]> git.hungrycats.org Git - linux/log
linux
22 years agoLinux 2.5.64 v2.5.64
Linus Torvalds [Tue, 4 Mar 2003 11:11:19 +0000 (03:11 -0800)]
Linux 2.5.64

22 years ago[PATCH] More IEEE1394 updates
Ben Collins [Tue, 4 Mar 2003 08:25:16 +0000 (00:25 -0800)]
[PATCH] More IEEE1394 updates

 - Cleanup Makefile.
 - Fix potentional crash in dv1394 module unload.
 - Why does ARM's memcpy not return void*? Oh well, work around it in
  eth1394 by not expecting standard behavior.

22 years agoFix up sunrpc for the dentry hash list changes.
Linus Torvalds [Tue, 4 Mar 2003 07:46:50 +0000 (23:46 -0800)]
Fix up sunrpc for the dentry hash list changes.

22 years agoMerge penguin.transmeta.com:/home/penguin/torvalds/repositories/kernel/hlist
Linus Torvalds [Tue, 4 Mar 2003 06:30:19 +0000 (22:30 -0800)]
Merge penguin.transmeta.com:/home/penguin/torvalds/repositories/kernel/hlist
into penguin.transmeta.com:/home/penguin/torvalds/repositories/kernel/linux

22 years ago[PATCH] dcache/inode hlist patchkit
Andi Kleen [Tue, 4 Mar 2003 06:28:32 +0000 (22:28 -0800)]
[PATCH] dcache/inode hlist patchkit

 - Inode and dcache Hash table only needs half the memory/cache because
   of using hlists.
 - Simplify dcache-rcu code.  With NULL end markers in the hlists
   is_bucket is not needed anymore.  Also the list walking code
   generates better code on x86 now because it doesn't need to dedicate
   a register for the list head.
 - Reorganize struct dentry to be more cache friendly.  All the state
   accessed for the hash walk is in one chunk now together with the
   inline name (all at the end)
 - Add prefetching for all the list walks.  Old hash lookup code didn't
   use it.
 - Some other minor cleanup.

22 years agoMerge bk://ldm.bkbits.net/linux-2.5-core
Linus Torvalds [Tue, 4 Mar 2003 03:09:35 +0000 (19:09 -0800)]
Merge bk://ldm.bkbits.net/linux-2.5-core
into home.transmeta.com:/home/torvalds/v2.5/linux

22 years agoMerge osdl.org:/home/mochel/src/kernel/devel/linux-2.5-virgin
Patrick Mochel [Tue, 4 Mar 2003 06:17:16 +0000 (00:17 -0600)]
Merge osdl.org:/home/mochel/src/kernel/devel/linux-2.5-virgin
into osdl.org:/home/mochel/src/kernel/devel/linux-2.5-core

22 years agosysfs: Register filesystems with sysfs.
Patrick Mochel [Tue, 4 Mar 2003 06:13:14 +0000 (00:13 -0600)]
sysfs: Register filesystems with sysfs.

Originally from Nikita Danliov.

- Add struct subsystem to struct file_system_type.
- Add fs_subsys_init() to explicitly call sysfs_init() very early.
- Register base fs subsystem then, too.
- Register filesystems' embedded subsystem when registered with fs core.

Resulting tree looks like:

# tree /sys/fs/
/sys/fs/
|-- bdev
|-- devpts
|-- eventpollfs
|-- ext2
|-- ext3
|-- futexfs
|-- iso9660
|-- pipefs
|-- proc
|-- ramfs
|-- rootfs
|-- sockfs
|-- sysfs
`-- tmpfs

22 years agodriver model: implement platform_match()
Patrick Mochel [Tue, 4 Mar 2003 04:20:36 +0000 (22:20 -0600)]
driver model: implement platform_match()

- For matching registered platform devices with platform drivers.

- Also, make sure platform_bus is initialized early, with the rest of the
  core stuff.

22 years agosysfs: initialize from fs/namespace.c::mnt_init().
Patrick Mochel [Tue, 4 Mar 2003 04:17:19 +0000 (22:17 -0600)]
sysfs: initialize from fs/namespace.c::mnt_init().

This ensures that sysfs is initialized very early, before anything can
possibily registered with it.

It happens early enough that all filesystems can registered with sysfs
(besides sysfs itself), which is coming next.

22 years agodriver model: Make initialization explicit.
Patrick Mochel [Tue, 4 Mar 2003 04:13:44 +0000 (22:13 -0600)]
driver model: Make initialization explicit.

- Call driver_init() from init/main.c::do_basic_setup().

