On Wed, 2024-07-03 at 11:45 +0200, Johannes Berg wrote: > On Fri, 2024-05-24 at 23:37 +0200, benja...@sipsolutions.net wrote: > > From: Benjamin Berg <benjamin.b...@intel.com> > > > > There should be no need for this. > > "should" ;-)
Hmm, I would have expected the previous patch is what breaks it. i.e. this line that I added: + /* Ensure the new MM is clean and nothing unwanted is mapped */ + unmap(new_id, 0, STUB_START); > This breaks things if glibc enables rseq. That might even be already > broken in the sense that it might corrupt memory that's put at the same > place the rseq was installed? But it at least it appears to still work > without this patch... Then it would make sense to me as the previous force_flush_all would have only unmapped up to TASK_SIZE. While this now uses STUB_START which clears more memory as TASK_SIZE has some alignment. > As we discussed, it might thus be better to fix up the execve() patch > and do that before this. Yeah, lets just solve this class of issues with execve() first. Benjamin
