Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | process-shared barrier support, based on discussion with bdonlan | Rich Felker | 2011-09-27 | 1 | -2/+11 |
| | | | | | | | | | | | | | this implementation is rather heavy-weight, but it's the first solution i've found that's actually correct. all waiters actually wait twice at the barrier so that they can synchronize exit, and they hold a "vm lock" that prevents changes to virtual memory mappings (and blocks pthread_barrier_destroy) until all waiters are finished inspecting the barrier. thus, it is safe for any thread to destroy and/or unmap the barrier's memory as soon as pthread_barrier_wait returns, without further synchronization. | ||||
* | consistency: change all remaining syscalls to use SYS_ rather than __NR_ prefix | Rich Felker | 2011-04-06 | 1 | -1/+1 |
| | |||||
* | global cleanup to use the new syscall interface | Rich Felker | 2011-03-20 | 1 | -2/+2 |
| | |||||
* | cleaning up syscalls in preparation for x86_64 port | Rich Felker | 2011-02-13 | 1 | -0/+4 |
| | | | | | | | | | - hide all the legacy xxxxxx32 name cruft in syscall.h so the actual source files can be clean and uniform across all archs. - cleanup llseek/lseek and mmap2/mmap handling for 32/64 bit systems - alternate implementation for nice if the target lacks nice syscall | ||||
* | initial check-in, version 0.5.0 v0.5.0 | Rich Felker | 2011-02-12 | 1 | -0/+18 |