Linus Torvalds a écrit :
On Sat, 5 May 2007, Eric Dumazet wrote:
But... what happens if the thread that was chosen exits from the loop in
ep_poll() with res = -EINTR (because of signal_pending(current))
Not a problem.
What happens is that an exclusive wake-up stops on the first entry in the
Christoph Lameter a écrit :
On Sat, 5 May 2007, Eric Dumazet wrote:
C. Introduces a slab_ops structure that allows a slab user to provide
operations on slabs.
Could you please make it const ?
Sure. Done.
thanks :)
All of this is really not necessary since the compiler knows how to
Jan,
Can You send output of x86info program and output of
lspci command? Longhaul wasn't working for You since 2.6.18 right?
I'm going to work now, but I will be available after 14:00 UTC.
If You have problem with longhaul+powersave there may be one thing
related. When I started to change
On Fri, May 04, 2007 at 03:15:57PM -0700, [EMAIL PROTECTED] wrote:
> 2. kick_object(void *)
> After SLUB has established references to the remaining objects in a slab it
> will drop all locks and then use kick_object on each of the objects for which
> we obtained a reference. The existence of the
Allocate/destroy a 'vmalloc' VM area: alloc_vm_area and free_vm_area
The alloc function ensures that page tables are constructed for the
region of kernel virtual address space and mapped into init_mm.
Signed-off-by: Jeremy Fitzhardinge <[EMAIL PROTECTED]>
Signed-off-by: Ian Pratt <[EMAIL
On Sat, 5 May 2007, Eric Dumazet wrote:
> > C. Introduces a slab_ops structure that allows a slab user to provide
> >operations on slabs.
>
> Could you please make it const ?
Sure. Done.
> > All of this is really not necessary since the compiler knows how to align
> > structures and we
[EMAIL PROTECTED] a écrit :
I originally intended this for the 2.6.23 development cycle but since there
is an aggressive push for SLUB I thought that we may want to introduce this
earlier.
Note that this covers new locking approaches that we may need to talk
over before going any further.
This
On Sat, 5 May 2007, Eric Dumazet wrote:
>
> But... what happens if the thread that was chosen exits from the loop in
> ep_poll() with res = -EINTR (because of signal_pending(current))
Not a problem.
What happens is that an exclusive wake-up stops on the first entry in the
wait-queue that it
On Sat, 5 May 2007, Pierre Ossman wrote:
> Pierre Ossman wrote:
> > Linus, please pull from
> >
> > git://git.kernel.org/pub/scm/linux/kernel/git/drzeus/mmc.git
> > for-linus
> >
>
> *ping*
*pong*.
Thanks for reminding me. I was away for a couple of days, missed some
emails, just
On Fri, 4 May 2007 23:34:08 -0400 Quentin Godfroy <[EMAIL PROTECTED]> wrote:
> By the way, is init 32 bits or 64 bits? It could break the ia32
> emulation thing, but not the 64bit native mode.
akpm2:/home/akpm> file /sbin/init
/sbin/init: ELF 64-bit LSB executable, AMD x86-64, version 1
Pierre Ossman wrote:
> Linus, please pull from
>
> git://git.kernel.org/pub/scm/linux/kernel/git/drzeus/mmc.git for-linus
>
*ping*
--
-- Pierre Ossman
Linux kernel, MMC maintainerhttp://www.kernel.org
PulseAudio, core developer http://pulseaudio.org
Quentin Godfroy wrote:
>> Won't this break with ET_DYN executables? And besides, isn't this the
>> same thing?
>>
>
> Indeed, I haven't seen that. For ET_DYN executables, it could be done a
> thing like load_addr+elf_ppnt->p_vaddr
Davi Arnaut a écrit :
Hi,
If multiple threads are parked on epoll_wait (on a single epoll fd) and
events become available, epoll performs a wake up of all threads of the
poll wait list, causing a thundering herd of processes trying to grab
the eventpoll lock.
This patch addresses this by using
>>> Switching from acpi_pm+performance to acpi_pm+ondemand also
>>> locks up after a few minutes.
