diff options
author | Stefan Liebler <stli@linux.vnet.ibm.com> | 2018-02-16 09:57:30 +0100 |
---|---|---|
committer | Stefan Liebler <stli@linux.vnet.ibm.com> | 2018-02-16 09:57:30 +0100 |
commit | 8724507385a591a71297c2da0e12001eae6bd88d (patch) | |
tree | d272a8418d373e7b99a58296ae37c0b672567357 /nptl/Makefile | |
parent | db9881ecd7e7278af3e6bb252a0b3015e275d7bd (diff) | |
download | glibc-8724507385a591a71297c2da0e12001eae6bd88d.tar.gz glibc-8724507385a591a71297c2da0e12001eae6bd88d.tar.xz glibc-8724507385a591a71297c2da0e12001eae6bd88d.zip |
Add runtime check if mutex will be elided in tst-mutex8 testcases.
An elided mutex don't fail destroy. Elision was disabled for the test nptl/tst-mutex8 in nptl/Makefile. Thus we can run tests which destroy a locked mutex. As elision is only disabled for tst-mutex8, the variants tst-mutex8-static, tst-mutexpi8 and tst-mutexpi8-static are still failing if lock elision is enabled. This patch adds a runtime check, if the checked type of mutex will be elided. This check is using TUNABLE_GET_FULL to determine if elision is enabled via the tunables framework. The pthread_mutex_destroy tests are only run if we dont't assume an elided mutex. This way, we can run the whole glibc testsuite with or without enabled lock elision. ChangeLog: * nptl/Makefile (tst-mutex8-ENV): Delete. * nptl/tst-mutex8.c (check_type): Add runtime check if mutex will be elided.
Diffstat (limited to 'nptl/Makefile')
-rw-r--r-- | nptl/Makefile | 4 |
1 files changed, 0 insertions, 4 deletions
diff --git a/nptl/Makefile b/nptl/Makefile index 6fc2c8bb6a..9340f9f699 100644 --- a/nptl/Makefile +++ b/nptl/Makefile @@ -726,10 +726,6 @@ endif $(objpfx)tst-compat-forwarder: $(objpfx)tst-compat-forwarder-mod.so -# Disable elision for tst-mutex8 so it can verify error case for -# destroying a mutex. -tst-mutex8-ENV = GLIBC_TUNABLES=glibc.elision.enable=0 - # The tests here better do not run in parallel ifneq ($(filter %tests,$(MAKECMDGOALS)),) .NOTPARALLEL: |