ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.0-test2/2.6.0-test2-mm1/
- More CPU scheduler tweaks.
- Some changes to the VM which are designed to further reduce the amount of
writeout which happens off the tail of the page LRU.
Some small benefits have been observed in the usual benchmarks. Needs
careful testing.
- Various other fixes from various people.
Changes since 2.6.0-test1-mm2:
-airo-fixes.patch
-ide-tcq-fix.patch
-speedstep-ich-timing-fix.patch
Merged
+cpumask_t-gcc-workaround-47.patch
More NR_CPUS > BITS_PER_LONG fixes
-fbdev-2.patch
Lots of rejects against Linus changes. Dropped for now.
+rootdisk-parsing-fix.patch
Fix "root=/dev/hda1" kernel boot option
+o8int.patch
+o9int.patch
+o10int.patch
CPU scheduler work.
-highpmd.patch
Dropped - too big.
-floppy-req-botched.patch
Dropped: wrong.
+alloc_bootmem_low_pages-ordering-fix.patch
bootmem fix for ia64
+floppy-smp-fixes.patch
Floppy driver SMP fixes
+1000HZ-time-accuracy-fix.patch
More accurate timekeeping
+sis-drm-fix.patch
SiS DRM driver fix
+signal-race-fix.patch
Signal handling race fix
+soundcard-devfs-fix.patch
Fix sound drivers with devfs
+6pack-hz-fix.patch
Fix the 6pack driver for HZ != 100
+devfs_lookup-revert-and-refix.patch
devfs fix fix
+write-mark_page_accessed.patch
Activate pages on the second pagecache write()
+zone-pressure.patch
+reclaim-mapped-pressure.patch
More accurate page unmapping in page reclaim
+vmscan-defer-writepage.patch
Give dirty pages another go around the inactive list before writing them
out.
+blacklist-asus-L3800C-dmi.patch
ASUS DMi fix
+force-CONFIG_INPUT.patch
Hopefully make migration from 2.4 easier.
+ipt_helper-build-fix.patch
netfilter build fix
+sb16-ioports-fix.patch
Soundblaster leaks resources
+nforce2-acpi-fixes.patch
ACPI fixes for nForce.
+select-xoffed-tty-fix.patch
Correctly handle select() against xoffed tty's
+conntrack-build-fix.patch
Another netfilter build fix
+arcnet-typo-fix.patch
arcnet fix
+ext3-commit-assertion-fix.patch
ext3 data=journal BUGfix
All 127 patches:
mm.patch
add -mmN to EXTRAVERSION
kgdb-ga.patch
kgdb stub for ia32 (George Anzinger's one)
kgdb-remove-cpu_callout_map.patch
kgdb: remove cpu_callout_map decls
kgdb-use-ggdb.patch
kgdb-ga-docco-fixes.patch
kgdb doc. edits/corrections
cpumask_t-1.patch
cpumask_t: allow more than BITS_PER_LONG CPUs
cpumask_t fix for s390
fix cpumask_t for s390
Fix cpumask changes for x86_64
fix cpumask_t for sparc64
cpumask_t-gcc-workaround-46.patch
cpumask_t: more gcc workarounds
cpumask_t-gcc-workaround-47.patch
cpumask_t gcc bug workarounds
cpumask-acpi-fix.patch
cpumask_t: build fix
kgdb-cpumask_t.patch
misc31.patch
misc fixes
selinux.patch
reslabify-pgds-and-pmds.patch
re-slabify i386 pgd's and pmd's
config_spinline.patch
uninline spinlocks for profiling accuracy.