>> Yep. Sounds like an ondemand issue. Thanks for verifying this for me.
>
> Nah, it also happens with cpufreq_powersave. I just need to check
> through some archives and try booting with
On 5/5/07, Florin Malita <[EMAIL PROTECTED]> wrote:
Hi Satyam,
Satyam Sharma wrote:
> Eeks ... no, wait. You found a (two, actually) bug alright, but fixed
> it wrong. When we fail a write, we *must* add it to the corrupted list
> and _then_ attempt to retry. So, the "if (++tries <= 5)" applies
On Fri, 04 May 2007, Shem Multinymous wrote:
> >+enum power_supply_type {
> >+ POWER_SUPPLY_TYPE_BATTERY = 0,
> >+ POWER_SUPPLY_TYPE_UPS,
> >+ POWER_SUPPLY_TYPE_AC,
> >+ POWER_SUPPLY_TYPE_USB,
> >+};
>
> How about dumb (non-USB) DC power? Any reason to distinguish it from
From: [EMAIL PROTECTED] (Eric W. Biederman)
Date: Fri, 04 May 2007 21:14:42 -0600
> David Miller <[EMAIL PROTECTED]> writes:
>
> > From: Rusty Russell <[EMAIL PROTECTED]>
> > Date: Sat, 05 May 2007 11:22:48 +1000
> >
> >> Hi Eric,
> >
> > To all of those who don't speak "rusty", "Hi Soandso"
On Fri, May 04, 2007 at 04:22:08PM -0700, Andrew Morton wrote:
> This patch kills my FC6 machine (using a config which was derived from RH's
> original):
>
> Freeing unused kernel memory: 368k freed
> Write protecting the kernel read-only data: 959k
> request_module: runaway loop modprobe
On Fri, May 04, 2007 at 04:31:49PM -0700, Jeremy Fitzhardinge wrote:
> Quentin Godfroy wrote:
> > + elf_ppnt = elf_phdata;
> > + for (i = 0; i< loc->elf_ex.e_phnum; i++, elf_ppnt++)
> > + if (elf_ppnt->p_type == PT_PHDR) {
> > + phdr_addr = elf_ppnt->p_vaddr;
> >
On Fri, 2007-05-04 at 20:53 -0600, Eric W. Biederman wrote:
> Rusty Russell <[EMAIL PROTECTED]> writes:
> The delicate part right now is that lguest is attempting to use the
> standard kernel entry point which does have a fixed ABI.
>
> If lguest uses that entry point in a hard to maintain way it
David Miller <[EMAIL PROTECTED]> writes:
> From: Rusty Russell <[EMAIL PROTECTED]>
> Date: Sat, 05 May 2007 11:22:48 +1000
>
>> Hi Eric,
>
> To all of those who don't speak "rusty", "Hi Soandso" means
> "NAK".
The question between Rusty and myself is not if we should remove
that code but when.
On Tue, 1 May 2007 13:43:18 -0700
"Cabot, Mason B" <[EMAIL PROTECTED]> wrote:
Hello all,
I've been testing the NAS performance of ext3/Openfiler 2.2 against
NTFS/WinXP and have found that NTFS significantly outperforms ext3 for
video workloads. The Windows CIFS client will attempt a
Rusty Russell <[EMAIL PROTECTED]> writes:
> Hi Eric,
>
> Well, I certainly don't recall that (that's not to say that someone
> didn't say it). Trying to meet the requirements of Xen, VMI and other
> future hypervisors lead to an awkward result; this is the main reason I
> started on
From: Rusty Russell <[EMAIL PROTECTED]>
Date: Sat, 05 May 2007 11:22:48 +1000
> Hi Eric,
To all of those who don't speak "rusty", "Hi Soandso" means
"NAK".
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info
On Fri, May 04, 2007 at 07:49:13PM +0400, Michael Tokarev wrote:
>
> How about providing a way to stop kernel (or filesystem) to make gaps
> in files instead? Like some ioctl(fd, FS_NOGAPS, 1) -- pretty much
> like 'doze has, just the opposite (on windows, this flag is "on" by
> default).
