Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

* Jiri Kosina wrote: > On Wed, 10 Aug 2016, Ingo Molnar wrote: > > > > ENTRY(__sw_hweight64) > > > -#ifdef CONFIG_X86_64 > > > + > > > > So this removes an #ifdef without removing the #else branch. > > Hi Ingo, > > this patch is obsolete; the real issue has already been

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

* Jiri Kosina wrote: > On Wed, 10 Aug 2016, Ingo Molnar wrote: > > > > ENTRY(__sw_hweight64) > > > -#ifdef CONFIG_X86_64 > > > + > > > > So this removes an #ifdef without removing the #else branch. > > Hi Ingo, > > this patch is obsolete; the real issue has already been fixed by 65ea11ec6

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

* Jiri Kosina wrote: > ENTRY(__sw_hweight64) > -#ifdef CONFIG_X86_64 > + So this removes an #ifdef without removing the #else branch. Help? Thanks, Ingo

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

* Jiri Kosina wrote: > ENTRY(__sw_hweight64) > -#ifdef CONFIG_X86_64 > + So this removes an #ifdef without removing the #else branch. Help? Thanks, Ingo

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

On Wed, 10 Aug 2016, Ingo Molnar wrote: > > ENTRY(__sw_hweight64) > > -#ifdef CONFIG_X86_64 > > + > > So this removes an #ifdef without removing the #else branch. Hi Ingo, this patch is obsolete; the real issue has already been fixed by 65ea11ec6 ("x86/hweight: Don't clobber %rdi") in Linus'

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

On Wed, 10 Aug 2016, Ingo Molnar wrote: > > ENTRY(__sw_hweight64) > > -#ifdef CONFIG_X86_64 > > + > > So this removes an #ifdef without removing the #else branch. Hi Ingo, this patch is obsolete; the real issue has already been fixed by 65ea11ec6 ("x86/hweight: Don't clobber %rdi") in Linus'

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

On Mon, 8 Aug 2016, Borislav Petkov wrote: > > Okay, scratch this thread, it's been just covered (and Linus already > > applied it) in > > > > > > https://lkml.kernel.org/r/1470677729-10561-1-git-send-email-ville.syrj...@linux.intel.com > > Can you confirm it fixes the issue on your

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

On Mon, 8 Aug 2016, Borislav Petkov wrote: > > Okay, scratch this thread, it's been just covered (and Linus already > > applied it) in > > > > > > https://lkml.kernel.org/r/1470677729-10561-1-git-send-email-ville.syrj...@linux.intel.com > > Can you confirm it fixes the issue on your

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

On Mon, Aug 08, 2016 at 08:48:44PM +0200, Jiri Kosina wrote: > Okay, scratch this thread, it's been just covered (and Linus already > applied it) in > > > https://lkml.kernel.org/r/1470677729-10561-1-git-send-email-ville.syrj...@linux.intel.com Can you confirm it fixes the issue on your

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

On Mon, Aug 08, 2016 at 08:48:44PM +0200, Jiri Kosina wrote: > Okay, scratch this thread, it's been just covered (and Linus already > applied it) in > > > https://lkml.kernel.org/r/1470677729-10561-1-git-send-email-ville.syrj...@linux.intel.com Can you confirm it fixes the issue on your

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

Okay, scratch this thread, it's been just covered (and Linus already applied it) in https://lkml.kernel.org/r/1470677729-10561-1-git-send-email-ville.syrj...@linux.intel.com -- Jiri Kosina SUSE Labs

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

Okay, scratch this thread, it's been just covered (and Linus already applied it) in https://lkml.kernel.org/r/1470677729-10561-1-git-send-email-ville.syrj...@linux.intel.com -- Jiri Kosina SUSE Labs

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

Hi Jiri, [auto build test ERROR on tip/auto-latest] [also build test ERROR on v4.8-rc1 next-20160808] [cannot apply to tip/x86/core] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] url:

Re: [PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

Hi Jiri, [auto build test ERROR on tip/auto-latest] [also build test ERROR on v4.8-rc1 next-20160808] [cannot apply to tip/x86/core] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] url:

[PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

From: Jiri Kosina The open-coded version of __sw_hweight32() and __sw_hweight64() are broken; the variable <-> register mapping in the comments doesn't match the registers being used, resulting in wrong calculations. This has a potential to demonstrate itself by various

[PATCH v2] x86/hweight: fix open-coded versions of 32bit and 64bit hweight calculations

From: Jiri Kosina The open-coded version of __sw_hweight32() and __sw_hweight64() are broken; the variable <-> register mapping in the comments doesn't match the registers being used, resulting in wrong calculations. This has a potential to demonstrate itself by various syptoms on machines