* major VM suckage with 2.4.10pre12 and 2.4.10pre13 and highmem, we will help test
@ 2001-09-21 20:15 HABBINGA,ERIK (HP-Loveland,ex1)
2001-09-21 21:00 ` Rik van Riel
2001-09-22 5:44 ` Linus Torvalds
0 siblings, 2 replies; 3+ messages in thread
From: HABBINGA,ERIK (HP-Loveland,ex1) @ 2001-09-21 20:15 UTC (permalink / raw)
To: HABBINGA,ERIK (HP-Loveland,ex1),
'linux-kernel@vger.kernel.org'
Kernel 2.4.10pre13 did not help our NFS SPEC testing on a machine with 4GB
RAM. Refer to my previous message about those results:
http://lists.insecure.org/linux-kernel/2001/Sep/3036.html
In a nutshell, kswapd starts grabbing 99% of the CPU for long stretches in
time, which causes us to drop NFS RPC connections, which causes performance
to suck.
We have access to a few machines with lots of memory (4-8GB), and we are
more than willing to test any patches the community can provide to help
solve this problem. It appears that some of the VM coders don't have access
to highmem machines. We can help you if you let us.
We hoped that the following patch would help with the 0-order allocation
failure message that we see, but we discovered that it has been included in
2.4.10-pre13: http://lists.insecure.org/linux-kernel/2001/Sep/3044.html
Sep 21 12:48:10 catdog1 kernel: __alloc_pages: 3-order allocation failed
(gfp=0x20/0) from c012e1e2
Sep 21 12:48:10 catdog1 kernel: __alloc_pages: 3-order allocation failed
(gfp=0x20/0) from c012e1e2
Sep 21 12:48:10 catdog1 kernel: __alloc_pages: 2-order allocation failed
(gfp=0x20/0) from c012e1e2
Sep 21 12:48:10 catdog1 kernel: __alloc_pages: 3-order allocation failed
(gfp=0x20/0) from c012e1e2
Sep 21 12:48:10 catdog1 kernel: __alloc_pages: 3-order allocation failed
(gfp=0x20/0) from c012e1e2
Sep 21 12:48:10 catdog1 kernel: __alloc_pages: 2-order allocation failed
(gfp=0x20/0) from c012e1e2
Sep 21 12:48:10 catdog1 kernel: __alloc_pages: 1-order allocation failed
(gfp=0x20/0) from c012e1e2
Sep 21 12:48:10 catdog1 kernel: __alloc_pages: 0-order allocation failed
(gfp=0x20/0) from c012e1e2
(there's plenty more where these came from on the 2.4.10-pre13 run)
Hardware:
- 4 processors, 4GB ram
- 45 fibre channel drives, set up in hardware RAID 0/1
- 2 direct Gigabit Ethernet connections between SPEC SFS prime client and
system under test
- reiserfs
- all NFS filesystems exported with sync,no_wdelay to insure O_SYNC writes
to storage
- NFS v3 UDP
- LVM
- linux-2.4.10-pre13
The spec results consist of the following data (only the first three numbers
are significant for this discussion)
- load. The load the SPEC prime client will try to get out of the system
under test. Measured in I/O's per second (IOPS).
- throughput. The load seen from the system under test. Measured in IOPS
- response time. Measured in milliseconds
- total operations
- elapsed time. Measured in seconds
- NFS version. 2 or 3
- Protocol. UDP (U) or TCP (T)
- file set size in megabytes
- number of clients
- number of SPEC SFS processes
- biod reads
- biod writes
- SPEC SFS version
"INVALID" means that too many (> 1%) of the RPC calls between the SPEC prime
client and the system under test failed. This is not a good thing
SPEC results before 2.4.10-pre13 test stopped:
500 600 1.2 180047 300 3 U 5070624 1 48 2 2
2.0
1000 992 1.4 297584 300 3 U 10141248 1 48 2 2
2.0
INVALID 1500 637 32.5 190544 299 3 U 15210624 1 48 2 2
2.0
SPEC results before 2.4.10-pre12 test stopped:
500 497 1.2 149134 300 3 U 5070624 1 48 2 2
2.0
1000 1001 1.5 299153 299 3 U 10141248 1 48 2 2
2.0
1500 1587 6.9 476068 300 3 U 15210624 1 48 2 2
2.0
INVALID 2000 795 24.5 237732 299 3 U 20281248 1 48 2 2
2.0
2.4.7 Results (only first three numbers shown, test continues fine for many
more tests)
500 497 1.2 149206 300 3 U 5070624 1 48 2 2
2.0
1000 1005 1.5 300503 299 3 U 10141248 1 48 2 2
2.0
1500 1502 1.3 449232 299 3 U 15210624 1 48 2 2
2.0
With 2.4.7 kernel, the test runs much longer, and INVALID is never seen.
