| Age | Commit message (Collapse) | Author |
|---|
|
Fixes STAR 9000953410: "perf callgraph profiling causing RCU stalls"
| perf record -g -c 15000 -e cycles /sbin/hackbench
|
| INFO: rcu_preempt self-detected stall on CPU
| 1: (1 GPs behind) idle=609/140000000000002/0 softirq=2914/2915 fqs=603
| Task dump for CPU 1:
in-kernel dwarf unwinder has a fast binary lookup and a fallback linear
search (which iterates thru each of ~11K entries) thus takes 2 orders of
magnitude longer (~3 million cycles vs. 2000). Routines written in hand
assembler lack dwarf info (as we don't support assembler CFI pseudo-ops
yet) fail the unwinder binary lookup, hit linear search, failing
nevertheless in the end.
However the linear search is pointless as binary lookup tables are created
from it in first place. It is impossible to have binary lookup fail while
succeed the linear search. It is pure waste of cycles thus removed by
this patch.
This manifested as RCU stalls / NMI watchdog splat when running
hackbench under perf with callgraph profiling. The triggering condition
was perf counter overflowing in routine lacking dwarf info (like memset)
leading to patheic 3 million cycle unwinder slow path and by the time it
returned new interrupts were already pending (Timer, IPI) and taken
rightaway. The original memset didn't make forward progress, system kept
accruing more interrupts and more unwinder delayes in a vicious feedback
loop, ultimately triggering the NMI diagnostic.
Cc: stable@vger.kernel.org
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
SYNC in __switch_to() is a historic relic and not needed at all.
- In UP context it is obviously useless, why would we want to stall
the core for all updates to stack memory of t0 to complete before
loading kernel mode callee registers from t1 stack's memory.
- In SMP, there could be potential race in which outgoing task could
be concurrently picked for running on a different core, thus writes
to stack here need to be visible before the reads from stack on
other core. Peter confirmed that generic schedular already has needed
barriers (by way of rq lock) so there is no need for additional arch
barrier.
This came up when Noam was trying to replace this SYNC with EZChip
specific hardware thread scheduling instruction for their platform
support.
Link: http://lkml.kernel.org/r/20151102092654.GM17308@twins.programming.kicks-ass.net
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: linux-kernel@vger.kernel.org
Cc: Noam Camus <noamc@ezchip.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Although kernel doesn't support the multiple IRQ priority levels provided
by HS38x core intc yet, ensure that the default prio value is used
anyways by relevant code.
SLEEP insn needs to be provided the IRQ priority level which can
interrupt it. This needs to be the default level which maynot
necessarily be 0 as assumed by current code.
This change allows a kernel with ARCV2_IRQ_DEF_PRIO = 1 to boot fine.
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
No semantical changes, prepares for ARCv2 specific change in next commit
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
When building kernel with buildroot built toolchain, CROSS_COMPILE
currently needs adjustment even if minor. This is because the defconfigs
prefer "arc-linux-uclibc-" prefix from hand built (non buildroot) toolchain
while buildroot provides "arc-buildroot-linux-uclibc-"
To avoid this use the common "arc-linux-" prefix which is provided by
buildroot and has also been in hand built tools for quite some time.
Signed-off-by: sujayraaj <sujayraaj@gmail.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
[vgupta: updated changelog]
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc
Pull ARC fixes from Vineet Gupta:
"Found a couple of brown paper bag bugs with the prev pull request
(including a SMP build breakage report from Guenter). Since these are
urgent I also decided to send over a bunch of other pending fixes
which could have otherwise waited an rc or two.
Summary:
- A bunch of brown paper bag bugs (MAINTAINERS list email, SMP build
failure)
- cpu_relax() now compiler barrier for UP as well
- handling of userspace Bus Errors for ARCompact builds"
* tag 'arc-4.4-rc1-part2' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc:
ARC: Fix silly typo in MAINTAINERS file
ARC: cpu_relax() to be compiler barrier even for UP
ARC: use ASL assembler mnemonic
ARC: [arcompact] Handle bus error from userspace as Interrupt not exception
ARC: remove extraneous header include
ARCv2: lib: memcpy: use local symbols
|
|
cpu_relax() on ARC has been barrier only for SMP (and no-op for UP). Per
recent discussions, it is safer to make it a compiler barrier
unconditionally.
Link: http://lkml.kernel.org/r/53A7D3AA.9020100@synopsys.com
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
ARCompact and ARCv2 only have ASL, while binutils used to support LSL as
a alias mnemonic.
Newer binutils (upstream) don't want to do that so replace it.
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Bus errors from userspace on ARCompact based cores are handled by core
as a high priority L2 interrupt but current code treated it as interrupt
Handling an interrupt like exception is certainly not going to go unnoticed.
