there's a new ptrace arch level feature in .28:
config X86_PTRACE_BTS
bool "Branch Trace Store"
it has broken fork() handling: the old DS area gets copied over into
a new task without clearing it.
Fixes exist but they came too late:
c5dee61: x86, bts: memory accounting
bf53de9: x86, bts: add fork and exit handling
and are queued up for v2.6.29. This shows that the facility is still not
tested well enough to release into a stable kernel - disable it for now and
reactivate in .29. In .29 the hardware-branch-tracer will use the DS/BTS
facilities too - hopefully resulting in better code.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix deadlock
This is in response to the following bug report:
Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=12100
Subject : resume (S2R) broken by Intel microcode module, on A110L
Submitter : Andreas Mohr <andi@lisas.de>
Date : 2008-11-25 08:48 (19 days old)
Handled-By : Dmitry Adamushko <dmitry.adamushko@gmail.com>
[ The deadlock scenario has been discovered by Andreas Mohr ]
I think I might have a logical explanation why the system:
(http://bugzilla.kernel.org/show_bug.cgi?id=12100)
might hang upon resuming, OTOH it should have likely hanged each and every time.
(1) possible deadlock in microcode_resume_cpu() if either 'if' section is
taken;
(2) now, I don't see it in spec. and can't experimentally verify it (newer
ucodes don't seem to be available for my Core2duo)... but logically-wise, I'd
think that when read upon resuming, the 'microcode revision' (MSR 0x8B) should
be back to its original one (we need to reload ucode anyway so it doesn't seem
logical if a cpu doesn't drop the version)... if so, the comparison with
memcmp() for the full 'struct cpu_signature' is wrong... and that's how one of
the aforementioned 'if' sections might have been triggered - leading to a
deadlock.
Obviously, in my tests I simulated loading/resuming with the ucode of the same
version (just to see that the file is loaded/re-loaded upon resuming) so this
issue has never popped up.
I'd appreciate if someone with an appropriate system might give a try to the
2nd patch (titled "fix a comparison && deadlock...").
In any case, the deadlock situation is a must-have fix.
Reported-by: Andreas Mohr <andi@lisas.de>
Signed-off-by: Dmitry Adamushko <dmitry.adamushko@gmail.com>
Tested-by: Andreas Mohr <andi@lisas.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: move the BTS buffer accounting to the mlock bucket
Add alloc_locked_buffer() and free_locked_buffer() functions to mm/mlock.c
to kalloc a buffer and account the locked memory to current.
Account the memory for the BTS buffer to the tracer.
Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: introduce new ptrace facility
Add arch_ptrace_untrace() function that is called when the tracer
detaches (either voluntarily or when the tracing task dies);
ptrace_disable() is only called on a voluntary detach.
Add ptrace_fork() and arch_ptrace_fork(). They are called when a
traced task is forked.
Clear DS and BTS related fields on fork.
Release DS resources and reclaim memory in ptrace_untrace(). This
releases resources already when the tracing task dies. We used to do
that when the traced task dies.
Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: remove annoying bootup printk
It's perfectly normal for no AMD GART to be present, e.g., if you have
Intel CPUs. None of the other iommu_init() functions makes noise when
it finds nothing.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Acked-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: clean up
Itroduce MCOUNT_SAVE/RESTORE_FRAME which allow us to
save a number of lines on source level.
Also fix a comment in ftrace.h.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix disabled MCE after resume
Don't prevent multiple initialization of MCEs.
Back from early prehistory mcheck_init() has a reentry check. Presumably
that was needed in very old kernels to prevent it entering twice.
But as Andreas points out this prevents CPU hotplug (and therefore resume)
to correctly reinitialize MCEs when a AP boots again after being
offlined.
Just drop the check.
Reported-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Tested-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: change the reporting of empty BTS records
Correctly report a cleared BTS record as invalid. Used to be reported
as branch from 0 to 0.
Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
Optimistically allocate a DS context. It is extremely unlikely that
one already existed. This simplifies the code a lot.
Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
VMI initialiation can relocate the fixmap, causing early_ioremap to
malfunction if it is initialized before the relocation. To fix this,
VMI activation is split into two phases; the detection, which must
happen before setting up ioremap, and the activation, which must happen
after parsing early boot parameters.
This fixes a crash on boot when VMI is enabled under VMware.
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Impact: move most important x86 irq entry-points to a separate subsection
Annotate do_IRQ and smp_apic_timer_interrupt to put them into the .irqentry.text
subsection. These function will so be recognized as hardirq entrypoints for the
function-graph-tracer. We could also annotate other irq entries but the others
are far less important but they can be added on request.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: let the function-graph-tracer be aware of the irq entrypoints
Add a new .irqentry.text section to store the irq entrypoints functions
inside the same section. This way, the tracer will be able to signal
an interrupts triggering on output by recognizing these entrypoints.
Also, make this section recordable for dynamic tracing.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: build fix
arch/x86/kernel/ds.c: In function 'ds_request':
arch/x86/kernel/ds.c:236: sorry, unimplemented: inlining failed in call to 'ds_get_context': recursive inlining
but the recursion here is scary ...
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
Move the BTS bits from ptrace.c into ds.c.
Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: make the ds code more debuggable
Turn BUG_ON's into WARN_ON_ONCE.
Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
these warnings:
arch/x86/kernel/paravirt-spinlocks.c: In function ‘default_spin_lock_flags’:
arch/x86/kernel/paravirt-spinlocks.c:12: warning: passing argument 1 of ‘__raw_spin_lock’ from incompatible pointer type
arch/x86/kernel/paravirt-spinlocks.c: At top level:
arch/x86/kernel/paravirt-spinlocks.c:11: warning: ‘default_spin_lock_flags’ defined but not used
showed that the prototype of default_spin_lock_flags() was confused about
what type spinlocks have.
the proper type on UP is raw_spinlock_t.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: Provide a way to pause the function graph tracer
As suggested by Steven Rostedt, the previous patch that prevented from
spinlock function tracing shouldn't use the raw_spinlock to fix it.
It's much better to follow lockdep with normal spinlock, so this patch
adds a new flag for each task to make the function graph tracer able
to be paused. We also can send an ftrace_printk whithout worrying of
the irrelevant traced spinlock during insertion.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: trace more functions
When the function graph tracer is configured, three more files are not
traced to prevent only four functions to be traced. And this impacts the
normal function tracer too.
arch/x86/kernel/process_64/32.c:
I had crashes when I let this file traced. After some debugging, I saw
that the "current" task point was changed inside__swtich_to(), ie:
"write_pda(pcurrent, next_p);" inside process_64.c Since the tracer store
the original return address of the function inside current, we had
crashes. Only __switch_to() has to be excluded from tracing.
kernel/module.c and kernel/extable.c:
Because of a function used internally by the function graph tracer:
__kernel_text_address()
To let the other functions inside these files to be traced, this patch
introduces the __notrace_funcgraph function prefix which is __notrace if
function graph tracer is configured and nothing if not.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86: fix early panic with boot option "nosmp"
x86/oprofile: fix Intel cpu family 6 detection
oprofile: fix CPU unplug panic in ppro_stop()
AMD IOMMU: fix possible race while accessing iommu->need_sync
AMD IOMMU: set device table entry for aliased devices
AMD IOMMU: struct amd_iommu remove padding on 64 bit
x86: fix broken flushing in GART nofullflush path
x86: fix dma_mapping_error for 32bit x86
Impact: fix boot crash with numcpus=0 on certain systems
Fix early exception in __get_smp_config with nosmp.
Bail out early when there is no MP table.
Reported-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Tested-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Alan Jenkins wrote:
> This is on an EeePC 701, /proc/cpuinfo as attached.
>
> Is this expected? Will the next release work?
>
> Thanks, Alan
>
> # opcontrol --setup --no-vmlinux
> cpu_type 'unset' is not valid
> you should upgrade oprofile or force the use of timer mode
>
> # opcontrol -v
> opcontrol: oprofile 0.9.4 compiled on Nov 29 2008 22:44:10
>
> # cat /dev/oprofile/cpu_type
> i386/p6
> # uname -r
> 2.6.28-rc6eeepc
Hi Alan,
Looking at the kernel driver code for oprofile it can return the "i386/p6" for
the cpu_type. However, looking at the user-space oprofile code there isn't the
matching entry in libop/op_cpu_type.c or the events/unit_mask files in
events/i386 directory.
The Intel AP-485 says this is a "Intel Pentium M processor model D". Seems like
the oprofile kernel driver should be identifying the processor as "i386/p6_mobile"
The driver identification code doesn't look quite right in nmi_init.c
http://git.kernel.org/?p=linux/kernel/git/sfr/linux-next.git;a=blob;f=arch/x86/oprofile/nmi_int.c;h=022cd41ea9b4106e5884277096e80e9088a7c7a9;hb=HEAD
has:
409 case 10 ... 13:
410 *cpu_type = "i386/p6";
411 break;
Referring to the Intel AP-485:
case 10 and 11 should produce "i386/piii"
case 13 should produce "i386/p6_mobile"
I didn't see anything for case 12.
Something like the attached patch. I don't have a celeron machine to verify that
changes in this area of the kernel fix thing.
-Will
Signed-off-by: William Cohen <wcohen@redhat.com>
Tested-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Acked-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
If oprofile statically compiled in kernel, a cpu unplug triggers
a panic in ppro_stop(), because a NULL pointer is dereferenced.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
The access to the iommu->need_sync member needs to be protected by the
iommu->lock. Otherwise this is a possible race condition. Fix it with
this patch.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
In some rare cases a request can arrive an IOMMU with its originial
requestor id even it is aliased. Handle this by setting the device table
entry to the same protection domain for the original and the aliased
requestor id.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Remove 16 bytes of padding from struct amd_iommu on 64bit builds
reducing its size to 120 bytes, allowing it to span one fewer
cachelines.
Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Impact: remove stale IOTLB entries
In the non-default nofullflush case the GART is only flushed when
next_bit wraps around. But it can happen that an unmap operation unmaps
memory which is behind the current next_bit location. If these addresses
are reused it may result in stale GART IO/TLB entries. Fix this by
setting the GART next_bit always behind an unmapped location.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Import: robustness checks
Add more checks in the function graph code to detect errors and
perhaps print out better information if a bug happens.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: feature, let entry function decide to trace or not
This patch lets the graph tracer entry function decide if the tracing
should be done at the end as well. This requires all function graph
entry functions return 1 if it should trace, or 0 if the return should
not be traced.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: better dumpstack output
I noticed in my crash dumps and even in the stack tracer that a
lot of functions listed in the stack trace are simply
return_to_handler which is ftrace graphs way to insert its own
call into the return of a function.
But we lose out where the actually function was called from.
This patch adds in hooks to the dumpstack mechanism that detects
this and finds the real function to print. Both are printed to
let the user know that a hook is still in place.
This does give a funny side effect in the stack tracer output:
Depth Size Location (80 entries)
----- ---- --------
0) 4144 48 save_stack_trace+0x2f/0x4d
1) 4096 128 ftrace_call+0x5/0x2b
2) 3968 16 mempool_alloc_slab+0x16/0x18
3) 3952 384 return_to_handler+0x0/0x73
4) 3568 -240 stack_trace_call+0x11d/0x209
5) 3808 144 return_to_handler+0x0/0x73
6) 3664 -128 mempool_alloc+0x4d/0xfe
7) 3792 128 return_to_handler+0x0/0x73
8) 3664 -32 scsi_sg_alloc+0x48/0x4a [scsi_mod]
As you can see, the real functions are now negative. This is due
to them not being found inside the stack.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: new ftrace_graph_stop function
While developing more features of function graph, I hit a bug that
caused the WARN_ON to trigger in the prepare_ftrace_return function.
Well, it was hard for me to find out that was happening because the
bug would not print, it would just cause a hard lockup or reboot.
The reason is that it is not safe to call printk from this function.
Looking further, I also found that it calls unregister_ftrace_graph,
which grabs a mutex and calls kstop machine. This would definitely
lock the box up if it were to trigger.
This patch adds a fast and safe ftrace_graph_stop() which will
stop the function tracer. Then it is safe to call the WARN ON.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: consistency change for function graph
This patch makes function graph record the mcount caller address
the same way the function tracer does.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
