3 comments
Not actually confirmed as compiler bugs yet.<p>Here’s a similar bug where OpenSSL’s inline asm was wrong and probably only GCC was smart enough to try to exploit the bug:<p><a href="https://github.com/openssl/openssl/pull/23233" rel="nofollow">https://github.com/openssl/openssl/pull/23233</a><p>Same for rust - could be UB in quiche that is getting exposed.
Can't even find an upstream compiler bug <i>report</i> not to mention a confirmation/patch. Doesn't look like the compiler output was analyzed at all either, they are just going by "-O3 results in valgrind warnings, -O2 doesn't".<p>It's not like compiler bugs are unheard of but they are rare enough that the base assumption should be that the bug is in your own code.
The social post links to this bug <a href="https://github.com/curl/curl/issues/23237" rel="nofollow">https://github.com/curl/curl/issues/23237</a>
Looks like a valid new optimization to me, which only fails on valgrind, which is a bit too strict with uninitialized values. In openssl's strlcpy. They'll deal with that.