This ensures that all the driver model subsystems are initialized before
any drivers or devices can be registered.

It nearly frees up the core and postcore initcall levels, making them
available for other kernel code to use freely.

22 years ago[PATCH] IEEE1394 ioctl allocation documentation
Ben Collins [Tue, 4 Mar 2003 02:58:48 +0000 (18:58 -0800)]
[PATCH] IEEE1394 ioctl allocation documentation

22 years agosysfs: fix warning.
Patrick Mochel [Tue, 4 Mar 2003 02:57:38 +0000 (20:57 -0600)]
sysfs: fix warning.

From Andrew Morton

22 years ago[PATCH] IEEE1394 updates
Ben Collins [Tue, 4 Mar 2003 02:55:45 +0000 (18:55 -0800)]
[PATCH] IEEE1394 updates

- Add ioctl32 compat function registration for video1394, dv1394 and
  amdtp.
- Convert SBP-2 driver to new SCSI hotplug host/device interfaces.
- Initial Async ISO (GASP) support.

22 years agopnp interface.c needs <linux/slab.h> for 'kfree()'
Linus Torvalds [Tue, 4 Mar 2003 01:13:00 +0000 (17:13 -0800)]
pnp interface.c needs <linux/slab.h> for 'kfree()'

22 years agoMerge bk://kernel.bkbits.net/davem/sparc-2.5
Linus Torvalds [Tue, 4 Mar 2003 00:42:43 +0000 (16:42 -0800)]
Merge bk://kernel.bkbits.net/davem/sparc-2.5
into home.transmeta.com:/home/torvalds/v2.5/linux

22 years ago[PATCH] sysfs build fix
Andrew Morton [Tue, 4 Mar 2003 00:26:42 +0000 (16:26 -0800)]
[PATCH] sysfs build fix

22 years ago[HOTPLUG]: drivers/base/hotplug.c needs linux/sched.h and linux/string.h
David S. Miller [Mon, 3 Mar 2003 20:27:42 +0000 (12:27 -0800)]
[HOTPLUG]: drivers/base/hotplug.c needs linux/sched.h and linux/string.h

22 years ago[SYSFS]: fs/sysfs/mount.c needs linux/init.h
David S. Miller [Mon, 3 Mar 2003 20:25:17 +0000 (12:25 -0800)]
[SYSFS]: fs/sysfs/mount.c needs linux/init.h

22 years ago[SPARC64]: Fix SMP boot failures.
Bruce D. Elliott [Mon, 3 Mar 2003 20:06:04 +0000 (12:06 -0800)]
[SPARC64]: Fix SMP boot failures.

22 years agosysfs: fix more merge breakage.
Patrick Mochel [Mon, 3 Mar 2003 11:11:30 +0000 (05:11 -0600)]
sysfs: fix more merge breakage.

22 years agosysfs: fixup merge
Patrick Mochel [Mon, 3 Mar 2003 10:59:28 +0000 (04:59 -0600)]
sysfs: fixup merge

22 years agocpufreq: convert to use new interface code.
Patrick Mochel [Mon, 3 Mar 2003 10:38:53 +0000 (04:38 -0600)]
cpufreq: convert to use new interface code.

- Create and register kobjects when adding a cpu to the cpufreq interface,
  instead of the terminally broken struct intf_data.

ACKed by Dominik Brodowski.

22 years ago[PATCH] Another bitop on boolean in pnpbios
John Levon [Mon, 3 Mar 2003 10:13:49 +0000 (02:13 -0800)]
[PATCH] Another bitop on boolean in pnpbios

22 years agoMerge bk://ldm.bkbits.net/linux-2.5-core
Linus Torvalds [Mon, 3 Mar 2003 10:13:05 +0000 (02:13 -0800)]
Merge bk://ldm.bkbits.net/linux-2.5-core
into home.transmeta.com:/home/torvalds/v2.5/linux

22 years ago[PATCH] small tty irq race fix
Nicolas Pitre [Mon, 3 Mar 2003 10:11:31 +0000 (02:11 -0800)]
[PATCH] small tty irq race fix

22 years agoMerge osdl.org:/home/mochel/src/kernel/devel/linux-2.5-virgin
Patrick Mochel [Mon, 3 Mar 2003 07:32:15 +0000 (01:32 -0600)]
Merge osdl.org:/home/mochel/src/kernel/devel/linux-2.5-virgin
into osdl.org:/home/mochel/src/kernel/devel/linux-2.5-core

22 years agodriver model: fix device interfaces.
Patrick Mochel [Mon, 3 Mar 2003 06:48:38 +0000 (00:48 -0600)]
driver model: fix device interfaces.

