Re: ule+smp: small optimization for turnstile priority lending

2012-11-09 Thread Andre Oppermann
On 07.11.2012 08:45, David Xu wrote: On 2012/11/07 14:17, Jeff Roberson wrote: On Wed, 7 Nov 2012, David Xu wrote: On 2012/11/06 19:03, Attilio Rao wrote: On 9/20/12, David Xu davi...@freebsd.org wrote: I found another scenario in taskqueue, in the function taskqueue_terminate, current

Re: ule+smp: small optimization for turnstile priority lending

2012-11-07 Thread Jeff Roberson
On Wed, 7 Nov 2012, David Xu wrote: On 2012/11/06 19:03, Attilio Rao wrote: On 9/20/12, David Xu davi...@freebsd.org wrote: On 2012/09/18 22:05, Andriy Gapon wrote: Here is a snippet that demonstrates the issue on a supposedly fully loaded 2-processor system: 136794 0 3670427870244462

Re: ule+smp: small optimization for turnstile priority lending

2012-11-06 Thread Attilio Rao
On 10/29/12, Attilio Rao atti...@freebsd.org wrote: On Mon, Oct 29, 2012 at 4:56 PM, Attilio Rao atti...@freebsd.org wrote: On Wed, Oct 3, 2012 at 3:05 PM, Andriy Gapon a...@freebsd.org wrote: on 20/09/2012 16:14 Attilio Rao said the following: On 9/20/12, Andriy Gapon a...@freebsd.org wrote:

Re: ule+smp: small optimization for turnstile priority lending

2012-11-06 Thread Attilio Rao
On 9/20/12, David Xu davi...@freebsd.org wrote: On 2012/09/18 22:05, Andriy Gapon wrote: Here is a snippet that demonstrates the issue on a supposedly fully loaded 2-processor system: 136794 0 3670427870244462 KTRGRAPH group:thread, id:Xorg tid 102818, state:running, attributes:

Re: ule+smp: small optimization for turnstile priority lending

2012-11-06 Thread David Xu
On 2012/11/06 19:03, Attilio Rao wrote: On 9/20/12, David Xu davi...@freebsd.org wrote: On 2012/09/18 22:05, Andriy Gapon wrote: Here is a snippet that demonstrates the issue on a supposedly fully loaded 2-processor system: 136794 0 3670427870244462 KTRGRAPH group:thread, id:Xorg tid

Re: ule+smp: small optimization for turnstile priority lending

2012-11-06 Thread David Xu
On 2012/11/07 14:17, Jeff Roberson wrote: On Wed, 7 Nov 2012, David Xu wrote: On 2012/11/06 19:03, Attilio Rao wrote: On 9/20/12, David Xu davi...@freebsd.org wrote: On 2012/09/18 22:05, Andriy Gapon wrote: Here is a snippet that demonstrates the issue on a supposedly fully loaded

Re: ule+smp: small optimization for turnstile priority lending

2012-10-29 Thread Attilio Rao
On Wed, Oct 3, 2012 at 3:05 PM, Andriy Gapon a...@freebsd.org wrote: on 20/09/2012 16:14 Attilio Rao said the following: On 9/20/12, Andriy Gapon a...@freebsd.org wrote: [snip] The patch works well as far as I can tell. Thank you! There is one warning with full witness enables but it appears

Re: ule+smp: small optimization for turnstile priority lending

2012-10-29 Thread Attilio Rao
On Mon, Oct 29, 2012 at 4:56 PM, Attilio Rao atti...@freebsd.org wrote: On Wed, Oct 3, 2012 at 3:05 PM, Andriy Gapon a...@freebsd.org wrote: on 20/09/2012 16:14 Attilio Rao said the following: On 9/20/12, Andriy Gapon a...@freebsd.org wrote: [snip] The patch works well as far as I can tell.

Re: ule+smp: small optimization for turnstile priority lending

2012-10-29 Thread Attilio Rao
On Mon, Oct 29, 2012 at 6:59 PM, Attilio Rao atti...@freebsd.org wrote: [ trimm ] This also let me wonder about the tdq_lowpri check in the self case, in general. Basically it forces sched_pickcpu() to select self if and only if the new thread to schedule has an higher priority than the

Re: ule+smp: small optimization for turnstile priority lending

2012-10-03 Thread Andriy Gapon
on 20/09/2012 16:14 Attilio Rao said the following: On 9/20/12, Andriy Gapon a...@freebsd.org wrote: [snip] The patch works well as far as I can tell. Thank you! There is one warning with full witness enables but it appears to be harmless (so far): Andriy, thanks a lot for your testing

