Xiaohui Xin and some other folks at Intel have been looking into what's
behind the performance hit of paravirt_ops when running native.
It appears that the hit is entirely due to the paravirtualized
spinlocks introduced by:
| commit 8efcbab674
| Date: Mon Jul 7 12:07:51 2008 -0700
|
| paravirt: introduce a "lock-byte" spinlock implementation
The extra call/return in the spinlock path is somehow
causing an increase in the cycles/instruction of somewhere around 2-7%
(seems to vary quite a lot from test to test). The working theory is
that the CPU's pipeline is getting upset about the
call->call->locked-op->return->return, and seems to be failing to
speculate (though I haven't seen anything definitive about the precise
reasons). This doesn't entirely make sense, because the performance
hit is also visible on unlock and other operations which don't involve
locked instructions. But spinlock operations clearly swamp all the
other pvops operations, even though I can't imagine that they're
nearly as common (there's only a .05% increase in instructions
executed).
If I disable just the pv-spinlock calls, my tests show that pvops is
identical to non-pvops performance on native (my measurements show that
it is actually about .1% faster, but Xiaohui shows a .05% slowdown).
Summary of results, averaging 10 runs of the "mmperf" test, using a
no-pvops build as baseline:
nopv Pv-nospin Pv-spin
CPU cycles 100.00% 99.89% 102.18%
instructions 100.00% 100.10% 100.15%
CPI 100.00% 99.79% 102.03%
cache ref 100.00% 100.84% 100.28%
cache miss 100.00% 90.47% 88.56%
cache miss rate 100.00% 89.72% 88.31%
branches 100.00% 99.93% 100.04%
branch miss 100.00% 103.66% 107.72%
branch miss rt 100.00% 103.73% 107.67%
wallclock 100.00% 99.90% 102.20%
The clear effect here is that the 2% increase in CPI is
directly reflected in the final wallclock time.
(The other interesting effect is that the more ops are
out of line calls via pvops, the lower the cache access
and miss rates. Not too surprising, but it suggests that
the non-pvops kernel is over-inlined. On the flipside,
the branch misses go up correspondingly...)
So, what's the fix?
Paravirt patching turns all the pvops calls into direct calls, so
_spin_lock etc do end up having direct calls. For example, the compiler
generated code for paravirtualized _spin_lock is:
<_spin_lock+0>: mov %gs:0xb4c8,%rax
<_spin_lock+9>: incl 0xffffffffffffe044(%rax)
<_spin_lock+15>: callq *0xffffffff805a5b30
<_spin_lock+22>: retq
The indirect call will get patched to:
<_spin_lock+0>: mov %gs:0xb4c8,%rax
<_spin_lock+9>: incl 0xffffffffffffe044(%rax)
<_spin_lock+15>: callq <__ticket_spin_lock>
<_spin_lock+20>: nop; nop /* or whatever 2-byte nop */
<_spin_lock+22>: retq
One possibility is to inline _spin_lock, etc, when building an
optimised kernel (ie, when there's no spinlock/preempt
instrumentation/debugging enabled). That will remove the outer
call/return pair, returning the instruction stream to a single
call/return, which will presumably execute the same as the non-pvops
case. The downsides arel 1) it will replicate the
preempt_disable/enable code at eack lock/unlock callsite; this code is
fairly small, but not nothing; and 2) the spinlock definitions are
already a very heavily tangled mass of #ifdefs and other preprocessor
magic, and making any changes will be non-trivial.
The other obvious answer is to disable pv-spinlocks. Making them a
separate config option is fairly easy, and it would be trivial to
enable them only when Xen is enabled (as the only non-default user).
But it doesn't really address the common case of a distro build which
is going to have Xen support enabled, and leaves the open question of
whether the native performance cost of pv-spinlocks is worth the
performance improvement on a loaded Xen system (10% saving of overall
system CPU when guests block rather than spin). Still it is a
reasonable short-term workaround.
