This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Other format: | [Raw text] |
Hi! SH needs the analog of what has been changed in commit 2983d85ee5492c4ab1f09ca554e47192fa5c97a2, originally submitted in <http://sourceware.org/ml/libc-hacker/2010-07/msg00000.html>, but not fixed for SH at that time. No other architecture is affected, as (apart From x86 and x86_64) SH is the only one to use thusly optimized code governed by __ASSUME_FUTEX_CLOCK_REALTIME, and only uses it in this very place. (So, now every usage in glibc and glibc-ports should be correct.) nptl/ 2012-01-24 Thomas Schwinge <thomas@codesourcery.com> * sysdeps/unix/sysv/linux/sh/lowlevellock.S (__lll_timedlock_wait): Check for timestamp before the Epoch. Index: nptl/sysdeps/unix/sysv/linux/sh/lowlevellock.S =================================================================== --- nptl/sysdeps/unix/sysv/linux/sh/lowlevellock.S (revision 355696) +++ nptl/sysdeps/unix/sysv/linux/sh/lowlevellock.S (working copy) @@ -244,6 +244,12 @@ __lll_timedlock_wait: bt .Lreltmo # endif + /* if (timeout->tv_sec < 0) return ETIMEDOUT; */ + mov.l @r6, r1 + cmp/pz r1 + bf/s 5f + mov #ETIMEDOUT, r0 + mov r4, r2 mov r5, r4 mov r7, r5 @@ -279,6 +285,7 @@ __lll_timedlock_wait: neg r0, r3 3: mov r3, r0 +5: rts mov.l @r15+, r12 GrÃÃe, Thomas
Attachment:
pgp00000.pgp
Description: PGP signature
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |