sched: Track and export per task [hard|soft]irq time
On Tue, May 25, 2010 at 2:13 AM, Balbir Singh <balbir(a)> wrote: * Venkatesh Pallipadi <venki(a)> [2010-05-24 17:11:19]: Currently, kernel does not have accounting mechanism for softirq and hardirq times at the task level. There is irq time info in kstat_cpu which is accumulate... 25 May 2010 17:58
BUG: using smp_processor_id() in preemptible [00000000] code: icedove-bin/5449
On Tue, 2010-05-25 at 11:07 -0700, Divyesh Shah wrote: Peter, this only affects some stats collection and no scheduling behavior should be affected by this. Besides this call at rq_init() to sched_clock() there are other calls in the request dispatch and the request completion path. Not sure if those are pr... 12 Jun 2010 06:06
[patch] padata: add parenthesis in MAX_SEQ_NR macro
MAX_SEQ_NR is used in padata_alloc_pd() like this: pd->max_seq_nr = (MAX_SEQ_NR / num_cpus) * num_cpus - 1; It needs parenthesis or the divide by num_cpus takes precedence over the subtraction. Signed-off-by: Dan Carpenter <error27(a)> --- Found by a static checker. I don't really know this ... 25 May 2010 17:58
[Regression] Crash in load_module() while freeing args
Hi, Recent -git kernels crash for me quite often during boot, reporting that kernel paging request could not be satisfied at load_module+0x18e4, which corresponds to line 2483 of kernel/module.c for the kernel in question (however the same call trace says IP is at precpu_modfree+0x1 at the moment of the crash, w... 25 May 2010 17:58
hrtimer: avoid double seqlock
Signed-off-by: Stanislaw Gruszka <stf_xl(a)> diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index b9b134b..5c69e99 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -89,7 +89,7 @@ static void hrtimer_get_softirq_time(struct hrtimer_cpu_base *base) do { seq = read_seqbegin(&xtime_lock); - x... 25 May 2010 16:52
[git pull] FireWire updates post 2.6.34
(Sorry, got Subject and Cc wrong in the previous mail.) Linus, please pull from the for-linus branch at git:// for-linus to receive the following IEEE 1394/ FireWire subsystem updates. You will get a trivial conflict in feature-removal-sche... 25 May 2010 16:52
genirq: Run irq handlers with interrupts disabled
Just noticed a minor side effect of this patch on /proc/stat. As int handlers run with ints disabled, timer interrupt based samples will no longer account for any irq time. As a result, irq time in /proc/stat, top, etc will be mostly zero (unless some handler is enabling ints). This is not a problem on s390, p... 25 May 2010 16:52
powernow-k8: section mismatch
Hi, I got the following with current git: WARNING: arch/x86/kernel/cpu/cpufreq/powernow-k8.o(.exit.text+0x1c): Section mismatch in reference from the function cleanup_module() to the variable The function __exit cleanup_module() references ... 25 May 2010 19:04
FLAT: allow arches to declare a larger alignment than the slab
On Tue, May 25, 2010 at 21:24, Mike Frysinger <vapier(a)> wrote: From: Jie Zhang <jie.zhang(a)> The recent commit 1f0ce8b3dd667dca7 which moved the ARCH_SLAB_MINALIGN default into the global header inadvertently broke FLAT for a bunch of systems.  Blackfin systems now fail on any FLAT exe... 26 May 2010 04:49
sata_nv times out for BD-ROM iHOS104-08
Hello, Alright, I don't have access to mcp51 but tested with mcp55 and could reproduce similar problem. It seems nIEN on mcp55 is stuck once set and iHOS104 doesn't set I on D2H FIS if nIEN is set on command H2D FIS, which the SATA standard specifically mandates not to do. So, the combination of buggy mcp55 ct... 3 Jun 2010 14:31