[ Impact: fix pvops performance regression when running native ]
Analysed-by: "Xin Xiaohui" <xiaohui.xin@intel.com>
Analysed-by: "Li Xin" <xin.li@intel.com>
Analysed-by: "Nakajima Jun" <jun.nakajima@intel.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Xen-devel <xen-devel@lists.xensource.com>
LKML-Reference: <4A0B62F7.5030802@goop.org>
[ fixed the help text ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jwessel/linux-2.6-kgdb:
kgdb: gdb documentation fix
kgdb,i386: use address that SP register points to in the exception frame
sysrq, intel_fb: fix sysrq g collision
* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
Revert "mm: add /proc controls for pdflush threads"
viocd: needs to depend on BLOCK
block: fix the bio_vec array index out-of-bounds test
* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:
powerpc: Fix PCI ROM access
powerpc/pseries: Really fix the oprofile CPU type on pseries
serial/nwpserial: Fix wrong register read address and add interrupt acknowledge.
powerpc/cell: Make ptcal more reliable
powerpc: Allow mem=x cmdline to work with 4G+
powerpc/mpic: Fix incorrect allocation of interrupt rev-map
powerpc: Fix oprofile sampling of marked events on POWER7
powerpc/iseries: Fix pci breakage due to bad dma_data initialization
powerpc: Fix mktree build error on Mac OS X host
powerpc/virtex: Fix duplicate level irq events.
powerpc/virtex: Add uImage to the default images list
powerpc/boot: add simpleImage.* to clean-files list
powerpc/8xx: Update defconfigs
powerpc/embedded6xx: Update defconfigs
powerpc/86xx: Update defconfigs
powerpc/85xx: Update defconfigs
powerpc/83xx: Update defconfigs
powerpc/fsl_soc: Remove mpc83xx_wdt_init, again
The s3c24xx_register_clock() function has been doing a test
on clk->owner to see if it is NULL, and then setting itself
as the owner if clk->owner == NULL.
This is not needed, arch/arm/plat-s3c/clock.c cannot be
compiled as a module, and even if it was, it should not be
playing with this field if it being registered from somewhere
else.
The best course of action is to remove this bit of
code completely.
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
The BAST support code is calling s3c_i2c0_set_platdata() from
the map_io() entry, instead of the bast_init() code. This causes
the registration to fail due to kmalloc() not being available
at the time.
This fixes the following error:
s3c_i2c0_set_platdata: no memory for platform data
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Fix unused code warning in arch/arm/plat-s3c24xx/dma.c if there
is no PM support enabled. The function to_dma_chan() should
be marked inline so that the compiler will eliminate it without
warning if it isn't used.
arch/arm/plat-s3c24xx/dma.c:1239: warning: 'to_dma_chan' defined but not used
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Fix compilation bug when debug was enabled
Reviewed-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Cleanup arm/plat-s3c64xx/include/plat/gpio-bank-h.h include file.
Using shift-left operation with value >32 is a bad habit.
Reviewed-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
The treatment of the SP register is different on x86_64 and i386.
This is a regression fix that lived outside the mainline kernel from
2.6.27 to now. The regression was a result of the original merge
consolidation of the i386 and x86_64 archs to x86.
The incorrectly reported SP on i386 prevented stack tracebacks from
working correctly in gdb.
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
The symbol 'floatx80_is_nan' prototype was defined
locally in fpa11_cprt.c when it was built outside the
file in softfloat-specialisze.
Move this into softfloat.h to fix the following sparse
warning:
softfloat-specialize:276:6: warning: symbol 'floatx80_is_nan' was not declared. Should it be static?
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Add header file decleration for 'ExtendedCPDO' in fpa11.h
to stop the following sparse warning:
extended_cpdo.c:90:14: warning: symbol 'ExtendedCPDO' was not declared. Should it be static?
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
This is a build fix, resyncing the DaVinci EVM ASoC board code
with the version in the DaVinci tree. That resync includes
support for the DM355 EVM, although that board isn't yet in
mainline.
(NOTE: also includes a bugfix to the platform_add_resources
call, recently sent by Chaithrika U S <chaithrika@ti.com> but
not yet merged into the DaVinci tree.)
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
A couple of issues crept in since about 2.6.27 related to accessing PCI
device ROMs on various powerpc machines.
First, historically, we don't allocate the ROM resource in the resource
tree. I'm not entirely certain of why, I susepct they often contained
garbage on x86 but it's hard to tell. This causes the current generic
code to always call pci_assign_resource() when trying to access the said
ROM from sysfs, which will try to re-assign some new address regardless
of what the ROM BAR was already set to at boot time. This can be a
problem on hypervisor platforms like pSeries where we aren't supposed
to move PCI devices around (and in fact probably can't).
