]> git.neil.brown.name Git - history.git/log
history.git
24 years ago[PATCH] BKL shifted into ->lookup()
Alexander Viro [Mon, 11 Feb 2002 18:01:17 +0000 (10:01 -0800)]
[PATCH] BKL shifted into ->lookup()

OK, here comes: ->lookup() had lost BKL, all in-tree instances of
->lookup() converted.

I'm adding Documentation/filesystems/porting - with the list of
API changes since 2.4.  Are you OK with that format?

(and yes, this sucker is *post*-compile ;-)

24 years ago[PATCH] discarded section problem
Richard Henderson [Mon, 11 Feb 2002 13:09:45 +0000 (05:09 -0800)]
[PATCH] discarded section problem

What should be happening with the references to the discarded .text.exit
section?  I see a __devexit_p mentioned in Documentation/pci.txt, but it
hasn't been implemented except for down inside ieee1394.

In any case, I need something like the following in order to build with
pre-release binutils 2.12.  If this sort of thing is acceptible I can
prepare a more comprehensive patch.

24 years ago[PATCH] video console fix up
David Mosberger [Mon, 11 Feb 2002 13:09:41 +0000 (05:09 -0800)]
[PATCH] video console fix up

Here is the last patch for today: it enables writecombined mappings
for ia64 in fbmem.c and gets rid of an ugly ia64 simulator workaround
in vgacon.c which isn't needed anymore.

--david

24 years ago[PATCH] fix for elf coredump deadlock
David Mosberger [Mon, 11 Feb 2002 13:09:37 +0000 (05:09 -0800)]
[PATCH] fix for elf coredump deadlock

This patch fixes a deadlock condition in the elf core dump that shows
on ia64 because ELF_CORE_COPY_REGS() needs to access user space (to
get a hold of the backing store of the stacked registers).  Marcelo
already accepted this into 2.4.17.

--david

24 years ago[PATCH] dma64_addr_t fix ups
David Mosberger [Mon, 11 Feb 2002 13:09:34 +0000 (05:09 -0800)]
[PATCH] dma64_addr_t fix ups

This patch fixes up two places whre dma64_addr_t is used incorrectly.
Note that pci_dev->dma_mask and the second argument to
blk_queue_bounce_limit() are both u64, so the old types clearly are
wrong (besides, dma64_addr_t is supposed to be used only with the
pci_dac_*() routines, as per DaveM's earlier mail).

--david

24 years ago[PATCH] drivers/char/pcwd.c
Rob Radez [Mon, 11 Feb 2002 13:09:30 +0000 (05:09 -0800)]
[PATCH] drivers/char/pcwd.c

This patch to drivers/char/pcwd.c against 2.5.4 does two things:
a) Makes one code snippet more consistent with the rest of the code, and
b) Makes it possible for this code to actually work

Nearly the same patch against 2.4 was reviewed by Alan, and, well, the
maintainer seems to have disappeared.  It's also looking like no one uses
this driver much either.

Regards,
Rob Radez

24 years ago[PATCH] updated version of VM_DATA_DEFAULT_FLAGS patch
David Mosberger [Mon, 11 Feb 2002 13:09:26 +0000 (05:09 -0800)]
[PATCH] updated version of VM_DATA_DEFAULT_FLAGS patch

Here is the latest version of the VM_DATA_DEFAULT_FLAGS patch
(relative to 2.5.4).

--david

24 years agoMerge bk://are.twiddle.net:8080/linux-alpha-2.5
Linus Torvalds [Mon, 11 Feb 2002 13:03:47 +0000 (05:03 -0800)]
Merge bk://are.twiddle.net:8080/linux-alpha-2.5
into home.transmeta.com:/home/torvalds/v2.5/linux

24 years agoMerge http://linuxusb.bkbits.net/linus-2.5
Linus Torvalds [Mon, 11 Feb 2002 13:03:04 +0000 (05:03 -0800)]
Merge http://linuxusb.bkbits.net/linus-2.5
into home.transmeta.com:/home/torvalds/v2.5/linux

24 years agoMerge master.kernel.org:/home/mingo/BK/linux-2.5
Linus Torvalds [Mon, 11 Feb 2002 13:01:48 +0000 (05:01 -0800)]
Merge master.kernel.org:/home/mingo/BK/linux-2.5
into home.transmeta.com:/home/torvalds/v2.5/linux

24 years agomerge to the -K3 scheduler.
Ingo Molnar [Mon, 11 Feb 2002 21:47:52 +0000 (22:47 +0100)]
merge to the -K3 scheduler.

24 years agoRemove nr_sectors from bio_end_io end I/O callback. It was a relic
Jens Axboe [Mon, 11 Feb 2002 13:38:30 +0000 (14:38 +0100)]
Remove nr_sectors from bio_end_io end I/O callback. It was a relic
from when completion was potentially called more than once to indicate
partial end I/O. These days bio->bi_end_io is _only_ called when I/O
has completed on the entire bio.

24 years ago[PATCH] (2.5.4) death of ->i_zombie
Alexander Viro [Mon, 11 Feb 2002 12:43:33 +0000 (04:43 -0800)]
[PATCH] (2.5.4) death of ->i_zombie

Rediffed to 2.5.4, documentation added.  This variant grabs
->s_vfs_rename_sem only for cross-directory renames.

24 years ago[PATCH] Optimized UP preempt fix
Robert Love [Mon, 11 Feb 2002 12:41:33 +0000 (04:41 -0800)]
[PATCH] Optimized UP preempt fix

I previously sent a patch by Mikael Pettersson to fix the UP+preempt
problem.  It seems from your BK repository you have not yet merged it;
if so, this patch takes a different approach which is optimal, removing
the unneeded conditional altogether in the UP case.  I have verified UP
and SMP are now correct.  Patch is against 2.5.4, please apply.

Robert Love

24 years agoMerged
Linus Torvalds [Mon, 11 Feb 2002 12:37:37 +0000 (04:37 -0800)]
Merged

24 years ago[PATCH] 2.5.4-pre6 apm compile fix
Stephen Rothwell [Mon, 11 Feb 2002 12:32:46 +0000 (04:32 -0800)]
[PATCH] 2.5.4-pre6 apm compile fix

Here is the patch against 2.5.4.  I have compiled this patch under
2.5.3, so it should still be OK.

This patch just resyncs the driver with 2.4.18-pre (which is what is
being testd by others).  The only outstanding known problem is some
very strange interaction with VMWARE.  But otherwise people seem
happy with the changes.

Original announcement to Dave Jones and Marcelo:

Update a couple of email addresses
Fix the idle handling (this is an improved version of the fix
that Alan Cox has in his -ac tree)
Notify user mode of suspend events before drivers (fix)
Make the idling percentage boot time configurable
Rename kapm-idled to kapmd

Credit to Andreas Steinmetz, Russell King, Thomas Hood and me.

More small updates to come.
--
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

24 years agoMerge http://gkernel.bkbits.net/alpha-2.5
Linus Torvalds [Mon, 11 Feb 2002 12:24:18 +0000 (04:24 -0800)]
Merge http://gkernel.bkbits.net/alpha-2.5
into home.transmeta.com:/home/torvalds/v2.5/linux

24 years agoAdd a couple #includes to fix the alpha build.
Jeff Garzik [Mon, 11 Feb 2002 16:27:43 +0000 (11:27 -0500)]
Add a couple #includes to fix the alpha build.

24 years agothread_saved_pc fix from akpm
Jens Axboe [Mon, 11 Feb 2002 11:15:42 +0000 (12:15 +0100)]
thread_saved_pc fix from akpm

24 years agoSmall batch of IDE code cleanups from Pavel Machek
Jens Axboe [Mon, 11 Feb 2002 10:44:06 +0000 (11:44 +0100)]
Small batch of IDE code cleanups from Pavel Machek

24 years agochange cciss driver version number. Applies to 2.5.4-pre3
Jens Axboe [Mon, 11 Feb 2002 10:40:31 +0000 (11:40 +0100)]
change cciss driver version number.  Applies to 2.5.4-pre3
-- steve.cameron@compaq.com

24 years agoMake cciss driver contribute to entropy pool.
Jens Axboe [Mon, 11 Feb 2002 10:40:07 +0000 (11:40 +0100)]
Make cciss driver contribute to entropy pool.
Applies to 2.5.4-pre3
-- steve.cameron@compaq.com

