Matthew Wilcox [Tue, 10 Oct 2006 14:01:21 +0000 (08:01 -0600)]
PCI: Replace HAVE_ARCH_PCI_MWI with PCI_DISABLE_MWI
pSeries is the only architecture left using HAVE_ARCH_PCI_MWI and it's
really inappropriate for its needs. It really wants to disable MWI
altogether. So here are a pair of stub implementations for pci_set_mwi
and pci_clear_mwi.
Also rename pci_generic_prep_mwi to pci_set_cacheline_size since that
better reflects what it does.
Signed-off-by: Matthew Wilcox <matthew@wil.cx> Cc: Paul Mackerras <paulus@samba.org> Acked-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Matthew Wilcox [Tue, 10 Oct 2006 14:01:20 +0000 (08:01 -0600)]
PCI: Use pci_generic_prep_mwi on sparc64
The setting of the CACHE_LINE_SIZE register in sparc64's pci
initialisation code isn't quite adequate as the device may have
incompatible requirements. The generic code tests for this, so switch
sparc64 over to using it.
Since sparc64 has different L1 cache line size and PCI cache line size,
it would need to override the generic code like i386 and ia64 do. We
know what the cache line size is at compile time though, so introduce a
new optional constant PCI_CACHE_LINE_BYTES.
Signed-off-by: Matthew Wilcox <matthew@wil.cx> Signed-off-by: David Miller <davem@davemloft.net> Acked-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Matthew Wilcox [Tue, 10 Oct 2006 14:01:19 +0000 (08:01 -0600)]
PCI: Use pci_generic_prep_mwi on ia64
The pci_generic_prep_mwi() code does everything that pcibios_prep_mwi()
does on ia64. All we need to do is be sure that pci_cache_line_size
is set appropriately, and we can delete pcibios_prep_mwi().
Using SMP_CACHE_BYTES as the default was wrong on uniprocessor machines
as it is only 8 bytes. The default in the generic code of L1_CACHE_BYTES
is at least as good.
Signed-off-by: Matthew Wilcox <matthew@wil.cx> Acked-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Alan Cox [Tue, 3 Oct 2006 23:41:26 +0000 (00:41 +0100)]
PCI: quirks: fix the festering mess that claims to handle IDE quirks
The number of permutations of crap we do is amazing and almost all of it
has the wrong effect in 2.6.
At the heart of this is the PCI SFF magic which says that compatibility
mode PCI IDE controllers use ISA IRQ routing and hard coded addresses
not the BAR values. The old quirks variously clears them, sets them,
adjusts them and then IDE ignores the result.
In order to drive all this garbage out and to do it portably we need to
handle the SFF rules directly and properly. Because we know the device
BAR 0-3 are not used in compatibility mode we load them with the values
that are implied (and indeed which many controllers actually
thoughtfully put there in this mode anyway).
This removes special cases in the IDE layer and libata which now knows
that bar 0/1/2/3 always contain the correct address. It means our
resource allocation map is accurate from boot, not "mostly accurate"
after ide is loaded, and it shoots lots of code. There is also lots more
code and magic constant knowledge to shoot once this is in and settled.
Been in my test tree for a while both with drivers/ide and with libata.
Wants some -mm shakedown in case I've missed something dumb or there are
corner cases lurking.
Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Shouldn't PCI-X state be saved/restored? No device really needs this
right now. qla24xx (fc HBA) and mthca (infiniband) don't do suspend,
and sky2 resets its tweaks when links are brought up.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Francois Romieu [Wed, 29 Nov 2006 21:21:33 +0000 (22:21 +0100)]
[PATCH] r8169: Fix iteration variable sign
This changes the type of variable "i" in rtl8169_init_one()
from "unsigned int" to "int". "i" is checked for < 0 later,
which can never happen for "unsigned". This results in broken
error handling.
Milan Svoboda [Wed, 29 Nov 2006 11:09:52 +0000 (12:09 +0100)]
[ARM] 3943/1: share declaration of struct pxa2xx_udc_mach_info between multiple platforms
Move declaration of struct pxa2xx_udc_mach_info from
include/asm-arm/arch-pxa/udc.h to new file
include/asm-arm/mach/udc_pxa2xx.h.
This allow us to use this structure with
multiple platforms - pxa and ixp4xx. USB
device controller used in pxa25x is the same
as controller used in ixp4xx.
Signed-off-by: Milan Svoboda <msvoboda@ra.rockwell.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patrick McHardy [Wed, 29 Nov 2006 04:10:21 +0000 (20:10 -0800)]
[NETFILTER]: ipt_REJECT: fix memory corruption
On devices with hard_header_len > LL_MAX_HEADER ip_route_me_harder()
reallocates the skb, leading to memory corruption when using the stale
tcph pointer to update the checksum.
Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Yasuyuki Kozakai [Mon, 27 Nov 2006 18:26:46 +0000 (10:26 -0800)]
[NETFILTER]: conntrack: fix refcount leak when finding expectation
All users of __{ip,nf}_conntrack_expect_find() don't expect that
it increments the reference count of expectation.
Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp> Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Yasuyuki Kozakai [Mon, 27 Nov 2006 18:25:59 +0000 (10:25 -0800)]
[NETFILTER]: nf_conntrack: fix the race on assign helper to new conntrack
The found helper cannot be assigned to conntrack after unlocking
nf_conntrack_lock. This tries to find helper to assign again.
Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp> Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Yasuyuki Kozakai [Mon, 27 Nov 2006 18:25:32 +0000 (10:25 -0800)]
[NETFILTER]: nfctnetlink: assign helper to newly created conntrack
This fixes the bug which doesn't assign helper to newly created
conntrack via nf_conntrack_netlink.
Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp> Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Linus Torvalds [Wed, 29 Nov 2006 04:54:40 +0000 (20:54 -0800)]
Merge branch 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev
* 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev:
[PATCH] libata: Fixup ata_sas_queuecmd to handle __ata_scsi_queuecmd failure
[PATCH] ahci: AHCI mode SATA patch for Intel ICH9
[PATCH] libata: don't schedule EH on wcache on/off if old EH
Linus Torvalds [Wed, 29 Nov 2006 01:28:41 +0000 (17:28 -0800)]
Merge branch 'for-linus' of git://one.firstfloor.org/home/andi/git/linux-2.6
* 'for-linus' of git://one.firstfloor.org/home/andi/git/linux-2.6:
[PATCH] x86-64: Use stricter in process stack check for unwinder
[PATCH] i386: Fix compilation with UP genericarch
[PATCH] x86-64: Fix warning in io_apic.c
[PATCH] x86-64: work around gcc4 issue with -Os in Dwarf2 stack unwind
[PATCH] x86_64: Align data segment to PAGE_SIZE boundary
Linus Torvalds [Wed, 29 Nov 2006 01:27:52 +0000 (17:27 -0800)]
Merge branch 'linus' of master.kernel.org:/pub/scm/linux/kernel/git/perex/alsa
* 'linus' of master.kernel.org:/pub/scm/linux/kernel/git/perex/alsa:
[ALSA] version 1.0.13
[ALSA] snd-emu10k1: Fix capture for one variant.
[ALSA] Fix hang-up at disconnection of usb-audio
[ALSA] hda: fix typo for xw4400 PCI sub-ID
[ALSA] hda: fix sigmatel dell system detection
[ALSA] Enable stereo line input for TAS codec
[ALSA] rtctimer: handle RTC interrupts with a tasklet
Joakim Tjernlund [Tue, 28 Nov 2006 23:11:52 +0000 (23:11 +0000)]
[PATCH] Fix Intel/Sharp command set erase suspend bug
When we sleep and wait for a suspended operation to be resumed, go
back and check until it's ready -- don't just continue after the first
time we're woken. This can cause file system corruption.
Andi Kleen [Tue, 28 Nov 2006 19:12:59 +0000 (20:12 +0100)]
[PATCH] x86-64: Use stricter in process stack check for unwinder
Previously it would check for alignment only, which could break
if the stack pointer was unaligned. Now explicitely check if the
stack pointer is in the stack page of the current process.
Jan Beulich [Tue, 28 Nov 2006 19:12:59 +0000 (20:12 +0100)]
[PATCH] x86-64: work around gcc4 issue with -Os in Dwarf2 stack unwind
This fixes a problem with gcc4 mis-compiling the stack unwind code under
-Os, which resulted in 'stuck' messages whenever an assembly routine was
encountered.
Matt Porter [Mon, 6 Nov 2006 10:20:38 +0000 (11:20 +0100)]
[ALSA] hda: fix sigmatel dell system detection
Fixes Dell system detection on 9200 codecs. The support
to detect certain Dell machines was merged in the
9205 table where it will be unused on the various Dell
9200-based codec systems. This moves the subsystem IDs
to the correct 9200 table.
Signed-off-by: Matt Porter <mporter@embeddedalley.com> Signed-off-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Jaroslav Kysela <perex@suse.cz>
Akinobu Mita [Mon, 27 Nov 2006 06:16:48 +0000 (15:16 +0900)]
selinux: fix dentry_open() error check
The return value of dentry_open() shoud be checked by IS_ERR().
Cc: Stephen Smalley <sds@tycho.nsa.gov> Cc: James Morris <jmorris@namei.org> Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Signed-off-by: James Morris <jmorris@namei.org>
Linus Torvalds [Mon, 27 Nov 2006 03:05:22 +0000 (19:05 -0800)]
Fix 'ALIGN()' macro, take 2
You wouldn't think that doing an ALIGN() macro that aligns something up
to a power-of-two boundary would be likely to have bugs, would you?
But hey, in the wonderful world of mixing integer types, you have to be
careful. This just makes sure that the alignment is interpreted in the
same type as the thing to be aligned.
Thanks to Roland Dreier, who noticed that the amso1100 driver got broken
by the previous fix (that just extended the mask to "unsigned long", but
was still broken in "unsigned long long" - it just happened to be the
same on 64-bit architectures).
Acked-by: Roland Dreier <rdreier@cisco.com> Cc: Andrew Morton <akpm@osdl.org> Cc: David Miller <davem@davemloft.net> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Kyle McMartin [Sun, 26 Nov 2006 23:56:56 +0000 (18:56 -0500)]
[PATCH] Fix incorrent type of flags in <asm/semaphore.h>
I still think using BUILD_BUG_ON() is unacceptable, especially given how
vague the error message was.
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
[ And I already removed gthe BUILD_BUG_ON() in the previous commit ] Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6:
[XFRM] STATE: Fix to respond error to get operation if no matching entry exists.
[NET]: Re-fix of doc-comment in sock.h
[6PACK]: Masking bug in 6pack driver.
[NET]: Fix kfifo_alloc() error check.
[UDP]: Make udp_encap_rcv use pskb_may_pull
[NETFILTER]: H.323 conntrack: fix crash with CONFIG_IP_NF_CT_ACCT
Russell King [Sat, 25 Nov 2006 16:19:23 +0000 (16:19 +0000)]
[ARM] Add PM_LEGACY defaults
Eliminate two warnings:
kernel/power/pm.c:205: warning: 'pm_register' is deprecated (declared at kernel/power/pm.c:64)
kernel/power/pm.c:206: warning: 'pm_send_all' is deprecated (declared at kernel/power/pm.c:180)
by updating defconfig files to contain a sensible PM_LEGACY default.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Hans Verkuil [Sat, 25 Nov 2006 12:40:28 +0000 (09:40 -0300)]
V4L/DVB (4885): Improve saa711x check
The old code would accept any device on the same i2c address as the
saa711x chips as an saa711x. However, this fails with saa717x chips,
which use that same address and so are misdetected as a saa7111. Now
check whether the chip is really a saa711x model.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Ira Snyder [Mon, 20 Nov 2006 10:20:48 +0000 (07:20 -0300)]
V4L/DVB (4849): Add missing spin_unlock to saa6588 decoder driver
Sparse noticed a lock imbalance in read_from_buf(). Further inspection shows
that the lock should not be held when the function exits.
This adds a spin_unlock_irqrestore(), so that every exit path of the
read_from_buf() function is consistent. The unlock was missing on an error
path.
Signed-off-by: Ira W. Snyder <devel@irasnyder.com> Signed-off-by: Hans J. Koch <koch@hjk-az.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
Oliver Endriss [Sun, 19 Nov 2006 05:15:37 +0000 (02:15 -0300)]
V4L/DVB (4840): Budget: diseqc_method module parameter for cards with subsystem-id 13c2:1003
New module parameter diseqc_method for cards with subsystem-id 13c2:1003.
- 0: unreliable method, can be used by all board revisions (default)
- 1: reliable method, works for newer board layouts only
The parameter has no effect for cards with other subsystem-ids.
Signed-off-by: Oliver Endriss <o.endriss@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
[XFRM] STATE: Fix to respond error to get operation if no matching entry exists.
When application uses XFRM_MSG_GETSA to get state entry through
netlink socket and kernel has no matching one, the application expects
reply message with error status by kernel.
Kernel doesn't send the message back in the case of Mobile IPv6 route
optimization protocols (i.e. routing header or destination options
header). This is caused by incorrect return code "0" from
net/xfrm/xfrm_user.c(xfrm_user_state_lookup) and it makes kernel skip
to acknowledge at net/netlink/af_netlink.c(netlink_rcv_skb).
This patch fix to reply ESRCH to application.
Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org> Signed-off-by: TAKAMIYA Noriaki <takamiya@po.ntts.co.jp> Signed-off-by: David S. Miller <davem@davemloft.net>
Paul Bonser [Fri, 24 Nov 2006 01:56:13 +0000 (17:56 -0800)]
[NET]: Re-fix of doc-comment in sock.h
Restoring old, correct comment for sk_filter_release, moving it to
where it should actually be, and changing new comment into proper
comment for sk_filter_rcu_free, where it actually makes sense.
The original fix submitted for this on Oct 23 mistakenly documented
the wrong function.
Signed-off-by: Paul Bonser <misterpib@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jean Delvare [Thu, 23 Nov 2006 19:48:28 +0000 (11:48 -0800)]
[6PACK]: Masking bug in 6pack driver.
Looks like a broken masking to me, binary not is used where bitwise
not was intended.
Signed-off-by: Jean Delvare <khali@linux-fr.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Olaf Kirch [Thu, 23 Nov 2006 04:11:42 +0000 (20:11 -0800)]
[UDP]: Make udp_encap_rcv use pskb_may_pull
Make udp_encap_rcv use pskb_may_pull
IPsec with NAT-T breaks on some notebooks using the latest e1000 chipset,
when header split is enabled. When receiving sufficiently large packets, the
driver puts everything up to and including the UDP header into the header
portion of the skb, and the rest goes into the paged part. udp_encap_rcv
forgets to use pskb_may_pull, and fails to decapsulate it. Instead, it
passes it up it to the IKE daemon.
Signed-off-by: Olaf Kirch <okir@suse.de> Signed-off-by: Jean Delvare <jdelvare@suse.de> Signed-off-by: David S. Miller <davem@davemloft.net>
Faidon Liambotis [Wed, 22 Nov 2006 05:46:02 +0000 (21:46 -0800)]
[NETFILTER]: H.323 conntrack: fix crash with CONFIG_IP_NF_CT_ACCT
H.323 connection tracking code calls ip_ct_refresh_acct() when
processing RCFs and URQs but passes NULL as the skb.
When CONFIG_IP_NF_CT_ACCT is enabled, the connection tracking core tries
to derefence the skb, which results in an obvious panic.
A similar fix was applied on the SIP connection tracking code some time
ago.
Signed-off-by: Faidon Liambotis <paravoid@debian.org> Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Reimplement execvp for our purposes - after we call fork() it is fundamentally
unsafe to use the kernel allocator - current is not valid there. So we simply
pass to our modified execvp() a preallocated buffer. This fixes a real bug
and works very well in testing (I've seen indirectly warning messages from the
forked thread - they went on the pipe connected to its stdout and where read
as a number by UML, when calling read_output(). I verified the obtained
number corresponded to "BUG:").