ppc64-bar-0-fix.patch
Allow PCI BARs that start at 0
ppc64-reloc_hide.patch
ppc64-semaphore-reimplementation.patch
ppc64: use the ia32 semaphore implementation
sym-do-160.patch
make the SYM driver do 160 MB/sec
ia64-percpu-revert.patch
revert percpu changes
x86_64-fixes.patch
x86_64 fixes
delay-ksoftirqd-fallback.patch
Try harded in IRQ context before falling back to ksoftirqd
ds-09-vicam-usercopy-fix.patch
vicam usercopy fix
buffer-debug.patch
buffer.c debugging
rcu-stats.patch
RCU statistics reporting
mtrr-hang-fix.patch
Fix mtrr-related hang
intel8x0-cleanup.patch
intel8x0 cleanups
bio-too-big-fix.patch
Fix raid "bio too big" failures
centrino-update.patch
update to speedstep-centrino.c
ppa-fix.patch
ppc fix
3c59x-pm-fix.patch
3c59x suspend/resume fix
dev_t-printing.patch
dev_t printing
rootdisk-parsing-fix.patch
fix "unable to mount root fs"
3c59x-eisa-fix.patch
non-MII 3c59x fix
slab-reclaim-accounting-fix.patch
kwsapd can free too much memory
timer-spin-fix.patch
ALSA locking fix
ak4xxx-fix.patch
fix snd-ice1724 module OOPS
less-kswapd-throttling.patch
stack-leak-fix.patch
info leak -- padded struct copied to user
unlock_buffer-barrier.patch
unlock_buffer() needs a barrier
linux-isp-2.patch
linux-isp-2-fix-again.patch
lost feral fix
feral-bounce-fix.patch
Feral driver - highmem issues
feral-bounce-fix-2.patch
Feral driver bouncing fix
list_del-debug.patch
list_del debug check
print-build-options-on-oops.patch
print a few config options on oops
show_task-free-stack-fix.patch
show_task() fix and cleanup
put_task_struct-debug.patch
ia32-mknod64.patch
mknod64 for ia32
ext2-64-bit-special-inodes.patch
ext2: support for 64-bit device nodes
ext3-64-bit-special-inodes.patch
ext3: support for 64-bit device nodes
64-bit-dev_t-kdev_t.patch
64-bit dev_t and kdev_t
64-bit-dev_t-other-archs.patch
enable 64-bit dev_t for other archs
oops-dump-preceding-code.patch
i386 oops output: dump preceding code
lockmeter.patch
invalidate_mmap_range.patch
Interface to invalidate regions of mmaps
aio-mm-refcounting-fix.patch
fix /proc mm_struct refcounting bug
aio-01-retry.patch
AIO: Core retry infrastructure
io_submit_one-EINVAL-fix.patch
Fix aio process hang on EINVAL
aio-02-lockpage_wq.patch
AIO: Async page wait
aio-03-fs_read.patch
AIO: Filesystem aio read
aio-04-buffer_wq.patch
AIO: Async buffer wait
aio-05-fs_write.patch
AIO: Filesystem aio write
aio-05-fs_write-fix.patch
aio-06-bread_wq.patch
AIO: Async block read
aio-06-bread_wq-fix.patch
aio-07-ext2getblk_wq.patch
AIO: Async get block for ext2
O_SYNC-speedup-2.patch
speed up O_SYNC writes
aio-09-o_sync.patch
aio O_SYNC
aio-10-BUG-fix.patch
AIO: fix a BUG
aio-11-workqueue-flush.patch
AIO: flush workqueues before destroying ioctx'es
aio-12-readahead.patch
AIO: readahead fixes
aio-dio-no-readahead.patch
aio O_DIRECT no readahead
lock_buffer_wq-fix.patch
lock_buffer_wq fix
unuse_mm-locked.patch
AIO: hold the context lock across unuse_mm
aio-take-task_lock.patch
From: Suparna Bhattacharya <[email protected]>
Subject: Re: 2.5.72-mm1 - Under heavy testing with AIO,.. vmstat seems to blow the kernel
aio-O_SYNC-fix.patch
Unify o_sync changes for aio and regular writes
aio-readahead-rework.patch
Unified page range readahead for aio and regular reads
truncate-pagefault-race-fix.patch
Fix vmtruncate race and distributed filesystem race
truncate-pagefault-race-fix-fix.patch
Make sure truncate fix has no race
printk-oops-mangle-fix.patch
disentangle printk's whilst oopsing on SMP
20-odirect_enable.patch
21-odirect_cruft.patch
22-read_proc.patch
23-write_proc.patch
24-commit_proc.patch
25-odirect.patch
nfs-O_DIRECT-always-enabled.patch
Force CONFIG_NFS_DIRECTIO
kjournald-PF_SYNCWRITE.patch
o1-interactivity.patch
CPU scheduler interactivity patch
