The LDT fixes in particular fix some potentially random strange behaviour.
And the alpha memmove() thing was a showstopper bug on alphas.
Linus
----
- pre7:
- x86 LDT handling fixes: revert some cleanups (the LDT really
doesn't act like a TLB context)
- Richard Henderson: alpha update (working memmove() from Ivan
Kokshaysky etc)
- Manfred: winbond-840.c net driver update (fix oops on module unload etc)
- Alan Cox: more synchronizations (with some fixes from Andrew Morton)
- pre6:
- Marc Joosen: BIOS int15/e820 memory query: don't assume %edx
unchanged by the BIOS. Fixes at least some IBM ThinkPads.
- Alan Cox: synchronize
- Marcelo Tosatti & me: properly sync dirty pages
- Andreas Dilger: proper ext2 compat flag checking
- pre5:
- NIIBE Yutaka: SuperH update
- Geert Uytterhoeven: m68k update
- David Miller: TCP RTO calc fix, UDP multicast fix etc
- Duncan Laurie: ServerWorks PIRQ routing definition.
- mm PageDirty cleanups, added sanity checks, and don't lose the bit.
- pre4:
- Christoph Rohland: shmfs cleanup
- Nicolas Pitre: don't forget loop.c flags
- Geert Uytterhoeven: new-style m68k Makefiles
- Neil Brown: knfsd cleanups, raid5 re-org
- Andrea Arkangeli: update to LVM-0.9
- LC Chang: sis900 driver doc update
- David Miller: netfilter oops fix
- Andrew Grover: acpi update
- pre3:
- Christian Jullien: smc9194: proper dev_kfree_skb_irq
- Cort Dougan: new-style PowerPC Makefiles
- Andrew Morton, Petr Vandrovec: fix run_task_queue
- Christoph Rohland: shmfs for shared memory handling
- pre2:
- Kai Germaschewski: ISDN update (including Makefiles)
- Jens Axboe: cdrom updates
- Petr Vandrovec; Matrox G450 support
- Bill Nottingham: fix FAT32 filesystems on 64-bit platforms
- David Miller: sparc (and other) Makefile fixup
- Andrea Arkangeli: alpha SMP TLB context fix (and cleanups)
- Niels Kristian Bech Jensen: checkconfig, USB warnings
- Andrew Grover: large ACPI update
- pre1:
- me: drop support for old-style Makefiles entirely. Big.
- me: check b_end_io at the IO submission path
- me: fix "ptep_mkdirty()" (so that swapoff() works correctly)
- fix fault case in copy_from_user() with a constant size, where
((size & 3) == 3)
>The LDT fixes in particular fix some potentially random strange behaviour.
>And the alpha memmove() thing was a showstopper bug on alphas.
And the network lockup bug...
--Ray
It looks like 2.4.0-test13-pre7 is a clear winner when running dbench 48
on my somewhat slow test machine (450 Mhz P-III, 192MB, IDE).
2.2.18 3.53307 MB/sec (NB=4.41633 MB/sec 35.3307 MBit/sec)
2.2.19-pre3 3.81213 MB/sec (NB=4.76516 MB/sec 38.1213 MBit/sec)
2.4.0-test13-pre5 4.06823 MB/sec (NB=5.08529 MB/sec 40.6823 MBit/sec)
2.4.0-test13-pre6 4.11353 MB/sec (NB=5.14192 MB/sec 41.1353 MBit/sec)
2.4.0-test13pre4-ac2 4.47376 MB/sec (NB=5.5922 MB/sec 44.7376 MBit/sec)
2.4.0-test13-pre7 6.3723 MB/sec (NB=7.96538 MB/sec 63.723 MBit/sec)
The tests were done under identical conditions, after fresh boot-up,
running KDE 2.0, one xterm, and xosview.
Here are a few selected lines from dmesg to put things in perspective.
Detected 448.810 MHz processor.
Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
hda: ST317221A, ATA DISK drive
Using r5 hash to sort names
reiserfs: using 3.5.x disk format
ReiserFS version 3.6.23
This is really looking great.
Steven
On Sat, 30 Dec 2000, Steven Cole wrote:
>
> It looks like 2.4.0-test13-pre7 is a clear winner when running dbench 48
> on my somewhat slow test machine (450 Mhz P-III, 192MB, IDE).
This is almost certainly purely due to changing (some would say "fixing")
the bdflush synchronous wait point.
No actual code was harmed in the making of this improvement.
But I will take full credit anyway.
Linus
Linus Torvalds wrote:
> On Sat, 30 Dec 2000, Steven Cole wrote:
> >
> > It looks like 2.4.0-test13-pre7 is a clear winner when running dbench 48
> > on my somewhat slow test machine (450 Mhz P-III, 192MB, IDE).
>
> This is almost certainly purely due to changing (some would say "fixing")
> the bdflush synchronous wait point.
>
After evaluating test13-pre7 with the quake 3 arena test,
I think it's even snappier than the previous champ, which
was test10 + low latency patches..
A most auspicious trend, if I might make so bold as
to state it in this forum.
jjs
On Sat, 30 Dec 2000, Linus Torvalds wrote:
> On Sat, 30 Dec 2000, Steven Cole wrote:
> >
> > It looks like 2.4.0-test13-pre7 is a clear winner when running dbench 48
> > on my somewhat slow test machine (450 Mhz P-III, 192MB, IDE).
>
> This is almost certainly purely due to changing (some would say "fixing")
> the bdflush synchronous wait point.
Nice:)
Did Rik's drop_behind performance fix make it in or can we look forward to
another jump in the dbench benchmarks?
Cheers
Chris
Chris Evans wrote:
> On Sat, 30 Dec 2000, Linus Torvalds wrote:
>
> > On Sat, 30 Dec 2000, Steven Cole wrote:
> > >
> > > It looks like 2.4.0-test13-pre7 is a clear winner when running dbench
> > > 48 on my somewhat slow test machine (450 Mhz P-III, 192MB, IDE).
> >
> > This is almost certainly purely due to changing (some would say
> > "fixing") the bdflush synchronous wait point.
>
> Nice:)
>
> Did Rik's drop_behind performance fix make it in or can we look forward to
> another jump in the dbench benchmarks?
And please do not forget marcello's swap clustering patch. I get a 13%
improvement on dbench with reiserfs when patched with it. From
conversations on kernelnewbies, Riel likes this one too.
Ed Tomlinson