Re: [PATCH] UV: fix incorrect tlb flush all issue

2012-09-26 Thread Andrew Morton
On Tue, 25 Sep 2012 07:34:52 -0500 Cliff Wickman wrote: > From: Cliff Wickman > > (this was sent as an ack on 9/13, but with incorrect title and sign-off) > > Ack. > But with the adjustment below. The 'end' argument was not declared long. > > I tested the patch on a UV. > It has the effect

Re: [PATCH] UV: fix incorrect tlb flush all issue

2012-09-26 Thread Ingo Molnar
* Cliff Wickman wrote: > From: Cliff Wickman > > (this was sent as an ack on 9/13, but with incorrect title and sign-off) > > Ack. > But with the adjustment below. The 'end' argument was not declared long. > > I tested the patch on a UV. > It has the effect of either clearing 1 or all TLBs

Re: [PATCH] UV: fix incorrect tlb flush all issue

2012-09-26 Thread Ingo Molnar
* Cliff Wickman c...@sgi.com wrote: From: Cliff Wickman c...@sgi.com (this was sent as an ack on 9/13, but with incorrect title and sign-off) Ack. But with the adjustment below. The 'end' argument was not declared long. I tested the patch on a UV. It has the effect of either

Re: [PATCH] UV: fix incorrect tlb flush all issue

2012-09-26 Thread Andrew Morton
On Tue, 25 Sep 2012 07:34:52 -0500 Cliff Wickman c...@sgi.com wrote: From: Cliff Wickman c...@sgi.com (this was sent as an ack on 9/13, but with incorrect title and sign-off) Ack. But with the adjustment below. The 'end' argument was not declared long. I tested the patch on a UV. It

[PATCH] UV: fix incorrect tlb flush all issue

2012-09-25 Thread Cliff Wickman
From: Cliff Wickman (this was sent as an ack on 9/13, but with incorrect title and sign-off) Ack. But with the adjustment below. The 'end' argument was not declared long. I tested the patch on a UV. It has the effect of either clearing 1 or all TLBs in a cpu. I added some debugging to test

[PATCH] UV: fix incorrect tlb flush all issue

2012-09-25 Thread Cliff Wickman
From: Cliff Wickman c...@sgi.com (this was sent as an ack on 9/13, but with incorrect title and sign-off) Ack. But with the adjustment below. The 'end' argument was not declared long. I tested the patch on a UV. It has the effect of either clearing 1 or all TLBs in a cpu. I added some

Re: [alex....@intel.com: Re: [PATCH] UV: fix incorrect tlb flush all issue]

2012-09-13 Thread Ingo Molnar
* Cliff Wickman wrote: > On Thu, Sep 13, 2012 at 05:53:10PM +0200, Ingo Molnar wrote: > > > > Ack? > > > > Thanks, > > > > Ingo > > Ack. > But with the adjustment below. The 'end' argument was not declared long. Ok, great - mind sending the updated patch properly under a new title

Re: [alex....@intel.com: Re: [PATCH] UV: fix incorrect tlb flush all issue]

2012-09-13 Thread Alex Shi
On 09/14/2012 05:20 AM, Cliff Wickman wrote: > On Thu, Sep 13, 2012 at 05:53:10PM +0200, Ingo Molnar wrote: >> >> Ack? >> >> Thanks, >> >> Ingo > > Ack. > But with the adjustment below. The 'end' argument was not declared long. > > I tested the patch on a UV. > It has the effect of either

Re: [alex....@intel.com: Re: [PATCH] UV: fix incorrect tlb flush all issue]

2012-09-13 Thread Cliff Wickman
On Thu, Sep 13, 2012 at 05:53:10PM +0200, Ingo Molnar wrote: > > Ack? > > Thanks, > > Ingo Ack. But with the adjustment below. The 'end' argument was not declared long. I tested the patch on a UV. It has the effect of either clearing 1 or all TLBs in a cpu. I added some debugging to

Re: [alex....@intel.com: Re: [PATCH] UV: fix incorrect tlb flush all issue]

2012-09-13 Thread Cliff Wickman
On Thu, Sep 13, 2012 at 05:53:10PM +0200, Ingo Molnar wrote: Ack? Thanks, Ingo Ack. But with the adjustment below. The 'end' argument was not declared long. I tested the patch on a UV. It has the effect of either clearing 1 or all TLBs in a cpu. I added some debugging to test for

Re: [alex....@intel.com: Re: [PATCH] UV: fix incorrect tlb flush all issue]

2012-09-13 Thread Alex Shi
On 09/14/2012 05:20 AM, Cliff Wickman wrote: On Thu, Sep 13, 2012 at 05:53:10PM +0200, Ingo Molnar wrote: Ack? Thanks, Ingo Ack. But with the adjustment below. The 'end' argument was not declared long. I tested the patch on a UV. It has the effect of either clearing 1 or all