This
"H. Peter Anvin" <[EMAIL PROTECTED]> writes:
> Eric W. Biederman wrote:
>>
>> I have this vague memory of liking 0x3c because if we do happen to use
>> more room then we intended the consequences are pretty benign.
>>
>> But that is a pretty minor consequence.
>>
I meant to say it is a pretty
Got something If I remove the atomics from both alloc and free then I
get a performance jump. But maybe also a runtime variation
Avoid the use of atomics in slab_alloc
About 5-7% performance gain. Or am I also seeing runtime variations?
What we do is add the last free field in the page
On Fri, 2007-05-04 at 09:07 -0600, Eric W. Biederman wrote:
> Rusty Russell <[EMAIL PROTECTED]> writes:
>
> > On Fri, 2007-05-04 at 08:13 -0600, Eric W. Biederman wrote:
> >> We don't have any working code, there are no in tree users.
> >
> > Hi Eric,
> >
> > Lack of in-tree code is
On Fri, 2007-05-04 at 08:58 -0700, Andrew Morton wrote:
> On Sat, 05 May 2007 00:37:30 +1000 Rusty Russell <[EMAIL PROTECTED]> wrote:
>
> > Of course, I expect that Andrew is about to push my patches to Linus
> > any day now... right Andrew? Then we don't need this argument.
>
> It would be
Once upon a time, Daniel J Blueman <[EMAIL PROTECTED]> said:
> On 4 May, 01:20, Chris Adams <[EMAIL PROTECTED]> wrote:
> >I've got a Thinkpad Z60m with an ExpressCard slot, and I got a Belkin
> >F5U250 GigE ExpressCard (Marvell 88E8053 chip using sky2 driver). It
> >appears that Linux only
Eric W. Biederman wrote:
>
> I have this vague memory of liking 0x3c because if we do happen to use
> more room then we intended the consequences are pretty benign.
>
> But that is a pretty minor consequence.
>
That's a dangerous assumption (besides, it's likely wrong, since there
are only two
"H. Peter Anvin" <[EMAIL PROTECTED]> writes:
> Eric W. Biederman wrote:
>>
>> My notes show 0x5c reserved for additional apm_bios_info, although
>> of the top of my head I don't know how realistic that is.
>>
>> 0x1e4 does look available.
>>
>> It has been a long time since I made that choice,
If you want to test some more: Here is a patch that removes the atomic ops
from the allocation patch. But I only see minor improvements on my amd64
box here.
Avoid the use of atomics in slab_alloc
This only increases netperf performance by 1%. Wonder why?
What we do is add the last free
Jan-Benedict Glaw wrote:
> On Fri, 2007-05-04 16:59:51 +0200, Bernd Schubert <[EMAIL PROTECTED]>
> wrote:
>> To see whats going on, I copied the entire / (so the initrd) into a
>> tmpfs
>> root, chrooted into it, also bind mounted the main / into this chroot
>> and
>> compared several times /bin
Theodore Tso wrote:
> On Fri, May 04, 2007 at 04:59:51PM +0200, Bernd Schubert wrote:
>>
>> I'm presently rather puzzled, if this is really a kernel bug, its a
>> big
>> bug.
>>
>> Summary: The system ramdisk (initrd) gets corrupted while running
>> mkfs.ext2 on a local sata disk partition.
>
> The URL? Well, I'm glad you asked:
>
> http://lguest.ozlabs.org/lguest-2.6.21-269.patch.gz
>
>
No, I have not. Thanks anyway, will try it. Finally. 8-)
But he was talking about me! ;-)
(kidding)
--
Glauber de Oliveira Costa.
"Free as in Freedom"
"The less confident you are, the more
On Thu, 3 May 2007, Dmitry Torokhov wrote:
>
> If you have not pulled yet please pull from:
>
> Â Â Â Â master.kernel.org:/pub/scm/linux/kernel/git/dtor/input.git
> for-linus
>
> because master branch will have extra stuff in the next minute or so.
Hmm. That thing had a conflict with the
Andrew Morton wrote:
On Thu, 03 May 2007 13:01:17 -0500
Paul Fulghum <[EMAIL PROTECTED]> wrote:
Add compat_ioctl handler to synclink_gt driver.
i386 allmodconfig:
make[3]: *** No rule to make target
`/usr/src/devel/usr/include/linux/.check.synclink.h', needed by
`__headerscheck'. Stop.
On Fri, 4 May 2007, Randy Dunlap wrote:
> > /* Perform the KICK callbacks to remove the objects */
> > for(p = addr; p < addr + s->objects * s->size; p += s->size)
>
> missed a space after "for".
Thanks but I was more hoping for a higher level of review. Locking
-
To unsubscribe
On Fri, 4 May 2007, Tim Chen wrote:
> On Thu, 2007-05-03 at 18:45 -0700, Christoph Lameter wrote:
> > H.. One potential issues are the complicated way the slab is
> > handled. Could you try this patch and see what impact it has?
> >
> The patch boost the throughput of TCP_STREAM test by 5%,
On Fri, 4 May 2007 16:03:43 -0700 (PDT) Christoph Lameter wrote:
> Fixes suggested by Andrew
>
> ---
> include/linux/slab.h | 12
> mm/slub.c| 32 +---
> 2 files changed, 33 insertions(+), 11 deletions(-)
>
> /* Perform the KICK
On Fri, May 04, 2007 at 04:59:51PM +0200, Bernd Schubert wrote:
>
> I'm presently rather puzzled, if this is really a kernel bug, its a big bug.
>
> Summary: The system ramdisk (initrd) gets corrupted while running
> mkfs.ext2 on a local sata disk partition.
What distribution are you using?
On Fri, 2007-05-04 at 16:59 -0700, Christoph Lameter wrote:
> >
> > to run the tests. The results are about the same as the non-NUMA case,
> > with slab about 5% better than slub.
>
> H... both tests were run in the same context? NUMA has additional
> overhead in other areas.
Both slab
On Thu, 2007-05-03 at 18:45 -0700, Christoph Lameter wrote:
> H.. One potential issues are the complicated way the slab is
> handled. Could you try this patch and see what impact it has?
>
The patch boost the throughput of TCP_STREAM test by 5%, for both slab
and slub. But slab is still 5%
Jesse Barnes wrote:
On Wednesday, May 2, 2007 4:54 pm Jesse Barnes wrote:
What happens if you take out the chipset register detection, does
the MCFG table give you the same result? Wonder if they're doing
something funny with start/end bus values or something in their
table. There's some code
Call percpu smp cacheline align interface.
Signed-off-by: Fenghua Yu <[EMAIL PROTECTED]>
Acked-by: Suresh Siddha <[EMAIL PROTECTED]>
diff -Nurp linux-2.6.21-rc7.0/arch/i386/kernel/init_task.c
linux-2.6.21-rc7.1/arch/i386/kernel/init_task.c
--- linux-2.6.21-rc7.0/arch/i386/kernel/init_task.c
Define percpu smp cacheline align interface
Signed-off-by: Fenghua Yu <[EMAIL PROTECTED]>
Acked-by: Suresh Siddha <[EMAIL PROTECTED]>
diff -Nurp linux-2.6.21-rc7.0/arch/alpha/kernel/vmlinux.lds.S
linux-2.6.21-rc7.1/arch/alpha/kernel/vmlinux.lds.S
---
This is follow-up for Suresh's runqueue align in smp patch at:
http://www.uwsg.iu.edu/hypermail/linux/kernel/0704.1/0340.html
The patches place all of smp cacheline aligned percpu data into
.data.percpu.cacheline_aligned_in_smp. Other percpu data is still in
data.percpu section. The patches
Nick Piggin wrote:
> I literally have about 4 or 5 new page flags I'd like to add today :) I
> can't of course, because we have very few spare ones left.
