powerpc/e6500: Add support for hardware threads

The general idea is that each core will release all of its
threads into the secondary thread startup code, which will
eventually wait in the secondary core holding area, for the
appropriate bit in the PACA to be set. The kick_cpu function
pointer will set that bit in the PACA, and thus "release"
the core/thread to boot. We also need to do a few things that
U-Boot normally does for CPUs (like enable branch prediction).

Signed-off-by: Andy Fleming <afleming@freescale.com>
[scottwood@freescale.com: various changes, including only enabling
 threads if Linux wants to kick them]
Signed-off-by: Scott Wood <scottwood@freescale.com>

8 files changed, 116 insertions, 10 deletions

diff --git a/arch/powerpc/include/asm/cputable.h b/arch/powerpc/include/asm/cputable.h
index bc23477..e91dec8 100644
--- a/arch/powerpc/include/asm/cputable.h
+++ b/arch/powerpc/include/asm/cputable.h
@@ -396,7 +396,7 @@ extern const char *powerpc_base_platform;
 	    CPU_FTR_L2CSR | CPU_FTR_LWSYNC | CPU_FTR_NOEXECUTE | \
 	    CPU_FTR_DBELL | CPU_FTR_POPCNTB | CPU_FTR_POPCNTD | \
 	    CPU_FTR_DEBUG_LVL_EXC | CPU_FTR_EMB_HV | CPU_FTR_ALTIVEC_COMP | \
-	    CPU_FTR_CELL_TB_BUG)
+	    CPU_FTR_CELL_TB_BUG | CPU_FTR_SMT)
 #define CPU_FTRS_GENERIC_32	(CPU_FTR_COMMON | CPU_FTR_NODSISRALIGN)
 
 /* 64-bit CPUs */
diff --git a/arch/powerpc/include/asm/ppc-opcode.h b/arch/powerpc/include/asm/ppc-opcode.h
index 3132bb9..e316dad 100644
--- a/arch/powerpc/include/asm/ppc-opcode.h
+++ b/arch/powerpc/include/asm/ppc-opcode.h
@@ -150,8 +150,10 @@
 #define PPC_INST_MCRXR_MASK		0xfc0007fe
 #define PPC_INST_MFSPR_PVR		0x7c1f42a6
 #define PPC_INST_MFSPR_PVR_MASK		0xfc1fffff
+#define PPC_INST_MFTMR			0x7c0002dc
 #define PPC_INST_MSGSND			0x7c00019c
 #define PPC_INST_MSGSNDP		0x7c00011c
+#define PPC_INST_MTTMR			0x7c0003dc
 #define PPC_INST_NOP			0x60000000
 #define PPC_INST_POPCNTB		0x7c0000f4
 #define PPC_INST_POPCNTB_MASK		0xfc0007fe
@@ -369,4 +371,11 @@
 #define TABORT(r)		stringify_in_c(.long PPC_INST_TABORT \
 					       | __PPC_RA(r))
 
+/* book3e thread control instructions */
+#define TMRN(x)			((((x) & 0x1f) << 16) | (((x) & 0x3e0) << 6))
+#define MTTMR(tmr, r)		stringify_in_c(.long PPC_INST_MTTMR | \
+					       TMRN(tmr) | ___PPC_RS(r))
+#define MFTMR(tmr, r)		stringify_in_c(.long PPC_INST_MFTMR | \
+					       TMRN(tmr) | ___PPC_RT(r))
+
 #endif /* _ASM_POWERPC_PPC_OPCODE_H */
diff --git a/arch/powerpc/include/asm/reg_booke.h b/arch/powerpc/include/asm/reg_booke.h
index da42982..1d65330 100644
--- a/arch/powerpc/include/asm/reg_booke.h
+++ b/arch/powerpc/include/asm/reg_booke.h
@@ -15,6 +15,8 @@
 #ifndef __ASM_POWERPC_REG_BOOKE_H__
 #define __ASM_POWERPC_REG_BOOKE_H__
 
+#include <asm/ppc-opcode.h>
+
 /* Machine State Register (MSR) Fields */
 #define MSR_GS_LG	28	/* Guest state */
 #define MSR_UCLE_LG	26	/* User-mode cache lock enable */
@@ -608,6 +610,13 @@
 /* Bit definitions for L1CSR2. */
 #define L1CSR2_DCWS	0x40000000	/* Data Cache write shadow */
 
+/* Bit definitions for BUCSR. */
+#define BUCSR_STAC_EN	0x01000000	/* Segment Target Address Cache */
+#define BUCSR_LS_EN	0x00400000	/* Link Stack */
+#define BUCSR_BBFI	0x00000200	/* Branch Buffer flash invalidate */
+#define BUCSR_BPEN	0x00000001	/* Branch prediction enable */
+#define BUCSR_INIT	(BUCSR_STAC_EN | BUCSR_LS_EN | BUCSR_BBFI | BUCSR_BPEN)
+
 /* Bit definitions for L2CSR0. */
 #define L2CSR0_L2E	0x80000000	/* L2 Cache Enable */
 #define L2CSR0_L2PE	0x40000000	/* L2 Cache Parity/ECC Enable */
@@ -731,5 +740,23 @@
 #define MMUBE1_VBE4		0x00000002
 #define MMUBE1_VBE5		0x00000001
 
+#define TMRN_IMSR0	0x120	/* Initial MSR Register 0 (e6500) */
+#define TMRN_IMSR1	0x121	/* Initial MSR Register 1 (e6500) */
+#define TMRN_INIA0	0x140	/* Next Instruction Address Register 0 */
+#define TMRN_INIA1	0x141	/* Next Instruction Address Register 1 */
+#define SPRN_TENSR	0x1b5	/* Thread Enable Status Register */
+#define SPRN_TENS	0x1b6	/* Thread Enable Set Register */
+#define SPRN_TENC	0x1b7	/* Thread Enable Clear Register */
+
+#define TEN_THREAD(x)	(1 << (x))
+
+#ifndef __ASSEMBLY__
+#define mftmr(rn)	({unsigned long rval; \
+			asm volatile(MFTMR(rn, %0) : "=r" (rval)); rval;})
+#define mttmr(rn, v)	asm volatile(MTTMR(rn, %0) : \
+				     : "r" ((unsigned long)(v)) \
+				     : "memory")
+#endif /* !__ASSEMBLY__ */
+
 #endif /* __ASM_POWERPC_REG_BOOKE_H__ */
 #endif /* __KERNEL__ */
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index a95145d..36ff6f03 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -180,6 +180,28 @@ exception_marker:
 #include "exceptions-64s.S"
 #endif
 
+#ifdef CONFIG_PPC_BOOK3E
+_GLOBAL(fsl_secondary_thread_init)
+	/* Enable branch prediction */
+	lis     r3,BUCSR_INIT@h
+	ori     r3,r3,BUCSR_INIT@l
+	mtspr   SPRN_BUCSR,r3
+	isync
+
+	/*
+	 * Fix PIR to match the linear numbering in the device tree.
+	 *
+	 * On e6500, the reset value of PIR uses the low three bits for
+	 * the thread within a core, and the upper bits for the core
+	 * number.  There are two threads per core, so shift everything
+	 * but the low bit right by two bits so that the cpu numbering is
+	 * continuous.
+	 */
+	mfspr	r3, SPRN_PIR
+	rlwimi	r3, r3, 30, 2, 30
+	mtspr	SPRN_PIR, r3
+#endif
+
 _GLOBAL(generic_secondary_thread_init)
 	mr	r24,r3
 
diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index 613a860..0448b1e 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -309,12 +309,10 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
 
 	/* Get physical cpuid */
 	intserv = of_get_flat_dt_prop(node, "ibm,ppc-interrupt-server#s", &len);
-	if (intserv) {
-		nthreads = len / sizeof(int);
-	} else {
-		intserv = of_get_flat_dt_prop(node, "reg", NULL);
-		nthreads = 1;
-	}
+	if (!intserv)
+		intserv = of_get_flat_dt_prop(node, "reg", &len);
+
+	nthreads = len / sizeof(int);
 
