Hugh Dickins [Thu, 4 Aug 2005 20:07:09 +0000 (13:07 -0700)]
[PATCH] fix VmSize and VmData after mremap
mremap's move_vma is applying __vm_stat_account to the old vma which may
have already been freed: move it to just before the do_munmap.
mremapping to and fro with CONFIG_DEBUG_SLAB=y showed /proc/<pid>/status
VmSize and VmData wrapping just like in kernel bugzilla #4842, and fixed by
this patch - worth including in 2.6.13, though not yet confirmed that it
fixes that specific report from Frank van Maarseveen.
John McCutchan [Thu, 4 Aug 2005 20:07:08 +0000 (13:07 -0700)]
[PATCH] inotify delete race fix
The included patch fixes a problem where a inotify client would receive a
delete event before the file was actually deleted. The bug affects both
dnotify & inotify.
Signed-off-by: John McCutchan <ttb@tentacle.dhs.org> Signed-off-by: Robert Love <rml@novell.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
David Howells [Thu, 4 Aug 2005 20:07:07 +0000 (13:07 -0700)]
[PATCH] Destruction of failed keyring oopses
The attached patch makes sure that a keyring that failed to instantiate
properly is destroyed without oopsing [CAN-2005-2099].
The problem occurs in three stages:
(1) The key allocator initialises the type-specific data to all zeroes. In
the case of a keyring, this will become a link in the keyring name list
when the keyring is instantiated.
(2) If a user (any user) attempts to add a keyring with anything other than
an empty payload, the keyring instantiation function will fail with an
error and won't add the keyring to the name list.
(3) The keyring's destructor then sees that the keyring has a description
(name) and tries to remove the keyring from the name list, which oopses
because the link pointers are both zero.
This bug permits any user to take down a box trivially.
Signed-Off-By: David Howells <dhowells@redhat.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
David Howells [Thu, 4 Aug 2005 20:07:06 +0000 (13:07 -0700)]
[PATCH] Error during attempt to join key management session can leave semaphore pinned
The attached patch prevents an error during the key session joining operation
from hanging future joins in the D state [CAN-2005-2098].
The problem is that the error handling path for the KEYCTL_JOIN_SESSION_KEYRING
operation has one error path that doesn't release the session management
semaphore. Further attempts to get the semaphore will then sleep for ever in
the D state.
This can happen in four situations, all involving an attempt to allocate a new
session keyring:
(1) ENOMEM.
(2) The users key quota being reached.
(3) A keyring name that is an empty string.
(4) A keyring name that is too long.
Any user may attempt this operation, and so any user can cause the problem to
occur.
Signed-Off-By: David Howells <dhowells@redhat.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Paul Mackerras [Thu, 4 Aug 2005 19:53:37 +0000 (12:53 -0700)]
[PATCH] ppc64: fix for kexec boot issue
The kexec boot is not successful on some power machines since all CPUs are
getting removed from global interrupt queue (GIQ) before kexec boot. Some
systems always expect at least one CPU in GIQ. Hence, this patch will make
sure that only secondary CPUs are removed from GIQ.
Signed-off-by: Haren Myneni <hbabu@us.ibm.com> Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
NeilBrown [Thu, 4 Aug 2005 19:53:34 +0000 (12:53 -0700)]
[PATCH] md: yet another attempt to get bitmap-based resync to do the right thing in all cases...
Firstly, R1BIO_Degraded was being set in a number of places in the resync
code, but is never used there, so get rid of those settings.
Then: When doing a resync, we want to clear the bit in the bitmap iff the
array will be non-degraded when the sync has completed. However the current
code would clear the bitmap if the array was non-degraded when the resync
*started*, which obviously isn't right (it is for 'resync' but not for
'recovery' - i.e. rebuilding a failed drive).
This patch calculated 'still_degraded' and uses the to tell bitmap_start_sync
whether this sync should clear the corresponding bit.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
NeilBrown [Thu, 4 Aug 2005 19:53:33 +0000 (12:53 -0700)]
[PATCH] md: always honour md bitmap being read from disk
The code currently will ignore the bitmap if the array seem to be in-sync.
This is wrong if the array is degraded, and probably wrong anyway. If the
bitmap says some chunks are not in in-sync, and the superblock says everything
IS in sync, then something is clearly wrong, and it is safer to trust the
bitmap.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] v4l: oopsfix for BTTV on badly behaved PCI chipsets
no_overlay bttv parameter implemented to fix OOPS on some PCI chipsets
(like some VIA) with these behaviors:
1) If pci_quicks does identify the chip as having troubles to
handle PCI2PCI transfers, no_overlay defaults to 1. The user may force
it to 0, to reenable (not recommended).
2) For newer chipsets not blacklisted, no_overlay=1 is provided as a
workaround until PCI chipset included on /drivers/pci/quirks.c
Thanks to Bodo Eggert <7eggert@gmx.de>
Signed-off-by: Michael Krufky <mkrufky@m1k.net> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Olof Johansson [Thu, 4 Aug 2005 19:53:29 +0000 (12:53 -0700)]
[PATCH] ppc64: Fix UP kernel build
CONFIG_KEXEC breaks UP builds because of a misspelled smp_release_cpus().
Also, the function isn't defined unless built with CONFIG_SMP but it is
needed if we are to go from a UP to SMP kernel. Enable it and document it.
Thanks to Steven Winiecki for reporting this and to Milton for remembering
how it's supposed to work and why.
Signed-off-by: Olof Johansson <olof@lixom.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] ide: fix kmalloc_node breakage in ide driver
Patch fixes oops caused by ide interfaces not on pci. pcibus_to_node
causes the kernel to crash otherwise. Patch also adds a BUG_ON to check if
hwif is NULL.
Michael Gernoth [Thu, 4 Aug 2005 19:43:40 +0000 (20:43 +0100)]
[PATCH] ARM: 2844/1: Add maintainer for Jornada 720
Patch from Michael Gernoth
As discussed on the handhelds.org Jornada mailinglist, I take over
maintainership of the currently unmaintained Jornada 720-port in
the mainline kernel.
Signed-off-by: Michael Gernoth <michael@gernoth.net> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
James Bottomley [Thu, 4 Aug 2005 18:28:40 +0000 (13:28 -0500)]
[SCSI] fix aic7xxx performance issues since 2.6.12-rc2
Several people noticed we dropped quite a bit on benchmark figures.
OK, it was my fault but unfortunately I discovered I ran out of brown
paper bags a while ago and forgot to reorder them.
The issue is that a construct introduced in the conversion of the
driver to use the transport class keyed off whether the block request
was tagged or not. However, the aic7xxx driver doesn't properly set
up the block layer TCQ (it uses the wrong API), so the driver now
things all requests are untagged and we keep it to a queue depth of a
single element. Oops.
The fix is to use the correct TCQ API.
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
This fixes five bugs in the key management syscall interface:
(1) add_key() returns 0 rather than EINVAL if the key type is "".
Checking the key type isn't "" should be left to lookup_user_key().
(2) request_key() returns ENOKEY rather than EPERM if the key type begins
with a ".".
lookup_user_key() can't do this because internal key types begin with a
".".
(3) Key revocation always returns 0, even if it fails.
(4) Key read can return EAGAIN rather than EACCES under some circumstances.
A key is permitted to by read by a process if it doesn't grant read
access, but it does grant search access and it is in the process's
keyrings. That search returns EAGAIN if it fails, and this needs
translating to EACCES.
(5) request_key() never adds the new key to the destination keyring if one is
supplied.
The wrong macro was being used to test for an error condition: PTR_ERR()
will always return true, whether or not there's an error; this should've
been IS_ERR().