(and it worked so far as we never saw a Bus error from userspace until
IPPK guys tested a DDR controller with ECC error detection etc hence
needed to explicitly trigger/handle such errors)
- So move mem_service exception handler from common code into ARCv2 code.
- In ARCompact code, define mem_service as L2 interrupt handler which
just drops down to pure kernel mode and goes of to enqueue SIGBUS
Reported-by: Nelson Pereira <npereira@synopsys.com>
Tested-by: Ana Martins <amartins@synopsys.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux
Pull DeviceTree updates from Rob Herring:
"A fairly large (by DT standards) pull request this time with the
majority being some overdue moving DT binding docs around to
consolidate similar bindings.
- DT binding doc consolidation moving similar bindings to common
locations. The majority of these are display related which were
scattered in video/, fb/, drm/, gpu/, and panel/ directories.
- Add new config option, CONFIG_OF_ALL_DTBS, to enable building all
dtbs in the tree for most arches with dts files (except powerpc for
now).
- OF_IRQ=n fixes for user enabled CONFIG_OF.
- of_node_put ref counting fixes from Julia Lawall.
- Common DT binding for wakeup-source and deprecation of all similar
bindings.
- DT binding for PXA LCD controller.
- Allow ignoring failed PCI resource translations in order to ignore
64-bit addresses on non-LPAE 32-bit kernels.
- Support setting the NUMA node from DT instead of only from parent
device.
- Couple of earlycon DT parsing fixes for address and options"
* tag 'devicetree-for-4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux: (45 commits)
MAINTAINERS: update DT binding doc locations
devicetree: add Sigma Designs vendor prefix
of: simplify arch_find_n_match_cpu_physical_id() function
Documentation: arm: Fixed typo in socfpga fpga mgr example
Documentation: devicetree: fix reference to legacy wakeup properties
Documentation: devicetree: standardize/consolidate on "wakeup-source" property
drivers: of: removing assignment of 0 to static variable
xtensa: enable building of all dtbs
mips: enable building of all dtbs
metag: enable building of all dtbs
metag: use common make variables for dtb builds
h8300: enable building of all dtbs
arm64: enable building of all dtbs
arm: enable building of all dtbs
arc: enable building of all dtbs
arc: use common make variables for dtb builds
of: add config option to enable building of all dtbs
of/fdt: fix error checking for earlycon address
of/overlay: add missing of_node_put
of/platform: add missing of_node_put
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull locking changes from Ingo Molnar:
"The main changes in this cycle were:
- More gradual enhancements to atomic ops: new atomic*_read_ctrl()
ops, synchronize atomic_{read,set}() ordering requirements between
architectures, add atomic_long_t bitops. (Peter Zijlstra)
- Add _{relaxed|acquire|release}() variants for inc/dec atomics and
use them in various locking primitives: mutex, rtmutex, mcs, rwsem.
This enables weakly ordered architectures (such as arm64) to make
use of more locking related optimizations. (Davidlohr Bueso)
- Implement atomic[64]_{inc,dec}_relaxed() on ARM. (Will Deacon)
- Futex kernel data cache footprint micro-optimization. (Rasmus
Villemoes)
- pvqspinlock runtime overhead micro-optimization. (Waiman Long)
- misc smaller fixlets"
* 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
ARM, locking/atomics: Implement _relaxed variants of atomic[64]_{inc,dec}
locking/rwsem: Use acquire/release semantics
locking/mcs: Use acquire/release semantics
locking/rtmutex: Use acquire/release semantics
locking/mutex: Use acquire/release semantics
locking/asm-generic: Add _{relaxed|acquire|release}() variants for inc/dec atomics
atomic: Implement atomic_read_ctrl()
atomic, arch: Audit atomic_{read,set}()
atomic: Add atomic_long_t bitops
futex: Force hot variables into a single cache line
locking/pvqspinlock: Kick the PV CPU unconditionally when _Q_SLOW_VAL
locking/osq: Relax atomic semantics
locking/qrwlock: Rename ->lock to ->wait_lock
locking/Documentation/lockstat: Fix typo - lokcing -> locking
locking/atomics, cmpxchg: Privatize the inclusion of asm/cmpxchg.h
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc
Pull ARC updates from Vineet Gupta:
- Support for new MM features in ARCv2 cores (THP, PAE40) Some generic
THP bits are touched - all ACKed by Kirill
- Platform framework updates to prepare for EZChip arrival (still in works)
- ARC Public Mailing list setup finally (linux-snps-arc@lists.infraded.org)
* tag 'arc-4.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc: (42 commits)
ARC: mm: PAE40 support
ARC: mm: PAE40: tlbex.S: Explicitify the size of pte_t
ARC: mm: PAE40: switch to using phys_addr_t for physical addresses
ARC: mm: HIGHMEM: populate high memory from DT
ARC: mm: HIGHMEM: kmap API implementation
ARC: mm: preps ahead of HIGHMEM support #2
ARC: mm: preps ahead of HIGHMEM support
ARC: mm: use generic macros _BITUL()/_AC()
ARC: mm: Improve Duplicate PD Fault handler
MAINTAINERS: Add public mailing list for ARC
ARC: Ensure DT mem base is same as what kernel is built with
ARC: boot: Non Master cpus only need to call EARLY_CPU_SETUP once
ARCv2: smp: [plat-*]: No need to explicitly call mcip_init_smp()
ARC: smp: Introduce smp hook @init_irq_cpu called for all cores
ARC: smp: Rename platform hook @init_smp -> @init_cpu_smp
ARCv2: smp: [plat-*]: No need to explicitly call mcip_init_early_smp()
ARC: smp: Introduce smp hook @init_early_smp for Master core
ARC: remove @init_time, @init_irq platform callbacks
ARC: smp: irqchip: handle IPI as percpu irq like timer
ARC: boot: Support Halt-on-reset and Run-on-reset SMP booting modes
...
