Commit graph

64,840 commits

Author SHA1 Message Date
Ivan Kokshaysky
74641f584d alpha: binfmt_aout fix
This fixes the problem introduced by commit 3bfacef412 (get rid of
special-casing the /sbin/loader on alpha): osf/1 ecoff binary segfaults
when binfmt_aout built as module.  That happens because aout binary
handler gets on the top of the binfmt list due to late registration, and
kernel attempts to execute the binary without preparatory work that must
be done by binfmt_loader.

Fixed by changing the registration order of the default binfmt handlers
using list_add_tail() and introducing insert_binfmt() function which
places new handler on the top of the binfmt list.  This might be generally
useful for installing arch-specific frontends for default handlers or just
for overriding them.

Signed-off-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Cc: Richard Henderson <rth@twiddle.net
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-05-02 15:36:10 -07:00
Ivan Kokshaysky
77b4cf5cb0 alpha: futex implementation
Signed-off-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Cc: Richard Henderson <rth@twiddle.net
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-05-02 15:36:10 -07:00
Ivan Kokshaysky
08a42e86bc alpha: exception table sorting
Exception fixups for sections other than .text (like one in futex_init())
break the natural ordering of fixup entries, so sorting is required.

Without that the result of the exception table search depends on phase of
the moon.

Signed-off-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Cc: Richard Henderson <rth@twiddle.net
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-05-02 15:36:10 -07:00
Ivan Kokshaysky
1ffb1c0c64 alpha: titan and marvel build fixes
These platforms got broken after u64 => 'long long' conversion.

Apparently that change was compile-tested with 'make allmodconfig', but it
doesn't include systems that depend on !ALPHA_LEGACY_START_ADDRESS.

Signed-off-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Cc: Richard Henderson <rth@twiddle.net
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-05-02 15:36:10 -07:00
Hirokazu Takata
9cd67243de m32r: use __stringify() macro in assembler.h
Replace #x with __stringify(x).
Also, #ifndef __STR is removed and undefine __STR macro at the beginning.

The __STR() macro is still remained, because the assembler.h might be
included from assembly codes as well as C codes.

Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
2009-05-02 22:38:21 +09:00
Hirokazu Takata
9c88b06dc7 m32r: build fix for __stringify macro
This patch fixes the following build error of 2.6.30-rc3-git2:

      AS      arch/m32r/kernel/head.o
    In file included from /include/linux/init.h:7,
                     from /arch/m32r/kernel/head.S:11:
    /include/linux/stringify.h:9: error: syntax error in macro parameter list
    /include/linux/stringify.h:10: error: syntax error in macro parameter list

This build error was caused at __HEAD macro in arch/m32r/kernel/head.S,
which uses __stringify() macro.

Remove -traditional option from EXTRA_AFLAGS for the m32r,
because the __stringify() macro depends on the gcc's variadic macro
extension function, due to commit:

    Make __stringify support variable argument macros too
    commit: 8f7c2c3731

Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
2009-05-02 21:43:26 +09:00
Cyrill Gorcunov
6f0aced639 x86, apic: use pr_ macro
Replace recenly appeared printk with pr_ macro
(the file already use a lot of them).

[ Impact: cleanup ]

Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
LKML-Reference: <20090501195425.GB4633@lenovo>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-01 21:56:25 +02:00
Thomas Gleixner
f9a196b8dc x86: initialize io_bitmap_base on 32bit
commit db949bba3c (x86-32: use non-lazy
io bitmap context switching) broke ioperm for 32bit because it removed
the lazy initialization of io_bitmap_base and did not set it to the
real bitmap offset.

[ Impact: fix non-working sys_ioperm() on 32-bit kernels ]

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2009-05-01 21:09:53 +02:00
Ingo Molnar
4420471f14 Merge branch 'x86/apic' into irq/numa
Conflicts:
	arch/x86/kernel/apic/io_apic.c

Merge reason: non-trivial interaction between ongoing work in io_apic.c
              and the NUMA migration feature in the irq tree.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-01 19:02:50 +02:00
Yinghai Lu
15e957d08d x86/irq: use move_irq_desc() in create_irq_nr()
move_irq_desc() will try to move irq_desc to the home node if
the allocated one is not correct, in create_irq_nr().

