, msr)) + return kvm_pmu_set_msr(vcpu, msr, data); if (!ignore_msrs) { pr_unimpl(vcpu, "unhandled wrmsr: 0x%x data %llx\n", msr, data); @@ -1865,10 +1828,6 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) case MSR_K8_SYSCFG: case MSR_K7_HWCR: case MSR_VM_HSAVE_PA: - case MSR_P6_PERFCTR0: - case MSR_P6_PERFCTR1: - case MSR_P6_EVNTSEL0: - case MSR_P6_EVNTSEL1: case MSR_K7_EVNTSEL0: case MSR_K7_PERFCTR0: case MSR_K8_INT_PENDING_MSG: @@ -1979,6 +1938,8 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) data = 0xbe702111; break; default: + if (kvm_pmu_msr(vcpu, msr)) + return kvm_pmu_get_msr(vcpu, msr, pdata); if (!ignore_msrs) { pr_unimpl(vcpu, "unhandled rdmsr: 0x%x\n", msr); return 1; @@ -2037,15 +1998,12 @@ static int msr_io(struct kvm_vcpu *vcpu, struct kvm_msrs __user *user_msrs, if (msrs.nmsrs >= MAX_IO_MSRS) goto out; - r = -ENOMEM; size = sizeof(struct kvm_msr_entry) * msrs.nmsrs; - entries = kmalloc(size, GFP_KERNEL); - if (!entries) + entries = memdup_user(user_msrs->entries, size); + if (IS_ERR(entries)) { + r = PTR_ERR(entries); goto out; - - r = -EFAULT; - if (copy_from_user(entries, user_msrs->entries, size)) - goto out_free; + } r = n = __msr_io(vcpu, &msrs, entries, do_msr); if (r < 0) @@ -2265,466 +2223,6 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) vcpu->arch.last_guest_tsc = kvm_x86_ops->read_l1_tsc(vcpu); } -static int is_efer_nx(void) -{ - unsigned long long efer = 0; - - rdmsrl_safe(MSR_EFER, &efer); - return efer & EFER_NX; -} - -static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu) -{ - int i; - struct kvm_cpuid_entry2 *e, *entry; - - entry = NULL; - for (i = 0; i < vcpu->arch.cpuid_nent; ++i) { - e = &vcpu->arch.cpuid_entries[i]; - if (e->function == 0x80000001) { - entry = e; - break; - } - } - if (entry && (entry->edx & (1 << 20)) && !is_efer_nx()) { - entry->edx &= ~(1 << 20); - printk(KERN_INFO "kvm: guest NX capability removed\n"); - } -} - -/* when an old userspace process fills a new kernel module */ -static int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu, - struct kvm_cpuid *cpuid, - struct kvm_cpuid_entry __user *entries) -{ - int r, i; - struct kvm_cpuid_entry *cpuid_entries; - - r = -E2BIG; - if (cpuid->nent > KVM_MAX_CPUID_ENTRIES) - goto out; - r = -ENOMEM; - cpuid_entries = vmalloc(sizeof(struct kvm_cpuid_entry) * cpuid->nent); - if (!cpuid_entries) - goto out; - r = -EFAULT; - if (copy_from_user(cpuid_entries, entries, - cpuid->nent * sizeof(struct kvm_cpuid_entry))) - goto out_free; - for (i = 0; i < cpuid->nent; i++) { - vcpu->arch.cpuid_entries[i].function = cpuid_entries[i].function; - vcpu->arch.cpuid_entries[i].eax = cpuid_entries[i].eax; - vcpu->arch.cpuid_entries[i].ebx = cpuid_entries[i].ebx; - vcpu->arch.cpuid_entries[i].ecx = cpuid_entries[i].ecx; - vcpu->arch.cpuid_entries[i].edx = cpuid_entries[i].edx; - vcpu->arch.cpuid_entries[i].index = 0; - vcpu->arch.cpuid_entries[i].flags = 0; - vcpu->arch.cpuid_entries[i].padding[0] = 0; - vcpu->arch.cpuid_entries[i].padding[1] = 0; - vcpu->arch.cpuid_entries[i].padding[2] = 0; - } - vcpu->arch.cpuid_nent = cpuid->nent; - cpuid_fix_nx_cap(vcpu); - r = 0; - kvm_apic_set_version(vcpu); - kvm_x86_ops->cpuid_update(vcpu); - update_cpuid(vcpu); - -out_free: - vfree(cpuid_entries); -out: - return r; -} - -static int kvm_vcpu_ioctl_set_cpuid2(struct kvm_vcpu *vcpu, - struct kvm_cpuid2 *cpuid, - struct kvm_cpuid_entry2 __user *entries) -{ - int r; - - r = -E2BIG; - if (cpuid->nent > KVM_MAX_CPUID_ENTRIES) - goto out; - r = -EFAULT; - if (copy_from_user(&vcpu->arch.cpuid_entries, entries, - cpuid->nent * sizeof(struct kvm_cpuid_entry2))) - goto out; - vcpu->arch.cpuid_nent = cpuid->nent; - kvm_apic_set_version(vcpu); - kvm_x86_ops->cpuid_update(vcpu); - update_cpuid(vcpu); - return 0; - -out: - return r; -} - -static int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu, - struct kvm_cpuid2 *cpuid, - struct kvm_cpuid_entry2 __user *entries) -{ - int r; - - r = -E2BIG; - if (cpuid->nent < vcpu->arch.cpuid_nent) - goto out; - r = -EFAULT; - if (copy_to_user(entries, &vcpu->arch.cpuid_entries, - vcpu->arch.cpuid_nent * sizeof(struct kvm_cpuid_entry2))) - goto out; - return 0; - -out: - cpuid->nent = vcpu->arch.cpuid_nent; - return r; -} - -static void cpuid_mask(u32 *word, int wordnum) -{ - *word &= boot_cpu_data.x86_capability[wordnum]; -} - -static void do_cpuid_1_ent(struct kvm_cpuid_entry2 *entry, u32 function, - u32 index) -{ - entry->function = function; - entry->index = index; - cpuid_count(entry->function, entry->index, - &entry->eax, &entry->ebx, &entry->ecx, &entry->edx); - entry->flags = 0; -} - -static bool supported_xcr0_bit(unsigned bit) -{ - u64 mask = ((u64)1 << bit); - - return mask & (XSTATE_FP | XSTATE_SSE | XSTATE_YMM) & host_xcr0; -} - -#define