> I would think the generic -fstack-protector testcases would already pick > this up. But apparently there doesn't exist any testcase for it! So > this failure mode wasn't detected either. Oh well. > > Not something I'll ask you to improve then!
Okay. So if everything is good, can the changes be merged ? Thanks Ayappan On Tue, Sep 2, 2025 at 3:17 PM Segher Boessenkool <seg...@kernel.crashing.org> wrote: > > On Tue, Sep 02, 2025 at 03:01:27PM +0530, Ayappan wrote: > > > Wow. So this shortcoming harks back to the origin of time. > > > > > > Why was it not discovered before, do you think? No one ever tried it? > > > > > > Should we have a testcase for it? _Do_ we have a testcase for it? > > > > I came to know about this recently when working with a build failure > > of Cpython in AIX. Recently in Cpython, the -fstack-protector flag is > > added and a compile > > check (-c) is added in configure to check whether it is accepted or > > not. In AIX , it > > passes during the compile check, and hence the flag is carried forward > > further but during the > > binary/library creation, it fails. If it would have failed during the > > compile check (-c) only, then the > > configure would have dropped the flag there only. > > Yeah, so arguably cpython is buggy here, too (it tests for something else > than what it then uses: it tests if the commandline flag is reecognised > at all, not if you can actually use it!) > > > # gcc -fstack-protector -c sample.c --> passes > > # gcc -fstack-protector sample.c > > collect2: fatal error: library libssp_nonshared not found > > compilation terminated. > > > > I am not sure whether we really need a testcase for this. > > I would think the generic -fstack-protector testcases would already pick > this up. But apparently there doesn't exist any testcase for it! So > this failure mode wasn't detected either. Oh well. > > Not something I'll ask you to improve then! > > > Segher
