Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

On Tue, Nov 08, 2016 at 03:54:55PM +0100, Ingo Molnar wrote: > This one you gave: > > > No affect on current hw - just a cleanup. > > Nothing in the existing changelog (including the title) explained that detail. Sounds to me you want to reintroduce the Impact: tagging. :-))) /me ducks Ok,

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

On Tue, Nov 08, 2016 at 03:54:55PM +0100, Ingo Molnar wrote: > This one you gave: > > > No affect on current hw - just a cleanup. > > Nothing in the existing changelog (including the title) explained that detail. Sounds to me you want to reintroduce the Impact: tagging. :-))) /me ducks Ok,

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

* Borislav Petkov wrote: > On Tue, Nov 08, 2016 at 11:29:09AM +0100, Ingo Molnar wrote: > > Looks good to me! Please also update the second patch with meta > > information and I can apply them. > > What exactly do you want to have there? I think the commit message is > pretty

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

* Borislav Petkov wrote: > On Tue, Nov 08, 2016 at 11:29:09AM +0100, Ingo Molnar wrote: > > Looks good to me! Please also update the second patch with meta > > information and I can apply them. > > What exactly do you want to have there? I think the commit message is > pretty explanatory.

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

On Tue, Nov 08, 2016 at 11:29:09AM +0100, Ingo Molnar wrote: > Looks good to me! Please also update the second patch with meta > information and I can apply them. What exactly do you want to have there? I think the commit message is pretty explanatory. -- Regards/Gruss, Boris. SUSE Linux

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

On Tue, Nov 08, 2016 at 11:29:09AM +0100, Ingo Molnar wrote: > Looks good to me! Please also update the second patch with meta > information and I can apply them. What exactly do you want to have there? I think the commit message is pretty explanatory. -- Regards/Gruss, Boris. SUSE Linux

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

* Borislav Petkov wrote: > On Tue, Nov 08, 2016 at 07:31:45AM +0100, Ingo Molnar wrote: > > So the point I tried to make is that to people doing -stable > > backporting decisions this description you just gave is much more > > valuable than the previous changelog. > > Ok, how's

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

* Borislav Petkov wrote: > On Tue, Nov 08, 2016 at 07:31:45AM +0100, Ingo Molnar wrote: > > So the point I tried to make is that to people doing -stable > > backporting decisions this description you just gave is much more > > valuable than the previous changelog. > > Ok, how's that below?

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

On Tue, Nov 08, 2016 at 07:31:45AM +0100, Ingo Molnar wrote: > So the point I tried to make is that to people doing -stable > backporting decisions this description you just gave is much more > valuable than the previous changelog. Ok, how's that below? I've integrated the gist of it in the

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

On Tue, Nov 08, 2016 at 07:31:45AM +0100, Ingo Molnar wrote: > So the point I tried to make is that to people doing -stable > backporting decisions this description you just gave is much more > valuable than the previous changelog. Ok, how's that below? I've integrated the gist of it in the

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

* Borislav Petkov wrote: > On Mon, Nov 07, 2016 at 03:07:46PM +0100, Ingo Molnar wrote: > > - cache domains might be seriously mixed up, resulting in serious drop in > >performance. > > > > - or domains might be partitioned 'wrong' but not catastrophically > > wrong,

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

* Borislav Petkov wrote: > On Mon, Nov 07, 2016 at 03:07:46PM +0100, Ingo Molnar wrote: > > - cache domains might be seriously mixed up, resulting in serious drop in > >performance. > > > > - or domains might be partitioned 'wrong' but not catastrophically > > wrong, resulting in a

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

On Mon, Nov 07, 2016 at 03:07:46PM +0100, Ingo Molnar wrote: > - cache domains might be seriously mixed up, resulting in serious drop in >performance. > > - or domains might be partitioned 'wrong' but not catastrophically > wrong, resulting in a minor performance drop (if at all)

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

On Mon, Nov 07, 2016 at 03:07:46PM +0100, Ingo Molnar wrote: > - cache domains might be seriously mixed up, resulting in serious drop in >performance. > > - or domains might be partitioned 'wrong' but not catastrophically > wrong, resulting in a minor performance drop (if at all)

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

* Borislav Petkov wrote: > On Mon, Nov 07, 2016 at 08:31:21AM +0100, Ingo Molnar wrote: > > > cpu_llc_id (Last Level Cache ID) derivation on AMD Fam17h has an > > > underflow bug when extracting the socket_id value. It starts from 0 > > How's this... > > > > so subtracting 1

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

* Borislav Petkov wrote: > On Mon, Nov 07, 2016 at 08:31:21AM +0100, Ingo Molnar wrote: > > > cpu_llc_id (Last Level Cache ID) derivation on AMD Fam17h has an > > > underflow bug when extracting the socket_id value. It starts from 0 > > How's this... > > > > so subtracting 1 from it will

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

On Mon, Nov 07, 2016 at 08:31:21AM +0100, Ingo Molnar wrote: > > cpu_llc_id (Last Level Cache ID) derivation on AMD Fam17h has an > > underflow bug when extracting the socket_id value. It starts from 0 How's this... > > so subtracting 1 from it will result in an invalid value. This breaks > >

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

On Mon, Nov 07, 2016 at 08:31:21AM +0100, Ingo Molnar wrote: > > cpu_llc_id (Last Level Cache ID) derivation on AMD Fam17h has an > > underflow bug when extracting the socket_id value. It starts from 0 How's this... > > so subtracting 1 from it will result in an invalid value. This breaks > >

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

* Borislav Petkov wrote: > Lemme clean up the commit message a bit more and add tags: > > --- > From: Yazen Ghannam > Date: Tue, 1 Nov 2016 11:51:02 -0500 > Subject: [PATCH] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems > > cpu_llc_id (Last Level Cache

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

* Borislav Petkov wrote: > Lemme clean up the commit message a bit more and add tags: > > --- > From: Yazen Ghannam > Date: Tue, 1 Nov 2016 11:51:02 -0500 > Subject: [PATCH] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems > > cpu_llc_id (Last Level Cache ID) derivation on AMD Fam17h has an >

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

Lemme clean up the commit message a bit more and add tags: --- From: Yazen Ghannam Date: Tue, 1 Nov 2016 11:51:02 -0500 Subject: [PATCH] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems cpu_llc_id (Last Level Cache ID) derivation on AMD Fam17h has an underflow bug when

Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

Lemme clean up the commit message a bit more and add tags: --- From: Yazen Ghannam Date: Tue, 1 Nov 2016 11:51:02 -0500 Subject: [PATCH] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems cpu_llc_id (Last Level Cache ID) derivation on AMD Fam17h has an underflow bug when extracting the socket_id

[PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

The current Fam17h cpu_llc_id (Last Level Cache ID) derivation has an underflow bug when extracting the socket_id value. The socket_id value starts from 0, so subtracting 1 will result in an underflow. This breaks scheduling topology later on since the cpu_llc_id will be incorrect. The APICID

[PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems

The current Fam17h cpu_llc_id (Last Level Cache ID) derivation has an underflow bug when extracting the socket_id value. The socket_id value starts from 0, so subtracting 1 will result in an underflow. This breaks scheduling topology later on since the cpu_llc_id will be incorrect. The APICID