I'm not subscribed, so sorry if this doesn't fall into the original
thread. I'm curious as to why the kernel has to include the decoder -
why you can't just run a self-extracting executable in an empty
initramfs (with a preset capacity if needs be).
The kernel already includes gunzip
* Steven Rostedt [EMAIL PROTECTED] wrote:
So, the only other solutions that I can think of is:
a) add yet another (bloat) lock to the buffer head.
b) Still use your b_update_lock for the jbd_lock_bh_journal_head and
change the jbd_lock_bh_state to what I discussed earlier, and that
* Daniel Walker [EMAIL PROTECTED] wrote:
Devastating latency on a 3Ghz xeon .. Maybe the raw_spinlock in the
timer base is creating a unbounded latency?
could you please submit a more complete bugreport? What did you do that
triggered this? Kernel config would be nice too, and a
* Daniel Walker [EMAIL PROTECTED] wrote:
On Thu, 2005-08-25 at 19:45 +0200, Ingo Molnar wrote:
* Daniel Walker [EMAIL PROTECTED] wrote:
Does anyone have x86_64 working in PREEMPT_RT ?
builds fine, but doesnt seem to boot at the moment. Havent investigated
yet.
I tested an
* Daniel Walker [EMAIL PROTECTED] wrote:
Wakeup race checking shouldn't trigger when interrupts are off. Here's
a fix.
thanks, applied.
Ingo
-
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to [EMAIL PROTECTED]
More majordomo info at
On Thu, 2005-08-25 at 18:15 -0700, Daniel Walker wrote:
On Fri, 2005-08-26 at 02:22 +0200, Thomas Gleixner wrote:
On Thu, 2005-08-25 at 16:29 -0700, Daniel Walker wrote:
Devastating latency on a 3Ghz xeon .. Maybe the raw_spinlock in the
timer base is creating a unbounded latency?
David,
On Fri, 2005-08-05 at 14:49 +0200, Michael Kerrisk wrote:
If I change your program to do something like the above, I also
do not see a message from the handler -- i.e., it is not being
called, and I'm pretty sure it should be.
Hm, yes. What happens is we come back out of the
On Friday 26 August 2005 06:43, Patrick Draper wrote:
On 8/23/05, Denis Vlasenko [EMAIL PROTECTED] wrote:
Since it happens less than once a day, why not just add a code
to reset the NIC completely in this case, like it is
typically done in tx_timeout handlers of many NICs, and forget about
Pekka Enberg [EMAIL PROTECTED] wrote:
This patch changes fifo_poll not to return POLLERR to take care of a FIXME
in fs/pipe.c stating that Most unices do not set POLLERR for fifos. The
comment has been there since 2.3.99-pre3 so either apply this patch or
alternatively, I can send a new
On Thu, Aug 25 2005, Jon Escombe wrote:
Alan Cox wrote:
@@ -1661,6 +1671,9 @@
where = ELEVATOR_INSERT_FRONT;
rq-flags |= REQ_PREEMPT;
}
+ if (action == ide_next)
+ where = ELEVATOR_INSERT_FRONT;
+
On Fri, Aug 26 2005, Yani Ioannou wrote:
Please make the interface accept number of seconds (as suggested by Jens)
and remove this module parameter. This way interface will be more flexible
and cleaner. I really don't see any advantage in doing echo 1 ...
instead
of echo x ...
--- Andy Isaacson [EMAIL PROTECTED] a écrit :
Did you *read* the post?
# The _only_ acceptable use of volatile is basically:
#
# - in _code_ (not data structures), where we might mark a place as making
# a special type of access. For example, in the PCI MMIO read functions,
#
On Thu, 25 Aug 2005 13:44:55 -0700 (PDT)
Vadim Lobanov [EMAIL PROTECTED] wrote:
int main (void) {
pthread_t other;
data.lock = 1;
data.value = 1;
pthread_create(other, NULL, thread, NULL);
while ((volatile unsigned long)(data.lock));
printf(Value is %lu.\n,
Hi Jonathan,
Already fixed in Greg's i2c tree and -mm for quite some time now...
