Re: Another preempt folding issue?

On 24.03.2014 18:39, Paolo Bonzini wrote: > Il 20/02/2014 16:50, Peter Zijlstra ha scritto: > > >> One thing I likely should do is to reinstall the exact same laptop > with 64bit > > >> kernel and userspace... maybe only 64bit kernel first... and make > > >> sure > on my >

Re: Another preempt folding issue?

On 24.03.2014 18:39, Paolo Bonzini wrote: Il 20/02/2014 16:50, Peter Zijlstra ha scritto: One thing I likely should do is to reinstall the exact same laptop with 64bit kernel and userspace... maybe only 64bit kernel first... and make sure on my side that this does not show up on

Re: Another preempt folding issue?

Il 20/02/2014 16:50, Peter Zijlstra ha scritto: > >> One thing I likely should do is to reinstall the exact same laptop with 64bit > >> kernel and userspace... maybe only 64bit kernel first... and make sure on my > >> side that this does not show up on 64bit, too. I took the word of reporters

Re: Another preempt folding issue?

Il 20/02/2014 16:50, Peter Zijlstra ha scritto: One thing I likely should do is to reinstall the exact same laptop with 64bit kernel and userspace... maybe only 64bit kernel first... and make sure on my side that this does not show up on 64bit, too. I took the word of reporters for

Re: Another preempt folding issue?

On Thu, Feb 20, 2014 at 04:38:13PM +0100, Stefan Bader wrote: > On 14.02.2014 18:21, Peter Zijlstra wrote: > > On Fri, Feb 14, 2014 at 06:02:32PM +0100, Stefan Bader wrote: > >> One thing I likely should do is to reinstall the exact same laptop with > >> 64bit > >> kernel and userspace... maybe

Re: Another preempt folding issue?

On 14.02.2014 18:21, Peter Zijlstra wrote: > On Fri, Feb 14, 2014 at 06:02:32PM +0100, Stefan Bader wrote: >> One thing I likely should do is to reinstall the exact same laptop with 64bit >> kernel and userspace... maybe only 64bit kernel first... and make sure on my >> side that this does not

Re: Another preempt folding issue?

On 14.02.2014 18:21, Peter Zijlstra wrote: On Fri, Feb 14, 2014 at 06:02:32PM +0100, Stefan Bader wrote: One thing I likely should do is to reinstall the exact same laptop with 64bit kernel and userspace... maybe only 64bit kernel first... and make sure on my side that this does not show up on

Re: Another preempt folding issue?

On Thu, Feb 20, 2014 at 04:38:13PM +0100, Stefan Bader wrote: On 14.02.2014 18:21, Peter Zijlstra wrote: On Fri, Feb 14, 2014 at 06:02:32PM +0100, Stefan Bader wrote: One thing I likely should do is to reinstall the exact same laptop with 64bit kernel and userspace... maybe only 64bit

Re: Another preempt folding issue?

On 14.02.2014 19:23, Stefan Bader wrote: > On 14.02.2014 18:33, Borislav Petkov wrote: >> On Fri, Feb 14, 2014 at 06:02:32PM +0100, Stefan Bader wrote: >>> Okaaay, I think I did what you asked. So yes, there is sse2 in the cpu >>> info. And >>> there is a mfence in the disassembly: >> >> Btw, I

Re: Another preempt folding issue?

On 14.02.2014 18:33, Borislav Petkov wrote: > On Fri, Feb 14, 2014 at 06:02:32PM +0100, Stefan Bader wrote: >> Okaaay, I think I did what you asked. So yes, there is sse2 in the cpu info. >> And >> there is a mfence in the disassembly: > > Btw, I just realized booting the kernel in the guest was

Re: Another preempt folding issue?

On Fri, Feb 14, 2014 at 06:02:32PM +0100, Stefan Bader wrote: > Okaaay, I think I did what you asked. So yes, there is sse2 in the cpu info. > And > there is a mfence in the disassembly: Btw, I just realized booting the kernel in the guest was a dumb idea, because, doh, the guest is not

Re: Another preempt folding issue?

On Fri, Feb 14, 2014 at 06:02:32PM +0100, Stefan Bader wrote: > One thing I likely should do is to reinstall the exact same laptop with 64bit > kernel and userspace... maybe only 64bit kernel first... and make sure on my > side that this does not show up on 64bit, too. I took the word of reporters

Re: Another preempt folding issue?