That large jump in response time on the 1500 IOPS run is also not seen with
2.4.7 kernel
The top screens from interesting parts of the latest NFS test are included
below. I'm willing to try any patches (or measurement tools, within reason)
anyone wants to provide to solve this problem.
Erik Habbinga
Hewlett Packard
12:33pm up 48 min, 2 users, load average: 8.59, 6.52, 4.77
220 processes: 218 sleeping, 2 running, 0 zombie, 0 stopped
CPU0 states: 0.0% user, 80.5% system, 0.1% nice, 18.4% idle
CPU1 states: 0.0% user, 83.3% system, 0.0% nice, 16.1% idle
CPU2 states: 0.2% user, 80.5% system, 0.2% nice, 18.1% idle
CPU3 states: 0.0% user, 79.4% system, 0.0% nice, 20.1% idle
Mem: 3924176K av, 3918500K used, 5676K free, 0K shrd, 96220K
buff
Swap: 1043900K av, 0K used, 1043900K free 3322056K
cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
7 root 19 0 0 0 0 SW 53.1 0.0 0:54 kswapd
9 root 12 0 0 0 0 DW 30.3 0.0 3:48 kupdated
320 root 10 0 0 0 0 SW 7.0 0.0 0:24 nfsd
335 root 9 0 0 0 0 SW 5.3 0.0 0:24 nfsd
352 root 9 0 0 0 0 SW 4.8 0.0 0:23 nfsd
554 root 17 0 1080 1080 768 R 4.8 0.0 1:12 top
392 root 9 0 0 0 0 SW 4.6 0.0 0:24 nfsd
391 root 9 0 0 0 0 SW 4.4 0.0 0:25 nfsd
325 root 9 0 0 0 0 DW 4.0 0.0 0:24 nfsd
359 root 9 0 0 0 0 SW 3.7 0.0 0:24 nfsd
293 root 9 0 0 0 0 SW 3.5 0.0 0:25 nfsd
361 root 9 0 0 0 0 SW 3.5 0.0 0:23 nfsd
374 root 9 0 0 0 0 SW 3.5 0.0 0:24 nfsd
327 root 10 0 0 0 0 SW 3.1 0.0 0:23 nfsd
388 root 9 0 0 0 0 SW 3.1 0.0 0:23 nfsd
12:33pm up 48 min, 2 users, load average: 8.39, 6.51, 4.78
220 processes: 217 sleeping, 3 running, 0 zombie, 0 stopped
CPU0 states: 0.0% user, 81.2% system, 0.0% nice, 18.3% idle
CPU1 states: 0.0% user, 82.3% system, 0.1% nice, 17.0% idle
CPU2 states: 0.0% user, 86.0% system, 0.0% nice, 13.5% idle
CPU3 states: 0.2% user, 86.2% system, 0.0% nice, 13.1% idle
Mem: 3924176K av, 3918436K used, 5740K free, 0K shrd, 96396K
buff
Swap: 1043900K av, 0K used, 1043900K free 3321192K
cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
7 root 17 0 0 0 0 RW 56.3 0.0 0:57 kswapd
9 root 9 0 0 0 0 DW 25.1 0.0 3:50 kupdated
294 root 9 0 0 0 0 SW 5.6 0.0 0:24 nfsd
303 root 9 0 0 0 0 SW 4.5 0.0 0:24 nfsd
334 root 9 0 0 0 0 SW 4.3 0.0 0:23 nfsd
312 root 9 0 0 0 0 SW 4.1 0.0 0:25 nfsd
308 root 9 0 0 0 0 SW 3.9 0.0 0:24 nfsd
403 root 9 0 0 0 0 SW 3.9 0.0 0:24 nfsd
554 root 15 0 1080 1080 768 R 3.9 0.0 1:12 top
302 root 9 0 0 0 0 SW 3.7 0.0 0:24 nfsd