 	/*
 	 * Now see if any of these threads match our boot cpu.
diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c
index e239df3..1bb4dcd 100644
--- a/arch/powerpc/kernel/setup-common.c
+++ b/arch/powerpc/kernel/setup-common.c
@@ -456,18 +456,20 @@ void __init smp_setup_cpu_maps(void)
 		intserv = of_get_property(dn, "ibm,ppc-interrupt-server#s",
 				&len);
 		if (intserv) {
-			nthreads = len / sizeof(int);
 			DBG("    ibm,ppc-interrupt-server#s -> %d threads\n",
 			    nthreads);
 		} else {
 			DBG("    no ibm,ppc-interrupt-server#s -> 1 thread\n");
-			intserv = of_get_property(dn, "reg", NULL);
+			intserv = of_get_property(dn, "reg", &len);
 			if (!intserv) {
 				cpu_be = cpu_to_be32(cpu);
 				intserv = &cpu_be;	/* assume logical == phys */
+				len = 4;
 			}
 		}
 
+		nthreads = len / sizeof(int);
+
 		for (j = 0; j < nthreads && cpu < nr_cpu_ids; j++) {
 			DBG("    thread %d -> cpu %d (hard id %d)\n",
 			    j, cpu, be32_to_cpu(intserv[j]));
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index ee082d7..6d06947 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -507,7 +507,11 @@ void __init setup_system(void)
 	check_smt_enabled();
 	setup_tlb_core_data();
 
-#ifdef CONFIG_SMP
+	/*
+	 * Freescale Book3e parts spin in a loop provided by firmware,
+	 * so smp_release_cpus() does nothing for them
+	 */
+#if defined(CONFIG_SMP) && !defined(CONFIG_PPC_FSL_BOOK3E)
 	/* Release secondary cpus out of their spinloops at 0x60 now that
 	 * we can map physical -> logical CPU ids
 	 */
diff --git a/arch/powerpc/platforms/85xx/smp.c b/arch/powerpc/platforms/85xx/smp.c
index ba093f5..d7c1e69 100644
--- a/arch/powerpc/platforms/85xx/smp.c
+++ b/arch/powerpc/platforms/85xx/smp.c
@@ -28,6 +28,7 @@
 #include <asm/dbell.h>
 #include <asm/fsl_guts.h>
 #include <asm/code-patching.h>
+#include <asm/cputhreads.h>
 
 #include <sysdev/fsl_soc.h>
 #include <sysdev/mpic.h>
@@ -168,6 +169,24 @@ static inline u32 read_spin_table_addr_l(void *spin_table)
 	return in_be32(&((struct epapr_spin_table *)spin_table)->addr_l);
 }
 
+#ifdef CONFIG_PPC64
+static void wake_hw_thread(void *info)
+{
+	void fsl_secondary_thread_init(void);
+	unsigned long imsr1, inia1;
+	int nr = *(const int *)info;
+
+	imsr1 = MSR_KERNEL;
+	inia1 = *(unsigned long *)fsl_secondary_thread_init;
+
+	mttmr(TMRN_IMSR1, imsr1);
+	mttmr(TMRN_INIA1, inia1);
+	mtspr(SPRN_TENS, TEN_THREAD(1));
+
+	smp_generic_kick_cpu(nr);
+}
+#endif
+
 static int smp_85xx_kick_cpu(int nr)
 {
 	unsigned long flags;
@@ -183,6 +202,31 @@ static int smp_85xx_kick_cpu(int nr)
 
 	pr_debug("smp_85xx_kick_cpu: kick CPU #%d\n", nr);
 
+#ifdef CONFIG_PPC64
+	/* Threads don't use the spin table */
+	if (cpu_thread_in_core(nr) != 0) {
+		int primary = cpu_first_thread_sibling(nr);
+
+		if (WARN_ON_ONCE(!cpu_has_feature(CPU_FTR_SMT)))
+			return -ENOENT;
+
+		if (cpu_thread_in_core(nr) != 1) {
+			pr_err("%s: cpu %d: invalid hw thread %d\n",
+			       __func__, nr, cpu_thread_in_core(nr));
+			return -ENOENT;
+		}
+
+		if (!cpu_online(primary)) {
+			pr_err("%s: cpu %d: primary %d not online\n",
+			       __func__, nr, primary);
+			return -ENOENT;
+		}
+
+		smp_call_function_single(primary, wake_hw_thread, &nr, 0);
+		return 0;
+	}
+#endif
+
 	np = of_get_cpu_node(nr, NULL);
 	cpu_rel_addr = of_get_property(np, "cpu-release-addr", NULL);