2.5 merge candidate list, final version. (End of "crunch time" series.)

2.5 merge candidate list, final version. (End of "crunch time" series.)

Post by Rob Landle » Tue, 29 Oct 2002 18:40:11



Hi Linus.

This list is the result of a week of scouring linux-kernel and posting
more or less daily versions soliciting feedback from everybody seriously
trying to get a patch into 2.5.  This is the ninth and final posting of
this list.

Previous versions, and the discussion they spawned, were here:
1.0) http://lists.insecure.org/lists/linux-kernel/2002/Oct/7006.html
1.1) http://lists.insecure.org/lists/linux-kernel/2002/Oct/7051.html
1.2) http://lists.insecure.org/lists/linux-kernel/2002/Oct/7363.html
1.3) http://lists.insecure.org/lists/linux-kernel/2002/Oct/7452.html
1.4) http://lists.insecure.org/lists/linux-kernel/2002/Oct/7827.html
1.5) http://lists.insecure.org/lists/linux-kernel/2002/Oct/8174.html
1.6) http://lists.insecure.org/lists/linux-kernel/2002/Oct/8787.html
1.7) http://lists.insecure.org/lists/linux-kernel/2002/Oct/9130.html

The vast majority of the 30 items in this list will probably be
dropped, but if this list results in explicit rejections instead of patches
getting missed or lost in the shuffle, it will have served its purpose.

On a side note, Dave Jones has started a 2.5 series development summary:
http://www.codemonkey.org.uk/post-halloween-2.5.txt

================================= Intro ====================================

The following features aim to be ready for submission to Linus by Monday,
October 28th, to be considered for inclusion (in 2.5.45) before the feature
freeze on Thursday, October 31 (halloween).

This list is just pending features trying to get in before feature freeze.
It's primarily for features that need more testing, or might otherwise get
forgotten in the rush.  If you want to know what's already gone in, or what's
being worked on for the next development cycle, check out:
http://kernelnewbies.org/status

Thanks to Rusty Russell and Guillaume Boissiere, whose respective 2.5 merge
candidate lists have been ruthlessly strip-mined in the process of
assembling this.  And to everybody who's emailed stuff.

============================ Pending features: =============================

1) Andrew Morton's -mm tree. (Andrew Morton, editor.)

Andrew Morton's -mm tree collates several other projects, including:

The ext2/ext3 Extended Attributes and Access Control Lists patch from Ted Tso
and Andreas Gruenbacher (ext23-*.patch), Page Table Sharing from Daniel
Phillips and Dave McCracken (shpte-ng.patch), a bunch of huge page upgrades
from Bill Irwin (hugetlb*.patch), the orlov allocator, Ingo's generic
nonlinear mappings...

Stuff.  Lots of stuff.

You can get Andrew Morton's MM tree from the following URL, including a
broken-out patches directory and a description file.  (The latest version
as of this writing is -mm6.)

http://www.zip.com.au/~akpm/linux/patches/2.5/2.5.44

----------------------------------------------------------------------------

2) Device mapper for Logical Volume Manager (LVM2)  (LVM2 team)  (in -ac)

Announce:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103536883428443&w=2

Download:
http://people.sistina.com/~thornber/patches/2.5-stable/

Home page:
http://www.sistina.com/products_lvm.htm

Note: this is in the 2.5-ac tree, available at:
http://www.kernel.org/pub/linux/kernel/people/alan/

----------------------------------------------------------------------------

3) EVMS (Enterprise Volume Management System) (IBM, Contact: Kevin Corry)

Fighting with LVM2 for a place in the tree, a bigger solution to a bigger
set of problems:

Home page:
http://sourceforge.net/projects/evms

Home page:
http://evms.sourceforge.net

Download:
http://evms.sourceforge.net/patches/

Some related discussions:
http://marc.theaimsgroup.com/?t=103359686900003&r=1&w=2
http://marc.theaimsgroup.com/?t=103439913000001&r=1&w=2
http://marc.theaimsgroup.com/?w=2&r=1&s=%5Bpatch%5D+evms+core&q=t

----------------------------------------------------------------------------

4) New kernel configuration system (Roman Zippel)