( This can happen on devices that are on different nodes that
  are using MSI, when drivers are loaded and unloaded randomly. )

v2: fix non-smp build
v3: add NUMA_IRQ_DESC to eliminate #ifdefs

[ Impact: improve irq descriptor locality on NUMA systems ]

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
LKML-Reference: <49F95EAE.2050903@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-01 19:01:12 +02:00
Fenghua Yu
aed5d5f4c5 Fix !CONFIG_DMAR build failure introduced by Intel IOMMU Pass Through Support
This updated patch should fix the compiling errors and remove the extern
iommu_pass_through from drivers/pci/intel-iommu.c file.

Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2009-05-01 16:44:47 +01:00
Michal Simek
c71a7a3875 microblaze: Fix return value for sys_ipc
Signed-off-by: John Linn <john.linn@xilinx.com>
Signed-off-by: Michal Simek <monstr@monstr.eu>
2009-05-01 16:38:14 +02:00
Tobias Klauser
595e8fabea microblaze: Storage class should be before const qualifier
The C99 specification states in section 6.11.5:

The placement of a storage-class specifier other than at the
beginning of the declaration specifiers in a declaration is an
obsolescent feature.

Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Michal Simek <monstr@monstr.eu>
2009-05-01 16:38:13 +02:00
Mark Brown
e3980b6a03 [ARM] SMDK6410: Hook regulator control of VDDARM up for WM1190-EV1
This allows the S3C CPUfreq driver to do DVFS.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 15:03:26 +01:00
Ben Dooks
dc8fc7edef [ARM] S3C2412: Add missing cache flush in suspend code
The alterations to the suspend code missed adding a
call to the cache flushing routines during the suspend
path of the S3C2412.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 12:28:16 +01:00
Ben Dooks
5ef316fbe7 [ARM] S3C64XX: Add S3C64XX_PA_IIS{0,1} to <mach/map.h>
Add the physical address of the two I2S channel register blocks.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
2009-05-01 12:28:16 +01:00
Peter Zijlstra
63a809a2dc perf_counter: fix nmi-watchdog interaction
When we don't have any perf-counters active, don't act like we know
what the NMI is for.

[ Impact: fix hard hang with nmi_watchdog=2 ]

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <20090501102533.109867793@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-01 13:23:44 +02:00
Ben Dooks
97c1b14523 [ARM] S3C: Move DMA channel management code to plat-s3c
Change the name of S3C2410_DMA_CHANNELS to S3C_DMA_CHANNELS in the process.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:39:07 +01:00
Ben Dooks
20934cdbaa [ARM] S3C24XX: dma.h should not include <mach/hardware.h>
The dma.h header does not need to include <mach/hardware.h> as there
is nothing it depends on. Remove the include and fixup the couple of
places where <mach/map.h> was not explicitly included (dragged in
via <mach/hardware.h>)

Note, we have to fix arch/arm/plat-s3c24xx/pm.c which seems to rely
on having <mach/hardware.h> included by this method.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:39:07 +01:00
Ben Dooks
c133c29011 [ARM] S3C: Make common <plat/dma.h> header
Move the common information from the <mach/dma.h> header into a new
shareable <plat/dma.h> header for re-use on the s3c64xx.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:39:07 +01:00
Ben Dooks
023b40cd10 [ARM] S3C24XX: Fix indentation in <plat/dma-regs.h>
The <plat/dma-regs.h> pre-date the invention of the TAB character,
so fix the indentation of the register defines.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:39:06 +01:00
Ben Dooks
44dc94045f [ARM] S3C24XX: DMA: Split hardware regs out of <mach/dma.h>
The <mach/dma.h> is for the driver API for the DMA system and should
not have anything relying on the CPU specific registers.

Remove the registers to <plat/dma-regs.h> for the code that really
needs to know about them.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:39:06 +01:00
Ben Dooks
8970ef47d5 [ARM] S3C24XX: Remove hardware specific registers from DMA
calls

The S3C24XX DMA API channel configuration registers are being passed
values comprised of register values which makes it hard to move the
API to cover both the S3C24XX and S3C64XX.

