linux-2.6
16 years agoMerge branch 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 28 Dec 2008 20:27:58 +0000 (12:27 -0800)] 
Merge branch 'sched-core-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (31 commits)
  sched: fix warning in fs/proc/base.c
  schedstat: consolidate per-task cpu runtime stats
  sched: use RCU variant of list traversal in for_each_leaf_rt_rq()
  sched, cpuacct: export percpu cpuacct cgroup stats
  sched, cpuacct: refactoring cpuusage_read / cpuusage_write
  sched: optimize update_curr()
  sched: fix wakeup preemption clock
  sched: add missing arch_update_cpu_topology() call
  sched: let arch_update_cpu_topology indicate if topology changed
  sched: idle_balance() does not call load_balance_newidle()
  sched: fix sd_parent_degenerate on non-numa smp machine
  sched: add uid information to sched_debug for CONFIG_USER_SCHED
  sched: move double_unlock_balance() higher
  sched: update comment for move_task_off_dead_cpu
  sched: fix inconsistency when redistribute per-cpu tg->cfs_rq shares
  sched/rt: removed unneeded defintion
  sched: add hierarchical accounting to cpu accounting controller
  sched: include group statistics in /proc/sched_debug
  sched: rename SCHED_NO_NO_OMIT_FRAME_POINTER => SCHED_OMIT_FRAME_POINTER
  sched: clean up SCHED_CPUMASK_ALLOC
  ...

16 years agoMerge branch 'tracing-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 28 Dec 2008 20:21:10 +0000 (12:21 -0800)] 
Merge branch 'tracing-core-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'tracing-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (241 commits)
  sched, trace: update trace_sched_wakeup()
  tracing/ftrace: don't trace on early stage of a secondary cpu boot, v3
  Revert "x86: disable X86_PTRACE_BTS"
  ring-buffer: prevent false positive warning
  ring-buffer: fix dangling commit race
  ftrace: enable format arguments checking
  x86, bts: memory accounting
  x86, bts: add fork and exit handling
  ftrace: introduce tracing_reset_online_cpus() helper
  tracing: fix warnings in kernel/trace/trace_sched_switch.c
  tracing: fix warning in kernel/trace/trace.c
  tracing/ring-buffer: remove unused ring_buffer size
  trace: fix task state printout
  ftrace: add not to regex on filtering functions
  trace: better use of stack_trace_enabled for boot up code
  trace: add a way to enable or disable the stack tracer
  x86: entry_64 - introduce FTRACE_ frame macro v2
  tracing/ftrace: add the printk-msg-only option
  tracing/ftrace: use preempt_enable_no_resched_notrace in ring_buffer_time_stamp()
  x86, bts: correctly report invalid bts records
  ...

Fixed up trivial conflict in scripts/recordmcount.pl due to SH bits
being already partly merged by the SH merge.

16 years agoMerge branch 'x86-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 28 Dec 2008 20:07:57 +0000 (12:07 -0800)] 
Merge branch 'x86-core-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'x86-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (246 commits)
  x86: traps.c replace #if CONFIG_X86_32 with #ifdef CONFIG_X86_32
  x86: PAT: fix address types in track_pfn_vma_new()
  x86: prioritize the FPU traps for the error code
  x86: PAT: pfnmap documentation update changes
  x86: PAT: move track untrack pfnmap stubs to asm-generic
  x86: PAT: remove follow_pfnmap_pte in favor of follow_phys
  x86: PAT: modify follow_phys to return phys_addr prot and return value
  x86: PAT: clarify is_linear_pfn_mapping() interface
  x86: ia32_signal: remove unnecessary declaration
  x86: common.c boot_cpu_stack and boot_exception_stacks should be static
  x86: fix intel x86_64 llc_shared_map/cpu_llc_id anomolies
  x86: fix warning in arch/x86/kernel/microcode_amd.c
  x86: ia32.h: remove unused struct sigfram32 and rt_sigframe32
  x86: asm-offset_64: use rt_sigframe_ia32
  x86: sigframe.h: include headers for dependency
  x86: traps.c declare functions before they get used
  x86: PAT: update documentation to cover pgprot and remap_pfn related changes - v3
  x86: PAT: add pgprot_writecombine() interface for drivers - v3
  x86: PAT: change pgprot_noncached to uc_minus instead of strong uc - v3
  x86: PAT: implement track/untrack of pfnmap regions for x86 - v3
  ...

16 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris...
Linus Torvalds [Sun, 28 Dec 2008 19:43:54 +0000 (11:43 -0800)] 
Merge branch 'for-linus' of git://git./linux/kernel/git/jmorris/security-testing-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6: (105 commits)
  SELinux: don't check permissions for kernel mounts
  security: pass mount flags to security_sb_kern_mount()
  SELinux: correctly detect proc filesystems of the form "proc/foo"
  Audit: Log TIOCSTI
  user namespaces: document CFS behavior
  user namespaces: require cap_set{ug}id for CLONE_NEWUSER
  user namespaces: let user_ns be cloned with fairsched
  CRED: fix sparse warnings
  User namespaces: use the current_user_ns() macro
  User namespaces: set of cleanups (v2)
  nfsctl: add headers for credentials
  coda: fix creds reference
  capabilities: define get_vfs_caps_from_disk when file caps are not enabled
  CRED: Allow kernel services to override LSM settings for task actions
  CRED: Add a kernel_service object class to SELinux
  CRED: Differentiate objective and effective subjective credentials on a task
  CRED: Documentation
  CRED: Use creds in file structs
  CRED: Prettify commoncap.c
  CRED: Make execve() take advantage of copy-on-write credentials
  ...

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
Linus Torvalds [Sun, 28 Dec 2008 19:43:22 +0000 (11:43 -0800)] 
Merge git://git./linux/kernel/git/herbert/crypto-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (57 commits)
  crypto: aes - Precompute tables
  crypto: talitos - Ack done interrupt in isr instead of tasklet
  crypto: testmgr - Correct comment about deflate parameters
  crypto: salsa20 - Remove private wrappers around various operations
  crypto: des3_ede - permit weak keys unless REQ_WEAK_KEY set
  crypto: sha512 - Switch to shash
  crypto: sha512 - Move message schedule W[80] to static percpu area
  crypto: michael_mic - Switch to shash
  crypto: wp512 - Switch to shash
  crypto: tgr192 - Switch to shash
  crypto: sha256 - Switch to shash
  crypto: md5 - Switch to shash
  crypto: md4 - Switch to shash
  crypto: sha1 - Switch to shash
  crypto: rmd320 - Switch to shash
  crypto: rmd256 - Switch to shash
  crypto: rmd160 - Switch to shash
  crypto: rmd128 - Switch to shash
  crypto: null - Switch to shash
  crypto: hash - Make setkey optional
  ...

