[lkml]   [2011]   [Nov]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH V2 2/2] module: Fix performance regression on modules with large symbol tables
>>> On 21.11.11 at 06:06, Rusty Russell <> wrote:
> On Thu, 17 Nov 2011 19:15:02 -0800, Kevin Cernekee <> wrote:
>> Potential memory savings from reusing EXACT strtab matches: 36,475 bytes
>> total; 35,432 from nvidia.ko. For comparison, discarding non-core
>> strtab entries saved about 91KB on my system.
> ...
>> Original scheme: 1.230s
>> With patch V2: 0.280s
>> With naive copying: 0.058s
> I'm deeply tempted. It's very simple, 46 lines shorter, preserves exact
> matches, and doesn't have any strange slowdowns on corner cases.
> But like Kevin, I could be convinced either way. Jan?

I'd be fine with the simpler solution, as long it is clearly understood
(and marked as intended to be that was for the time being) that it
has room for improvement, should it ever turn out to yield more
significant savings in other contexts.


 \ /
  Last update: 2011-11-21 09:11    [W:0.084 / U:4.340 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site