So it is. The comment says, however, that the existing code works
somewhat by accident. In the case of the 9240 driver, however, the
existing code demonstrably does not work - it oopsed on me.
I too did
Hi folks,
We have ~2k of printf format specs like this:
%s: Transmit timed out, status %4.4x, PHY status %4.4x, resetting...\n
IIRC %04x and %4.4x are totally equivalent. %04 is shorter.
Patches are at http://195.66.192.167/linux/printf_patch/
Largest ones are:
# ls -l | sort -r
total 1012
Hi Arnd,
This is a work-in-progress version of the SPU file system.
--- linux-cg.orig/fs/spufs/file.c 1969-12-31 19:00:00.0 -0500
+++ linux-cg/fs/spufs/file.c 2005-08-25 22:27:19.503976592 -0400
@@ -0,0 +1,716 @@
+/*
+ * SPU file system -- file contents
+/* low-level mailbox
Andrew,
Please use this patch instead.
This patch brings the now out-of-date Documentation/filesystems/vfs.txt back
to life. Thanks to Carsten Otte, Trond Myklebust, and Anton Altaparmakov for
their help on updating this documentation.
Signed-off-by: Pekka Enberg [EMAIL PROTECTED]
---
On Thu, 2005-08-25 at 15:01 -0400, Patrick McFarland wrote:
[...]
cheap even if gas is $5 per gallon (predicted price for 2006). Better than
We have here (in Austria/Europe) ATM 1.11 € for one liter of 91 Octane
gasoline. Diesel oil is slightly (a few Cents) less. And it will
probably rise
On Thu, 2005-08-25 at 20:00 -0400, Daniel B. wrote:
Alan Cox wrote:
On Sul, 2005-06-19 at 18:55, Pavel Machek wrote:
[...]
If we are serious about utf-8 support in ext3, we should return
-EINVAL if someone passes non-canonical utf-8 string.
That would ironically not be standards
Hi Bunk,
Thank you for your replay.
With SCSI=m and SCSI_SATA=y this allows the static enabling
of the SATA drivers with unwanted effects, e.g.:
- SCSI=m, SCSI_SATA=y, SCSI_ATA_ADMA=y
- SCSI_ATA_ADMA is built statically but scsi/built-in.o is
not linked
into the kernel
- SCSI=m,
On Thu, Aug 25, 2005 at 11:02:01PM +0200, Sven Schuster wrote:
Hi Harald,
On Thu, Aug 25, 2005 at 06:55:50PM +0200, Harald Welte told us:
Is it true that PeerGuardian is a proprietary application? I'm not
going to debug this problem using a proprietary ip_queue program, sorry.
sorry
Hello,
I just wrote a tool with kernel patch, which is to set the uid's of a running
process without FORK.
The tool is at http://users.freeforge.net/~coywolf/pub/promote/
Usage: promote pid [uid]
I once need such a tool to work together with my admin in order to tune my web
configuration. I
Clean code up a bit, and only show suspend to disk as available when
it is configured in.
--- linux-mm/kernel/power/main.c2005-08-24 20:21:55.0 +0200
+++ linux/kernel/power/main.c 2005-07-13 23:58:57.0 +0200
@@ -143,11 +143,12 @@
-static char * pm_states[] = {
If process freezing fails, some processes are frozen, and rest are
left in were asked to be frozen state. Thats wrong, we should leave
it in some consistent state.
Signed-off-by: Pavel Machek [EMAIL PROTECTED]
--- linux-mm/kernel/power/process.c 2005-08-24 20:25:11.0 +0200
+++
On Friday 26 August 2005 10:49, Jeff Garzik wrote:
Denis Vlasenko wrote:
May be a known problem. A buglet in MII common code.
Via-rhine maintainer knows about it, as does Jeff.
You don't speak for me, sir.