o2int.patch
O2int 0307041440 for 2.5.74-mm1
o3int.patch
O3int interactivity for 2.5.74-mm2
o4int.patch
O4int interactivity
o5int-2.patch
O5int for interactivity
o6int.patch
O6int for interactivity
o6.1int.patch
O6.1int
o7int.patch
O7int for interactivity
o8int.patch
O8int for interactivity
o9int.patch
O9int for interactivity
o10int.patch
O10int for interactivity
sched-balance-tuning.patch
CPU scheduler balancing fix
synaptics-reset-fix.patch
synaptics driver reset fix
ext3-block-allocation-cleanup.patch
nfs-revert-backoff.patch
nfs: revert backoff changes
ext3-elide-inode-block-reading.patch
ext3: avoid reading empty inode blocks
ext3_getblk-race-fix.patch
Fix race in ext3_getblk
ext3_write_super-speedup.patch
ext3: don't start a commit in write_super()
alloc_bootmem_low_pages-ordering-fix.patch
fix alloc_bootmem_low_pages
floppy-smp-fixes.patch
floppy smp fixes
1000HZ-time-accuracy-fix.patch
missing #if for 1000 HZ
sis-drm-fix.patch
SiS RM fix
signal-race-fix.patch
signal handling race condition causing reboot hangs
soundcard-devfs-fix.patch
soundcard.c devfs fix
6pack-hz-fix.patch
6PACK asumes HZ=100
devfs_lookup-revert-and-refix.patch
devfs_lookup stack corruption fix rework
write-mark_page_accessed.patch
use mark_page_accessed() in the write() path
zone-pressure.patch
vmscan: decaying average of zone pressure
reclaim-mapped-pressure.patch
vmscan: use zone_pressure for page unmapping decisions
vmscan-defer-writepage.patch
vmscan: give dirty referenced pages another pass around the LRU
blacklist-asus-L3800C-dmi.patch
add ASUS l3800P to DMI black list
force-CONFIG_INPUT.patch
Force CONFIG_INPUT if CONFIG_VT set
ipt_helper-build-fix.patch
Fix ipt_helper compilation
sb16-ioports-fix.patch
Fix /proc/ioports with sb16
nforce2-acpi-fixes.patch
ACPI patch which fixes all my IRQ problems on nforce2
select-xoffed-tty-fix.patch
fix select() with an xoffed tty
conntrack-build-fix.patch
fix ip_conntrack_core.h compile error
arcnet-typo-fix.patch
typo in drivers/net/arcnet/com20020-isa.c
ext3-commit-assertion-fix.patch
ext3: fix commit assertion failure
El Sun, 27 Jul 2003 23:37:16 -0700 Andrew Morton <[email protected]> escribi?:
> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.0-test2/2.6.0-test2-mm1/
>
> - More CPU scheduler tweaks.
O10 feels great here; behaviour under heavy load (make -jbignumber) is great;
gcc doesn't starves the rest of the processes; it still allows X/xchat/xmms/etc
do some work and the system remains usable; mp3 doesn't skip too much (only
when it tries to swapin some big process like galeon but i find that normal; before
this the same load in the past starved anything not classified as "compiler").
On Tue, 29 Jul 2003 10:38, Diego Calleja Garc?a wrote:
> El Sun, 27 Jul 2003 23:37:16 -0700 Andrew Morton <[email protected]> escribi?:
> > ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.0-test2
> >/2.6.0-test2-mm1/
> >
> > - More CPU scheduler tweaks.
>
> O10 feels great here; behaviour under heavy load (make -jbignumber) is
> great; gcc doesn't starves the rest of the processes; it still allows
> X/xchat/xmms/etc do some work and the system remains usable; mp3 doesn't
> skip too much (only when it tries to swapin some big process like galeon
> but i find that normal; before this the same load in the past starved
> anything not classified as "compiler").
Thanks. The swap or even heavy vm thing affecting interactivity is something
I've been thinking about for some time but I've yet to figure out a good way
to feed back the vm's activity into the scheduler in a meaningful way.
Con
Hi,
Some things I noticed:
1- Seems like you missed out framebuffer patch ( there was a little s/</<=
patch ) so we don't see any penguin at startup anymore.
