Page faults in kernel address space between PAGE_OFFSET up to VMALLOC_START should not try to map as vmalloc. Fix rarely endless page faults inside mount_block_root for root filesystem at boot time. Signed-off-by: Henry Nestler <henry.nestler@gmail.com> --- All 32bit kernels up to 2.6.25 can fail into this hole. I can not present this under native linux kernel. I see, that the 64bit has fixed the problem. I copied the same lines into 32bit part. Recorded debugs are from coLinux kernel...

It seems that commit 305d2a1a ("libata: unify mechanism to request follow-up SRST") causes sata_nv not to see the HD on a system I have here; with current git (80750147), I get this in my bootlog: [ 2.425108] Driver 'sd' needs updating - please use bus_type methods [ 2.435751] ACPI: PCI Interrupt Link [LSA0] enabled at IRQ 23 [ 2.441508] ACPI: PCI Interrupt 0000:00:05.0[A] -> Link [LSA0] -> GSI 23 (level, low) -> IRQ 23 [ 2.452929] sata_nv 0000:00:05.0: Using SWNCQ mode [ 2....

Hello Linux kernel crew, [Consider this more as a datapoint then a bug report, as after one network and one sata/southbridge issues showing up interminnently, the ASRock motherboard involved will be scrapped for a different one] The integrated NVidia sata controller and/or the hard-drive has failed during operation with the following output: Apr 22 23:36:54 backupserver kernel: [91202.294632] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (t...

Running 'make oldconfig' I just noticed that PCIEASPM defaults to 'y' in Kconfig even though the feature is both experimental and the help text recommends that if you are unsure you say 'n'. It seems to me that this really should default to 'n', not 'y' at the moment. The following patch makes that change. Please consider applying. Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com> --- Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/pcie/...

Quite frankly, I'd actually prefer to just reinstate "dev" to the irq registration instead. Why? Because that field is how we are able to track multiple interrupt registrations that share an IRQ. Now, the "request_irq()" logic has a special rule that actually tests that NULL is a valid cookie if the IRQF_SHARED bit isn't set, but isn't it a nice thing to double-check This, in contrast, *really* sucks as a cookie. In fact, it's useless. If there are multiple tp3780i instances on the sa...

Ok, so using the pointer to inside a specific pSettings field is fine. But can you also explain to me why that insane driver does this: static irqreturn_t UartInterrupt(int irq, void *dev_id) { - int irqno = (int)(unsigned long) dev_id; + unsigned short *irqno = dev_id; ... *irqno, dev_id); instead of just ignoring "dev_id" entirely, and then just using that "irq" argument directly? Linus --

Ok, that's just not going to happen. What's the upside? Really? I can tell you the downsides: - tons of huge patches with ugly churn - total disaster when it comes to any drivers that are maintained over multiple versions and/or out-of-tree like the DRI stuff so those upsides had better be really big. Linus --

Hi all, I dont't know where this device is comming from and what it does or is supposed to do. Hardware: Asus P5W-DH Delux Kernel: Stock 2.6.25 Distribution: Gentoo Linux See attached files for more information. Maybe related to: http://bugzilla.kernel.org/show_bug.cgi?id=9010 http://bugzilla.kernel.org/show_bug.cgi?id=8923 can anybody comment on this: [ 1.252722] ata2: SATA max UDMA/133 irq_stat 0x00400040, connection status changed irq 1274 [ 4.783300] ata2: SATA link ...

From: Jesper Juhl <jesper.juhl@gmail.com> The Coverity checker spotted that we do not check the return value of match_strdup() in fs/hfsplus/options.c::hfsplus_parse_options(). This is bad since match_strdup() does a memory allocation internally which can fail. If it does fail it will return NULL and in that case we will pass the NULL pointer on to load_nls() which will eventually dereference it --> BOOM! Much better to check the return value, fail gracefully and log an error message in c...

Using different strings in the printk wastes memory. --

Hello, I have released linux 0.01-3.3 on january 1, 2008. And I often be asked 2 question : the 1st was, "I can't build linux-0.01-3.3, why ?" =================================== Well linux 0.01 was very well written, except 1 thing in the Makefile : To get file size of the kernel image, Linus Torvalds used the following ( and ugly ) command : "ls -l tools/system | grep system | cut -c31-37" which may be invalid for you. So the unofficial linux-0.01-3.5 ( which was released on the end of ...

* Replace usages of MAX_NUMNODES with nr_node_ids in kernel/sched.c, where appropriate. This saves some allocated space as well as many wasted cycles going through node entries that are non-existent. For inclusion into sched-devel/latest tree. Based on: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git + sched-devel/latest .../mingo/linux-2.6-sched-devel.git Signed-off-by: Mike Travis <travis@sgi.com> --- kernel/sched.c | 18 +++++++++--------- ...

* Introduce a new PER_CPU macro called "EARLY_PER_CPU". This is used by some per_cpu variables that are initialized and accessed before there are per_cpu areas allocated. ["Early" in respect to per_cpu variables is "earlier than the per_cpu areas have been setup".] This patchset adds these new macros: DEFINE_EARLY_PER_CPU(_type, _name, _initvalue) EXPORT_EARLY_PER_CPU_SYMBOL(_name) DECLARE_EARLY_PER_CPU(_type, _name) early_per_cpu_ptr(_name) early_per_cpu_map(_n...

* Increase the limit of NR_CPUS to 4096 and introduce a boolean called "MAXSMP" which when set (e.g. "allyesconfig"), will set NR_CPUS = 4096 and NODES_SHIFT = 9 (512). Changed max setting for NODES_SHIFT from 15 to 9 to accurately reflect the real limit. * Introduce a new PER_CPU macro called "EARLY_PER_CPU". This is used by some per_cpu variables that are initialized and accessed before there are per_cpu areas allocated. * Add a flag "arch_provides_topology_pointer...

Limit this driver to ARCH_PXA, noticed by breakage on allyesconfig builds on 32-bit x86. Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> --- Linus, please apply and Russell can fix it up however he likes later in the week. Many people are hitting this. drivers/mfd/Kconfig | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig index 2566479..bd29c61 100644 --- a/drivers/mfd/Kconfig +++ b/drivers/mfd/Kconfig @@ -32...

Hmm. While I in general support the notion of trying to compile drivers on as wide a variety as hardware as possible, if that HTC_PASIC3 thing really is a PXA-only piece of hardware, I don't really see the point of not making the config file accurately represent that. That said, as far as I can tell, the compile failure is because of this line: #include <asm/arch/pxa-regs.h> and I cannot for the life of me see _why_ it tries to include that header file. It seems to compile fi...

I really don't see what you are talking about, and nobody else has that problem. With patches, once you send them, they are sent, and you can't fix it. With git, once you send my the "please pull", it's sent, and you can't fix it. And with either, you can update the queue later. I don't understand AT ALL why you think they are different. In fact, if anything, git trees are a lot more flexible. With git, what you can do is fix up the tree at any time, and send me an email saying "...

Well, first I need to gain the permission of Andrew to post his private message. I'm not being subborn here - I _do_ _not_ reproduce private Talk to Andrew then. -- Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of: --

In which case, since it's likely I won't get a reply in the next hour (which'll delay my response by 24 hours) let me paraphrase what Andrew said. Andrew believes that it is beneficial to have other architectures, particularly x86, build other architectures drivers. Meanwhile, pHilipp Zabel believes this hardware not to be ARM specific. So, the majority concensus in the three way discussion was that it should remain visible, and the (unnecessary) include and its dependents be removed. Hav...

Hello, The following patch fixes a [probable] copy & paste mistake in airprime.c. Instead of unlocking an acquired mutex, the actual code tries to lock it again. Signed-off-by: Leonardo Chiquitto <lchiquitto@novell.com> Cc: Greg Kroah-Hartman <gregkh@suse.de> --- Leonardo Index: kernel/drivers/usb/serial/airprime.c =================================================================== --- kernel.orig/drivers/usb/serial/airprime.c +++ kernel/drivers/usb/serial/airprime.c @@...

This set of four patches moves XPC and XPNET to drivers/misc/sgi-xp in preparation for enabling SGI UV X86_64 support. This version is prompted by reported build problems with certain configs, and a desire to keep things simple by only moving XPC/XPNET as is and only addressing Lindent and checkpatch.pl issues (not x86_64 issues) at this time. --

Ran patches through scripts/Lindent (part 2). Signed-off-by: Dean Nelson <dcn@sgi.com> --- drivers/misc/sgi-xp/xpc_channel.c | 390 ++++++++++++---------------------- drivers/misc/sgi-xp/xpc_main.c | 329 ++++++++++++---------------- drivers/misc/sgi-xp/xpnet.c | 109 +++------ 3 files changed, 322 insertions(+), 506 deletions(-) Index: linux-2.6/drivers/misc/sgi-xp/xpc_channel.c =================================================================== --- linux-2.6.orig/driver...

Move XPC and XPNET from arch/ia64/sn/kernel to drivers/misc/sgi-xp. Signed-off-by: Dean Nelson <dcn@sgi.com> --- arch/ia64/Kconfig | 11 ----------- arch/ia64/sn/kernel/Makefile | 7 +------ drivers/misc/Kconfig | 12 ++++++++++++ drivers/misc/Makefile | 1 + drivers/misc/sgi-xp/Makefile | 11 +++++++++++ {include/asm-ia64/sn => drivers/m...

I've now dropped the change to the uncached allocator. You'll need to update this patch with the changes to the users of uncached_{alloc,free}_page. Using the older one-page version your patches to move XP and XPC all apply, and I see no build errors (I don't have a system on which to check that this works). -Tony --

This removes the special-case handling for PTRACE_GETSIGINFO and PTRACE_SETSIGINFO from x86_64's sys32_ptrace. The generic compat_ptrace_request code handles these. Signed-off-by: Roland McGrath <roland@redhat.com> --- arch/x86/kernel/ptrace.c | 30 +----------------------------- 1 files changed, 1 insertions(+), 29 deletions(-) diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 1edbb13..b564116 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c ...