|
|
Otherwise perf profiles don't charge tme to memcpy
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
This is the first working implementation of 40-bit physical address
extension on ARCv2.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
That way a single flip of phys_addr_t to 64 bit ensures all places
dealing with physical addresses get correct data
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Implement kmap* API for ARC.
This enables
- permanent kernel maps (pkmaps): :kmap() API
- fixmap : kmap_atomic()
We use a very simple/uniform approach for both (unlike some of the other
arches). So fixmap doesn't use the customary compile time address stuff.
The important semantic is sleep'ability (pkmap) vs. not (fixmap) which
the API guarantees.
Note that this patch only enables highmem for subsequent PAE40 support
as there is no real highmem for ARC in pure 32-bit paradigm as explained
below.
ARC has 2:2 address split of the 32-bit address space with lower half
being translated (virtual) while upper half unstranslated
(0x8000_0000 to 0xFFFF_FFFF). kernel itself is linked at base of
unstranslated space (i.e. 0x8000_0000 onwards), which is mapped to say
DDR 0x0 by external Bus Glue logic (outside the core). So kernel can
potentially access 1.75G worth of memory directly w/o need for highmem.
(the top 256M is taken by uncached peripheral space from 0xF000_0000 to
0xFFFF_FFFF)
In PAE40, hardware can address memory beyond 4G (0x1_0000_0000) while
the logical/virtual addresses remain 32-bits. Thus highmem is required
for kernel proper to be able to access these pages for it's own purposes
(user space is agnostic to this anyways).
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Explicit'ify that all memory added so far is low memory
Nothing semantical
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Before we plug in highmem support, some of code needs to be ready for it
- copy_user_highpage() needs to be using the kmap_atomic API
- mk_pte() can't assume page_address()
- do_page_fault() can't assume VMALLOC_END is end of kernel vaddr space
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
- Move the verbosity knob from .data to .bss by using inverted logic
- No need to readout PD1 descriptor
- clip the non pfn bits of PD0 to avoid clipping inside the loop
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
With prev fixes, all cores now start via common entry point @stext which
already calls EARLY_CPU_SETUP for all cores - so no need to invoke it
again
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
MCIP now registers it's own per cpu setup routine (for IPI IRQ request)
using smp_ops.init_irq_cpu().
So no need for platforms to do that. This now completely decouples
platforms from MCIP.
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Note this is not part of platform owned static machine_desc,
but more of device owned plat_smp_ops (rather misnamed) which a IPI
provider or some such typically defines.
This will help us seperate out the IPI registration from platform
specific init_cpu_smp() into device specific init_irq_cpu()
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
This conveys better that it is called for each cpu
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
MCIP now registers it's own probe callback with smp_ops.init_early_smp()
which is called by ARC common code, so no need for platforms to do that.
This decouples the platforms and MCIP and helps confine MCIP details
to it's own file.
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
This adds a platform agnostic early SMP init hook which is called on
Master core before calling setup_processor()
setup_arch()
smp_init_cpus()
smp_ops.init_early_smp()
...
setup_processor()
How this helps:
- Used for one time init of certain SMP centric IP blocks, before
calling setup_processor() which probes various bits of core,
possibly including this block
- Currently platforms need to call this IP block init from their
init routines, which doesn't make sense as this is specific to ARC
core and not platform and otherwise requires copy/paste in all
(and hence a possible point of failure)
e.g. MCIP init is called from 2 platforms currently (axs10x and sim)
which will go away once we have this.