Second, our code that generates the PCI tree from the OF device-tree
(instead of doing config space probing) which we mostly use on pseries
at the moment, didn't set the (new) flag IORESOURCE_SIZEALIGN on any
resource. That means that any attempt at re-assigning such a resource
with pci_assign_resource() would fail due to resource_alignment()
returning 0.
This fixes this by doing these two things:
- The code that calculates resource flags based on the OF device-node
is improved to set IORESOURCE_SIZEALIGN on any valid BAR, and while at
it also set IORESOURCE_READONLY for ROMs since we were lacking that too
- We now allocate ROM resources as part of the resource tree. However
to limit the chances of nasty conflicts due to busted firmwares, we
only do it on the second pass of our two-passes allocation scheme,
so that all valid and enabled BARs get precedence.
This brings pSeries back the ability to access PCI ROMs via sysfs (and
thus initialize various video cards from X etc...).
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
My previous pach for fixing the oprofile CPU type got somewhat mismerged
(by my fault) when it collided with another related patch. This should
finally (fingers crossed) fix the whole thing.
We make sure we keep the -old- oprofile type and CPU type whenever
one of them was specified in the first pass through the function.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
There have been a series of checkstops on QS21 related to
ptcal being set up incorrectly. On systems that only
have memory on a single node, ptcal fails when it gets
a pointer to memory on the remote node.
Moreover, agressive prefetching in memcpy and other
functions may accidentally touch the first cache line
of the page that we reserve for ptcal, which causes
an ECC checkstop.
We now allocate pages only from the specified node, moves the
ptcal area into the middle of the allocated page to avoid
potential prefetch problems and prints the address of the
ptcal area to facilitate diagnostics.
Signed-off-by: Gerhard Stenzel <gerhard.stenzel@de.ibm.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
We're currently choking on mem=4g (and above) due to memory_limit
being specified as an unsigned long. Make memory_limit
phys_addr_t to fix this.
Signed-off-by: Becky Bruce <beckyb@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Before when we were setting up the irq host map for mpic we passed in
just isu_size for the size of the linear map. However, for a number of
mpic implementations we have no isu (thus pass in 0) and will end up
with a no linear map (size = 0). This causes us to always call
irq_find_mapping() from mpic_get_irq().
By moving the allocation of the host map to after we've determined the
number of sources we can actually benefit from having a linear map for
the non-isu users that covers all the interrupt sources.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Description
-----------
Change ppc64 oprofile kernel driver to use the SLOT bits (MMCRA[37:39]only on
older processors where those bits are defined.
Background
----------
The performance monitor unit of the 64-bit POWER processor family has the
ability to collect accurate instruction-level samples when profiling on marked
events (i.e., "PM_MRK_<event-name>"). In processors prior to POWER6, the MMCRA
register contained "slot information" that the oprofile kernel driver used to
adjust the value latched in the SIAR at the time of a PMU interrupt. But as of
POWER6, these slot bits in MMCRA are no longer necessary for oprofile to use,
since the SIAR itself holds the accurate sampled instruction address. With
POWER6, these MMCRA slot bits were zero'ed out by hardware so oprofile's use of
these slot bits was, in effect, a NOP. But with POWER7, these bits are no
longer zero'ed out; however, they serve some other purpose rather than slot
information. Thus, using these bits on POWER7 to adjust the SIAR value results
in samples being attributed to the wrong instructions. The attached patch
changes the oprofile kernel driver to ignore these slot bits on all newer
processors starting with POWER6.
Signed-off-by: Maynard Johnson <maynardj@us.ibm.com>
Signed-off-by: Michael Wolf <mjw@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Commit 4fc665b88a "powerpc: Merge 32 and
64-bit dma code" made changes to the PCI initialisation code that added
an assignment to archdata.dma_data but only for 32 bit code. Commit
7eef440a54 "powerpc/pci: Cosmetic cleanups
of pci-common.c" removed the conditional compilation. Unfortunately,
the iSeries code setup the archdata.dma_data before that assignment was
done - effectively overwriting the dma_data with NULL.
Fix this up by moving the iSeries setup of dma_data into a
pci_dma_dev_setup callback.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The mktree utility defines some variables as "uint", although this is not a
standard C type, and so cross-compiling on Mac OS X fails. Change this to
"unsigned int".
Signed-off-by: Timur Tabi <timur@freescale.com>
Acked-by: Josh Boyer <jwboyer@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The interrupt controller was not handling level interrupts correctly
such that duplicate interrupts were happening. This fixes the problem
and adds edge type interrupts which are needed in Xilinx hardware.