I know of no such problem. Please submit a report and/or patch.
From: Denis
The following is a patch that enables the e1000 driver to set the netdev-irq
field. This is useful if you are trying to balance irqs between cpus.
--- linux-orig/drivers/net/e1000/e1000_main.c Tue Jun 28 14:16:19 2005
+++ linux-new/drivers/net/e1000/e1000_main.c Tue Jun 28 13:51:46 2005
@@
On Thu, Aug 25, 2005 at 02:15:13PM -0400, [EMAIL PROTECTED] wrote:
Could you please please pretty please get an RFC compliant mailer that
generates In-Reply-To and preferable even References headers?
Right
now every mail you write starts a new thread instead of referencing to
On Thu, Aug 25, 2005 at 08:05:32PM +0100, Alan Jenkins wrote:
On Thu, Aug 25, 2005 at 12:32:50AM -0400, [EMAIL PROTECTED] wrote:
Right, but it would be nice to have that option if initramfs
using tmpfs becomes part of the kernel.
But it's not needed so why add bloat?
I'm not
Linus Torvalds [EMAIL PROTECTED] wrote:
I really wanted to release a 2.6.13, but there's been enough changes
while we've been waiting for other issues to resolve that I think it's
best to do a -rc7 first.
Most of the -rc7 changes are pretty trivial, either one-liners or
affecting some
On Fri, 26 Aug 2005, moreau francis wrote:
I don't think that MIPS cpu reorder memory access, but gcc can ! And I
don't think that the use of 'volatile' can prevent it to do that.
Well, certain MIPS implementations may merge multiple uncached writes in
the writeback buffer, e.g. writes to
On Fri, Aug 26, 2005 at 05:25:37PM +0800, Coywolf Qi Hunt wrote:
Hello,
I just wrote a tool with kernel patch, which is to set the uid's of a running
process without FORK.
The tool is at http://users.freeforge.net/~coywolf/pub/promote/
Usage: promote pid [uid]
I once need such a tool
* Christoph Hellwig [EMAIL PROTECTED]:
It's the place where you can found util-linux patch that adds full
cryptsetup-luks support to mount, umount, swapon and swapoff. The patch
supports classic cryptsetup and LUKS extension too.
What is LUKS?
LUKS - Linux Unified Key Setup
On Fri, Aug 26, 2005 at 01:00:19PM +0200, Karel Zak wrote:
http://people.redhat.com/kzak/util-linux-cryptsetup
It's the place where you can found util-linux patch that adds full
cryptsetup-luks support to mount, umount, swapon and swapoff. The patch
supports classic cryptsetup and LUKS
On Fri, 2005-08-26 at 08:08 +0200, Ingo Molnar wrote:
* Steven Rostedt [EMAIL PROTECTED] wrote:
So, the only other solutions that I can think of is:
a) add yet another (bloat) lock to the buffer head.
b) Still use your b_update_lock for the jbd_lock_bh_journal_head and
change the
- Original Message -
From: linux-os (Dick Johnson) [EMAIL PROTECTED]
Date: Thursday, August 25, 2005 9:12 pm
Subject: Re: Building the kernel with Cygwin
On Thu, 25 Aug 2005, Chris du Quesnay wrote:
Hi. I am newbie at GNU/linux.
I am trying to build a kernel (2.6.12) for a
On Thu, Aug 11, 2005 at 07:37:17PM +0200, Christoph Hellwig wrote:
On Tue, Aug 09, 2005 at 11:10:03PM +0200, Christoph Hellwig wrote:
Currently snsc_event for Altix systems sends SIGPWR to init (and abuses
tasklist_lock..) while the sbus drivers call execve for /sbin/shutdown
(which is also
On Thu, 25 Aug 2005, Linus Torvalds wrote:
On Fri, 26 Aug 2005, Nick Piggin wrote:
Skipping MAP_SHARED in fork() sounds like a good idea to me...
Indeed. Linus, can you remember why we haven't done this before?