On 14.02.2014 15:47, Borislav Petkov wrote: > On Fri, Feb 14, 2014 at 03:24:09PM +0100, Stefan Bader wrote: >> Actually, this code just makes so much more sense if I let objdump do >> relocation info... > > Ok, we're pretty sure you have an MFENCE there in resched_task but can > you confirm it

Re: Another preempt folding issue? (maybe bisect)

On Fri, Feb 14, 2014 at 04:21:32PM +0100, Borislav Petkov wrote: > Oh, and just in case this is relatively easy to reproduce and in case we > don't have any other idea, bisection might be another option. I'm not > saying you should do it right away - I'm just putting it on the table... I'm fairly

Re: Another preempt folding issue? (maybe bisect)

On Fri, Feb 14, 2014 at 04:28:16PM +0100, Stefan Bader wrote: > Oh yeah, bisection is nearly as entertaining as doing my tax records. > Hm, on the other hand those will have to be done at some point too... > :-P Ah, tax records, crap, I have those lying around in the corner since forever... I

Re: Another preempt folding issue? (maybe bisect)

On 14.02.2014 16:21, Borislav Petkov wrote: > Oh, and just in case this is relatively easy to reproduce and in case we > don't have any other idea, bisection might be another option. I'm not > saying you should do it right away - I'm just putting it on the table... > > :-) > > :-) > Oh yeah,

Re: Another preempt folding issue? (maybe bisect)

Oh, and just in case this is relatively easy to reproduce and in case we don't have any other idea, bisection might be another option. I'm not saying you should do it right away - I'm just putting it on the table... :-) :-) -- Regards/Gruss, Boris. Sent from a fat crate under my desk.

Re: Another preempt folding issue?

On Fri, Feb 14, 2014 at 03:24:09PM +0100, Stefan Bader wrote: > Actually, this code just makes so much more sense if I let objdump do > relocation info... Ok, we're pretty sure you have an MFENCE there in resched_task but can you confirm it please. First, does /proc/cpuinfo have the "sse2"

Re: Another preempt folding issue?

On Thu, Feb 13, 2014 at 07:26:05PM +0100, Peter Zijlstra wrote: > On Thu, Feb 13, 2014 at 07:03:56PM +0100, Stefan Bader wrote: > > Yeah... not sure the interleaved source helps or not ... > > It did, thanks! Stefan, can you also send sched/core.s? I'm particularly interested in what

Re: Another preempt folding issue?

On Fri, Feb 14, 2014 at 11:55:25AM +0100, Stefan Bader wrote: > Ok, I think I now got a log of the actual issue. It seems cpu#1 missed out on > handling a reschedule interrupt but did send one to cpu#0 and on cpu#0 while > handling the interrupt the tif flag was not set (yet?) but then when it is,

Re: Another preempt folding issue?

On Fri, Feb 14, 2014 at 12:24:42PM +0100, Stefan Bader wrote: > Oh and one thing I was wondering. Not sure I do understand it right... When > initially converting to percpu counts, you changed the 32bit assembly like > that: > > --- a/arch/x86/kernel/entry_32.S > +++ b/arch/x86/kernel/entry_32.S

Re: Another preempt folding issue?

On 13.02.2014 19:25, Peter Zijlstra wrote: > On Thu, Feb 13, 2014 at 06:00:19PM +0100, Stefan Bader wrote: >> On 12.02.2014 12:54, Peter Zijlstra wrote: >>> On Wed, Feb 12, 2014 at 12:09:29PM +0100, Stefan Bader wrote: Something else here I run a kernel with CONFIG_PREEMPT not set and NR_CPUS

Re: Another preempt folding issue?

On 13.02.2014 19:25, Peter Zijlstra wrote: > On Thu, Feb 13, 2014 at 06:00:19PM +0100, Stefan Bader wrote: >> On 12.02.2014 12:54, Peter Zijlstra wrote: >>> On Wed, Feb 12, 2014 at 12:09:29PM +0100, Stefan Bader wrote: Something else here I run a kernel with CONFIG_PREEMPT not set and NR_CPUS

Re: Another preempt folding issue?

On 13.02.2014 19:25, Peter Zijlstra wrote: On Thu, Feb 13, 2014 at 06:00:19PM +0100, Stefan Bader wrote: On 12.02.2014 12:54, Peter Zijlstra wrote: On Wed, Feb 12, 2014 at 12:09:29PM +0100, Stefan Bader wrote: Something else here I run a kernel with CONFIG_PREEMPT not set and NR_CPUS limited

Re: Another preempt folding issue?

On 13.02.2014 19:25, Peter Zijlstra wrote: On Thu, Feb 13, 2014 at 06:00:19PM +0100, Stefan Bader wrote: On 12.02.2014 12:54, Peter Zijlstra wrote: On Wed, Feb 12, 2014 at 12:09:29PM +0100, Stefan Bader wrote: Something else here I run a kernel with CONFIG_PREEMPT not set and NR_CPUS limited

Re: Another preempt folding issue?

On Fri, Feb 14, 2014 at 12:24:42PM +0100, Stefan Bader wrote: Oh and one thing I was wondering. Not sure I do understand it right... When initially converting to percpu counts, you changed the 32bit assembly like that: --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@

Re: Another preempt folding issue?

On Fri, Feb 14, 2014 at 11:55:25AM +0100, Stefan Bader wrote: Ok, I think I now got a log of the actual issue. It seems cpu#1 missed out on handling a reschedule interrupt but did send one to cpu#0 and on cpu#0 while handling the interrupt the tif flag was not set (yet?) but then when it is,

Re: Another preempt folding issue?

On Thu, Feb 13, 2014 at 07:26:05PM +0100, Peter Zijlstra wrote: On Thu, Feb 13, 2014 at 07:03:56PM +0100, Stefan Bader wrote: Yeah... not sure the interleaved source helps or not ... It did, thanks! Stefan, can you also send sched/core.s? I'm particularly interested in what resched_task()

Re: Another preempt folding issue?

On Fri, Feb 14, 2014 at 03:24:09PM +0100, Stefan Bader wrote: Actually, this code just makes so much more sense if I let objdump do relocation info... Ok, we're pretty sure you have an MFENCE there in resched_task but can you confirm it please. First, does /proc/cpuinfo have the sse2 string?

Re: Another preempt folding issue? (maybe bisect)

Oh, and just in case this is relatively easy to reproduce and in case we don't have any other idea, bisection might be another option. I'm not saying you should do it right away - I'm just putting it on the table... :-) :-) -- Regards/Gruss, Boris. Sent from a fat crate under my desk.

Re: Another preempt folding issue? (maybe bisect)

On 14.02.2014 16:21, Borislav Petkov wrote: Oh, and just in case this is relatively easy to reproduce and in case we don't have any other idea, bisection might be another option. I'm not saying you should do it right away - I'm just putting it on the table... :-) :-) Oh yeah, bisection

Re: Another preempt folding issue? (maybe bisect)

On Fri, Feb 14, 2014 at 04:28:16PM +0100, Stefan Bader wrote: Oh yeah, bisection is nearly as entertaining as doing my tax records. Hm, on the other hand those will have to be done at some point too... :-P Ah, tax records, crap, I have those lying around in the corner since forever... I need

Re: Another preempt folding issue? (maybe bisect)

On Fri, Feb 14, 2014 at 04:21:32PM +0100, Borislav Petkov wrote: Oh, and just in case this is relatively easy to reproduce and in case we don't have any other idea, bisection might be another option. I'm not saying you should do it right away - I'm just putting it on the table... I'm fairly

Re: Another preempt folding issue?

On 14.02.2014 15:47, Borislav Petkov wrote: On Fri, Feb 14, 2014 at 03:24:09PM +0100, Stefan Bader wrote: Actually, this code just makes so much more sense if I let objdump do relocation info... Ok, we're pretty sure you have an MFENCE there in resched_task but can you confirm it please.

Re: Another preempt folding issue?

On Fri, Feb 14, 2014 at 06:02:32PM +0100, Stefan Bader wrote: One thing I likely should do is to reinstall the exact same laptop with 64bit kernel and userspace... maybe only 64bit kernel first... and make sure on my side that this does not show up on 64bit, too. I took the word of reporters

Re: Another preempt folding issue?

On Fri, Feb 14, 2014 at 06:02:32PM +0100, Stefan Bader wrote: Okaaay, I think I did what you asked. So yes, there is sse2 in the cpu info. And there is a mfence in the disassembly: Btw, I just realized booting the kernel in the guest was a dumb idea, because, doh, the guest is not baremetal.

Re: Another preempt folding issue?

