Re: Linux 2.4.22-pre3

Re: Linux 2.4.22-pre3

[Margit Schubert-While]

[-- Attachment #1: Type: text/plain, Size: 51 bytes --]

Attached fixes Aschwin's problem.


         Margit

[-- Attachment #2: vicampatch --]
[-- Type: application/octet-stream, Size: 454 bytes --]

diff -Naur linux-2.4.22pre3/drivers/usb/vicam.c linux-2.4.22pre3mw0/drivers/usb/vicam.c
--- linux-2.4.22pre3/drivers/usb/vicam.c     2003-07-06 08:24:31.000000000 +0200
+++ linux-2.4.22pre3mw0/drivers/usb/vicam.c  2003-07-06 11:58:01.000000000 +0200
@@ -763,8 +763,8 @@
 static void
 vicam_close(struct video_device *dev)
 {
-	DBG("close\n");
 	struct vicam_camera *cam = (struct vicam_camera *) dev->priv;
+	DBG("close\n");
 
 
 	if (cam->is_removed) {

Re: Linux 2.4.22-pre3

[Marcelo Tosatti]

On Fri, 11 Jul 2003, Jim Gifford wrote:

> Ok,
>     Here is what I got from my check. Snort is the culprit. When it starts,
> it runs at 40mb. Each hour it adds 20 mb. So by the time the system is up
> for hours, it starts cosuming memory. Today after I restartd my system, I
> noticed every hour it was adding a mb of RAM.
>
> It started at 40mb and after 4 hours is up to 45mb.
>
> Thanx all for your help and assistance.

Ok, thanks for your report.

Try -pre5 8)

Re: Linux 2.4.22-pre3

[Jim Gifford]

Ok,
    Here is what I got from my check. Snort is the culprit. When it starts,
it runs at 40mb. Each hour it adds 20 mb. So by the time the system is up
for hours, it starts cosuming memory. Today after I restartd my system, I
noticed every hour it was adding a mb of RAM.

It started at 40mb and after 4 hours is up to 45mb.

Thanx all for your help and assistance.

----- Original Message ----- 
From: "Marcelo Tosatti" <marcelo@conectiva.com.br>
To: "Jim Gifford" <maillist@jg555.com>
Cc: "lkml" <linux-kernel@vger.kernel.org>
Sent: Friday, July 11, 2003 10:06 AM
Subject: Re: Linux 2.4.22-pre3


>
> Right. Waiting for more results.
>
> Please always forawrd lkml okay?
>
> On Fri, 11 Jul 2003, Jim Gifford wrote:
>
> > It locked up last night, but no error information. But I did notice a
> > massive memory leak. I will follow up on that today, (756MB).
> >
> > ----- Original Message -----
> > From: "Marcelo Tosatti" <marcelo@conectiva.com.br>
> > To: "Jim Gifford" <maillist@jg555.com>
> > Cc: <riel@redhat.com>; "Andrea Arcangeli" <andrea@suse.de>
> > Sent: Friday, July 11, 2003 5:56 AM
> > Subject: Re: Linux 2.4.22-pre3
> >
> >
> > > y
> > >
> > > On Thu, 10 Jul 2003, Jim Gifford wrote:
> > >
> > > > I have a question, look at the buffers. Should the buffers be
increasing
> > in
> > > > this manner??
> > >
> > > I think its correct - its just caching more data after time passes.
> > >
> > > RIel, Andrea?
> > >
> > > >
> > > > When I started the system this was my reading
> > > >
> > > > top - 01:39:10 up 5 min, 1 user, load average: 3.70, 1.78, 0.72
> > > > Tasks: 106 total, 3 running, 103 sleeping, 0 stopped, 0 zombie
> > > > Cpu0 : 30.8% user, 13.5% system, 55.8% nice, 0.0% idle
> > > > Cpu1 : 39.7% user, 18.6% system, 41.7% nice, 0.0% idle
> > > > Mem: 1033896k total, 195548k used, 838348k free, 7188k buffers
> > > > Swap: 265060k total, 0k used, 265060k free, 45132k cached
> > > >
> > > > Here is my current reading after being up 13 hours.
> > > >
> > > > top - 14:41:57 up 13:06,  2 users,  load average: 2.92, 3.26, 3.11
> > > > Tasks: 118 total,   5 running, 113 sleeping,   0 stopped,   0 zombie
> > > >  Cpu0 :  32.5% user,  10.8% system,  56.7% nice,   0.0% idle
> > > >  Cpu1 :  27.4% user,  12.1% system,  60.5% nice,   0.0% idle
> > > > Mem:   1033672k total,   577568k used,   456104k free,   137452k
buffers
> > > > Swap:   265060k total,        0k used,   265060k free,   167232k
cached
> > >
> > >
> >
>

Re: Linux 2.4.22-pre3

[Marcelo Tosatti]

Right. Waiting for more results.

Please always forawrd lkml okay?

On Fri, 11 Jul 2003, Jim Gifford wrote:

> It locked up last night, but no error information. But I did notice a
> massive memory leak. I will follow up on that today, (756MB).
>
> ----- Original Message -----
> From: "Marcelo Tosatti" <marcelo@conectiva.com.br>
> To: "Jim Gifford" <maillist@jg555.com>
> Cc: <riel@redhat.com>; "Andrea Arcangeli" <andrea@suse.de>
> Sent: Friday, July 11, 2003 5:56 AM
> Subject: Re: Linux 2.4.22-pre3
>
>
> > y
> >
> > On Thu, 10 Jul 2003, Jim Gifford wrote:
> >
> > > I have a question, look at the buffers. Should the buffers be increasing
> in
> > > this manner??
> >
> > I think its correct - its just caching more data after time passes.
> >
> > RIel, Andrea?
> >
> > >
> > > When I started the system this was my reading
> > >
> > > top - 01:39:10 up 5 min, 1 user, load average: 3.70, 1.78, 0.72
> > > Tasks: 106 total, 3 running, 103 sleeping, 0 stopped, 0 zombie
> > > Cpu0 : 30.8% user, 13.5% system, 55.8% nice, 0.0% idle
> > > Cpu1 : 39.7% user, 18.6% system, 41.7% nice, 0.0% idle
> > > Mem: 1033896k total, 195548k used, 838348k free, 7188k buffers
> > > Swap: 265060k total, 0k used, 265060k free, 45132k cached
> > >
> > > Here is my current reading after being up 13 hours.
> > >
> > > top - 14:41:57 up 13:06,  2 users,  load average: 2.92, 3.26, 3.11
> > > Tasks: 118 total,   5 running, 113 sleeping,   0 stopped,   0 zombie
> > >  Cpu0 :  32.5% user,  10.8% system,  56.7% nice,   0.0% idle
> > >  Cpu1 :  27.4% user,  12.1% system,  60.5% nice,   0.0% idle
> > > Mem:   1033672k total,   577568k used,   456104k free,   137452k buffers
> > > Swap:   265060k total,        0k used,   265060k free,   167232k cached
> >
> >
>

Re: Linux 2.4.22-pre3

[Benjamin Herrenschmidt]

On Fri, 2003-07-11 at 15:03, Geert Uytterhoeven wrote:
> On Sat, 5 Jul 2003, Marcelo Tosatti wrote:
> > Summary of changes from v2.4.22-pre2 to v2.4.22-pre3
> > ============================================
> > Benjamin Herrenschmidt <benh@kernel.crashing.org>:
> >   o ppc32: Update adbhid driver
> 
> This change breaks the build for Mac/m68k (cfr. 2.5.x). The patch below cures
> that, cfr. the similar so-far-unapplied patch for 2.5.x (it's CONFIG_ALL_PPC in
> 2.4.x and CONFIG_PPC_PMAC in 2.5.x, right)?

Right, I always forget you also use adbhid nowadays... sorry about that.

Ben.

Re: Linux 2.4.22-pre3

[Geert Uytterhoeven]

On Sat, 5 Jul 2003, Marcelo Tosatti wrote:
> Summary of changes from v2.4.22-pre2 to v2.4.22-pre3
> ============================================
> Benjamin Herrenschmidt <benh@kernel.crashing.org>:
>   o ppc32: Update adbhid driver

This change breaks the build for Mac/m68k (cfr. 2.5.x). The patch below cures
that, cfr. the similar so-far-unapplied patch for 2.5.x (it's CONFIG_ALL_PPC in
2.4.x and CONFIG_PPC_PMAC in 2.5.x, right)?

--snip--

ADB HID: Exclude PowerMac-specific things on classic Macs

--- linux-2.4.x/drivers/macintosh/adbhid.c	Tue Jul  8 13:30:28 2003
+++ linux-m68k-2.4.x/drivers/macintosh/adbhid.c	Fri Jul 11 14:39:56 2003
@@ -44,7 +44,9 @@
 #include <linux/pmu.h>
 
 #include <asm/machdep.h>
+#ifdef CONFIG_ALL_PPC
 #include <asm/pmac_feature.h>
+#endif
 
 #ifdef CONFIG_PMAC_BACKLIGHT
 #include <asm/backlight.h>
@@ -158,6 +160,7 @@
 		return;
 	case 0x3f: /* ignore Powerbook Fn key */
 		return;
+#ifdef CONFIG_ALL_PPC
 	case 0x7e: /* Power key on PBook 3400 needs remapping */
 		switch(pmac_call_feature(PMAC_FTR_GET_MB_INFO,
 			NULL, PMAC_MB_INFO_MODEL, 0)) {
@@ -167,6 +170,7 @@
 			keycode = 0x7f;
 		}
 		break;
+#endif /* CONFIG_ALL_PPC */
 	}
 
 	if (adbhid[id]->keycode[keycode])

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: Linux 2.4.22-pre3

[Larry McVoy]

On Tue, Jul 08, 2003 at 03:18:17PM -0400, Jeff Garzik wrote:
> On Tue, Jul 08, 2003 at 08:05:21PM +0200, Adrian Bunk wrote:
> > On Sun, Jul 06, 2003 at 06:05:27PM -0700, Larry McVoy wrote:
> > > On Sun, Jul 06, 2003 at 10:00:34PM -0300, Marcelo Tosatti wrote:
> > > > On Sun, 6 Jul 2003, Ben Collins wrote:
> > > > > Any chance you could be consistent in tagging the -pre's? Neither pre2,
> > > > > nor pre3 is tagged in BK, and thus, not tagged in CVS/SVN either.
> > > > 
> > > > I guess I have tagged -pre2 and -pre3:
> > > > 
> > > > Maybe I'm missing something?
> > > 
> > > Hmm.   Ben, look again in the CVS tree and make sure that the tags aren't
> > > there.  Maybe the converter screwed up?  
> > 
> > -pre2 and -pre3 are also missing at
> >   http://ftp.kernel.org/pub/linux/kernel/v2.4/testing/cset/
> 
> hrm.  Well, it's definitely tagged correctly in Marcelo's main 2.4 BK
> repo.

I think I've found the bug - it's in the code that collapses multiple 
changesets into one CVS checkin.  It looks like we are picking up 
tags only if the tag was on the last changeset in the sequence instead
of any changeset in the sequence.  We're fixing it.
-- 
---
Larry McVoy              lm at bitmover.com          http://www.bitmover.com/lm

Re: Linux 2.4.22-pre3

[Ben Collins]

On Thu, Jul 10, 2003 at 08:15:07AM -0700, Larry McVoy wrote:
> On Tue, Jul 08, 2003 at 03:18:17PM -0400, Jeff Garzik wrote:
> > On Tue, Jul 08, 2003 at 08:05:21PM +0200, Adrian Bunk wrote:
> > > On Sun, Jul 06, 2003 at 06:05:27PM -0700, Larry McVoy wrote:
> > > > On Sun, Jul 06, 2003 at 10:00:34PM -0300, Marcelo Tosatti wrote:
> > > > > On Sun, 6 Jul 2003, Ben Collins wrote:
> > > > > > Any chance you could be consistent in tagging the -pre's? Neither pre2,
> > > > > > nor pre3 is tagged in BK, and thus, not tagged in CVS/SVN either.
> > > > > 
> > > > > I guess I have tagged -pre2 and -pre3:
> > > > > 
> > > > > Maybe I'm missing something?
> > > > 
> > > > Hmm.   Ben, look again in the CVS tree and make sure that the tags aren't
> > > > there.  Maybe the converter screwed up?  
> > > 
> > > -pre2 and -pre3 are also missing at
> > >   http://ftp.kernel.org/pub/linux/kernel/v2.4/testing/cset/
> > 
> > hrm.  Well, it's definitely tagged correctly in Marcelo's main 2.4 BK
> > repo.
> 
> I think I've found the bug - it's in the code that collapses multiple 
> changesets into one CVS checkin.  It looks like we are picking up 
> tags only if the tag was on the last changeset in the sequence instead
> of any changeset in the sequence.  We're fixing it.