Signed-off-by: John Linn <john.linn@xilinx.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
It is common to use U-Boot on Xilinx Virtex platforms. This patch
ensures that CONFIG_DEFAULT_UIMAGE is selected for virtex
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
Locking of irq_desc is now done in irq_set_affinity; don't lock it again
in chip specific set_affinity function.
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
When init is started it is SIGNAL_UNKILLABLE. If it were to get an
address error, we would try to send it SIGBUS, but it would be ignored
and the faulting instruction restarted. This results in an endless
loop.
We need to use force_sig() instead so it will actually die and give us
some useful information.
Reported-by: Florian Fainelli <florian@openwrt.org>
Signed-off-by: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
This patch modifies parameter of octeon_cvmcount_read() from 'void' to
'struct clocksource *cs', which fixes compile warning for incompatible
parameter type.
Signed-off-by: Coly Li <coly.li@suse.de>
Cc: David Daney <ddaney@caviumnetworks.com>
Cc: Ingo Molnar <mingo@elte.hu>
Reviewed-by: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
The inline assembler used on 32-bit kernels was using the "h" constraint
which was considered dangerous and removed for gcc 4.4.0.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Commit 351336929c (kernel.org) rsp.
b3594a089f1c17ff919f8f78505c3f20e1f6f8ce (linux-mips.org):
> From: Chris Dearman <chris@mips.com>
> Date: Wed, 19 Sep 2007 00:58:24 +0100
> Subject: [PATCH] [MIPS] Allow setting of the cache attribute at run time.
>
> Slightly tacky, but there is a precedent in the sparc archirecture code.
introduces the variable _page_cachable_default, which defaults to zero and.
is used to create the prototype PTE for __kmap_atomic in
arch/mips/mm/init.c:kmap_init before initialization in
arch/mips/mm/c-r4k.c:coherency_setup, so the default value of 0 will be
used as the CCA of kmap atomic pages which on many processors is not a
defined CCA value and may result in writes to kmap_atomic pages getting
corrupted. Debugged by Jon Fraser (jfraser@broadcom.com).
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
The RAMROOT function was a successful but non-portable attempt to append
the root filesystem to the end of the kernel image. The preferred and
portable solution is to use an initramfs instead.
Signed-off-by: Shane McDonald <mcdonald.shane@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
I don't think that in 15 years of Linux/MIPS the zero division checking
code generated by gcc by default has ever caught anything.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
There have been a number of compile problems with the msp71xx configuration
ever since it was included in the linux-mips.org repository. This patch
resolves compilation problems with attempting to reset the board using
non-existent GPIO routines.
This patch has been compile-tested against the current HEAD.
Signed-off-by: Shane McDonald <mcdonald.shane@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
There have been a number of compile problems with the msp71xx configuration
ever since it was included in the linux-mips.org repository. This patch
resolves the "multiple definition of plat_timer_setup" problem, and creates
the required get_c0_compare_int function.
This patch has been compile-tested against the current HEAD.
Signed-off-by: Shane McDonald <mcdonald.shane@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
There was already a define for NMI_OFFSET in asm/sn/addr.h, which now
clashes with linux/hardirq.h. Rename the one in sn/addr.h to fix IP27
builds..
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Beyond the requirements of the architecture standard Cavium also supports
8k and 32k pages.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Acked-by: David Daney <ddaney@caviumnetworks.com>
Addition of -fwrapv option in 2.6.29 discloses possible overflow with
signed arithmetics. For example, result of "a * 6 / 12" (int a =
400000000) is 200000000 without -fwrapv but -157913941 with -fwrapv.
Change some variable to unsigned to avoid such overflows.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Synchronize dma_map_page/dma_unmap_page and dma_map_single/dma_unmap_single.
This will reduce unnecessary writebacks and invalidates.
[Ralf: make dma_unmap_page an inline function.]
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
This is useful for IDT RC32332, RC32334 and NEC VR5500 processors which do
not implement the full MIPS32 / MIPS64 architecture.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Quoting from Loongson2FUserGuide.pdf:
5.22.1 Hazards
The processor detects most of the pipeline hazards in hardware, including
CP0 hazards and load hazards. No NOP instructions are required to correct
instruction sequences.
