The latest version of the -mjc branch has been released. It is
available at:
ftp://ftp.kernel.org/pub/linux/kernel/people/mjc/linux-2.4/patch-2.4.18-pre8-mjc.bz2
This latest release is an immense performance enhancement. :)
--------------------------------------------------------------
(I lost a bunch of my original patches with credits for some of the code
listed here. This is a complete list, however, of all the code in this
release of the -mjc branch.)
2.4.18-pre7-ac3 (Alan Cox et al)
Reverse Mapping VM Patch #12c (Rik van Riel)
No Buffer Wait Patch (RvR)
O(1) Scheduler K2 (Ingo Molnar)
IRQrate (IM)
Preemptible Kernel Patch (Robert M. Love)
Break Selected Locks (RML)
Fibonacci Hashing: (prime for 32-bit) (William Lee Irwin III)
-Inode hash (WLI)
-Page cache hash (WLI)
-UID hash (dynamic) (WLI)
-PID hash (dynamic) (WLI)
OOM trigger increase (WLI)
Change VM policy to overcommit (WLI)
Page aging adjustments (WLI)
Swap-storm prevention (hysterisis) (WLI)
Netfilter Header Order Fix (WLI)
proc_pid_statm() cleanup (WLI)
task_nice export (WLI)
Bootmem Rewrite (WLI)
Mountcache size reduction (WLI, Andi Kleen, Al Viro)
Slab search optimization (Momchil Velikov)
Radix tree page cache search (MV)
Cross Compile fix on GCC3 (MV)
Volatile xtime (MV)
UserMode Linux (Jeff Dike)
Loop Mounting Deadlock Fix (Andrew Morton)
Read Latency Fix #2 (AKPM)
slab cache estimate optimization (Balbir Singh)
x86 Syscall stall fix (Alexander Khripin)
Pollselect speedup (Manfred Spraul)
Bdflush parameters (Stephen van der Krantz)
ECC Memory (linux-ecc)
-config option et al (me)
Nanosleep (?)
mmap -ENOMEM fix (?)
x86 fast pte (?)
cache flush update (?)
TCP Connection Tracking update (?)
AMD Elan Updates (Robert Schwiebel)
binfmt_elf checks (?)
3.5GB Address Space (?)
SiS5513 updates (?)
RivaFB Software Blanking (?)
Rage128 Update (?)
lm_sensors (lm_sensors team)
AMD vcool patch (Daniel Nofftz)
ia64 DMA update (?)
proc-file-read documentation (Thomas Hood)
------
Michael J. Cohen
OhDarn.net
Hi Michael,
Your patch list is very nice. Is there any chance to add XFS patch, also?
If so, it would be wonderful for me and the others , I think, who uses XFS
and want to use these patches, too. Because, if i patch my kernel with XFS,
i can't add other patches, like yours.
thnx
----- Original Message -----
From: Michael Cohen <[email protected]>
To: <[email protected]>
Sent: Thursday, February 07, 2002 11:17 AM
Subject: [ANNOUNCE] 2.4.18-pre8-mjc
> The latest version of the -mjc branch has been released. It is
> available at:
>
ftp://ftp.kernel.org/pub/linux/kernel/people/mjc/linux-2.4/patch-2.4.18-pre8
-mjc.bz2
>
On 2/7/02 5:03 AM, "Tarkan Erimer" <[email protected]> wrote:
> Your patch list is very nice. Is there any chance to add XFS patch, also?
Shawn Starr has been working on this.
> If so, it would be wonderful for me and the others , I think, who uses XFS
> and want to use these patches, too. Because, if i patch my kernel with XFS,
> i can't add other patches, like yours.
Personally I think XFS ought to stay in fs/xfs already and stop encroaching
on the rest of the kernel. It's also HUGE. But hey, if Shawn wants to do
it, let him do it.
------
Michael Cohen
On Thu, Feb 07, 2002 at 04:17:54AM -0500, Michael Cohen wrote:
> SiS5513 updates (?)
Lionel Bouton <[email protected]>
Which patch did you take in ?
Post 20020110_1 patches add support to old chipsets (pre-ATA66) but are known to
have buggy chipset detection code.
I've just received a new ECS K7S5AL (old one broken) so I'll be able to test the code myself
again (currently memtest86 testing).
Expect debugged code next week.
Full patch history available at:
http://inet6.dyn.dhs.org/sponsoring/sis5513/index.html
LB.
> proc-file-read documentation (Thomas Hood)
Please note that the patch you included is obsolete.
(It was a first draft, and contained a couple of errors.)
Get the latest version of the patch via the link on this
page:
http://panopticon.csustan.edu/thood/pnpbios.html
Thanks.
Thomas
On 07 Feb 2002 04:17:54 -0500
Michael Cohen <[email protected]> wrote:
| Volatile xtime (MV)
The change in include/sched.h seems to miss its counter-part in kernel/timer.c, hence a type mismatch leading to a compilation failure.
--- linux/kernel/timer.c Thu Feb 7 23:44:00 2002
+++ linux/kernel/timer.c~ Thu Feb 7 23:27:47 2002
@@ -34,7 +34,7 @@
long tick = (1000000 + HZ/2) / HZ; /* timer interrupt period */
/* The current time */
-struct timeval xtime __attribute__ ((aligned (16)));
+volatile struct timeval xtime __attribute__ ((aligned (16)));
/* Don't completely fail for HZ > 500. */
int tickadj = 500/HZ ? : 1; /* microsecs */
Regards,
Thibaut
I got some miscellaneous errors/warnings upon compiling (this is with an
SMP build, .config is available if needed). I don't need these modules
(I was just doing my standard "compile as much as possible" routine), so
I just disabled them and continued on my way.
gcc -D__KERNEL__ -I/home/bstrand/src/linux-2.4.17/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2
-fomit-frame-pointeri2c-tsunami.c:35: asm/hwrpb.h: No such file or directory
i2c-tsunami.c:36: asm/core_tsunami.h: No such file or directory
gcc -D__KERNEL__ -I/home/bstrand/src/linux-2.4.17/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointergcc
-D__KERNEL__ -I/home/bstrand/src/linux-2.4.17/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2
-fomit-frame-pointeri2c-tsunami.c: In function `writempd':
i2c-tsunami.c:72: `TSUNAMI_cchip' undeclared (first use in this function)
i2c-tsunami.c:72: (Each undeclared identifier is reported only once
i2c-tsunami.c:72: for each function it appears in.)
i2c-tsunami.c: In function `readmpd':
i2c-tsunami.c:78: `TSUNAMI_cchip' undeclared (first use in this function)
i2c-tsunami.c:79: warning: control reaches end of non-void function
i2c-tsunami.c: In function `i2c_tsunami_init':
i2c-tsunami.c:156: `hwrpb' undeclared (first use in this function)
i2c-tsunami.c:156: `ST_DEC_TSUNAMI' undeclared (first use in this function)
i2c-tsunami.c:160: `TSUNAMI_cchip' undeclared (first use in this function)
gcc -D__KERNEL__ -I/home/bstrand/src/linux-2.4.17/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointerr128_cce.c:
In function `r128_cce_packet':
r128_cce.c:1023: warning: unused variable `size'
r128_cce.c:1021: warning: unused variable `buffer'
r128_cce.c:1019: warning: unused variable `dev_priv'
gcc -D__KERNEL__ -I/home/bstrand/src/linux-2.4.17/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointeri810_dma.c:
In function `i810_free_page':
i810_dma.c:296: warning: passing arg 1 of `wake_up_page_Rsmp_4acd422b'
makes pointer from integer without a cast
Regards,
Brian
Replying to Michael Cohen:
> The latest version of the -mjc branch has been released. It is
> available at:
> ftp://ftp.kernel.org/pub/linux/kernel/people/mjc/linux-2.4/patch-2.4.18-pre8-mjc.bz2
Excellent!
I tried to build up some by myself, but you appeared :)))
here is small addon to your patch - some netfilter modules, route_me_harder
fix, and tulip vlan patch
ftp://stingr.net/pub/linux/patch-2.4.18-pre8-mjc-s0.gz
--
Paul P 'Stingray' Komkoff 'Greatest' Jr // (icq)23200764 // (irc)Spacebar
PPKJ1-RIPE // (smtp)[email protected] // (http)stingr.net // (pgp)0xA4B4ECA4
On Thu, 7 Feb 2002, Michael Cohen wrote:
> On 2/7/02 5:03 AM, "Tarkan Erimer" <[email protected]> wrote:
>
> > Your patch list is very nice. Is there any chance to add XFS patch, also?
>
> Shawn Starr has been working on this.
>
> > If so, it would be wonderful for me and the others , I think, who uses XFS
> > and want to use these patches, too. Because, if i patch my kernel with XFS,
> > i can't add other patches, like yours.
>
> Personally I think XFS ought to stay in fs/xfs already and stop encroaching
> on the rest of the kernel. It's also HUGE. But hey, if Shawn wants to do
> it, let him do it.
The size is less an issue than having hooks all over the kernel, where
other features can/will break and be broken.
--
bill davidsen <[email protected]>
CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.