Re: [alex....@intel.com: Re: [PATCH] UV: fix incorrect tlb flush all issue]

2012-09-13 Thread Ingo Molnar
* Cliff Wickman c...@sgi.com wrote: On Thu, Sep 13, 2012 at 05:53:10PM +0200, Ingo Molnar wrote: Ack? Thanks, Ingo Ack. But with the adjustment below. The 'end' argument was not declared long. Ok, great - mind sending the updated patch properly under a new title (the

Re: [PATCH] UV: fix incorrect tlb flush all issue

2012-09-09 Thread Alex Shi
On 09/07/2012 03:10 PM, Jan Beulich wrote: On 07.09.12 at 07:37, Alex Shi wrote: >> @@ -1113,7 +1114,10 @@ const struct cpumask *uv_flush_tlb_others(const >> struct >> cpumask *cpumask, >> >> record_send_statistics(stat, locals, hubs, remotes, bau_desc); >> >> -

Re: [PATCH] UV: fix incorrect tlb flush all issue

2012-09-09 Thread Alex Shi
On 09/07/2012 03:10 PM, Jan Beulich wrote: On 07.09.12 at 07:37, Alex Shi alex@intel.com wrote: @@ -1113,7 +1114,10 @@ const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask, record_send_statistics(stat, locals, hubs, remotes, bau_desc); -

Re: [PATCH] UV: fix incorrect tlb flush all issue

2012-09-07 Thread Jan Beulich
>>> On 07.09.12 at 07:37, Alex Shi wrote: > @@ -1113,7 +1114,10 @@ const struct cpumask *uv_flush_tlb_others(const struct > cpumask *cpumask, > > record_send_statistics(stat, locals, hubs, remotes, bau_desc); > > - bau_desc->payload.address = start; > + if (!end) So despite

Re: [PATCH] UV: fix incorrect tlb flush all issue

2012-09-07 Thread Jan Beulich
On 07.09.12 at 07:37, Alex Shi alex@intel.com wrote: @@ -1113,7 +1114,10 @@ const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask, record_send_statistics(stat, locals, hubs, remotes, bau_desc); - bau_desc-payload.address = start; + if (!end) So

Re: [PATCH] UV: fix incorrect tlb flush all issue

2012-09-06 Thread Alex Shi
On 09/07/2012 07:11 AM, Andrew Morton wrote: > On Fri, 24 Aug 2012 16:57:35 +0800 > Alex Shi wrote: > >> The flush tlb optimization code has logical issue on UV platform. >> It doesn't flush the full range at all, since it simply >> ignores its 'end' parameter (and hence also the "all"

Re: [PATCH] UV: fix incorrect tlb flush all issue

2012-09-06 Thread Andrew Morton
On Fri, 24 Aug 2012 16:57:35 +0800 Alex Shi wrote: > The flush tlb optimization code has logical issue on UV platform. > It doesn't flush the full range at all, since it simply > ignores its 'end' parameter (and hence also the "all" indicator) > in uv_flush_tlb_others() function. > > This patch

Re: [PATCH] UV: fix incorrect tlb flush all issue

2012-09-06 Thread Andrew Morton
On Fri, 24 Aug 2012 16:57:35 +0800 Alex Shi alex@intel.com wrote: The flush tlb optimization code has logical issue on UV platform. It doesn't flush the full range at all, since it simply ignores its 'end' parameter (and hence also the all indicator) in uv_flush_tlb_others() function.

Re: [PATCH] UV: fix incorrect tlb flush all issue

2012-09-06 Thread Alex Shi
On 09/07/2012 07:11 AM, Andrew Morton wrote: On Fri, 24 Aug 2012 16:57:35 +0800 Alex Shi alex@intel.com wrote: The flush tlb optimization code has logical issue on UV platform. It doesn't flush the full range at all, since it simply ignores its 'end' parameter (and hence also the all

[PATCH] UV: fix incorrect tlb flush all issue

2012-08-24 Thread Alex Shi
The flush tlb optimization code has logical issue on UV platform. It doesn't flush the full range at all, since it simply ignores its 'end' parameter (and hence also the "all" indicator) in uv_flush_tlb_others() function. This patch fixed this issue, but untested due to hardware leaking.

[PATCH] UV: fix incorrect tlb flush all issue

2012-08-24 Thread Alex Shi
The flush tlb optimization code has logical issue on UV platform. It doesn't flush the full range at all, since it simply ignores its 'end' parameter (and hence also the all indicator) in uv_flush_tlb_others() function. This patch fixed this issue, but untested due to hardware leaking.