src-test2/lib/libpthread, branch release/5.2.0_cvs FreeBSD source tree https://cgit-dev.freebsd.org/src-test2/atom?h=release%2F5.2.0_cvs 2004-01-10T05:53:29Z This commit was manufactured by cvs2svn to create tag 2004-01-10T05:53:29Z cvs2svn cvs2svn@FreeBSD.org 2004-01-10T05:53:29Z urn:sha1:58c9ff44c8041f394c663cb9640e9b77fe07bb13 'RELENG_5_2_0_RELEASE'. This commit was manufactured to restore the state of the 5.2-RELEASE image. Releases prior to 5.3-RELEASE are omitting the secure/ and crypto/ subdirs. Apply a second fix for stack alignment with libkse. This time, enter the 2003-12-05T01:41:43Z Peter Wemm peter@FreeBSD.org 2003-12-05T01:41:43Z urn:sha1:30a62d30f4abee3b3052b55ba8db30065ac5d5f6 UTS with the stack correctly aligned. Also, while here, use an indirect jump rather than the pushq/ret hack. This fixes threaded apps that use floating point for me, although it hasn't solved all the problems. It is an improvement though. Preservation of the 128 byte red zone hasn't been resolved yet. Approved by: re (scottl) Eliminate two pushl by using call instruction directly, this really 2003-11-29T14:25:43Z David Xu davidxu@FreeBSD.org 2003-11-29T14:25:43Z urn:sha1:508f442784dfe720871827063aeeb6c971a1c51f helps branch predict a lot for INTEL P4. Approved by: re (scottl) 1.Macro optimizing KSE_LOCK_ACQUIRE and THR_LOCK_ACQUIRE to use static fall 2003-11-29T14:22:29Z David Xu davidxu@FreeBSD.org 2003-11-29T14:22:29Z urn:sha1:170422c2ef6a210a6e29b278d1ac9be50cdacb7c through branch predict as suggested in INTEL IA32 optimization guide. 2.Allocate siginfo arrary separately to avoid pthread to be allocated at 2K boundary, which hits L1 address alias problem and causes context switch to be slow down. 3.Simplify context switch code by removing redundant code, code size is reduced, so it is expected to run faster. Reviewed by: deischen Approved by: re (scottl) Remove surplus mmap() call for stack guard page in init_private, it is done 2003-11-29T14:10:02Z David Xu davidxu@FreeBSD.org 2003-11-29T14:10:02Z urn:sha1:5a8fe60d7e677d822305195ee569a3edd0ce3b3d in init_main_thread. Also don't initialize lock and lockuser again for initial thread, it is already done by _thr_alloc(). Reviewed by: deischen Approved by: re (scottl) Back out last change and go back to using KSE locks instead of thread 2003-11-16T15:01:26Z Daniel Eischen deischen@FreeBSD.org 2003-11-16T15:01:26Z urn:sha1:5303e946074a81543103f8716f4eca7168eebb3a locks until we know why this breaks ia64. Reported by: marcel If a thread in critical region got a synchronous signal, according current 2003-11-09T00:37:14Z David Xu davidxu@FreeBSD.org 2003-11-09T00:37:14Z urn:sha1:0e17930dd7bfaf070b3d047c80c4f57e8de86404 signal handling mode, there is no chance to handle the signal, something must be wrong in the library, just call kse_thr_interrupt to dump its core. I have the code for a long time, but forgot to commit it. Use THR lock instead of KSE lock to avoid scheduler be blocked in spinlock. 2003-11-08T06:07:04Z David Xu davidxu@FreeBSD.org 2003-11-08T06:07:04Z urn:sha1:38a53c620691c8f75fc4d21713a7e27bd10b4ae2 Reviewed by: deischen style(9) 2003-11-05T18:19:24Z Daniel Eischen deischen@FreeBSD.org 2003-11-05T18:19:24Z urn:sha1:7a1192c1d37d834a4049eb6d9082cef62e2d47fb Reviewed by: bde Don't declare the malloc lock; use the declaration provided in libc. 2003-11-05T18:18:45Z Daniel Eischen deischen@FreeBSD.org 2003-11-05T18:18:45Z urn:sha1:94db4dd759dc22f1927f945a3611281d99ebbdf4 Noticed by: bde