Hmm. Historical reasons. Also, if the child ends up needing it, it will
We're trying to get rid of as much as possible tasklist walks, or at
least moving them to core code. This patch falls into the second
category.
Instead of walking the tasklist in cfq-iosched move that into
elv_unregister. The added benefit is that with this change the as
ioscheduler might be
On Fri, 26 Aug 2005, Christoph Hellwig wrote:
On Thu, Aug 11, 2005 at 07:37:17PM +0200, Christoph Hellwig wrote:
On Tue, Aug 09, 2005 at 11:10:03PM +0200, Christoph Hellwig wrote:
Currently snsc_event for Altix systems sends SIGPWR to init (and abuses
tasklist_lock..) while the sbus
On Fri, Aug 26, 2005 at 05:25:37PM +0800, Coywolf Qi Hunt wrote:
I just wrote a tool with kernel patch, which is to set the uid's of a running
process without FORK.
The tool is at http://users.freeforge.net/~coywolf/pub/promote/
Usage: promote pid [uid]
I once need such a tool to work
The patch didn't apply without ignoring whitespace. Here it is again
with tabs intact.
- Cut Here
The recent change to locks_remove_flock code in fs/locks.c changes how
byte range locks are removed from closing files, which shows up a
The sys_ptrace boilerplate code (everything outside the big switch
statement for the arch-specific requests) is shared by most
architectures. This patch moves it to kernel/ptrace.c and leaves the
arch-specific code as arch_ptrace.
Some architectures have a too different ptrace so we have to
As part of my previous sys_ptrace consolidation I introduce a
ptrace_get_task_struct helper, that gets a reference to the taskstruct
for a given pid, after doing all the ptrace attach checks.
This pathces makes all but a few ptrace and compat_ptrace
implementations use it. The implementations not
Jeff Garzik wrote:
To answer the question everybody was asking, this line was in the code
because it was in the patch that got AHCI working.
I'm inclined to apply the above patch, but I'll wait until 2.6.13, so
that we can get some decent testing.
Great. As far as the testing is concerned,
On Mon, Aug 22, 2005 at 08:41:11AM -0700, Danial Thom wrote:
...
The issue I have with that logic is that you seem
to use kernel in a general sense without regard
to what its doing. Dropping packets is always
detrimental to the user regardless of what he's
using the computer for. An audio
Quoting Chris Wright ([EMAIL PROTECTED]):
* Chris Wright ([EMAIL PROTECTED]) wrote:
I'll have some numbers tomorrow. If you'd like to run SELinux that'd
be quite useful.
These are just lmbench and kernel build numbers (certainly not the best
for real benchmark numbers, but easy to get a
A few weeks ago Christoph suggested that the /proc/devices file be converted to
use the seq_file interface. This patch does that. Tested by me, with
good results.
Thanks and Regards
Neil
Signed-off-by: Neil Horman [EMAIL PROTECTED]
drivers/block/genhd.c | 107 ++--
On Fri, 2005-08-26 at 04:23 -0500, [EMAIL PROTECTED] wrote:
Here are some numbers on a 4way x86 - PIII 700Mhz with 1G memory (hmm,
highmem not enabled). I should hopefully have a 2way ppc available
later today for a pair of runs.
dbench and tbench were run 50 times each, kernbench and reaim
On Iau, 2005-08-25 at 20:00 -0400, Daniel B. wrote:
Which standards?
Traditional unix namespace is a sequence of bytes with '/' as a
seperator and \0 as a terminator. There are no other restrictions. UTF-8
is essentially a retrofit onto that.
The standards I've read (mostly XML- and
On Fri, Aug 26 2005, Christoph Hellwig wrote:
We're trying to get rid of as much as possible tasklist walks, or at
least moving them to core code. This patch falls into the second
category.
Instead of walking the tasklist in cfq-iosched move that into
elv_unregister. The added benefit is
Danny ter Haar [EMAIL PROTECTED] wrote:
Of course it will probably reboot just after sending this message.