24 years agoReplace calls to suser() with capable(). Move those checks to be
Jens Axboe [Mon, 11 Feb 2002 10:39:44 +0000 (11:39 +0100)]
Replace calls to suser() with capable().  Move those checks to be
as late as possible to avoid accounting overcharging processes with
privilege usage.  Applies to 2.5.4-pre3
-- steve.cameron@compaq.com

24 years agoReplace awful schedule_timeout polling code with
Jens Axboe [Mon, 11 Feb 2002 10:39:04 +0000 (11:39 +0100)]
Replace awful schedule_timeout polling code with
completions.  Applies to 2.5.4-pre3
-- steve.cameron@compaq.com

24 years agoFix for cciss driver where I had passed the wrong
Jens Axboe [Mon, 11 Feb 2002 10:38:04 +0000 (11:38 +0100)]
Fix for cciss driver where I had passed the wrong
first parameter to grok_partitions in the ioctl for
registering a new disk.

24 years agoadd more build config files to ignore list
Jens Axboe [Mon, 11 Feb 2002 10:36:35 +0000 (11:36 +0100)]
add more build config files to ignore list

24 years agosetup_str[] only used in modular builds.
Jens Axboe [Mon, 11 Feb 2002 10:35:01 +0000 (11:35 +0100)]
setup_str[] only used in modular builds.

24 years agoThis patch fixes a bug that appears when you have more than 16 physical
Jens Axboe [Mon, 11 Feb 2002 10:33:09 +0000 (11:33 +0100)]
This patch fixes a bug that appears when you have more than 16 physical
LUNs attached to a cciss controller, and a tape drive is beyond the 16th
LUN.  In such a case, the tape drive would not be accessible without this
patch.  Applies to 2.5.4-pre3.  -- steve.cameron@compaq.com

24 years agoUpdate Alpha SMP for the new scheduler and preempt api change.
Richard Henderson [Mon, 11 Feb 2002 06:45:52 +0000 (22:45 -0800)]
Update Alpha SMP for the new scheduler and preempt api change.

24 years agoMerge
Jeff Garzik [Mon, 11 Feb 2002 06:30:43 +0000 (01:30 -0500)]
Merge

24 years agoremoved 'typedef' from the kl5kusb105 usb-serial driver.
Greg Kroah-Hartman [Mon, 11 Feb 2002 05:55:54 +0000 (21:55 -0800)]
removed 'typedef' from the kl5kusb105 usb-serial driver.

24 years agoremoved 'typedef' from the Keyspan usb-serial drivers.
Greg Kroah-Hartman [Mon, 11 Feb 2002 05:55:23 +0000 (21:55 -0800)]
removed 'typedef' from the Keyspan usb-serial drivers.

24 years agoremoved 'typedef' from the IO Networks Edgeport usb-serial driver.
Greg Kroah-Hartman [Mon, 11 Feb 2002 05:54:53 +0000 (21:54 -0800)]
removed 'typedef' from the IO Networks Edgeport usb-serial driver.

24 years agoremoved 'typedef' from the ftdi_sio usb-serial driver.
Greg Kroah-Hartman [Mon, 11 Feb 2002 05:54:20 +0000 (21:54 -0800)]
removed 'typedef' from the ftdi_sio usb-serial driver.

24 years agoremoved 'typedef' from the Digi Acceleport usb-serial driver.
Greg Kroah-Hartman [Mon, 11 Feb 2002 05:53:29 +0000 (21:53 -0800)]
removed 'typedef' from the Digi Acceleport usb-serial driver.

24 years agopatch by Simon Evans <spse@secret.org.uk> that adds a Konica USB webcam driver
Greg Kroah-Hartman [Mon, 11 Feb 2002 05:52:50 +0000 (21:52 -0800)]
patch by Simon Evans <spse@secret.org.uk> that adds a Konica USB webcam driver

24 years agopatch from David Probnell, updating the USB error-codes.txt file
Greg Kroah-Hartman [Mon, 11 Feb 2002 05:52:10 +0000 (21:52 -0800)]
patch from David Probnell, updating the USB error-codes.txt file

24 years agopatch from Peter Osterlund <petero2@telia.com> to fix usb-storage debug code
Greg Kroah-Hartman [Mon, 11 Feb 2002 05:51:32 +0000 (21:51 -0800)]
patch from Peter Osterlund <petero2@telia.com> to fix usb-storage debug code
compile problem.

