mirror of
https://github.com/jart/cosmopolitan.git
synced 2025-05-28 00:02:28 +00:00
Make pthread mutexes more scalable
pthread_mutex_lock() now uses a better algorithm which goes much faster in multithreaded environments that have lock contention. This comes at the cost of adding some fixed-cost overhead to mutex invocations. That doesn't matter for Cosmopolitan because our core libraries all encode locking operations as NOP instructions when in single-threaded mode. Overhead only applies starting the moment you first call clone().
This commit is contained in:
parent
7de2f229a7
commit
7ff0ea8c13
32 changed files with 410 additions and 112 deletions
3
third_party/dlmalloc/dlmalloc.c
vendored
3
third_party/dlmalloc/dlmalloc.c
vendored
|
@ -25,8 +25,7 @@
|
|||
#define HAVE_MMAP 1
|
||||
#define HAVE_MREMAP 0
|
||||
#define HAVE_MORECORE 0
|
||||
#define USE_LOCKS 1
|
||||
#define USE_SPIN_LOCKS 1
|
||||
#define USE_LOCKS 2
|
||||
#define MORECORE_CONTIGUOUS 0
|
||||
#define MALLOC_INSPECT_ALL 1
|
||||
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue