> -Original Message-
> From: Matt Fleming [mailto:m...@console-pimps.org]
> Sent: Wednesday, November 27, 2013 6:45 AM
> To: Madper Xie
> Cc: 'linux-kernel@vger.kernel.org'; linux-...@vger.kernel.org;
> matt.flem...@intel.com; Seiji Aguchi; rich...@nod
> Seiji - are you okay with this? If so I'll just make a quick patch now.
>
I'm OK with it.
Seiji
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
P
Commit-ID: d34603b07c4255b2b00a546d34f297ccd50ae4c6
Gitweb: http://git.kernel.org/tip/d34603b07c4255b2b00a546d34f297ccd50ae4c6
Author: Seiji Aguchi
AuthorDate: Wed, 30 Oct 2013 16:39:03 -0400
Committer: H. Peter Anvin
CommitDate: Fri, 8 Nov 2013 14:15:49 -0800
x86, trace: Add page
Commit-ID: 25c74b10bacead867478480170083f69cfc0db48
Gitweb: http://git.kernel.org/tip/25c74b10bacead867478480170083f69cfc0db48
Author: Seiji Aguchi
AuthorDate: Wed, 30 Oct 2013 16:37:00 -0400
Committer: H. Peter Anvin
CommitDate: Fri, 8 Nov 2013 14:15:45 -0800
x86, trace: Register
Commit-ID: ac7956e2699380b8b10146ec2ba8cbe43a03ff7a
Gitweb: http://git.kernel.org/tip/ac7956e2699380b8b10146ec2ba8cbe43a03ff7a
Author: Seiji Aguchi
AuthorDate: Wed, 30 Oct 2013 16:37:47 -0400
Committer: H. Peter Anvin
CommitDate: Fri, 8 Nov 2013 14:15:47 -0800
x86, trace: Delete
Commit-ID: 959c071f0974cda7702d7574647de7ad9259eb57
Gitweb: http://git.kernel.org/tip/959c071f0974cda7702d7574647de7ad9259eb57
Author: Seiji Aguchi
AuthorDate: Wed, 30 Oct 2013 16:36:08 -0400
Committer: H. Peter Anvin
CommitDate: Fri, 8 Nov 2013 14:15:44 -0800
x86, trace: Remove
> How does efivars backend handle "unlink(2)" in the pstore file system.
> pstore will call the backend->erase function passing the "id". The
> backend should then erase the right record from persistent storage.
>
> With the ((timestamp * 100 + part) * 100 + count function - you can
> easily rev
> What about feeding the bytes of all three integers into a non-cryptographic
> hash function?
> Using this way you get a cheap unique id.
It is reasonable to me.
Seiji
> >> +{
> >> + char id_str[64];
> >> + u64 id = 0;
> >> +
> >> + sprintf(id_str, "%lu%u%d", timestamp, part, count);
> >> + if (kstrtoull(id_str, 10, &id))
> >> + pr_warn("efi-pstore: failed to generate id\n");
> >> + return id;
> >> +}
> >
> > This is ju
> I also like option 1 ... but I think the "id" should be a persistent value for
> a given saved record. So some func(timestamp, part, count) would be a
> good idea. If we try using "sequential" numbers - and don't manage to
> clear out /sys/fs/pstore each time - then we may have the same "dmesg"
er.kernel.org
> [mailto:linux-efi-ow...@vger.kernel.org] On Behalf Of Madper Xie
> Sent: Wednesday, October 30, 2013 11:01 PM
> To: Luck, Tony
> Cc: Seiji Aguchi; Madper Xie; keesc...@chromium.org; ccr...@android.com;
> an...@enomsg.org; linux-...@vger.kernel.org; linux-
> ker.
> Ah - I was expecting that the backend driver would have a unique "id" for
> each record stored ... but is seems that this isn't true for efivars.
>
So, do you mean efivars should fix to use the "id" in a proper way?
I acked Madper's patch 2/2 earlier today, but when I look at your test result,
Currently irq vector handlers for tracing are registered in both set_intr_gate()
and __trace_alloc_intr_gate() in alloc_intr_gate().
But, we don't need to do that twice.
So, let's delete __trace_alloc_intr_gate().
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/d
comes from the page fault events.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/trace/exceptions.h | 52 +
arch/x86/mm/Makefile| 2 ++
arch/x86/mm/fault.c | 13 +
3 files changed, 67 insertions(+)
create mode
_fault() to add tracepoints
in a subsequent patch.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/desc.h| 28 +++-
arch/x86/include/asm/hw_irq.h | 3 +++
arch/x86/include/asm/segment.h | 3 +++
arch/x86/include/asm/traps.h | 20
arch/x86/ke
Change from v3:
- Separate modifications to make review easy.
- Refactor implementations registering exception/irq_vector
handers. (Patch 1, 2, 3)
This series introduce page fault tracepoints.
Detailed descriptions are explained in each patch.
Any comments are welcome.
Seiji Aguchi (4
Move set_intr_gate() into a macro by removing __alloc_intr_gate().
The purpose is to avoid failing a kernel build after applying
a subsequent patch which changes set_intr_gate() to macro.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/desc.h | 7 +--
1 file changed, 1 insertion(+), 6
[] SyS_mount+0x83/0xc0
[1.158207] [] system_call_fastpath+0x16/0x1b
[1.158207] ---[ end trace 61981bc62de9f6f4 ]---
Signed-off-by: Seiji Aguchi
---
drivers/firmware/efi/efi-pstore.c | 143 +++---
drivers/firmware/efi/efivars.c| 12 ++--
drivers/firmwar
> On Fri, 18 Oct, at 10:30:58PM, Seiji Aguchi wrote:
> > The scanning and deleting logic is still needed. In case an entry(A)
> > is found, the pointer is saved to psi->data. And efi_pstore_read()
> > passes the entry(A) to a pstore filesystem by releasing
> > __e
> -Original Message-
> From: Madper Xie [mailto:c...@redhat.com]
> Sent: Wednesday, October 30, 2013 5:45 AM
> To: tony.l...@intel.com; keesc...@chromium.org; ccr...@android.com;
> an...@enomsg.org; Seiji Aguchi
> Cc: linux-...@vger.kernel.org; linux-kernel@vger.k
> -Original Message-
> From: Madper Xie [mailto:c...@redhat.com]
> Sent: Wednesday, October 30, 2013 5:45 AM
> To: tony.l...@intel.com; keesc...@chromium.org; ccr...@android.com;
> an...@enomsg.org; Seiji Aguchi
> Cc: linux-...@vger.kernel.org; linux-kernel@vger.k
To distinguish all events more precisely, "count" is needed to add to the name.
Please see the commit below.
755d4fe46529018ae45bc7c86df682de45ace764
Seiji
> -Original Message-
> From: Madper Xie [mailto:c...@redhat.com]
> Sent: Monday, October 28, 2013 9:22 AM
>
Matt,
> It seems to me that because you're no longer dropping __efivars->lock
> when reading from the EFI variable store, you actually don't need all
> the ->scanning and ->deleting logic because anything that sets those
> flags runs to completion while holding the lock.
The scanning and deleting
ctober 17, 2013 11:07 AM
> To: Madper Xie
> Cc: Seiji Aguchi; linux-kernel@vger.kernel.org; linux-...@vger.kernel.org;
> matt.flem...@intel.com; tony.l...@intel.com; Tomoki
> Sekiyama; dle-deve...@lists.sourceforge.net
> Subject: Re: [PATCH v3] efivars,efi-pstore: Hold off deletion of s
Hi Madper,
I tested this patch on 3.12-rc4.
Could you please send me the log when you failed to apply?
Seiji
> -Original Message-
> From: Madper Xie [mailto:c...@redhat.com]
> Sent: Thursday, October 17, 2013 1:54 AM
> To: Seiji Aguchi
> Cc: linux-kernel@vger.ke
> There's also an "either/or" choice between using efi-capsule with pstore, and
> the
> traditional kexec/kdump method for getting a memory dump from a crash. We
> have to go through a reset to save the capsule - but we don't want a reset for
> kexec. Perhaps we can pass the reset parameters th
> +#define efi_call_reset_virt(f, args...)
