| 8120f5e2 | 04-Jan-2013 |
Matthew Dillon <dillon@apollo.backplane.com> |
libc - Add poor man's cache coloring optimization to nmalloc module.
* A series of large allocations in excess of 32KB will be offset by 4K from
each other. This fixes performance issues on Sandy
libc - Add poor man's cache coloring optimization to nmalloc module.
* A series of large allocations in excess of 32KB will be offset by 4K from
each other. This fixes performance issues on SandyBridge and later cpus
related to large matrix operations.
This eats an extra 4K of VM for such allocations but does not eat any
additional real memory.
* Greatly improves large FP matrix benchmarks. Real-world effects are more
questionable.
* The Sandybridge and later cpus use a virtually indexed, physically tagged
L1 cache, and tend to be sensitive to substantially different memory
addresses winding up on the same cache line. Matrix operations (primarily
benchmarks) can cause these sorts of effects.
Reported-by: alexh
show more ...
|
| 20b2da2c | 25-May-2012 |
Venkatesh Srinivas <me@endeavour.zapto.org> |
libc -- dmalloc: Call malloc_init as-needed, rather than via ctor (#2)
This commit is a second revision of
e12d3396c777165504d60d2a1408dcd7cb63660d; for details, see the original
commit message.
Th
libc -- dmalloc: Call malloc_init as-needed, rather than via ctor (#2)
This commit is a second revision of
e12d3396c777165504d60d2a1408dcd7cb63660d; for details, see the original
commit message.
That commit was reverted quickly, as it broke pthreads; this revision
does not suffer from that problem, as it preserves the __constructor
logic for malloc_init.
Reverts: 4018c6eddd57f4abf9134690cbfa46c9d7103558 (Revert libc ...)
Reported-by: marino@
Closes-bug: 2305
show more ...
|