Re: ule+smp: small optimization for turnstile priority lending

2012-09-20 Thread David Xu
On 2012/09/18 22:05, Andriy Gapon wrote: Here is a snippet that demonstrates the issue on a supposedly fully loaded 2-processor system: 136794 0 3670427870244462 KTRGRAPH group:thread, id:Xorg tid 102818, state:running, attributes: prio:122 136793 0 3670427870241000 KTRGRAPH group:thread,

Re: ule+smp: small optimization for turnstile priority lending

2012-09-20 Thread Andriy Gapon
on 19/09/2012 10:33 Attilio Rao said the following: It is hard for me to tell if this is subject to same issues because I do not entirely understand where the locking was happening in your patch. Can you try testing this with your already KTR instrumented test and possibly report? The patch

Re: ule+smp: small optimization for turnstile priority lending

2012-09-20 Thread Andriy Gapon
on 20/09/2012 14:04 Andriy Gapon said the following: on 19/09/2012 10:33 Attilio Rao said the following: It is hard for me to tell if this is subject to same issues because I do not entirely understand where the locking was happening in your patch. Can you try testing this with your already

Re: ule+smp: small optimization for turnstile priority lending

2012-09-20 Thread Andriy Gapon
on 20/09/2012 14:04 Andriy Gapon said the following: on 19/09/2012 10:33 Attilio Rao said the following: It is hard for me to tell if this is subject to same issues because I do not entirely understand where the locking was happening in your patch. Can you try testing this with your already

Re: ule+smp: small optimization for turnstile priority lending

2012-09-20 Thread Attilio Rao
On 9/20/12, Andriy Gapon a...@freebsd.org wrote: on 19/09/2012 10:33 Attilio Rao said the following: It is hard for me to tell if this is subject to same issues because I do not entirely understand where the locking was happening in your patch. Can you try testing this with your already KTR

Re: ule+smp: small optimization for turnstile priority lending

2012-09-19 Thread Andriy Gapon
on 19/09/2012 02:01 Attilio Rao said the following: On Tue, Sep 18, 2012 at 8:07 PM, Andriy Gapon a...@freebsd.org wrote: on 18/09/2012 19:50 Attilio Rao said the following: On 9/18/12, Andriy Gapon a...@freebsd.org wrote: Here is a snippet that demonstrates the issue on a supposedly fully

Re: ule+smp: small optimization for turnstile priority lending

2012-09-19 Thread Attilio Rao
On Wed, Sep 19, 2012 at 7:03 AM, Andriy Gapon a...@freebsd.org wrote: on 19/09/2012 02:01 Attilio Rao said the following: On Tue, Sep 18, 2012 at 8:07 PM, Andriy Gapon a...@freebsd.org wrote: on 18/09/2012 19:50 Attilio Rao said the following: On 9/18/12, Andriy Gapon a...@freebsd.org wrote:

Re: ule+smp: small optimization for turnstile priority lending

2012-09-18 Thread Attilio Rao
On 9/18/12, Andriy Gapon a...@freebsd.org wrote: Here is a snippet that demonstrates the issue on a supposedly fully loaded 2-processor system: 136794 0 3670427870244462 KTRGRAPH group:thread, id:Xorg tid 102818, state:running, attributes: prio:122 136793 0 3670427870241000 KTRGRAPH

Re: ule+smp: small optimization for turnstile priority lending

2012-09-18 Thread Andriy Gapon
on 18/09/2012 19:50 Attilio Rao said the following: On 9/18/12, Andriy Gapon a...@freebsd.org wrote: Here is a snippet that demonstrates the issue on a supposedly fully loaded 2-processor system: 136794 0 3670427870244462 KTRGRAPH group:thread, id:Xorg tid 102818, state:running,

Re: ule+smp: small optimization for turnstile priority lending

2012-09-18 Thread Jeff Roberson
On Tue, 18 Sep 2012, Andriy Gapon wrote: on 18/09/2012 19:50 Attilio Rao said the following: On 9/18/12, Andriy Gapon a...@freebsd.org wrote: Here is a snippet that demonstrates the issue on a supposedly fully loaded 2-processor system: 136794 0 3670427870244462 KTRGRAPH group:thread,

Re: ule+smp: small optimization for turnstile priority lending

2012-09-18 Thread Attilio Rao
On Tue, Sep 18, 2012 at 8:07 PM, Andriy Gapon a...@freebsd.org wrote: on 18/09/2012 19:50 Attilio Rao said the following: On 9/18/12, Andriy Gapon a...@freebsd.org wrote: Here is a snippet that demonstrates the issue on a supposedly fully loaded 2-processor system: 136794 0