16 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6
Linus Torvalds [Sun, 28 Dec 2008 19:41:32 +0000 (11:41 -0800)] 
Merge branch 'for-linus' of git://git./linux/kernel/git/tiwai/sound-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6: (367 commits)
  ALSA: ASoC: fix a typo in omp-pcm.c
  ASoC: Fix DSP formats in SSM2602 audio codec
  ASoC: Fix incorrect DSP format in OMAP McBSP DAI and affected drivers
  ALSA: hda: fix incorrect mixer index values for 92hd83xx
  ALSA: hda: dinput_mux check
  ALSA: hda - Add quirk for another HP dv7
  ALSA: ASoC - Add missing __devexit annotation to wm8350.c
  ALSA: ASoc: DaVinci: davinci-evm use dsp_b mode
  ALSA: ASoC: DaVinci: i2s, evm, pass same value to codec and cpu_dai
  ALSA: ASoC: tlv320aic3x add dsp_a
  ALSA: ASoC: DaVinci: document I2S limitations
  ALSA: ASoC: DaVinci: davinci-i2s clean up
  ALSA: ASoC: DaVinci: davinci-i2s clean up
  ALSA: ASoC: DaVinci: davinci-i2s add comments to explain polarity
  ALSA: ASoC: DaVinci: davinvi-evm, make requests explicit
  ALSA: ca0106 - disable 44.1kHz capture
  ALSA: ca0106 - Add missing card->private_data initialization
  ALSA: ca0106 - Check ac97 availability at PM
  ALSA: hda - Power up always when no jack detection is available
  ALSA: hda - Fix unused variable warnings in patch_sigmatel.c
  ...

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6
Linus Torvalds [Sun, 28 Dec 2008 19:39:19 +0000 (11:39 -0800)] 
Merge git://git./linux/kernel/git/lethal/sh-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6: (132 commits)
  sh: oprofile: Fix up the module build.
  sh: add UIO support for JPU on SH7722.
  serial: sh-sci: Fix up port pinmux for SH7366.
  sh: mach-rsk: Use uImage generation by default for rsk7201/7203.
  sh: mach-sh03: Fix up pata_platform build breakage.
  sh: enable deferred io LCDC on Migo-R
  video: sh_mobile_lcdcfb deferred io support
  video: deferred io with physically contiguous memory
  video: deferred io cleanup
  video: fix deferred io fsync()
  sh: add LCDC interrupt configuration to AP325 and Migo-R
  sh_mobile_lcdc: use FB_SYS helpers instead of FB_CFB
  sh: split coherent pages
  sh: dma: Kill off ISA DMA wrapper.
  sh: Conditionalize the code dumper on CONFIG_DUMP_CODE.
  sh: Kill off the unused SH_ALPHANUMERIC debug option.
  sh: Enable skipping of bss on debug platforms for sh32 also.
  doc: Update sh cpufreq documentation.
  sh: mrshpc_setup_windows() needs to be inline.
  serial: sh-sci: sci_poll_get_char() is only used by CONFIG_CONSOLE_POLL.
  ...

16 years agoMerge branch 'sched/urgent'; commit 'v2.6.28' into sched/core
Ingo Molnar [Thu, 25 Dec 2008 12:42:23 +0000 (13:42 +0100)] 
Merge branch 'sched/urgent'; commit 'v2.6.28' into sched/core

16 years agoMerge branches 'tracing/ftrace', 'tracing/hw-branch-tracing' and 'tracing/ring-buffer...
Ingo Molnar [Thu, 25 Dec 2008 12:11:00 +0000 (13:11 +0100)] 
Merge branches 'tracing/ftrace', 'tracing/hw-branch-tracing' and 'tracing/ring-buffer'; commit 'v2.6.28' into tracing/core

16 years agosched, trace: update trace_sched_wakeup()
Peter Zijlstra [Tue, 16 Dec 2008 07:07:03 +0000 (08:07 +0100)] 
sched, trace: update trace_sched_wakeup()

Impact: extend the wakeup tracepoint with the info whether the wakeup was real

Add the information needed to distinguish 'real' wakeups from 'false'
wakeups.

Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agoMerge branch 'x86/tsc' into tracing/core
Ingo Molnar [Thu, 25 Dec 2008 11:48:18 +0000 (12:48 +0100)] 
Merge branch 'x86/tsc' into tracing/core

Merge it to resolve this incidental conflict between the BTS fixes/cleanups
and changes in x86/tsc:

Conflicts:
arch/x86/kernel/cpu/intel.c

16 years agoMerge branches 'x86/pat2' and 'x86/fpu'; commit 'v2.6.28' into x86/core
Ingo Molnar [Thu, 25 Dec 2008 10:50:41 +0000 (11:50 +0100)] 
Merge branches 'x86/pat2' and 'x86/fpu'; commit 'v2.6.28' into x86/core

16 years agox86: traps.c replace #if CONFIG_X86_32 with #ifdef CONFIG_X86_32
Jaswinder Singh [Tue, 23 Dec 2008 16:20:11 +0000 (21:50 +0530)] 
x86: traps.c replace #if CONFIG_X86_32 with #ifdef CONFIG_X86_32

Impact: cleanup, avoid warning on X86_64

Fixes this warning on X86_64:

  CC      arch/x86/kernel/traps.o
  arch/x86/kernel/traps.c:695:5: warning: "CONFIG_X86_32" is not defined

Signed-off-by: Jaswinder Singh <jaswinder@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agoMerge branch 'topic/udev-id-rename' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:34 +0000 (11:40 +0100)] 
Merge branch 'topic/udev-id-rename' into to-push

16 years agoMerge branch 'topic/sound-core-fix' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:33 +0000 (11:40 +0100)] 
Merge branch 'topic/sound-core-fix' into to-push

16 years agoMerge branch 'topic/snd-hrtimer' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:32 +0000 (11:40 +0100)] 
Merge branch 'topic/snd-hrtimer' into to-push

16 years agoMerge branch 'topic/pcxhr-update' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:31 +0000 (11:40 +0100)] 
Merge branch 'topic/pcxhr-update' into to-push

16 years agoMerge branch 'topic/oxygen' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:30 +0000 (11:40 +0100)] 
Merge branch 'topic/oxygen' into to-push

16 years agoMerge branch 'topic/jack-mechanical' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:29 +0000 (11:40 +0100)] 
Merge branch 'topic/jack-mechanical' into to-push

16 years agoMerge branch 'topic/hda' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:28 +0000 (11:40 +0100)] 
Merge branch 'topic/hda' into to-push

16 years agoMerge branch 'topic/cs5535audio' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:28 +0000 (11:40 +0100)] 
Merge branch 'topic/cs5535audio' into to-push

16 years agoMerge branch 'topic/convert-tasklet' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:27 +0000 (11:40 +0100)] 
Merge branch 'topic/convert-tasklet' into to-push

16 years agoMerge branch 'topic/ca0106' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:26 +0000 (11:40 +0100)] 
Merge branch 'topic/ca0106' into to-push

16 years agoMerge branch 'topic/audigy-capture-boost' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:26 +0000 (11:40 +0100)] 
Merge branch 'topic/audigy-capture-boost' into to-push

16 years agoMerge branch 'topic/asoc' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:25 +0000 (11:40 +0100)] 
Merge branch 'topic/asoc' into to-push

16 years agoMerge branch 'topic/aoa' into to-push
Takashi Iwai [Thu, 25 Dec 2008 10:40:24 +0000 (11:40 +0100)] 
Merge branch 'topic/aoa' into to-push

16 years agotracing/ftrace: don't trace on early stage of a secondary cpu boot, v3
Frederic Weisbecker [Wed, 24 Dec 2008 22:30:02 +0000 (23:30 +0100)] 
tracing/ftrace: don't trace on early stage of a secondary cpu boot, v3

Impact: fix a crash/hard-reboot on certain configs while enabling cpu runtime

On some archs, the boot of a secondary cpu can have an early fragile state.
On x86-64, the pda is not initialized on the first stage of a cpu boot but
it is needed to get the cpu number and the current task pointer. This data
is needed during tracing. As they were dereferenced at this stage, we got a
crash while tracing a cpu being enabled at runtime.