- Remove struct intf_data.
  It was silly, and gave us an unbalanced interface.
  Instead, when removing a device_interface, all of a class's devices are
  iterated over, and intf->remove_device() is called for each. (Instead
  of the device's list of interfaces it belongs to).
  Interfaces are expected to tell if they support the device being removed.

- Remove struct device::intf_list.

- Protect classes' device list access with devclass_sem, which is taken
  when a device or interface is added or removed. This allows access to the
  list w/o taking the class's lock, which allows subordinate objects to be
  registered (e.g. the interface itself, or objects the interfaces create
  for a device).

- This should fix a deadlock when devices are added to interfaces, which
  calls intf->add_device(), then interface_add_data(), which call
  kobject_register() and hangs, since it tries to take the class's lock
  again.

22 years agokobject: fix oops in to_kset()
Patrick Mochel [Mon, 3 Mar 2003 03:59:26 +0000 (21:59 -0600)]
kobject: fix oops in to_kset()

Check whether kset is NULL or not before doing container_of().

From Dominik Brodowski

22 years agodriver model: remove <linux/sched.h> from include/linux/device.h
Patrick Mochel [Mon, 3 Mar 2003 03:48:58 +0000 (21:48 -0600)]
driver model: remove <linux/sched.h> from include/linux/device.h

From Matt Wilcox

22 years agosysfs: fix oops in directory removal.
Patrick Mochel [Mon, 3 Mar 2003 03:43:37 +0000 (21:43 -0600)]
sysfs: fix oops in directory removal.

If a file that doesn't exist was looked up in a directory, and that
directory is later removed, sysfs would reap the negative dentrys along
with the valid ones.

Fix is to manually remove the dentrys from the parent's list under the
dcache_lock, then check if they're valid with dget_locked().

This ensures all the dentrys are removed, valid and invalid, and we don't
reap anything we shouldn't.

22 years ago[PATCH] Spelling fixes from spell-fix.pl
Steven Cole [Mon, 3 Mar 2003 00:34:23 +0000 (16:34 -0800)]
[PATCH] Spelling fixes from spell-fix.pl

This patch contains the first fruits of the automatic spell checking and
correcting scripts written by Dan Kegel and Matthias Schniedermeyer.
Those scripts are available here: http://www.kegel.com/kerspell/

This patch fixes these six words from the first five lines of
spell-fix.txt.  280 lines remaining.

  accesible   -> accessible
  accesing    -> accessing
  accomodate  -> accommodate
  acommodate  -> accommodate
  Acknowlege  -> Acknowledge
  acknoledged -> acknowledged

22 years agoMerge bk://kernel.bkbits.net/davem/net-2.5
Linus Torvalds [Mon, 3 Mar 2003 00:33:50 +0000 (16:33 -0800)]
Merge bk://kernel.bkbits.net/davem/net-2.5
into home.transmeta.com:/home/torvalds/v2.5/linux

22 years ago[SCTP]: net/sctp/proc.c needs linux/init.h
David S. Miller [Sun, 2 Mar 2003 18:13:09 +0000 (10:13 -0800)]
[SCTP]: net/sctp/proc.c needs linux/init.h

22 years ago[ATM SUNI]: suni_init should not be __init and remove mod inc/dec.
Chas Williams [Sun, 2 Mar 2003 17:49:48 +0000 (09:49 -0800)]
[ATM SUNI]: suni_init should not be __init and remove mod inc/dec.

22 years ago[IPV6]: More C99 initializers.
Hideaki Yoshifuji [Sun, 2 Mar 2003 17:47:53 +0000 (09:47 -0800)]
[IPV6]: More C99 initializers.

22 years ago[IPSEC]: Make sure ESP output pads Null Encryption properly.
Joy Latten [Sun, 2 Mar 2003 17:45:33 +0000 (09:45 -0800)]
[IPSEC]: Make sure ESP output pads Null Encryption properly.

22 years ago[BRIDGE]: C99 patches for net/bridge/netfilter.
Art Haas [Sun, 2 Mar 2003 17:43:48 +0000 (09:43 -0800)]
[BRIDGE]: C99 patches for net/bridge/netfilter.

22 years ago[NETFILTER]: Add new ip6tables matches.
Harald Welte [Sun, 2 Mar 2003 17:40:56 +0000 (09:40 -0800)]
[NETFILTER]: Add new ip6tables matches.