Announcement:
http://lists.insecure.org/lists/linux-kernel/2002/Oct/9043.html

Download:
http://www.xs4all.nl/~zippel/lc/

Linus has actually looked fairly favorably on this one so far:
http://lists.insecure.org/lists/linux-kernel/2002/Oct/3250.html

And an AOL for it:

http://lists.insecure.org/lists/linux-kernel/2002/Oct/8255.html

----------------------------------------------------------------------------

5) Linux Trace Toolkit (LTT) (Karim Yaghmour)

Announce:
http://lists.insecure.org/lists/linux-kernel/2002/Oct/7016.html

Download:
http://opersys.com/ftp/pub/LTT/ExtraPatches/patch-ltt-linux-2.5.44-va...

User tools:
http://opersys.com/ftp/pub/LTT/TraceToolkit-0.9.6pre2.tgz

----------------------------------------------------------------------------

6) Kernel Probes (IBM, contact: Vamsi Krishna S)

Kprobes announcement:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103528410215211&w=2

Download Base Kprobes Patch:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103528425615302&q=raw

KProbes->DProbes support patches:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103528454215523&q=raw
http://marc.theaimsgroup.com/?l=linux-kernel&m=103528454015520&q=raw
http://marc.theaimsgroup.com/?l=linux-kernel&m=103528485415813&q=raw

Download gzipped tarball patches from official IBM site:
http://www-124.ibm.com/linux/patches/?project_id=141

KProbes home page:
http://www-124.ibm.com/linux/projects/kprobes

A good explanation of the difference between kprobes, dprobes,
and kernel hooks:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103532874900445&w=2

Clarification: just kprobes is being submitted for 2.5.45, (and
optionally some basic dprobes support,) but not the whole of dprobes:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103536827928012&w=2
And also:
http://lists.insecure.org/lists/linux-kernel/2002/Oct/9160.html

----------------------------------------------------------------------------

7) Posix support and High Resolution timers (George Anzinger)

George has a patch that provides posix support, and on top of that a
patch to provide high resolution timers.  He talks about it here:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103562196700924&w=2

The project's home page is here:
http://high-res-timers.sourceforge.net/

Downloadable patches may be found here:
http://sourceforge.net/project/showfiles.php?group_id=20460&release_i...

And descriptions for each patch (in the order they should be applied) are
available here (although in this case the archive truncates the patches
themselves, use the download link above):
http://marc.theaimsgroup.com/?l=linux-kernel&m=103553654329827&w=2
http://marc.theaimsgroup.com/?l=linux-kernel&m=103557676007653&w=2
http://marc.theaimsgroup.com/?l=linux-kernel&m=103557677207693&w=2
http://marc.theaimsgroup.com/?l=linux-kernel&m=103558349714128&w=2

Linus had concerns with this one a while back:
http://lists.insecure.org/lists/linux-kernel/2002/Oct/3463.html

----------------------------------------------------------------------------

8) Posix timers alternate implementation (Jim Houston)

Jim Houston has an alternate patch to provide posix support (but not
high resolution timers on top of it, yet).

http://marc.theaimsgroup.com/?l=linux-kernel&m=103549000027416&q=raw

----------------------------------------------------------------------------

9) Linux Kernel Crash Dumps (Matt Robinson, LKCD team)

Announce:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103553563728914&w=2
And again:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103536576625905&w=2

Download:
http://lkcd.sourceforge.net/download/latest/

----------------------------------------------------------------------------

10) Rewrite of the console layer (James Simmons)

Announcement:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103487329526903&w=2

Home page:
http://linuxconsole.sourceforge.net/

Downloadable patch:
http://phoenix.infradead.org/~jsimmons/fbdev.diff.gz

Bitkeeper tree:
bk://fbdev.bkbits.net/fbdev-2.5

----------------------------------------------------------------------------

11) IPv6 upgrades and crypto API (Yoshifuji Hideyaki)

The Usagi ipv6 upgrades have been available for a while, and their
author would like to see them in 2.5:

README:
ftp://ftp.linux-ipv6.org/pub/usagi/patch/ipsec/README.IPSEC

Downloadable patch here:
ftp://ftp.linux-ipv6.org/pub/usagi/patch/ipsec/ipsec-2.5.43-ALL-03.pa...