24 years agoMerge
Linus Torvalds [Mon, 11 Feb 2002 03:09:28 +0000 (19:09 -0800)]
Merge

24 years agoMerge master.kernel.org:/home/axboe/linus-merge-2.5
Linus Torvalds [Mon, 11 Feb 2002 02:58:36 +0000 (18:58 -0800)]
Merge master.kernel.org:/home/axboe/linus-merge-2.5
into home.transmeta.com:/home/torvalds/v2.5/linux

24 years agobio_endio doesn't take nr_sectors argument anymore.
Jens Axboe [Mon, 11 Feb 2002 15:05:59 +0000 (16:05 +0100)]
bio_endio doesn't take nr_sectors argument anymore.

24 years agoMerge penguin:v2.5/linux
Linus Torvalds [Mon, 11 Feb 2002 02:40:32 +0000 (18:40 -0800)]
Merge penguin:v2.5/linux
into home.transmeta.com:/home/torvalds/v2.5/linux

24 years agoMerge master.kernel.org:/home/davem/BK/davem-2.5
Linus Torvalds [Mon, 11 Feb 2002 00:53:09 +0000 (16:53 -0800)]
Merge master.kernel.org:/home/davem/BK/davem-2.5
into penguin.transmeta.com:/home/penguin/torvalds/repositories/kernel/linux

24 years agoFixes for premature thread_info changeset.
Richard Henderson [Sun, 10 Feb 2002 23:37:45 +0000 (15:37 -0800)]
Fixes for premature thread_info changeset.
Minor warning removal.

24 years agoUpdate Alpha UP for thread_info and scheduler changes.
Richard Henderson [Sun, 10 Feb 2002 21:42:53 +0000 (13:42 -0800)]
Update Alpha UP for thread_info and scheduler changes.

24 years agoAdd writev support to TUN driver.
David S. Miller [Sun, 10 Feb 2002 15:27:01 +0000 (07:27 -0800)]
Add writev support to TUN driver.
From Eddie C. Dost

24 years agoNetfilter bugfixes from Harald and Paul Russell.
David S. Miller [Sun, 10 Feb 2002 15:24:13 +0000 (07:24 -0800)]
Netfilter bugfixes from Harald and Paul Russell.

24 years agoSplit protocol specific information out from struct sock.
David S. Miller [Sun, 10 Feb 2002 15:08:02 +0000 (07:08 -0800)]
Split protocol specific information out from struct sock.
Work done by Arnaldo Carvalho de Melo.

24 years agoClean up sparc64 build.
David S. Miller [Sun, 10 Feb 2002 14:32:15 +0000 (06:32 -0800)]
Clean up sparc64 build.

24 years agoRemove warning in /proc inode conversions.
Linus Torvalds [Sun, 10 Feb 2002 12:39:28 +0000 (04:39 -0800)]
Remove warning in /proc inode conversions.

24 years agoupdate version v2.5.4
Linus Torvalds [Sun, 10 Feb 2002 11:24:03 +0000 (03:24 -0800)]
update version

24 years ago[PATCH] 2.5.4-pre6 compile fix for i386/kernel/signal.c
Udo A. Steinberg [Sun, 10 Feb 2002 10:45:52 +0000 (02:45 -0800)]
[PATCH] 2.5.4-pre6 compile fix for i386/kernel/signal.c

Fixe a compiler warning in signal.c due to a missing prototype for
"do_coredump".

24 years ago[PATCH] 2.5.4-pre6 apm compile fix
Udo A. Steinberg [Sun, 10 Feb 2002 10:45:47 +0000 (02:45 -0800)]
[PATCH] 2.5.4-pre6 apm compile fix

Make apm compile properly and without warnings.