Thanks.

-- 
Debian     - http://www.debian.org/
Linux 1394 - http://www.linux1394.org/
Subversion - http://subversion.tigris.org/
Deqo       - http://www.deqo.com/

Re: Linux 2.4.22-pre3

[Marcelo Tosatti]

Jim,

Hum, so there are no memory problems.

Do you have the NMI oopser on? It might help us.

Also, do you remember when it started to crash like this? (which kernel
started to show the problem)

On Wed, 9 Jul 2003, Jim Gifford wrote:

> I ran memtest86 for 48 hours with no errors. (This was done about to week
> ago)
> ----- Original Message -----
> From: "Marcelo Tosatti" <marcelo@conectiva.com.br>
> To: "Jim Gifford" <maillist@jg555.com>
> Cc: "Alan Cox" <alan@lxorguk.ukuu.org.uk>; "lkml"
> <linux-kernel@vger.kernel.org>
> Sent: Wednesday, July 09, 2003 10:33 AM
> Subject: Re: Linux 2.4.22-pre3
>
>
> >
> >
> > On Tue, 8 Jul 2003, Jim Gifford wrote:
> >
> > > ----- Original Message -----
> > > From: "Alan Cox" <alan@lxorguk.ukuu.org.uk>
> > > To: "Jim Gifford" <maillist@jg555.com>
> > > Cc: "Marcelo Tosatti" <marcelo@conectiva.com.br>; "lkml"
> > > <linux-kernel@vger.kernel.org>
> > > Sent: Tuesday, July 08, 2003 3:23 PM
> > > Subject: Re: Linux 2.4.22-pre3
> > >
> > >
> > > > On Maw, 2003-07-08 at 22:36, Jim Gifford wrote:
> > > > > Server is a multi-purpose box. Apache, Courier Mail, Proftp, and
> SSH.
> > > (ran
> > > > > on version 2.4.19 with no issues, problems started with 2.4.20.)
> > > > >
> > > > > Compiled using GCC 3.3.
> > > >
> > > > So the problem started before you switched compilers you were using to
> > > > build kernels ?
> > >
> > > The problem was also pre GCC 3.3. I also have tried to compile the
> kernel
> > > with GCC 2.95.3 I still get lock ups also.
> >
> > Have you tried memtest on the box? It might be memory corruption (??).
> >
> > Please confirm your memory is fine (with memtest), then we continue trying
> > to find the problem, OK?

Re: Linux 2.4.22-pre3

[Marcelo Tosatti]

On Tue, 8 Jul 2003, Jim Gifford wrote:

> ----- Original Message -----
> From: "Alan Cox" <alan@lxorguk.ukuu.org.uk>
> To: "Jim Gifford" <maillist@jg555.com>
> Cc: "Marcelo Tosatti" <marcelo@conectiva.com.br>; "lkml"
> <linux-kernel@vger.kernel.org>
> Sent: Tuesday, July 08, 2003 3:23 PM
> Subject: Re: Linux 2.4.22-pre3
>
>
> > On Maw, 2003-07-08 at 22:36, Jim Gifford wrote:
> > > Server is a multi-purpose box. Apache, Courier Mail, Proftp, and SSH.
> (ran
> > > on version 2.4.19 with no issues, problems started with 2.4.20.)
> > >
> > > Compiled using GCC 3.3.
> >
> > So the problem started before you switched compilers you were using to
> > build kernels ?
>
> The problem was also pre GCC 3.3. I also have tried to compile the kernel
> with GCC 2.95.3 I still get lock ups also.

Have you tried memtest on the box? It might be memory corruption (??).

Please confirm your memory is fine (with memtest), then we continue trying
to find the problem, OK?

Re: Linux 2.4.22-pre3

[Jim Gifford]

----- Original Message ----- 
From: "Alan Cox" <alan@lxorguk.ukuu.org.uk>
To: "Jim Gifford" <maillist@jg555.com>
Cc: "Marcelo Tosatti" <marcelo@conectiva.com.br>; "lkml"
<linux-kernel@vger.kernel.org>
Sent: Tuesday, July 08, 2003 3:23 PM
Subject: Re: Linux 2.4.22-pre3


> On Maw, 2003-07-08 at 22:36, Jim Gifford wrote:
> > Server is a multi-purpose box. Apache, Courier Mail, Proftp, and SSH.
(ran
> > on version 2.4.19 with no issues, problems started with 2.4.20.)
> >
> > Compiled using GCC 3.3.
>
> So the problem started before you switched compilers you were using to
> build kernels ?
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>
The problem was also pre GCC 3.3. I also have tried to compile the kernel
with GCC 2.95.3 I still get lock ups also.

Re: Linux 2.4.22-pre3

[Alan Cox]

On Maw, 2003-07-08 at 22:36, Jim Gifford wrote:
> Server is a multi-purpose box. Apache, Courier Mail, Proftp, and SSH. (ran
> on version 2.4.19 with no issues, problems started with 2.4.20.)
> 
> Compiled using GCC 3.3.

So the problem started before you switched compilers you were using to
build kernels ?

Re: Linux 2.4.22-pre3

[Jim Gifford]

----- Original Message ----- 
From: "Marcelo Tosatti" <marcelo@conectiva.com.br>
To: "Jim Gifford" <maillist@jg555.com>
Cc: "lkml" <linux-kernel@vger.kernel.org>
Sent: Tuesday, July 08, 2003 1:38 PM
Subject: Re: Linux 2.4.22-pre3


>
> Are you using SMP? What drivers are you using and what your workload is?
>
> On Tue, 8 Jul 2003, Jim Gifford wrote:
>
> > Still receiving lockups. It it is about 2 days and 10 hours before the
> > system just freezes. I can use the sysrq keys to safely shutdown. No
error
> > messages present in any logs.
> >
> > As I stated in previous emails, when I run sync every hour, this
prevents
> > the lock ups. I will try this under the 2.4.22-pre3.
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>

Here is in the information you requested, probably more than you need.

Running a Dell Poweredge 4200 with 2 Pentium II 333 with a Dell
Perc2(Megaraid) and adaptec. 1GB of ECC/EDO RAM. Adaptec 6922 Network
Adapter

Using Megaraid Driver 1.18i (from lsi logic's web site)
Using Adaptec Driver included in pre3.
Using Tulip driver for ethernet included in pre3
Using usb-ohci driver for usb included in pre3
Utilizing ext3 file system on 3 logical drives.

Server is a multi-purpose box. Apache, Courier Mail, Proftp, and SSH. (ran
on version 2.4.19 with no issues, problems started with 2.4.20.)

Compiled using GCC 3.3.

Here is the procinfo information. With all services running.

Linux 2.4.22-pre3 (root@server) (gcc 3.3) #1 SMP Tue Jul 8 12:17:09 PDT 2003
2CPU [server.jg555.com]

Memory:      Total        Used        Free      Shared     Buffers
Cached
Mem:       1033680      224568      809112           0       13604
60280
Swap:      1060280           0     1060280

Bootup: Tue Jul  8 14:05:29 2003    Load average: 3.19 3.15 2.47 8/114 22182

user  :       0:17:49.33  34.6%  page in :    67027  disk 1:     1563r
3187w
nice  :       0:24:10.61  46.9%  page out:    76204  disk 2:     2202r
1396w
system:       0:06:05.34  11.8%  swap in :        1  disk 3:     3377r
1780w
idle  :       0:03:26.38   6.7%  swap out:        0  disk 4:        4r
0w
uptime:       0:25:45.82         context :   838418

irq  0:    154583 timer                 irq  6:         1
irq  1:         3 keyboard              irq  8:        96 rtc
irq  2:         0 cascade [4]           irq  9:     13516 megaraid
irq  3:         4 serial                irq 11:      3417 eth0, eth1
irq  4:        62 serial                irq 12:         7 PS/2 Mouse
irq  5:         0 usb-ohci              irq 15:       132 aic7xxx, aic7xxx

ver_linux output
Linux server 2.4.22-pre3 #1 SMP Tue Jul 8 12:17:09 PDT 2003 i686 unknown
unknown GNU/Linux

Gnu C                  3.3
Gnu make               3.80
util-linux             2.11z
mount                  2.11z
modutils               2.4.25
e2fsprogs              1.33
PPP                    2.4.1
Linux C Library        2.3.2
Dynamic linker (ldd)   2.3.2
Linux C++ Library      5.0.4
Procps                 3.1.9
Net-tools              1.60
Kbd                    1.08
Sh-utils               5.0
Modules Loaded         dazuko ipt_TCPMSS ipt_TOS ipt_string ipt_psd
ipt_pkttype ipt_unclean ipt_state ipt_REJECT ipt_LOG ipt_limit
iptable_mangle iptable_nat iptable_filter ip_tables ip_conntrack_h323
ip_conntrack_mms ip_conntrack_irc ip_conntrack_ftp ip_conntrack tulip rtc
usb-ohci usbcore aic7xxx megaraid sd_mod scsi_mod

Re: Linux 2.4.22-pre3

[Marcelo Tosatti]

Are you using SMP? What drivers are you using and what your workload is?

On Tue, 8 Jul 2003, Jim Gifford wrote:

> Still receiving lockups. It it is about 2 days and 10 hours before the
> system just freezes. I can use the sysrq keys to safely shutdown. No error
> messages present in any logs.
>
> As I stated in previous emails, when I run sync every hour, this prevents
> the lock ups. I will try this under the 2.4.22-pre3.

Re: Linux 2.4.22-pre3

[Jeff Garzik]

On Tue, Jul 08, 2003 at 08:05:21PM +0200, Adrian Bunk wrote:
> On Sun, Jul 06, 2003 at 06:05:27PM -0700, Larry McVoy wrote:
> > On Sun, Jul 06, 2003 at 10:00:34PM -0300, Marcelo Tosatti wrote:
> > > On Sun, 6 Jul 2003, Ben Collins wrote:
> > > > Any chance you could be consistent in tagging the -pre's? Neither pre2,
> > > > nor pre3 is tagged in BK, and thus, not tagged in CVS/SVN either.
> > > 
> > > I guess I have tagged -pre2 and -pre3:
> > > 
> > > Maybe I'm missing something?
> > 
> > Hmm.   Ben, look again in the CVS tree and make sure that the tags aren't
> > there.  Maybe the converter screwed up?  
> 
> -pre2 and -pre3 are also missing at
>   http://ftp.kernel.org/pub/linux/kernel/v2.4/testing/cset/

hrm.  Well, it's definitely tagged correctly in Marcelo's main 2.4 BK
repo.

	Jeff

Re: Linux 2.4.22-pre3

[Adrian Bunk]

On Sun, Jul 06, 2003 at 06:05:27PM -0700, Larry McVoy wrote:
> On Sun, Jul 06, 2003 at 10:00:34PM -0300, Marcelo Tosatti wrote:
> > On Sun, 6 Jul 2003, Ben Collins wrote:
> > > Any chance you could be consistent in tagging the -pre's? Neither pre2,
> > > nor pre3 is tagged in BK, and thus, not tagged in CVS/SVN either.
> > 
> > I guess I have tagged -pre2 and -pre3:
> > 
> > Maybe I'm missing something?
> 
> Hmm.   Ben, look again in the CVS tree and make sure that the tags aren't
> there.  Maybe the converter screwed up?  

-pre2 and -pre3 are also missing at
  http://ftp.kernel.org/pub/linux/kernel/v2.4/testing/cset/

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: Linux 2.4.22-pre3

[Jim Gifford]

Still receiving lockups. It it is about 2 days and 10 hours before the
system just freezes. I can use the sysrq keys to safely shutdown. No error
messages present in any logs.

As I stated in previous emails, when I run sync every hour, this prevents
the lock ups. I will try this under the 2.4.22-pre3.

----- Original Message ----- 
From: "Marcelo Tosatti" <marcelo@conectiva.com.br>
To: "lkml" <linux-kernel@vger.kernel.org>
Sent: Saturday, July 05, 2003 6:02 PM
Subject: Linux 2.4.22-pre3


>
> Hi,
>
> Here goes -pre3. It contains a lot of updates and fixes all over.
>
> We should have increased interactivity under heavy IO (users with
> interactivity problems please test and report results).
>
> Well, detailed changelog here.
>
>
> Summary of changes from v2.4.22-pre2 to v2.4.22-pre3
> ============================================
>
> <abbotti@mev.co.uk>:
>   o USB: several ftdi_sio driver patches
>
> <alex_williamson@hp.com[helgaas]>:
>   o ia64: Poll for CPEs on all CPUs, improve check for # of CPEs logged
>   o ia64: Switch to polling for CMCs if they happen too fast
>   o ia64: add wmb in sba_iommu to guarantee IOPDIR updates are visible
>   o ia64: fix timer interrupts getting lost
>
> <arun.sharma@intel.com[helgaas]>:
>   o ia64: fix IA-32 emulation of msgctl()
>   o ia64: define rlim_cur/rlim_max as unsigned
>   o ia64: fix IA-32 version of shmctl()
>   o ia64: ia32 semctl check for bad command
>   o ia64: Patch by Arun Sharma: In the absence of the patch, this system
call fails:
>   o ia64: Fix SMP FPH handling.  From 2.5 patch by Asit Mallick, David
Mosberger, Arun Sharma.
>   o ia64: work around race conditions in ia32 support code
>   o ia64: IA-32 support patch: msgsnd/msgrcv return value off by 4
>   o ia64: IA-32 support patch: munmap should return EINVAL if size == 0
>   o ia64: IA-32 support patch: mmap should return ENOMEM
>
> <baldrick@wanadoo.fr>:
>   o USB speedtouch: use common CRC library
>
> <bdschuym@pandora.be>:
>   o [NETFILTER]: Missing return in arp_packet_match()
>   o [NETFILTER]: Add arptables mangle module
>
> <bjorn_helgaas@hp.com[helgaas]>:
>   o ia64: Export pm_idle
>   o ia64: sys32_sysinfo: update to current struct sysinfo (add totalhigh,
freehigh, mem_unit).
>   o ia64: Make struct sysinfo32 internal padding explicit
>   o ia64: Make CONFIG_SYSCTL control sys32_sysctl as well.  Based on a
patch from Peter Chubb.
>   o ia64: ia64_fetch_and_add(), xchg(), ia64_cmpxchg(), etc
>   o ia64: Update default configs
>   o ia64: iosapic: remove find_iosapic duplication
>   o ia64: iosapic: simplify ISA IRQ init
>   o ia64: iosapic: self-documenting polarity/trigger arguments
>   o ia64: iosapic: Remove gratuitous differences with 2.5 (whitespace, C99
initializers, printk levels, etc).
>   o ia64: Use printk severity-levels where appropriate
>   o ia64: cleanup unwind.c warnings (from David's 2.5 change)
>   o ia64: mca.c whitespace changes and dead code removal from 2.5
>   o ia64: sba_iommu: whitespace and comment changes to align with 2.5
>   o ia64: sba_iommu: prefetch_spill_page alignment with 2.5
>   o ia64: sba_iommu: printk text and other trivial changes to align with
2.5
>   o io4
>   o ia64: sba_iommu: make sure devices are at least 32-bit capable (from
2.5)
>   o ia64: sba_iommu: Combine HWP0001 and HWP0004 ACPI claim (from 2.5
changes by Alex Williamson).
>   o ia64: sba_iommu: remove workarounds for broken, never released,
firmware that didn't program IBASE/IMASK correctly.
>   o ia64: remove cpu_is_online local defs, in favor of a 2.5-style
cpu_online
>   o ia64: Remove unused variable from acpi.c
>   o ia64: sba_iommu: fix warning and use old-style ACPI typedef
>   o ia64: whitespace and trivial changes in mca.c
>   o ia64: palinfo whitespace changes to match 2.5
>   o ia64: simplify syscalls with force_successful_syscall_return()
>   o ia64: Remove unused acpi_get_addr_space() interface
>   o ia64: Wrap pal.h with #ifdef __KERNEL__ to solve userland compilation
issues (including <linux/modules.h>).
>   o ia64: Don't blindly probe PCI buses (probe only those reported by
ACPI)
>   o ia64: pci warning for unavailable resources
>   o ia64: TLB flushing fixes - don't use smp_call_function in
context-switch path.
>   o ia64: Disable interrupts during context switch
>   o ia64: ptrace whitespace changes to follow 2.5
>   o ia64: add hugetlb and cmd649 IDE to configs
>   o ia64: Export SAL error records in
/proc/sal/{mca,init,cmc,cpe}/{event,data}
>   o ia64: Rename EFI systab tags (no spaces, etc, for easier parsing)
>   o ia64: Ignore empty address ranges from _CRS to workaround buggy Big
Sur firmware.
>
> <chad_smith@hp.com[helgaas]>:
>   o ia64: expose pointers from EFI system table in /proc
>
> <chas@cmd.nrl.navy.mil>:
>   o [ATM]: remove iovcnt member in struct atm_skb
>
> <chas@cmf.nrl.navy.mil>:
>   o [ATM]: Backport HE driver fixes from 2.5.x
>   o [ATM]: ixmicro puts esi in different location
>   o [ATM]: lock neighbor entry during update in clip.c
>   o [ATM]: make sub skb->cb is clear before upcall to network
>   o [ATM]: eliminate ATM_PDU_OVHD, ops->free_rx_skb and ops->alloc_tx
>   o [ATM]: make clip buildable as a module
>
> <dave@thedillows.org>:
>   o Use a non-zero rx_copybreak to avoid charging a full MTU to the socket
on tiny packets.
>   o Fix misreporting of card type and spurious "already scheduled"
messages
>
> <david@csse.uwa.edu.au>:
>   o USB: usb-uhci fix for one-shot interrupt problem
>   o USB: usb-ohci handling of one-shot interrupt transfers
>
> <davidm@hpl.hp.com[helgaas]>:
>   o ia64: For SIGSEGV triggered by NaT page, set si_addr to faulting data
address, not the faulting IP.
>
> <davidm@tiger.hpl.hp.com[helgaas]>:
>   o ia64: Make ia64_fetch_and_add() simpler to optimize so lib/rwsem.c can
be optimized properly.
>   o ia64: Implement pcibios_set_mwi() and define HAVE_ARCH_PCI_MWI to
ensure that PCI line-size gets programmed properly.  Based
>   o ia64; Improve debug output from kernel unwinder.  Based on patch by
Keith Owens.  (Ported to 2.4 by Bjorn Helgaas).
>   o ia64: In kernel unwinder, replace dump_info_pt() with
get_scratch_regs() and reformat to make it fit in 100 columns.
>   o ia64: Add unwcheck.sh script contributed by Harish Patil.  It checks
the unwind info for consistency (well, just the obvious stuff, but it's a
start).
>   o ia64: Minor cleanups.  (From 2.5 by Bjorn Helgaas)
>   o ia64: Make signal deliver work when the current register frame is
incomplete (as a result of a faulting mandatory RSE load).
>   o ia64: Correct region_start calculation in kernel unwinder
>   o ia64: clean up unneeded test in kernel unwinder
>   o ia64: More vmlinux.lds.S cleanups (__start/__end inside sections)
>   o ia64: Minor fixes
>   o ia64: Two small MCA fixes
>   o ia64: Sync itc after interrupts enabled
>   o ia64: Sync sys32_ipc() with x86 counter-part
>   o ia64: Patch by Arun Sharma: In brl_emu.c, a 64 bit value was being
assigned to an int.
>   o ia64: Minor whitespace & formatting fixups in asm-ia64/sal.h
>   o ia64: Fix SAL processor-log info handling.  Based on patch by Keith
Owens.
>   o ia64: Manual merge of Keith Owen's patch to avoid deadlock on
ia64_sal_mc_rendez().  Also prefix local-variables in SAL macros to avoid
name collisions.
>   o ia64: dump the min-state area in the MCA INIT platform handler
>   o ia64: Update platform INIT handler to print a backtrace
>   o ia64: Consolidate backtrace printing in a single routine
(ia64_do_show_stack())
>   o ia64: fix /proc/.../vm_info memory attributes
>   o ia64: Fix printing of memory attributes
>   o mca.c
>   o ia64: Fix INIT copying of banked registers
>   o ia64: ptrace: don't let reading NaT bits for R4-R7 overwrite the value
we're intending to write; get_rnat & put_rnat cleanups.
>   o ia64: Fix ptrace() RNaT accessors
>   o ia64: Fix page-fault handler so it handles not-present translations
for region 5 (patch by John Marvin).
>   o ia64: Fix unwinder so core-dumps work again.  Without this patch, most
scratch-regs came out wrong.
>   o ia64: Fixups for GCC v3.3
>
> <davidm@wailua.hpl.hp.com[helgaas]>:
>   o ia64: Change struct ia64_fpreg so it will get 16-byte alignment with
all ia64 compilers, not just GCC.
>   o ia64: Don't output backspaces in palinfo output
>
> <eranian@hpl.hp.com[helgaas]>:
>   o ia64: perfmon update to v1.4
>   o ia64: perfmon fixes for system-wide monitoring overflow, opcode
matcher, and force PMC[89] bit 2 on.
>   o ia64: perfmon update
>   o ia64: perfmon TLB_* and ALAT event fix
>
> <garyhade@us.ibm.com[helgaas]>:
>   o ia64: fix sysinfo(2) memory value truncation for 32-bit apps
>
> <grigouze@noos.fr>:
>   o USB: zaurus SL-C700
>
> <james@cobaltmountain.com[helgaas]>:
>   o include_asm-ia64_sal.h, typo: the the
>
> <jbarnes@sgi.com[helgaas]>:
>   o ia64: ACPI fix for no PCI
>
> <jes@wildopensource.com[helgaas]>:
>   o ia64: don't try to synchronize ITCs on ITC_DRIFT platforms
>
> <jgarzik@pobox.com>:
>   o fix Via pci irq routing
>
> <jh@sgi.com[helgaas]>:
>   o ia64: SGI SN update
>   o ia64: SN2 update 030528
>   o ia64: SN2 update 030630
>
> <jsm@udlkern.fc.hp.com[helgaas]>:
>   o ia64: don't let PTRACE_POKEDATA write the NaT bits of syscall args
>
> <judd@jpilot.org>:
>   o USB: visor.h[c] USB device IDs
>
> <kaos@ocs.com.au[helgaas]>:
>   o ia64: fix unwinder to call get_scratch_regs() only when really needed
>
> <kaos@sgi.com[helgaas]>:
>   o ia64: fix scratch-regs handling in kernel unwinder
>   o ia64: unwind.c - allow unw_access_gr(r0)
>   o ia64: Trivial stack-size correction in mca.c
>   o ia64: mca rendezvous fix
>   o ia64: Hold modlist_lock while searching exception tables
>   o ia64: Handle SAL rejection of MCA rendezvous timeout value
>
> <kenneth.w.chen@intel.com[helgaas]>:
>   o ia64: rwsem using atomic primitive
>
> <kpc-usbdev@gelato.uiuc.edu>:
>   o USB: Desknote/ECS UCR-61S2B card reader (2.4.21 patched)
>
> <lethal@linux-sh.org>:
>   o SH64 Merge
>   o Add SH-5 support to SH-SCI
>   o Add SH-5 support to tulip_core
>   o Update MAINTAINERS for sh/sh64
>   o SH-5 DMAC Support
>   o sh64 PCI DMA coherency fixups
>   o sh64: Fix SHMBLA compile error
>   o sh64: Add an onchip_unmap() to clean up after
>   o sh64: tlbmiss handler updates
>   o sh64: Don't startup the irq in make_intc_irq()
>   o sh64: Add workarounds for cache aliasing issues
>   o sh64: Cleanup sleep usage
>   o sh64: Fix PTRACE_POKEUSR to ignore changes of privileged
>   o sh64: Make memcpy safe on SH5-101 cut2
>   o sh64: export more needed symbols
>   o sh64: Fixes for Cayman LEDs
>
> <mk@linux-ipv6.org>:
>   o [CRYPTO]: Update deflate dependencies
>
> <mkp@mkp.net[helgaas]>:
>   o ia64: declare ia64_sal_handler_init non-static
>
> <mort@wildopensource.com[helgaas]>:
>   o ia64: print ISR for FPSWA faults
>   o ia64: runtime platform detection for 2.5
>
> <richard.curnow@superh.com>:
>   o Ensure that the 'unlink' XDR structures are correctly aligned on
64-bit architectures.
>
> <romieu@fr.zoreil.com>:
>   o [NETFILTER]: Fix leaks in error paths of ip_recent_ctrl
>
> <rusty@rustcorp.com.au[helgaas]>:
>   o Designated initializers for ia64
>
> <schwab@suse.de[helgaas]>:
>   o ia64: fix unwinder bug in unw_access_gr()
>   o ia64: Fix request_module from ia32 process
>   o ia64: make sys32_ptrace() use ptrace_check_attach()
>
> <shemminger@osdl.org>:
>   o [BRIDGE]: Ethernet bridge fixes
>
> <shmulik.hen@intel.com>:
>   o Fix load balance problem with high UDP Tx stress
>   o Fix 802.3ad long fail over with high UDP Tx stress
>   o [netdrvr bonding] Fix change active for ALB/TLB
>
> <sv@sw.com.sg[helgaas]>:
>   o ia64: improve show_trace_task() portability
>
> <venkatesh.pallipadi@intel.com[helgaas]>:
>   o ia64: IA-32 emulation patch: ptrace get_FPREGS bug fix
>
> <will@sowerbutts.com>:
>   o USB: Update for the powermate driver to work with newer devices
>
> Adam J. Richter <adam@yggdrasil.com>:
>   o [CRYPTO]: Simplify crypto memory allocation
>
> Adrian Bunk <bunk@fs.tum.de>:
>   o postfix a constant in efi.h with ULL
>
> Alan Cox <alan@lxorguk.ukuu.org.uk>:
>   o [NETFILTER]: Fix nat_helper warnings with gcc 3.3
>   o [NET]: Add EDP2 ethernet protocol ID
>   o [SPARC]: d_path() can return an error code, must handle it
>   o Optimise FAT handling using the prev_free info as DOS does
>   o PATH: add hfsplus file system (stands alone)
>   o NLS config.in for hfsplus
>   o config.in for HFSPLUS
>   o makefile for HFSPLUS
>   o fix leak in llc 802
>   o fix decnet gcc 3.3 build
>   o add xapic checking define
>   o add the extra cpu bit test flags
>   o remove io_apic_modify - this doesnt work on some APICs
>   o add the MSR's for IA32 perf ctl
>   o fix false sharing of mm info
>   o we moved these so this copy can go
>   o collated copy of Geerts patches for m68k headers
>   o add a flag so we can forbid APM idling