Dave Miller is doing a major ipv6 layer rewrite, but no patch has been sent
to the list yet.  Ironically, James Morris has a Crypto API patch on top of
Dave Miller's tree:

Announce:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103559983324080&w=2

Download:
http://samba.org/~jamesm/crypto/

----------------------------------------------------------------------------

12) MMU-less processor support (Greg Ungerer)

Most recent announcement (with links):
http://marc.theaimsgroup.com/?l=linux-kernel&m=103578189421588&w=2

A version of this is in the 2.5-ac tree.  An announcement of a patch against
that is here:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103578338922236&w=2

----------------------------------------------------------------------------

13) sys_epoll (I.E. /dev/poll) (Davide Libenzi)

Announce:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103542994232004&w=2

Homepage:
http://www.xmailserver.org/linux-patches/nio-improve.html

Download:
http://www.xmailserver.org/linux-patches/sys_epoll-2.5.44-last.diff

Linus participated repeatedly in a thread on this one too, expressing
concerns which (hopefully) have been addressed.  See:
http://lists.insecure.org/lists/linux-kernel/2002/Oct/6428.html

----------------------------------------------------------------------------

14) CD Recording/sgio patches (Jens Axboe)

Announce:
http://lists.insecure.org/lists/linux-kernel/2002/Oct/8060.html

Download patch:
http://www.kernel.org/pub/linux/kernel/people/axboe/patches/v2.5/2.5....

This should be in Alan Cox's tree as of 2.4.44-ac4.
...

read more »

 
 
 

2.5 merge candidate list, final version. (End of "crunch time" series.)

Post by Nikita Danilo » Tue, 29 Oct 2002 19:00:14


Quote:Rob Landley writes:

 > Hi Linus.
 >
 > This list is the result of a week of scouring linux-kernel and posting
 > more or less daily versions soliciting feedback from everybody seriously
 > trying to get a patch into 2.5.  This is the ninth and final posting of
 > this list.
 >
 > Previous versions, and the discussion they spawned, were here:
 > 1.0) http://lists.insecure.org/lists/linux-kernel/2002/Oct/7006.html
 > 1.1) http://lists.insecure.org/lists/linux-kernel/2002/Oct/7051.html
 > 1.2) http://lists.insecure.org/lists/linux-kernel/2002/Oct/7363.html
 > 1.3) http://lists.insecure.org/lists/linux-kernel/2002/Oct/7452.html
 > 1.4) http://lists.insecure.org/lists/linux-kernel/2002/Oct/7827.html
 > 1.5) http://lists.insecure.org/lists/linux-kernel/2002/Oct/8174.html

[...]

 > http://www.kernel.org/pub/linux/kernel/people/rusty/patches/Hotcpu/ho...
 >
 > ----------------------------------------------------------------------------
 >
 > 30) Reiser4.
 >
 > I don't have a patch yet, but Hans Reiser is very insistent that this
 > will be ready by halloween.  (VERY insistent.)  I'll let him speak for
 > himself:
 > http://lists.insecure.org/lists/linux-kernel/2002/Oct/8793.html
 >
 > And again (promises, promises):
 > http://lists.insecure.org/lists/linux-kernel/2002/Oct/9082.html
 >
 > Still no patch at the time of this writing, though.  In theory it
 > should show up here:
 > http://namesys.com/download.html

Snapshot is available at http://www.namesys.com/snapshots/:

reiser4 proper:
http://www.namesys.com/snapshots/reiser4-2002.10.24.tar.gz

necessary changes to the core kernel, plus some UML patches, plus some
patches for debugging:
http://www.namesys.com/snapshots/reiser4-core-2002.10.24.diff

utils:
http://www.namesys.com/snapshots/reiser4progs-2002.10.24.tar.gz

I shall produce newer snapshot to-morrow.

It does work, but code is not production quality yet. Do *NOT* put
anything close to critical data on it.

