Kernel

[PATCH 0/2] Fix migration races in rmap_walk() V6
Patch 1 of this series is the biggest change. Instead of the trylock+retry logic, it finds the "root" anon_vma and locks all anon_vmas encountered. As long as walkers taking multiple locks use the same order, there is no deadlock. Stress-tests based on compaction have been running a while with these patches applied... 6 May 2010 11:55

[PATCH 1/7] oprofile/x86: fix uninitialized counter usage during cpu hotplug
This fixes a NULL pointer dereference that is triggered when taking a cpu offline after oprofile was initialized, e.g.: $ opcontrol --init $ opcontrol --start-daemon $ opcontrol --shutdown $ opcontrol --deinit $ echo 0 > /sys/devices/system/cpu/cpu1/online See the crash dump below. Though the counter ha... 6 May 2010 11:55

[PATCH 0/7] oprofile/x86: fix cpu hotplug
This patch series fixes oprofile cpu hotplug. The code was broken and causing crashes due to uninitialized counter usage. This can be only fixed with a code rework. The first patch is to avoid kernel crashes during cpu hotplug. This is a small fix for linux-stable. Ingo, I will send a separate pull request. -R... 6 May 2010 11:55

[PATCH 4/7] oprofile/x86: stop disabled counters in nmi handler
This patch adds checks to the nmi handler. Now samples are only generated and counters reenabled, if the counters are running. Otherwise the counters are stopped, if oprofile is using the nmi. In other cases it will ignore the nmi notification. Cc: Andi Kleen <andi(a)firstfloor.org> Signed-off-by: Robert Richter <... 6 May 2010 11:55

[PATCH 3/7] oprofile/x86: protect cpu hotplug sections
This patch reworks oprofile cpu hotplug code as follows: Introduce ctr_running variable to check, if counters are running or not. The state must be known for taking a cpu on or offline and when switching counters during counter multiplexing. Protect on_each_cpu() sections with get_online_cpus()/put_online_cpu()... 6 May 2010 11:55

[PATCH 6/7] oprofile/x86: notify cpus only when daemon is running
This patch moves the cpu notifier registration from nmi_init() to nmi_setup(). The corresponding unregistration function is now in nmi_shutdown(). Thus, the hotplug code is only active, if the oprofile daemon is running. Cc: Andi Kleen <andi(a)firstfloor.org> Signed-off-by: Robert Richter <robert.richter(a)amd.com> ... 6 May 2010 11:55

[PATCH] mm,compaction: Do not schedule work on other CPUs for compaction
Migration normally requires a call to migrate_prep() as a preparation step. This schedules work on all CPUs for pagevecs to be drained. This makes sense for move_pages and memory hot-remove but is unnecessary for memory compaction. To avoid queueing work on multiple CPUs, this patch introduces migrate_prep_local... 6 May 2010 11:55

linux-next: Tree for May 6 (acpi: PM=n)
On Thu, 6 May 2010 15:15:02 +1000 Stephen Rothwell wrote: Hi all, Changes since 20100505: When CONFIG_PM is not set: drivers/built-in.o: In function `acpi_init': bus.c:(.init.text+0x2d84): undefined reference to `pm_flags' bus.c:(.init.text+0x2d91): undefined reference to `pm_flags' --- ~Randy... 6 May 2010 19:51

btusb patch, add_apple_MacBookPro62
Hello Marcel, the Apple MBP model 6.2 (and maybe other recent MBPs) use an Apple branded BCM2046 USB bluetooth device that works well [1] with your generic driver. The attached patch just adds the vendor and device id - please apply. Cheers, -bf [1] http://ubuntuforums.org/showthread.php?t=1469437 ... 19 May 2010 18:15

perf_events: ctx_flexible_sched_in() not maximizing PMU utilization
On Thu, 2010-05-06 at 16:03 +0200, Stephane Eranian wrote: Hi, Looking at ctx_flexible_sched_in(), the logic is that if group_sched_in() fails for a HW group, then no other HW group in the list is even tried. I don't understand this restriction. Groups are independent of each other. The failure of o... 14 May 2010 11:42