I remember Rik saying that if need be he can (try to?) think of a method
to implement it without a page flag.
--
➧ Ulrich Drepper ➧ Red Hat,
This patch is a rollup of all the core pieces of the Xen
implementation, including:
- booting and setup
- pagetable setup
- privileged instructions
- segmentation
- multicall batching
Signed-off-by: Jeremy Fitzhardinge <[EMAIL PROTECTED]>
Signed-off-by: Chris Wright <[EMAIL PROTECTED]>
Cc:
On Fri, 4 May 2007, Tim Chen wrote:
> On Fri, 2007-05-04 at 11:27 -0700, Christoph Lameter wrote:
>
> >
> > Not sure where to go here. Increasing the per cpu slab size may hold off
> > the issue up to a certain cpu cache size. For that we would need to
> > identify which slabs create the
On Thu, 05 Apr 2007 17:01:02 +0900
Yasunori Goto <[EMAIL PROTECTED]> wrote:
> Hello.
>
> This is to fix many section mismatches of code related to memory hotplug.
> I checked compile with memory hotplug on/off on ia64 and x86-64 box.
>
> ..
>
>
This accounts for the time Xen steals from our VCPUs. This accounting
gets run on each timer interrupt, just as a way to get it run
relatively often, and when interesting things are going on.
Stolen time is not really used by much in the kernel; it is reported
in /proc/stats, and that's about
Disable interrupts between allocating a multicall entry and actually
issuing it, to prevent an interrupt from coming in, allocating and
initializing further multicall entries, and then issuing them all,
including the partially completed one.
Signed-off-by: Jeremy Fitzhardinge <[EMAIL PROTECTED]>
Add a new mm function apply_to_page_range() which applies a given
function to every pte in a given virtual address range in a given mm
structure. This is a generic alternative to cut-and-pasting the Linux
idiomatic pagetable walking code in every place that a sequence of
PTEs must be accessed.
Stolen time should never be negative; if it ever is, it probably
indicates some other bug. However, if it does happen, then its better
to just clamp it at zero, rather than trying to account for it as a
huge positive number.
Signed-off-by: Jeremy Fitzhardinge <[EMAIL PROTECTED]>
Acked-by: Chris
The block device frontend driver allows the kernel to access block
devices exported exported by a virtual machine containing a physical
block device driver.
Signed-off-by: Ian Pratt <[EMAIL PROTECTED]>
Signed-off-by: Christian Limpach <[EMAIL PROTECTED]>
Signed-off-by: Chris Wright <[EMAIL
Put config options for Xen after the core pieces are in place.
Signed-off-by: Jeremy Fitzhardinge <[EMAIL PROTECTED]>
Signed-off-by: Chris Wright <[EMAIL PROTECTED]>
---
arch/i386/Kconfig |2 ++
arch/i386/xen/Kconfig | 11 +++
2 files changed, 13 insertions(+)
Xen has a notion of pinned pagetables, which are pagetables that
remain read-only to the guest and are validated by the hypervisor.
This makes context switches much cheaper, because the hypervisor
doesn't need to revalidate the pagetable each time.
This patch adds a PG_pinned flag for pagetable
Hi forward this message to dri-devel Mailing List, where you could find
more tester on i815 DRI drives .
I hope I don't had made a loop :)
Forwarded Message
From: Andreas Mohr <[EMAIL PROTECTED]>
To: Pavel Machek <[EMAIL PROTECTED]>
Cc: Andrew Morton <[EMAIL PROTECTED]>, [EMAIL
This patchs adds the mechanism to allow us to patch inline versions of
common operations.
The implementations of the direct-access versions save_fl, restore_fl,
irq_enable and irq_disable are now in assembler, and the same code is
used for both out of line and inline uses.
Signed-off-by: Jeremy
The "netif_" prefix is used in the Linux network stack, so rename the
Xen structures to xen_netif_ to avoid confusion (and potential
collision).