2- Con's patch makes KDE's sound daemon skip ( aRts ) when using Juk ( KDE
JukeBox ) [ to skip just minimize/maximize any window fast ] . Seems like
problem is at aRts decoding as mplayer -ao arts works fine without skips.
Regards,
/ismail
On Tuesday 29 July 2003 07.33, ismail (cartman) donmez wrote:
> Hi,
>
> Some things I noticed:
>
> -- snip --
>
> 2- Con's patch makes KDE's sound daemon skip ( aRts ) when using Juk ( KDE
> JukeBox ) [ to skip just minimize/maximize any window fast ] . Seems like
> problem is at aRts decoding as mplayer -ao arts works fine without skips.
>
Is mplayer suid root? (allow SCHED_FIFO/RR usage)
To get the equivalent function you need artswrapper to be suid root,
and "Run soundserver with realtime priority" feature enabled in aRTs config.
[Dangerous with multiuser setups - user can add a plug in that loops forever
with higher priority than all other processes -> dead system.
The SCHED_SOFTRR patch would be a very welcome addition.
a) No need for suid root, artswrapper tries to use SCHED_FIFO
Davide - SCHED_FIFO should also be handled as a SCHED_SOFTRR request!
Not only SCHED_RR (SCHED_FIFO is more frequently used!)
b) Automatic prevention of overuse
]
If this is not the case - my mplayer is not suid root.
aRTs adds some overhead to allow mixing audio from several applications.
(even if this is not necessary on some audio boards like SB Live!)
To get the same kind of overhead you should run mplayer with output through
arts...
artsdsp mplayer ...
The final thing to check is KDE Audio Buffer Size (if this is small then it
won't matter if mplayer uses a big buffer...)
/RogerL
--
Roger Larsson
Skellefte?
Sweden
On Tuesday 29 July 2003 22:59, Roger Larsson wrote:
> Is mplayer suid root? (allow SCHED_FIFO/RR usage)
> To get the equivalent function you need artswrapper to be suid root,
> and "Run soundserver with realtime priority" feature enabled in aRTs
> config.
>
No and aRts is running as root, I will check buffer size thing.
/cartman
On Sun, Jul 27, 2003 at 11:37:16PM -0700, Andrew Morton wrote:
>
> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.0-test2/2.6.0-test2-mm1/
>
> - More CPU scheduler tweaks.
>
> - Some changes to the VM which are designed to further reduce the amount of
> writeout which happens off the tail of the page LRU.
>
> Some small benefits have been observed in the usual benchmarks. Needs
> careful testing.
>
> - Various other fixes from various people.
Hi there Andrew.
I have 2.6.0-test2-mm1 running on two (XP2600 & PII300 named 10.6) machines in this
scenario, and a pIII500 named fs running 2.4.20-rmap15e-fs196-1.4, and a
PIII550 named 10.4 running 2.4.21-rc1-rmap15g.
I have the xp2600 copying large directory trees from each of the three
systems over the network.
I get nfs errors from the activity with the 2.4 systems, but not from the
slowest system that's also running 2.6.
I also got an error "NFS: giant filename in readdir (len
792b9a88)!" but it doesn't attribute it to a certain client...
I see this all the time on my 2.4 nfs clients too, so it's not 2.6 specific.
Though it was interesting that there weren't any errors reported from the
2.6 nfs server.
Anyway, I'd like to know what the trouble is with this that I've been having
for a while now.
