jemalloc 3 performance vs. mozjemalloc

Matthew Hall mhall at
Wed Feb 4 14:27:09 PST 2015

On Wed, Feb 04, 2015 at 06:51:53PM +0900, Mike Hommey wrote:
> The problem here is that very low-level things are involved, and taking
> a test-case out of context changes the very conditions of those low-level
> things. As I wrote in the start of this thread, I did see things out of
> context that may or may not matter, and it turns out issue #192 is the
> one thing that made the most difference because of its impact on page
> faults. All the other things I did, like tweaking the config (enabling
> tcache, adjusting lg_dirty_mult, ...), returning to 3.6 instead of dev,
> etc. didn't do much.

It makes me wonder if a simpler setup of ARM Linux running the test cases on 
an embedded board like an RPIV2 w/ quad-core, BeagleBoard, etc. would be able 
to repro the issue or not. Then you wouldn't have to worry about all the 
Android hacks getting in the way of the test.


More information about the jemalloc-discuss mailing list