22 years ago[NETFILTER]: Fix typo in ftp conntrack helper.
Harald Welte [Sun, 2 Mar 2003 17:27:23 +0000 (09:27 -0800)]
[NETFILTER]: Fix typo in ftp conntrack helper.

22 years ago[NETFILTER]: Fix conntrack bug introduced by list_del change.
Harald Welte [Sun, 2 Mar 2003 17:25:50 +0000 (09:25 -0800)]
[NETFILTER]: Fix conntrack bug introduced by list_del change.

22 years ago[NETFILTER]: fix NAT ICMP reply translation of inner packet.
Harald Welte [Sun, 2 Mar 2003 17:24:05 +0000 (09:24 -0800)]
[NETFILTER]: fix NAT ICMP reply translation of inner packet.

22 years ago[NETFILTER]: Fix icmp-type all problem in iptables.
Harald Welte [Sun, 2 Mar 2003 17:21:38 +0000 (09:21 -0800)]
[NETFILTER]: Fix icmp-type all problem in iptables.

22 years ago[ATM FORE200E]: Fix build.
Chas Williams [Sun, 2 Mar 2003 17:18:42 +0000 (09:18 -0800)]
[ATM FORE200E]: Fix build.

22 years ago[ATM]: Let upper layer k now lec supports multicast.
Chas Williams [Sun, 2 Mar 2003 17:16:52 +0000 (09:16 -0800)]
[ATM]: Let upper layer k now lec supports multicast.

22 years ago[NETFILTER]: Switch over to new-style module refcounting, help from Christoph Hellwig.
Rusty Russell [Sun, 2 Mar 2003 17:15:46 +0000 (09:15 -0800)]
[NETFILTER]: Switch over to new-style module refcounting, help from Christoph Hellwig.

22 years ago[ATM]: Get minimum frame size right in lec.c
Chas Williams [Sun, 2 Mar 2003 17:13:53 +0000 (09:13 -0800)]
[ATM]: Get minimum frame size right in lec.c

22 years ago[IPV6]: Make sure temporary addresses are regenerated properly.
Hideaki Yoshifuji [Sun, 2 Mar 2003 17:12:02 +0000 (09:12 -0800)]
[IPV6]: Make sure temporary addresses are regenerated properly.

22 years ago[ATM]: Fix mispatch.
David S. Miller [Sun, 2 Mar 2003 17:10:24 +0000 (09:10 -0800)]
[ATM]: Fix mispatch.

22 years ago[ATM]: Use skb_pull instead of direct skb mangling.
Chas Williams [Sun, 2 Mar 2003 17:09:03 +0000 (09:09 -0800)]
[ATM]: Use skb_pull instead of direct skb mangling.

22 years ago[NET]: Remove 2.0/2.2 compat code from netfilter, approved by Rusty.
Christoph Hellwig [Sun, 2 Mar 2003 17:07:05 +0000 (09:07 -0800)]
[NET]: Remove 2.0/2.2 compat code from netfilter, approved by Rusty.

22 years ago[IPV6]: Export in6addr_{any,loopback} to modules.
Bruce Allan [Sun, 2 Mar 2003 17:02:36 +0000 (09:02 -0800)]
[IPV6]: Export in6addr_{any,loopback} to modules.

22 years agoResolve conflicts.
David S. Miller [Sun, 2 Mar 2003 17:01:01 +0000 (09:01 -0800)]
Resolve conflicts.

22 years ago[SPARC64]: Use EXTRA_CFLAGS instead of mangling CFLAGS directly.
David S. Miller [Sun, 2 Mar 2003 16:49:35 +0000 (08:49 -0800)]
[SPARC64]: Use EXTRA_CFLAGS instead of mangling CFLAGS directly.

22 years ago[NET]: Move fc_type_trans into generic code.
Christoph Hellwig [Sun, 2 Mar 2003 16:47:43 +0000 (08:47 -0800)]
[NET]: Move fc_type_trans into generic code.

22 years ago[SPARC32/64]: Expand ioctl size field in backwards-compatible way.
Pete Zaitcev [Sun, 2 Mar 2003 16:45:12 +0000 (08:45 -0800)]
[SPARC32/64]: Expand ioctl size field in backwards-compatible way.

22 years ago[SPARC64]: update us3_cpufreq to support userspace governor.
Dominik Brodowski [Sun, 2 Mar 2003 16:34:51 +0000 (08:34 -0800)]
[SPARC64]: update us3_cpufreq to support userspace governor.

