diff options
| author | Leah Rowe <leah@libreboot.org> | 2026-03-06 23:50:06 +0000 |
|---|---|---|
| committer | Leah Rowe <leah@libreboot.org> | 2026-03-06 23:50:06 +0000 |
| commit | 79c972367e93c42d9a2b14e32bd6e82f5c64de9f (patch) | |
| tree | ed308747a64077b518a41bb80c3ee2fe0e3f75f6 | |
| parent | 6e47315192cf9bd05eb0d228a2b4b145513d933d (diff) | |
util/nvmutil: reset errno on /dev/random fallback
i didn't catch this before. stale error state would
have passed through, even on ultimate success.
Signed-off-by: Leah Rowe <leah@libreboot.org>
| -rw-r--r-- | util/nvmutil/nvmutil.c | 13 |
1 files changed, 13 insertions, 0 deletions
diff --git a/util/nvmutil/nvmutil.c b/util/nvmutil/nvmutil.c index e4fe6c72..1a6f4687 100644 --- a/util/nvmutil/nvmutil.c +++ b/util/nvmutil/nvmutil.c @@ -358,7 +358,20 @@ open_dev_urandom(void) /* * Fall back to /dev/random on old platforms * where /dev/urandom does not exist. + * + * In general, we can assume that any system + * that would have urandom, probably has the + * other device (/dev/urandom). + * + * Given that on many modern systems, urandom + * and random are essentially the same, this + * also provides some fault-tolerance there, + * while still ultimately being strict under + * fallback conditions, thus: */ + + errno = 0; + rname = oldrandom; xopen(&rfd, rname, O_RDONLY, &st_rfd); } |