These values can be calculated from knowing which device the channel
is connected to, so remove them from the two calls s3c2410_dma_config
and s3c2410_dma_devconfig.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:39:06 +01:00
Ben Dooks
5a9eb8da8b [ARM] S3C24XX: Move plat/dma.h
Move the platform dma.h to dma-plat.h to ensure it doen't get
confused with plat/dma.h

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:39:06 +01:00
Kwangwoo Lee
1f26a8a0fd [PATCH] S3C64XX: Basic A&W6410 board support patch V2
A&W6410 board uses Samsung S3C6410 SoC and it is registered with 2183.
Framebuffer and ethernet devices are supported currently.

Unnecessary header file is removed.
Fix to use __raw_writel() and __raw_readl() for accessing mapped address.

Thank you for your comments.

Signed-off-by: Kwangwoo Lee <kwangwoo.lee@gmail.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:38:46 +01:00
Mark Brown
f53aee29be [ARM] SMDK6410: Mark all supplies as always_on
Since no consumers are currently configured for the SMDK6410 mark all
the supplies on the board as being always_on, ensuring interoperability
with future regulator API changes to disable unused regulators.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:38:46 +01:00
Mark Brown
db9256f33c [ARM] SMDK6410: Use active high IRQ for the WM8350 on the WM1190-EV1
Using an active high IRQ ensures that the WM8350 interrupt handling
does not spin when used with a SMDK6410 which has not had R20 removed
and R21 fitted to connect EINT12 to the PMIC module rather thant the
fixed regulators on the CPU module.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:38:46 +01:00
Kyungmin Park
87c4122f17 [ARM] S3C6410: Basic support for NCP board (v2)
This is a basic support for NCP board based on s3c6410.

Only enables the serial. also remove empty i2c device.

Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:38:46 +01:00
Mark Brown
ecc558acab [ARM] SMDK6410: Support WM1190-EV1 PMIC board
The SMDK6410 supports pluggable PMIC boards. One such board is the
Wolfson Microelectronics 1190-EV1 for the WM8350 PMICs. This patch
introduces initial support for this module. Further patches will
add additional integration with the system as support for the
S3C6410 and other devices on the system improves.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:38:45 +01:00
Mark Brown
3056ea0afb [ARM] SMDK6410: Add support for SMSC9115 ethernet controller
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:38:45 +01:00
Ben Dooks
d6662c351a [ARM] SMDK6410: Ensure LCD settings are setup
Ensure that the LCD output type is RGB and that
the modem interface is not bypassing the LCD
block. This ensures the LCD interface output
gets to the pins in the correct format.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-05-01 11:38:45 +01:00
Sam Ravnborg
c520f72172 avr32: drop unused CLEAN_FILES
Left over from before we used arch/avr32/include

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
2009-05-01 10:54:00 +02:00
Benjamin Herrenschmidt
0203d6ec4e powerpc: Fix setting of oprofile cpu type
commit 2657dd4e30 introduced a
bug where we would now always override the "real" oprofile CPU
type with the "compatible" one provided by a pseudo-PVR in the
device-tree which is incorrect and breaks oprofile on all current
configs since the "compatible" ones aren't yet recognized.

This fixes it.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2009-05-01 15:12:05 +10:00
Michael Wolf
79af6c49a9 powerpc adjust oprofile_cpu_type version 3
Oprofile is changing the naming it is using for the compatibility modes.
Instead of having compat-power<x>, oprofile will go to family naming
convention and use ibm-compat-v<x>.  Currently only ibm-compat-v1 will
be defined.
The notion of compatibility events just started with POWER6. So there is
no way that any other tool could exist that is using these
oprofile_cpu_type strings we want to change.

Signed-off-by: Mike Wolf <mjw@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2009-05-01 15:12:05 +10:00
Thomas Gleixner
3c56999eec Merge branch 'core/signal' into perfcounters/core
This is necessary to avoid the conflict of syscall numbers.