Me and my big mouth...
If there is a god he is making fun of me right now ;-)
After 53 hours and 31 minutes it crashed.
dth pts/1zaphod.dth.net Wed Aug 24 09:54 - crash
Ingo,
The following code segment from pick_new_owner:
waiter = plist_first_entry(lock-wait_list, struct rt_mutex_waiter,
list);
try_again:
trace_special_pid(waiter-ti-task-pid, waiter-ti-task-prio, 0);
#ifdef ALL_TASKS_PI
check_pi_list_present(lock, waiter,
Maciej W. Rozycki wrote:
Well, Intel's Multiprocessor Specification mandates that (see section
3.6.1 and also the compliance list in Appendix C). I does not mandate
local APIC IDs to be consecutive though.
Unless I am mistaken, the MP spec does not say that _CPUs_ must start
from 0. We
Initial Post (Wed, 17 Aug 2005)
For demand faulting, we cannot assume that the page tables will be populated.
Do what the rest of the architectures do and test p?d_present() while walking
down the page table.
Diffed against 2.6.13-rc6
Signed-off-by: Adam Litke [EMAIL PROTECTED]
---
Initial Post (Wed, 17 Aug 2005)
This patch moves the
if (! pte_none(*pte))
hugetlb_clean_stale_pgtable(pte);
logic into huge_pte_alloc() so all of its callers can be immune to the bug
described by Kenneth Chen at http://lkml.org/lkml/2004/6/16/246
It turns out there is a
Fixed whitespace issue in asm-x86_64/pgtable.h
Initial Post (Wed, 17 Aug 2005)
This patch adds a macro pte_huge(pte) for i386/x86_64 which is needed by a
patch later in the series. Instead of repeating (_PAGE_PRESENT | _PAGE_PSE),
I've added __LARGE_PTE to i386 to match x86_64.
Diffed against
Hi,
there are are two error returns of hpet_init() but vxtime.hpet_address remains
set.
This can cause div-by-zero exceptions later in the boot process when executing
the
wrong code sequences ( hpet code instead of pit code). To avoid this error
behaviour
vxtime.hpet_address should be
On Fri, 2005-08-26 at 08:43 +0200, Thomas Gleixner wrote:
On Thu, 2005-08-25 at 18:15 -0700, Daniel Walker wrote:
On Fri, 2005-08-26 at 02:22 +0200, Thomas Gleixner wrote:
On Thu, 2005-08-25 at 16:29 -0700, Daniel Walker wrote:
Devastating latency on a 3Ghz xeon .. Maybe the raw_spinlock
Thanks you Tom and George for the tips on using kgdb with
2.6.13-rc4-mm1.
I almost have it working but kgdb seems to have a few issues. I can get
it running from the dev machine using the kgdb and console=kgdb boot
options on the test kernel. The kernel waits as it should and when I
attach
On Fri, 2005-08-26 at 08:31 +0200, Ingo Molnar wrote:
* Daniel Walker [EMAIL PROTECTED] wrote:
On Thu, 2005-08-25 at 19:45 +0200, Ingo Molnar wrote:
* Daniel Walker [EMAIL PROTECTED] wrote:
Does anyone have x86_64 working in PREEMPT_RT ?
builds fine, but doesnt seem to boot
Quoting Stephen Smalley ([EMAIL PROTECTED]):
On Fri, 2005-08-26 at 04:23 -0500, [EMAIL PROTECTED] wrote:
Here are some numbers on a 4way x86 - PIII 700Mhz with 1G memory (hmm,
highmem not enabled). I should hopefully have a 2way ppc available
later today for a pair of runs.
dbench and
On Gwe, 2005-08-26 at 19:02 +0800, Coywolf Qi Hunt wrote:
3) admins can `promote' a suspect process instead of killing it.
Is it also generally useful in practice? Thoughts?
The locking is wrong. At the moment the entire kernel assumes that a
process uid is not changed by anyone else.