The added use of __cant_sleep() is not a new bug since __cant_sleep() is
already used in the same function - passing an atomicity parameter would be
better but it would require huge change, stating that this function must not
be called in atomic context and can sleep is a better idea (will make sure of
this gradually).
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it> Acked-by: Jeff Dike <jdike@addtoit.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[akpm@osdl.org: two compile fixes] Signed-off-by: Arjan van de Ven <arjan@linux.intel.com> Signed-off-by: Jiri Kosina <jikos@jikos.cz> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Alexey Dobriyan [Sat, 25 Nov 2006 19:09:36 +0000 (11:09 -0800)]
[PATCH] Enforce "unsigned long flags;" when spinlocking
Make it break or warn if you pass to spin_lock_irqsave() and friends
something different from "unsigned long flags;". Suprisingly large amount
of these was caught by recent commit c53421b18f205c5f97c604ae55c6a921f034b0f6 and others.
Idea is largely from FRV typechecking. Suggestions from Andrew Morton.
All stupid typos in first version fixed.
Passes allmodconfig on i386, x86_64, alpha, arm as well as my usual config.
Note #1: checking with sparse is still needed, because a driver can save
and pass around flags or something. So far patch is very intrusive.
Note #2: techically, we should break only if
sizeof(flags) < sizeof(unsigned long),
however, the more pain for getting suspicious code into kernel,
the better.
Jeremy Higdon [Sat, 25 Nov 2006 19:09:33 +0000 (11:09 -0800)]
[PATCH] sgiioc4: Disable module unload
This patch removes a module_exit function that sgiioc4 should not have had.
It seems that the IDE layer doesn't support submodule unloading. sgiioc4 was
the only driver in drivers/ide/pci that had an exit function. After an
unload, the devices would stay around and the next attempt to reference would
crash...
Signed-off-by: Jeremy Higdon <jeremy@sgi.com> Acked-by: "Bartlomiej Zolnierkiewicz" <bzolnier@gmail.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Roman Zippel [Sat, 25 Nov 2006 19:09:32 +0000 (11:09 -0800)]
[PATCH] fix menuconfig colours with TERM=vt100
On Mon, 13 Nov 2006, Phil Oester wrote:
> In commit 350b5b76384e77bcc58217f00455fdbec5cac594, the default menuconfig
> color scheme was changed to bluetitle. This breaks the highlighting
> of the selected item for me with TERM=vt100. The only way I can see
> which item is selected is via:
>
> make MENUCONFIG_COLOR=mono menuconfig
>
> Which restores the pre-2.6.19 white on black highlighting.
Fix.
Cc: Phil Oester <kernel@linuxace.com> 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>
David Brownell [Sat, 25 Nov 2006 19:09:28 +0000 (11:09 -0800)]
[PATCH] rtc class locking bugfixes
I got a lockdep warning when running "rtctest" so I though it'd be good
to see what was up.
- The warning was for rtc->irq_task_lock, gotten from rtc_update_irq()
by irq handlerss ... but in a handful of other cases, grabbed without
blocking IRQs.
- Some callers to rtc_update_irq() were not ensuring IRQs were blocked,
yet the routine expects that; make sure all callers block IRQs.
It would appear that RTC API tests haven't been part of anyone's kernel
regression test suite recently, at least not with lockdep running.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Acked-by: Alessandro Zummo <a.zummo@towertech.it> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
David Brownell [Sat, 25 Nov 2006 19:09:27 +0000 (11:09 -0800)]
[PATCH] rtc framework handles periodic irqs
The RTC framework has an irq_set_freq() method that should be used to manage
the periodic IRQ frequency, but the current ioctl logic doesn't know how to do
that. This patch teaches it how.
This means that drivers implementing irq_set_freq() will automatically support
RTC_IRQP_{READ,SET} ioctls; that logic doesn't need duplication within the
driver.
[akpm@osdl.org: export rtc_irq_set_freq] Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Acked-by: Alessandro Zummo <a.zummo@towertech.it> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This updates the RTC documentation to summarize the two APIs now available:
the old PC/AT one, and the new RTC class drivers. It also updates the
included "rtctest.c" file to better meet Linux style guidelines, and to work
with the new RTC drivers.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Acked-by: Alessandro Zummo <a.zummo@towertech.it> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Miklos Szeredi [Sat, 25 Nov 2006 19:09:20 +0000 (11:09 -0800)]
[PATCH] fuse: fix Oops in lookup
Fix bug in certain error paths of lookup routines. The request object was
reused for sending FORGET, which is illegal. This bug could cause an Oops
in 2.6.18. In earlier versions it might silently corrupt memory, but this
is very unlikely.
These error paths are never triggered by libfuse, so this wasn't noticed
even with the 2.6.18 kernel, only with a filesystem using the raw kernel
interface.
Thanks to Russ Cox for the bug report and test filesystem.
Thomas Chou [Sat, 25 Nov 2006 19:09:18 +0000 (11:09 -0800)]
[PATCH] initramfs: handle more than one source dir or file list
Fix bug 7401.
Handle more than one source dir or file list to the initramfs gen scripts.
The Kconfig help for INITRAMFS_SOURCE claims that you can specify multiple
space-separated sources in order to allow unprivileged users to build an
image. There are two bugs in the current implementation that prevent this
from working.
First, we pass "file1 dir2" to the gen_initramfs_list.sh script, which it
obviously can't open.
Second, gen_initramfs_list.sh -l outputs multiple definitions for
deps_initramfs -- one for each argument.
Signed-off-by: Thomas Chou <thomas@wytron.com.tw> Cc: Sam Ravnborg <sam@ravnborg.org> Acked-by: Matthew Wilcox <matthew@wil.cx> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Mel Gorman [Thu, 23 Nov 2006 12:01:41 +0000 (12:01 +0000)]
[PATCH] x86_64: fix bad page state in process 'swapper'
find_min_pfn_for_node() and find_min_pfn_with_active_regions() both
depend on a sorted early_node_map[]. However, sort_node_map() is being
called after fin_min_pfn_with_active_regions() in
free_area_init_nodes().
In most cases, this is ok, but on at least one x86_64, the SRAT table
caused the E820 ranges to be registered out of order. This gave the
wrong values for the min PFN range resulting in some pages not being
initialised.
This patch sorts the early_node_map in find_min_pfn_for_node(). It has
been boot tested on x86, x86_64, ppc64 and ia64.
Dave Jones [Thu, 23 Nov 2006 01:42:01 +0000 (20:42 -0500)]
[PATCH] Correct bound checking from the value returned from _PPC method.
processor_perflib.c::acpi_processor_ppc_notifier() check if the value
returned by the processor's _PPC method is 0 and return failed if so.
This is wrong since 0 indicate that the bios think the processor can go
to the highest frequency. This patch for example fix the HP NX 6125 to
allow its highest frequency to be available.
Signed-off-by: Bruno Ducrot <ducrot@poupinou.org> Cc: "Pallipadi, Venkatesh" <venkatesh.pallipadi@intel.com> Signed-off-by: Dave Jones <davej@redhat.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Linus Torvalds [Thu, 23 Nov 2006 17:17:16 +0000 (09:17 -0800)]
Merge branch 'for-linus' of master.kernel.org:/home/rmk/linux-2.6-arm
* 'for-linus' of master.kernel.org:/home/rmk/linux-2.6-arm:
[ARM] 3941/1: [Jornada7xx] - Addition to MAINTAINERS
[ARM] 3942/1: ARM: comment: consistent_sync should not be called directly
[ARM] ebsa110: fix warnings generated by asm/arch/io.h
[ARM] 3933/1: Source drivers/ata/Kconfig
Manuel Lauss [Wed, 22 Nov 2006 13:51:32 +0000 (14:51 +0100)]
[PATCH] make au1xxx-ide compile again
The Au1xx IDE controller driver doesn't compile:
CC drivers/ide/mips/au1xxx-ide.o
/linux-2.6.19-rc6-work/drivers/ide/mips/au1xxx-ide.c:480: error: conflicting types for 'auide_ddma_tx_callback'
include2/asm/mach-au1x00/au1xxx_ide.h:174: error: previous declaration of 'auide_ddma_tx_callback' was here
/linux-2.6.19-rc6-work/drivers/ide/mips/au1xxx-ide.c:486: error: conflicting types for 'auide_ddma_rx_callback'
include2/asm/mach-au1x00/au1xxx_ide.h:176: error: previous declaration of 'auide_ddma_rx_callback' was here
Signed-off-by: Manuel Lauss <mano@roarinelk.homelinux.net> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Thiemo Seufer [Thu, 16 Nov 2006 22:13:54 +0000 (22:13 +0000)]
[MIPS] Hack for SB1 cache issues
Removing flush_icache_page a while ago broke SB1 which was using an empty
flush_data_cache_page function. This glues things well enough so a more
efficient but also more intrusive solution can be found later.
Linus Torvalds [Wed, 22 Nov 2006 22:55:29 +0000 (14:55 -0800)]
[AGP] Allocate AGP pages with GFP_DMA32 by default
Not all graphic page remappers support physical addresses over the 4GB
mark for remapping, so while some do (the AMD64 GART always did, and I
just fixed the i965 to do so properly), we're safest off just forcing
GFP_DMA32 allocations to make sure graphics pages get allocated in the
low 32-bit address space by default.
AGP sub-drivers that really care, and can do better, could just choose
to implement their own allocator (or we could add another "64-bit safe"
default allocator for their use), but quite frankly, you're not likely
to care in practice.
So for now, this trivial change means that we won't be allocating pages
that we can't map correctly by mistake on x86-64.
[ On traditional 32-bit x86, this could never happen, because GFP_KERNEL
would never allocate any highmem memory anyway ]
Acked-by: Andi Kleen <ak@suse.de> Acked-by: Dave Jones <davej@redhat.com> Cc: Eric Anholt <eric@anholt.net> Cc: Keith Packard <keithp@keithp.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Dan Williams [Tue, 21 Nov 2006 21:57:23 +0000 (22:57 +0100)]
[ARM] 3942/1: ARM: comment: consistent_sync should not be called directly
/*
* Note: Drivers should NOT use this function directly, as it will break
* platforms with CONFIG_DMABOUNCE.
* Use the driver DMA support - see dma-mapping.h (dma_sync_*)
*/
Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Linus Torvalds [Wed, 22 Nov 2006 17:37:54 +0000 (09:37 -0800)]
[AGP] Fix intel 965 AGP memory mapping function
This introduces a i965-specific "mask_memory()" function that knows
about the extended physical addresses that the i965 supports. This
allows us to correctly map in physical memory in the >4GB range into the
GTT.
Also simplify/clean-up the i965 case for the aperture sizing by just
returning the fixed 512kB size from "fetch_size()". We don't really
care that not all of the aperture may be visible - the only thing that
cares about the aperture size is the Intel "stolen memory" calculation,
which depends on the fixed size.
Cc: Keith Packard <keithp@keithp.com> Cc: Eric Anholt <eric@anholt.net> Cc: Dave Jones <davej@redhat.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Linus Torvalds [Wed, 22 Nov 2006 17:32:06 +0000 (09:32 -0800)]
Don't call "note_interrupt()" with irq descriptor lock held
This reverts commit f72fa707604c015a6625e80f269506032d5430dc, and solves
the problem that it tried to fix by simply making "__do_IRQ()" call the
note_interrupt() function without the lock held, the way everybody else
does.
It should be noted that all interrupt handling code must never allow the
descriptor actors to be entered "recursively" (that's why we do all the
magic IRQ_PENDING stuff in the first place), so there actually is
exclusion at that much higher level, even in the absense of locking.
projects / linux-2.6 / log