Thanks
xp2600:
00:00.0 Host bridge: VIA Technologies, Inc. VT8377 [KT400 AGP] Host Bridge
00:01.0 PCI bridge: VIA Technologies, Inc. VT8235 PCI Bridge
00:0a.0 VGA compatible controller: Trident Microsystems TGUI 9660/938x/968x
(rev d3)
00:0e.0 RAID bus controller: CMD Technology Inc PCI0680 (rev 02)
00:10.0 USB Controller: VIA Technologies, Inc. USB (rev 80)
00:10.1 USB Controller: VIA Technologies, Inc. USB (rev 80)
00:10.2 USB Controller: VIA Technologies, Inc. USB (rev 80)
00:10.3 USB Controller: VIA Technologies, Inc. USB 2.0 (rev 82)
00:11.0 ISA bridge: VIA Technologies, Inc. VT8235 ISA Bridge
00:11.1 IDE interface: VIA Technologies, Inc. VT82C586/B/686A/B PIPC Bus
Master IDE (rev 06)
00:12.0 Ethernet controller: VIA Technologies, Inc. VT6102 [Rhine-II] (rev 74)
xp2600 syslog:
Aug 8 16:55:18 srv-lr2600 kernel: nfs: server 10.4 OK
Aug 8 16:56:22 srv-lr2600 kernel: nfs: server 10.4 not responding, still
trying
Aug 8 16:56:22 srv-lr2600 kernel: nfs: server 10.4 OK
Aug 8 16:56:47 srv-lr2600 kernel: nfs: server fs not responding, still trying
Aug 8 16:56:47 srv-lr2600 kernel: nfs: server fs OK
Aug 8 16:56:50 srv-lr2600 kernel: nfs: server fs not responding, still trying
Aug 8 16:56:50 srv-lr2600 kernel: nfs: server fs OK
Aug 8 16:57:15 srv-lr2600 kernel: nfs: server 10.4 not responding, still
trying
Aug 8 16:57:15 srv-lr2600 kernel: nfs: server 10.4 OK
Aug 8 16:57:17 srv-lr2600 kernel: nfs: server 10.4 not responding, still
trying
Aug 8 16:57:17 srv-lr2600 kernel: nfs: server 10.4 OK
Aug 8 16:57:21 srv-lr2600 kernel: nfs: server 10.4 not responding, still
trying
Aug 8 16:57:21 srv-lr2600 kernel: nfs: server 10.4 OK
Aug 8 16:57:31 srv-lr2600 kernel: nfs: server 10.4 not responding, still
trying
Aug 8 16:57:31 srv-lr2600 kernel: nfs: server 10.4 not responding, still
trying
Aug 8 16:57:31 srv-lr2600 kernel: nfs: server 10.4 OK
Aug 8 16:57:31 srv-lr2600 kernel: nfs: server 10.4 OK
Aug 8 16:57:51 srv-lr2600 kernel: nfs: server fs not responding, still trying
Aug 8 16:57:51 srv-lr2600 kernel: nfs: server fs OK
Aug 8 16:58:42 srv-lr2600 kernel: nfs: server fs not responding, still trying
Aug 8 16:58:42 srv-lr2600 kernel: nfs: server fs OK
Aug 8 16:58:54 srv-lr2600 kernel: nfs: server 10.4 not responding, still
trying
Aug 8 16:58:54 srv-lr2600 kernel: nfs: server 10.4 OK
Aug 8 16:59:07 srv-lr2600 kernel: nfs: server 10.4 not responding, still
trying
Aug 8 16:59:07 srv-lr2600 kernel: nfs: server 10.4 OK
Aug 8 16:59:50 srv-lr2600 kernel: nfs: server fs not responding, still trying
Aug 8 16:59:50 srv-lr2600 kernel: nfs: server fs OK
Aug 8 16:59:51 srv-lr2600 kernel: nfs: server fs not responding, still trying
Aug 8 16:59:51 srv-lr2600 kernel: nfs: server fs OK
Aug 8 17:00:58 srv-lr2600 kernel: nfs: server fs not responding, still trying
Aug 8 17:00:58 srv-lr2600 kernel: nfs: server fs OK
Aug 8 17:01:22 srv-lr2600 kernel: nfs: server 10.4 not responding, still
trying
Aug 8 17:01:22 srv-lr2600 kernel: nfs: server 10.4 OK
Aug 8 17:01:26 srv-lr2600 kernel: NFS: giant filename in readdir (len
792b9a88)!
