[PING^2][PATCH] Use single shared memory block pool for all pool allocators

Ping. On 08/03/2015 11:40 AM, Mikhail Maltsev wrote: > On Jul 26, 2015, at 11:50 AM, Andi Kleen wrote: >> I've been compiling gcc with tcmalloc to do a similar speedup. It would be >> interesting to compare that to your patch. > I repeated the test with TCMalloc and

Re: [PATCH] Use single shared memory block pool for all pool allocators

On Jul 26, 2015, at 11:50 AM, Andi Kleen a...@firstfloor.org wrote: I've been compiling gcc with tcmalloc to do a similar speedup. It would be interesting to compare that to your patch. I repeated the test with TCMalloc and jemalloc. TCMalloc shows nice results, though it required some tweaks:

Re: [PATCH] Use single shared memory block pool for all pool allocators

On Sun, Jul 26, 2015 at 9:09 PM, pins...@gmail.com wrote: On Jul 26, 2015, at 11:50 AM, Andi Kleen a...@firstfloor.org wrote: Mikhail Maltsev malts...@gmail.com writes: Hi, all! Recently I did some profiling of GCC to find hotspots and areas of possible performance improvement among

Re: [PATCH] Use single shared memory block pool for all pool allocators

Mikhail Maltsev malts...@gmail.com writes: Hi, all! Recently I did some profiling of GCC to find hotspots and areas of possible performance improvement among them. glibc malloc(3) is one of (perhaps known) I've been compiling gcc with tcmalloc to do a similar speedup. It would be interesting

[PATCH] Use single shared memory block pool for all pool allocators

Hi, all! Recently I did some profiling of GCC to find hotspots and areas of possible performance improvement among them. glibc malloc(3) is one of (perhaps known) hotspots. It seemed rather strange to me that pool allocators call malloc(3) and free(3) rather often, and spend considerable time in

Re: [PATCH] Use single shared memory block pool for all pool allocators

On Jul 26, 2015, at 11:50 AM, Andi Kleen a...@firstfloor.org wrote: Mikhail Maltsev malts...@gmail.com writes: Hi, all! Recently I did some profiling of GCC to find hotspots and areas of possible performance improvement among them. glibc malloc(3) is one of (perhaps known) I've