Signed-off-by: Jeremy Fitzhardinge <[EMAIL PROTECTED]>
Signed-off-by: Chris Wright <[EMAIL PROTECTED]>
---
include/xen/interface/io/netif.h | 18
Rik van Riel wrote:
Nick Piggin wrote:
Rik van Riel wrote:
With lazy freeing of anonymous pages through MADV_FREE, performance of
the MySQL sysbench workload more than doubles on my quad-core system.
OK, I've run some tests on a 16 core Opteron system, both sysbench with
MySQL 5.33 (set
An experimental patch for Xen allows guests to place their vcpu_info
structs anywhere. We try to use this to place the vcpu_info into the
PDA, which allows direct access.
If this works, then switch to using direct access operations for
irq_enable, disable, save_fl and restore_fl.
Signed-off-by:
Ulrich Drepper wrote:
Nick Piggin wrote:
What I found is that, on this system, MADV_FREE performance improvement
was in the noise when you look at it on top of the MADV_DONTNEED glibc
and down_read(mmap_sem) patch in sysbench.
I don't want to judge the numbers since I cannot but I want to
This patch uses the lazy-mmu hooks to batch mmu operations where
possible. This is primarily useful for batching operations applied to
active pagetables, which happens during mprotect, munmap, mremap and
the like (mmap does not do bulk pagetable operations, so it isn't
helped).
Signed-off-by:
If we receive a partially checksummed packed, we need to work out how
much of it was checksummed based on its protocol. The ideal would be
that Xen could tell us how much is checksummed, but in the meantime
we'll make do with this.
[ This is a separate patch for review; will be folded into
Add Xen 'grant table' driver which allows granting of access to
selected local memory pages by other virtual machines and,
symmetrically, the mapping of remote memory pages which other virtual
machines have granted access to.
This driver is a prerequisite for many of the Xen virtual device
Make the appropriate hypercalls to halt and reboot the virtual machine.
Signed-off-by: Jeremy Fitzhardinge <[EMAIL PROTECTED]>
Acked-by: Chris Wright <[EMAIL PROTECTED]>
---
arch/i386/xen/enlighten.c | 43 +++
arch/i386/xen/smp.c |4 +---
2
When setting up the initial pagetable, which includes mappings of all
low physical memory, ignore a mapping which tries to set the RW bit on
an RO pte. An RO pte indicates a page which is part of the current
pagetable, and so it cannot be allowed to become RW.
Once xen_pagetable_setup_done is
Xen implements interrupts in terms of event channels. This patch maps
an event through an event channel to an irq, and then feeds it through
the normal interrupt path, via a Xen irq_chip implementation.
Signed-off-by: Jeremy Fitzhardinge <[EMAIL PROTECTED]>
Signed-off-by: Chris Wright <[EMAIL
On Sat, 5 May 2007 01:22:05 +0200
Andi Kleen <[EMAIL PROTECTED]> wrote:
> > 2.6.21:
> >
> > akpm2:/home/akpm# opreport -l /boot/vmlinux-$(uname -r) | head -50
> > opreport error: No sample file found: try running opcontrol --dump
> > or specify a session containing sample files
>
> For me it
Xen pagetable handling, including the machinery to implement direct
pagetables. This means that pte update operations are intercepted so
that pseudo-physical addresses can be converted to machine addresses.
It also implements late pinning/early unpinning so that pagetables are
initialized while
Hi Andi,
This series of patches implements the Xen paravirt-ops interface.
It applies to 2.6.21-git3 + ff patches-2.6.21-git3-070501-1.tar.gz.
Changes since the last posting:
- reviews of xenbus (me), netfront (hch, rusty, herbert xu) and
blockfront (hch), with most comments addressed.
Add Xen support for preemption. This is mostly a cleanup of existing
preempt_enable/disable calls, or just comments to explain the current
usage.
Signed-off-by: Jeremy Fitzhardinge <[EMAIL PROTECTED]>
Acked-by: Chris Wright <[EMAIL PROTECTED]>
---
arch/i386/xen/Kconfig |2
The network device frontend driver allows the kernel to access network
devices exported exported by a virtual machine containing a physical
network device driver.
