Insert the Local APIC and IO APIC(s) into the resource tree. It allows the
APIC resources to be visible within /proc/iomem. The patch also takes into
account IO APIC(s) mapped in the PCI space by deferring the insertion until
after PCI has allocated its necessary resources.
Signed-off-by: Aaron Durbin <adurbin@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@muc.de>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
When a spinlock lockup occurs, arrange for the NMI code to emit an all-cpu
backtrace, so we get to see which CPU is holding the lock, and where.
Cc: Andi Kleen <ak@muc.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Here is a patch (used by perfmon2) to detect the presence of the
Precise Event Based Sampling (PEBS) feature for Intel 64-bit processors.
The patch also adds the cpu_has_pebs macro.
changelog:
- adds X86_FEATURE_PEBS
- adds cpu_has_pebs to test for X86_FEATURE_PEBS
Signed-off-by: stephane eranian <eranian@hpl.hp.com>
Signed-off-by: Andi Kleen <ak@suse.de>
The unwinder has some extra newlines, which eat up loads of screen
space when it spews. (See https://bugzilla.redhat.com/bugzilla/attachment.cgi?id=137900
for a nasty example).
warning_symbol-> and warning-> already printk a newline, so don't add one
in the strings passed to them.
[AK: redone for new code]
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
* sanitize prototypes, annotate
* usual ntohs->shift
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
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.
Ported from i386.
Signed-off-by: Andi Kleen <ak@suse.de>
Commit 2c8c0e6b8d ("[PATCH] Convert x86-64
to early param") broke the earlyprintk=...,keep feature.
This restores that functionality. Tested on x86_64. Must-have for
v2.6.19, no risk.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Pass the work_struct pointer to the work function rather than context data.
The work function can use container_of() to work out the data.
For the cases where the container of the work_struct may go away the moment the
pending bit is cleared, it is made possible to defer the release of the
structure by deferring the clearing of the pending bit.
To make this work, an extra flag is introduced into the management side of the
work_struct. This governs auto-release of the structure upon execution.
Ordinarily, the work queue executor would release the work_struct for further
scheduling or deallocation by clearing the pending bit prior to jumping to the
work function. This means that, unless the driver makes some guarantee itself
that the work_struct won't go away, the work function may not access anything
else in the work_struct or its container lest they be deallocated.. This is a
problem if the auxiliary data is taken away (as done by the last patch).
However, if the pending bit is *not* cleared before jumping to the work
function, then the work function *may* access the work_struct and its container
with no problems. But then the work function must itself release the
work_struct by calling work_release().
In most cases, automatic release is fine, so this is the default. Special
initiators exist for the non-auto-release case (ending in _NAR).
Signed-Off-By: David Howells <dhowells@redhat.com>
Separate delayable work items from non-delayable work items be splitting them
into a separate structure (delayed_work), which incorporates a work_struct and
the timer_list removed from work_struct.
The work_struct struct is huge, and this limits it's usefulness. On a 64-bit
architecture it's nearly 100 bytes in size. This reduces that by half for the
non-delayable type of event.
Signed-Off-By: David Howells <dhowells@redhat.com>
o Explicitly align data segment to PAGE_SIZE boundary otherwise depending on
config options and tool chain it might be placed on a non PAGE_SIZE aligned
boundary and vmlinux loaders like kexec fail when they encounter a
PT_LOAD type segment which is not aligned to PAGE_SIZE boundary.
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
o Explicitly align data segment to PAGE_SIZE boundary otherwise depending on
config options and tool chain it might be placed on a non PAGE_SIZE aligned
boundary and vmlinux loaders like kexec fail when they encounter a
PT_LOAD type segment which is not aligned to PAGE_SIZE boundary.
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
This is to fix compile error of x86-64 memory hotplug without any NUMA
option.
