https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113596
--- Comment #4 from John Sanpe <sanpeqf at gmail dot com> --- (In reply to Andrew Pinski from comment #1) > Why do you think this is a bug? > > You are forcing a function which calls alloca to be inlined, GCC DOES not > normally inline functions which call alloca due to not restoring the stack > pointer after the return of the inlined function. I think most people's logic should be that exiting the function will release the memory of alloca, and clang does the same.