412 root 9 0 0 0 0 SW 3.7 0.0 0:23 nfsd
310 root 9 0 0 0 0 SW 3.6 0.0 0:24 nfsd
404 root 9 0 0 0 0 SW 3.6 0.0 0:24 nfsd
340 root 9 0 0 0 0 SW 3.4 0.0 0:24 nfsd
295 root 9 0 0 0 0 SW 3.2 0.0 0:24 nfsd
12:33pm up 48 min, 2 users, load average: 8.68, 6.60, 4.81
220 processes: 214 sleeping, 6 running, 0 zombie, 0 stopped
CPU0 states: 0.0% user, 94.0% system, 0.1% nice, 5.5% idle
CPU1 states: 0.0% user, 94.0% system, 0.2% nice, 5.4% idle
CPU2 states: 0.0% user, 94.0% system, 0.0% nice, 5.6% idle
CPU3 states: 0.1% user, 93.4% system, 0.0% nice, 6.1% idle
Mem: 3924176K av, 3919232K used, 4944K free, 0K shrd, 96472K
buff
Swap: 1043900K av, 0K used, 1043900K free 3321548K
cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
7 root 15 0 0 0 0 RW 59.6 0.0 1:01 kswapd
9 root 12 0 0 0 0 DW 22.5 0.0 3:51 kupdated
554 root 12 0 1080 1080 768 R 19.4 0.0 1:14 top
325 root 9 0 0 0 0 SW 4.3 0.0 0:25 nfsd
361 root 9 0 0 0 0 SW 4.1 0.0 0:24 nfsd
336 root 9 0 0 0 0 DW 3.8 0.0 0:24 nfsd
394 root 9 0 0 0 0 SW 3.6 0.0 0:24 nfsd
302 root 9 0 0 0 0 SW 3.4 0.0 0:25 nfsd
370 root 9 0 0 0 0 SW 3.4 0.0 0:24 nfsd
311 root 10 0 0 0 0 SW 3.3 0.0 0:24 nfsd
412 root 9 0 0 0 0 SW 3.3 0.0 0:23 nfsd
368 root 9 0 0 0 0 SW 3.1 0.0 0:24 nfsd
371 root 9 0 0 0 0 SW 3.1 0.0 0:25 nfsd
375 root 9 0 0 0 0 SW 3.1 0.0 0:24 nfsd
397 root 9 0 0 0 0 SW 3.1 0.0 0:24 nfsd
12:33pm up 48 min, 2 users, load average: 8.87, 6.71, 4.87
220 processes: 215 sleeping, 5 running, 0 zombie, 0 stopped
CPU0 states: 0.0% user, 94.6% system, 0.1% nice, 5.0% idle
CPU1 states: 0.0% user, 94.1% system, 0.0% nice, 5.6% idle
CPU2 states: 0.0% user, 93.5% system, 0.4% nice, 5.5% idle
CPU3 states: 1.4% user, 93.4% system, 0.0% nice, 4.5% idle
Mem: 3924176K av, 3919148K used, 5028K free, 0K shrd, 96268K
buff
Swap: 1043900K av, 0K used, 1043900K free 3321116K
cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
7 root 14 0 0 0 0 RW 81.3 0.0 1:07 kswapd
554 root 18 0 1080 1080 768 R 23.8 0.0 1:15 top
9 root 9 0 0 0 0 DW 22.0 0.0 3:53 kupdated
326 root 9 0 0 0 0 SW 4.6 0.0 0:23 nfsd
351 root 10 0 0 0 0 SW 4.3 0.0 0:24 nfsd
374 root 9 0 0 0 0 SW 4.3 0.0 0:25 nfsd
289 root 9 0 0 0 0 SW 4.2 0.0 0:24 nfsd
331 root 9 0 0 0 0 RW 3.7 0.0 0:25 nfsd
406 root 9 0 0 0 0 SW 3.7 0.0 0:25 nfsd
397 root 9 0 0 0 0 SW 3.5 0.0 0:25 nfsd
332 root 9 0 0 0 0 SW 3.4 0.0 0:24 nfsd
353 root 9 0 0 0 0 SW 3.3 0.0 0:25 nfsd
346 root 9 0 0 0 0 SW 3.1 0.0 0:25 nfsd
402 root 9 0 0 0 0 SW 3.1 0.0 0:25 nfsd
412 root 10 0 0 0 0 SW 3.1 0.0 0:24 nfsd
12:33pm up 49 min, 2 users, load average: 8.88, 6.75, 4.89
220 processes: 214 sleeping, 6 running, 0 zombie, 0 stopped
CPU0 states: 0.1% user, 98.3% system, 0.3% nice, 0.6% idle
CPU1 states: 0.0% user, 98.1% system, 0.0% nice, 1.5% idle
CPU2 states: 0.0% user, 99.1% system, 0.0% nice, 0.5% idle
CPU3 states: 0.0% user, 97.3% system, 0.1% nice, 2.2% idle
Mem: 3924176K av, 3918888K used, 5288K free, 0K shrd, 96076K
buff
Swap: 1043900K av, 0K used, 1043900K free 3320648K
cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
7 root 18 0 0 0 0 RW 99.9 0.0 1:14 kswapd
9 root 9 0 0 0 0 RW 18.0 0.0 3:54 kupdated
554 root 19 0 1080 1080 768 R 17.6 0.0 1:17 top
410 root 9 0 0 0 0 SW 3.8 0.0 0:24 nfsd
289 root 9 0 0 0 0 SW 3.7 0.0 0:24 nfsd
295 root 9 0 0 0 0 DW 3.7 0.0 0:24 nfsd
342 root 9 0 0 0 0 SW 3.7 0.0 0:24 nfsd
401 root 9 0 0 0 0 SW 3.7 0.0 0:24 nfsd
369 root 9 0 0 0 0 SW 3.5 0.0 0:25 nfsd
396 root 9 0 0 0 0 SW 3.5 0.0 0:24 nfsd
322 root 9 0 0 0 0 SW 3.4 0.0 0:24 nfsd
361 root 9 0 0 0 0 SW 3.2 0.0 0:24 nfsd
291 root 9 0 0 0 0 SW 3.1 0.0 0:25 nfsd
301 root 9 0 0 0 0 SW 3.1 0.0 0:24 nfsd
318 root 9 0 0 0 0 SW 3.1 0.0 0:24 nfsd
12:33pm up 49 min, 2 users, load average: 8.73, 6.75, 4.90
220 processes: 216 sleeping, 4 running, 0 zombie, 0 stopped
CPU0 states: 0.1% user, 99.1% system, 0.1% nice, 0.3% idle
CPU1 states: 2.0% user, 97.4% system, 0.1% nice, 0.1% idle
CPU2 states: 0.1% user, 99.4% system, 0.0% nice, 0.1% idle
CPU3 states: 0.0% user, 99.5% system, 0.1% nice, 0.0% idle
Mem: 3924176K av, 3919156K used, 5020K free, 0K shrd, 96028K
buff
Swap: 1043900K av, 0K used, 1043900K free 3320772K
cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
7 root 17 0 0 0 0 RW 99.9 0.0 1:21 kswapd
9 root 9 0 0 0 0 DW 24.3 0.0 3:56 kupdated
554 root 19 0 1080 1080 768 R 17.3 0.0 1:18 top
388 root 9 0 0 0 0 SW 4.9 0.0 0:24 nfsd
415 root 9 0 0 0 0 SW 4.9 0.0 0:25 nfsd
381 root 9 0 0 0 0 SW 4.7 0.0 0:24 nfsd
398 root 9 0 0 0 0 DW 4.3 0.0 0:24 nfsd
396 root 9 0 0 0 0 SW 4.1 0.0 0:25 nfsd
341 root 9 0 0 0 0 SW 4.0 0.0 0:25 nfsd
401 root 9 0 0 0 0 SW 4.0 0.0 0:25 nfsd
294 root 9 0 0 0 0 SW 3.8 0.0 0:25 nfsd
347 root 9 0 0 0 0 SW 3.2 0.0 0:25 nfsd
352 root 9 0 0 0 0 SW 3.2 0.0 0:24 nfsd
303 root 9 0 0 0 0 SW 3.1 0.0 0:24 nfsd
402 root 9 0 0 0 0 SW 3.1 0.0 0:25 nfsd
...and this contiues for quite awhile...