F(x) bit(X86_FEATURE_##x) - -static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, - u32 index, int *nent, int maxnent) -{ - unsigned f_nx = is_efer_nx() ? F(NX) : 0; -#ifdef CONFIG_X86_64 - unsigned f_gbpages = (kvm_x86_ops->get_lpage_level() == PT_PDPE_LEVEL) - ? F(GBPAGES) : 0; - unsigned f_lm = F(LM); -#else - unsigned f_gbpages = 0; - unsigned f_lm = 0; -#endif - unsigned f_rdtscp = kvm_x86_ops->rdtscp_supported() ? F(RDTSCP) : 0; - - /* cpuid 1.edx */ - const u32 kvm_supported_word0_x86_features = - F(FPU) | F(VME) | F(DE) | F(PSE) | - F(TSC) | F(MSR) | F(PAE) | F(MCE) | - F(CX8) | F(APIC) | 0 /* Reserved */ | F(SEP) | - F(MTRR) | F(PGE) | F(MCA) | F(CMOV) | - F(PAT) | F(PSE36) | 0 /* PSN */ | F(CLFLSH) | - 0 /* Reserved, DS, ACPI */ | F(MMX) | - F(FXSR) | F(XMM) | F(XMM2) | F(SELFSNOOP) | - 0 /* HTT, TM, Reserved, PBE */; - /* cpuid 0x80000001.edx */ - const u32 kvm_supported_word1_x86_features = - F(FPU) | F(VME) | F(DE) | F(PSE) | - F(TSC) | F(MSR) | F(PAE) | F(MCE) | - F(CX8) | F(APIC) | 0 /* Reserved */ | F(SYSCALL) | - F(MTRR) | F(PGE) | F(MCA) | F(CMOV) | - F(PAT) | F(PSE36) | 0 /* Reserved */ | - f_nx | 0 /* Reserved */ | F(MMXEXT) | F(MMX) | - F(FXSR) | F(FXSR_OPT) | f_gbpages | f_rdtscp | - 0 /* Reserved */ | f_lm | F(3DNOWEXT) | F(3DNOW); - /* cpuid 1.ecx */ - const u32 kvm_supported_word4_x86_features = - F(XMM3) | F(PCLMULQDQ) | 0 /* DTES64, MONITOR */ | - 0 /* DS-CPL, VMX, SMX, EST */ | - 0 /* TM2 */ | F(SSSE3) | 0 /* CNXT-ID */ | 0 /* Reserved */ | - 0 /* Reserved */ | F(CX16) | 0 /* xTPR Update, PDCM */ | - 0 /* Reserved, DCA */ | F(XMM4_1) | - F(XMM4_2) | F(X2APIC) | F(MOVBE) | F(POPCNT) | - 0 /* Reserved*/ | F(AES) | F(XSAVE) | 0 /* OSXSAVE */ | F(AVX) | - F(F16C) | F(RDRAND); - /* cpuid 0x80000001.ecx */ - const u32 kvm_supported_word6_x86_features = - F(LAHF_LM) | F(CMP_LEGACY) | 0 /*SVM*/ | 0 /* ExtApicSpace */ | - F(CR8_LEGACY) | F(ABM) | F(SSE4A) | F(MISALIGNSSE) | - F(3DNOWPREFETCH) | 0 /* OSVW */ | 0 /* IBS */ | F(XOP) | - 0 /* SKINIT, WDT, LWP */ | F(FMA4) | F(TBM); - - /* cpuid 0xC0000001.edx */ - const u32 kvm_supported_word5_x86_features = - F(XSTORE) | F(XSTORE_EN) | F(XCRYPT) | F(XCRYPT_EN) | - F(ACE2) | F(ACE2_EN) | F(PHE) | F(PHE_EN) | - F(PMM) | F(PMM_EN); - - /* cpuid 7.0.ebx */ - const u32 kvm_supported_word9_x86_features = - F(SMEP) | F(FSGSBASE) | F(ERMS); - - /* all calls to cpuid_count() should be made on the same cpu */ - get_cpu(); - do_cpuid_1_ent(entry, function, index); - ++*nent; - - switch (function) { - case 0: - entry->eax = min(entry->eax, (u32)0xd); - break; - case 1: - entry->edx &= kvm_supported_word0_x86_features; - cpuid_mask(&entry->edx, 0); - entry->ecx &= kvm_supported_word4_x86_features; - cpuid_mask(&entry->ecx, 4); - /* we support x2apic emulation even if host does not support - * it since we emulate x2apic in software */ - entry->ecx |= F(X2APIC); - break; - /* function 2 entries are STATEFUL. That is, repeated cpuid commands - * may return different values. This forces us to get_cpu() before - * issuing the first command, and also to emulate this annoying behavior - * in kvm_emulate_cpuid() using KVM_CPUID_FLAG_STATE_READ_NEXT */ - case 2: { - int t, times = entry->eax & 0xff; - - entry->flags |= KVM_CPUID_FLAG_STATEFUL_FUNC; - entry->flags |= KVM_CPUID_FLAG_STATE_READ_NEXT; - for (t = 1; t < times && *nent < maxnent; ++t) { - do_cpuid_1_ent(&entry[t], function, 0); - entry[t].flags |= KVM_CPUID_FLAG_STATEFUL_FUNC; - ++*nent; - } - break; - } - /* function 4 has additional index. */ - case 4: { - int i, cache_type; - - entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX; - /* read more entries until cache_type is zero */ - for (i = 1; *nent < maxnent; ++i) { - cache_type = entry[i - 1].eax & 0x1f; - if (!cache_type) - break; - do_cpuid_1_ent(&entry[i], function, i); - entry[i].flags |= - KVM_CPUID_FLAG_SIGNIFCANT_INDEX; - ++*nent; - } - break; - } - case 7: { - entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX; - /* Mask ebx against host capbability word 9 */ - if (index == 0) { - entry->ebx &= kvm_supported_word9_x86_features; - cpuid_mask(&entry->ebx, 9); - } else - entry->ebx = 0; - entry->eax = 0; - entry->ecx = 0; - entry->edx = 0; - break; - } - case 9: - break; - /* function 0xb has additional index. */ - case 0xb: { - int i, level_type; - - entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX; - /* read more entries until level_type is zero */ - for (i = 1; *nent < maxnent; ++i) { - level_type = entry[i - 1].ecx & 0xff00; - if (!level_type) - break; - do_cpuid_1_ent(&entry[i], function, i); - entry[i].flags |= - KVM_CPUID_FLAG_SIGNIFCANT_INDEX; - ++*nent; - } - break; - } - case 0xd: { - int idx, i; - - entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX; - for (idx = 1, i = 1; *nent < maxnent && idx < 