 >
 > Or perhaps here:
 > ftp://ftp.lugoj.org/pub/reiserfs/devlinux.com/pub/namesys/reiserfs-fo...
 >
 > In the meantime, all I can find on Reiser4 is some kind of hybrid
 > marketing brochure/design document thing:
 > http://www.reiserfs.org/v4/v4.html
 >
 > Did I mention Hans was insistent?  The man can make puppy eyes through email.
 > It's quite impressive.

Works without email too. :-)

 > --
 > http://penguicon.sf.net - Terry Pratchett, Eric Raymond, Pete Abrams, Illiad,
 > CmdrTaco, liquid nitrogen ice cream, and caffienated jello.  Well why not?

Nikita.

 > -
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

2.5 merge candidate list, final version. (End of "crunch time" series.)

Post by Tomas Szep » Wed, 30 Oct 2002 14:40:05


Quote:> Snapshot is available at http://www.namesys.com/snapshots/:

> http://www.namesys.com/snapshots/reiser4-2002.10.24.tar.gz
> http://www.namesys.com/snapshots/reiser4-core-2002.10.24.diff

2.5.44:
Producing a module I get a couple unresolved symbols,
and trying to build directly into the kernel results in

...
ld -m elf_i386 -e stext -T arch/i386/vmlinux.lds.s arch/i386/kernel/head.o arch/i386/kernel/init_task.o  init/built-in.o --start-group  arch/i386/kernel/built-in.o  arch/i386/mm/built-in.o  arch/i386/mach-generic/built-in.o  kernel/built-in.o  mm/built-in.o  fs/built-in.o  ipc/built-in.o  security/built-in.o  lib/lib.a  arch/i386/lib/lib.a  drivers/built-in.o  sound/built-in.o  arch/i386/pci/built-in.o  net/built-in.o --end-group  -o vmlinux
fs/built-in.o(.init.text+0x1381): In function `init_reiser4':
: undefined reference to `local symbols in discarded section .exit.text'
make[1]: *** [vmlinux] Error 1
make: *** [vmlinux] Error 2

Quote:> http://www.namesys.com/snapshots/reiser4progs-2002.10.24.tar.gz

... is missing the "configure" script.
Generating one with autoconf 2.53 doesn't seem to work, either.

T.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

2.5 merge candidate list, final version. (End of "crunch time" series.)

Post by Hans Reise » Thu, 31 Oct 2002 10:30:11


We are going to submit a patch appropriate for inclusion as an
experimental FS on Halloween.   I hope you will forgive our pushing the
limit timewise, it is not by choice, but the algorithms we used to more
than double reiserfs V3 performance were, quite frankly, hard to code.

A description of it can be found at www.namesys.com/v4/fast_reiser4.html.

I am going to ask persons to volunteer to produce independent
benchmarks.  I am told by Oleg and Zam that it is now faster than in the
benchmarks in the above draft paper, and I hope to see this verified by
persons outside Namesys.

I hope you will also find our atomic transactions infrastructure
interesting.  It allows us to guarantee each system call to reiserfs is
fully atomic, and in later versions we hope to use it to make atomicity
available to user space so that multiple fs operations can be guaranteed
atomic.  This will hopefully help to eliminate a major class of security
holes, as well as provide significantly better protection against user
data corruption by crashes than any current filesystem.  As you can see,
it does not seem to substantially harm performance, or at least we
assume that our more than doubling performance means that performance
was not harmed.;-)

Finally, and I think in the longterm most importantly, we have
implemented a full plugin infrastructure for reiser4.  This will make it
possible for us to implement database and search engine functionality
one plugin at a time.  (See www.namesys.com/whitepaper.html for a
description of those semantics.)  We hope you will find the semantics
much more fun than tacking SQL onto the FS like OFS is going to do.;-)

We don't know how fast OFS will be, but we hope you will enjoy watching
us raise the bar a few feet.  We are very e*d by more than doubling
performance.  The read performance improvements are more than I expected.  