Aug 8 17:01:59 srv-lr2600 kernel: nfs: server 10.4 not responding, still
trying
Aug 8 17:01:59 srv-lr2600 last message repeated 2 times
Aug 8 17:01:59 srv-lr2600 kernel: nfs: server 10.4 OK
Aug 8 17:01:59 srv-lr2600 last message repeated 2 times
Aug 8 17:02:11 srv-lr2600 kernel: nfs: server fs not responding, still trying
Aug 8 17:02:11 srv-lr2600 kernel: nfs: server fs OK
vmstat:
0 2 0 7552 1478928 32724 0 0 5 4774 5856 1674 0 14 0 86
0 1 0 97024 1389524 32764 0 0 0 4735 6046 1734 0 14 0 85
0 1 0 8200 1478228 32860 0 0 0 5824 6110 1744 0 15 0 85
0 2 0 7040 1478104 34016 0 0 4 5292 6112 1770 0 15 0 85
0 2 0 192460 1284440 39600 0 0 0 6996 8789 3082 0 24 0 75
0 2 0 15092 1454596 45632 0 0 3 7851 9797 3301 0 28 0 72
0 3 0 53296 1407600 53768 0 0 3 8697 8521 3085 0 28 0 72
0 5 0 6696 1450928 56264 0 0 2 7903 10747 3853 0 32 0 68
0 7 0 7656 1451404 56308 0 0 1 9772 11195 4079 0 36 0 64
0 6 0 35712 1422992 56668 0 0 1 10428 11105 4117 0 37 0 63
0 5 0 7320 1446636 60580 0 0 47 9266 10373 3909 0 37 0 63
0 5 0 8068 1445088 61300 0 0 2 7959 10872 4021 0 33 0 67
0 5 0 6864 1446200 61076 0 0 0 9735 10713 3971 0 35 0 64
0 8 0 32300 1414148 66872 0 0 0 8824 11027 4200 0 36 0 64
0 7 0 5788 1436968 68768 0 0 0 9944 11230 4332 0 39 0 61
0 5 0 5784 1438292 67292 0 0 0 10764 11122 4184 0 39 0 61
0 5 0 6360 1434940 70472 0 0 5 11552 10707 4052 0 38 2 59
0 5 0 7532 1434312 70156 0 0 0 7860 11021 4111 0 35 0 65
0 4 0 7528 1435300 69288 0 0 5 10776 11244 4242 0 39 0 61
0 6 0 7080 1436852 68300 0 0 0 9002 11108 4143 0 36 0 64
0 3 0 7176 1439128 65928 0 0 0 11320 11264 4228 0 39 0 60
0 5 0 58536 1388564 65304 0 0 2 8709 10707 3988 0 35 0 64
0 3 0 31676 1416804 63980 0 0 0 9620 11173 4110 0 37 0 63
0 5 0 6360 1443248 62880 0 0 0 11070 11306 4156 0 39 0 61
0 5 0 6936 1443752 62012 0 0 0 8965 11151 4124 0 36 0 64
0 4 0 7528 1445472 59796 0 0 0 11077 11293 4196 0 39 0 61
0 5 0 15852 1434576 62800 0 0 0 10615 11199 4158 0 38 0 62
0 6 0 7224 1442660 63412 0 0 0 9675 11185 4127 0 37 0 63
0 5 0 7144 1444804 61592 0 0 0 9662 11146 4129 0 37 0 63
0 4 0 7580 1446844 59108 0 0 0 10283 11309 4144 0 38 0 62
0 6 0 4972 1452864 56176 0 0 1 10481 11256 4159 0 38 0 62
0 5 0 6424 1454372 53964 0 0 0 10686 11190 4066 0 37 0 63
0 6 0 5324 1459288 50328 0 0 0 9117 11137 4063 0 35 0 65
0 5 0 6412 1460396 48228 0 0 0 10657 11244 4159 0 38 0 62
0 6 0 5004 1463616 46492 0 0 6 9593 11127 4095 0 36 0 64
0 4 0 5132 1463024 46496 0 0 0 10694 11166 4133 0 38 0 62
[killing linux-mm in CC]
On Fri, Aug 08, 2003 at 05:28:17PM -0700, Mike Fedyk wrote:
> Though it was interesting that there weren't any errors reported from the
> 2.6 nfs server.
Ok, the 2.6 client is having trouble with the 2.6 server also, so that
wasn't anything special then.
I have some tcpdump traces from one of the nfs servers that shows the pauses
that are happening in the transmission if anyone would like to take a look.
I'm not sure what else I can do to find out why it is pausing.
Any ideas?