24 years agoMerge master.kernel.org:/home/davem/BK/davem-2.5
Linus Torvalds [Sun, 10 Feb 2002 09:49:24 +0000 (01:49 -0800)]
Merge master.kernel.org:/home/davem/BK/davem-2.5
into home.transmeta.com:/home/torvalds/v2.5/linux

24 years agoMerge penguin:v2.5/linux
Linus Torvalds [Sun, 10 Feb 2002 09:33:31 +0000 (01:33 -0800)]
Merge penguin:v2.5/linux
into home.transmeta.com:/home/torvalds/v2.5/linux

24 years ago[PATCH] check copy_from_user return codes in serial drivers.
Dave Jones [Sun, 10 Feb 2002 08:44:58 +0000 (00:44 -0800)]
[PATCH] check copy_from_user return codes in serial drivers.

forward-ported from 2.4

24 years ago[PATCH] preempt-kernel on UP compile fix
Robert Love [Sun, 10 Feb 2002 07:46:03 +0000 (23:46 -0800)]
[PATCH] preempt-kernel on UP compile fix

A recent scheduler change looks to of affected this, and I missed it
since I did my last testing on SMP only.  Please, apply.

Robert Love

24 years ago[PATCH] preemptible kernel documentation, etc v2.5.4-pre6
Robert Love [Sun, 10 Feb 2002 04:59:32 +0000 (20:59 -0800)]
[PATCH] preemptible kernel documentation, etc

Linus,

The attached patch adds a Documentation/preempt-locking.txt file which
describes the new locking rules wrt preemptive kernels (ie, watch
per-CPU data, etc).  It also updates a CREDITS entry and adds some
comments.

Patch is against 2.5.4-pre5, please apply.

Robert Love

24 years ago[PATCH] yet another lseek fix
Robert Love [Sun, 10 Feb 2002 04:58:05 +0000 (20:58 -0800)]
[PATCH] yet another lseek fix

Another lseek fix, spotted by Dave Hansen @ IBM.  Missing a
corresponding lock_kernel for an unlock_kernel.  Patch against
2.5.4-pre5, please apply.  Thanks,

Robert Love

24 years agoAutomerge
Linus Torvalds [Sun, 10 Feb 2002 04:48:39 +0000 (20:48 -0800)]
Automerge

24 years agoMerge vaio.(none):/home/kai/kernel/linux-2.5.isdn.link_order
Kai Germaschewski [Sun, 10 Feb 2002 16:55:34 +0000 (17:55 +0100)]
Merge vaio.(none):/home/kai/kernel/linux-2.5.isdn.link_order
into vaio.(none):/home/kai/kernel/linux-2.5.isdn

24 years agoLink ISDN after USB. Otherwise, when builtin, the ST5481 USB ISDN
Kai Germaschewski [Sun, 10 Feb 2002 16:54:15 +0000 (17:54 +0100)]
Link ISDN after USB. Otherwise, when builtin, the ST5481 USB ISDN
adapter driver initializes before the USB layer -> Oops

24 years agorename built-in ISDN object to vmlinux-obj.o
Kai Germaschewski [Sun, 10 Feb 2002 16:46:09 +0000 (17:46 +0100)]
rename built-in ISDN object to vmlinux-obj.o

(was isdn.a before, but there's no need to use an archive here)

24 years agoadd missing "\n" in printk statements.
Kai Germaschewski [Sun, 10 Feb 2002 16:38:43 +0000 (17:38 +0100)]
add missing "\n" in printk statements.

24 years agoremove CONFIG_ISDN entries from arch/Config.help - The real entry is
Kai Germaschewski [Sun, 10 Feb 2002 16:37:03 +0000 (17:37 +0100)]
remove CONFIG_ISDN entries from arch/Config.help - The real entry is
in drivers/isdn/Config.help

24 years agoMerge hostme.bitkeeper.com:/ua/repos/g/gkernel/linus-2.5
Jeff Garzik [Sun, 10 Feb 2002 04:42:19 +0000 (20:42 -0800)]
Merge hostme.bitkeeper.com:/ua/repos/g/gkernel/linus-2.5
into hostme.bitkeeper.com:/ua/repos/g/gkernel/alpha-2.5

24 years agoAutomerge
Linus Torvalds [Sun, 10 Feb 2002 04:35:40 +0000 (20:35 -0800)]
Automerge

24 years agoUpdate alpha defconfig to the options available in 2.5.4-pre5.
Jeff Garzik [Sat, 9 Feb 2002 21:43:15 +0000 (16:43 -0500)]
Update alpha defconfig to the options available in 2.5.4-pre5.

24 years agoMerge http://gkernel.bkbits.net/net-drivers-2.5
Jeff Garzik [Sat, 9 Feb 2002 21:13:46 +0000 (16:13 -0500)]
Merge http://gkernel.bkbits.net/net-drivers-2.5
into rum.normnet.org:/spare/vanilla/net-drivers-2.5

24 years agoMerge rum.normnet.org:/spare/vanilla/linus-2.5
Jeff Garzik [Sat, 9 Feb 2002 21:03:34 +0000 (16:03 -0500)]
Merge rum.normnet.org:/spare/vanilla/linus-2.5
into rum.normnet.org:/spare/vanilla/net-drivers-2.5

24 years agoClean up sparc64 build
David S. Miller [Sat, 9 Feb 2002 20:58:07 +0000 (12:58 -0800)]
Clean up sparc64 build

24 years agoVarious bugfixes to sync with 2.4 version:
James Bottomley [Sat, 9 Feb 2002 16:23:57 +0000 (11:23 -0500)]
Various bugfixes to sync with 2.4 version:

- Fix bug in scripts that was causing CDRW hang
- add sense buffer length initialisation
- update lasi700 to new driver model and fix rmmod hang
- add correct MCA reporting to D700 driver

24 years agoAdd CONFIG_PREEMPTION.
David S. Miller [Sat, 9 Feb 2002 16:07:33 +0000 (08:07 -0800)]
Add CONFIG_PREEMPTION.

24 years agoSparc64 preemption support.
David S. Miller [Sat, 9 Feb 2002 16:03:53 +0000 (08:03 -0800)]
Sparc64 preemption support.

24 years agoMerge http://gkernel.bkbits.net/linus-2.5
Jeff Garzik [Sat, 9 Feb 2002 15:11:54 +0000 (07:11 -0800)]
Merge http://gkernel.bkbits.net/linus-2.5
into hostme.bitkeeper.com:/ua/repos/g/gkernel/net-drivers-2.5

24 years agoBuild fixes for sparc64 outside of sparc64 specific files.
David S. Miller [Sat, 9 Feb 2002 14:53:50 +0000 (06:53 -0800)]
Build fixes for sparc64 outside of sparc64 specific files.

24 years agoSparc64 thread_info implementation.
David S. Miller [Sat, 9 Feb 2002 14:45:28 +0000 (06:45 -0800)]
Sparc64 thread_info implementation.

24 years ago[PATCH] ->s_id cleanups
Alexander Viro [Sat, 9 Feb 2002 14:19:23 +0000 (06:19 -0800)]
[PATCH] ->s_id cleanups

Patch converts a bunch of places that hadn't been caught when
->s_id was introduced.  Please, apply.

24 years ago[PATCH] affs cleanup
Alexander Viro [Sat, 9 Feb 2002 14:18:13 +0000 (06:18 -0800)]
[PATCH] affs cleanup

affs digs in blk_size[] to find the size of device (it
has its analog of superblock living in the middle of disk).  That
information is available in ->s_bdev->bd_inode->i_size - no need
to mess with major/minor/arrays to get it.

24 years ago[PATCH] /proc/modules cleanup (seq_file, again)
Alexander Viro [Sat, 9 Feb 2002 14:18:08 +0000 (06:18 -0800)]
[PATCH] /proc/modules cleanup (seq_file, again)

/proc/modules switched to use of seq_file, cleaned up.

24 years ago[PATCH] /proc/slabinfo cleanup (seq_...)
Alexander Viro [Sat, 9 Feb 2002 14:18:05 +0000 (06:18 -0800)]
[PATCH] /proc/slabinfo cleanup (seq_...)

Straightforward rewrite to seq_file.