Signed-off-by: Zhang Le <r0bertz@gentoo.org>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
The compat.h does not need seccomp.h since TIF_32BIT was moved to
thread_info.h
This fixes a build error of 64-bit kernel without CONFIG_SECCOMP.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Acked-by: : David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
The LTP timerfd01 test is failing (blocking forever) on the 32-bit ABIs. We
need to use the compat_* wrappers for these system calls.
Signed-off-by: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
I have taken Wu Zhangjin's and Philippe Vachon's version as references,
did a little modification and tested on 16K page size kernel. It works
well.
Unfornately although it already has defined cpu_has_dc_aliases as 1, 4k
page size still not working. More work needed here.
Signed-off-by: Zhang Le <r0bertz@gentoo.org>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
The following symbols are needlessly defined global: cpuerr_irq and
memerr_irq. This patch makes the symbols static.
Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@movial.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
The file arch/mips/sgi-ip32/ip32-berr.c needlessly defines the function
ip32_be_handler() as global, and this patch makes it static.
Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@movial.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
The file arch/mips/mm/sc-rm7k.c needlessly defines two global symbols:
rm7k_sc_ops
rm7k_tcache_enabled
This patch makes these symbols static.
Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@movial.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
au1xxx_power_dev_t? is never defined; get rid of all PM stuff as well
since it is not in the driver source anyway.
Signed-off-by: Manuel Lauss <mano@roarinelk.homelinux.net>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Commit 779e7d41ad created a name collision
in SMTC builds. The attached patch corrects this in a a
not-too-terribly-ugly manner. Note that the SMTC case has to come
first, because CEVT_R4K will also be true.
Signed-off-by: Kevin D. Kissell <kevink@paralogos.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/czankel/xtensa-2.6:
xtensa: Fix linker script to include .literal sections
xtensa: update s6105_defconfig for ccount calibration
xtensa: implement ccount calibration for s6000
xtensa: fix wrong extern declaration renamed in code using it
xtensa: register gpio chip before use
xtensa: always use correct stack pointer for stack traces
xtensa: Fix checksum header file
xtensa: Fix architecture specific Kconfig
* 'kvm-updates/2.6.30' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: SVM: Remove port 80 passthrough
KVM: Make EFER reads safe when EFER does not exist
KVM: Fix NX support reporting
KVM: SVM: Fix cross vendor migration issue with unusable bit
After upgrading from gcc 4.2.2 to 4.4.0, the function graph tracer broke.
Investigating, I found that in the asm that replaces the return value,
gcc was using the same register for the old value as it was for the
new value.
mov (addr), old
mov new, (addr)
But if old and new are the same register, we clobber new with old!
I first thought this was a bug in gcc 4.4.0 and reported it:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40132
Andrew Pinski responded (quickly), saying that it was correct gcc behavior
and the code needed to denote old as an "early clobber".
Instead of "=r"(old), we need "=&r"(old).
[Impact: keep function graph tracer from breaking with gcc 4.4.0 ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
mmu.c needs to #include module.h to prevent these warnings:
arch/x86/xen/mmu.c:239: warning: data definition has no type or storage class
arch/x86/xen/mmu.c:239: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL'
arch/x86/xen/mmu.c:239: warning: parameter names (without types) in function declaration
[ Impact: cleanup ]
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Acked-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Commit 9b8de7479d ("FRV: Fix the section
attribute on UP DECLARE_PER_CPU()") cleaned up DECLARE/DEFINE_PER_CPU()
macros and in the process made alpha percpu.h include
include/asm-generic/percpu.h which breaks compilation due to duplicate
definitions.
Remove inclusion of generic asm helper file and define whatever necessary
in alpha header proper.
In the longer term, percpu definitions will be unified and all these
little subtlties will be removed.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: David Howells <dhowells@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
With the clkdev, musb_core.c needs to register clock with name "ick".
Once all the platforms using the musb driver have been converted
to use clockdev, the clock name does not need to be passed
from the low-level init code.
Signed-off-by: Tony Lindgren <tony@atomide.com>
SPI driver will do unhandled fault on OMAP2420 if trying to probe
non-existing SPI busses. Register those additional busses runtime only
for cpus having them.
Signed-off-by: Jarkko Nikula <jhnikula@gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Fix "tusb6010 init error 5, -19" and compilation warning from function
tusb6010_platform_retime "warning: 'sysclk_ps' is used uninitialized in this
function".
I suppose commit c094ba34b8f780885d029ce3c2715a194b780e5d was meant to test
for zero fclk_ps instead of sysclk_ps.