This change only adds the hooks but they are empty for now. Next commit
will populate them and remove the explicit init calls from platforms.
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
These are not in use for ARC platforms. Moreover DT mechanims exist to
probe them w/o explicit platform calls.
- clocksource drivers can use CLOCKSOURCE_OF_DECLARE()
- intc IRQCHIP_DECLARE() calls + cascading inside DT allows external
intc to be probed automatically
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
The reason this was not done so far was lack of genuine IPI_IRQ for
ARC700, as we don't have a SMP version of core yet (which might change
soon thx to EZChip). Nevertheles to increase the build coverage, we
need to allow CONFIG_SMP for ARC700 and still be able to run it on a
UP platform (nsim or AXS101) with a UP Device Tree (SMP-on-UP)
The build itself requires some define for IPI_IRQ and even a dummy
value is fine since that code won't run anyways.
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
For Run-on-reset, non masters need to spin wait. For Halt-on-reset they
can jump to entry point directly.
Also while at it, made reset vector handler as "the" entry point for
kernel including host debugger based boot (which uses the ELF header
entry point)
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Enable building all dtb files when CONFIG_OF_ALL_DTBS is enabled. The dtbs
are not really dependent on a platform being enabled or any other kernel
config, so for testing coverage it is convenient to build all of the dtbs.
This builds all dts files in the tree, not just targets listed.
Signed-off-by: Rob Herring <robh@kernel.org>
Acked-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Use dtb-y and always make variables to build dtbs instead of explicit
dtbs rule. This is in preparation to support building all dtbs.
Signed-off-by: Rob Herring <robh@kernel.org>
Acked-by: Vineet Gupta <vgupta@synopsys.com>
|
|
DesignWare MMC Controller's transfer mode should be decided
at runtime instead of compile-time. So we remove this config
option and read dw_mmc's register to select DMA master.
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Acked-by: Vineet Gupta <vgupta@synopsys.com>
Signed-off-by: Jaehoon Chung <jh80.chung@samsung.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
|
|
For non halt-on-reset case, all cores start of simultaneously in @stext.
Master core0 proceeds with kernel boot, while other spin-wait on
@wake_flag being set by master once it is ready. So NO hardware assist
is needed for master to "kick" the others.
This patch moves this soft implementation out of mcip.c (as there is no
hardware assist) into common smp.c
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
This frees up some bits to hold more high level info such as PAE being
present, w/o increasing the size of already bloated cpuinfo struct
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
It was generating warnings when called as write_aux_reg(x, paddr >> 32)
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
This is done by improving the laddering logic !
Before:
if Exception
goto excep_or_pure_k_ret
if !Interrupt(L2)
goto l1_chk
else
INTERRUPT_EPILOGUE 2
l1_chk:
if !Interrupt(L1) (i.e. pure kernel mode)
goto excep_or_pure_k_ret
else
INTERRUPT_EPILOGUE 1
excep_or_pure_k_ret:
EXCEPTION_EPILOGUE
Now:
if !Interrupt(L1 or L2) (i.e. exception or pure kernel mode)
goto excep_or_pure_k_ret
; guaranteed to be an interrupt
if !Interrupt(L2)
goto l1_ret
else
INTERRUPT_EPILOGUE 2
; by virtue of above, no need to chk for L1 active
l1_ret:
INTERRUPT_EPILOGUE 1
excep_or_pure_k_ret:
EXCEPTION_EPILOGUE
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
The requirement is to
- Reenable Exceptions (AE cleared)
- Reenable Interrupts (E1/E2 set)
We need to do wiggle these bits into ERSTATUS and call RTIE.
Prev version used the pre-exception STATUS32 as starting point for what
goes into ERSTATUS. This required explicit fixups of U/DE/L bits.
Instead, use the current (in-exception) STATUS32 as starting point.
Being in exception handler U/DE/L can be safely assumed to be correct.
Only AE/E1/E2 need to be fixed.
So the new implementation is slightly better
-Avoids read form memory
-Is 4 bytes smaller for the typical 1 level of intr configuration
-Depicts the semantics more clearly
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Historically this was done by ARC IDE driver, which is long gone.
IRQ core is pretty robust now and already checks if IRQs are enabled
in hard ISRs. Thus no point in checking this in arch code, for every
call of irq enabled.
Further if some driver does do that - let it bring down the system so we
notice/fix this sooner than covering up for sucker
This makes local_irq_enable() - for L1 only case atleast simple enough
so we can inline it.
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Implement the TLB flush routine to evict a sepcific Super TLB entry,
vs. moving to a new ASID on every such flush.
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
|
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