On 14.02.2014 18:33, Borislav Petkov wrote: On Fri, Feb 14, 2014 at 06:02:32PM +0100, Stefan Bader wrote: Okaaay, I think I did what you asked. So yes, there is sse2 in the cpu info. And there is a mfence in the disassembly: Btw, I just realized booting the kernel in the guest was a dumb

Re: Another preempt folding issue?

On 14.02.2014 19:23, Stefan Bader wrote: On 14.02.2014 18:33, Borislav Petkov wrote: On Fri, Feb 14, 2014 at 06:02:32PM +0100, Stefan Bader wrote: Okaaay, I think I did what you asked. So yes, there is sse2 in the cpu info. And there is a mfence in the disassembly: Btw, I just realized

Re: Another preempt folding issue?

On Thu, Feb 13, 2014 at 07:03:56PM +0100, Stefan Bader wrote: > Yeah... not sure the interleaved source helps or not ... It did, thanks! -- 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

Re: Another preempt folding issue?

On Thu, Feb 13, 2014 at 06:00:19PM +0100, Stefan Bader wrote: > On 12.02.2014 12:54, Peter Zijlstra wrote: > > On Wed, Feb 12, 2014 at 12:09:29PM +0100, Stefan Bader wrote: > >> Something else here I run a kernel with CONFIG_PREEMPT not set and NR_CPUS > >> limited to 8 (for the 32bit kernel). So

Re: Another preempt folding issue?

On Thu, Feb 13, 2014 at 06:00:19PM +0100, Stefan Bader wrote: > On 12.02.2014 12:54, Peter Zijlstra wrote: > > On Wed, Feb 12, 2014 at 12:09:29PM +0100, Stefan Bader wrote: > >> Something else here I run a kernel with CONFIG_PREEMPT not set and NR_CPUS > >> limited to 8 (for the 32bit kernel). So

Re: Another preempt folding issue?

On 12.02.2014 12:54, Peter Zijlstra wrote: > On Wed, Feb 12, 2014 at 12:09:29PM +0100, Stefan Bader wrote: >> Something else here I run a kernel with CONFIG_PREEMPT not set and NR_CPUS >> limited to 8 (for the 32bit kernel). So the default apic driver is used. >> Since >>

Re: Another preempt folding issue?

On 12.02.2014 12:54, Peter Zijlstra wrote: On Wed, Feb 12, 2014 at 12:09:29PM +0100, Stefan Bader wrote: Something else here I run a kernel with CONFIG_PREEMPT not set and NR_CPUS limited to 8 (for the 32bit kernel). So the default apic driver is used. Since default_send_IPI_mask_logical is

Re: Another preempt folding issue?

On Thu, Feb 13, 2014 at 06:00:19PM +0100, Stefan Bader wrote: On 12.02.2014 12:54, Peter Zijlstra wrote: On Wed, Feb 12, 2014 at 12:09:29PM +0100, Stefan Bader wrote: Something else here I run a kernel with CONFIG_PREEMPT not set and NR_CPUS limited to 8 (for the 32bit kernel). So the

Re: Another preempt folding issue?

On Thu, Feb 13, 2014 at 06:00:19PM +0100, Stefan Bader wrote: On 12.02.2014 12:54, Peter Zijlstra wrote: On Wed, Feb 12, 2014 at 12:09:29PM +0100, Stefan Bader wrote: Something else here I run a kernel with CONFIG_PREEMPT not set and NR_CPUS limited to 8 (for the 32bit kernel). So the

Re: Another preempt folding issue?

On Thu, Feb 13, 2014 at 07:03:56PM +0100, Stefan Bader wrote: Yeah... not sure the interleaved source helps or not ... It did, thanks! -- 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

Re: Another preempt folding issue?

On Wed, Feb 12, 2014 at 12:09:29PM +0100, Stefan Bader wrote: > Something else here I run a kernel with CONFIG_PREEMPT not set and NR_CPUS > limited to 8 (for the 32bit kernel). So the default apic driver is used. Since > default_send_IPI_mask_logical is only used from there, I assume the trace

Re: Another preempt folding issue?

On Wed, Feb 12, 2014 at 11:40:17AM +0100, Borislav Petkov wrote: > Also what I'm wondering about and what's not clear from Stefan's reply > is whether this is purely a 32-bit issue, i.e. a 32-bit host running a > 64-bit qemu running a 32-bit iso or what is it? > > Or do we have reports for both

Re: Another preempt folding issue?