Conflicts:
	arch/x86/ia32/ia32entry.S
	arch/x86/include/asm/unistd_32.h
	arch/x86/include/asm/unistd_64.h

Fixes up the borked syscall numbers of perfcounters versus
preadv/pwritev as well.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2009-04-30 21:16:49 +02:00
Catalin Marinas
0516e4643c [ARM] 5489/1: ARM errata: Data written to the L2 cache can be overwritten with stale data
This patch is a workaround for the 460075 Cortex-A8 (r2p0) erratum. It
configures the L2 cache auxiliary control register so that the Write
Allocate mode for the L2 cache is disabled.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2009-04-30 20:13:00 +01:00
Catalin Marinas
855c551f5b [ARM] 5490/1: ARM errata: Processor deadlock when a false hazard is created
This patch adds a workaround for the 458693 Cortex-A8 (r2p0)
erratum. It sets the corresponding bits in the auxiliary control
register so that the PLD instruction becomes a NOP.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2009-04-30 20:12:59 +01:00
Catalin Marinas
7ce236fcd6 [ARM] 5487/1: ARM errata: Stale prediction on replaced interworking branch
This patch adds the workaround for the 430973 Cortex-A8 (r1p0..r1p2)
erratum. The BTAC/BTB is now flushed at every context switch.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2009-04-30 20:12:50 +01:00
Catalin Marinas
9cba3ccc8f [ARM] 5488/1: ARM errata: Invalidation of the Instruction Cache operation can fail
This patch implements the recommended workaround for erratum 411920
(ARM1136, ARM1156, ARM1176).

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2009-04-30 20:12:47 +01:00
Remis Lima Baima
bf293c17b2 x86: added 'ifndef _ASM_X86_IOMAP_H' to iomap.h
iomap.h misses the include guards.

[ Impact: cleanup ]

Signed-off-by: Remis Lima Baima <remis.developer@googlemail.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
LKML-Reference: <200904301836.23885.arnd@arndb.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-30 19:26:11 +02:00
Thomas Gleixner
12d161147f x86: hookup sys_rt_tgsigqueueinfo
Make the new sys_rt_tgsigqueueinfo available for x86.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2009-04-30 19:24:24 +02:00
Sam Ravnborg
83c4832683 x86: boot/compressed/vmlinux.lds.S: fix build of bzImage with 64 bit compiler
Jesper reported that he saw following build issue:

 > ld:arch/x86/boot/compressed/vmlinux.lds:9: syntax error
 > make[2]: *** [arch/x86/boot/compressed/vmlinux] Error 1
 > make[1]: *** [arch/x86/boot/compressed/vmlinux] Error 2
 > make: *** [bzImage] Error 2

CPP defines the symbol "i386" to "1".
Undefine this to fix it.

[ Impact: build fix with certain tool chains ]

Reported-by: Jesper Dangaard Brouer <jdb@comx.dk>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
LKML-Reference: <alpine.LFD.2.00.0904260958190.3101@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-30 12:36:50 +02:00
John Wright
2f65dd475c x86: gettimeofday() vDSO: fix segfault when tv == NULL
According to the gettimeofday(2) manual:

       If either tv or tz is NULL, the corresponding structure is not
       set or returned.

Since it is legal to give NULL as the tv argument, the code should make
sure tv is not NULL before trying to dereference it.

This issue manifests itself on x86_64 when vdso=0 is not on the kernel
command-line and libc uses the vDSO for gettimeofday() (e.g. glibc >=
2.7).  A simple reproducer:

  #include <stdio.h>
  #include <sys/time.h>

  int main(void)
  {
      struct timezone tz;

      gettimeofday(NULL, &tz);

      return 0;
  }

See http://bugs.debian.org/466491 for more details.

[ Impact: fix gettimeofday(NULL, &tz) segfault ]

Signed-off-by: John Wright <john.wright@hp.com>
Cc: Andi Kleen <ak@suse.de>
Cc: John Wright <john.wright@hp.com>
LKML-Reference: <1241037121-14805-1-git-send-email-john.wright@hp.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-30 12:31:45 +02:00
Pekka Enberg
9518e0e435 x86: move per-cpu mmu_gathers to mm/init.c
[ Impact: cleanup ]

Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
LKML-Reference: <1240923650.1982.22.camel@penberg-laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-30 10:12:37 +02:00
Pekka Enberg
2b72394e40 x86: move max_pfn_mapped and max_low_pfn_mapped to setup.c
This patch moves the max_pfn_mapped and max_low_pfn_mapped global
variables to kernel/setup.c where they're initialized.

[ Impact: cleanup ]

Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
LKML-Reference: <1240923649.1982.21.camel@penberg-laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-30 10:12:36 +02:00
Magnus Damm
f425752fc6 sh: remove old CMT driver
This patch removes the old CMT driver (CONFIG_SH_CMT/timer-cmt.c)

As replacement, select the sh_cmt driver with CONFIG_SH_TIMER_CMT
and configure timer channel using platform data.