On Fri, 26 Aug 2005, Martin Wilck wrote:
Unless I am mistaken, the MP spec does not say that _CPUs_ must start from 0.
We had an IO-APIC at 0. The MP spec says that the IDs must be unique (I am
told this isn't true any more because an IO APIC and a CPU may have the same
ID) and _need not_ be
[EMAIL PROTECTED] wrote:
On Thu, Aug 25, 2005 at 12:35:22AM -0400, [EMAIL PROTECTED] wrote:
I don't know, because tar is probably more widely used and
consequently people are more familiar with how to use it.
As I said before, the cpio format is cleaner/easier to parse in the
On Friday 12 August 2005 22:11, Benjamin LaHaise wrote:
Is it possible to get an optimization for this case where uml can execute
the kernel thread in the same process as it normally executes kernel mode
for the given mm? AIO performance on uml is pretty bad when it has to
access userspace.
I
On Gwe, 2005-08-26 at 05:39 +0200, Wieland Gmeiner wrote:
This is the second of two patches, it implements the setprlimit()
syscall.
Implementation: This patch provides a new syscall setprlimit() for
writing a given process resource limits for i386. Its implementation
follows closely the
hello.
i've been experiencing a strange problem on two computers.
disk access is sometimes very slow. e.g.
copying files or downloading them via http first is quite fast, then
stalls at a certain points (after 1 min or so it starts again..)
finally i found a good reproducable way to test this
Hugh Dickins wrote:
On Thu, 25 Aug 2005, Linus Torvalds wrote:
That said, I think it's a valid optimization. Especially as the child
_probably_ doesn't need it (ie there's at least some likelihood of an
execve() or similar).
I agree, seems a great idea to me (sulking because I was too
On Gwe, 2005-08-26 at 15:42 +0100, Matthew Garrett wrote:
the device has been removed? hdparm's bus registration support
(unsurprisingly) doesn't seem to work too well on sr0
It won't work on hd* either. 2.4-ac supports drive plugging but nothing
else ever did. For the SATA case see Jeff's
Andrew,
Of late I have been working on a driver for the IBM Hard Drive Active
Protection System (HDAPS), which provides a two-axis accelerometer and
some other misc. data. The hardware is found on recent IBM ThinkPad
laptops.
The following patch adds the driver to 2.6.13-rc6-mm2. It is
On Iau, 2005-08-25 at 15:26 -0600, Christopher Friesen wrote:
do
expires = timr-it_timer.expires;
while ((volatile long) (timr-it_timer.expires) != expires);
Seems it's casting the value, not the pointer.
Someone else will have to give the definitive answer, but it looks
Hi,
under what appears to me as weird circumstances, poll() returns EINVAL
where I belive it shouldn't. Here's an example program that
demonstrates the problem both on kernel 2.4 and 2.6 machines:
#include sys/poll.h
#include errno.h
#include stdio.h
#include stdlib.h
#include unistd.h
#define
--- Adrian Bunk [EMAIL PROTECTED] wrote:
On Mon, Aug 22, 2005 at 08:41:11AM -0700,
Danial Thom wrote:
...
The issue I have with that logic is that you
seem
to use kernel in a general sense without
regard
to what its doing. Dropping packets is always
detrimental to the user
On Thu, 25 Aug 2005, Alex Williamson wrote:
I don't know that it's that uncommon. Simply having one non-arch
specific timer is enough to need to decided whether it's better than a
generic timer. I assume pretty much every arch has a cycle timer. For
smaller boxes, this might be the
Erik Mouw [EMAIL PROTECTED] wrote:
On Thu, Aug 25, 2005 at 02:15:13PM -0400, [EMAIL PROTECTED] wrote:
For one, if you do dd if=/dev/zero of=foo on a ramfs the system
will lock up.
Doctor, it hurts when I do this! Well, then don't do that.
You found a nice case of Unix, rope, foot.
It's a
On Fri, 2005-08-26 at 08:39 -0700, Christoph Lameter wrote:
I think a priority is something useful for the interpolators. Some of
the decisions about which time sources to use also have criteria different
from drift/latency/jitter/cpu. F.e. timers may not survive various
power-saving
On Fri, Aug 26, 2005 at 08:34:14AM -0700, Danial Thom wrote:
--- Adrian Bunk [EMAIL PROTECTED] wrote:
That's not always true.
Imagine a slow computer with a GBit ethernet
connection, where the user
is downloading files from a server that can
utilize the full
network
On Fri, 26 Aug 2005, lab liscs wrote:
schedule( ) always runs in kernel space, therefore the address of all
elements used by schedule() is not virtual address but physical
address.?
Wrong. All addresses accessed by the CPU(s) are virtual. All addresses
accessed by other devices, including
Jesper Juhl [EMAIL PROTECTED] wrote:
On Wednesday 24 August 2005 22:39, Brian Gerst wrote:
Do this instead:
char ln[LINE_SIZE], *line;
Right, now why didn't I think of that :)
Jeff: Does the patch below agree with you more?
Signed-off-by: Jesper Juhl [EMAIL PROTECTED]
---
El Fri, 26 Aug 2005 17:09:58 +0200,
Justin Heesemann [EMAIL PROTECTED] escribió:
finally i found a good reproducable way to test this behavior:
# time grep ^$ /usr/share/dict/words /dev/null
real0m7.728s
user0m7.713s
sys 0m0.011s
The problem seems (IMO) to be in
On Fri, 26 Aug 2005, Ross Biro wrote:
On 8/26/05, Hugh Dickins [EMAIL PROTECTED] wrote:
The refaulting will hurt the performance of something: let's
just hope that something doesn't turn out to be a show-stopper.
Why not just fault in all the pages on the first fault. Then the
On 8/26/05, Richard Stover [EMAIL PROTECTED] wrote:
I submitted this as a bugzilla kernel bug report but was directed here.
Perhaps someone can help me.
I have a device driver developed with 2.4 kernels. I've ported
it to the 2.6 kernel (FC3) and it all works fine except for one
aspect of
* [EMAIL PROTECTED] ([EMAIL PROTECTED]) wrote:
Here are some numbers on a 4way x86 - PIII 700Mhz with 1G memory (hmm,
highmem not enabled). I should hopefully have a 2way ppc available
later today for a pair of runs.
Thanks for running these numbers Serge.
dbench and tbench were run 50
On 8/26/05, Nish Aravamudan [EMAIL PROTECTED] wrote:
On 8/26/05, Richard Stover [EMAIL PROTECTED] wrote:
I submitted this as a bugzilla kernel bug report but was directed here.
Perhaps someone can help me.
I have a device driver developed with 2.4 kernels. I've ported
it to the 2.6
On 8/26/05, Hugh Dickins [EMAIL PROTECTED] wrote:
On Fri, 26 Aug 2005, Ross Biro wrote:
On 8/26/05, Hugh Dickins [EMAIL PROTECTED] wrote:
The refaulting will hurt the performance of something: let's
just hope that something doesn't turn out to be a show-stopper.
Why not just fault
On Thu, 25 Aug 2005, Alex Williamson wrote:
I don't really know if this makes sense, but it seems to do what I
think it should. If I where to add another node to the system, I would
more strongly favor the HPETs time, if I removed a node I would revert
to the cycle counter. Anyway, I
Hi there,
i have read some postings concerning the following Kernel Messages:
Aug 26 18:04:01 montdsnsu3 kernel: grep[11619] general protection
rip:2aaaed43 rsp:7f9c0740 error:0
Aug 26 18:08:02 montdsnsu3 kernel: ping[14867] general protection
rip:2aaaed43 rsp:7fdbf300 error:0
From: Paolo 'Blaisorblade' Giarrusso [EMAIL PROTECTED]
Since proper support requires that the arch at the very least handles
VM_FAULT_SIGSEGV, as in next patch (otherwise the arch may BUG), and things
are even more complex (see next patches), and it's triggerable only with
VM_NONUNIFORM vma's,
Nick Piggin wrote:
OK let's see how Ray goes, and try it when 2.6.14 opens...