> \
> +({ \
> + unsigned long __flags; \
> + bool __nmi = in_nmi();
Thank you for reviewing.
> > http://marc.info/?l=linux-mm&m=136807959830182&w=2
> > http://marc.info/?l=linux-mm&m=136807959130175&w=2
> >
>
> For permanence, please use links of the form:
>
> http://lkml.kernel.org/r/message-id
>
> (Yes, they currently point to marc.info, but can
Peter,
Any comment?
Seiji
> -Original Message-
> From: linux-kernel-ow...@vger.kernel.org
> [mailto:linux-kernel-ow...@vger.kernel.org] On Behalf Of Seiji Aguchi
> Sent: Monday, September 09, 2013 5:56 PM
> To: linux-kernel@vger.kernel.org; x...@kernel.org
> Cc: h.
Matt,
I submitted a v3 patch based on my comment below..
Seiji
> -Original Message-
> From: linux-efi-ow...@vger.kernel.org
> [mailto:linux-efi-ow...@vger.kernel.org] On Behalf Of Seiji Aguchi
> Sent: Wednesday, October 09, 2013 12:37 PM
> To: Matt Fleming
>
__alloc_percpu+0x10/0x20
[1.158207] [] vfs_kern_mount+0x63/0xf0
[1.158207] [] do_mount+0x23e/0xa20
[1.158207] [] ? strndup_user+0x4b/0xf0
[1.158207] [] SyS_mount+0x83/0xc0
[1.158207] [] system_call_fastpath+0x16/0x1b
[1.158207] ---[ end trace 61981bc62de9f6f4 ]---
Si
Thank you for reviewing.
In my understanding, your point is that all accesses to efivar_entry should be
done while holding __efivars->lock.
> > @@ -88,8 +103,9 @@ static int efi_pstore_read_func(struct efivar_entry
> > *entry, void *data)
> > return 0;
> >
> > entry->var.DataSize
Are there anyone who can review this bugfix?
Seiji
> -Original Message-
> From: linux-kernel-ow...@vger.kernel.org
> [mailto:linux-kernel-ow...@vger.kernel.org] On Behalf Of Seiji Aguchi
> Sent: Friday, September 27, 2013 4:24 PM
> To: linux-kernel@vger.ke
] ---[ end trace 61981bc62de9f6f4 ]---
Signed-off-by: Seiji Aguchi
---
drivers/firmware/efi/efi-pstore.c | 145 +++---
drivers/firmware/efi/efivars.c| 3 +-
drivers/firmware/efi/vars.c | 39 +++---
include/linux/efi.h | 4 +-
4 files
efivars works fine with this v2 patch.
Tested-by: Seiji Aguchi
> -Original Message-
> From: Aruna Balakrishnaiah [mailto:ar...@linux.vnet.ibm.com]
> Sent: Thursday, September 12, 2013 2:51 AM
> To: linuxppc-...@ozlabs.org; tony.l...@intel.com; Seiji Aguchi;
&g
d my "Tested-by" to all three patches.
Tested-by: Seiji Aguchi
N�r��yb�X��ǧv�^�){.n�+{zX����ܨ}���Ơz�&j:+v���zZ+��+zf���h���~i���z� �w���?�&�)ߢf��^jǫy�m��@A�a���
0��h���i
register trace handler to
trace IDT and non-trace handler to normal IDT.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/desc.h | 33 +
arch/x86/include/asm/hw_irq.h | 14 -
arch/x86/include/asm/trace/exceptions.h | 52 ++
> -Original Message-
> From: Steven Rostedt [mailto:rost...@goodmis.org]
> Sent: Friday, September 06, 2013 12:50 PM
> To: Seiji Aguchi
> Cc: linux-kernel@vger.kernel.org; x...@kernel.org; h...@zytor.com;
> mi...@elte.hu; b...@alien8.de; t...@linutronix.de;
&g
Any comment?