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: major VM suckage with 2.4.10pre12 and 2.4.10pre13 and highmem, we will help test
2001-09-21 20:15 major VM suckage with 2.4.10pre12 and 2.4.10pre13 and highmem, we will help test HABBINGA,ERIK (HP-Loveland,ex1)
@ 2001-09-21 21:00 ` Rik van Riel
2001-09-22 5:44 ` Linus Torvalds
1 sibling, 0 replies; 3+ messages in thread
From: Rik van Riel @ 2001-09-21 21:00 UTC (permalink / raw)
To: HABBINGA,ERIK (HP-Loveland,ex1); +Cc: 'linux-kernel@vger.kernel.org'
On Fri, 21 Sep 2001, HABBINGA,ERIK (HP-Loveland,ex1) wrote:
> Kernel 2.4.10pre13 did not help our NFS SPEC testing on a machine with
> 4GB RAM. Refer to my previous message about those results:
> http://lists.insecure.org/linux-kernel/2001/Sep/3036.html
>
> In a nutshell, kswapd starts grabbing 99% of the CPU for long
> stretches in time, which causes us to drop NFS RPC connections, which
> causes performance to suck.
I'm curious, how do recent -ac kernels perform here ?
If you have the time, could you test 2.4.9-ac13 plain
and 2.4.9-ac13 with my page aging and launder patches
from http://www.surriel.com/patches/ ? ;)
cheers,
Rik
--
IA64: a worthy successor to i860.
http://www.surriel.com/ http://distro.conectiva.com/
Send all your spam to aardvark@nl.linux.org (spam digging piggy)
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: major VM suckage with 2.4.10pre12 and 2.4.10pre13 and highmem, we will help test
2001-09-21 20:15 major VM suckage with 2.4.10pre12 and 2.4.10pre13 and highmem, we will help test HABBINGA,ERIK (HP-Loveland,ex1)
2001-09-21 21:00 ` Rik van Riel
@ 2001-09-22 5:44 ` Linus Torvalds
1 sibling, 0 replies; 3+ messages in thread
From: Linus Torvalds @ 2001-09-22 5:44 UTC (permalink / raw)
To: linux-kernel
In article <F341E03C8ED6D311805E00902761278C04728F82@xfc04.fc.hp.com>,
HABBINGA,ERIK (HP-Loveland,ex1) <erik_habbinga@hp.com> wrote:
>Kernel 2.4.10pre13 did not help our NFS SPEC testing on a machine with 4GB
>RAM. Refer to my previous message about those results:
>http://lists.insecure.org/linux-kernel/2001/Sep/3036.html
>
>In a nutshell, kswapd starts grabbing 99% of the CPU for long stretches in
>time, which causes us to drop NFS RPC connections, which causes performance
>to suck.
Duh.
No wonder.
There's this test in page_alloc.c that tests for atomic allocations the
wrong way - it uses "(gfp_mask & __GFP_HIGH)", when the test _should_ be
"!(gfp_mask & __GFP_WAIT)".
Which causes regular kernel allocations to eat into the reserved memory
pool. Which in turn makes the _real_ atomic allocations really rather
unhappy (resulting in lost packets - which pretty much guarantees sucky
NFS performance ;).
And it probably doesn't make kswapd that happy either...
You can either fix that single line, or try out pre14 which cleans up
some other things too..
Linus
-----
pre14:
- Richard Gooch: devfs update
- Andrea Arcangeli: clean up/fix ramdisk handling now that it's in page cache
- Al Viro: follow up the above with initrd cleanups
- Keith Owens: get rid of drivers/scsi/53c700-mem.c file
- Trond Myklebust: RPC over TCP race fix
- Greg KH: USB update (ohci understands USB_ZERO_PACKET)
- me: clean up reference bit handling, fix silly GFP_ATOMIC allocation bug
pre13:
- Manfred Spraul: /proc/pid/maps cleanup (and bugfix for non-x86)
- Al Viro: "block device fs" - cleanup of page cache handling
- Hugh Dickins: VM/shmem cleanups and swap search speedup
- David Miller: sparc updates, soc driver typo fix, net updates
- Jeff Garzik: network driver updates (dl2k, yellowfin and tulip)
- Neil Brown: knfsd cleanups and fixues
- Ben LaHaise: zap_page_range merge from -ac
pre12:
- Alan Cox: much more merging
- Pete Zaitcev: ymfpci race fixes
- Andrea Arkangeli: VM race fix and OOM tweak.
- Arjan Van de Ven: merge RH kernel fixes
- Andi Kleen: use more readable 'likely()/unlikely()' instead of __builtin_expect()
- Keith Owens: fix 64-bit ELF types
- Gerd Knorr: mark more broken PCI bridges, update btaudio driver
- Paul Mackerras: powermac driver update
- me: clean up PTRACE_DETACH to use common infrastructure
pre11:
- Neil Brown: md cleanups/fixes
- Andrew Morton: console locking merge
- Andrea Arkangeli: major VM merge
pre10:
- Alan Cox: continued merging
- Mingming Cao: make msgrcv/shmat check the queue/segment ID's properly
- Greg KH: USB serial init failure fix, Xircom serial converter driver
- Neil Brown: nsfd/raid/md/lockd cleanups
- Ingo Molnar: multipath RAID personality, raid xor update
- Hugh Dickins/Marcelo Tosatti: swapin read-ahead race fix
- Vojtech Pavlik: fix up some of the infrastructure for x86-64
- Robert Love: AMD 761 AGP GART support
- Jens Axboe: fix SCSI-generic queue handling race
- me: be sane about page reference bits
pre9:
- Greg KH: start migration to new "min()/max()"
- Roman Zippel: move affs over to "min()/max()".
- Vojtech Pavlik: VIA update (make sure not to IRQ-unmask a vt82c576)
- Jan Kara: quota bug-fix (don't decrement quota for non-counted inode)
- Anton Altaparmakov: more NTFS updates
- Al Viro: make nosuid/noexec/nodev be per-mount flags, not per-filesystem
- Alan Cox: merge input/joystick layer differences, driver and alpha merge
- Keith Owens: scsi Makefile cleanup
- Trond Myklebust: fix oopsable race in locking code
- Jean Tourrilhes: IrDA update
pre8:
- Christoph Hellwig: clean up personality handling a bit
- Robert Love: update sysctl/vm documentation
- make the three-argument (that everybody hates) "min()" be "min_t()",
and introduce a type-anal "min()" that complains about arguments of
different types.
pre7:
- Alan Cox: big driver/mips sync
- Andries Brouwer, Christoph Hellwig: more gendisk fixups
- Tobias Ringstrom: tulip driver workaround for DC21143 erratum
pre6:
- Jens Axboe: remove trivially dead io_request_lock usage
- Andrea Arcangeli: softirq cleanup and ARM fixes. Slab cleanups
- Christoph Hellwig: gendisk handling helper functions/cleanups
- Nikita Danilov: reiserfs dead code pruning
- Anton Altaparmakov: NTFS update to 1.1.18
- firestream network driver: patch reverted on authors request
- NIIBE Yutaka: SH architecture update
- Paul Mackerras: PPC cleanups, PPC8xx update.