>   o add the ide_register_driver defines
>   o add EDP2 protocol id
>   o update fat docs - we now use the field
>   o bring PCI_IDS back into sync
>   o add new entry to sisfb types
>   o support cramfs initrd
>   o add timedop stub for IPC=n
>   o assorted module race fixe
>   o dont corrupt utsname on failed copy
>   o fix make rpm
>   o dont idle if forbid_idle set
>   o large scale DMI table updates
>   o merge long standing reboot fix form -ac
>   o fix up semops and return, allow timedop
>   o fix error in vm86 fixups
>   o add semtimedop to ia64 emu too
>   o fix up gcc 3.3 bits
>   o copy the right data in mips emulation
>   o collected m68k core diffs
>   o typo fix
>   o fix iphase leak
>   o bump cciss to new vendor driver
>   o Jens floppy locking fixes
>   o add comtrol note in case we need to know in the future
>   o & v && fixes in sysrq.c
>   o update sonypi driver
>   o parisc gsc driver sync
>   o fix config.in bits for IDE
>   o make IDE modularisable
>   o fix ide dma timeout bugs
>   o make pnpide module happy
>   o Herbert's fix for ide proc oops
>   o make pdc4030 module happy
>   o add generic support for toshiba piccolo
>   o fix hpt speed bits
>   o fix promise sx6000 newer board problems
>   o clean up older pdc
>   o siimage updates, add aar-1210sa
>   o SiS IDE updates
>   o hptraid updates
>   o small setup-pci cleanups
>   o d_path can return an error code, must handle it
>   o update motion eye drivers
>   o fix leak in octagon
>   o new 3c59x. plus handle power bits
>   o typo fix in atari_pamsnet driver
>   o fix ma600 gcc 3.3
>   o minor m68k fixes
>   o fix leak in aironet4500_cs
>   o fix plip hang on ifdown/ifup
>   o update sonic
>   o update orinoco drivers
>   o update pci.ids
>   o add cirrus support to i82092
>   o fix rsrc manager
>   o pci routing for ti cardbus
>   o update aacraid
>   o aic7xxx allow db4
>   o gdth register failure path
>   o update scsi tape docs
>   o megaraid broke config tools
>   o send_diag wants long timeout default
>   o let the ide layer fail commands
>   o resync scsi blacklist
>   o new segate bios string
>   o update scsi tape driver
>   o remove noise
>   o fix copy from user bug in cmpci
>   o update AC97 codec core
>   o switch cards to new ac97_audio
>   o switch i810 to generalised digital out, new ac97
>   o ac97 updates
>   o fix long standing doc typo
>   o update trident, fix printks, new ac97
>   o Update via audio - fix problems esd, mpg321
>   o update to new ac97_codec
>   o core fbcon fixes
>   o update vesafb memory handling for big cards
>   o update sis fb drivers
>   o add semtimedop to x86 headers
>   o update ac97 codec headers
>   o declare semtimedop function
>   o add scripts ready to merge kconfig
>   o update cciss docs to match new driver
>   o add vram to vesafb docs
>   o CMD640 update
>   o (new) Turn on the IDE modular stuff in the Makefile
>   o (resend) collected semaphore fixes and semtimedop
>   o make i810 audio compile
>
> Alex Williamson <alex_williamson@hp.com>:
>   o ia64: CMC deadlock fix
>
> Andi Kleen <ak@muc.de>:
>   o Personality fixes for x86-64
>   o x86-64 merge
>   o Support exception-trace sysctl for x86-64
>   o non executable stack support for x86-64
>
> Andrew Morton <akpm@digeo.com>:
>   o [CRYPTO]: Fix memcpy/memset args
>
> Ben Collins <bcollins@debian.org>:
>   o Update IEEE1394 (r972)
>
> Benjamin Herrenschmidt <benh@kernel.crashing.org>:
>   o ppc32: support for 750FX rev2 CPU
>   o ppc32: Enable use of USB2 on recent G4s
>   o ppc32: Update PowerMac motherboard support
>   o ppc32: Update swim3 floppy driver
>   o ppc32: Add TotalImpact briQ panel driver
>   o ppc32: Add a "query" function to core ADB
>   o ppc32: Update adbhid driver
>   o ppc32: Update battery calculation code & via-pmu
>   o ppc32: Minimal ethtool for bmac and mace
>   o ppc32: Fix a problem with both gmac and sungem
>
> Bjorn Helgaas <bjorn_helgaas@hp.com>:
>   o ia64: chmod +x unwcheck.sh script
>   o ia64: iosapic: make pcat_compat system property
>   o ia64: iosapic: rationalize __init/__devinit
>   o ia64: Export io_space so drivers using legacy I/O ports can insmod
>   o ia64: brl_emu.c: use temporary variable to avoid gcc3.1 warning
>   o ia64: remove incorrect and redundant "cpu not responding" message
>   o ia64: Update configs
>   o ia64: pci.c: Trivial changes to follow 2.5
>   o ia64: sba_iommu: use seq_file
>   o ia64: acpi: handle vendor resources more generically
>   o Move UP cpu_online definition to <linux/smp.h>
>   o Cset exclude:
rohit.seth@intel.com[helgaas]|ChangeSet|20030623203306|58862
>
> Chris Mason <mason@suse.com>:
>   o Fix potential IO hangs and increase interactiveness during heavy IO
>
> Christoph Hellwig <hch@lst.de>:
>   o [CRYPTO-2.4]: Missing ULL postfixes and statics
>
> David S. Miller <davem@nuts.ninka.net>:
>   o [BK]: Add *~ to ignore regexps
>   o [CRYPTO]: kunmap does not return a value
>   o [CRYPTO]: Build/warning fixups
>   o [CRYPTO]: Clean up header file usage
>   o [CRYPTO]: Include kernel.h in crypto.h
>   o [CRYPTO]: Allocate work buffers instead of using kstack
>   o [CRYPTO]: Make sha256.c more palatable to GCCs optimizers
>   o [CRYPTO]: internal.h needs init.h
>   o [CRYPTO]: Use appropriate defaults if AH/ESP is enabled
>   o [CRYPTO-2.4]: Add dummy kmap_types.h header for sparc64
>   o [CRYPTO]: Include linux/errno.h as appropriate
>   o [CRYPTO-2.4]: module_name does not exist in 2.4.x
>   o [CRYPTO-2.4]: const static --> static const
>   o [CRYPTO]: deflate.c needs slab.h
>   o [CRYPTO-2.4]: Fix condition typos in crypto/Config.in
>   o [CRYPTO-2.4]: Emulate module_name semantics correctly to avoid OOPS
>   o [CRYPTO-2.4]: Make sure crypto config is before lib config on ia64
>   o [NET]: net/bluetooth/cmtp/core.c needs linux/init.h
>   o [NET]: Scale DST/ipv6 intervals like we did for ipv4
>   o [SPARC64]: Fix build error from OBP parsing patch
>   o [SPARC64]: Update defconfig
>
> Erik Andersen <andersen@codepoet.org>:
>   o fix 2.4.22-pre broken x86 math-emu
>
> Greg Kroah-Hartman <greg@kroah.com>:
>   o USB: remove stupid conversions and use of floating point from aiptek.c
>   o USB: 2.4 fix UHCI debug kmalloc() usage
>   o USB: add support for 50 baud to io_edgeport.c
>   o USB: pl2303: report CTS and DSR status changes to userspace
>   o Cset exclude: cweidema@indiana.edu|ChangeSet|20030620002017|05386
>   o USB: compiler fixes for previous vicam patches
>
> Hugh Dickins <hugh@veritas.com>:
>   o remove shmem info->sem
>   o shmem_getpage absorb _locked
>   o shmem_getpage read,cache,write
>   o shmem truncation swizzled
>   o shmem account metablocks
>   o shmem_file_write and _read
>   o init_tmpfs shm_mnt error
>   o shmem whitespace only
>   o shmem misc minor mods
>   o swapoff loopable tmpfs
>   o shmem mount percentile size
>   o shmem_removepage replace recalc_inode
>   o loop file use highmem
>   o madvise_willneed check readpage
>   o shmem_file_write precheck_file_write
>   o mremap VM_LOCKED move_vma
>   o shmem loopable tmpfs [again]
>
> Ivan Kokshaysky <ink@jurassic.park.msu.ru>:
>   o alpha: fix generic kernel build (Jay Estabrook)
>   o alpha: finalize Sable/Lynx support (Jay Estabrook)
>
> James Morris <jmorris@intercode.com.au>:
>   o [CRYPTO]: Add initial crypto api subsystem
>   o [CRYPTO]: Cleanups based upon feedback from Rusty and jgarzik
>   o [CRYPTO]: Cleanups based upon feedback from Rusty and jgarzik
>   o [CRYPTO]: Use try_inc_mod_count and semaphore for alg list
>   o [CRYPTO]: Use kmod to try to autoload modules
>   o [CRYPTO]: Bug fixes and cleanups
>   o [CRYPTO]: More bug fixes and cleanups
>   o [CRYPTO]: Add MD4
>   o [CRYPTO]: Algorithm lookup API change plus bug fixes
>   o [CRYPTO]: Run tcrypt through lindent, plus doc update
>   o [CRYPTO]: Assert that interfaces are called on correct cipher type
>   o [CRYPTO]: Cleanups and more consistency checks
>   o [CRYPTO]: Update to IV get/set interface
>   o [CRYPTO]: Add some documentation
>   o [CRYPTO]: Fix some credits
>   o [CRYPTO]: Cleanups based upon suggestions by Jeff Garzik
>   o [CRYPTO]: Uninline some functions to save some bloat
>   o [CRYPTO]: Cleanups based upon feedback from jgarzik
>   o [CRYPTO]: Add crypto_alg_available interface
>   o [CRYPTO]: Rework HMAC interface
>   o [CRYPTO]: Add SHA256 plus bug fixes
>   o [CRYPTO]: Add blowfish algorithm
>   o [CRYPTO]: minor updates
>   o [CRYPTO] kstack cleanup (v0.28)
>   o [CRYPTO] Add maintainers entry
>   o [CRYPTO] Minor doc update
>   o [CRYPTO]: Add null algorithms and minor cleanups
>   o [CRYPTO]: Kill stray CRYPTO_ALG_TYPE_COMP
>   o [CRYPTO]: Add twofish algorithm
>   o [CRYPTO]: Add serpent algorithm
>   o [CRYPTO]: Documentation update
>   o [CRYPTO]: Dont compile procfs stuff if procfs is not enabled
>   o [CRYPTO]: Add AES algorithm
>   o [CRYPTO]: More credits for AES
>   o [CRYPTO]: Add support for SHA-386 and SHA-512
>   o [CRYPTO] remove superfluous goto from des module init exception path
>   o [CRYPTO] Add AES and MD4 to tcrypto crypto_alg_available() test
>   o [CRYPTO]: in/out scatterlist support for ciphers
>   o [CRYPTO]: Move km_types out of header
>   o [CRYPTO]: Add encrypt_iv() and decrypt_iv() methods
>   o [CRYPTO]: Eliminate crypto_tfm.crt_ctx, from Adam Richter
>   o [CRYPTO]: Documentation updates
>   o [CRYPTO]: Make use of crypto_exit_ops() during crypto_free_tfm()
>   o [CRYPTO]: Add Deflate algorithm to crypto API
>   o [CRYPTO]: deflate module: workaround zlib bug
>   o [CRYPTO]: Fix config dependencies
>
> Jeff Garzik <jgarzik@redhat.com>:
>   o [CRYPTO]: Kill accidental double memset
>   o [netdrvr 8139too] fix debug printk
>
> Linus Torvalds <torvalds@transmeta.com>:
>   o The crypto auto-load should be enabled if crypto is enabled
>
> Marcelo Tosatti <marcelo@freak.distro.conectiva>:
>   o Cset exclude: rusty@rustcorp.com.au|ChangeSet|20030625201246|52045
>   o Added missing BROKEN_PNP_BIOS declaration
>   o Changed EXTRAVERSION to -pre3
>
> Martin Schwidefsky <schwidefsky@de.ibm.com>:
>   o s390 base update
>   o s390 common i/o layer fixes
>   o s390 dasd driver update
>   o s390 31 bit compat
>   o s390 documentation update
>   o Add Configure.help entries for s390 options
>   o s390 3215 driver update
>   o s390 ctc network driver update
>   o s390 iucv network driver
>   o s390 defconfigs update
>   o console semaphore fix
>
> Matt Domsch <matt_domsch@dell.com>:
>   o ia64: efivars fix by Matt Domsch and Peter Chubb
>
> Olaf Hering <olh@suse.de>:
>   o missing asm-ppc64/kmap_types.h
>
> Oleg Drokin <green@angband.namesys.com>:
>   o reiserfs: Relocated journal support by Edward Shushkin & Vladimir
Saveliev
>   o reiserfs: speed up large file holes creation
>   o reiserfs: Make most of the reiserfs warning messages to print what
device they relate to
>
> Oliver Neukum <oliver@neukum.org>:
>   o USB: disconnect of v4l devices in 2.4
>   o USB: fix to previous vicam patch
>
> Peter Chubb <peter@chubb.wattle.id.au>:
>   o ia64: declare test_bit() arg as "const"
>
> Roger Luethi <rl@hellgate.ch>:
>   o [netdrvr via-rhine] via-rhine 1.18-rc1: Fix Rhine-I regression
>
> Russell King <rmk@arm.linux.org.uk>:
>   o ARM merge part 1 - arch/arm
>   o ARM merge part 2 - include/asm-arm
>   o ARM merge part 3 - drivers/acorn
>
> Rusty Russell <rusty@rustcorp.com.au>:
>   o 2.5.43 export _end
>
> Scott Feldman <scott.feldman@intel.com>:
>   o Remove CAP_NET_ADMIN check for SIOCETHTOOL's
>
> Tom Callaway <tcallawa@redhat.com>:
>   o [SPARC64]: Fix OBP version parsing on newer systems
>
> Trond Myklebust <trond.myklebust@fys.uio.no>:
>   o A patch by Chuck Lever that cleans up the RPC socket slot allocation
code.
>   o A patch by Chuck Lever with further cleanups of the RPC socket slot
allocation code.
>   o Another patch by Chuck Lever that ensures that the PG_uptodate bit
gets set when the entire page gets written by nfs_writepage_sync()
>   o A patch by Patrice Dumas to implement nlmsvc_proc_granted_res
>   o A patch by Patrice Dumas to add a check in order to ensure that we
really were requesting a blocking lock when we get a reply from the server
asking us to block.
>   o A patch to ensures that blocks which are not going to time out are
placed last on the nlm_block list (problem reported by Olaf Kirch).
>   o Add standard spinlocks to protect the socket from being released by
one CPU while the other is in a soft interrupt.
>   o Fix a race: Ensure that requests retry if the remote server
disconnects us while we're inside xprt_transmit().
>   o Don't use an RPC child process when reconnecting to a TCP server
>   o Ensure that if we need to reconnect the socket, we also resend the
entire message.
>   o Fix a TCP client corruption problem affecting resent requests
>   o Ensure that the lockd clients always use one of the reserved ports
>   o Replace buggy version of xdr_shift_buf() with the version from 2.5.x
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>