64; ++idx) { - do_cpuid_1_ent(&entry[i], function, idx); - if (entry[i].eax == 0 || !supported_xcr0_bit(idx)) - continue; - entry[i].flags |= - KVM_CPUID_FLAG_SIGNIFCANT_INDEX; - ++*nent; - ++i; - } - break; - } - case KVM_CPUID_SIGNATURE: { - char signature[12] = "KVMKVMKVM\0\0"; - u32 *sigptr = (u32 *)signature; - entry->eax = 0; - entry->ebx = sigptr[0]; - entry->ecx = sigptr[1]; - entry->edx = sigptr[2]; - break; - } - case KVM_CPUID_FEATURES: - entry->eax = (1 << KVM_FEATURE_CLOCKSOURCE) | - (1 << KVM_FEATURE_NOP_IO_DELAY) | - (1 << KVM_FEATURE_CLOCKSOURCE2) | - (1 << KVM_FEATURE_ASYNC_PF) | - (1 << KVM_FEATURE_CLOCKSOURCE_STABLE_BIT); - - if (sched_info_on()) - entry->eax |= (1 << KVM_FEATURE_STEAL_TIME); - - entry->ebx = 0; - entry->ecx = 0; - entry->edx = 0; - break; - case 0x80000000: - entry->eax = min(entry->eax, 0x8000001a); - break; - case 0x80000001: - entry->edx &= kvm_supported_word1_x86_features; - cpuid_mask(&entry->edx, 1); - entry->ecx &= kvm_supported_word6_x86_features; - cpuid_mask(&entry->ecx, 6); - break; - case 0x80000008: { - unsigned g_phys_as = (entry->eax >> 16) & 0xff; - unsigned virt_as = max((entry->eax >> 8) & 0xff, 48U); - unsigned phys_as = entry->eax & 0xff; - - if (!g_phys_as) - g_phys_as = phys_as; - entry->eax = g_phys_as | (virt_as << 8); - entry->ebx = entry->edx = 0; - break; - } - case 0x80000019: - entry->ecx = entry->edx = 0; - break; - case 0x8000001a: - break; - case 0x8000001d: - break; - /*Add support for Centaur's CPUID instruction*/ - case 0xC0000000: - /*Just support up to 0xC0000004 now*/ - entry->eax = min(entry->eax, 0xC0000004); - break; - case 0xC0000001: - entry->edx &= kvm_supported_word5_x86_features; - cpuid_mask(&entry->edx, 5); - break; - case 3: /* Processor serial number */ - case 5: /* MONITOR/MWAIT */ - case 6: /* Thermal management */ - case 0xA: /* Architectural Performance Monitoring */ - case 0x80000007: /* Advanced power management */ - case 0xC0000002: - case 0xC0000003: - case 0xC0000004: - default: - entry->eax = entry->ebx = entry->ecx = entry->edx = 0; - break; - } - - kvm_x86_ops->set_supported_cpuid(function, entry); - - put_cpu(); -} - -#undef F - -static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, - struct kvm_cpuid_entry2 __user *entries) -{ - struct kvm_cpuid_entry2 *cpuid_entries; - int limit, nent = 0, r = -E2BIG; - u32 func; - - if (cpuid->nent < 1) - goto out; - if (cpuid->nent > KVM_MAX_CPUID_ENTRIES) - cpuid->nent = KVM_MAX_CPUID_ENTRIES; - r = -ENOMEM; - cpuid_entries = vmalloc(sizeof(struct kvm_cpuid_entry2) * cpuid->nent); - if (!cpuid_entries) - goto out; - - do_cpuid_ent(&cpuid_entries[0], 0, 0, &nent, cpuid->nent); - limit = cpuid_entries[0].eax; - for (func = 1; func <= limit && nent < cpuid->nent; ++func) - do_cpuid_ent(&cpuid_entries[nent], func, 0, - &nent, cpuid->nent); - r = -E2BIG; - if (nent >= cpuid->nent) - goto out_free; - - do_cpuid_ent(&cpuid_entries[nent], 0x80000000, 0, &nent, cpuid->nent); - limit = cpuid_entries[nent - 1].eax; - for (func = 0x80000001; func <= limit && nent < cpuid->nent; ++func) - do_cpuid_ent(&cpuid_entries[nent], func, 0, - &nent, cpuid->nent); - - - - r = -E2BIG; - if (nent >= cpuid->nent) - goto out_free; - - /* Add support for Centaur's CPUID instruction. */ - if (boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR) { - do_cpuid_ent(&cpuid_entries[nent], 0xC0000000, 0, - &nent, cpuid->nent); - - r = -E2BIG; - if (nent >= cpuid->nent) - goto out_free; - - limit = cpuid_entries[nent - 1].eax; - for (func = 0xC0000001; - func <= limit && nent < cpuid->nent; ++func) - do_cpuid_ent(&cpuid_entries[nent], func, 0, - &nent, cpuid->nent); - - r = -E2BIG; - if (nent >= cpuid->nent) - goto out_free; - } - - do_cpuid_ent(&cpuid_entries[nent], KVM_CPUID_SIGNATURE, 0, &nent, - cpuid->nent); - - r = -E2BIG; - if (nent >= cpuid->nent) - goto out_free; - - do_cpuid_ent(&cpuid_entries[nent], KVM_CPUID_FEATURES, 0, &nent, - cpuid->nent); - - r = -E2BIG; - if (nent >= cpuid->nent) - goto out_free; - - r = -EFAULT; - if (copy_to_user(entries, cpuid_entries, - nent * sizeof(struct kvm_cpuid_entry2))) - goto out_free; - cpuid->nent = nent; - r = 0; - -out_free: - vfree(cpuid_entries); -out: - return r; -} - static int kvm_vcpu_ioctl_get_lapic(struct kvm_vcpu *vcpu, struct kvm_lapic_state *s) { @@ -3042,13 +2540,12 @@ long kvm_arch_vcpu_ioctl(struct file *filp, r = -EINVAL; if (!vcpu->arch.apic) goto out; - u.lapic = kmalloc(sizeof(struct kvm_lapic_state), GFP_KERNEL); - r = -ENOMEM; - if (!u.lapic) - goto out; - r = -EFAULT; - if (copy_from_user(u.lapic, argp, sizeof(struct kvm_lapic_state))) + u.lapic = memdup_user(argp, sizeof(*u.lapic)); + if (IS_ERR(u.lapic)) { + r = PTR_ERR(u.lapic); goto out; + } + r = kvm_vcpu_ioctl_set_lapic(vcpu, u.lapic); if (r) goto out; @@ -3227,14 +2724,11 @@ long