* * *
use skb.cb for storing private data
Netfront's use of nh.raw and h.raw for storing page+offset is a bit
hinky, and it breaks
Implement xen_sched_clock, which returns the number of ns the current
vcpu has been actually in an unstolen state (ie, running or blocked,
vs runnable-but-not-running, or offline) since boot.
Signed-off-by: Jeremy Fitzhardinge <[EMAIL PROTECTED]>
Acked-by: Chris Wright <[EMAIL PROTECTED]>
Cc:
This is a fairly straightforward Xen implementation of smp_ops.
Xen has its own IPI mechanisms, and has no dependency on any
APIC-based IPI. The smp_ops hooks and the flush_tlb_others pv_op
allow a Xen guest to avoid all APIC code in arch/i386 (the only apic
operation is a single apic_read for
Implement a Xen back-end for hvc console.
From: Gerd Hoffmann <[EMAIL PROTECTED]>
Signed-off-by: Jeremy Fitzhardinge <[EMAIL PROTECTED]>
Signed-off-by: Chris Wright <[EMAIL PROTECTED]>
Cc: Olof Johansson <[EMAIL PROTECTED]>
---
arch/i386/xen/events.c |3 -
drivers/Makefile |3 +
Time is implemented by using a clocksource which is driven from the
hypervisor's nanosecond timebase. Xen implements time by
extrapolating from known timestamps using the tsc; the hypervisor is
responsible for making sure that the tsc is constant rate and
synchronized between vcpus.
Add early printk support via hvc console, enable using
"earlyprintk=xen" on the kernel command line.
From: Gerd Hoffmann <[EMAIL PROTECTED]>
Signed-off-by: Jeremy Fitzhardinge <[EMAIL PROTECTED]>
Signed-off-by: Chris Wright <[EMAIL PROTECTED]>
Acked-by: Ingo Molnar <[EMAIL PROTECTED]>
---
Add the "nosegneg" fake capabilty to the vsyscall page notes. This is
used by the runtime linker to select a glibc version which then
disables negative-offset accesses to the thread-local segment via
%gs. These accesses require emulation in Xen (because segments are
truncated to protect the
On May 4 2007 23:20, David Johnson wrote:
>
>longhaul: VIA C3 'Nehemiah C' [C5P] CPU detected. Powersaver supported.
>longhaul: Using ACPI support.
>
>It seems that longhaul on my system is 'using ACPI support' whereas on yours
>it is 'using northbridge support'. I'm getting lockups after
On Fri, 2007-05-04 at 11:27 -0700, Christoph Lameter wrote:
>
> Not sure where to go here. Increasing the per cpu slab size may hold off
> the issue up to a certain cpu cache size. For that we would need to
> identify which slabs create the performance issue.
>
> One easy way to check that
Hi,
If multiple threads are parked on epoll_wait (on a single epoll fd) and
events become available, epoll performs a wake up of all threads of the
poll wait list, causing a thundering herd of processes trying to grab
the eventpoll lock.
This patch addresses this by using exclusive waiters (wake
On 5/4/07, Davide Libenzi <[EMAIL PROTECTED]> wrote:
This is a pretty specific case, that is not very typical to find in the
usual common event loop dispatch application design.
This is where you are very wrong. Yes, it's rare in the Unix world
because non-trivial programs cannot implement
On May 4 2007 15:49, john stultz wrote:
>
>> Switching from acpi_pm+performance to acpi_pm+ondemand also
>> locks up after a few minutes.
>
>Yep. Sounds like an ondemand issue. Thanks for verifying this for me.