Signed-off-by: Jarkko Nikula <jarkko.nikula@nokia.com>
Cc: Roel Kluin <roel.kluin@gmail.com>
Tested-by: Kalle Valo <kalle.valo@iki.fi>
Signed-off-by: Tony Lindgren <tony@atomide.com>
GPIO de-bounce clocks don't have any impact on the module idle state, so
the clock code should not wait for the module to enable after the de-bounce
clocks are enabled.
Problem found by Kevin Hilman <khilman@deeprootsystems.com>.
Signed-off-by: Paul Walmsley <paul@pwsan.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Fix this build error when CONFIG_BLOCK is not set:
drivers/cdrom/viocd.c: In function 'viocd_blk_open':
drivers/cdrom/viocd.c:156: error: dereferencing pointer to incomplete type
drivers/cdrom/viocd.c: In function 'viocd_blk_release':
drivers/cdrom/viocd.c:162: error: dereferencing pointer to incomplete type
drivers/cdrom/viocd.c: In function 'viocd_blk_ioctl':
drivers/cdrom/viocd.c:170: error: dereferencing pointer to incomplete type
drivers/cdrom/viocd.c: In function 'viocd_blk_media_changed':
drivers/cdrom/viocd.c:176: error: dereferencing pointer to incomplete type
...
Signed-off-by: Alexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Fix resembles implementation from Marc Gauthier and Piet Denaly:
In the Xtensa architecture, assembly generates literals which must always
precede the code (the L32R instruction that loads them only uses negative
PC-relative offsets). For any *.text section, literals are placed in a
corresponding *.literal section. The linker script (vmlinux.lds) must
place these in the correct order. It must also combine them, when the
*.text section can be larger than L32R's 256 kB range.
For example, this doesn't work: *(.literal) *(.text) because L32R
instructions at the end of .text can't reach the literals.
The linker can solve this if they are combined in parentheses, like this:
*(.literal .text)
because it is now allowed mix literals in .text to bring them in range.
None of this is done by standard vmlinux.lds.h macros such as TEXT_TEXT
and INIT_TEXT. To avoid replicating the logic of that header file, we
instead post-process the generated linker script to convert *(xxx.text)
to *(xxx.literal xxx.text) for the following text sections:
.text .ref.text .*init.text .*exit.text .text.*
using a sed script. To do this we must override the default rule for
vmlinux.lds (see scripts/Makefile.build and the top-level Makefile)
to insert this extra step.
Signed-off-by: Marc Gauthier <marc@tensilica.com>
Signed-off-by: Pete Delaney <piet@tensilica.com>
Signed-off-by: Chris Zankel <chris@zankel.net>
The previous patch enabled ccount calibration for the s6000 variant.
This patch updates the defconfig for the s6105 platform to reflect this
change.
Signed-off-by: Oskar Schirmer <os@emlix.com>
Signed-off-by: Chris Zankel <chris@zankel.net>
Calculate core frequency from timers at boot time
instead of assuming a fixed frequency. This is
useful as the true frequency is set up by the
boot loader, thus variable.
Signed-off-by: Oskar Schirmer <os@emlix.com>
Signed-off-by: Chris Zankel <chris@zankel.net>
The variable ccount_nsec has been renamed to nsec_per_ccount
in arch/xtensa/kernel/time.c in 2b8aea74 (2007-08-05),
but the fix failed to rename the variable in
arch/xtensa/include/asm/timex.h as well.
Signed-off-by: Oskar Schirmer <os@emlix.com>
Signed-off-by: Chris Zankel <chris@zankel.net>
Platform initialization sets up the LED heartbeat that is controlled
via GPIO. Requesting the GPIO pins fails, however, as the chip is
only initialized later by a device_initcall().
Fix this up by exporting the initialization function. Let the
platform set up the chip before it starts using it.
Signed-off-by: Johannes Weiner <jw@emlix.com>
Signed-off-by: Chris Zankel <chris@zankel.net>
Commit '28a0ce7 xtensa: use correct stack pointer for stack traces'
changed the stack tracer from always reading the stack pointer
register to always using the saved value in the task descriptor.
The author was too dense to consider the fact that the saved stack
value is stale for a running process und thus unusable for 'current'.
What we do now is to use the stack pointer register (a1) for when the
task is unknown - we can't help it then - or when the task is
'current'. For everything else use the saved stack pointer value
contained in the task descriptor.