24 years ago[PATCH] fix for typo in ext2_new_inode()
Alexander Viro [Sat, 9 Feb 2002 14:18:01 +0000 (06:18 -0800)]
[PATCH] fix for typo in ext2_new_inode()

Please, apply.  Otherwise we eat quota like there's no tomorrow ;-/

24 years agoMerge rum.normnet.org:/spare/vanilla/linus-2.5
Jeff Garzik [Sat, 9 Feb 2002 13:21:38 +0000 (08:21 -0500)]
Merge rum.normnet.org:/spare/vanilla/linus-2.5
into rum.normnet.org:/spare/vanilla/net-drivers-2.5

24 years ago[PATCH] Make cardbus device_register work
Peter Osterlund [Sat, 9 Feb 2002 12:20:19 +0000 (04:20 -0800)]
[PATCH] Make cardbus device_register work

It compiles in -pre5 but doesn't work unless you also apply the patch
below. Without this patch, bus_id will be empty which makes
device_register fail.

24 years agodefconfig:
Linus Torvalds [Sat, 9 Feb 2002 04:17:24 +0000 (20:17 -0800)]
defconfig:
  update

24 years agoMerge home.transmeta.com:/home/torvalds/v2.5/preempt
Linus Torvalds [Sat, 9 Feb 2002 05:55:50 +0000 (21:55 -0800)]
Merge home.transmeta.com:/home/torvalds/v2.5/preempt
into home.transmeta.com:/home/torvalds/v2.5/linux

24 years ago[PATCH] includes cleanup, 2nd try.
Dave Jones [Sat, 9 Feb 2002 05:47:00 +0000 (21:47 -0800)]
[PATCH] includes cleanup, 2nd try.

Big bits first, I'll redo the smaller bits tomorrow after some sleep.
Same as last time, rediffed against pre5

24 years agoupdate version
Linus Torvalds [Sat, 9 Feb 2002 05:35:09 +0000 (21:35 -0800)]
update version

24 years agoMerge penguin:v2.5/linux
Linus Torvalds [Sat, 9 Feb 2002 05:17:57 +0000 (21:17 -0800)]
Merge penguin:v2.5/linux
into home.transmeta.com:/home/torvalds/v2.5/linux

24 years ago[PATCH] 2.5.4pre5 scsi/aha1542.c & DMA changes
Paul Gortmaker [Sat, 9 Feb 2002 05:17:14 +0000 (21:17 -0800)]
[PATCH] 2.5.4pre5 scsi/aha1542.c & DMA changes

There are a couple of errors in the DMA changes to the
aha1542.c driver that appeared in 2.5.4pre3:

- a typo (BUF vs BUS)
- a macro handed a page instead of a scatterlist
- printk format length mismatch

Paul.

24 years agoMake AGP on x86 work temporarily while waiting for the proper fix.
Linus Torvalds [Sat, 9 Feb 2002 04:43:54 +0000 (20:43 -0800)]
Make AGP on x86 work temporarily while waiting for the proper fix.

24 years ago[PATCH] driverfs - preparation to fixes
Alexander Viro [Sat, 9 Feb 2002 03:27:14 +0000 (19:27 -0800)]
[PATCH] driverfs - preparation to fixes

Make dentry and ->i_sem manipulations in driverfs explicit.  Something
is seriously rotten there and real fixes will take non-trivial work
(e.g, I suspect that we actually ought to kill driverfs_rename() - at
least cross-directory one for regular files, maybe completely).  This
patch simply cleans the ground.

24 years ago[PATCH] fs/ext2/balloc.c cleanup
Alexander Viro [Sat, 9 Feb 2002 03:23:50 +0000 (19:23 -0800)]
[PATCH] fs/ext2/balloc.c cleanup

Mostly cosmetical cleanup - sane variable names, etc.
That's pretty much it for balloc.c cleanups - stuff after that
will be about decent threading.

