summaryrefslogtreecommitdiffstats
path: root/Modules/_operator.c
diff options
context:
space:
mode:
authorMiss Islington (bot) <31488909+miss-islington@users.noreply.github.com>2020-11-21 09:12:21 (GMT)
committerGitHub <noreply@github.com>2020-11-21 09:12:21 (GMT)
commit97136d71a78a4b6b816f7e14acc52be426efcb6f (patch)
tree15f723a869c7c40d2f01e97e755b59f7ccc3acc6 /Modules/_operator.c
parent0762e09eb14269b38f60e1b58d2c7f36056a4694 (diff)
downloadcpython-97136d71a78a4b6b816f7e14acc52be426efcb6f.zip
cpython-97136d71a78a4b6b816f7e14acc52be426efcb6f.tar.gz
cpython-97136d71a78a4b6b816f7e14acc52be426efcb6f.tar.bz2
bpo-40791: Make compare_digest more constant-time. (GH-20444)
* bpo-40791: Make compare_digest more constant-time. 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>
Diffstat (limited to 'Modules/_operator.c')
-rw-r--r--Modules/_operator.c2
1 files changed, 1 insertions, 1 deletions
diff --git a/Modules/_operator.c b/Modules/_operator.c
index 5aa229f..a816548 100644
--- a/Modules/_operator.c
+++ b/Modules/_operator.c
@@ -735,7 +735,7 @@ _tscmp(const unsigned char *a, const unsigned char *b,
volatile const unsigned char *left;
volatile const unsigned char *right;
Py_ssize_t i;
- unsigned char result;
+ volatile unsigned char result;
/* loop count depends on length of b */
length = len_b;