- attached patch-22
[OpenSSL-1.0.2] TNF local patch - OPENSSL_rdtsc() uses non-atomic ops under 32bit PowerPC
original commit message:
http://mail-index.netbsd.org/source-changes/2012/07/27/msg036014.html
Comments (3)
-
reporter -
reporter - changed title to [OpenSSL-1.0.2] TNF local patch - OPENSSL_rdtsc() uses non-atomic ops under 32bit PowerPC
-
reporter - changed status to resolved
BUGFIX: Issue
- OPENSSL_rdtsc() uses non-atomic ops under 32bit PowerPC.#163dist/crypto/ppccpuid.pl: don't read TBU, OPENSSL_rdtsc's return type is uint32_t and TBU is useless and discarded. lib/libcrypto/arch/powerpc{,64}/ppccpuid.S: regen and drop N local patch.
TODO: 1.0.2 doesn't support some evbppc's CPU(eg. e500v2) lacks mftb instruction. to support these cpu, rewrite some code using mfspr instruction and time base register=268.
this change taken from openssl's master branch(modified by me): https://github.com/openssl/openssl/commit/81eae077ce679c1d7d29e19991bf055e4888a2fc https://github.com/openssl/openssl/commit/c8f370485c43729db44b680e41e875ddd7f3108c
original commit message:
crpyto/ppccpuid.pl: add FPU probe and fix OPENSSL_rdtsc.
Reviewed-by: Kurt Roeckx kurt@openssl.org
PPC: Try out if mftb works before using it
If this fails try out if mfspr268 works.
Use OPENSSL_ppccap=0x20 for enabling mftb, OPENSSL_ppccap=0x40 for enabling mfspr268, and OPENSSL_ppccap=0 for enabling neither.
Fixes #8012
Reviewed-by: Paul Dale paul.dale@oracle.com (Merged from #8043)
→ <<cset 852e73837e5237086debb5e865b026b99a764ebd>>
- Log in to comment