- me: reverse broken bootdata allocation patch that went into pre5
pre5:
- Merge with Alan
- Trond Myklebust: NFS fixes - kmap and root inode special case
- Al Viro: more superblock cleanups, inode leak in rd.c, minix
directories in page cache
- Paul Mackerras: clean up rubbish from sl82c105.c
- Neil Brown: md/raid cleanups, NFS filehandles
- Johannes Erdfelt: USB update (usb-2.0 support, visor fix, Clie fix,
pl2303 driver update)
- David Miller: sparc and net update
- Eric Biederman: simplify and correct bootdata allocation - don't
overwrite ramdisks
- Tim Waugh: support multiple SuperIO devices, parport doc updates
pre4:
- Hugh Dickins: swapoff cleanups and speedups
- Matthew Dharm: USB storage update
- Keith Owens: Makefile fixes
- Tom Rini: MPC8xx build fix
- Nikita Danilov: reiserfs update
- Jakub Jelinek: ELF loader fix for ET_DYN
- Andrew Morton: reparent_to_init() for kernel threads
- Christoph Hellwig: VxFS and SysV updates, vfs_permission fix
pre3:
- Johannes Erdfelt, Oliver Neukum: USB printer driver race fix
- John Byrne: fix stupid i386-SMP irq stack layout bug
- Andreas Bombe, me: yenta IO window fix
- Neil Brown: raid1 buffer state fix
- David Miller, Paul Mackerras: fix up sparc and ppc respectively for kmap/kbd_rate
- Matija Nalis: umsdos fixes, and make it possible to boot up with umsdos
- Francois Romieu: fix bugs in dscc4 driver
- Andy Grover: new PCI config space access functions (eventually for ACPI)
- Albert Cranford: fix incorrect e2fsprog data from ver_linux script
- Dave Jones: re-sync x86 setup code, fix macsonic kmalloc use
- Johannes Erdfelt: remove obsolete plusb USB driver
- Andries Brouwer: fix USB compact flash version info, add blksize ioctls
pre2:
- Al Viro: block device cleanups
- Marcelo Tosatti: make bounce buffer allocations more robust (it's ok
for them to do IO, just not cause recursive bounce IO. So allow them)
- Anton Altaparmakov: NTFS update (1.1.17)
- Paul Mackerras: PPC update (big re-org)
- Petko Manolov: USB pegasus driver fixes
- David Miller: networking and sparc updates
- Trond Myklebust: Export atomic_dec_and_lock
- OGAWA Hirofumi: find and fix umsdos "filldir" users that were broken
by the 64-bit-cleanups. Fix msdos warnings.
- Al Viro: superblock handling cleanups and race fixes
- Johannes Erdfelt++: USB updates
pre1:
- Jeff Hartmann: DRM AGP/alpha cleanups
- Ben LaHaise: highmem user pagecopy/clear optimization
- Vojtech Pavlik: VIA IDE driver update
- Herbert Xu: make cramfs work with HIGHMEM pages
- David Fennell: awe32 ram size detection improvement
- Istvan Varadi: umsdos EMD filename bug fix
- Keith Owens: make min/max work for pointers too
- Jan Kara: quota initialization fix
- Brad Hards: Kaweth USB driver update (enable, and fix endianness)
- Ralf Baechle: MIPS updates
- David Gibson: airport driver update
- Rogier Wolff: firestream ATM driver multi-phy support
- Daniel Phillips: swap read page referenced set - avoid swap thrashing
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2001-09-22 5:46 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2001-09-21 20:15 major VM suckage with 2.4.10pre12 and 2.4.10pre13 and highmem, we will help test HABBINGA,ERIK (HP-Loveland,ex1)
2001-09-21 21:00 ` Rik van Riel
2001-09-22 5:44 ` Linus Torvalds
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox