<html> <head> <base href="https://bugs.freedesktop.org/"> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - piglit.spec.glsl-1_50.compiler.vs-redeclares-pervertex-out-before-global-redeclaration" href="https://bugs.freedesktop.org/show_bug.cgi?id=102685#c1">Comment # 1</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - piglit.spec.glsl-1_50.compiler.vs-redeclares-pervertex-out-before-global-redeclaration" href="https://bugs.freedesktop.org/show_bug.cgi?id=102685">bug 102685</a> from <a class="email" href="mailto:itoral@igalia.com" title="Iago Toral <itoral@igalia.com>"> Iago Toral</a> <pre>Hi Mark, sorry about this, I am not sure why I didn't see this before, I was sure that I had sent this patch to CI before sending it for review :-/ Anyway, the problem is not with piglit, there is some rather unsafe code right prior to the place where this commit operated that can invalidate the memory of 'var' in one specific scenario (triggered by these parser tests) that makes any access to that variable from that point forward potentially unsafe: the variable ends up pointing to freed memory so you get undefined behavior, so the crash is not even guaranteed to happen. I think this is quite unfortunate, so I'll be sending a couple of patches, one to fix this regression and one or two more to make that implementation safer so we don't see more instances of this type of problems in the future.</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the QA Contact for the bug.</li> </ul> </body> </html>