Nah, it also happens with cpufreq_powersave. I just need to check
through some
Quentin Godfroy wrote:
> + elf_ppnt = elf_phdata;
> + for (i = 0; i< loc->elf_ex.e_phnum; i++, elf_ppnt++)
> + if (elf_ppnt->p_type == PT_PHDR) {
> + phdr_addr = elf_ppnt->p_vaddr;
>
Won't this break with ET_DYN executables? And besides, isn't this the
On Sat, 2007-05-05 at 01:18 +0200, Andi Kleen wrote:
> On Friday 04 May 2007 23:29:04 john stultz wrote:
> > One of the 2.6.21 regressions was Guilherme's problem seeing his box
> > lock up when the system detected an unstable TSC and dropped back to
> > using the HPET.
> >
> > In digging deeper,
On Fri, 4 May 2007 10:09:21 -0400
Quentin Godfroy <[EMAIL PROTECTED]> wrote:
> On a dynamic ELF executable, the current kernel loader gives to the
> interpreter (in the AUXV vector) the AT_PHDR argument as :
> offset_of_phdr_in_file + first address.
>
> It can be wrong for an executable where
Hi Satyam,
Satyam Sharma wrote:
Eeks ... no, wait. You found a (two, actually) bug alright, but fixed
it wrong. When we fail a write, we *must* add it to the corrupted list
and _then_ attempt to retry. So, the "if (++tries <= 5)" applies to
"if (!err) goto retry;" and not to the
On Friday 04 May 2007 23:44:08 Andrew Morton wrote:
> On Fri, 04 May 2007 14:29:04 -0700
> john stultz <[EMAIL PROTECTED]> wrote:
>
> > One of the 2.6.21 regressions was Guilherme's problem seeing his box
> > lock up when the system detected an unstable TSC and dropped back to
> > using the HPET.
On Friday 04 May 2007 23:33:47 Andrew Morton wrote:
> On Fri, 4 May 2007 13:42:12 -0700
>
> 2.6.20:
>
> akpm2:/home/akpm> opcontrol --start-daemon
> /usr/bin/opcontrol: line 1098: /dev/oprofile/0/enabled: No such file or
> directory
> /usr/bin/opcontrol: line 1098: /dev/oprofile/0/event: No
On Friday 04 May 2007 23:29:04 john stultz wrote:
> One of the 2.6.21 regressions was Guilherme's problem seeing his box
> lock up when the system detected an unstable TSC and dropped back to
> using the HPET.
>
> In digging deeper, we found the HPET is not actually incrementing on
> this system.
Eric W. Biederman wrote:
>
> My notes show 0x5c reserved for additional apm_bios_info, although
> of the top of my head I don't know how realistic that is.
>
> 0x1e4 does look available.
>
> It has been a long time since I made that choice, and I do see that
> looking at struct screen_info I
On 5/4/07, Rafael J. Wysocki <[EMAIL PROTECTED]> wrote:
The change of the hibernation/suspend code ordering made before 2.6.21 has
caused some systems to have problems related to ACPI. In particular, the
'platform' hibernation mode doesn't work any more on some systems.
It seems that
Fixes suggested by Andrew
---
include/linux/slab.h | 12
mm/slub.c| 32 +---
2 files changed, 33 insertions(+), 11 deletions(-)
Index: slub/mm/slub.c
===
---
Antonino:
Can you try two tests (with my patch applied):
1. comment out the tty_flush_buffer() call in tty_ldisc_flush() and test
2. uncomment (reenable) the above call and comment out the
tty_flush_buffer() call in tty_ioctl() and test
Thanks,
Paul
-
To unsubscribe from this list: send the
Hi,
I attach a diff against 2.6.21 for adding wpa support for airo driver.
In then end of 2005 I manage to make work wpa but the code was really
ugly. I manage to find some time to clean it.
To support wpa, a new interface of the firmware should be used. This
interface is incompatible with
On Fri, 2007-05-04 at 23:02 +0200, Jan Engelhardt wrote:
> On May 4 2007 13:37, john stultz wrote:
> >>
> >> I found that setting the cpufreq governor to ondemand making the box
> >> lock up solid in 2.6.20.2 and 2.6.21 after a few seconds. Sysrq
> >> does not work anymore, and the last
1 - 100 of 1169 matches
Mail list logo