Signed-off-by: Johannes Weiner <jw@emlix.com>
Signed-off-by: Chris Zankel <chris@zankel.net>
We need to add a "memory" dependency (barrier) in assembly macros
that access (read or write) memory. Otherwise, the compiler might
ill-optimize the order of memory accesses.
Signed-off-by: Chris Zankel <chris@zankel.net>
Move a misplace endmenu marker to enable platform options and
disable PCI and automatic calibrating for the XT2K board. The
on-board PCI bridge is somewhat broken, anyway, and the
calibrating relies on some whacky usage of the serial port.
Signed-off-by: Chris Zankel <chris@zankel.net>
Use ®s->sp instead of regs for getting the top of stack in kernel mode.
(on x86-64, regs->sp always points the top of stack)
[ Impact: Oprofile decodes only stack for backtracing on i386 ]
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
[ v2: rename the API to kernel_stack_pointer(), move variable inside ]
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: systemtap@sources.redhat.com
Cc: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Jan Blunck <jblunck@suse.de>
Cc: Christoph Hellwig <hch@infradead.org>
LKML-Reference: <20090511210300.17332.67549.stgit@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Print reserved memory only if it was actually reserved.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
- the byte operand constraints were wrong for 32-bit
- the to-op's input operands weren't properly parenthesized
[ Impact: fix possible miscompilation or build failure ]
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
KVM optimizes guest port 80 accesses by passthing them through to the host.
Some AMD machines die on port 80 writes, allowing the guest to hard-lock the
host.
Remove the port passthrough to avoid the problem.
Cc: stable@kernel.org
Reported-by: Piotr Jaroszyński <p.jaroszynski@gmail.com>
Tested-by: Piotr Jaroszyński <p.jaroszynski@gmail.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
found one system where cpu address line is 44bits, mtrr printout
is not right:
[ 0.000000] MTRR variable ranges enabled:
[ 0.000000] 0 base 0 00000000 mask FF0 00000000 write-back
[ 0.000000] 1 base 10 00000000 mask FFF 80000000 write-back
[ 0.000000] 2 base 0 80000000 mask FFF 80000000 uncachable
[ 0.000000] 3 base 0 7F800000 mask FFF FF800000 uncachable
Li Zefan and Frederic pointed out the high_width could be -4 some how.
It turns out when phys_addr is 44bit, size_or_mask will be
ffffffff,00000000 so ffs(size_or_mask) will be 0.
Try to check low 32 bit, to get correct high_width.
Signed-off-by: Yinghai Lu <yinghai@kerne.org>
Also-analyzed-by: Frederic Weisbecker <fweisbec@gmail.com>
Also-analyzed-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Zhaolei <zhaolei@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
LKML-Reference: <4A026540.8060504@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Some processors don't have EFER; don't oops if userspace wants us to
read EFER when we check NX.
Cc: stable@kernel.org
Signed-off-by: Avi Kivity <avi@redhat.com>
AMDs VMCB does not have an explicit unusable segment descriptor field,
so we emulate it by using "not present". This has to be setup before
the fixups, because this field is used there.
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (26 commits)
bonding: fix panic if initialization fails
IXP4xx: complete Ethernet netdev setup before calling register_netdev().
IXP4xx: use "ENODEV" instead of "ENOSYS" in module initialization.
ipvs: Fix IPv4 FWMARK virtual services
ipv4: Make INET_LRO a bool instead of tristate.
net: remove stale reference to fastroute from Kconfig help text
net: update skb_recycle_check() for hardware timestamping changes
bnx2: Fix panic in bnx2_poll_work().
net-sched: fix bfifo default limit
igb: resolve panic on shutdown when SR-IOV is enabled
wimax: oops: wimax_dev_add() is the only one that can initialize the state
wimax: fix oops if netlink fails to add attribute
Bluetooth: Move dev_set_name() to a context that can sleep
netfilter: ctnetlink: fix wrong message type in user updates
netfilter: xt_cluster: fix use of cluster match with 32 nodes
netfilter: ip6t_ipv6header: fix match on packets ending with NEXTHDR_NONE
netfilter: add missing linux/types.h include to xt_LED.h
mac80211: pid, fix memory corruption
mac80211: minstrel, fix memory corruption
cfg80211: fix comment on regulatory hint processing
...
* 'fixes-for-linus' of git://git.monstr.eu/linux-2.6-microblaze:
microblaze: Fix return value for sys_ipc
microblaze: Storage class should be before const qualifier