Some other archs like ia64 can have such kind of issue too.

Changes on v2:

We dropped the previous solution of a per-arch called function to guess the
current state of a cpu. That could slow down the tracing.

This patch removes the -pg flag on arch/x86/kernel/cpu/common.c where
the low level cpu boot functions exist, on start_secondary() and a helper
function used at this stage.

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agoMerge branch 'next' into for-linus
James Morris [Thu, 25 Dec 2008 00:40:09 +0000 (11:40 +1100)] 
Merge branch 'next' into for-linus

16 years agocrypto: aes - Precompute tables
Herbert Xu [Thu, 25 Dec 2008 00:05:13 +0000 (11:05 +1100)] 
crypto: aes - Precompute tables

The tables used by the various AES algorithms are currently
computed at run-time.  This has created an init ordering problem
because some AES algorithms may be registered before the tables
have been initialised.

This patch gets around this whole thing by precomputing the tables.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: talitos - Ack done interrupt in isr instead of tasklet
Lee Nipper [Sat, 20 Dec 2008 06:09:25 +0000 (17:09 +1100)] 
crypto: talitos - Ack done interrupt in isr instead of tasklet

Previous commit for interrupt mitigation moved the done interrupt
acknowlegement from the isr to the talitos_done tasklet.
This patch moves the done interrupt acknowledgement back
into the isr so that done interrupts will always be acknowledged.
This covers the case for acknowledging interrupts for channel done processing
that has actually already been completed by the tasklet prior to fielding
a pending interrupt.

Signed-off-by: Lee Nipper <lee.nipper@freescale.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: testmgr - Correct comment about deflate parameters
Geert Uytterhoeven [Thu, 18 Dec 2008 06:17:46 +0000 (17:17 +1100)] 
crypto: testmgr - Correct comment about deflate parameters

The comment for the deflate test vectors says the winbits parameter is 11,
while the deflate module actually uses -11 (a negative window bits parameter
enables the raw deflate format instead of the zlib format).
Correct this, to avoid confusion about the format used.

Signed-off-by: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: salsa20 - Remove private wrappers around various operations
Harvey Harrison [Wed, 17 Dec 2008 05:53:49 +0000 (16:53 +1100)] 
crypto: salsa20 - Remove private wrappers around various operations

ROTATE -> rol32
XOR was always used with the same destination, use ^=
PLUS/PLUSONE use ++ or +=

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: des3_ede - permit weak keys unless REQ_WEAK_KEY set
Jarod Wilson [Wed, 17 Dec 2008 05:51:13 +0000 (16:51 +1100)] 
crypto: des3_ede - permit weak keys unless REQ_WEAK_KEY set

While its a slightly insane to bypass the key1 == key2 ||
key2 == key3 check in triple-des, since it reduces it to the
same strength as des, some folks do need to do this from time
to time for backwards compatibility with des.

My own case is FIPS CAVS test vectors. Many triple-des test
vectors use a single key, replicated 3x. In order to get the
expected results, des3_ede_setkey() needs to only reject weak
keys if the CRYPTO_TFM_REQ_WEAK_KEY flag is set.

Also sets a more appropriate RES flag when a weak key is found.

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: sha512 - Switch to shash
Adrian-Ken Rueegsegger [Wed, 17 Dec 2008 05:49:02 +0000 (16:49 +1100)] 
crypto: sha512 - Switch to shash

This patch changes sha512 and sha384 to the new shash interface.

Signed-off-by: Adrian-Ken Rueegsegger <ken@codelabs.ch>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: sha512 - Move message schedule W[80] to static percpu area
Adrian-Ken Rueegsegger [Wed, 17 Dec 2008 05:47:52 +0000 (16:47 +1100)] 
crypto: sha512 - Move message schedule W[80] to static percpu area

The message schedule W (u64[80]) is too big for the stack. In order
for this algorithm to be used with shash it is moved to a static
percpu area.

Signed-off-by: Adrian-Ken Rueegsegger <ken@codelabs.ch>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: michael_mic - Switch to shash
Adrian-Ken Rueegsegger [Sun, 7 Dec 2008 11:35:38 +0000 (19:35 +0800)] 
crypto: michael_mic - Switch to shash

This patch changes michael_mic to the new shash interface.

Signed-off-by: Adrian-Ken Rueegsegger <ken@codelabs.ch>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: wp512 - Switch to shash
Adrian-Ken Rueegsegger [Sun, 7 Dec 2008 11:34:37 +0000 (19:34 +0800)] 
crypto: wp512 - Switch to shash

This patch changes wp512, wp384 and wp256 to the new shash interface.

Signed-off-by: Adrian-Ken Rueegsegger <ken@codelabs.ch>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: tgr192 - Switch to shash
Adrian-Ken Rueegsegger [Wed, 3 Dec 2008 11:58:32 +0000 (19:58 +0800)] 
crypto: tgr192 - Switch to shash

This patch changes tgr192, tgr160 and tgr128 to the new shash interface.

Signed-off-by: Adrian-Ken Rueegsegger <ken@codelabs.ch>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: sha256 - Switch to shash
Adrian-Ken Rueegsegger [Wed, 3 Dec 2008 11:57:49 +0000 (19:57 +0800)] 
crypto: sha256 - Switch to shash

This patch changes sha256 and sha224 to the new shash interface.

Signed-off-by: Adrian-Ken Rueegsegger <ken@codelabs.ch>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: md5 - Switch to shash
Adrian-Ken Rueegsegger [Wed, 3 Dec 2008 11:57:12 +0000 (19:57 +0800)] 
crypto: md5 - Switch to shash

This patch changes md5 to the new shash interface.

Signed-off-by: Adrian-Ken Rueegsegger <ken@codelabs.ch>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: md4 - Switch to shash
Adrian-Ken Rueegsegger [Wed, 3 Dec 2008 11:55:27 +0000 (19:55 +0800)] 
crypto: md4 - Switch to shash

This patch changes md4 to the new shash interface.

Signed-off-by: Adrian-Ken Rueegsegger <ken@codelabs.ch>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: sha1 - Switch to shash
Adrian-Ken Rueegsegger [Tue, 2 Dec 2008 13:08:20 +0000 (21:08 +0800)] 
crypto: sha1 - Switch to shash

This patch changes sha1 to the new shash interface.

Signed-off-by: Adrian-Ken Rueegsegger <ken@codelabs.ch>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: rmd320 - Switch to shash
Herbert Xu [Sat, 8 Nov 2008 02:11:09 +0000 (10:11 +0800)] 
crypto: rmd320 - Switch to shash

This patch changes rmd320 to the new shash interface.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: rmd256 - Switch to shash
Herbert Xu [Sat, 8 Nov 2008 01:58:10 +0000 (09:58 +0800)] 
crypto: rmd256 - Switch to shash

This patch changes rmd256 to the new shash interface.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: rmd160 - Switch to shash
Herbert Xu [Sat, 8 Nov 2008 01:18:51 +0000 (09:18 +0800)] 
crypto: rmd160 - Switch to shash

This patch changes rmd160 to the new shash interface.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: rmd128 - Switch to shash
Herbert Xu [Sat, 8 Nov 2008 01:10:40 +0000 (09:10 +0800)] 
crypto: rmd128 - Switch to shash

This patch changes rmd128 to the new shash interface.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: null - Switch to shash
Herbert Xu [Sat, 8 Nov 2008 00:09:56 +0000 (08:09 +0800)] 
crypto: null - Switch to shash

This patch changes digest_null to the new shash interface.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hash - Make setkey optional
Herbert Xu [Sat, 8 Nov 2008 00:56:57 +0000 (08:56 +0800)] 
crypto: hash - Make setkey optional

Since most cryptographic hash algorithms have no keys, this patch
makes the setkey function optional for ahash and shash.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: testmgr - Validate output length in (de)compression tests
Geert Uytterhoeven [Fri, 28 Nov 2008 12:51:28 +0000 (20:51 +0800)] 
crypto: testmgr - Validate output length in (de)compression tests

When self-testing (de)compression algorithms, make sure the actual size of
the (de)compressed output data matches the expected output size.
Otherwise, in case the actual output size would be smaller than the expected
output size, the subsequent buffer compare test would still succeed, and no
error would be reported.

Signed-off-by: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: remove uses of __constant_{endian} helpers
Harvey Harrison [Fri, 28 Nov 2008 12:49:19 +0000 (20:49 +0800)] 
crypto: remove uses of __constant_{endian} helpers

Base versions handle constant folding just fine.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: testmgr - Fix error flow of test_comp
Ingo Molnar [Tue, 25 Nov 2008 15:19:24 +0000 (23:19 +0800)] 
crypto: testmgr - Fix error flow of test_comp

This warning:

  crypto/testmgr.c: In function â€˜test_comp’:
  crypto/testmgr.c:829: warning: â€˜ret’ may be used uninitialized in this function

triggers because GCC correctly notices that in the ctcount == 0 &&
dtcount != 0 input condition case this function can return an undefined
value, if the second loop fails.

Remove the shadowed 'ret' variable from the second loop that was probably
unintended.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hifn_795x - Use softirq kernel mapping in bh context
Evgeniy Polyakov [Tue, 25 Nov 2008 15:17:10 +0000 (23:17 +0800)] 
crypto: hifn_795x - Use softirq kernel mapping in bh context

Use KM_SOFTIRQ instead of KM_IRQ in tasklet context.
Added bug_on on input no-page condition.

Signed-off-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hifn_795x - Fix queue management
Evgeniy Polyakov [Mon, 24 Nov 2008 14:04:39 +0000 (22:04 +0800)] 
crypto: hifn_795x - Fix queue management

Fix queue management. Change ring size and perform its check not
one after another descriptor, but using stored pointers to the last
checked descriptors.

Signed-off-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hifn_795x - Remove some unused cruft
Patrick McHardy [Mon, 24 Nov 2008 14:03:37 +0000 (22:03 +0800)] 
crypto: hifn_795x - Remove some unused cruft

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hifn_795x - Fix queue processing
Patrick McHardy [Mon, 24 Nov 2008 14:02:55 +0000 (22:02 +0800)] 
crypto: hifn_795x - Fix queue processing

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hifn_795x - Fix request context corruption
Patrick McHardy [Mon, 24 Nov 2008 14:01:42 +0000 (22:01 +0800)] 
crypto: hifn_795x - Fix request context corruption

HIFN uses the transform context to store per-request data, which breaks
when more than one request is outstanding. Move per request members from
struct hifn_context to a new struct hifn_request_context and convert
the code to use this.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hifn_795x - Don't copy src sg list
Patrick McHardy [Mon, 24 Nov 2008 14:00:49 +0000 (22:00 +0800)] 
crypto: hifn_795x - Don't copy src sg list

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hifn_795x - Fix DMA setup
Patrick McHardy [Mon, 24 Nov 2008 13:59:25 +0000 (21:59 +0800)] 
crypto: hifn_795x - Fix DMA setup

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: ansi_cprng - fix inverted DT increment routine
Jarod Wilson [Mon, 24 Nov 2008 13:20:13 +0000 (21:20 +0800)] 
crypto: ansi_cprng - fix inverted DT increment routine

The ANSI X9.31 PRNG docs aren't particularly clear on how to increment DT,
but empirical testing shows we're incrementing from the wrong end. A 10,000
iteration Monte Carlo RNG test currently winds up not getting the expected
result.

From http://csrc.nist.gov/groups/STM/cavp/documents/rng/RNGVS.pdf :

# CAVS 4.3
# ANSI931 MCT
[X9.31]
[AES 128-Key]

COUNT = 0
Key = 9f5b51200bf334b5d82be8c37255c848
DT = 6376bbe52902ba3b67c925fa701f11ac
V = 572c8e76872647977e74fbddc49501d1
R = 48e9bd0d06ee18fbe45790d5c3fc9b73

Currently, we get 0dd08496c4f7178bfa70a2161a79459a after 10000 loops.

Inverting the DT increment routine results in us obtaining the expected result
of 48e9bd0d06ee18fbe45790d5c3fc9b73. Verified on both x86_64 and ppc64.

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agolibcrc32c: Select CRYPTO in Kconfig
Herbert Xu [Thu, 13 Nov 2008 14:05:13 +0000 (22:05 +0800)] 
libcrc32c: Select CRYPTO in Kconfig

Selecting CRYPTO_CRC32C is not enough as CRYPTO which CRYPTO_CRC32C
depends on may be disabled.  This patch adds the select on CRYPTO.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: ansi_cprng - Avoid incorrect extra call to _get_more_prng_bytes
Jarod Wilson [Thu, 13 Nov 2008 14:03:20 +0000 (22:03 +0800)] 
crypto: ansi_cprng - Avoid incorrect extra call to _get_more_prng_bytes

While working with some FIPS RNGVS test vectors yesterday, I discovered a
little bug in the way the ansi_cprng code works right now.

For example, the following test vector (complete with expected result)
from http://csrc.nist.gov/groups/STM/cavp/documents/rng/RNGVS.pdf ...

Key = f3b1666d13607242ed061cabb8d46202
DT = e6b3be782a23fa62d71d4afbb0e922fc
V = f0000000000000000000000000000000
R = 88dda456302423e5f69da57e7b95c73a

...when run through ansi_cprng, yields an incorrect R value
of e2afe0d794120103d6e86a2b503bdfaa.

If I load up ansi_cprng w/dbg=1 though, it was fairly obvious what was
going wrong:

----8<----
getting 16 random bytes for context ffff810033fb2b10
Calling _get_more_prng_bytes for context ffff810033fb2b10
Input DT: 00000000: e6 b3 be 78 2a 23 fa 62 d7 1d 4a fb b0 e9 22 fc
Input I: 00000000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
Input V: 00000000: f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
tmp stage 0: 00000000: e6 b3 be 78 2a 23 fa 62 d7 1d 4a fb b0 e9 22 fc
tmp stage 1: 00000000: f4 8e cb 25 94 3e 8c 31 d6 14 cd 8a 23 f1 3f 84
tmp stage 2: 00000000: 8c 53 6f 73 a4 1a af d4 20 89 68 f4 58 64 f8 be
Returning new block for context ffff810033fb2b10
Output DT: 00000000: e7 b3 be 78 2a 23 fa 62 d7 1d 4a fb b0 e9 22 fc
Output I: 00000000: 04 8e cb 25 94 3e 8c 31 d6 14 cd 8a 23 f1 3f 84
Output V: 00000000: 48 89 3b 71 bc e4 00 b6 5e 21 ba 37 8a 0a d5 70
New Random Data: 00000000: 88 dd a4 56 30 24 23 e5 f6 9d a5 7e 7b 95 c7 3a
Calling _get_more_prng_bytes for context ffff810033fb2b10
Input DT: 00000000: e7 b3 be 78 2a 23 fa 62 d7 1d 4a fb b0 e9 22 fc
Input I: 00000000: 04 8e cb 25 94 3e 8c 31 d6 14 cd 8a 23 f1 3f 84
Input V: 00000000: 48 89 3b 71 bc e4 00 b6 5e 21 ba 37 8a 0a d5 70
tmp stage 0: 00000000: e7 b3 be 78 2a 23 fa 62 d7 1d 4a fb b0 e9 22 fc
tmp stage 1: 00000000: 80 6b 3a 8c 23 ae 8f 53 be 71 4c 16 fc 13 b2 ea
tmp stage 2: 00000000: 2a 4d e1 2a 0b 58 8e e6 36 b8 9c 0a 26 22 b8 30
Returning new block for context ffff810033fb2b10
Output DT: 00000000: e8 b3 be 78 2a 23 fa 62 d7 1d 4a fb b0 e9 22 fc
Output I: 00000000: c8 e2 01 fd 9f 4a 8f e5 e0 50 f6 21 76 19 67 9a
Output V: 00000000: ba 98 e3 75 c0 1b 81 8d 03 d6 f8 e2 0c c6 54 4b
New Random Data: 00000000: e2 af e0 d7 94 12 01 03 d6 e8 6a 2b 50 3b df aa
returning 16 from get_prng_bytes in context ffff810033fb2b10
----8<----

The expected result is there, in the first "New Random Data", but we're
incorrectly making a second call to _get_more_prng_bytes, due to some checks
that are slightly off, which resulted in our original bytes never being
returned anywhere.

One approach to fixing this would be to alter some byte_count checks in
get_prng_bytes, but it would mean the last DEFAULT_BLK_SZ bytes would be
copied a byte at a time, rather than in a single memcpy, so a slightly more
involved, equally functional, and ultimately more efficient way of fixing this
was suggested to me by Neil, which I'm submitting here. All of the RNGVS ANSI
X9.31 AES128 VST test vectors I've passed through ansi_cprng are now returning
the expected results with this change.

Signed-off-by: Jarod Wilson <jarod@redhat.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: md4 - Use ARRAY_SIZE
Julia Lawall [Thu, 13 Nov 2008 13:19:04 +0000 (21:19 +0800)] 
crypto: md4 - Use ARRAY_SIZE

ARRAY_SIZE is more concise to use when the size of an array is divided by
the size of its type or the size of its first element.

The semantic patch that makes this change is as follows:
(http://www.emn.fr/x-info/coccinelle/)

// <smpl>
@i@
@@

#include <linux/kernel.h>

@depends on i using "paren.iso"@
type T;
T[] E;
@@

- (sizeof(E)/sizeof(T))
+ ARRAY_SIZE(E)
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agolibcrc32c: Add crc32c_le macro
Herbert Xu [Tue, 11 Nov 2008 04:20:06 +0000 (12:20 +0800)] 
libcrc32c: Add crc32c_le macro

The bnx2x driver actually uses the crc32c_le name so this patch
restores the crc32c_le symbol through a macro.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agolibcrc32c: Fix "crc32c undefined" compilation error
Adrian-Ken Rueegsegger [Tue, 11 Nov 2008 04:14:00 +0000 (12:14 +0800)] 
libcrc32c: Fix "crc32c undefined" compilation error

The latest shash changes leave crc32c undefined:

[...]
Building modules, stage 2.
  MODPOST 1381 modules
  ERROR: "crc32c" [net/sctp/sctp.ko] undefined!
  ERROR: "crc32c" [net/ipv4/netfilter/nf_nat_proto_sctp.ko] undefined!

Adding EXPORT_SYMBOL(crc32c) to lib/libcrc32c.c fixes the compile error.
This patch has been compile-tested only.

Signed-off-by: Adrian-Ken Rueegsegger <rueegsegger@swiss-it.ch>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agolibcrc32c: Move implementation to crypto crc32c
Herbert Xu [Fri, 7 Nov 2008 07:11:47 +0000 (15:11 +0800)] 
libcrc32c: Move implementation to crypto crc32c

This patch swaps the role of libcrc32c and crc32c.  Previously
the implementation was in libcrc32c and crc32c was a wrapper.
Now the code is in crc32c and libcrc32c just calls the crypto
layer.

The reason for the change is to tap into the algorithm selection
capability of the crypto API so that optimised implementations
such as the one utilising Intel's CRC32C instruction can be
used where available.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: crc32c - Test descriptor context format
Herbert Xu [Fri, 7 Nov 2008 06:58:52 +0000 (14:58 +0800)] 
crypto: crc32c - Test descriptor context format

This patch adds a test for the requirement that all crc32c algorithms
shall store the partial result in the first four bytes of the descriptor
context.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: crc32c-intel - Switch to shash
Herbert Xu [Thu, 6 Nov 2008 08:56:41 +0000 (16:56 +0800)] 
crypto: crc32c-intel - Switch to shash

This patch changes crc32c-intel to the new shash interface.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: crc32c - Switch to shash
Herbert Xu [Tue, 9 Sep 2008 07:23:07 +0000 (17:23 +1000)] 
crypto: crc32c - Switch to shash

This patch changes crc32c to the new shash interface.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hash - Export shash through hash
Herbert Xu [Sun, 31 Aug 2008 12:21:09 +0000 (22:21 +1000)] 
crypto: hash - Export shash through hash

This patch allows shash algorithms to be used through the old hash
interface.  This is a transitional measure so we can convert the
underlying algorithms to shash before converting the users across.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: api - Call type show function before legacy for proc
Herbert Xu [Thu, 6 Nov 2008 06:39:16 +0000 (14:39 +0800)] 
crypto: api - Call type show function before legacy for proc

This patch makes /proc/crypto call the type-specific show function
if one is present before calling the legacy show functions for
cipher/digest/compress.  This allows us to reuse the type values
for those legacy types.  In particular, hash and digest will share
one type value while shash is phased in as the default hash type.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hash - Add import/export interface
Herbert Xu [Sun, 2 Nov 2008 13:38:11 +0000 (21:38 +0800)] 
crypto: hash - Add import/export interface

It is often useful to save the partial state of a hash function
so that it can be used as a base for two or more computations.

The most prominent example is HMAC where all hashes start from
a base determined by the key.  Having an import/export interface
means that we only have to compute that base once rather than
for each message.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hash - Export shash through ahash
Herbert Xu [Sun, 31 Aug 2008 08:52:18 +0000 (18:52 +1000)] 
crypto: hash - Export shash through ahash

This patch allows shash algorithms to be used through the ahash
interface.  This is required before we can convert digest algorithms
over to shash.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hash - Add shash interface
Herbert Xu [Sun, 31 Aug 2008 05:47:27 +0000 (15:47 +1000)] 
crypto: hash - Add shash interface

The shash interface replaces the current synchronous hash interface.
It improves over hash in two ways.  Firstly shash is reentrant,
meaning that the same tfm may be used by two threads simultaneously
as all hashing state is stored in a local descriptor.

The other enhancement is that shash no longer takes scatter list
entries.  This is because shash is specifically designed for
synchronous algorithms and as such scatter lists are unnecessary.

All existing hash users will be converted to shash once the
algorithms have been completely converted.

There is also a new finup function that combines update with final.
This will be extended to ahash once the algorithm conversion is
done.

This is also the first time that an algorithm type has their own
registration function.  Existing algorithm types will be converted
to this way in due course.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: api - Rebirth of crypto_alloc_tfm
Herbert Xu [Sat, 20 Sep 2008 21:52:53 +0000 (06:52 +0900)] 
crypto: api - Rebirth of crypto_alloc_tfm

This patch reintroduces a completely revamped crypto_alloc_tfm.
The biggest change is that we now take two crypto_type objects
when allocating a tfm, a frontend and a backend.  In fact this
simply formalises what we've been doing behind the API's back.

For example, as it stands crypto_alloc_ahash may use an
actual ahash algorithm or a crypto_hash algorithm.  Putting
this in the API allows us to do this much more cleanly.

The existing types will be converted across gradually.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: api - Move type exit function into crypto_tfm
Herbert Xu [Sun, 14 Sep 2008 01:19:03 +0000 (18:19 -0700)] 
crypto: api - Move type exit function into crypto_tfm

The type exit function needs to undo any allocations done by the type
init function.  However, the type init function may differ depending
on the upper-level type of the transform (e.g., a crypto_blkcipher
instantiated as a crypto_ablkcipher).

So we need to move the exit function out of the lower-level
structure and into crypto_tfm itself.

As it stands this is a no-op since nobody uses exit functions at
all.  However, all cases where a lower-level type is instantiated
as a different upper-level type (such as blkcipher as ablkcipher)
will be converted such that they allocate the underlying transform
and use that instead of casting (e.g., crypto_ablkcipher casted
into crypto_blkcipher).  That will need to use a different exit
function depending on the upper-level type.

This patch also allows the type init/exit functions to call (or not)
cra_init/cra_exit instead of always calling them from the top level.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: ansi_cprng - Allow resetting of DT value
Neil Horman [Wed, 5 Nov 2008 04:13:14 +0000 (12:13 +0800)] 
crypto: ansi_cprng - Allow resetting of DT value

This is a patch that was sent to me by Jarod Wilson, marking off my
outstanding todo to allow the ansi cprng to set/reset the DT counter value in a
cprng instance.  Currently crytpo_rng_reset accepts a seed byte array which is
interpreted by the ansi_cprng as a {V key} tuple.  This patch extends that tuple
to now be {V key DT}, with DT an optional value during reset.  This patch also
fixes  a bug we noticed in which the offset of the key area of the seed is
started at DEFAULT_PRNG_KSZ rather than DEFAULT_BLK_SZ as it should be.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: padlock - Avoid resetting cword on successive operations
Herbert Xu [Sun, 31 Aug 2008 05:58:45 +0000 (15:58 +1000)] 
crypto: padlock - Avoid resetting cword on successive operations

Resetting the control word is quite expensive.  Fortunately this
isn't an issue for the common operations such as CBC and ECB as
the whole operation is done through a single call.  However, modes
such as LRW and XTS have to call padlock over and over again for
one operation which really hurts if each call resets the control
word.

This patch uses an idea by Sebastian Siewior to store the last
control word used on a CPU and only reset the control word if
that changes.

Note that any task switch automatically resets the control word
so we only need to be accurate with regard to the stored control
word when no task switches occur.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: crc32c-intel - Update copyright head
Kent Liu [Fri, 31 Oct 2008 08:52:58 +0000 (16:52 +0800)] 
crypto: crc32c-intel - Update copyright head

The original copyright head for crc32c-intel.c is incorrect. Please merge
the patch to update it.

Signed-Off-By: Kent Liu <kent.liu@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: talitos - Preempt overflow interrupts off-by-one fix
Vishnu Suresh [Mon, 20 Oct 2008 13:06:18 +0000 (21:06 +0800)] 
crypto: talitos - Preempt overflow interrupts off-by-one fix

In commit ec6644d6325b5a38525f1d5b20fd4bf7db05cf2a "crypto: talitos - Preempt
overflow interrupts", the test in atomic_inc_not_zero was interpreted by the
author to be applied after the increment operation (not before). This off-by-one
fix prevents overflow error interrupts from occurring when requests are frequent
and large enough to do so.

Signed-off-by: Vishnu Suresh <Vishnu@freescale.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: camellia - use kernel-provided bitops, unaligned access
Harvey Harrison [Sun, 12 Oct 2008 12:40:12 +0000 (20:40 +0800)] 
crypto: camellia - use kernel-provided bitops, unaligned access

Remove the private implementation of 32-bit rotation and unaligned
access with byteswapping.

As a bonus, fixes sparse warnings:
crypto/camellia.c:602:2: warning: cast to restricted __be32
crypto/camellia.c:603:2: warning: cast to restricted __be32
crypto/camellia.c:604:2: warning: cast to restricted __be32
crypto/camellia.c:605:2: warning: cast to restricted __be32
crypto/camellia.c:710:2: warning: cast to restricted __be32
crypto/camellia.c:711:2: warning: cast to restricted __be32
crypto/camellia.c:712:2: warning: cast to restricted __be32
crypto/camellia.c:713:2: warning: cast to restricted __be32
crypto/camellia.c:714:2: warning: cast to restricted __be32
crypto/camellia.c:715:2: warning: cast to restricted __be32
crypto/camellia.c:716:2: warning: cast to restricted __be32
crypto/camellia.c:717:2: warning: cast to restricted __be32

[Thanks to Tomoyuki Okazaki for spotting the typo]
Tested-by: Carlo E. Prelz <fluido@fluido.as>
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: testmgr - Trigger a panic when self test fails in FIPS mode
Neil Horman [Sun, 12 Oct 2008 12:36:51 +0000 (20:36 +0800)] 
crypto: testmgr - Trigger a panic when self test fails in FIPS mode

The FIPS specification requires that should self test for any supported
crypto algorithm fail during operation in fips mode, we need to prevent
the use of any crypto functionality until such time as the system can
be re-initialized.  Seems like the best way to handle that would be
to panic the system if we were in fips mode and failed a self test.
This patch implements that functionality.  I've built and run it
successfully.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: talitos - Perform auth check in h/w if on sec 2.1 and above
Kim Phillips [Sun, 12 Oct 2008 12:33:14 +0000 (20:33 +0800)] 
crypto: talitos - Perform auth check in h/w if on sec 2.1 and above

SEC version 2.1 and above adds the capability to do the IPSec ICV
memcmp in h/w. Results of the cmp are written back in the descriptor
header, along with the done status.  A new callback is added that
checks these ICCR bits instead of performing the memcmp on the core,
and is enabled by h/w capability.

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
After testing on different parts, another condition was added
before using h/w auth check because different
SEC revisions require different handling.

The SEC 3.0 allows a more flexible link table where
the auth data can span separate link table entries.
The SEC 2.4/2.1 does not support this case.
So a test was added in the decrypt routine
for a fragmented case; the h/w auth check is disallowed for
revisions not having the extent in the link table;
in this case the hw auth check is done by software.

A portion of a previous change for SEC 3.0 link table handling
was removed since it became dead code with the hw auth check supported.

This seems to be the best compromise for using hw auth check
on supporting SEC revisions; it keeps the link table logic
simpler for the fragmented cases.

Signed-off-by: Lee Nipper <lee.nipper@freescale.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: talitos - Implement done interrupt mitigation
Lee Nipper [Sun, 12 Oct 2008 12:29:34 +0000 (20:29 +0800)] 
crypto: talitos - Implement done interrupt mitigation

In talitos_interrupt, upon one done interrupt, mask further done interrupts,
and ack only any error interrupt.
In talitos_done, unmask done interrupts after completing processing.
In flush_channel, ack each done channel processed.
Keep done overflow interrupts masked because even though each pkt
is ack'ed, a few done overflows still occur.

Signed-off-by: Lee Nipper <lee.nipper@freescale.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: talitos - Pass correct interrupt status to error handler
Kim Phillips [Sun, 12 Oct 2008 12:19:35 +0000 (20:19 +0800)] 
crypto: talitos - Pass correct interrupt status to error handler

Since we ack early, the re-read interrupt status in talitos_error
may be already updated with a new value.  Pass the error ISR value
directly in order to report and handle the error based on the correct
error status.

Also remove unused error tasklet.

Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Lee Nipper <lee.nipper@freescale.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agocrypto: hifn - Disable driver when physical size exceeds 32 bits
Evgeniy Polyakov [Sun, 12 Oct 2008 12:14:15 +0000 (20:14 +0800)] 
crypto: hifn - Disable driver when physical size exceeds 32 bits

On Tue, Sep 23, 2008 at 08:06:32PM +0200, Dimitri Puzin (max@psycast.de) wrote:
> With this patch applied it still doesn't work as expected. The overflow
> messages are gone however syslog shows
> [  120.924266] hifn0: abort: c: 0, s: 1, d: 0, r: 0.
> when doing cryptsetup luksFormat as in original e-mail. At this point
> cryptsetup hangs and can't be killed with -SIGKILL. I've attached
> SysRq-t dump of this condition.

Yes, I was wrong with the patch: HIFN does not support 64-bit addresses
afaics.

Attached patch should not allow HIFN to be registered on 64-bit arch, so
crypto layer will fallback to the software algorithms.

Signed-off-by: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
16 years agoLinux 2.6.28 v2.6.28
Linus Torvalds [Wed, 24 Dec 2008 23:26:37 +0000 (15:26 -0800)] 
Linux 2.6.28

Happy holidays..

16 years agoRevert "x86: disable X86_PTRACE_BTS"
Ingo Molnar [Wed, 24 Dec 2008 20:08:37 +0000 (21:08 +0100)] 
Revert "x86: disable X86_PTRACE_BTS"

This reverts commit 40f15ad8aadff5ebb621b17a6f303ad2cd3f847d.

The CONFIG_X86_PTRACE_BTS bugs have been fixed via:

 c5dee61: x86, bts: memory accounting
 bf53de9: x86, bts: add fork and exit handling

Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agoMerge branch 'linus' into tracing/hw-branch-tracing
Ingo Molnar [Wed, 24 Dec 2008 20:08:26 +0000 (21:08 +0100)] 
Merge branch 'linus' into tracing/hw-branch-tracing

16 years agoMerge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab...
Linus Torvalds [Wed, 24 Dec 2008 18:24:52 +0000 (10:24 -0800)] 
Merge branch 'for_linus' of git://git./linux/kernel/git/mchehab/linux-2.6

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6:
  V4L/DVB (9920): em28xx: fix NULL pointer dereference in call to VIDIOC_INT_RESET command
  V4L/DVB (9908a): MAINTAINERS: mark linux-uvc-devel as subscribers only
  V4L/DVB (9906): v4l2-compat: test for unlocked_ioctl as well.
  V4L/DVB (9885): drivers/media Kconfig's: fix bugzilla #12204
  V4L/DVB (9875): gspca - main: Fix vidioc_s_jpegcomp locking.
  V4L/DVB (9781): [PATCH] Cablestar 2 I2C retries (fix CableStar2 support)
  V4L/DVB (9780): dib0700: Stop repeating after user stops pushing button

16 years agoMerge branch 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Wed, 24 Dec 2008 18:24:14 +0000 (10:24 -0800)] 
Merge branch 'x86-fixes-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86: disable X86_PTRACE_BTS

16 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6
Linus Torvalds [Wed, 24 Dec 2008 18:23:21 +0000 (10:23 -0800)] 
Merge branch 'for-linus' of git://git./linux/kernel/git/tiwai/sound-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6:
  ALSA: hda - Add missing terminators in patch_sigmatel.c

16 years agoMerge branch 'fix/hda' into topic/hda
Takashi Iwai [Wed, 24 Dec 2008 10:04:08 +0000 (11:04 +0100)] 
Merge branch 'fix/hda' into topic/hda

16 years agoALSA: hda - Add missing terminators in patch_sigmatel.c
Herton Ronaldo Krzesinski [Tue, 23 Dec 2008 18:53:00 +0000 (16:53 -0200)] 
ALSA: hda - Add missing terminators in patch_sigmatel.c

Signed-off-by: Herton Ronaldo Krzesinski <herton@mandriva.com.br>
Cc: stable@kernel.org
Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agox86: disable X86_PTRACE_BTS
Ingo Molnar [Wed, 24 Dec 2008 09:49:51 +0000 (10:49 +0100)] 
x86: disable X86_PTRACE_BTS

there's a new ptrace arch level feature in .28:

  config X86_PTRACE_BTS
  bool "Branch Trace Store"

it has broken fork() handling: the old DS area gets copied over into
a new task without clearing it.