Re: Linux 2.4.22-pre3

[Lukasz Trabinski]

On Mon, 7 Jul 2003, Chris Mason wrote:

> The changes in -pre3 won't cut down on the load reported by top/ps.  The
> goal is to cut down on latencies on the box as a whole.  So the load
> will still be as high, but you'll still be able to do an ls without
> waiting indefinitely for the results.

How long can i wait 1-5 minutes?

> So, if you are still seeing bad response times, I can send you a patch
> to try in private mail (it has been posted here a few times already).

Please send, Thank You!

-- 
*[ Łukasz Trąbiński ]*
SysAdmin @wsisiz.edu.pl

Re: Linux 2.4.22-pre3

[Chris Mason]

On Mon, 2003-07-07 at 15:34, Lukasz Trabinski wrote:
> In article <Pine.LNX.4.55L.0307052151180.21992@freak.distro.conectiva> you wrote:
> > 
> > Hi,
> > 
> > Here goes -pre3. It contains a lot of updates and fixes all over.
> > 
> > We should have increased interactivity under heavy IO (users with
> > interactivity problems please test and report results).
> 
> Well, I have still big load (80-120) when slocate running on my big ext3 HOME
> area ( 12x23GB partitions ). If You need, I can send to you snapshots from
> SysRq.

The changes in -pre3 won't cut down on the load reported by top/ps.  The
goal is to cut down on latencies on the box as a whole.  So the load
will still be as high, but you'll still be able to do an ls without
waiting indefinitely for the results.

So, if you are still seeing bad response times, I can send you a patch
to try in private mail (it has been posted here a few times already).

-chris

Re: Linux 2.4.22-pre3

[Lukasz Trabinski]

In article <Pine.LNX.4.55L.0307052151180.21992@freak.distro.conectiva> you wrote:
> 
> Hi,
> 
> Here goes -pre3. It contains a lot of updates and fixes all over.
> 
> We should have increased interactivity under heavy IO (users with
> interactivity problems please test and report results).

Well, I have still big load (80-120) when slocate running on my big ext3 HOME
area ( 12x23GB partitions ). If You need, I can send to you snapshots from
SysRq.

-- 
LT

Re: Linux 2.4.22-pre3

[Larry McVoy]

On Sun, Jul 06, 2003 at 10:00:34PM -0300, Marcelo Tosatti wrote:
> On Sun, 6 Jul 2003, Ben Collins wrote:
> > Any chance you could be consistent in tagging the -pre's? Neither pre2,
> > nor pre3 is tagged in BK, and thus, not tagged in CVS/SVN either.
> 
> I guess I have tagged -pre2 and -pre3:
> 
> Maybe I'm missing something?

Hmm.   Ben, look again in the CVS tree and make sure that the tags aren't
there.  Maybe the converter screwed up?  
-- 
---
Larry McVoy              lm at bitmover.com          http://www.bitmover.com/lm

Re: Linux 2.4.22-pre3

[Marcelo Tosatti]

On Sun, 6 Jul 2003, Ben Collins wrote:

> On Sat, Jul 05, 2003 at 10:02:09PM -0300, Marcelo Tosatti wrote:
> >
> > Hi,
> >
> > Here goes -pre3. It contains a lot of updates and fixes all over.
>
> Any chance you could be consistent in tagging the -pre's? Neither pre2,
> nor pre3 is tagged in BK, and thus, not tagged in CVS/SVN either.

I guess I have tagged -pre2 and -pre3:

ChangeSet@1.1024, 2003-07-05 20:59:23-03:00, mason@suse.com
  [PATCH] Fix potential IO hangs and increase interactiveness during heavy
IO

  io-stalls-10:


  ===== drivers/block/ll_rw_blk.c 1.45 vs edited =====
  TAG: v2.4.22-pre3


-----

ChangeSet@1.1011, 2003-06-26 18:25:43-03:00, trini@kernel.crashing.org
  [PATCH] Add /proc/sys/kernel/l3cr

  Hello.  The following patch is from Mark Greer.  This adds read-only
  support for the L3 cache register found in the MPC745x line of CPUs.

  ===== arch/ppc/kernel/ppc_htab.c 1.7 vs edited =====
  TAG: v2.4.22-pre2

Maybe I'm missing something?

Re: Linux 2.4.22-pre3

[Ben Collins]

On Sun, Jul 06, 2003 at 06:05:27PM -0700, Larry McVoy wrote:
> On Sun, Jul 06, 2003 at 10:00:34PM -0300, Marcelo Tosatti wrote:
> > On Sun, 6 Jul 2003, Ben Collins wrote:
> > > Any chance you could be consistent in tagging the -pre's? Neither pre2,
> > > nor pre3 is tagged in BK, and thus, not tagged in CVS/SVN either.
> > 
> > I guess I have tagged -pre2 and -pre3:
> > 
> > Maybe I'm missing something?
> 
> Hmm.   Ben, look again in the CVS tree and make sure that the tags aren't
> there.  Maybe the converter screwed up?  

Doesn't show up in linux-2.4/ChangeSet,v as a tag.

-- 
Debian     - http://www.debian.org/
Linux 1394 - http://www.linux1394.org/
Subversion - http://subversion.tigris.org/
Deqo       - http://www.deqo.com/

Re: Linux 2.4.22-pre3

[Matthias Andree]

On Sat, 05 Jul 2003, Marcelo Tosatti wrote:

> Here goes -pre3. It contains a lot of updates and fixes all over.