CC arch/x86_64/mm/init.o
arch/x86_64/mm/init.c:501: error: redefinition of 'memory_add_physaddr_to_nid'
include/linux/memory_hotplug.h:71: error: previous definition of 'memory_add_phys
addr_to_nid' was here
arch/x86_64/mm/init.c:509: error: redefinition of 'memory_add_physaddr_to_nid'
arch/x86_64/mm/init.c:501: error: previous definition of 'memory_add_physaddr_to_
nid' was here
I confirmed compile completion with !NUMA, (NUMA & !ACPI_NUMA),
or (NUMA & ACPI_NUMA).
Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Acked-by: Andi Kleen <ak@suse.de>
Cc: "Randy.Dunlap" <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The scheduler on Andreas Friedrich's hyperthreading system stopped
working properly: the scheduler would never move tasks to another CPU!
The lask known working kernel was 2.6.8.
After a couple of attempts to corner the bug, the following smoking gun
was found:
BIOS reported wrong ACPI idfor the processor
CPU#1: set_cpus_allowed(), swapper:1, 3 -> 2
[<c0103bbe>] show_trace_log_lvl+0x34/0x4a
[<c0103ceb>] show_trace+0x2c/0x2e
[<c01045f8>] dump_stack+0x2b/0x2d
[<c0116a77>] set_cpus_allowed+0x52/0xec
[<c0101d86>] cpu_idle_wait+0x2e/0x100
[<c0259c57>] acpi_processor_power_exit+0x45/0x58
[<c0259752>] acpi_processor_remove+0x46/0xea
[<c025c6fb>] acpi_start_single_object+0x47/0x54
[<c025cee5>] acpi_bus_register_driver+0xa4/0xd3
[<c04ab2d7>] acpi_processor_init+0x57/0x77
[<c01004d7>] init+0x146/0x2fd
[<c0103a87>] kernel_thread_helper+0x7/0x10
a quick look at cpu_idle_wait() shows how broken that code is
on i386: it changes the init task's affinity map but never
restores it ...
and because all userspace tasks get forked by init, they all
inherited that single-CPU affinity mask. x86_64 cloned this
bug too.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Andreas Friedrich <andreas.friedrich@fujitsu-siemens.com>
Cc: Wolfgang Erig <Wolfgang.Erig@fujitsu-siemens.com>
Cc: Andrew Morton <akpm@osdl.org>
Cc: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
the new dwarf2 unwinder crashes while trying to dump the stack:
Leftover inexact backtrace:
Unable to handle kernel paging request at ffffffff82800000 RIP:
[<ffffffff8026cf26>] dump_trace+0x35b/0x3d2
PGD 203027 PUD 205027 PMD 0
Oops: 0000 [2] PREEMPT SMP
CPU 0
Modules linked in:
Pid: 30, comm: khelper Not tainted 2.6.19-rc6-rt1 #11
RIP: 0010:[<ffffffff8026cf26>] [<ffffffff8026cf26>] dump_trace+0x35b/0x3d2
RSP: 0000:ffff81003fb9d848 EFLAGS: 00010006
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
RDX: 0000000000000000 RSI: ffffffff805b3520 RDI: 0000000000000000
RBP: ffffffff827ffff9 R08: ffffffff80aad000 R09: 0000000000000005
R10: ffffffff80aae000 R11: ffffffff8037961b R12: ffff81003fb9d858
R13: 0000000000000000 R14: ffffffff80598460 R15: ffffffff80ab1fc0
FS: 0000000000000000(0000) GS:ffffffff806c4200(0000) knlGS:0000000000000000
CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b
CR2: ffffffff82800000 CR3: 0000000000201000 CR4: 00000000000006e0
this crash happened because it did not sanitize the dwarf2 data it
got, and got an unaligned stack pointer - which happily walked past
the process stack (and eventually reached the end of kernel memory
and pagefaulted there) due to this naive iteration condition:
HANDLE_STACK (((long) stack & (THREAD_SIZE-1)) != 0);
note that i386 is alot more conservative when it comes to trusting
stack pointers:
static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
{
return p > (void *)tinfo &&
p < (void *)tinfo + THREAD_SIZE - 3;
}
but the x86_64 code did not take this bit of i386 code.
The fix is to align the stack pointer.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Andi Kleen <ak@suse.de>
Cc: Jan Beulich <jbeulich@novell.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
on x86_64, the CONFIG_CC_STACKPROTECTOR build fails if used in a
distcc setup that has "CC" defined to "distcc gcc":
gcc: gcc: linker input file unused because linking not done
gcc: gcc: linker input file unused because linking not done
gcc: gcc: linker input file unused because linking not done
this is because the gcc-x86_64-has-stack-protector.sh script
has a 2-parameters assumption. Fix this by passing $(CC) as
a single parameter.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Please-Use-Me-More: make randconfig
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Komuro reports that ISA interrupts do not work after a disable_irq(),
causing some PCMCIA drivers to not work, with messages like
eth0: Asix AX88190: io 0x300, irq 3, hw_addr xx:xx:xx:xx:xx:xx
eth0: found link beat
eth0: autonegotiation complete: 100baseT-FD selected
eth0: interrupt(s) dropped!
eth0: interrupt(s) dropped!
eth0: interrupt(s) dropped!
...
Linus Torvalds <torvalds@osdl.org> said:
"Now, edge-triggered interrupts are a _lot_ harder to mask, because the
Intel APIC is an unbelievable piece of sh*t, and has the edge-detect logic
_before_ the mask logic, so if a edge happens _while_ the device is
masked, you'll never ever see the edge ever again (unmasking will not
cause a new edge, so you simply lost the interrupt).
So when you "mask" an edge-triggered IRQ, you can't really mask it at all,
because if you did that, you'd lose it forever if the IRQ comes in while
you masked it. Instead, we're supposed to leave it active, and set a flag,
and IF the IRQ comes in, we just remember it, and mask it at that point
instead, and then on unmasking, we have to replay it by sending a
self-IPI."
This trivial patch solves the problem.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Ingo Molnar <mingo@redhat.com>
Acked-by: Komuro <komurojun-mbn@nifty.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When another interrupt happens in exit_idle the exit idle notifier
could be called an incorrect number of times.
Add a test_and_clear_bit_pda and use it handle the bit
atomically against interrupts to avoid this.
Pointed out by Stephane Eranian
Signed-off-by: Andi Kleen <ak@suse.de>
The vgetcpu per CPU initialization previously relied on CPU hotplug
events for all CPUs to initialize the per CPU state. That only
worked only on kernels with CONFIG_HOTPLUG_CPU enabled. On the
others some CPUs didn't get their state initialized properly
and vgetcpu wouldn't work.
Change the initialization sequence to instead run in a normal
initcall (which runs after the normal CPU bootup) and initialize
all running CPUs there. Later hotplug CPUs are still handled
with an hotplug notifier.
This actually simplifies the code somewhat.
Signed-off-by: Andi Kleen <ak@suse.de>
Timer overrides are normally disabled on Nvidia board because
they are commonly wrong, except on new ones with HPET support.
Unfortunately there are quite some Asus boards around that
don't have HPET, but need a timer override.
We don't know yet how to handle this transparently,
but at least add a command line option to force the timer override
and let them boot.
Cc: len.brown@intel.com
Signed-off-by: Andi Kleen <ak@suse.de>
x86_64: setup saved_max_pfn correctly
2.6.19-rc4 has broken CONFIG_CRASH_DUMP support on x86_64. It is impossible
to read out the kernel contents from /proc/vmcore because saved_max_pfn is set
to zero instead of the max_pfn value before the user map is setup.
This happens because saved_max_pfn is initialized at parse_early_param() time,
and at this time no active regions have been registered. save_max_pfn is setup
from e820_end_of_ram(), more exact find_max_pfn_with_active_regions() which
returns 0 because no regions exist.
This patch fixes this by registering before and removing after the call
to e820_end_of_ram().
Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
This can happen on kexec kernels with some configurations, in particularly
on Unisys ES7000 systems.
Analysis by Amul Shah
Cc: Amul Shah <amul.shah@unisys.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Stephen Tweedie, Herbert Xu, and myself have been struggling with a very
nasty bug in Xen. But it also pointed out a small bug in the x86_64
kernel boot setup.
The GDT limit being setup by the initial bzImage code when entering into
protected mode is way too big. The comment by the code states that the
size of the GDT is 2048, but the actual size being set up is much bigger
(32768). This happens simply because of one extra '0'.
Instead of setting up a 0x800 size, 0x8000 is set up. On bare metal this
is fine because the CPU wont load any segments unless they are
explicitly used. But unfortunately, this breaks Xen on vmx FV, since it
(for now) blindly loads all the segments into the VMCS if they are less
than the gdt limit. Since the real mode segments are around 0x3000, we are
getting junk into the VMCS and that later causes an exception.
Stephen Tweedie has written up a patch to fix the Xen side and will be
submitting that to those folks. But that doesn't excuse the GDT limit
being a magnitude too big.
AK: changed to compute true gdt size in assembler, fixed comment
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Andi Kleen <ak@suse.de>
ptrace(PTRACE_[SG]ET_THREAD_AREA) calls from ia32 code
should be passed onto the x86_64 implementation.
The default case in sys32_ptrace used to call to sys_ptrace(), but is
now EINVAL. This patch fixes a regression caused by that changed.
Signed-off-by: Mike McCormack <mike@codeweavers.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Fix partial page check in e820_register_active_regions to ensure
partial pages are
not being marked as active in the memory pool.
Signed-off-by: Aaron Durbin <adurbin@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
This refactoring actually optimizes the code a little by caching the value
that we think the device is programmed with instead of reading it back from
the hardware. Which simplifies the code a little and should speed things up a
bit.
This patch introduces the concept of a ht_irq_msg and modifies the
architecture read/write routines to update this code.
There is a minor consistency fix here as well as x86_64 forgot to initialize
the htirq as masked.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Andi Kleen <ak@suse.de>
Acked-by: Bryan O'Sullivan <bos@pathscale.com>
Cc: <olson@pathscale.com>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This is the x86-64 version of f9dadfa71b
that did the same thing on i386.
Since the "mask" bit is in the low word, when we write a new entry, we
need to write the high word first, before we potentially unmask it.
The exception is when we actually want to mask the interrupt, in which
case we want to write the low word first to make sure that the high word
doesn't change while the interrupt routing is still active.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This is just commit 130fe05dbc ported to
x86-64, for all the same reasons. It cleans up the IO-APIC accesses in
order to then fix the ordering issues.
We move the accessor functions (that were only used by io_apic.c) out of
a header file, and use proper memory-mapped accesses rather than making
up our own "volatile" pointers.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The recent change to make x86_64 support i386 binaries compiled
with -mregparm=3 only covered signal handlers without SA_SIGINFO.
(the 3-arg "real-time" ones) This is useful for klibc at least.
Signed-off-by: Albert Cahalan <acahalan@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add a vmlinux.lds.h helper macro for defining the eight-level initcall table,
teach all the architectures to use it.
This is a prerequisite for a patch which performs initcall synchronisation for
multithreaded-probing.
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
[ Added AVR32 as well ]
Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When I generalized __assign_irq_vector I failed to pay attention
to what happens when you access a per cpu data structure for
a cpu that is not online. It is an undefined case making any
code that does it have undefined behavior as well.
The code still needs to be able to allocate a vector across cpus
that are not online to properly handle combinations like lowest
priority interrupt delivery and cpu_hotplug. Not that we can do
that today but the infrastructure shouldn't prevent it.
So this patch updates the places where we touch per cpu data
to only touch online cpus, it makes cpu vector allocation
an atomic operation with respect to cpu hotplug, and it updates
the cpu start code to properly initialize vector_irq so we
don't have inconsistencies.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andi Kleen <ak@suse.de>
There is no reason to remember a per cpu position of which vector
to try. Keeping a global position is simpler and more likely to
result in a global vector allocation even if I don't need or require
it. For level triggered interrupts this means we are less likely to
acknowledge another cpus irq, and cause the level triggered irq to
harmlessly refire.
This simplification makes it easier to only access data structures
of online cpus, by having fewer special cases to deal with.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andi Kleen <ak@suse.de>
This patch increases the timeout for PCI split transactions on PHB1 on
the first Calgary to work around an issue with the aic94xx
adapter. Fixes kernel.org bugzilla #7180
(http://bugzilla.kernel.org/show_bug.cgi?id=7180)
Based on excellent debugging and a patch by Darrick J. Wong
<djwong@us.ibm.com>
Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Darrick J. Wong <djwong@us.ibm.com>
There was a typo in the C3 latency test to decide of the TSC
should be used or not. It used the C2 latency threshold, not the
C3 one. Fix that.
This should fix the time on various dual core laptops.
Signed-off-by: Andi Kleen <ak@suse.de>
This patch increases the timeout for PCI split transactions on PHB1 on
the first Calgary to work around an issue with the aic94xx
adapter. Fixes kernel.org bugzilla #7180
(http://bugzilla.kernel.org/show_bug.cgi?id=7180)
Based on excellent debugging and a patch by Darrick J. Wong
<djwong@us.ibm.com>
Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Darrick J. Wong <djwong@us.ibm.com>
There was a typo in the C3 latency test to decide of the TSC
should be used or not. It used the C2 latency threshold, not the
C3 one. Fix that.
This should fix the time on various dual core laptops.
Signed-off-by: Andi Kleen <ak@suse.de>
By default route the 8254 over the 8259 and only disable
it on ATI boards where this causes double timer interrupts.
This should unbreak some Nvidia boards where the timer doesn't
seem to tick of it isn't enabled in the 8259. At least one
VIA board also seemed to have a little trouble with the disabled
8259.
For 2.6.20 we'll try both dynamically without black listing, but I think
for .19 this is the safer approach because it has been already well tested
in earlier kernels. This also makes the x86-64 behaviour the same
as i386.
Command line options can change all this of course.
Signed-off-by: Andi Kleen <ak@suse.de>
o A recent change to vmlinux.ld.S file broke kexec as now resulting vmlinux
program headers are overlapping in physical address space.
o Now all the vsyscall related sections are placed after data and after
that mostly init data sections are placed. To avoid physical overlap
among phdrs, there are three possible solutions.
- Place vsyscall sections also in data phdrs instead of user
- move vsyscal sections after init data in bss.
- create another phdrs say data.init and move all the sections
after vsyscall into this new phdr.
o This patch implements the third solution.
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Magnus Damm <magnus@valinux.co.jp>
Cc: Andi Kleen <ak@suse.de>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
TARGET_CPUS is the default irq routing poicy. It specifies which cpus the
kernel should aim an irq at. In physflat delivery mode we can route an irq to
a single cpu. But that doesn't mean our default policy should only be a
single cpu is allowed.
By allowing the irq routing code to select from multiple cpus this enables
systems with more irqs then we can service on a single processor to actually
work.
I just audited and tested the code and irqbalance doesn't care, and the
io_apic.c doesn't care if we have extra cpus in the mask. Everything will use
or assume we are using the lowest numbered cpu in the mask if we can't use
them all.
So this should result in no behavior changes except on systems that need it.
Thanks for YH Lu for spotting this problem in his testing.
Cc: Yinghai Lu <yinghai.lu@amd.com>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Jan convinced me that it was unnecessary because the assembly stubs do
this already on the stack.
Cc: jbeulich@novell.com
Signed-off-by: Andi Kleen <ak@suse.de>
Thanks to YH Lu for spotting this. It appears I missed this function when I
refactored allocate_irq_vector and introduced irq_domain, with the result that
all retriggered irqs would go to cpu 0 even if we were not prepared to receive
them there.
While reviewing YH's patch I also noticed that this function was missing
locking, and since I am now reading two values from two diffrent arrays that
looks like a race we might be able to hit in the real world.
Cc: Yinghai Lu <yinghai.lu@amd.com>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
This patch:
- out of range system calls failing to return -ENOSYS under
system call tracing
[AK: split out from another patch by Jan as separate bugfix]
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Andrew Morton