Fixes exist but they came too late:

  c5dee61: x86, bts: memory accounting
  bf53de9: x86, bts: add fork and exit handling

and are queued up for v2.6.29. This shows that the facility is still not
tested well enough to release into a stable kernel - disable it for now and
reactivate in .29. In .29 the hardware-branch-tracer will use the DS/BTS
facilities too - hopefully resulting in better code.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agox86: PAT: fix address types in track_pfn_vma_new()
H. Peter Anvin [Tue, 23 Dec 2008 18:10:40 +0000 (10:10 -0800)] 
x86: PAT: fix address types in track_pfn_vma_new()

Impact: cleanup, fix warning

This warning:

 arch/x86/mm/pat.c: In function track_pfn_vma_copy:
 arch/x86/mm/pat.c:701: warning: passing argument 5 of follow_phys from incompatible pointer type

Triggers because physical addresses are resource_size_t, not u64.

This really matters when calling an interface like follow_phys() which
takes a pointer to a physical address -- although on x86, being
littleendian, it would generally work anyway as long as the memory region
wasn't completely uninitialized.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
16 years agoparisc: disable UP-optimized flush_tlb_mm
Kyle McMartin [Tue, 23 Dec 2008 13:44:30 +0000 (08:44 -0500)] 
parisc: disable UP-optimized flush_tlb_mm

flush_tlb_mm's "optimized" uniprocessor case of allocating a new
context for userspace is exposing a race where we can suddely return
to a syscall with the protection id and space id out of sync, trapping
on the next userspace access.

Debugged-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Tested-by: Helge Deller <deller@gmx.de>
Signed-off-by: Kyle McMartin <kyle@mcmartin.ca>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge branch 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied...
Linus Torvalds [Wed, 24 Dec 2008 01:01:40 +0000 (17:01 -0800)] 
Merge branch 'drm-fixes' of git://git./linux/kernel/git/airlied/drm-2.6

* 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6:
  drm/radeon: fix correctness of irq_enabled check for radeon.

16 years agoedac: fix edac core deadlock when removing a device
Harry Ciao [Tue, 23 Dec 2008 21:57:16 +0000 (13:57 -0800)] 
edac: fix edac core deadlock when removing a device

When deleting an edac device, we have to wait for its edac_dev.work to be
completed before deleting the whole edac_dev structure.  Since we have no
idea which work in current edac_poller's workqueue is the work we are
conerned about, we wait for all work in the edac_poller's workqueue to be
proceseed.  This is done via flush_cpu_workqueue() which inserts a
wq_barrier into the tail of the workqueue and then sleeping on the
completion of this wq_barrier.  The edac_poller will wake up sleepers when
it is found.

EDAC core creates only one kernel worker thread, edac_poller, to run the
works of all current edac devices.  They share the same callback function
of edac_device_workq_function(), which would grab the mutex of
device_ctls_mutex first before it checks the device.  This is exactly
where edac_poller and rmmod would have a great chance to deadlock.

In below call trace of rmmod > ... >
edac_device_del_device >
edac_device_workq_teardown > flush_workqueue > flush_cpu_workqueue,

device_ctls_mutex would have already been grabbed by
edac_device_del_device().  So, on one hand rmmod would sleep on the
completion of a wq_barrier, holding device_ctls_mutex; on the other hand
edac_poller would be blocked on the same mutex when it's running any one
of works of existing edac evices(Note, this edac_dev.work is likely to be
totally irrelevant to the one that is being removed right now)and never
would have a chance to run the work of above wq_barrier to wake rmmod up.

edac_device_workq_teardown() should not be called within the critical
region of device_ctls_mutex.  Just like is done in edac_pci_del_device()
and edac_mc_del_mc(), where edac_pci_workq_teardown() and
edac_mc_workq_teardown() are called after related mutex are released.

Moreover, an edac_dev.work should check first if it is being removed.  If
this is the case, then it should bail out immediately.  Since not all of
existing edac devices are to be removed, this "shutting flag" should be
contained to edac device being removed.  The current edac_dev.op_state can
be used to serve this purpose.

The original deadlock problem and the solution have been witnessed and
tested on actual hardware.  Without the solution, rmmod an edac driver
would result in below deadlock:

root@localhost:/root> rmmod mv64x60_edac
EDAC DEBUG: mv64x60_dma_err_remove()
EDAC DEBUG: edac_device_del_device()
EDAC DEBUG: find_edac_device_by_dev()

(hang for a moment)

INFO: task edac-poller:2030 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
edac-poller   D 00000000     0  2030      2
Call Trace:
[df159dc0] [c0071e3c] free_hot_cold_page+0x17c/0x304 (unreliable)
[df159e80] [c000a024] __switch_to+0x6c/0xa0
[df159ea0] [c03587d8] schedule+0x2f4/0x4d8
[df159f00] [c03598a8] __mutex_lock_slowpath+0xa0/0x174
[df159f40] [e1030434] edac_device_workq_function+0x28/0xd8 [edac_core]
[df159f60] [c003beb4] run_workqueue+0x114/0x218
[df159f90] [c003c674] worker_thread+0x5c/0xc8
[df159fd0] [c004106c] kthread+0x5c/0xa0
[df159ff0] [c0013538] original_kernel_thread+0x44/0x60
INFO: task rmmod:2062 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
rmmod         D 0ff2c9fc     0  2062   1839
Call Trace:
[df119c00] [c0437a74] 0xc0437a74 (unreliable)
[df119cc0] [c000a024] __switch_to+0x6c/0xa0
[df119ce0] [c03587d8] schedule+0x2f4/0x4d8
[df119d40] [c03591dc] schedule_timeout+0xb0/0xf4

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agocgroups: avoid accessing uninitialized data in failure path
Li Zefan [Tue, 23 Dec 2008 21:57:14 +0000 (13:57 -0800)] 
cgroups: avoid accessing uninitialized data in failure path

If cgroup_get_rootdir() failed, free_cg_links() will be called in the
failure path, but tmp_cg_links hasn't been initialized at that time.

I introduced this bug in the 2.6.27 merge window.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agocgroups: suppress bogus warning messages
Sharyathi Nagesh [Tue, 23 Dec 2008 21:57:12 +0000 (13:57 -0800)] 
cgroups: suppress bogus warning messages

Remove spurious warning messages that are thrown onto the console during
cgroup operations.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Sharyathi Nagesh <sharyathi@in.ibm.com>
Acked-by: Serge E. Hallyn <serge@hallyn.com>
Cc: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>