kvm_arch_vcpu_ioctl(struct file *filp, break; } case KVM_SET_XSAVE: { - u.xsave = kzalloc(sizeof(struct kvm_xsave), GFP_KERNEL); - r = -ENOMEM; - if (!u.xsave) - break; - - r = -EFAULT; - if (copy_from_user(u.xsave, argp, sizeof(struct kvm_xsave))) - break; + u.xsave = memdup_user(argp, sizeof(*u.xsave)); + if (IS_ERR(u.xsave)) { + r = PTR_ERR(u.xsave); + goto out; + } r = kvm_vcpu_ioctl_x86_set_xsave(vcpu, u.xsave); break; @@ -3255,15 +2749,11 @@ long kvm_arch_vcpu_ioctl(struct file *filp, break; } case KVM_SET_XCRS: { - u.xcrs = kzalloc(sizeof(struct kvm_xcrs), GFP_KERNEL); - r = -ENOMEM; - if (!u.xcrs) - break; - - r = -EFAULT; - if (copy_from_user(u.xcrs, argp, - sizeof(struct kvm_xcrs))) - break; + u.xcrs = memdup_user(argp, sizeof(*u.xcrs)); + if (IS_ERR(u.xcrs)) { + r = PTR_ERR(u.xcrs); + goto out; + } r = kvm_vcpu_ioctl_x86_set_xcrs(vcpu, u.xcrs); break; @@ -3460,16 +2950,59 @@ static int kvm_vm_ioctl_reinject(struct kvm *kvm, return 0; } +/** + * write_protect_slot - write protect a slot for dirty logging + * @kvm: the kvm instance + * @memslot: the slot we protect + * @dirty_bitmap: the bitmap indicating which pages are dirty + * @nr_dirty_pages: the number of dirty pages + * + * We have two ways to find all sptes to protect: + * 1. Use kvm_mmu_slot_remove_write_access() which walks all shadow pages and + * checks ones that have a spte mapping a page in the slot. + * 2. Use kvm_mmu_rmap_write_protect() for each gfn found in the bitmap. + * + * Generally speaking, if there are not so many dirty pages compared to the + * number of shadow pages, we should use the latter. + * + * Note that letting others write into a page marked dirty in the old bitmap + * by using the remaining tlb entry is not a problem. That page will become + * write protected again when we flush the tlb and then be reported dirty to + * the user space by copying the old bitmap. + */ +static void write_protect_slot(struct kvm *kvm, + struct kvm_memory_slot *memslot, + unsigned long *dirty_bitmap, + unsigned long nr_dirty_pages) +{ + /* Not many dirty pages compared to # of shadow pages. */ + if (nr_dirty_pages < kvm->arch.n_used_mmu_pages) { + unsigned long gfn_offset; + + for_each_set_bit(gfn_offset, dirty_bitmap, memslot->npages) { + unsigned long gfn = memslot->base_gfn + gfn_offset; + + spin_lock(&kvm->mmu_lock); + kvm_mmu_rmap_write_protect(kvm, gfn, memslot); + spin_unlock(&kvm->mmu_lock); + } + kvm_flush_remote_tlbs(kvm); + } else { + spin_lock(&kvm->mmu_lock); + kvm_mmu_slot_remove_write_access(kvm, memslot->id); + spin_unlock(&kvm->mmu_lock); + } +} + /* * Get (and clear) the dirty memory log for a memory slot. */ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log) { - int r, i; + int r; struct kvm_memory_slot *memslot; - unsigned long n; - unsigned long is_dirty = 0; + unsigned long n, nr_dirty_pages; mutex_lock(&kvm->slots_lock); @@ -3477,43 +3010,41 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, if (log->slot >= KVM_MEMORY_SLOTS) goto out; - memslot = &kvm->memslots->memslots[log->slot]; + memslot = id_to_memslot(kvm->memslots, log->slot); r = -ENOENT; if (!memslot->dirty_bitmap) goto out; n = kvm_dirty_bitmap_bytes(memslot); - - for (i = 0; !is_dirty && i < n/sizeof(long); i++) - is_dirty = memslot->dirty_bitmap[i]; + nr_dirty_pages = memslot->nr_dirty_pages; /* If nothing is dirty, don't bother messing with page tables. */ - if (is_dirty) { + if (nr_dirty_pages) { struct kvm_memslots *slots, *old_slots; - unsigned long *dirty_bitmap; + unsigned long *dirty_bitmap, *dirty_bitmap_head; - dirty_bitmap = memslot->dirty_bitmap_head; - if (memslot->dirty_bitmap == dirty_bitmap) - dirty_bitmap += n / sizeof(long); - memset(dirty_bitmap, 0, n); + dirty_bitmap = memslot->dirty_bitmap; + dirty_bitmap_head = memslot->dirty_bitmap_head; + if (dirty_bitmap == dirty_bitmap_head) + dirty_bitmap_head += n / sizeof(long); + memset(dirty_bitmap_head, 0, n); r = -ENOMEM; - slots = kzalloc(sizeof(struct kvm_memslots), GFP_KERNEL); + slots = kmemdup(kvm->memslots, sizeof(*kvm->memslots), GFP_KERNEL); if (!slots) goto out; - memcpy(slots, kvm->memslots, sizeof(struct kvm_memslots)); - slots->memslots[log->slot].dirty_bitmap = dirty_bitmap; - slots->generation++; + + memslot = id_to_memslot(slots, log->slot); + memslot->nr_dirty_pages = 0; + memslot->dirty_bitmap = dirty_bitmap_head; + update_memslots(slots, NULL); old_slots = kvm->memslots; rcu_assign_pointer(kvm->memslots, slots); synchronize_srcu_expedited(&kvm->srcu); - dirty_bitmap = old_slots->memslots[log->slot].dirty_bitmap; kfree(old_slots); - spin_lock(&kvm->mmu_lock); - kvm_mmu_slot_remove_write_access(kvm, log->slot); - spin_unlock(&kvm->mmu_lock); + write_protect_slot(kvm, memslot, dirty_bitmap, nr_dirty_pages); r = -EFAULT; if (copy_to_user(log->dirty_bitmap, dirty_bitmap, n)) @@ -3658,14 +3189,14 @@ long kvm_arch_vm_ioctl(struct file *filp, } case KVM_GET_IRQCHIP: { /* 0: PIC master, 1: PIC slave, 2: IOAPIC */ - struct kvm_irqchip *chip = kmalloc(sizeof(*chip), GFP_KERNEL); + struct kvm_irqchip *chip; - r = -ENOMEM; - if (!chip) + chip = memdup_user(argp, sizeof(*chip)); + if (IS_ERR(chip)) { + r = PTR_ERR(chip); goto out; - r = -EFAULT; - if (copy_from_user(chip, argp, sizeof *chip)) - goto get_irqchip_out; + } + r = -ENXIO; if (!irqchip_in_kernel(kvm)) goto get_irqchip_out; @@ -3684,14 +3215,14 @@ long kvm_arch_vm_ioctl(struct file *filp, } case KVM_SET_IRQCHIP: { /* 0: PIC master, 1: PIC slave, 2: IOAPIC */ - struct kvm_irqchip *chip = kmalloc(sizeof(*chip), GFP_KERNEL); + struct kvm_irqchip *chip; - r = -ENOMEM; - if (!chip) + chip = memdup_user(argp, sizeof(*chip)); + if (IS_ERR(chip)) { + r = PTR_ERR(chip); goto out; - r = -EFAULT; - if (copy_from_user(chip, argp, sizeof *chip)) - goto set_irqchip_out; + } + r = -ENXIO; if (!irqchip_in_kernel(kvm)) goto set_irqchip_out; @@ -3898,12 +3429,7 @@ void kvm_get_segment(struct kvm_vcpu *vcpu, kvm_x86_ops->get_segment(vcpu, var, seg); } -static gpa_t translate_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access) -{ - return gpa; -} - -static gpa_t translate_nested_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access) +gpa_t translate_nested_gpa(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access) { gpa_t t_gpa; struct x86_exception exception; @@ -4087,7 +3613,7 @@ int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa, ret = kvm_write_guest(vcpu->kvm, gpa, val, bytes); if (ret < 0) return 0; - kvm_mmu_pte_write(vcpu, gpa, val, bytes, 1); + kvm_mmu_pte_write(vcpu, gpa, val, bytes); return 1; } @@ -4324,7 +3850,7 @@ static int emulator_cmpxchg_emulated(struct x86_emulate_ctxt *ctxt, if (!exchanged) return X86EMUL_CMPXCHG_FAILED; - kvm_mmu_pte_write(vcpu, gpa, new, bytes, 1); + kvm_mmu_pte_write(vcpu, gpa, new, bytes); return X86EMUL_CONTINUE; @@ -4349,32 +3875,24 @@ static int kernel_pio(struct kvm_vcpu *vcpu, void *pd) return r; } - -static int emulator_pio_in_emulated(struct x86_emulate_ctxt *ctxt, - int size, unsigned short port, void *val, - unsigned int count) +static int emulator_pio_in_out(struct kvm_vcpu *vcpu, int size, + unsigned short port, void *val, + unsigned int count, bool in) { - struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); - - if (vcpu->arch.pio.count) - goto data_avail; - - trace_kvm_pio(0, port, size, count); + trace_kvm_pio(!in, port, size, count); vcpu->arch.pio.port = port; - vcpu->arch.pio.in = 1; + vcpu->arch.pio.in = in; vcpu->arch.pio.count = count; vcpu->arch.pio.size = size; if (!kernel_pio(vcpu, vcpu->arch.pio_data)) { - data_avail: - memcpy(val, vcpu->arch.pio_data, size * count); vcpu->arch.pio.count = 0; return 1; } vcpu->run->exit_reason = KVM_EXIT_IO; - vcpu->run->io.direction = KVM_EXIT_IO_IN; + vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT; vcpu->run->io.size = size; vcpu->run->io.data_offset = KVM_PIO_PAGE_OFFSET * PAGE_SIZE; vcpu->run->io.count = count; @@ -4383,36 +3901,37 @@ static int emulator_pio_in_emulated(struct x86_emulate_ctxt *ctxt, return 0; } -static int emulator_pio_out_emulated(struct x86_emulate_ctxt *ctxt, - int size, unsigned short port, - const void *val, unsigned int count) +static int emulator_pio_in_emulated(struct x86_emulate_ctxt *ctxt, + int size, unsigned short port, void *val, + unsigned int count) { struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); + int ret; - trace_kvm_pio(1, port, size, count); - - vcpu->arch.pio.port = port; - vcpu->arch.pio.in = 0; - vcpu->arch.pio.count = count; - vcpu->arch.pio.size = size; - - memcpy(vcpu->arch.pio_data, val, size * count); + if (vcpu->arch.pio.count) + goto data_avail; - if (!kernel_pio(vcpu, vcpu->arch.pio_data)) { + ret = emulator_pio_in_out(vcpu, size, port, val, count, true); + if (ret) { +data_avail: + memcpy(val, vcpu->arch.pio_data, size * count); vcpu->arch.pio.count = 0; return 1; } - vcpu->run->exit_reason = KVM_EXIT_IO; - vcpu->run->io.direction = KVM_EXIT_IO_OUT; - vcpu->run->io.size = size; - vcpu->run->io.data_offset = KVM_PIO_PAGE_OFFSET * PAGE_SIZE; - vcpu->run->io.count = count; - vcpu->run->io.port = port; - return 0; } +static int emulator_pio_out_emulated(struct x86_emulate_ctxt *ctxt, + int size, unsigned short port, + const void *val, unsigned int count) +{ + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); + + memcpy(vcpu->arch.pio_data, val, size * count); + return emulator_pio_in_out(vcpu, size, port, (void *)val, count, false); +} + static unsigned long get_segment_base(struct kvm_vcpu *vcpu, int seg) { return kvm_x86_ops->get_segment_base(vcpu, seg); @@ -4627,6 +4146,12 @@ static int emulator_set_msr(struct x86_emulate_ctxt *ctxt, return kvm_set_msr(emul_to_vcpu(ctxt), msr_index, data); } +static int emulator_read_pmc(struct x86_emulate_ctxt *ctxt, + u32 pmc, u64 *pdata) +{ + return kvm_pmu_read_pmc(emul_to_vcpu(ctxt), pmc, pdata); +} + static void emulator_halt(struct x86_emulate_ctxt *ctxt) { emul_to_vcpu(ctxt)->arch.halt_request = 1; @@ -4679,6 +4204,7 @@ static struct x86_emulate_ops emulate_ops = { .set_dr = emulator_set_dr, .set_msr = emulator_set_msr, .get_msr = emulator_get_msr, + .read_pmc = emulator_read_pmc, .halt = emulator_halt, .wbinvd = emulator_wbinvd, .fix_hypercall = emulator_fix_hypercall, @@ -4836,6 +4362,50 @@ static bool reexecute_instruction(struct kvm_vcpu *vcpu, gva_t gva) return false; } +static bool retry_instruction(struct x86_emulate_ctxt *ctxt, + unsigned long cr2, int emulation_type) +{ + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); + unsigned long last_retry_eip, last_retry_addr, gpa = cr2; + + last_retry_eip = vcpu->arch.last_retry_eip; + last_retry_addr = vcpu->arch.last_retry_addr; + + /* + * If the emulation is caused by #PF and it is non-page_table + * writing instruction, it means the VM-EXIT is caused by shadow + * page protected, we can zap the shadow page and retry this + * instruction directly. + * + * Note: if the guest uses a non-page-table modifying instruction + * on the PDE that points to the instruction, then we will unmap + * the instruction and go to an infinite loop. So, we cache the + * last retried eip and the last fault address, if we meet the eip + * and the address again, we can break out of the potential infinite + * loop. + */ + vcpu->arch.last_retry_eip = vcpu->arch.last_retry_addr = 0; + + if (!(emulation_type & EMULTYPE_RETRY)) + return false; + + if (x86_page_table_writing_insn(ctxt)) + return false; + + if (ctxt->eip == last_retry_eip && last_retry_addr == cr2) + return false; + + vcpu->arch.last_retry_eip = ctxt->eip; + vcpu->arch.last_retry_addr = cr2; + + if (!vcpu->arch.mmu.direct_map) + gpa = kvm_mmu_gva_to_gpa_write(vcpu, cr2, NULL); + + kvm_mmu_unprotect_page(vcpu->kvm, gpa >> PAGE_SHIFT); + + return true; +} + int x86_emulate_instruction(struct kvm_vcpu *vcpu, unsigned long cr2, int emulation_type, @@ -4877,6 +4447,9 @@ int x86_emulate_instruction(struct kvm_vcpu *vcpu, return EMULATE_DONE; } + if (retry_instruction(ctxt, cr2, emulation_type)) + return EMULATE_DONE; + /* this is needed for vmware backdoor interface to work since it changes registers values during IO operation */ if (vcpu->arch.emulate_regs_need_sync_from_vcpu) { @@ -5095,17 +4668,17 @@ static void kvm_timer_init(void) static DEFINE_PER_CPU(struct kvm_vcpu *, current_vcpu); -static int kvm_is_in_guest(void) +int kvm_is_in_guest(void) { - return percpu_read(current_vcpu) != NULL; + return __this_cpu_read(current_vcpu) != NULL; } static int kvm_is_user_mode(void) { int user_mode = 3; - if (percpu_read(current_vcpu)) - user_mode = kvm_x86_ops->get_cpl(percpu_read(current_vcpu)); + if (__this_cpu_read(current_vcpu)) + user_mode = kvm_x86_ops->get_cpl(__this_cpu_read(current_vcpu)); return user_mode != 0; } @@ -5114,8 +4687,8 @@ static unsigned long kvm_get_guest_ip(void) { unsigned long ip = 0; - if (percpu_read(current_vcpu)) - ip = kvm_rip_read(percpu_read(current_vcpu)); + if (__this_cpu_read(current_vcpu)) + ip = kvm_rip_read(__this_cpu_read(current_vcpu)); return ip; } @@ -5128,13 +4701,13 @@ static struct perf_guest_info_callbacks kvm_guest_cbs = { void kvm_before_handle_nmi(struct kvm_vcpu *vcpu) { - percpu_write(current_vcpu, vcpu); + __this_cpu_write(current_vcpu, vcpu); } EXPORT_SYMBOL_GPL(kvm_before_handle_nmi); void kvm_after_handle_nmi(struct kvm_vcpu *vcpu) { - percpu_write(current_vcpu, NULL); + __this_cpu_write(current_vcpu, NULL); } EXPORT_SYMBOL_GPL(kvm_after_handle_nmi); @@ -5233,15 +4806,6 @@ int kvm_emulate_halt(struct kvm_vcpu *vcpu) } EXPORT_SYMBOL_GPL(kvm_emulate_halt); -static inline gpa_t hc_gpa(struct kvm_vcpu *vcpu, unsigned long a0, - unsigned long a1) -{ - if (is_long_mode(vcpu)) - return a0; - else - return a0 | ((gpa_t)a1 << 32); -} - int kvm_hv_hypercall(struct kvm_vcpu *vcpu) { u64 param, ingpa, outgpa, ret; @@ -5337,9 +4901,6 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu) case KVM_HC_VAPIC_POLL_IRQ: ret = 0; break; - case KVM_HC_MMU_OP: - r = kvm_pv_mmu_op(vcpu, a0, hc_gpa(vcpu, a1, a2), &ret); - break; default: ret = -KVM_ENOSYS; break; @@ -5369,125 +4930,6 @@ int emulator_fix_hypercall(struct x86_emulate_ctxt *ctxt) return emulator_write_emulated(ctxt, rip, instruction, 3, NULL); } -static int move_to_next_stateful_cpuid_entry(struct kvm_vcpu *vcpu, int i) -{ - struct kvm_cpuid_entry2 *e = &vcpu->arch.cpuid_entries[i]; - int j, nent = vcpu->arch.cpuid_nent; - - e->flags &= ~KVM_CPUID_FLAG_STATE_READ_NEXT; - /* when no next entry is