On 12.02.2014 11:40, Borislav Petkov wrote: > On Wed, Feb 12, 2014 at 11:37:13AM +0100, Peter Zijlstra wrote: >>> Another reporter also saw this on an AMD and said it could not be >>> reproduced on >>> the same hardware and the same software versions when using 64bit instead >>> of 32. >>> >>>

Re: Another preempt folding issue?

On Wed, Feb 12, 2014 at 11:37:13AM +0100, Peter Zijlstra wrote: > > Another reporter also saw this on an AMD and said it could not be > > reproduced on > > the same hardware and the same software versions when using 64bit instead > > of 32. > > > > In my case on a 32bit installation I will see

Re: Another preempt folding issue?

On Wed, Feb 12, 2014 at 09:20:24AM +0100, Stefan Bader wrote: > On 11.02.2014 20:45, Peter Zijlstra wrote: > > On Tue, Feb 11, 2014 at 07:34:51PM +0100, Stefan Bader wrote: > >> Hi Peter, > >> > >> I am currently looking at a weird issue that manifest itself when trying > >> to run > >> kvm

Re: Another preempt folding issue?

On 11.02.2014 20:45, Peter Zijlstra wrote: > On Tue, Feb 11, 2014 at 07:34:51PM +0100, Stefan Bader wrote: >> Hi Peter, >> >> I am currently looking at a weird issue that manifest itself when trying to >> run >> kvm enabled qemu on a i386 host (v3.13 kernel, oh and potentially important >> the

Re: Another preempt folding issue?

On 11.02.2014 20:45, Peter Zijlstra wrote: On Tue, Feb 11, 2014 at 07:34:51PM +0100, Stefan Bader wrote: Hi Peter, I am currently looking at a weird issue that manifest itself when trying to run kvm enabled qemu on a i386 host (v3.13 kernel, oh and potentially important the cpu is 64bit

Re: Another preempt folding issue?

On Wed, Feb 12, 2014 at 09:20:24AM +0100, Stefan Bader wrote: On 11.02.2014 20:45, Peter Zijlstra wrote: On Tue, Feb 11, 2014 at 07:34:51PM +0100, Stefan Bader wrote: Hi Peter, I am currently looking at a weird issue that manifest itself when trying to run kvm enabled qemu on a i386

Re: Another preempt folding issue?

On Wed, Feb 12, 2014 at 11:37:13AM +0100, Peter Zijlstra wrote: Another reporter also saw this on an AMD and said it could not be reproduced on the same hardware and the same software versions when using 64bit instead of 32. In my case on a 32bit installation I will see this on

Re: Another preempt folding issue?

On 12.02.2014 11:40, Borislav Petkov wrote: On Wed, Feb 12, 2014 at 11:37:13AM +0100, Peter Zijlstra wrote: Another reporter also saw this on an AMD and said it could not be reproduced on the same hardware and the same software versions when using 64bit instead of 32. In my case on a

Re: Another preempt folding issue?

On Wed, Feb 12, 2014 at 11:40:17AM +0100, Borislav Petkov wrote: Also what I'm wondering about and what's not clear from Stefan's reply is whether this is purely a 32-bit issue, i.e. a 32-bit host running a 64-bit qemu running a 32-bit iso or what is it? Or do we have reports for both 32-bit

Re: Another preempt folding issue?

On Wed, Feb 12, 2014 at 12:09:29PM +0100, Stefan Bader wrote: Something else here I run a kernel with CONFIG_PREEMPT not set and NR_CPUS limited to 8 (for the 32bit kernel). So the default apic driver is used. Since default_send_IPI_mask_logical is only used from there, I assume the trace you

Re: Another preempt folding issue?

On Tue, Feb 11, 2014 at 07:34:51PM +0100, Stefan Bader wrote: > Hi Peter, > > I am currently looking at a weird issue that manifest itself when trying to > run > kvm enabled qemu on a i386 host (v3.13 kernel, oh and potentially important > the > cpu is 64bit capable, so qemu-system-x86_64 is

Re: Another preempt folding issue?

On Tue, Feb 11, 2014 at 07:34:51PM +0100, Stefan Bader wrote: Hi Peter, I am currently looking at a weird issue that manifest itself when trying to run kvm enabled qemu on a i386 host (v3.13 kernel, oh and potentially important the cpu is 64bit capable, so qemu-system-x86_64 is called).