If multiple CMT channels are enabled using platform data, use the
earlytimer parameter on the kernel command line to select channel.
For instance, use "earlytimer=sh_cmt.0" to select the first channel.

To verify which timer is being used, look at printouts or the timer
irq count in /proc/interrupts.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-04-30 13:21:14 +09:00
Magnus Damm
698aa99da5 sh: sh2/sh2a 16-bit CMT platform data
This patch adds 16-bit cmt platform data for the following cpus:
 - sh7619 (2 channels)
 - sh7203/sh7263 (2 channels)
 - sh7206 (2 channels)

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-04-30 13:20:37 +09:00
Magnus Damm
e6be3a2586 sh: pass through ioremap() for non-mmu processors.
All 32-bit SuperH processors currently go through __ioremap_mode()
and check for IO_TRAPPED and directly mapped segments. With this
patch we simplify the MMU less case with a pass through version of
 __ioremap_mode() which just returns the physical address.

The effects of this is change are:
 - fix non-MMU ioremap() of high address hardware blocks (sh7203 CMT)
 - make sure IO_TRAPPED is not selected

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-04-30 12:56:37 +09:00
Robert Richter
43f6201a22 perf_counter, x86: rename bitmasks to ->used_mask and ->active_mask
Standardize on explicitly mentioning '_mask' in fields that
are not plain flags but masks. This avoids typos like:

       if (cpuc->used)

(which could easily slip through review unnoticed), while if a
typo looks like this:

       if (cpuc->used_mask)

it might get noticed during review.

[ Impact: cleanup ]

Signed-off-by: Robert Richter <robert.richter@amd.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1241016956-24648-1-git-send-email-robert.richter@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-29 22:19:36 +02:00
Paul Mackerras
ab7ef2e50a perf_counter: powerpc: allow use of limited-function counters
POWER5+ and POWER6 have two hardware counters with limited functionality:
PMC5 counts instructions completed in run state and PMC6 counts cycles
in run state.  (Run state is the state when a hardware RUN bit is 1;
the idle task clears RUN while waiting for work to do and sets it when
there is work to do.)

These counters can't be written to by the kernel, can't generate
interrupts, and don't obey the freeze conditions.  That means we can
only use them for per-task counters (where we know we'll always be in
run state; we can't put a per-task counter on an idle task), and only
if we don't want interrupts and we do want to count in all processor
modes.

Obviously some counters can't go on a limited hardware counter, but there
are also situations where we can only put a counter on a limited hardware
counter - if there are already counters on that exclude some processor
modes and we want to put on a per-task cycle or instruction counter that
doesn't exclude any processor mode, it could go on if it can use a
limited hardware counter.

To keep track of these constraints, this adds a flags argument to the
processor-specific get_alternatives() functions, with three bits defined:
one to say that we can accept alternative event codes that go on limited
counters, one to say we only want alternatives on limited counters, and
one to say that this is a per-task counter and therefore events that are
gated by run state are equivalent to those that aren't (e.g. a "cycles"
event is equivalent to a "cycles in run state" event).  These flags
are computed for each counter and stored in the counter->hw.counter_base
field (slightly wonky name for what it does, but it was an existing
unused field).

Since the limited counters don't freeze when we freeze the other counters,
we need some special handling to avoid getting skew between things counted
on the limited counters and those counted on normal counters.  To minimize
this skew, if we are using any limited counters, we read PMC5 and PMC6
immediately after setting and clearing the freeze bit.  This is done in
a single asm in the new write_mmcr0() function.

The code here is specific to PMC5 and PMC6 being the limited hardware
counters.  Being more general (e.g. having a bitmap of limited hardware
counter numbers) would have meant more complex code to read the limited
counters when freezing and unfreezing the normal counters, with
conditional branches, which would have increased the skew.  Since it
isn't necessary for the code to be more general at this stage, it isn't.

This also extends the back-ends for POWER5+ and POWER6 to be able to
handle up to 6 counters rather than the 4 they previously handled.

Signed-off-by: Paul Mackerras <paulus@samba.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Robert Richter <robert.richter@amd.com>
LKML-Reference: <18936.19035.163066.892208@cargo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-29 14:58:35 +02:00