(If any of you attempt to replicate the benchmarks, please be sure to
use reiser4 readdir order for writes to reiser4 (that means don't use
tarballs made from ext2), and to use the latest hard drives and fast
processors with udma 5 turned on.  We are quite sensitive to transfer
speed since we do a good job of avoiding seeks.  We are sensitive to
readdir order because we sort directory entries (which is necessary for
having efficient large directory lookups)).   In reiser4.1 we will ship
a repacker, and then it won't matter what order you do writes in so long
as the repacker gets a chance to run at night.

Cheers,

Hans


>Rob Landley writes:
> > Hi Linus.

> > This list is the result of a week of scouring linux-kernel and posting
> > more or less daily versions soliciting feedback from everybody seriously
> > trying to get a patch into 2.5.  This is the ninth and final posting of
> > this list.

> > Previous versions, and the discussion they spawned, were here:
> > 1.0) http://www.veryComputer.com/
> > 1.1) http://www.veryComputer.com/
> > 1.2) http://www.veryComputer.com/
> > 1.3) http://www.veryComputer.com/
> > 1.4) http://www.veryComputer.com/
> > 1.5) http://www.veryComputer.com/

>[...]

> > http://www.veryComputer.com/

> > ----------------------------------------------------------------------------

> > 30) Reiser4.

> > I don't have a patch yet, but Hans Reiser is very insistent that this
> > will be ready by halloween.  (VERY insistent.)  I'll let him speak for
> > himself:
> > http://www.veryComputer.com/

> > And again (promises, promises):
> > http://www.veryComputer.com/

> > Still no patch at the time of this writing, though.  In theory it
> > should show up here:
> > http://www.veryComputer.com/

>Snapshot is available at http://www.veryComputer.com/:

>reiser4 proper:
>http://www.veryComputer.com/

>necessary changes to the core kernel, plus some UML patches, plus some
>patches for debugging:
>http://www.veryComputer.com/

>utils:
>http://www.veryComputer.com/

>I shall produce newer snapshot to-morrow.

>It does work, but code is not production quality yet. Do *NOT* put
>anything close to critical data on it.

> > Or perhaps here:
> > ftp://ftp.lugoj.org/pub/reiserfs/devlinux.com/pub/namesys/reiserfs-fo...

> > In the meantime, all I can find on Reiser4 is some kind of hybrid
> > marketing brochure/design document thing:
> > http://www.veryComputer.com/

> > Did I mention Hans was insistent?  The man can make puppy eyes through email.
> > It's quite impressive.

>Works without email too. :-)

> > --
> > http://www.veryComputer.com/ - Terry Pratchett, Eric Raymond, Pete Abrams, Illiad,
> > CmdrTaco, liquid nitrogen ice cream, and caffienated jello.  Well why not?

>Nikita.

> > -
>-
>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

>More majordomo info at  http://www.veryComputer.com/
>Please read the FAQ at  http://www.veryComputer.com/

--
Hans

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://www.veryComputer.com/
Please read the FAQ at  http://www.veryComputer.com/

 
 
 

2.5 merge candidate list, final version. (End of "crunch time" series.)

Post by Jeff Garzi » Thu, 31 Oct 2002 10:50:12



> We are going to submit a patch appropriate for inclusion as an
> experimental FS on Halloween.   I hope you will forgive our pushing
> the limit timewise, it is not by choice, but the algorithms we used to
> more than double reiserfs V3 performance were, quite frankly, hard to
> code.

Does your merge change the core code at all?  Does it add new syscalls?

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

2.5 merge candidate list, final version. (End of "crunch time" series.)

Post by Andrew Morto » Thu, 31 Oct 2002 11:40:12




> > We are going to submit a patch appropriate for inclusion as an
> > experimental FS on Halloween.   I hope you will forgive our pushing
> > the limit timewise, it is not by choice, but the algorithms we used to
> > more than double reiserfs V3 performance were, quite frankly, hard to
> > code.

> Does your merge change the core code at all?  Does it add new syscalls?

Their changes are tiny, and sensible.  See
http://www.namesys.com/snapshots/2002.10.29/

But I'd like to ask about the status of reiser3 support.