It still lacks the update of the "shortlog" (aka. lk-changelog.pl)
:-(

bk clone bk://kernel.bkbits.net/torvalds/tools/
bk get shortlog
feed your changelog through shortlog

Re: Linux 2.4.22-pre3

[Ben Collins]

On Sat, Jul 05, 2003 at 10:02:09PM -0300, Marcelo Tosatti wrote:
> 
> Hi,
> 
> Here goes -pre3. It contains a lot of updates and fixes all over.

Any chance you could be consistent in tagging the -pre's? Neither pre2,
nor pre3 is tagged in BK, and thus, not tagged in CVS/SVN either.

It just makes it easier when tracking down regressions to have known
points of reference common across BK/CVS/SVN/tar+diff.

-- 
Debian     - http://www.debian.org/
Linux 1394 - http://www.linux1394.org/
Subversion - http://subversion.tigris.org/
Deqo       - http://www.deqo.com/

Re: Linux 2.4.22-pre3

[Margit Schubert-While]

[-- Attachment #1: Type: text/plain, Size: 1031 bytes --]

Erros/unresolved as below.
Attached patch fixes sdla_chdlc, sbni and mark_page_accessed.
The sonyapi I don't know. Seems like we should export acpi_disabled in
kernel/setup.c.
The comx is an ongoing fight :-)

Margit


sdla_chdlc.c:594:43: missing terminating " character
sdla_chdlc.c: In function `wpc_init':
sdla_chdlc.c:595: error: parse error before "Failed"
sdla_chdlc.c:595: error: stray '\' in program
sdla_chdlc.c:595:68: missing terminating " character
make[3]: *** [sdla_chdlc.o] Error 1

sbni.c: In function `calc_crc32':
sbni.c:1558: error: asm-specifier for variable `_crc' conflicts with asm 
clobber list
make[3]: *** [sbni.o] Error 1

depmod: *** Unresolved symbols in 
/lib/modules/2.4.22-pre3/kernel/drivers/char/sonypi.o
depmod:         acpi_disabled
depmod: *** Unresolved symbols in 
/lib/modules/2.4.22-pre3/kernel/drivers/net/wan/comx.o
depmod:         proc_get_inode
depmod: *** Unresolved symbols in 
/lib/modules/2.4.22-pre3/kernel/fs/hfsplus/hfsplus.o
depmod:         mark_page_accessed


         Margit

[-- Attachment #2: patchpre3 --]
[-- Type: application/octet-stream, Size: 2006 bytes --]

diff -Naur linux-2.4.22pre2/drivers/net/wan/sbni.c linux-2.4.22pre2mw0/drivers/net/wan/sbni.c
--- linux-2.4.22pre2/drivers/net/wan/sbni.c	2002-11-29 00:53:14.000000000 +0100
+++ linux-2.4.22pre2mw0/drivers/net/wan/sbni.c	2003-07-04 18:26:26.000000000 +0200
@@ -1552,13 +1552,13 @@
 static u32
 calc_crc32( u32  crc,  u8  *p,  u32  len )
 {
-	register u32  _crc __asm ( "ax" );
+	register u32  _crc;
 	_crc = crc;
 	
 	__asm __volatile (
 		"xorl	%%ebx, %%ebx\n"
-		"movl	%1, %%esi\n" 
-		"movl	%2, %%ecx\n" 
+		"movl	%2, %%esi\n" 
+		"movl	%3, %%ecx\n" 
 		"movl	$crc32tab, %%edi\n"
 		"shrl	$2, %%ecx\n"
 		"jz	1f\n"
@@ -1594,7 +1594,7 @@
 		"jnz	0b\n"
 
 	"1:\n"
-		"movl	%2, %%ecx\n"
+		"movl	%3, %%ecx\n"
 		"andl	$3, %%ecx\n"
 		"jz	2f\n"
 
@@ -1619,9 +1619,9 @@
 		"xorb	2(%%esi), %%bl\n"
 		"xorl	(%%edi,%%ebx,4), %%eax\n"
 	"2:\n"
-		:
-		: "a" (_crc), "g" (p), "g" (len)
-		: "ax", "bx", "cx", "dx", "si", "di"
+		: "=a" (_crc)
+		: "0" (_crc), "g" (p), "g" (len)
+		: "bx", "cx", "dx", "si", "di"
 	);
 
 	return  _crc;
diff -Naur linux-2.4.22pre2/drivers/net/wan/sdla_chdlc.c linux-2.4.22pre2mw0/drivers/net/wan/sdla_chdlc.c
--- linux-2.4.22pre2/drivers/net/wan/sdla_chdlc.c	2002-11-29 00:53:14.000000000 +0100
+++ linux-2.4.22pre2mw0/drivers/net/wan/sdla_chdlc.c	2003-07-04 17:20:19.000000000 +0200
@@ -591,8 +591,7 @@
 	
 
 		if (chdlc_set_intr_mode(card, APP_INT_ON_TIMER)){
-			printk (KERN_INFO "%s: 
-				Failed to set interrupt triggers!\n",
+			printk (KERN_INFO "%s: Failed to set interrupt triggers!\n",
 				card->devname);
 			return -EIO;	
         	}
diff -Naur linux-2.4.22pre2/mm/filemap.c linux-2.4.22pre2mw0/mm/filemap.c
--- linux-2.4.22pre2/mm/filemap.c	2003-06-30 09:52:45.000000000 +0200
+++ linux-2.4.22pre2mw0/mm/filemap.c	2003-07-05 15:22:40.000000000 +0200
@@ -1338,6 +1338,8 @@
 		SetPageReferenced(page);
 }
 
+EXPORT_SYMBOL(mark_page_accessed);
+
 /*
  * This is a generic file read routine, and uses the
  * inode->i_op->readpage() function for the actual low-level

Re: Linux 2.4.22-pre3

[Aschwin Marsman]

On Sat, 5 Jul 2003, Marcelo Tosatti wrote:

> Hi,
> 
> Here goes -pre3. It contains a lot of updates and fixes all over.
> 
> We should have increased interactivity under heavy IO (users with
> interactivity problems please test and report results).

Compilation breaks for me:

gcc -D__KERNEL__ -I/usr/src/linux-2.4.21/include -Wall -Wstrict-prototypes 
 -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common -fomit-frame-pointer 
 -pipe -mpreferred-stack-boundary=2 -march=i686 -DMODULE -DMODVERSIONS 
 -include /usr/src/linux-2.4.21/include/linux/modversions.h  -nostdinc 
 -iwithprefix include -DKBUILD_BASENAME=vicam  -c -o vicam.o vicam.c
vicam.c: In function `vicam_close':
vicam.c:767: parse error before `struct'
vicam.c:770: `cam' undeclared (first use in this function)
vicam.c:770: (Each undeclared identifier is reported only once
vicam.c:770: for each function it appears in.)
make[2]: *** [vicam.o] Error 1
make[2]: Leaving directory `/usr/src/linux-2.4.21/drivers/usb'
make[1]: *** [_modsubdir_usb] Error 2
make[1]: Leaving directory `/usr/src/linux-2.4.21/drivers'
make: *** [_mod_drivers] Error 2

Two patches are included that might be a cause, although it looks like
they were included to solve compilation problems:

>   o USB: compiler fixes for previous vicam patches
>   o USB: fix to previous vicam patch
 
I changed CONFIG_USB_VICAM from a module to no, because I don't need
the driver.

Have a nice weekend,
 
Aschwin Marsman
 
--
aYniK Software Solutions         all You need is Knowledge
P.O. box 134                     NL-7600 AC Almelo - the Netherlands
a.marsman@aYniK.com              http://www.aYniK.com

Linux 2.4.22-pre3

[Marcelo Tosatti]

Hi,

Here goes -pre3. It contains a lot of updates and fixes all over.

We should have increased interactivity under heavy IO (users with
interactivity problems please test and report results).

Well, detailed changelog here.


Summary of changes from v2.4.22-pre2 to v2.4.22-pre3
============================================

<abbotti@mev.co.uk>:
  o USB: several ftdi_sio driver patches

<alex_williamson@hp.com[helgaas]>:
  o ia64: Poll for CPEs on all CPUs, improve check for # of CPEs logged
  o ia64: Switch to polling for CMCs if they happen too fast
  o ia64: add wmb in sba_iommu to guarantee IOPDIR updates are visible
  o ia64: fix timer interrupts getting lost

<arun.sharma@intel.com[helgaas]>:
  o ia64: fix IA-32 emulation of msgctl()
  o ia64: define rlim_cur/rlim_max as unsigned
  o ia64: fix IA-32 version of shmctl()
  o ia64: ia32 semctl check for bad command
  o ia64: Patch by Arun Sharma: In the absence of the patch, this system call fails:
  o ia64: Fix SMP FPH handling.  From 2.5 patch by Asit Mallick, David Mosberger, Arun Sharma.
  o ia64: work around race conditions in ia32 support code
  o ia64: IA-32 support patch: msgsnd/msgrcv return value off by 4
  o ia64: IA-32 support patch: munmap should return EINVAL if size == 0
  o ia64: IA-32 support patch: mmap should return ENOMEM

<baldrick@wanadoo.fr>:
  o USB speedtouch: use common CRC library

<bdschuym@pandora.be>:
  o [NETFILTER]: Missing return in arp_packet_match()
  o [NETFILTER]: Add arptables mangle module

<bjorn_helgaas@hp.com[helgaas]>:
  o ia64: Export pm_idle
  o ia64: sys32_sysinfo: update to current struct sysinfo (add totalhigh, freehigh, mem_unit).
  o ia64: Make struct sysinfo32 internal padding explicit
  o ia64: Make CONFIG_SYSCTL control sys32_sysctl as well.  Based on a patch from Peter Chubb.
  o ia64: ia64_fetch_and_add(), xchg(), ia64_cmpxchg(), etc
  o ia64: Update default configs
  o ia64: iosapic: remove find_iosapic duplication
  o ia64: iosapic: simplify ISA IRQ init
  o ia64: iosapic: self-documenting polarity/trigger arguments
  o ia64: iosapic: Remove gratuitous differences with 2.5 (whitespace, C99 initializers, printk levels, etc).
  o ia64: Use printk severity-levels where appropriate
  o ia64: cleanup unwind.c warnings (from David's 2.5 change)
  o ia64: mca.c whitespace changes and dead code removal from 2.5
  o ia64: sba_iommu: whitespace and comment changes to align with 2.5
  o ia64: sba_iommu: prefetch_spill_page alignment with 2.5
  o ia64: sba_iommu: printk text and other trivial changes to align with 2.5
  o io4
  o ia64: sba_iommu: make sure devices are at least 32-bit capable (from 2.5)
  o ia64: sba_iommu: Combine HWP0001 and HWP0004 ACPI claim (from 2.5 changes by Alex Williamson).
  o ia64: sba_iommu: remove workarounds for broken, never released, firmware that didn't program IBASE/IMASK correctly.
  o ia64: remove cpu_is_online local defs, in favor of a 2.5-style cpu_online
  o ia64: Remove unused variable from acpi.c
  o ia64: sba_iommu: fix warning and use old-style ACPI typedef
  o ia64: whitespace and trivial changes in mca.c
  o ia64: palinfo whitespace changes to match 2.5
  o ia64: simplify syscalls with force_successful_syscall_return()
  o ia64: Remove unused acpi_get_addr_space() interface
  o ia64: Wrap pal.h with #ifdef __KERNEL__ to solve userland compilation issues (including <linux/modules.h>).
  o ia64: Don't blindly probe PCI buses (probe only those reported by ACPI)
  o ia64: pci warning for unavailable resources
  o ia64: TLB flushing fixes - don't use smp_call_function in context-switch path.
  o ia64: Disable interrupts during context switch
  o ia64: ptrace whitespace changes to follow 2.5
  o ia64: add hugetlb and cmd649 IDE to configs
  o ia64: Export SAL error records in /proc/sal/{mca,init,cmc,cpe}/{event,data}
  o ia64: Rename EFI systab tags (no spaces, etc, for easier parsing)
  o ia64: Ignore empty address ranges from _CRS to workaround buggy Big Sur firmware.