24 years ago[PATCH] ext2_new_block() cleanup
Alexander Viro [Sat, 9 Feb 2002 03:20:52 +0000 (19:20 -0800)]
[PATCH] ext2_new_block() cleanup

in ext2_new_block() inode->i_sb can't be NULL.  Dead code removed.

merged returns in ext2_new_block()

in ext2_new_block() added a local variable (sbi) for sb->u.ext2_sb;
users converted.

calculation of prealloc_goal moved into beginning of function.

new local variable (block).  Initialized to 0, set to the number of
allocated block only when we are known to be successful.
ext2_new_block() returns the value of 'block' now.  As the result, out:
is gone - it's identical to out_unlock now.

check for block being chosen past the end of disk had been moved up No
sense to check that after we had done the preallocations...

quota allocation taken into the beginning of ext2_new_block().  we put
the number of blocks we'd reserved into dq_alloc and then decrement it
each time old code would do DQUOT_ALLOC/increment when it would do
DQUOT_FREE.  On the exit we release dq_alloc blocks I.e.  all
allocations are pushed into the beginning and all freeing - to the end.

now that we had moved the last possible point of failure up, we can do
the same with assignment of final result to block.

code that does allocation within a group had been taken out into a
helper - grab_block(map, size, goal).

decrementing fs-wide free blocks count moved into the beginning of
function.  We try to reserve up to dq_alloc blocks - amount we _can_
grab is limited by the actual free space and root-reserved percentage.
That logics had been taken into a helper - reserve_blocks(sb, amount).
Number of blocks we had managed to reserve is put into es_alloc.  When
we actually grab a block, es_alloc is decremented.  Upon the exit we
release the leftovers.  Old code didn't honor 'reserved for root' limit
when doing prealloc - now that had been fixed.

EXT2_BLOCKS_PER_GROUP(sb) put into a local variable (group_size)

similar "reserve and release leftovers in the end" scheme for group
(i.e.  we try to reserve as much as we want in the group, then count how
much had been used up and release the leftovers in the end).

cleaned up the preallocation loop.

24 years ago[PATCH] (5/5) ext2_free_blocks() cleanup
Alexander Viro [Sat, 9 Feb 2002 03:17:09 +0000 (19:17 -0800)]
[PATCH] (5/5) ext2_free_blocks() cleanup

minor cleanup: update 'freed' only after we finish with group

24 years ago[PATCH] (4/5) ext2_free_blocks() cleanup
Alexander Viro [Sat, 9 Feb 2002 03:17:05 +0000 (19:17 -0800)]
[PATCH] (4/5) ext2_free_blocks() cleanup

in ext2_free_blocks() inode->i_sb can't be NULL; dead code removed.

24 years ago[PATCH] (3/5) ext2_free_blocks() cleanup
Alexander Viro [Sat, 9 Feb 2002 03:17:00 +0000 (19:17 -0800)]
[PATCH] (3/5) ext2_free_blocks() cleanup

new helper - group_release_blocks(desc, bh, count); updates free blocks
count for given group ext2_free_blocks now counts the blocks freed in
group and updates the per-group free blocks count at once.

24 years ago[PATCH] (2/5) ext2_free_blocks() cleanup
Alexander Viro [Sat, 9 Feb 2002 03:16:56 +0000 (19:16 -0800)]
[PATCH] (2/5) ext2_free_blocks() cleanup

we postpone releasing the quota until the end of ext2_free_blocks() and
do that at once.

24 years ago[PATCH] (1/5) ext2_free_blocks() cleanup
Alexander Viro [Sat, 9 Feb 2002 03:16:52 +0000 (19:16 -0800)]
[PATCH] (1/5) ext2_free_blocks() cleanup

new helper function - release_blocks(sb, n); in ext2_free_blocks() we
count the blocks we'd freed and update the fs-wide count of free blocks
in the end - just before unlocking superblock.  instead of doing that
step at a time.

reviewed by ext2 folks

24 years ago[PATCH] Re: [PATCH] Preemptible Kernel for 2.5
Robert Love [Sat, 9 Feb 2002 03:11:35 +0000 (19:11 -0800)]
[PATCH] Re: [PATCH] Preemptible Kernel for 2.5

On Sat, 2002-02-09 at 01:43, Linus Torvalds wrote:

> That will clean up all your issues with header file ordering.

You are right, it did.  I removed all the sched.h dependencies and this
reduced the size of the patch greatly.  I now use current_thread_info()
and none of the header or include hackery from before.  I've tested this
with and without preemption enabled with success.

I appreciate your help with this.

Again, this is a minimal i386-only patch.  I have other arches,
documentation, etc.  Patch against 2.5.4-pre5.  Enjoy,

Robert Love