Typo: depricated -> deprecated Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de> --- commit b7ec07d26c9fd915fd32b759aa05a1ead8c432f5 tree 6f6dafac4072fddd670a97f68cbd64bf505d0408 parent 43dcafe3f7bfc07bdaf55d353fd4c5112c5f2be6 author Rolf Eike Beer <eike-kernel@sf-tec.de> Sun, 07 Oct 2007 10:29:39 +0200 committer Rolf Eike Beer <eike-kernel@sf-tec.de> Sun, 07 Oct 2007 10:29:39 +0200 drivers/acpi/Kconfig | 2 +- drivers/net/natsemi.c | 2 +- ...

Ah, indeed. That means introducing a new "really congested" threshold for kswapd is *dangerous*. I realized this later on, and am now heading for another direction. The basic idea is to - rotate pdflush issued writeback pages for kswapd; - use the more precise zone_rotate_wait() to throttle kswapd. The code is a quick hack and not tested yet. Early comments are more than welcome. Fengguang --- include/linux/mmzone.h | 1 + mm/filemap.c | 5 ++++- mm/page_alloc.c |...

- Various source code cleanups (no functional changes) - Shrink binary size by refactoring some code - One K8 optimization that is unfortunately not measurable. But it sounded good in theory! @) (if you have a better scheduling latency benchmark than lmbench3 it would be cool to try) - One real bug fix for a obscure problem All against the sched-devel git tree with d024f67c652f7f0519d2a1906b646286abbb2e48 head. Future wishlist: someone document all the magic numbers in the load balancer c...

Replace a particularly ugly ifdef with an inline and a new macro. Also split up the function to be easier readable. Signed-off-by: Andi Kleen <ak@suse.de> Index: linux-2.6-sched-devel/kernel/sched.c =================================================================== --- linux-2.6-sched-devel.orig/kernel/sched.c +++ linux-2.6-sched-devel/kernel/sched.c @@ -74,6 +74,12 @@ unsigned long long __attribute__((weak)) return (unsigned long long)jiffies * (1000000000 / HZ); } +#ifdef CONF...

These functions were largely cut'n'pasted. This moves the common code into single helpers instead. Advantage is about 1k less code on x86-64 and 91 lines of code removed. It adds one function call to the non timeout version of the functions; i don't expect this to be measurable. Signed-off-by: Andi Kleen <ak@suse.de> Index: linux-2.6-sched-devel/kernel/sched.c =================================================================== --- linux-2.6-sched-devel.orig/kernel/sched.c +++ linux-2.6...

Some CPUs like K8 cannot predict indirect calls. A common optimization in object oriented languages is to check for the most common call target and then call it directly; otherwise do an indirect call. This patch does this manually for sched_fair calls in sched.c In theory -- it doesn't seem to happen currently -- this would allow the compiler to even inline parts of sched_fair.c into the callers. I only did it only for fast paths (wake up and schedule) because the generated code is little ...

Hi, IDE/ATA wizzards are kindly requested to have a look at: http://bugzilla.kernel.org/show_bug.cgi?id=4580#c21 as I have no idea of what we can do about the "hard drive password vs suspend" problem, if anything. Greetings, Rafael -

FWIW, ACPI is turned on in libata-dev.git#upstream, and as soon as Andrew does his next update, it should be turned on there too. (it might even be in his last -mm update, not sure) Jeff -

Doesn't appear to be enough. I can't figure out why (since it appears write_cache_pages bubbles up directly to sys_msync), but with that patch applied, in my test case[1], msync returns -1 EIO. However, with the exact same kernel without that patch applied, msync returns 524288 (AOP_WRITEPAGE_ACTIVATE). But as your patch specifically flips 524288 to 0, I can't figure out how it eventually returns -1 EIO. Ryan [1] "apt-get check" on a unionfs2 mount backed by tmpfs over cdrom, standard live...

Hi, I found this in an `allyesconfig' build: CC drivers/dma/ioatdma.o drivers/dma/ioatdma.c: In function #ioat_dma_free_chan_resources#: drivers/dma/ioatdma.c:247: warning: unused variable #chanctrl# Hannes Signed-off-by: Johannes Weiner <hannes-kernel@saeurebad.de> diff --git a/drivers/dma/ioatdma.c b/drivers/dma/ioatdma.c index 41b18c5..d9db64b 100644 --- a/drivers/dma/ioatdma.c +++ b/drivers/dma/ioatdma.c @@ -244,7 +244,6 @@ static void ioat_dma_free_chan_resources(...

According to vfs.txt, ->writepage() may return AOP_WRITEPAGE_ACTIVATE back to the VFS/VM. Indeed some filesystems such as tmpfs can return AOP_WRITEPAGE_ACTIVATE; and stackable file systems (e.g., Unionfs) also return AOP_WRITEPAGE_ACTIVATE if the lower f/s returned it. Anyway, some Ubuntu users of Unionfs reported that msync(2) sometimes returns AOP_WRITEPAGE_ACTIVATE (decimal 524288) back to userland. Therefore, some user programs fail, esp. if they're written such as this: err = msync...