Signed-Off-By: David Howells <dhowells@redhat.com> Signed-Off-By: Linus Torvalds <torvalds@osdl.org>
This removes the calls to device_suspend() from the shutdown path that
were added sometime during 2.6.13-rc*. They aren't working properly on
a number of configs (I got reports from both ppc powerbook users and x86
users) causing the system to not shutdown anymore.
I think it isn't the right approach at the moment anyway. We have
already a shutdown() callback for the drivers that actually care about
shutdown and the suspend() code isn't yet in a good enough shape to be
so much generalized. Also, the semantics of suspend and shutdown are
slightly different on a number of setups and the way this was patched in
provides little way for drivers to cleanly differenciate. It should
have been at least a different message.
For 2.6.13, I think we should revert to 2.6.12 behaviour and have a
working suspend back.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Russell King [Thu, 4 Aug 2005 13:17:33 +0000 (14:17 +0100)]
[PATCH] ARM: Fix ARM fault handler for get_user_pages() fixes.
The ARM fault handler is optimised to make the fast path, err, fast.
The renumbering of the VM_FAULT_* codes broke this because numbers
were used instead of the definitions. Fix this.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Len Brown [Thu, 4 Aug 2005 04:17:42 +0000 (00:17 -0400)]
When a merge does not work automatically, git prevents
commit from running until a change has been made in
the destination. In this instance the desired result
was to choose the destination version of the file
and ignore the source version, but git would not
allow that.
Here I added a blank line to let git commit think
I resolved a merge conflict.
Luming Yu [Wed, 3 Aug 2005 21:38:04 +0000 (17:38 -0400)]
[ACPI] revert Embedded Controller to polling-mode by default (ala 2.6.12)
Burst mode isn't ready for prime time,
but can be enabled for test via "ec_burst=1"
Signed-off-by: Luming Yu <luming.yu@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
Ian Campbell [Wed, 3 Aug 2005 19:34:52 +0000 (20:34 +0100)]
[PATCH] ARM: 2833/2: Remove support for WDIOF_MAGICCLOSE from sa1100-wdt
Patch from Ian Campbell
On PXA255 there is no way to disable the watchdog. Turning off OIER[E3]
as suggested in the existing comment does not work.
I posted a note to the ARM mailing list a little while ago asking for
opinions from people using SA1100. There was one reponse from Nico who
believes that the SA1100 is the same as the PXA255 in this respect.
You also asked me to involve the watchdog maintainer which I tried to
do but didn't hear anything back. There are only a couple of other
drivers which can't stop the watchdog and there seems to be no
consistancy regarding printing an error etc. I decided to print
something since that matches the case for all the other drivers when
NOWAYOUT is turned on.
Also, I changed the device .name to "watchdog" like most of the other
watchdogs. udev uses it as the device name (by default) and spaces etc.
get in the way.
Superceded 2833/1 because 2.6.13-rc4 caused rejects.
Signed-off-by: Ian Campbell <icampbell@arcom.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Deepak Saxena [Wed, 3 Aug 2005 18:58:21 +0000 (19:58 +0100)]
[PATCH] ARM: 2835/1: Add UPF_SKIP_TEST to IXP4xx serial ports
Patch from Deepak Saxena
This allows the serial driver autconf to work properly on all the IXP
serial ports. W/o it we basically put the serial port in an unrecoverable
state and lose console.
Signed-off-by: Deepak Saxena <dsaxena@plexity.net> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Catalin Marinas [Wed, 3 Aug 2005 18:53:25 +0000 (19:53 +0100)]
[PATCH] ARM: 2841/1: Fix VFP +/-0 case for doubles addition
Patch from Catalin Marinas
The IEEE 754 standard specifies that the result of (x - x), where x is
a valid number, should be -0 if the rounding mode is towards minus
infinity or +0 otherwise.
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Deepak Saxena [Wed, 3 Aug 2005 18:49:17 +0000 (19:49 +0100)]
[PATCH] ARM: 2839/1: Remove XScale cache and TLB locking code
Patch from Deepak Saxena
The XScale locking code is not something that has been validated
on 2.6 and needs to be replaced with a more generic API to use
with other ARMs that support locking features.
Signed-off-by: Deepak Saxena <dsaxena@plexity.net> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Richard Purdie [Wed, 3 Aug 2005 18:49:17 +0000 (19:49 +0100)]
[PATCH] ARM: 2837/2: Re: ARM: Make NWFPE preempt safe
Patch from Richard Purdie
NWFPE used global variables which meant it wasn't safe for use with
preemptive kernels. This patch removes them and communicates the
information between functions in a preempt safe manner. Generation
of some exceptions was broken and this has also been corrected.
Tests with glibc's maths test suite show no change in the results
before/after this patch.
Signed-off-by: Richard Purdie <rpurdie@rpsys.net> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Ben Dooks [Wed, 3 Aug 2005 18:49:16 +0000 (19:49 +0100)]
[PATCH] ARM: 2832/1: BAST - limit clock-rate for IIC bus
Patch from Ben Dooks
The default clock rate does not specify a maximum, so the
default of 400KHz is used. This rate is too fast for the PMU
on the EB2410ITX, so we now specify platform data with a rate
of around 100KHz.
Signed-off-by: Ben Dooks <ben-linux@fluff.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Nick Piggin [Wed, 3 Aug 2005 10:24:01 +0000 (20:24 +1000)]
[PATCH] fix get_user_pages bug
Checking pte_dirty instead of pte_write in __follow_page is problematic
for s390, and for copy_one_pte which leaves dirty when clearing write.
So revert __follow_page to check pte_write as before, and make
do_wp_page pass back a special extra VM_FAULT_WRITE bit to say it has
done its full job: once get_user_pages receives this value, it no longer
requires pte_write in __follow_page.
But most callers of handle_mm_fault, in the various architectures, have
switch statements which do not expect this new case. To avoid changing
them all in a hurry, make an inline wrapper function (using the old
name) that masks off the new bit, and use the extended interface with
double underscores.
Yes, we do have a call to do_wp_page from do_swap_page, but no need to
change that: in rare case it's needed, another do_wp_page will follow.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
[ Cleanups by Nick Piggin ] Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[ACPI] PCI interrupt link suspend/resume - revert to 2.6.12 behaviour
This patch disables the PCI Interrupt Link refernece counts,
which should not co-exist with the 2.6.12 irq_router.resume
method or else a double acpi_pci_link_set() could result
on resume.
Signed-off-by: David Shaohua Li <shaohua.li@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
Haren Myneni [Wed, 3 Aug 2005 05:08:18 +0000 (15:08 +1000)]
[PATCH] Xmon bug fix for soft-reset
For soft reset during system hang, got an error "CPU did not take
control" for some CPUs even though they responded to soft-reset (called
SystemReset, die and called debugger - xmon). First these CPUs entered
into xmon by IPI callback and then got a soft-reset exception and
re-entered into xmon again. The first CPU which re-entered into xmon got
the output lock and made into xmon successfully without unlocking.
Hence, the next CPU(s) which re-entered into xmon try to acquire a lock
(get_output_lock). Therefore, we can not view state of those CPU(s).
[This is a simple, very low risk, obvious fix for an obvious bug, and
should go into 2.6.13. -- paulus]
Signed-off-by: Haren Myneni <hbabu@us.ibm.com> Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Ivan Kokshaysky [Tue, 2 Aug 2005 23:09:03 +0000 (03:09 +0400)]
[PATCH] ACPI: increase PCIBIOS_MIN_IO on x86
We have increased PCIBIOS_MIN_IO to 0x4000, but still want
motherboard resources to be allocated properly. So we need
to state 0x1000 (according to the comment) limit explicitely.
Signed-off-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Ivan Kokshaysky [Tue, 2 Aug 2005 22:59:47 +0000 (02:59 +0400)]
[PATCH] increase PCIBIOS_MIN_IO on x86
There is a number of x86 laptops that have some non-PCI IO ports
in the 0x1000-0x1fff range, and it's quite hard to control the correct
order of resource allocation between PCI and other subsystems controlling
these ports. Especially with modular kernel.
So just increase PCIBIOS_MIN_IO to 0x4000 to prevent any new PCI
resource allocations in the problematic range (this limitation must
apply _only_ to the root bus resources - see Linus' change in
pci_bus_alloc_resource). As PCIBIOS_MIN_IO and PCIBIOS_MIN_CARDBUS_IO
are the same now on i386 and x86-64, we can remove the latter.
Signed-off-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Linus Torvalds [Tue, 2 Aug 2005 21:55:40 +0000 (14:55 -0700)]
pci: make bus resource start address override minimum IO address
The reason we have PCIBIOS_MIN_IO and PCIBIOS_MIN_CARDBUS_IO is because
we want to protect badly documented motherboard PCI resources and thus
don't want to allocate new resources in low IO/MEM space.
However, if we have already discovered a PCI bridge with a specified
resource base, that should override that decision.
This change will allow us to move the "careful" region upwards without
resulting in problems allocating resources in low mappings. This was
brought on by us having allocated a bus resource at 0x1000, conflicting
with a undocumented VAIO Sony PI resources.
Jens Axboe [Tue, 2 Aug 2005 18:08:02 +0000 (20:08 +0200)]
[PATCH] cfq-iosched: fix problem with barriers and max_depth == 1
CFQ will currently stall when using write barriers and the default
max_depth setting of 1, since we artificially need a depth of 2 when
pre-pending the first flush. So never deny the barrier request going to
the device.
This is a regression since 2.6.12, it was found in SUSE testing.
James Bottomley [Tue, 2 Aug 2005 14:32:17 +0000 (09:32 -0500)]
[SCSI] aic7xxx: final fixes for DT handling
The aic7xxx can support Data Group transfers at periods > 12.5, so
eliminate that restriction. Additionally wide is a requirement for DT
so ensure wide is set if users request DT.
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Paul Mackerras [Tue, 2 Aug 2005 11:51:36 +0000 (21:51 +1000)]
[PATCH] Obvious bugfix for yenta resource allocation
Recent changes (well, dating from 12 July) have broken cardbus on my
powerbook: I get 3 messages saying "no resource of type xxx available,
trying to continue", and if I plug in my wireless card, it complains
that there are no resources allocated to the card. This all worked in
2.6.12.
Looking at the code in yenta_socket.c, function yenta_allocate_res,
it's obvious what is wrong: if we get to line 639 (i.e. there wasn't a
usable preassigned resource), we will always flow through to line 668,
which is the printk that I was seeing, even if a resource was
successfully allocated. It looks to me as though there should be a
return statement after the two config_writel's in each of the 3
branches of the if statements, so that the function returns after
successfully setting up the resource.
The patch below adds these return statements, and with this patch,
cardbus works on my powerbook once again.
Kai Makisara [Tue, 2 Aug 2005 09:21:51 +0000 (12:21 +0300)]
[SCSI] Fix SCSI tape oops at module removal
Removing the SCSI tape module results in an oops in class_device_destroy if
any devices are present. The patch at the end of this message fixes the bug
by moving class_destroy() later in exit_st() so that the class still exists
when devices are removed. (The bug is old but class_simple_device_remove() did
nothing when the class did not exist.)
The patch also fixes a "class leak" in init_st() error path.
I would like to get this into 2.6.13 but it may be too late?
Signed-off-by: Kai Makisara <kai.makisara@kolumbus.fi> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Mike Kravetz [Tue, 2 Aug 2005 04:11:48 +0000 (21:11 -0700)]
[PATCH] ppc64: POWER 4 fails to boot with NUMA
If CONFIG_NUMA is set, some POWER 4 systems will fail to boot. This is
because of special processing needed to handle invalid node IDs (0xffff) on
POWER 4. My previous patch to handle memory 'holes' within nodes forgot to
add this special case for POWER 4 in one place.
In reality, I'm not sure that configuring the kernel for NUMA on POWER 4 makes
much sense. Are there POWER 4 based systems with NUMA characteristics that
are presented by the firmware? But, distros want one kernel for all systems
so NUMA is on by default in their kernels. The patch handles those cases.
Signed-off-by: Mike Kravetz <kravetz@us.ibm.com> Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Rusty Russell [Tue, 2 Aug 2005 04:11:47 +0000 (21:11 -0700)]
[PATCH] Module per-cpu alignment cannot always be met
The module code assumes noone will ever ask for a per-cpu area more than
SMP_CACHE_BYTES aligned. However, as these cases show, gcc asks sometimes
asks for 32-byte alignment for the per-cpu section on a module, and if
CONFIG_X86_L1_CACHE_SHIFT is 4, we hit that BUG_ON(). This is obviously an
unusual combination, as there have been few reports, but better to warn
than die.
Eric Dumazet [Tue, 2 Aug 2005 04:11:43 +0000 (21:11 -0700)]
[PATCH] sys_set_mempolicy() doesnt check if mode < 0
A kernel BUG() is triggered by a call to set_mempolicy() with a negative
first argument. This is because the mode is declared as an int, and the
validity check doesnt check < 0 values. Alternatively, mode could be
declared as unsigned int or unsigned long.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Hugh Dickins [Tue, 2 Aug 2005 04:11:42 +0000 (21:11 -0700)]
[PATCH] x86_64: access of some bad address
x86_64 has a large sparse gate area between VSYSCALL_START and
VSYSCALL_END, not all of it presently backed by pmds. Alexander Nyberg has
found that in some circumstances gdb may try to ptrace here, and hit
get_user_pages BUG_ON. It seems odd that gdb should be accessing here, but
it certainly shouldn't crash in this way: relax BUG_ON to -EFAULT. Fixes
kernel bugzilla #4801.
Roman Zippel [Tue, 2 Aug 2005 04:11:40 +0000 (21:11 -0700)]
[PATCH] hfs: don't dirty unchanged inode
If inode size hasn't changed, don't do anything further in truncate, which
also prevents a dirty inode, what might upset some readonly devices quite
badly.
Signed-off-by: Roman Zippel <zippel@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Mark Haverkamp [Tue, 2 Aug 2005 04:11:37 +0000 (21:11 -0700)]
[PATCH] aacraid: Fix for controller load based timeouts
Martin Drab found that he could get aacraid timeouts with high load on his
controller / disk drive combinations. After some experimentation Mark
Salyzyn has come up with a patch to reduce the default max_sectors to
something that will keep the controller from being overloaded and will
eliminate the timeout issues.
Signed-off-by: Mark Haverkamp <markh@osdl.org> Cc: James Bottomley <James.Bottomley@steeleye.com> Acked-by: Mark Salyzyn <mark_salyzyn@adaptec.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Hirokazu Takata [Tue, 2 Aug 2005 04:11:35 +0000 (21:11 -0700)]
[PATCH] m32r: Fix local-timer event handling
There was a scheduling problem of the m32r SMP kernel; A process rarely
stopped and gave no responding but the other process have been handled by
the other CPU still lives, then if we did something in the other terminal
or something like that, the stopped process came back to life and continued
its operation... (ex. LMbench: lat_sig)
In the m32r SMP kernel, a local-timer event is delivered by using an
IPI(inter processor interrupts); LOCAL_TIMER_IPI. And a function
smp_send_timer() is prepared to send the LOCAL_TIMER_IPI from the current
CPU to the other CPUs.
The funtion smp_send_timer() was placed and used in do_IRQ() in
former times (before 2.6.10-rc3-mm1 kernel), however, it was
unintentionally removed when arch/m32r/kernel/irq.c was modified to
employ the generic hardirq framework (CONFIG_GENERIC_HARDIRQ) in
my previous patch.
[PATCH 2.6.10-rc3-mm1] m32r: Use generic hardirq framework
http://www.ussg.iu.edu/hypermail/linux/kernel/0412.2/0358.html
Kumar Gala [Mon, 1 Aug 2005 05:34:51 +0000 (22:34 -0700)]
[PATCH] ppc32: Mark boards that don't build as BROKEN
Marked APUS and GEMINI as BROKEN since they do not build at the platform
level. We have requested that the maintainers of these boards/platforms
fix them by the time 2.6.15 is released or we plan on concerning them
unmaintained and thus removing them.
Signed-off-by: Kumar Gala <kumar.gala@freescale.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Denis Vlasenko [Mon, 1 Aug 2005 05:34:50 +0000 (22:34 -0700)]
[PATCH] silence cs89x0
cs89x0 talks a lot at boot. Seems like debug leftover. This patch
downgrades printks to KERN_DEBUG. While we're at it, make these messages a
bit less obscure.
Cc: Jeff Garzik <jgarzik@pobox.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The code that sets the altivec capability of the CPU based on firmware
informations can enable altivec when the kernel has CONFIG_ALTIVEC
disabled. This results in "interesting" crashes.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- BTTV and SAA7134 version numbers incremented to reflect changes.
- pci_dma_supported() is called after pci_set_dma_mask() which
already did check that for us. This patch removes the unneeded call to
pci_dma_supported() at bttv-driver.c
- Ensure a sufficient I2C bus idle time between 2 messages for
saa7134-i2c.c
- It is important to write at first to MO_GP3_IO for cx88-tvaudio.c
- Use try_to_freeze() instead of refrigerator at msp3400.c
- Recognizing the MFPE05-2 Tuner at tveeprom.c
- Add new parameter to help identify radio chipsets at tuner module:
show_i2c=1 will show 16 reading bytes from detected tuners.
- BTTV does generate some Unimplemented IOCTL log at tuner module:
0x40046d11(dir=1,tp=0x6d,nr=17,sz=4) means that it is sending
MSP3400 calls to non-msp3400 tuners. Warning eliminated.
VIDIOSAUDIO is also called, so debug messages updated. It is still
requiring IOCTL implementation.
- Added two more tuners.
- Add support for the SVideo input on the GDI Black Gold.
Signed-off-by: Peter Missel <peter.missel@onlinehome.de> Signed-off-by: Graham Bevan <graham.bevan@ntlworld.com> Signed-off-by: Torsten Seeboth <Torsten.Seeboth@t-online.de> Signed-off-by: Hartmut Hackmann <hartmut.hackmann@t.online.de> Signed-off-by: Tobias Klauser <tklauser@nuerscht.ch> Signed-off-by: Michael Krufky <mkrufky@m1k.net> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Eric Lammerts [Mon, 1 Aug 2005 05:34:42 +0000 (22:34 -0700)]
[PATCH] disable addres space randomization default on transmeta CPUs
We know that the randomisation slows down some workloads on Transmeta CPUs
by quite large amounts. We think it's because the CPU needs to recode the
same x86 instructions when they pop up at a different virtual address after
a fork+exec.
So disable randomization by default on those CPUs.
Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Linus Torvalds [Mon, 1 Aug 2005 19:37:54 +0000 (12:37 -0700)]
Revert ACPI interrupt resume changes
If there are devices that use interrupts over a suspend event, ACPI must
restore the PCI interrupt links on resume. Anything else breaks any
device that hasn't been converted to the new (dubious) PM rules.
Drivers that need the irq free/re-aquire sequence can be done one by one
independently of this one.
Linus Torvalds [Mon, 1 Aug 2005 18:14:49 +0000 (11:14 -0700)]
Fix get_user_pages() race for write access
There's no real guarantee that handle_mm_fault() will always be able to
break a COW situation - if an update from another thread ends up
modifying the page table some way, handle_mm_fault() may end up
requiring us to re-try the operation.
That's normally fine, but get_user_pages() ended up re-trying it as a
read, and thus a write access could in theory end up losing the dirty
bit or be done on a page that had not been properly COW'ed.
This makes get_user_pages() always retry write accesses as write
accesses by making "follow_page()" require that a writable follow has
the dirty bit set. That simplifies the code and solves the race: if the
COW break fails for some reason, we'll just loop around and try again.
[PATCH] tridentfb: Fix scrolling artifacts during disk IO
Reported by: Jochen Hein (Bugzilla Bug 4312)
When there is disk I/O happening, the framebuffer has a little snow on
the screen. Once I/O has finished, no garbage remains on screen.
This bug was explained by: Knut Petersen
Most important is CRTC register 2f, signal quality is also improved for
higher vclk values by changing set_vclk() according to the X drivers and
cyblafb.c
The fix is to set the performance register (0x2f) with a more stable
value.
[PATCH] tridentfb: Fix scrolling artifacts if acceleration is enabled
Reported by: Jochen Hein (Bugzilla Bug 4386)
booting leaves the end of long lines in the last line on screen when
scrolling. When X is running, scrolling puts garbage on the screen
(looks like X data) Console switch fixes the screen. Behaviour seems to
be identical with noaccel and without on the video=tridentfb parameter
in lilo.conf.
This bug was explained by: Knut_Petersen
Acceleration is broken for all BLADE 3D chips for all versions of kernel
2.6 except for 32bit modes. Most important reason is that the u32 col
parameter of the graphics engine needs the color value replicated to all
u8 of the u32 (8bit modes) and to both u16 of the u32.
Fix color value passed to graphics engine, verified by the reporter.
Ingo Molnar [Mon, 1 Aug 2005 11:39:13 +0000 (13:39 +0200)]
[PATCH] remove sys_set_zone_reclaim()
This removes sys_set_zone_reclaim() for now. While i'm sure Martin is
trying to solve a real problem, we must not hard-code an incomplete and
insufficient approach into a syscall, because syscalls are pretty much
for eternity. I am quite strongly convinced that this syscall must not
hit v2.6.13 in its current form.
Firstly, the syscall lacks basic syscall design: e.g. it allows the
global setting of VM policy for unprivileged users. (!) [ Imagine an
Oracle installation and a SAP installation on the same NUMA box fighting
over the 'optimal' setting for this flag. What will they do? Will they
try to set the flag to their own preferred value every second or so? ]
Secondly, it was added based on a single datapoint from Martin:
where Martin characterizes the numbers the following way:
' Run-to-run variability for "make -j" is huge, so these numbers aren't
terribly useful except to see that with reclaim the benchmark still
finishes in a reasonable amount of time. '
in other words: the fundamental problem has likely not been solved, only
a tendential move into the right direction has been observed, and a
handful of numbers were picked out of a set of hugely variable results,
without showing the variability data. How much variance is there
run-to-run?
I'd really suggest to first walk the walk and see what's needed to get
stable & predictable kernel compilation numbers on that NUMA box, before
adding random syscalls to tune a particular aspect of the VM ... which
approach might not even matter once the whole picture has been analyzed
and understood!
The third, most important point is that the syscall exposes VM tuning
internals in a completely unstructured way. What sense does it make to
have a _GLOBAL_ per-node setting for 'should we go to another node for
reclaim'? If then it might make sense to do this per-app, via numalib or
so.
The change is minimalistic in that it doesnt remove the syscall and the
underlying infrastructure changes, only the user-visible changes. We
could perhaps add a CAP_SYS_ADMIN-only sysctl for this hack, a'ka
/proc/sys/vm/swappiness, but even that looks quite counterproductive
when the generic approach is that we are trying to reduce the number of
external factors in the VM balance picture.
[PATCH] pcmcia: fix multiple insertion of multifunction cards
The ordering of setting and clearing device_add_pending went wrong on some
occasions, causing multifunction cards only to be handled correctly on the
first insertion, not on subsequent ones.
projects / linux-2.6 / log