Re: [PATCH] [RFC] TAINT_PERFORMANCE

2014-08-19 Thread Mike Galbraith
On Wed, 2014-08-20 at 01:26 +0300, Kirill A. Shutemov wrote: > On Tue, Aug 19, 2014 at 02:26:04PM -0700, Dave Hansen wrote: > > + TAINT_PERF_IF(SCHEDSTATS); > > Is SCHEDSTATS really harmful? If your config is minimalist, a tad. If your config is.. Rubenesque (distro), you probably won't noti

Re: [PATCH] [RFC] TAINT_PERFORMANCE

2014-08-19 Thread Christoph Lameter
On Tue, 19 Aug 2014, Laura Abbott wrote: > I nominate CONFIG_DEBUG_PAGEALLOC, CONFIG_SLUB_DEBUG, > CONFIG_SLUB_DEBUG_ON as well since I've wasted days debugging > supposed performance issues where those were on. CONFIG_SLUB_DEBUG is not enabling debugging. It just includes the code to do so at ke

Re: [PATCH] [RFC] TAINT_PERFORMANCE

2014-08-19 Thread Laura Abbott
On 8/19/2014 2:26 PM, Dave Hansen wrote: > From: Dave Hansen > > I have more than once myself been the victim of an accidentally- > enabled kernel config option being mistaken for a true > performance problem. > > I'm sure I've also taken profiles or performance measurements > and assumed they w

Re: [PATCH] [RFC] TAINT_PERFORMANCE

2014-08-19 Thread Kirill A. Shutemov
On Tue, Aug 19, 2014 at 02:26:04PM -0700, Dave Hansen wrote: > > From: Dave Hansen > > I have more than once myself been the victim of an accidentally- > enabled kernel config option being mistaken for a true > performance problem. > > I'm sure I've also taken profiles or performance measuremen

[PATCH] [RFC] TAINT_PERFORMANCE

2014-08-19 Thread Dave Hansen
From: Dave Hansen I have more than once myself been the victim of an accidentally- enabled kernel config option being mistaken for a true performance problem. I'm sure I've also taken profiles or performance measurements and assumed they were real-world when really I was measuing the performanc