22 years ago[SPARC]: Fix Kconfig typo.
Pete Zaitcev [Sun, 2 Mar 2003 16:33:56 +0000 (08:33 -0800)]
[SPARC]: Fix Kconfig typo.

22 years ago[SPARC]: Kill GFP_DMA in iommu code.
Pete Zaitcev [Sun, 2 Mar 2003 16:33:24 +0000 (08:33 -0800)]
[SPARC]: Kill GFP_DMA in iommu code.

22 years agoMerge nuts.ninka.net:/home/davem/src/BK/sparcwork-2.5
David S. Miller [Sun, 2 Mar 2003 15:36:16 +0000 (07:36 -0800)]
Merge nuts.ninka.net:/home/davem/src/BK/sparcwork-2.5
into nuts.ninka.net:/home/davem/src/BK/sparc-2.5

22 years ago[SPARC64]: Solaris module asm needs asm/thread_info.h
David S. Miller [Sun, 2 Mar 2003 15:25:11 +0000 (07:25 -0800)]
[SPARC64]: Solaris module asm needs asm/thread_info.h

22 years ago[SPARC64]: Update defconfig.
David S. Miller [Sun, 2 Mar 2003 15:24:05 +0000 (07:24 -0800)]
[SPARC64]: Update defconfig.

22 years ago[FRAMEBUFFER]: Convert TCX driver to new APIs.
David S. Miller [Sun, 2 Mar 2003 15:19:36 +0000 (07:19 -0800)]
[FRAMEBUFFER]: Convert TCX driver to new APIs.

22 years ago[FRAMEBUFFER]: Convert P9100 driver to new APIs.
David S. Miller [Sun, 2 Mar 2003 14:52:03 +0000 (06:52 -0800)]
[FRAMEBUFFER]: Convert P9100 driver to new APIs.

22 years ago[FRAMEBUFFER]: Convert cg14 driver to new APIs.
David S. Miller [Sun, 2 Mar 2003 14:28:47 +0000 (06:28 -0800)]
[FRAMEBUFFER]: Convert cg14 driver to new APIs.

22 years agoMerge http://linux-isdn.bkbits.net/linux-2.5.make
Linus Torvalds [Sun, 2 Mar 2003 13:05:27 +0000 (05:05 -0800)]
Merge http://linux-isdn.bkbits.net/linux-2.5.make
into home.transmeta.com:/home/torvalds/v2.5/linux

22 years agokbuild: remove dependency on compile.h
Kai Germaschewski [Sun, 2 Mar 2003 16:55:36 +0000 (10:55 -0600)]
kbuild: remove dependency on compile.h

We had a dummy dependency on include/linux/compile.h, but it really caused
more trouble than benefits. It's not actually needed for the module
postprocessing, it was only put there to make sure we recognize when
gcc changed under us. However, we really can only do so much, and
the rest of kbuild won't notice a changed gcc either, so if the user replaces
gcc during a build, he just can't rely on the build doing the right thing.

The common cases are still covered, anyway. When the command to invoke
gcc changes ("CC=gcc32") we notice, and when the path to

/usr/lib/gcc-lib/i386-redhat-linux/2.96/include/stdarg.h

changes (which luckily contains the "2.96", we'll notice and handle that,
too.

22 years ago[PATCH] kbuild: Top-level Makefile, trivial tidy up
Sam Ravnborg [Sun, 2 Mar 2003 16:38:09 +0000 (10:38 -0600)]
[PATCH] kbuild: Top-level Makefile, trivial tidy up

1) Remove dep from "make help", it is no longer useful
2) replace Generating with GEN when generating version.h

22 years ago[PATCH] kbuild: do not run split-include for all compilations
Sam Ravnborg [Sun, 2 Mar 2003 16:36:48 +0000 (10:36 -0600)]
[PATCH] kbuild: do not run split-include for all compilations

When a rule in the top-level Makefile includes scripts as one
of the prerequisites it inherits FORCE, and thus is always build.
include/linux/autoconf.h recently included scripts hereby forcing
split-include to be run for each compilation.

Fix all rules that lists scripts as a prerequisite but did not list FORCE.
Fixed by listing the executable needed direct.

22 years agodo_mounts: Fix CONFIG_BLK_DEV_MD=m case
Kai Germaschewski [Sun, 2 Mar 2003 16:33:44 +0000 (10:33 -0600)]
do_mounts: Fix CONFIG_BLK_DEV_MD=m case

We don't really have a nice way to say "compile this when CONFIG_FOO
is y, don't otherwise".