found, the current entry[i] is reselected */ - for (j = i + 1; ; j = (j + 1) % nent) { - struct kvm_cpuid_entry2 *ej = &vcpu->arch.cpuid_entries[j]; - if (ej->function == e->function) { - ej->flags |= KVM_CPUID_FLAG_STATE_READ_NEXT; - return j; - } - } - return 0; /* silence gcc, even though control never reaches here */ -} - -/* find an entry with matching function, matching index (if needed), and that - * should be read next (if it's stateful) */ -static int is_matching_cpuid_entry(struct kvm_cpuid_entry2 *e, - u32 function, u32 index) -{ - if (e->function != function) - return 0; - if ((e->flags & KVM_CPUID_FLAG_SIGNIFCANT_INDEX) && e->index != index) - return 0; - if ((e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC) && - !(e->flags & KVM_CPUID_FLAG_STATE_READ_NEXT)) - return 0; - return 1; -} - -struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu, - u32 function, u32 index) -{ - int i; - struct kvm_cpuid_entry2 *best = NULL; - - for (i = 0; i < vcpu->arch.cpuid_nent; ++i) { - struct kvm_cpuid_entry2 *e; - - e = &vcpu->arch.cpuid_entries[i]; - if (is_matching_cpuid_entry(e, function, index)) { - if (e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC) - move_to_next_stateful_cpuid_entry(vcpu, i); - best = e; - break; - } - } - return best; -} -EXPORT_SYMBOL_GPL(kvm_find_cpuid_entry); - -int cpuid_maxphyaddr(struct kvm_vcpu *vcpu) -{ - struct kvm_cpuid_entry2 *best; - - best = kvm_find_cpuid_entry(vcpu, 0x80000000, 0); - if (!best || best->eax < 0x80000008) - goto not_found; - best = kvm_find_cpuid_entry(vcpu, 0x80000008, 0); - if (best) - return best->eax & 0xff; -not_found: - return 36; -} - -/* - * If no match is found, check whether we exceed the vCPU's limit - * and return the content of the highest valid _standard_ leaf instead. - * This is to satisfy the CPUID specification. - */ -static struct kvm_cpuid_entry2* check_cpuid_limit(struct kvm_vcpu *vcpu, - u32 function, u32 index) -{ - struct kvm_cpuid_entry2 *maxlevel; - - maxlevel = kvm_find_cpuid_entry(vcpu, function & 0x80000000, 0); - if (!maxlevel || maxlevel->eax >= function) - return NULL; - if (function & 0x80000000) { - maxlevel = kvm_find_cpuid_entry(vcpu, 0, 0); - if (!maxlevel) - return NULL; - } - return kvm_find_cpuid_entry(vcpu, maxlevel->eax, index); -} - -void kvm_emulate_cpuid(struct kvm_vcpu *vcpu) -{ - u32 function, index; - struct kvm_cpuid_entry2 *best; - - function = kvm_register_read(vcpu, VCPU_REGS_RAX); - index = kvm_register_read(vcpu, VCPU_REGS_RCX); - kvm_register_write(vcpu, VCPU_REGS_RAX, 0); - kvm_register_write(vcpu, VCPU_REGS_RBX, 0); - kvm_register_write(vcpu, VCPU_REGS_RCX, 0); - kvm_register_write(vcpu, VCPU_REGS_RDX, 0); - best = kvm_find_cpuid_entry(vcpu, function, index); - - if (!best) - best = check_cpuid_limit(vcpu, function, index); - - if (best) { - kvm_register_write(vcpu, VCPU_REGS_RAX, best->eax); - kvm_register_write(vcpu, VCPU_REGS_RBX, best->ebx); - kvm_register_write(vcpu, VCPU_REGS_RCX, best->ecx); - kvm_register_write(vcpu, VCPU_REGS_RDX, best->edx); - } - kvm_x86_ops->skip_emulated_instruction(vcpu); - trace_kvm_cpuid(function, - kvm_register_read(vcpu, VCPU_REGS_RAX), - kvm_register_read(vcpu, VCPU_REGS_RBX), - kvm_register_read(vcpu, VCPU_REGS_RCX), - kvm_register_read(vcpu, VCPU_REGS_RDX)); -} -EXPORT_SYMBOL_GPL(kvm_emulate_cpuid); - /* * Check if userspace requested an interrupt window, and that the * interrupt window is open. @@ -5648,6 +5090,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) int r; bool req_int_win = !irqchip_in_kernel(vcpu->kvm) && vcpu->run->request_interrupt_window; + bool req_immediate_exit = 0; if (vcpu->requests) { if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu)) @@ -5687,7 +5130,12 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) record_steal_time(vcpu); if (kvm_check_request(KVM_REQ_NMI, vcpu)) process_nmi(vcpu); - + req_immediate_exit = + kvm_check_request(KVM_REQ_IMMEDIATE_EXIT, vcpu); + if (kvm_check_request(KVM_REQ_PMU, vcpu)) + kvm_handle_pmu_event(vcpu); + if (kvm_check_request(KVM_REQ_PMI, vcpu)) + kvm_deliver_pmi(vcpu); } r = kvm_mmu_reload(vcpu); @@ -5738,6 +5186,9 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); + if (req_immediate_exit) + smp_send_reschedule(vcpu->cpu); + kvm_guest_enter(); if (unlikely(vcpu->arch.switch_db_regs)) { @@ -5943,10 +5394,6 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) if (r <= 0) goto out; - if (kvm_run->exit_reason == KVM_EXIT_HYPERCALL) - kvm_register_write(vcpu, VCPU_REGS_RAX, - kvm_run->hypercall.ret); - r = __vcpu_run(vcpu); out: @@ -6148,7 +5595,7 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu, mmu_reset_needed |= kvm_read_cr4(vcpu) != sregs->cr4; kvm_x86_ops->set_cr4(vcpu, sregs->cr4); if (sregs->cr4 & X86_CR4_OSXSAVE) - update_cpuid(vcpu); + kvm_update_cpuid(vcpu); idx = srcu_read_lock(&vcpu->kvm->srcu); if (!is_long_mode(vcpu) && is_pae(vcpu)) { @@ -6425,6 +5872,8 @@ int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu) kvm_async_pf_hash_reset(vcpu); vcpu->arch.apf.halted = false; + kvm_pmu_reset(vcpu); + return kvm_x86_ops->vcpu_reset(vcpu); } @@ -6473,10 +5922,6 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) kvm = vcpu->kvm; vcpu->arch.emulate_ctxt.ops = &emulate_ops; - vcpu->arch.walk_mmu = &vcpu->arch.mmu; - vcpu->arch.mmu.root_hpa = INVALID_PAGE; - vcpu->arch.mmu.translate_gpa = translate_gpa; - vcpu->arch.nested_mmu.translate_gpa = translate_nested_gpa; if (!irqchip_in_kernel(kvm) || kvm_vcpu_is_bsp(vcpu)) vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE; else @@ -6513,6 +5958,7 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) goto fail_free_mce_banks; kvm_async_pf_hash_reset(vcpu); + kvm_pmu_init(vcpu); return 0; fail_free_mce_banks: @@ -6531,6 +5977,7 @@ void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu) { int idx; + kvm_pmu_destroy(vcpu); kfree(vcpu->arch.mce_banks); kvm_free_lapic(vcpu); idx = srcu_read_lock(&vcpu->kvm->srcu); diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h index d36fe23..cb80c29 100644 --- a/arch/x86/kvm/x86.h +++ b/arch/x86/kvm/x86.h @@ -33,9 +33,6 @@ static inline bool kvm_exception_is_soft(unsigned int nr) return (nr == BP_VECTOR) || (nr == OF_VECTOR); } -struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu, - u32 function, u32 index); - static inline bool is_protmode(struct kvm_vcpu *vcpu) { return kvm_read_cr0_bits(vcpu, X86_CR0_PE); @@ -125,4 +122,6 @@ int kvm_write_guest_virt_system(struct x86_emulate_ctxt *ctxt, gva_t addr, void *val, unsigned int bytes, struct x86_exception *exception); +extern u64 host_xcr0; + #endif diff --git a/arch/x86/xen/Kconfig b/arch/x86/xen/Kconfig index 26c731a..fdce49c 100644 --- a/arch/x86/xen/Kconfig +++ b/arch/x86/xen/Kconfig @@ -29,7 +29,8 @@ config XEN_PVHVM config XEN_MAX_DOMAIN_MEMORY int - default 128 + default 500 if X86_64 + default 64 if X86_32 depends on XEN help This only affects the sizing of some bss arrays, the unused @@ -48,3 +49,4 @@ config XEN_DEBUG_FS help Enable statistics output and various tuning options in debugfs. Enabling this option may incur a significant performance overhead. + diff --git a/arch/x86/xen/grant-table.c b/arch/x86/xen/grant-table.c index 5a40d24..3a5f55d 100644 --- a/arch/x86/xen/grant-table.c +++ b/arch/x86/xen/grant-table.c @@ -54,6 +54,20 @@ static int map_pte_fn(pte_t *pte, struct page *pmd_page, return 0; } +/* + * This function is used to map shared frames to store grant status. It is + * different from map_pte_fn above, the frames type here is uint64_t. + */ +static int map_pte_fn_status(pte_t *pte, struct page *pmd_page, + unsigned long addr, void *data) +{ + uint64_t **frames = (uint64_t **)data; + + set_pte_at(&init_mm, addr, pte, mfn_pte((*frames)[0], PAGE_KERNEL)); + (*frames)++; + return 0; +} + static int unmap_pte_fn(pte_t *pte, struct page *pmd_page, unsigned long addr, void *data) { @@ -64,10 +78,10 @@ static int unmap_pte_fn(pte_t *pte, struct page *pmd_page, int arch_gnttab_map_shared(unsigned long *frames, unsigned long nr_gframes, unsigned long max_nr_gframes, - struct grant_entry **__shared) + void **__shared) { int rc; - struct grant_entry *shared = *__shared; + void *shared = *__shared; if (shared == NULL) { struct vm_struct *area = @@ -83,8 +97,30 @@ int arch_gnttab_map_shared(unsigned long *frames, unsigned long nr_gframes, return rc; } -void arch_gnttab_unmap_shared(struct grant_entry *shared, - unsigned long nr_gframes) +int arch_gnttab_map_status(uint64_t *frames, unsigned long nr_gframes, + unsigned long max_nr_gframes, + grant_status_t **__shared) +{ + int rc; + grant_status_t *shared = *__shared; + + if (shared == NULL) { + /* No need to pass in PTE as we are going to do it + * in apply_to_page_range anyhow. */ + struct vm_struct *area = + alloc_vm_area(PAGE_SIZE * max_nr_gframes, NULL); + BUG_ON(area == NULL); + shared = area->addr; + *__shared = shared; + } + + rc = apply_to_page_range(&init_mm, (unsigned long)shared, + PAGE_SIZE * nr_gframes, + map_pte_fn_status, &frames); + return rc; +} + +void arch_gnttab_unmap(void *shared, unsigned long nr_gframes) { apply_to_page_range(&init_mm, (unsigned long)shared, PAGE_SIZE * nr_gframes, unmap_pte_fn, NULL); diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index f4bf8aa..58a0e46 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -1852,7 +1852,7 @@ pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd, xen_write_cr3(__pa(initial_page_table)); memblock_reserve(__pa(xen_start_info->pt_base), - xen_start_info->nr_pt_frames * PAGE_SIZE)); + xen_start_info->nr_pt_frames * PAGE_SIZE); return initial_page_table; } |