Chris had patches *ages* ago to convert it to use direct-to-BIO for
reads, and writes should be done as well.  reiserfs3 is still using
buffer-head-based IO for bulk reads and writes.  That's a 25-30% hit
in CPU cost, and all the old ZONE_NORMAL-full-of-buffer_heads
problems.

Any plans to get that work finished off?
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

2.5 merge candidate list, final version. (End of "crunch time" series.)

Post by Nikita Danilo » Thu, 31 Oct 2002 11:40:13


Quote:Jeff Garzik writes:


 >
 > > We are going to submit a patch appropriate for inclusion as an
 > > experimental FS on Halloween.   I hope you will forgive our pushing
 > > the limit timewise, it is not by choice, but the algorithms we used to
 > > more than double reiserfs V3 performance were, quite frankly, hard to
 > > code.
 >
 >
 >
 > Does your merge change the core code at all?  Does it add new syscalls?
 >

Here is a list of things reiser4 needs from the core:

 (1) export of generic_forget_inode()

 (2) export of page_cache_readahead()

 (3) export of remove_from_page_cache()

 (4) export of fsync_super()

 (5) patch to allow safe sharing of ->journal_info pointer in task_struct
     by several users

(1)-(4) are only necessary to compile reiser4 as module.

(5) is good not only for reiser4. Without it different file systems
using ->journal_info cannot "co-exist peacefully".

No new system calls are introduced (yet).

Nikita.

 >
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

2.5 merge candidate list, final version. (End of "crunch time" series.)

Post by Nikita Danilo » Thu, 31 Oct 2002 12:20:15


Quote:Andrew Morton writes:


 > >

 > >
 > > > We are going to submit a patch appropriate for inclusion as an
 > > > experimental FS on Halloween.   I hope you will forgive our pushing
 > > > the limit timewise, it is not by choice, but the algorithms we used to
 > > > more than double reiserfs V3 performance were, quite frankly, hard to
 > > > code.
 > >
 > > Does your merge change the core code at all?  Does it add new syscalls?
 > >
 >
 > Their changes are tiny, and sensible.  See
 > http://www.namesys.com/snapshots/2002.10.29/
 >
 > But I'd like to ask about the status of reiser3 support.
 >
 > Chris had patches *ages* ago to convert it to use direct-to-BIO for
 > reads, and writes should be done as well.  reiserfs3 is still using
 > buffer-head-based IO for bulk reads and writes.  That's a 25-30% hit
 > in CPU cost, and all the old ZONE_NORMAL-full-of-buffer_heads
 > problems.

Are you talking about buffer_heads_over_limit magic inside vmscan.c? Is
it to avoid deadlock with buffer_head allocation in ->vm_writeback?

 >
 > Any plans to get that work finished off?

Chris had done great work indeed (plus data logging, plus quotas, plus
ACLs by Jeff Mahoney, etc). Unfortunately we were too busy with reiser4
here lately. Hopefully, a lot of stuff will be merged into reiserfs3.x
after Halloween.

Nikita.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

2.5 merge candidate list, final version. (End of "crunch time" series.)

Post by Chris Maso » Thu, 31 Oct 2002 19:10:24





> > > We are going to submit a patch appropriate for inclusion as an
> > > experimental FS on Halloween.   I hope you will forgive our pushing
> > > the limit timewise, it is not by choice, but the algorithms we used to
> > > more than double reiserfs V3 performance were, quite frankly, hard to
> > > code.

> > Does your merge change the core code at all?  Does it add new syscalls?

> Their changes are tiny, and sensible.  See
> http://www.namesys.com/snapshots/2002.10.29/

> But I'd like to ask about the status of reiser3 support.

> Chris had patches *ages* ago to convert it to use direct-to-BIO for
> reads, and writes should be done as well.  reiserfs3 is still using
> buffer-head-based IO for bulk reads and writes.  That's a 25-30% hit
> in CPU cost, and all the old ZONE_NORMAL-full-of-buffer_heads
> problems.

> Any plans to get that work finished off?

Very much so, but I wanted it on top of the data journaling stuff to
make porting that easier.  But, in the interest in getting something
more stable and faster out the door asap, I'll pull just the
direct-to-BIO stuff out.

-chris

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/