Alternatives are:

obj-$(subst m,,$(CONFIG_FOO)) := foo.o

or

obj-$(CONFIG_FOO) := foo.o
obj-m :=

or

obj-y := do_foo.o
do_foo-$(CONFIG_FOO) := foo.o

I chose the last one, though I'm not particularly happy with either.

22 years ago[PATCH] fix Coverted -> Converted
Steven Cole [Sun, 2 Mar 2003 12:44:00 +0000 (04:44 -0800)]
[PATCH] fix Coverted -> Converted

This patch fixes what might have been a joke, but wasn't.

 Coverted -> Converted

One down, 285 to go.

22 years ago[PATCH] cciss: add passthrough ioctl
Andrew Morton [Sun, 2 Mar 2003 12:38:36 +0000 (04:38 -0800)]
[PATCH] cciss: add passthrough ioctl

Patch from Stephen Cameron <steve.cameron@hp.com>

Add new big passthrough ioctl to allow large buffers.  Used by e.g.  online
array controller firmware flash utility.

22 years ago[PATCH] cciss: add CCISS_GETLUNINFO ioctl
Andrew Morton [Sun, 2 Mar 2003 12:38:23 +0000 (04:38 -0800)]
[PATCH] cciss: add CCISS_GETLUNINFO ioctl

Patch from Stephen Cameron <steve.cameron@hp.com>

Add CCISS_GETLUNINFO ioctl.

This ioctl returns the LUNID, number of partitions, and current number of
opens on a logical volume.  Used by the array config utility or any app that
needs to send passthrough commands to a particular logical disk.

22 years ago[PATCH] cciss: add cmd_type to sendcmd parameters
Andrew Morton [Sun, 2 Mar 2003 12:38:15 +0000 (04:38 -0800)]
[PATCH] cciss: add cmd_type to sendcmd parameters

Patch from Stephen Cameron <steve.cameron@hp.com>

* add cmd_type to sendcmd parameters to allow scsi messages to be
  sent down.
* factor out duplicated code into fill_cmd function.

22 years ago[PATCH] cciss: retry bus resets
Andrew Morton [Sun, 2 Mar 2003 12:38:09 +0000 (04:38 -0800)]
[PATCH] cciss: retry bus resets

Patch from Stephen Cameron <steve.cameron@hp.com>

Make cciss driver retry 3rd party bus reset aborted commands up to 3 times.
(ported to 2.5 by me, original patch by Charles White)

This is needed for a multi port storage box that can have multiple hosts
connected to it, or be used in a multipath configuration.  In certain
configurations SCSI bus resets initiated by one host may affect another host.

22 years ago[PATCH] cciss: fix unlikely startup problem
Andrew Morton [Sun, 2 Mar 2003 12:38:02 +0000 (04:38 -0800)]
[PATCH] cciss: fix unlikely startup problem

Patch from Stephen Cameron <steve.cameron@hp.com>

* Make cciss driver wait longer for board to enter simple mode to
  handle an unlikely corner case.  (Hot replace of failed 144GB
  drive in RAID 5 set at just the wrong time prior to driver loading
  can make the board take a longer than usual time to go into "simple
  mode".  Without the patch, the driver gives up too early, and consequently
  doesn't work.  (A reboot will generally "fix" it.)  This patch avoids
  the problem.
* Fix a couple ioctls to return EAGAIN instead of inappropriate EFAULT.

22 years ago[PATCH] hugetlb: fix MAP_FIXED handling
Andrew Morton [Sun, 2 Mar 2003 12:37:54 +0000 (04:37 -0800)]
[PATCH] hugetlb: fix MAP_FIXED handling

Patch from Rohit Seth

ia64 reserves virtual address region 4 (any address which starts with 0x4)
for huge pages.

Apparently, for hardware reasons, we do not wish to allow mappings with
other page sizes to appear in that region.

This patch prevents the user from being able to place regular MAP_FIXED
mappings into region 4 on ia64.  It is a no-op for ia32.

22 years ago[PATCH] hugetlbpage documentation update
Andrew Morton [Sun, 2 Mar 2003 12:37:48 +0000 (04:37 -0800)]
[PATCH] hugetlbpage documentation update

Patch from Rohit Seth.  Updates the hugetlb page documentation.

22 years ago[PATCH] remove dead hugetlb_key forward decl
Andrew Morton [Sun, 2 Mar 2003 12:37:41 +0000 (04:37 -0800)]
[PATCH] remove dead hugetlb_key forward decl

Patch from Rohit Seth.  Remove a dead declaration.

22 years ago[PATCH] allow CONFIG_SWAP=n for i386
Andrew Morton [Sun, 2 Mar 2003 12:37:34 +0000 (04:37 -0800)]
[PATCH] allow CONFIG_SWAP=n for i386

Patch from Christoph Hellwig <hch@sgi.com>

There's a bunch of minor fixes needed to disable the swap code for systems
with mmu.

22 years ago[PATCH] fix typo in arch/i386/kernel/mpparse.c in printk
Andrew Morton [Sun, 2 Mar 2003 12:37:26 +0000 (04:37 -0800)]
[PATCH] fix typo in arch/i386/kernel/mpparse.c in printk

Patch from "Vitezslav Samel" <samel@mail.cz>

Obvious fix of printk level (wrong since 2.5.62).

22 years ago[PATCH] ext2: clear ext3 htree flag on directories
Andrew Morton [Sun, 2 Mar 2003 12:37:19 +0000 (04:37 -0800)]
[PATCH] ext2: clear ext3 htree flag on directories

Forward port of a change which Ted made to 2.4's ext2.

HTREE backwards compatibility patch.

"I thought (and assumed) this patch had been applied to both the ext2 and
 ext3 filesystems in the 2.4 kernel.  It turns out it had only made it into
 the ext3 filesystem code.  This means that if an HTREE-enabled filesystem is
 mounted using ext2, it will corrupt the filesystem as far as e2fsck and an
 ext3 htree-enabled kernel is concerned.  (The corruption won't cause any
 data loss, but it will cause e2fsck and an ext3-htree kernel to omit a lot
 of warning messages.)"

22 years ago[PATCH] clean up redundant code for alloc_pages
Andrew Morton [Sun, 2 Mar 2003 12:37:13 +0000 (04:37 -0800)]
[PATCH] clean up redundant code for alloc_pages

Patch from Matthew Dobson <colpatch@us.ibm.com>

Consolidate alloc_pages() and alloc_pages_node().  There is no change in
sizeof(vmlinux).

22 years ago[PATCH] Don't panic if TSC is enabled and notsc is used
Andrew Morton [Sun, 2 Mar 2003 12:37:07 +0000 (04:37 -0800)]
[PATCH] Don't panic if TSC is enabled and notsc is used

Patch from: john stultz <johnstul@us.ibm.com>

The kernel will currently panic if it was built with CONFIG_X86_TSC and
someone boots it with `notsc'.

Change it to just print a warning.

22 years ago[PATCH] fix preempt-issues with smp_call_function()
Andrew Morton [Sun, 2 Mar 2003 12:36:59 +0000 (04:36 -0800)]
[PATCH] fix preempt-issues with smp_call_function()

Patch from Thomas Schlichter <schlicht@uni-mannheim.de>

Based on a patch from Dave Jones.

It converts a large number of instances of:

smp_call_function(foo);
foo();

into

on_each_cpu(foo);

and in doing so fixes up the preempt-unsafeness of the first version.

22 years ago[PATCH] fix presto_get_sb() return value and oops.
Andrew Morton [Sun, 2 Mar 2003 12:36:51 +0000 (04:36 -0800)]
[PATCH] fix presto_get_sb() return value and oops.

Patch from Maneesh Soni <maneesh@in.ibm.com>

It's supposed to return an ERR_PTR value on error, not NULL.

22 years ago[PATCH] loop: Fix OOM and oops
Andrew Morton [Sun, 2 Mar 2003 12:36:44 +0000 (04:36 -0800)]
[PATCH] loop: Fix OOM and oops

The loop driver takes a copy of the data which it is writing.  When this
happens on the try_to_free_pages() path, loop can easily consume ALL memory
and bio_copy() will fail to allocate a page.

Loop forgets to check the bio_copy() return value and oopses.

Fix this by dropping PF_MEMALLOC and throttling to the block writeout speed.

The patch exports blk_congestion_wait() to modules for this.  This is a
needed export: several filesystems have a "try to allocate and yield if it
failed" loop and blk_congestion_wait() is a more appropriate way of
implementing the sleep in this situation.

22 years ago[PATCH] deadline IO scheduler dispatching fix
Andrew Morton [Sun, 2 Mar 2003 12:36:38 +0000 (04:36 -0800)]
[PATCH] deadline IO scheduler dispatching fix

Patch from Nick Piggin <piggin@cyberone.com.au>

Closes a window in which we can get an invalid (already dispatched) request
in the next_drq cache.

This fixed an oops which was reported by Janet Morgan.

22 years ago[PATCH] fix e100 for big-endian machines
Andrew Morton [Sun, 2 Mar 2003 12:36:31 +0000 (04:36 -0800)]
[PATCH] fix e100 for big-endian machines

Patch from Anton Blanchard <anton@samba.org>

e100 is performing a header checksum calculation which is

a) duplication of the core kernel's calculation of the same and

b) incorrect on big-endian machines.

This fix has been tested on ia32 and pppc64 and acked by davem, jgarzik
and Scott Feldman

22 years ago[PATCH] spelling fixes
Andrew Morton [Sun, 2 Mar 2003 12:36:23 +0000 (04:36 -0800)]
[PATCH] spelling fixes

s/seperate/separate/

22 years ago[ARM] SA11x0 PCMCIA 8
Russell King [Sun, 2 Mar 2003 21:00:15 +0000 (21:00 +0000)]
[ARM] SA11x0 PCMCIA 8

Fix socket memory mapping bug - we need to take account of the
requested card offset when returning the address for ioremap.

22 years ago[ARM] SA11x0 PCMCIA 7
Russell King [Sun, 2 Mar 2003 20:56:17 +0000 (20:56 +0000)]
[ARM] SA11x0 PCMCIA 7

Request PCMCIA socket resources without marking them busy - this
allows future modifications to pcmcia to allow it to claim the
CIS resources for statically mapped sockets.

22 years ago[ARM] SA11x0 PCMCIA 6
Russell King [Sun, 2 Mar 2003 20:48:47 +0000 (20:48 +0000)]
[ARM] SA11x0 PCMCIA 6

Sanitise our pcmcia interrupt handling.  Make sure we only use
set_irq_type after the interrupt has been registered.

Register the socket handling interrupts using the function name
rather than some pointer passed by the generic code.

Let the generic sa11x0 code handle the cards interrupt masking.

22 years ago[ARM] SA11x0 PCMCIA 5
Russell King [Sun, 2 Mar 2003 20:16:48 +0000 (20:16 +0000)]
[ARM] SA11x0 PCMCIA 5

Remove get_irq_info method, passing the information via the init
method instead.  The IRQ allocated to a socket never changes on
these platforms, so the extra code was just bloat.

22 years ago[ARM] SA11x0 PCMCIA 4
Russell King [Sun, 2 Mar 2003 20:09:27 +0000 (20:09 +0000)]
[ARM] SA11x0 PCMCIA 4

Convert all SA11x0 drivers to register with the Linux driver model,
so they work with the pcmcia layer in 2.5.63.

22 years ago[ARM] SA11x0 PCMCIA 3
Russell King [Sun, 2 Mar 2003 19:51:33 +0000 (19:51 +0000)]
[ARM] SA11x0 PCMCIA 3

Make low level socket_status() method take the socket index as an
argument in the same way that our other methods do.  socket_status()
now only returns the status of the requested socket, not all sockets.

22 years ago[ARM] SA11x0 PCMCIA 2
Russell King [Sun, 2 Mar 2003 19:42:53 +0000 (19:42 +0000)]
[ARM] SA11x0 PCMCIA 2

Make low-level configure_socket() method take the socket number as
a parameter rather than embedded in a structure.

22 years ago[ARM] SA11x0 PCMCIA 1
Russell King [Sun, 2 Mar 2003 19:32:19 +0000 (19:32 +0000)]
[ARM] SA11x0 PCMCIA 1

Make SA11x0 generic pcmcia implementation fit better with Dominik's
work - pass around our socket structure rather than the socket number
itself.  Reference socket operations using the socket structure
rather than a static variable.

22 years ago[PATCH] Fix USB address setting
Alan Stern [Sun, 2 Mar 2003 01:48:38 +0000 (17:48 -0800)]
[PATCH] Fix USB address setting

You wouldn't expect that an innocent two-line patch would practically
destroy the entire functionality of the USB subsystem.  But that's what my
last patch did.  I didn't realize until I saw it in Greg's list of patches
forwarded to Linus; that patch includes a mistakenly unreverted prior
change.  This explains the problems I was having yesterday, and probably
also the problems a great many people are having today.

Anyway, here's the correction -- it puts things back to what they should
have been in the first place.  Greg, please apply this and send it to
Linus for his tree as soon as reasonably possible.

22 years ago[PATCH] Fix cardbus build problem
Russell King [Sat, 1 Mar 2003 10:24:05 +0000 (02:24 -0800)]
[PATCH] Fix cardbus build problem

This is a temporary patch which should get cardbus working again.
In order to use pci_remove_behind_bridge(), we need to separately
allocate the pci_dev structures.