<chad_smith@hp.com[helgaas]>:
  o ia64: expose pointers from EFI system table in /proc

<chas@cmd.nrl.navy.mil>:
  o [ATM]: remove iovcnt member in struct atm_skb

<chas@cmf.nrl.navy.mil>:
  o [ATM]: Backport HE driver fixes from 2.5.x
  o [ATM]: ixmicro puts esi in different location
  o [ATM]: lock neighbor entry during update in clip.c
  o [ATM]: make sub skb->cb is clear before upcall to network
  o [ATM]: eliminate ATM_PDU_OVHD, ops->free_rx_skb and ops->alloc_tx
  o [ATM]: make clip buildable as a module

<dave@thedillows.org>:
  o Use a non-zero rx_copybreak to avoid charging a full MTU to the socket on tiny packets.
  o Fix misreporting of card type and spurious "already scheduled" messages

<david@csse.uwa.edu.au>:
  o USB: usb-uhci fix for one-shot interrupt problem
  o USB: usb-ohci handling of one-shot interrupt transfers

<davidm@hpl.hp.com[helgaas]>:
  o ia64: For SIGSEGV triggered by NaT page, set si_addr to faulting data address, not the faulting IP.

<davidm@tiger.hpl.hp.com[helgaas]>:
  o ia64: Make ia64_fetch_and_add() simpler to optimize so lib/rwsem.c can be optimized properly.
  o ia64: Implement pcibios_set_mwi() and define HAVE_ARCH_PCI_MWI to ensure that PCI line-size gets programmed properly.  Based
  o ia64; Improve debug output from kernel unwinder.  Based on patch by Keith Owens.  (Ported to 2.4 by Bjorn Helgaas).
  o ia64: In kernel unwinder, replace dump_info_pt() with get_scratch_regs() and reformat to make it fit in 100 columns.
  o ia64: Add unwcheck.sh script contributed by Harish Patil.  It checks the unwind info for consistency (well, just the obvious stuff, but it's a start).
  o ia64: Minor cleanups.  (From 2.5 by Bjorn Helgaas)
  o ia64: Make signal deliver work when the current register frame is incomplete (as a result of a faulting mandatory RSE load).
  o ia64: Correct region_start calculation in kernel unwinder
  o ia64: clean up unneeded test in kernel unwinder
  o ia64: More vmlinux.lds.S cleanups (__start/__end inside sections)
  o ia64: Minor fixes
  o ia64: Two small MCA fixes
  o ia64: Sync itc after interrupts enabled
  o ia64: Sync sys32_ipc() with x86 counter-part
  o ia64: Patch by Arun Sharma: In brl_emu.c, a 64 bit value was being assigned to an int.
  o ia64: Minor whitespace & formatting fixups in asm-ia64/sal.h
  o ia64: Fix SAL processor-log info handling.  Based on patch by Keith Owens.
  o ia64: Manual merge of Keith Owen's patch to avoid deadlock on ia64_sal_mc_rendez().  Also prefix local-variables in SAL macros to avoid name collisions.
  o ia64: dump the min-state area in the MCA INIT platform handler
  o ia64: Update platform INIT handler to print a backtrace
  o ia64: Consolidate backtrace printing in a single routine (ia64_do_show_stack())
  o ia64: fix /proc/.../vm_info memory attributes
  o ia64: Fix printing of memory attributes
  o mca.c
  o ia64: Fix INIT copying of banked registers
  o ia64: ptrace: don't let reading NaT bits for R4-R7 overwrite the value we're intending to write; get_rnat & put_rnat cleanups.
  o ia64: Fix ptrace() RNaT accessors
  o ia64: Fix page-fault handler so it handles not-present translations for region 5 (patch by John Marvin).
  o ia64: Fix unwinder so core-dumps work again.  Without this patch, most scratch-regs came out wrong.
  o ia64: Fixups for GCC v3.3

<davidm@wailua.hpl.hp.com[helgaas]>:
  o ia64: Change struct ia64_fpreg so it will get 16-byte alignment with all ia64 compilers, not just GCC.
  o ia64: Don't output backspaces in palinfo output

<eranian@hpl.hp.com[helgaas]>:
  o ia64: perfmon update to v1.4
  o ia64: perfmon fixes for system-wide monitoring overflow, opcode matcher, and force PMC[89] bit 2 on.
  o ia64: perfmon update
  o ia64: perfmon TLB_* and ALAT event fix

<garyhade@us.ibm.com[helgaas]>:
  o ia64: fix sysinfo(2) memory value truncation for 32-bit apps

<grigouze@noos.fr>:
  o USB: zaurus SL-C700

<james@cobaltmountain.com[helgaas]>:
  o include_asm-ia64_sal.h, typo: the the

<jbarnes@sgi.com[helgaas]>:
  o ia64: ACPI fix for no PCI

<jes@wildopensource.com[helgaas]>:
  o ia64: don't try to synchronize ITCs on ITC_DRIFT platforms

<jgarzik@pobox.com>:
  o fix Via pci irq routing

<jh@sgi.com[helgaas]>:
  o ia64: SGI SN update
  o ia64: SN2 update 030528
  o ia64: SN2 update 030630

<jsm@udlkern.fc.hp.com[helgaas]>:
  o ia64: don't let PTRACE_POKEDATA write the NaT bits of syscall args

<judd@jpilot.org>:
  o USB: visor.h[c] USB device IDs

<kaos@ocs.com.au[helgaas]>:
  o ia64: fix unwinder to call get_scratch_regs() only when really needed

<kaos@sgi.com[helgaas]>:
  o ia64: fix scratch-regs handling in kernel unwinder
  o ia64: unwind.c - allow unw_access_gr(r0)
  o ia64: Trivial stack-size correction in mca.c
  o ia64: mca rendezvous fix
  o ia64: Hold modlist_lock while searching exception tables
  o ia64: Handle SAL rejection of MCA rendezvous timeout value

<kenneth.w.chen@intel.com[helgaas]>:
  o ia64: rwsem using atomic primitive

<kpc-usbdev@gelato.uiuc.edu>:
  o USB: Desknote/ECS UCR-61S2B card reader (2.4.21 patched)

<lethal@linux-sh.org>:
  o SH64 Merge
  o Add SH-5 support to SH-SCI
  o Add SH-5 support to tulip_core
  o Update MAINTAINERS for sh/sh64
  o SH-5 DMAC Support
  o sh64 PCI DMA coherency fixups
  o sh64: Fix SHMBLA compile error
  o sh64: Add an onchip_unmap() to clean up after
  o sh64: tlbmiss handler updates
  o sh64: Don't startup the irq in make_intc_irq()
  o sh64: Add workarounds for cache aliasing issues
  o sh64: Cleanup sleep usage
  o sh64: Fix PTRACE_POKEUSR to ignore changes of privileged
  o sh64: Make memcpy safe on SH5-101 cut2
  o sh64: export more needed symbols
  o sh64: Fixes for Cayman LEDs

<mk@linux-ipv6.org>:
  o [CRYPTO]: Update deflate dependencies

<mkp@mkp.net[helgaas]>:
  o ia64: declare ia64_sal_handler_init non-static

<mort@wildopensource.com[helgaas]>:
  o ia64: print ISR for FPSWA faults
  o ia64: runtime platform detection for 2.5

<richard.curnow@superh.com>:
  o Ensure that the 'unlink' XDR structures are correctly aligned on 64-bit architectures.

<romieu@fr.zoreil.com>:
  o [NETFILTER]: Fix leaks in error paths of ip_recent_ctrl

<rusty@rustcorp.com.au[helgaas]>:
  o Designated initializers for ia64

<schwab@suse.de[helgaas]>:
  o ia64: fix unwinder bug in unw_access_gr()
  o ia64: Fix request_module from ia32 process
  o ia64: make sys32_ptrace() use ptrace_check_attach()

<shemminger@osdl.org>:
  o [BRIDGE]: Ethernet bridge fixes

<shmulik.hen@intel.com>:
  o Fix load balance problem with high UDP Tx stress
  o Fix 802.3ad long fail over with high UDP Tx stress
  o [netdrvr bonding] Fix change active for ALB/TLB

<sv@sw.com.sg[helgaas]>:
  o ia64: improve show_trace_task() portability

<venkatesh.pallipadi@intel.com[helgaas]>:
  o ia64: IA-32 emulation patch: ptrace get_FPREGS bug fix

<will@sowerbutts.com>:
  o USB: Update for the powermate driver to work with newer devices

Adam J. Richter <adam@yggdrasil.com>:
  o [CRYPTO]: Simplify crypto memory allocation

Adrian Bunk <bunk@fs.tum.de>:
  o postfix a constant in efi.h with ULL

Alan Cox <alan@lxorguk.ukuu.org.uk>:
  o [NETFILTER]: Fix nat_helper warnings with gcc 3.3
  o [NET]: Add EDP2 ethernet protocol ID
  o [SPARC]: d_path() can return an error code, must handle it
  o Optimise FAT handling using the prev_free info as DOS does
  o PATH: add hfsplus file system (stands alone)
  o NLS config.in for hfsplus
  o config.in for HFSPLUS
  o makefile for HFSPLUS
  o fix leak in llc 802
  o fix decnet gcc 3.3 build
  o add xapic checking define
  o add the extra cpu bit test flags
  o remove io_apic_modify - this doesnt work on some APICs
  o add the MSR's for IA32 perf ctl
  o fix false sharing of mm info
  o we moved these so this copy can go
  o collated copy of Geerts patches for m68k headers
  o add a flag so we can forbid APM idling
  o add the ide_register_driver defines
  o add EDP2 protocol id
  o update fat docs - we now use the field
  o bring PCI_IDS back into sync
  o add new entry to sisfb types
  o support cramfs initrd
  o add timedop stub for IPC=n
  o assorted module race fixe
  o dont corrupt utsname on failed copy
  o fix make rpm
  o dont idle if forbid_idle set
  o large scale DMI table updates
  o merge long standing reboot fix form -ac
  o fix up semops and return, allow timedop
  o fix error in vm86 fixups
  o add semtimedop to ia64 emu too
  o fix up gcc 3.3 bits
  o copy the right data in mips emulation
  o collected m68k core diffs
  o typo fix
  o fix iphase leak
  o bump cciss to new vendor driver
  o Jens floppy locking fixes
  o add comtrol note in case we need to know in the future
  o & v && fixes in sysrq.c
  o update sonypi driver
  o parisc gsc driver sync
  o fix config.in bits for IDE
  o make IDE modularisable
  o fix ide dma timeout bugs
  o make pnpide module happy
  o Herbert's fix for ide proc oops
  o make pdc4030 module happy
  o add generic support for toshiba piccolo
  o fix hpt speed bits
  o fix promise sx6000 newer board problems
  o clean up older pdc
  o siimage updates, add aar-1210sa
  o SiS IDE updates
  o hptraid updates
  o small setup-pci cleanups
  o d_path can return an error code, must handle it
  o update motion eye drivers
  o fix leak in octagon
  o new 3c59x. plus handle power bits
  o typo fix in atari_pamsnet driver
  o fix ma600 gcc 3.3
  o minor m68k fixes
  o fix leak in aironet4500_cs
  o fix plip hang on ifdown/ifup
  o update sonic
  o update orinoco drivers
  o update pci.ids
  o add cirrus support to i82092
  o fix rsrc manager
  o pci routing for ti cardbus
  o update aacraid
  o aic7xxx allow db4
  o gdth register failure path
  o update scsi tape docs
  o megaraid broke config tools
  o send_diag wants long timeout default
  o let the ide layer fail commands
  o resync scsi blacklist
  o new segate bios string
  o update scsi tape driver
  o remove noise
  o fix copy from user bug in cmpci
  o update AC97 codec core
  o switch cards to new ac97_audio
  o switch i810 to generalised digital out, new ac97
  o ac97 updates
  o fix long standing doc typo
  o update trident, fix printks, new ac97
  o Update via audio - fix problems esd, mpg321
  o update to new ac97_codec
  o core fbcon fixes
  o update vesafb memory handling for big cards
  o update sis fb drivers
  o add semtimedop to x86 headers
  o update ac97 codec headers
  o declare semtimedop function
  o add scripts ready to merge kconfig
  o update cciss docs to match new driver
  o add vram to vesafb docs
  o CMD640 update
  o (new) Turn on the IDE modular stuff in the Makefile
  o (resend) collected semaphore fixes and semtimedop
  o make i810 audio compile