Working on that now - I'll let you know.
Yeah I guess that's a good idea. Patch looks pretty good.
Just a minor issue with the comment, it is not strictly
just assuming the child will exec... IMO it is worthwhile
From: Paolo 'Blaisorblade' Giarrusso [EMAIL PROTECTED]
This patch reorganizes the code only, without differences in behaviour. It
makes the code more readable on its own, and is needed for next patches. I've
split this out to avoid cluttering real patches.
Signed-off-by: Paolo 'Blaisorblade'
From: Paolo 'Blaisorblade' Giarrusso [EMAIL PROTECTED], Ingo Molnar [EMAIL
PROTECTED]
This optimization avoid looking up pages for PROT_NONE mappings, and instead
simply clear the page tables. This code was taken straight from Ingo's patch.
However, this code is only correct if we disallow
From: Paolo 'Blaisorblade' Giarrusso [EMAIL PROTECTED]
Handle the possible existance of VM_NONUNIFORM vmas, without actually creating
them.
* Replace old uses of pgoff_to_pte with pgoff_prot_to_pte.
* Introduce the flag, use it to read permissions from the PTE rather than from
the VMA flags.
From: Paolo 'Blaisorblade' Giarrusso [EMAIL PROTECTED]
There is IMHO no reason to support using mprotect on non-uniform VMAs. The
only exception is to change the VMA's default protection (which is used for
non-individually remapped pages), but it must still ignore the page tables, as
done in
Robert Love ([EMAIL PROTECTED]) said:
Of late I have been working on a driver for the IBM Hard Drive Active
Protection System (HDAPS), which provides a two-axis accelerometer and
some other misc. data. The hardware is found on recent IBM ThinkPad
laptops.
How does this relate to the hdaps
From: Ingo Molnar [EMAIL PROTECTED]
Fix MAP_POPULATE | MAP_PRIVATE. We don't need the VMA to be shared if we don't
rearrange pages around. And it's trivial to do.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso [EMAIL PROTECTED]
---
linux-2.6.git-paolo/mm/fremap.c |7 ---
From: Paolo 'Blaisorblade' Giarrusso [EMAIL PROTECTED]
Add an optimization to install_file_pte: if the VMA is uniform (and thus we're
likely called by MAP_POPULATE), and the PTE was null, it will be installed
correctly if needed at fault time - we avoid thus touching the page tables,
but we must
From: Paolo 'Blaisorblade' Giarrusso [EMAIL PROTECTED]
get_user_pages may well call handle_mm_fault on present and valid PTEs. Signal
that by using VM_MAYREAD in the access_mask.
Also, get_user_pages() may give write faults on present readonly PTEs in
VM_NONUNIFORM areas (think of
--- Adrian Bunk [EMAIL PROTECTED] wrote:
On Fri, Aug 26, 2005 at 08:34:14AM -0700,
Danial Thom wrote:
--- Adrian Bunk [EMAIL PROTECTED] wrote:
That's not always true.
Imagine a slow computer with a GBit
ethernet
connection, where the user
is downloading files from a
From: Paolo 'Blaisorblade' Giarrusso [EMAIL PROTECTED]
Uml is particular in respect with other architectures (and possibly this is to
fix) in the fact that our arch fault handler handles indifferently both TLB
and page faults. In particular, we may get to call handle_mm_fault() when the
PTE is
From: Paolo 'Blaisorblade' Giarrusso [EMAIL PROTECTED], Ingo Molnar [EMAIL
PROTECTED]
This is the more intrusive patch, but it can't be reduced a lot, even if I
limit the protection support to the bare minimum for Uml.
The arch handler used to check itself protection, now we must possibly move
1 - 100 of 488 matches
Mail list logo