Re: kmail - it usually behaves well with the patches as long as you turn off word wrapping and use "Insert file..." to insert them. -- Dmitry -

Security fixes since 2.6.16.54: - CVE-2005-0504: fix buffer overflow in the moxa driver - CVE-2007-2453: random fixes - CVE-2007-3104: store sysfs inode nrs in s_ino to avoid readdir oopses - CVE-2007-3105: random: fix bound check ordering - CVE-2007-3848: Reset current->pdeath_signal on SUID binary execution - CVE-2007-4571: Convert snd-page-alloc proc file to use seq_file - CVE-2007-4573: x86_64: Zero extend all registers after ptrace in 32bit entry path Location: [ message continues ]

Subject: Driver core: fix SYSF_DEPRECATED breakage for nested classdevs From: Dmitry Torokhov <dmitry.torokhov@gmail.com> We should only reparent to a class former class devices that form the base of class hierarchy. Nested devices should still grow from their real parents. Signed-off-by: Dmitry Torokhov <dtor@mail.ru> Tested-by: Andrey Borzenkov <arvidjaar@mail.ru> Tested-by: Anssi Hannula <anssi.hannula@gmail.com> --- Linus, Please consider applying this patch bef...

And indeed, reverting this commit with 'patch -R' makes things work again. Jeff -

The command is Fedora 7 initscripts... (typing in by hand) /etc/sysconfig/network-scripts/network-functions-ipv6: line 246: 1760 Killed LC_ALL=C /sbin/ip $options NULL pointer dereference at 0x000003f8 rtnetlink_rcv_msg inet6_rtm_newroute netlink_run_queue rtnetlink_rcv netlink_data_ready netlink_sendskb netlink_sendmsg sock_send_msg mntput_no_expire autoremove_wake_function find_lock_page zone_statistics get_page_from_freelist __alloc_pages move_addr_to_kernel verify_iovec ...

>} In Document/RCU/whatisRCU.txt Note that the value returned by rcu_dereference() is valid only within the enclosing RCU read-side critical section. For example, the following is -not- legal: rcu_read_lock(); p = rcu_dereference(head.next); rcu_read_unlock(); x = p->address; rcu_read_lock(); y = p->data; rcu_read_unlock(); Holding a reference from one RCU read-side critical section to another is just as illegal as holding a reference from one lock-based critic...

I'm sorry,I think I got your idea, if you do not need ptr in my_task_that_may_sleep(), why you need to grab a reference to "ptr". If it is ok to use "ptr" after rcu_read_ulock(). The basic idea behind RCU is to split updates into "removal" and "reclamation" phases. If you memory region pointed by "ptr" will not "reclamation" in sleep, it is ok Regards. -

Still think that is a very BAD (tm) idea. Esp since with the help of RCU there is hardly any performance loss of not doing proper deletes. -

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de> --- drivers/firewire/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) Index: linux/drivers/firewire/Kconfig =================================================================== --- linux.orig/drivers/firewire/Kconfig +++ linux/drivers/firewire/Kconfig @@ -11,7 +11,8 @@ config FIREWIRE This is the "Juju" FireWire stack, a new alternative implementation designed for robustness and simplicity. You can build e...

When root_port_reset() in ohci-hub.c polls for the end of the reset, it puts no limit on the loop and will only exit the loop when either the RH_PS_PRS bit clears or the register returns all 1's (the latter condition is a recent addition). If for some reason the bit never clears, we sit here forever and never exit the loop. I actually hit this on one of my machines, and I'm trying to track down what's happening. Regardless of why my machine is doing this, there absolutely should be some upp...

From: David Brownell <david-b@pacbell.net> I'm 700 patches deep with an additionally large backlog of patches to apply for the networking and sparc64 tree. I don't have the time, which is why I reported the problem to the OHCI maintainer instead of letting it slip through At a minimum you should exit the loop and print out a warning messages and try to continue even without trying to reset the whole controller if that will take some developer effort. Anything is better than just ha...

I'm not sure that's a valid assumption in light of things like sharing pagetables between processes etc etc.. (granted, that one is out of tree right now but I still hope it'll go in some day:) -

On Sun, 7 Oct 2007 10:17:47 -0700 s/implemented/merged/ :) IN fact shared pagetables are already there for hugepages. no I'm not saying that. I'm just saying that I'm worried about the locking robustness of your trick in general. -

Good point. It's not a problem for me because they're two ptes in different pagetables, whereas my multiple ptes are always within the same pagetable. Someone wanting to lock multiple ptes from multiple pagetables would have a more complex locking order problem. (Or if we end up sharing ptes between pagetables it might get tricky.) J -