https://sourceware.org/bugzilla/show_bug.cgi?id=34062
--- Comment #7 from Takao Sato <takaosato1997 at gmail dot com> --- (In reply to Eric Botcazou from comment #6) > > Regardless of how the project chooses to classify this — whether as a > > security defect or a stability issue — we believe the underlying enforcement > > gap that Alan described in comment #2 is worth addressing, as it affects > > users who may not be aware that IA-64 backend code is active in their > > builds. > > But almost any change made to a complex software has the potential to > introduce issues that were not present before it; given that the IA-64 > back-end gets essentially no testing in the wild nowadays because the > architecture is dead, fixing this problem that no human being has noticed in > two decades could end up breaking the back-end without anyone noticing. > Obsolete software should be left dying in peace instead of being fed to > fuzzers without proper consideration. Thank you for the explanation, Eric. I understand and respect the project's position — the risk of introducing regressions in untested legacy code is a legitimate concern, and we appreciate the honest assessment. Thank you as well to Alan for the technical insight in comment #2. I hope to continue contributing to binutils security and look forward to collaborating on future findings where the impact may be more clearly actionable. We will do our best to ensure that future reports are well-scoped and focused on issues the project finds worth addressing. Best regards, Takao Sato -- You are receiving this mail because: You are on the CC list for the bug.
