diff options
| author | David Goldblatt <davidgoldblatt@fb.com> | 2017-10-03 18:03:02 -0700 |
|---|---|---|
| committer | David Goldblatt <davidtgoldblatt@gmail.com> | 2017-10-04 18:37:23 -0700 |
| commit | 1245faae9052350a96dbcb22de7979bca566dbec (patch) | |
| tree | 19043580b26ec4953bab4435e44206d71e7db697 /configure.ac | |
| parent | 7c6c99b8295829580c506067495a23c07436e266 (diff) | |
Power: disable the CPU_SPINWAIT macro.
Quoting from https://github.com/jemalloc/jemalloc/issues/761 :
[...] reading the Power ISA documentation[1], the assembly in [the CPU_SPINWAIT
macro] isn't correct anyway (as @marxin points out): the setting of the
program-priority register is "sticky", and we never undo the lowering.
We could do something similar, but given that we don't have testing here in the
first place, I'm inclined to simply not try. I'll put something up reverting the
problematic commit tomorrow.
[1] Book II, chapter 3 of the 2.07B or 3.0B ISA documents.
Diffstat (limited to 'configure.ac')
| -rw-r--r-- | configure.ac | 7 |
1 files changed, 3 insertions, 4 deletions
diff --git a/configure.ac b/configure.ac index ab2f41ae..f957377b 100644 --- a/configure.ac +++ b/configure.ac @@ -381,6 +381,7 @@ dnl CPU-specific settings. CPU_SPINWAIT="" case "${host_cpu}" in i686|x86_64) + HAVE_CPU_SPINWAIT=1 if test "x${je_cv_msvc}" = "xyes" ; then AC_CACHE_VAL([je_cv_pause_msvc], [JE_COMPILABLE([pause instruction MSVC], [], @@ -399,13 +400,11 @@ case "${host_cpu}" in fi fi ;; - powerpc*) - AC_DEFINE_UNQUOTED([HAVE_ALTIVEC], [ ]) - CPU_SPINWAIT='__asm__ volatile("or 31,31,31")' - ;; *) + HAVE_CPU_SPINWAIT=0 ;; esac +AC_DEFINE_UNQUOTED([HAVE_CPU_SPINWAIT], [$HAVE_CPU_SPINWAIT]) AC_DEFINE_UNQUOTED([CPU_SPINWAIT], [$CPU_SPINWAIT]) case "${host_cpu}" in |