Alex Williamson <alex_williamson@hp.com>:
  o ia64: CMC deadlock fix

Andi Kleen <ak@muc.de>:
  o Personality fixes for x86-64
  o x86-64 merge
  o Support exception-trace sysctl for x86-64
  o non executable stack support for x86-64

Andrew Morton <akpm@digeo.com>:
  o [CRYPTO]: Fix memcpy/memset args

Ben Collins <bcollins@debian.org>:
  o Update IEEE1394 (r972)

Benjamin Herrenschmidt <benh@kernel.crashing.org>:
  o ppc32: support for 750FX rev2 CPU
  o ppc32: Enable use of USB2 on recent G4s
  o ppc32: Update PowerMac motherboard support
  o ppc32: Update swim3 floppy driver
  o ppc32: Add TotalImpact briQ panel driver
  o ppc32: Add a "query" function to core ADB
  o ppc32: Update adbhid driver
  o ppc32: Update battery calculation code & via-pmu
  o ppc32: Minimal ethtool for bmac and mace
  o ppc32: Fix a problem with both gmac and sungem

Bjorn Helgaas <bjorn_helgaas@hp.com>:
  o ia64: chmod +x unwcheck.sh script
  o ia64: iosapic: make pcat_compat system property
  o ia64: iosapic: rationalize __init/__devinit
  o ia64: Export io_space so drivers using legacy I/O ports can insmod
  o ia64: brl_emu.c: use temporary variable to avoid gcc3.1 warning
  o ia64: remove incorrect and redundant "cpu not responding" message
  o ia64: Update configs
  o ia64: pci.c: Trivial changes to follow 2.5
  o ia64: sba_iommu: use seq_file
  o ia64: acpi: handle vendor resources more generically
  o Move UP cpu_online definition to <linux/smp.h>
  o Cset exclude: rohit.seth@intel.com[helgaas]|ChangeSet|20030623203306|58862

Chris Mason <mason@suse.com>:
  o Fix potential IO hangs and increase interactiveness during heavy IO

Christoph Hellwig <hch@lst.de>:
  o [CRYPTO-2.4]: Missing ULL postfixes and statics

David S. Miller <davem@nuts.ninka.net>:
  o [BK]: Add *~ to ignore regexps
  o [CRYPTO]: kunmap does not return a value
  o [CRYPTO]: Build/warning fixups
  o [CRYPTO]: Clean up header file usage
  o [CRYPTO]: Include kernel.h in crypto.h
  o [CRYPTO]: Allocate work buffers instead of using kstack
  o [CRYPTO]: Make sha256.c more palatable to GCCs optimizers
  o [CRYPTO]: internal.h needs init.h
  o [CRYPTO]: Use appropriate defaults if AH/ESP is enabled
  o [CRYPTO-2.4]: Add dummy kmap_types.h header for sparc64
  o [CRYPTO]: Include linux/errno.h as appropriate
  o [CRYPTO-2.4]: module_name does not exist in 2.4.x
  o [CRYPTO-2.4]: const static --> static const
  o [CRYPTO]: deflate.c needs slab.h
  o [CRYPTO-2.4]: Fix condition typos in crypto/Config.in
  o [CRYPTO-2.4]: Emulate module_name semantics correctly to avoid OOPS
  o [CRYPTO-2.4]: Make sure crypto config is before lib config on ia64
  o [NET]: net/bluetooth/cmtp/core.c needs linux/init.h
  o [NET]: Scale DST/ipv6 intervals like we did for ipv4
  o [SPARC64]: Fix build error from OBP parsing patch
  o [SPARC64]: Update defconfig

Erik Andersen <andersen@codepoet.org>:
  o fix 2.4.22-pre broken x86 math-emu

Greg Kroah-Hartman <greg@kroah.com>:
  o USB: remove stupid conversions and use of floating point from aiptek.c
  o USB: 2.4 fix UHCI debug kmalloc() usage
  o USB: add support for 50 baud to io_edgeport.c
  o USB: pl2303: report CTS and DSR status changes to userspace
  o Cset exclude: cweidema@indiana.edu|ChangeSet|20030620002017|05386
  o USB: compiler fixes for previous vicam patches

Hugh Dickins <hugh@veritas.com>:
  o remove shmem info->sem
  o shmem_getpage absorb _locked
  o shmem_getpage read,cache,write
  o shmem truncation swizzled
  o shmem account metablocks
  o shmem_file_write and _read
  o init_tmpfs shm_mnt error
  o shmem whitespace only
  o shmem misc minor mods
  o swapoff loopable tmpfs
  o shmem mount percentile size
  o shmem_removepage replace recalc_inode
  o loop file use highmem
  o madvise_willneed check readpage
  o shmem_file_write precheck_file_write
  o mremap VM_LOCKED move_vma
  o shmem loopable tmpfs [again]

Ivan Kokshaysky <ink@jurassic.park.msu.ru>:
  o alpha: fix generic kernel build (Jay Estabrook)
  o alpha: finalize Sable/Lynx support (Jay Estabrook)

James Morris <jmorris@intercode.com.au>:
  o [CRYPTO]: Add initial crypto api subsystem
  o [CRYPTO]: Cleanups based upon feedback from Rusty and jgarzik
  o [CRYPTO]: Cleanups based upon feedback from Rusty and jgarzik
  o [CRYPTO]: Use try_inc_mod_count and semaphore for alg list
  o [CRYPTO]: Use kmod to try to autoload modules
  o [CRYPTO]: Bug fixes and cleanups
  o [CRYPTO]: More bug fixes and cleanups
  o [CRYPTO]: Add MD4
  o [CRYPTO]: Algorithm lookup API change plus bug fixes
  o [CRYPTO]: Run tcrypt through lindent, plus doc update
  o [CRYPTO]: Assert that interfaces are called on correct cipher type
  o [CRYPTO]: Cleanups and more consistency checks
  o [CRYPTO]: Update to IV get/set interface
  o [CRYPTO]: Add some documentation
  o [CRYPTO]: Fix some credits
  o [CRYPTO]: Cleanups based upon suggestions by Jeff Garzik
  o [CRYPTO]: Uninline some functions to save some bloat
  o [CRYPTO]: Cleanups based upon feedback from jgarzik
  o [CRYPTO]: Add crypto_alg_available interface
  o [CRYPTO]: Rework HMAC interface
  o [CRYPTO]: Add SHA256 plus bug fixes
  o [CRYPTO]: Add blowfish algorithm
  o [CRYPTO]: minor updates
  o [CRYPTO] kstack cleanup (v0.28)
  o [CRYPTO] Add maintainers entry
  o [CRYPTO] Minor doc update
  o [CRYPTO]: Add null algorithms and minor cleanups
  o [CRYPTO]: Kill stray CRYPTO_ALG_TYPE_COMP
  o [CRYPTO]: Add twofish algorithm
  o [CRYPTO]: Add serpent algorithm
  o [CRYPTO]: Documentation update
  o [CRYPTO]: Dont compile procfs stuff if procfs is not enabled
  o [CRYPTO]: Add AES algorithm
  o [CRYPTO]: More credits for AES
  o [CRYPTO]: Add support for SHA-386 and SHA-512
  o [CRYPTO] remove superfluous goto from des module init exception path
  o [CRYPTO] Add AES and MD4 to tcrypto crypto_alg_available() test
  o [CRYPTO]: in/out scatterlist support for ciphers
  o [CRYPTO]: Move km_types out of header
  o [CRYPTO]: Add encrypt_iv() and decrypt_iv() methods
  o [CRYPTO]: Eliminate crypto_tfm.crt_ctx, from Adam Richter
  o [CRYPTO]: Documentation updates
  o [CRYPTO]: Make use of crypto_exit_ops() during crypto_free_tfm()
  o [CRYPTO]: Add Deflate algorithm to crypto API
  o [CRYPTO]: deflate module: workaround zlib bug
  o [CRYPTO]: Fix config dependencies

Jeff Garzik <jgarzik@redhat.com>:
  o [CRYPTO]: Kill accidental double memset
  o [netdrvr 8139too] fix debug printk

Linus Torvalds <torvalds@transmeta.com>:
  o The crypto auto-load should be enabled if crypto is enabled

Marcelo Tosatti <marcelo@freak.distro.conectiva>:
  o Cset exclude: rusty@rustcorp.com.au|ChangeSet|20030625201246|52045
  o Added missing BROKEN_PNP_BIOS declaration
  o Changed EXTRAVERSION to -pre3

Martin Schwidefsky <schwidefsky@de.ibm.com>:
  o s390 base update
  o s390 common i/o layer fixes
  o s390 dasd driver update
  o s390 31 bit compat
  o s390 documentation update
  o Add Configure.help entries for s390 options
  o s390 3215 driver update
  o s390 ctc network driver update
  o s390 iucv network driver
  o s390 defconfigs update
  o console semaphore fix

Matt Domsch <matt_domsch@dell.com>:
  o ia64: efivars fix by Matt Domsch and Peter Chubb

Olaf Hering <olh@suse.de>:
  o missing asm-ppc64/kmap_types.h

Oleg Drokin <green@angband.namesys.com>:
  o reiserfs: Relocated journal support by Edward Shushkin & Vladimir Saveliev
  o reiserfs: speed up large file holes creation
  o reiserfs: Make most of the reiserfs warning messages to print what device they relate to

Oliver Neukum <oliver@neukum.org>:
  o USB: disconnect of v4l devices in 2.4
  o USB: fix to previous vicam patch

Peter Chubb <peter@chubb.wattle.id.au>:
  o ia64: declare test_bit() arg as "const"

Roger Luethi <rl@hellgate.ch>:
  o [netdrvr via-rhine] via-rhine 1.18-rc1: Fix Rhine-I regression

Russell King <rmk@arm.linux.org.uk>:
  o ARM merge part 1 - arch/arm
  o ARM merge part 2 - include/asm-arm
  o ARM merge part 3 - drivers/acorn

Rusty Russell <rusty@rustcorp.com.au>:
  o 2.5.43 export _end

Scott Feldman <scott.feldman@intel.com>:
  o Remove CAP_NET_ADMIN check for SIOCETHTOOL's

Tom Callaway <tcallawa@redhat.com>:
  o [SPARC64]: Fix OBP version parsing on newer systems

Trond Myklebust <trond.myklebust@fys.uio.no>:
  o A patch by Chuck Lever that cleans up the RPC socket slot allocation code.
  o A patch by Chuck Lever with further cleanups of the RPC socket slot allocation code.
  o Another patch by Chuck Lever that ensures that the PG_uptodate bit gets set when the entire page gets written by nfs_writepage_sync()
  o A patch by Patrice Dumas to implement nlmsvc_proc_granted_res
  o A patch by Patrice Dumas to add a check in order to ensure that we really were requesting a blocking lock when we get a reply from the server asking us to block.
  o A patch to ensures that blocks which are not going to time out are placed last on the nlm_block list (problem reported by Olaf Kirch).
  o Add standard spinlocks to protect the socket from being released by one CPU while the other is in a soft interrupt.
  o Fix a race: Ensure that requests retry if the remote server disconnects us while we're inside xprt_transmit().
  o Don't use an RPC child process when reconnecting to a TCP server
  o Ensure that if we need to reconnect the socket, we also resend the entire message.
  o Fix a TCP client corruption problem affecting resent requests
  o Ensure that the lockd clients always use one of the reserved ports
  o Replace buggy version of xdr_shift_buf() with the version from 2.5.x