Staging
v0.8.1
v0.8.1
https://github.com/python/cpython
Revision db95802bdfac4d13db3e2a391ec7b9e2f8d92dbe authored by Miss Islington (bot) on 22 November 2020, 17:33:09 UTC, committed by GitHub on 22 November 2020, 17:33:09 UTC
The existing volatile `left`/`right` pointers guarantee that the reads will all occur, but does not guarantee that they will be _used_. So a compiler can still short-circuit the loop, saving e.g. the overhead of doing the xors and especially the overhead of the data dependency between `result` and the reads. That would change performance depending on where the first unequal byte occurs. This change removes that optimization. (This is change GH-1 from https://bugs.python.org/issue40791 .) (cherry picked from commit 31729366e2bc09632e78f3896dbce0ae64914f28) Co-authored-by: Devin Jeanpierre <jeanpierreda@google.com>
1 parent 9ae1742
Tip revision: db95802bdfac4d13db3e2a391ec7b9e2f8d92dbe authored by Miss Islington (bot) on 22 November 2020, 17:33:09 UTC
bpo-40791: Make compare_digest more constant-time. (GH-23438)
bpo-40791: Make compare_digest more constant-time. (GH-23438)
Tip revision: db95802
File | Mode | Size |
---|---|---|
NEWS.d | ||
ACKS | -rw-r--r-- | 26.6 KB |
HISTORY | -rw-r--r-- | 1.3 MB |
Porting | -rw-r--r-- | 63 bytes |
README | -rw-r--r-- | 1.4 KB |
README.AIX | -rw-r--r-- | 4.7 KB |
README.coverity | -rw-r--r-- | 845 bytes |
README.valgrind | -rw-r--r-- | 4.5 KB |
SpecialBuilds.txt | -rw-r--r-- | 9.7 KB |
coverity_model.c | -rw-r--r-- | 4.1 KB |
gdbinit | -rw-r--r-- | 4.7 KB |
indent.pro | -rw-r--r-- | 557 bytes |
python-config.in | -rw-r--r-- | 1.9 KB |
python-config.sh.in | -rw-r--r-- | 2.7 KB |
python-wing3.wpr | -rw-r--r-- | 555 bytes |
python-wing4.wpr | -rw-r--r-- | 835 bytes |
python-wing5.wpr | -rw-r--r-- | 835 bytes |
python.man | -rw-r--r-- | 17.8 KB |
python.pc.in | -rw-r--r-- | 293 bytes |
svnmap.txt | -rw-r--r-- | 4.1 MB |
valgrind-python.supp | -rw-r--r-- | 8.6 KB |
vgrindefs | -rw-r--r-- | 500 bytes |
Computing file changes ...