> -Original Message-
> From: linux-kernel-ow...@vger.kernel.org
> [mailto:linux-kernel-ow...@vger.kernel.org] On Behalf Of Seiji Aguchi
> Sent: Friday, August 23, 2013 11:38 AM
> To: linux-kernel@vger.kernel.org; x...@kernel.org
> Cc: h...@zytor.com; rost.
..@intel.com]
> Sent: Wednesday, September 04, 2013 12:11 PM
> To: Aruna Balakrishnaiah; Seiji Aguchi
> Cc: jkeni...@linux.vnet.ibm.com; keesc...@chromium.org;
> mah...@linux.vnet.ibm.com; ccr...@android.com; linux-
> ker...@vger.kernel.org; linuxppc-...@ozlabs.org; cbouatmai...@gmail.com
Aruna,
Sorry for the late response.
> Seiji,
>
> Could you let us know the efivars buffer size with which the pstore is
> registered when
> the failure occurred.
I looked into the issue today.
I added some debug message just before pstore_compress().
As you can see below, the buffer size is a
ler to
trace IDT and non-trace handler to normal IDT.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/desc.h | 33 +
arch/x86/include/asm/hw_irq.h | 14 -
arch/x86/include/asm/trace/exceptions.h | 51 +
a
> -Original Message-
> From: Luck, Tony [mailto:tony.l...@intel.com]
> Sent: Thursday, August 22, 2013 7:17 PM
> To: Seiji Aguchi; Aruna Balakrishnaiah; linuxppc-...@ozlabs.org;
> linux-kernel@vger.kernel.org; keesc...@chromium.org
> Cc: jkeni...@linux.vnet.ibm.com; an
> * callback from kmsg_dump. (s2,l2) has the most recently
> * written bytes, older bytes are in (s1,l1). Save as much
> @@ -148,23 +243,56 @@ static void pstore_dump(struct kmsg_dumper *dumper,
> char *dst;
> unsigned long size;
> int hsize;
> +
> -Original Message-
> From: linux-kernel-ow...@vger.kernel.org
> [mailto:linux-kernel-ow...@vger.kernel.org] On Behalf Of Aruna Balakrishnaiah
> Sent: Friday, August 16, 2013 9:18 AM
> To: linuxppc-...@ozlabs.org; tony.l...@intel.com;
> linux-kernel@vger.kernel.org; keesc...@chromium.o
> Printing the regs pointer is rather useless. This is specific for x86,
> why not print the ip of where it happened and the faulting address
> itself?
Thank you for reviewing.
I will change the regs pointer to ip.
> Note, you only need to change the TP_printk() to do that. For
> efficiency reaso
]---
Signed-off-by: Seiji Aguchi
---
drivers/firmware/efi/efi-pstore.c | 145 +++---
drivers/firmware/efi/efivars.c| 3 +-
drivers/firmware/efi/vars.c | 39 +++---
include/linux/efi.h | 4 +-
4 files changed, 151 insertions(+), 40
Any comment?
> -Original Message-
> From: Seiji Aguchi [mailto:seiji.agu...@hds.com]
> Sent: Tuesday, July 30, 2013 6:53 PM
> To: linux-kernel@vger.kernel.org; x...@kernel.org
> Cc: h...@zytor.com; rost...@goodmis.org; mi...@elte.hu; b...@alien8.de;
> t...@li
aw() to register just non-trace handler to IDT.
This is used at boot time which tracing is disabled.
- Make set_intr_gate() macro so that it can register trace handler to
trace IDT and non-trace handler to normal IDT.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/desc.h
> -Original Message-
> From: H. Peter Anvin [mailto:h...@zytor.com]
> Sent: Friday, July 05, 2013 12:41 PM
> To: Thomas Gleixner
> Cc: Dave Jones; Linus Torvalds; Linux Kernel; Ingo Molnar; Peter Zijlstra;
> Seiji Aguchi
> Subject: Re: Yet more softlockups.
>
Commit-ID: 4787c368a9bca39e173d702389ee2eaf0520abc1
Gitweb: http://git.kernel.org/tip/4787c368a9bca39e173d702389ee2eaf0520abc1
Author: Seiji Aguchi
AuthorDate: Fri, 28 Jun 2013 14:02:11 -0400
Committer: Ingo Molnar
CommitDate: Tue, 2 Jul 2013 09:52:31 +0200
x86/tracing: Add irq_enter
Ingo,
Please merge this patch to your tree.
It is a bug fix of irq_vector tracepoints patch.
Seiji
> -Original Message-
> From: Steven Rostedt [mailto:rost...@goodmis.org]
> Sent: Friday, June 28, 2013 2:16 PM
> To: Seiji Aguchi
> Cc: linux-kernel@vger.kernel.org; x..
50.771947] OK
[ 50.772099] Testing event reschedule_entry: OK
Signed-off-by: Seiji Aguchi
---
arch/x86/kernel/smp.c | 29 ++---
1 files changed, 18 insertions(+), 11 deletions(-)
diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
index f4fe0b8..cdaa347 100644
---
> > +void smp_trace_reschedule_interrupt(struct pt_regs *regs)
> > +{
>
> Sure, but we should probably add a comment here to why
> smp_trace_reschedule_interrupt() is different than
> smp_reschedule_interrupt().
>
> /*
>* Need to call irq_enter() before calling the trace point.
>
> > diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
> > index f4fe0b8..b959056 100644
> > --- a/arch/x86/kernel/smp.c
> > +++ b/arch/x86/kernel/smp.c
> > @@ -268,9 +268,11 @@ void smp_reschedule_interrupt(struct pt_regs *regs)
> > void smp_trace_reschedule_interrupt(struct pt_regs *regs
Any comments?
> -Original Message-
> From: linux-kernel-ow...@vger.kernel.org
> [mailto:linux-kernel-ow...@vger.kernel.org] On Behalf Of Seiji Aguchi
> Sent: Monday, June 24, 2013 4:21 PM
> To: linux-kernel@vger.kernel.org; x...@kernel.org; rost...@goodmis.org;
> h...@
I just sent the fix below.
[PATCH -tip] x86,trace: Add rcu_irq_enter/exit() in
smp_trace_reschedule_interrupt()
> -Original Message-
> From: Fengguang Wu [mailto:fengguang...@intel.com]
> Sent: Saturday, June 22, 2013 8:36 PM
> To: fengguang...@intel.com; Seiji Aguchi
&g
t;79d30489>] ? repair_env_string+0x51/0x51
[ 50.731214] [<79d302c3>] i386_start_kernel+0x79/0x7d
[ 50.771947] OK
[ 50.772099] Testing event reschedule_entry: OK
Signed-off-by: Seiji Aguchi
---
arch/x86/kernel/smp.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
dif
Commit-ID: 33e5ff634f07dec26b7ed1fd7f9e32978fe1f2b2
Gitweb: http://git.kernel.org/tip/33e5ff634f07dec26b7ed1fd7f9e32978fe1f2b2
Author: Seiji Aguchi
AuthorDate: Sat, 22 Jun 2013 07:33:30 -0400
Committer: Ingo Molnar
CommitDate: Sun, 23 Jun 2013 11:41:36 +0200
x86/tracing: Add config
> +static inline void load_current_idt(void)
> +{
> + unsigned long flags;
> +
> + local_irq_save(flags);
> + if (is_debug_idt_enabled())
> + load_debug_idt();
> + else
> + load_idt((const struct desc_ptr *)&idt_descr);
> + local_irq_restore
n function `trace_thermal_interrupt':
(.entry.text+0x132b): undefined reference to `smp_trace_thermal_interrupt'
In this case, trace_threshold_interrupt/trace_thermal_interrupt are not needed
to define.
So, add config option checking to their definitions in entry_64.S.
Signed-off-by: Seiji A
Commit-ID: cf910e83ae23692fdeefc7e506e504c4c468d38a
Gitweb: http://git.kernel.org/tip/cf910e83ae23692fdeefc7e506e504c4c468d38a
Author: Seiji Aguchi
AuthorDate: Thu, 20 Jun 2013 11:46:53 -0400
Committer: H. Peter Anvin
CommitDate: Thu, 20 Jun 2013 22:25:34 -0700
x86, trace: Add irq
Commit-ID: eddc0e922a3530e0f22cef170229bcae3a7d5e31
Gitweb: http://git.kernel.org/tip/eddc0e922a3530e0f22cef170229bcae3a7d5e31
Author: Seiji Aguchi
AuthorDate: Thu, 20 Jun 2013 11:45:17 -0400
Committer: H. Peter Anvin
CommitDate: Thu, 20 Jun 2013 22:25:01 -0700
x86, trace: Introduce
Commit-ID: 629f4f9d59a27d8e58aa612e886e6a9a63ea7aeb
Gitweb: http://git.kernel.org/tip/629f4f9d59a27d8e58aa612e886e6a9a63ea7aeb
Author: Seiji Aguchi
AuthorDate: Thu, 20 Jun 2013 11:45:44 -0400
Committer: H. Peter Anvin
CommitDate: Thu, 20 Jun 2013 22:25:13 -0700
x86: Rename variables
only when debugging
is disabled.
In addition, the new IDT is created only when CONFIG_TRACING is enabled to
avoid being
used for other purposes.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/desc.h | 72 -
arch/x86/include/asm/entry_arch.h
Rename variables for debugging to describe meaning of them precisely.
Also, introduce a generic way to switch IDT by checking a current state,
debug on/off.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/desc.h | 47 +++--
arch/x86/kernel/cpu
esn't work, either.
If irq_enter() is nested, it may have a time penalty because it has to check if
it
was already called or not. The time penalty is not desired in performance
sensitive
paths even if it is tiny.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/apic.h
_timer_vector
- reschedule_vector
- call_function_vector
- call_function_single_vector
- irq_work_entry_vector
- error_apic_vector
- thermal_apic_vector
- threshold_apic_vector
- spurious_apic_vector
- x86_platform_ipi_vector
Seiji Aguchi (3):
trace,x86: Introd
This unfortunately
causes a lot of wasted duplicate code created.
By adding a DEFINE_EVENT_FN(), these events can still use a
DECLARE_EVENT_CLASS() and then define their own triggers.
Signed-off-by: Steven Rostedt
Signed-off-by: Seiji Aguchi
---
include/linux/tracepoint.h |2 ++
include/
> On Tue, 2013-06-18 at 15:01 -0400, Seiji Aguchi wrote:
> > /*
> > * the load_current_idt() is called with interrupt disabled by
> > local_irq_save()
> > * to avoid races. That way the IDT will always be set back to the
> > expected
> >
only when debugging
is disabled.
In addition, the new IDT is created only when CONFIG_TRACING is enabled to
avoid being
used for other purposes.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/desc.h | 72 -
arch/x86/include/asm/entry_arch.h
Rename variables for debugging to describe meaning of them precisely.
Also, introduce a generic way to switch IDT by checking a current state,
debug on/off.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/desc.h | 47 +++--
arch/x86/kernel/cpu
esn't work, either.
If irq_enter() is nested, it may have a time penalty because it has to check if
it
was already called or not. The time penalty is not desired in performance
sensitive
paths even if it is tiny.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/apic.h
oducing
irq_vector_entry/exit.
so that we can enable them independently.
- local_timer_vector
- reschedule_vector
- call_function_vector
- call_function_single_vector
- irq_work_entry_vector
- error_apic_vector
- thermal_apic_vector
- threshold_apic_vector
- spurious_apic_vect
This unfortunately
causes a lot of wasted duplicate code created.
By adding a DEFINE_EVENT_FN(), these events can still use a
DECLARE_EVENT_CLASS() and then define their own triggers.
Signed-off-by: Steven Rostedt
Signed-off-by: Seiji Aguchi
---
include/linux/tracepoint.h |2 ++
include/
>
> This patchset breaks building with SGI UV support enabled:
>
> /home/hpa/kernel/distwork/arch/x86/platform/uv/tlb_uv.c: In function
> ‘uv_bau_init’:
> /home/hpa/kernel/distwork/arch/x86/platform/uv/tlb_uv.c:2130:1: error:
> ‘trace_uv_bau_message_intr1’ undeclared (first
> use in this function
Any comments?
> -Original Message-
> From: linux-kernel-ow...@vger.kernel.org
> [mailto:linux-kernel-ow...@vger.kernel.org] On Behalf Of Seiji Aguchi
> Sent: Friday, June 07, 2013 4:49 PM
> To: linux-kernel@vger.kernel.org; rost...@goodmis.org; h...@zytor.com
> Cc: t...
only when debugging
is disabled.
In addition, the new IDT is created only when CONFIG_TRACING is enabled to
avoid being
used for other purposes.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/desc.h | 72 -
arch/x86/include/asm/entry_arch.h
Rename variables for debugging to describe meaning of them precisely.
Also, introduce a generic way to switch IDT by checking a current state,
debug on/off.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/desc.h | 47 +++--
arch/x86/kernel/cpu
esn't work, either.
If irq_enter() is nested, it may have a time penalty because it has to check if
it
was already called or not. The time penalty is not desired in performance
sensitive
paths even if it is tiny.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/apic.h
ork_entry_vector
- error_apic_vector
- thermal_apic_vector
- threshold_apic_vector
- spurious_apic_vector
- x86_platform_ipi_vector
Seiji Aguchi (3):
trace,x86: Introduce entering/exiting_irq()
x86: Rename variables for debugging
trace,x86: Add irq vector tracepoints
Steven Rostedt (1):
tracin
This unfortunately
causes a lot of wasted duplicate code created.
By adding a DEFINE_EVENT_FN(), these events can still use a
DECLARE_EVENT_CLASS() and then define their own triggers.
Signed-off-by: Steven Rostedt
Signed-off-by: Seiji Aguchi
---
include/linux/tracepoint.h |2 ++
include/
Steven,
Thank you for reviewing.
I will create a patch just renaming variables and adding the
is_debug_idt_enabled() and load_debug_idt().
Seiji
> -Original Message-
> From: Steven Rostedt [mailto:rost...@goodmis.org]
> Sent: Thursday, June 06, 2013 4:01 PM
> To: Seiji
> -Original Message-
> From: Steven Rostedt [mailto:rost...@goodmis.org]
> Sent: Thursday, June 06, 2013 2:26 PM
> To: Seiji Aguchi
> Cc: linux-kernel@vger.kernel.org; h...@zytor.com; t...@linutronix.de;
> mi...@elte.hu; b...@alien8.de; linux-e...@vger.kernel.org;
>
_table and debug_stack_use_ctr to
debug_idt_descr,
debug_idt_table, and debug_idt_ctr to describe meaning of the variables
precisely.
In addtion, the new IDT is created only when CONFIG_TRACING is enabled to avoid
being
used for other purposes.
Signed-off-by: Seiji Aguchi
---
arch/x86/i
exit.
so that we can enable them independently.
- local_timer_vector
- reschedule_vector
- call_function_vector
- call_function_single_vector
- irq_work_entry_vector
- error_apic_vector
- thermal_apic_vector
- threshold_apic_vector
- spurious_apic_vector
- x86_platfor
But it doesn't work, either.
If irq_enter() is nested, it may have a time penalty because it has to check if
it
was already called or not. The time penalty is not desired in performance
sensitive
paths even if it is tiny.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/api
This unfortunately
causes a lot of wasted duplicate code created.
By adding a DEFINE_EVENT_FN(), these events can still use a
DECLARE_EVENT_CLASS() and then define their own triggers.
Signed-off-by: Steven Rostedt
Signed-off-by: Seiji Aguchi
---
include/linux/tracepoint.h |2 ++
include/
OK, I will rename debug_stack_use_ctr to debug_idt_ctr.
Thanks.
Seiji
> -Original Message-
> From: Steven Rostedt [mailto:rost...@goodmis.org]
> Sent: Tuesday, June 04, 2013 4:58 PM
> To: Seiji Aguchi
> Cc: H. Peter Anvin; linux-kernel@vger.kernel.org; x...@k
Steven,
>
> Hmm, maybe we can do it. Have two counters, a debug_idt_ctr and a
> trace_idt_ctr, then have a function that basically does this:
>
> if (this_cpu_read(debug_idt_ctr))
I think we can use "debug_stack_use_ctr" for the checking.
Is it correct?
Or, do I need to introduce a new d
> > The point is that the IDT address itself should not be mutable state if
> > it can be at all avoided.
>
> Hmm, maybe we can do it. Have two counters, a debug_idt_ctr and a
> trace_idt_ctr, then have a function that basically does this:
>
> if (this_cpu_read(debug_idt_ctr))
>
> Yeah, I believe this does work. But you probably should add a comment
> like the following:
OK. I will add some comment above " extern atomic_long_t
current_idt_descr_ptr;".
>
> /*
> * The current_idt_descr_ptr can only be set out of interrupt context
> * to avoid races.
I will introduc
in; LKML; Thomas Gleixner; Frederic Weisbecker; Andrew
> Morton; Seiji Aguchi; Borislav Petkov (b...@alien8.de)
> Subject: [PATCH (v2 - fixup)] x86: Have debug/nmi restore the IDT it replaced
>
> I've been playing with Seiji's patches:
>
> https://lkml.org/lkml/2013
e() called at regstering time of irq_vector handlers.
- Switch IDT to new one at enabling TP time.
- Restore to an original IDT at disabling TP time.
The new IDT is created only when CONFIG_TRACING is enabled to avoid being used
for other purposes.
Signed-off-by: Seiji Aguchi
---
arch/x86/i
use it has to check if
it
was already called or not. The time penalty is not desired in performance
sensitive
paths even if it is tiny.
Signed-off-by: Seiji Aguchi
---
arch/x86/include/asm/apic.h | 27 +++
1 files changed, 27 insertions(+), 0 deletions(-)
diff --git
This unfortunately
causes a lot of wasted duplicate code created.
By adding a DEFINE_EVENT_FN(), these events can still use a
DECLARE_EVENT_CLASS() and then define their own triggers.
Signed-off-by: Steven Rostedt
Signed-off-by: Seiji Aguchi
---
include/linux/tracepoint.h |2 ++
include/
nction_single_vector
- irq_work_entry_vector
- error_apic_vector
- thermal_apic_vector
- threshold_apic_vector
- spurious_apic_vector
- x86_platform_ipi_vector
Please see descriptions in each patch.
Seiji Aguchi (2):
trace,x86: Introduce entering/exiting_irq()
trace,x86: Add irq vector tracepo
anks. I will update this patchset as above.
Seiji
> -Original Message-
> From: Steven Rostedt [mailto:rost...@goodmis.org]
> Sent: Thursday, May 23, 2013 3:55 PM
> To: Seiji Aguchi
> Cc: linux-kernel@vger.kernel.org; x...@kernel.org; h...@zytor.com; Thomas
> Glei
ussed in another thread as you posted "[PATCH]
x86: Have debug/nmi restore the IDT it replaced"?
Anyway, I will update my patch in accordance with your comments.
Seiji
> -----Original Message-
> From: Steven Rostedt [mailto:rost...@goodmis.org]
> Sent: Thursday
Steven, Peter
When can I get feedback?
Seiji
> -Original Message-
> From: Seiji Aguchi
> Sent: Friday, April 05, 2013 3:16 PM
> To: linux-kernel@vger.kernel.org; x...@kernel.org; rost...@goodmis.org;
> h...@zytor.com
> Cc: Thomas Gleixner (t...@linutronix.de); &
1 - 100 of 299 matches
Mail list logo
