On Wed, 2016-11-02 at 09:52 +0100, Svante Signell wrote: > On Tue, 2016-11-01 at 16:20 +0100, Samuel Thibault wrote: > > > > Svante Signell, on Tue 01 Nov 2016 16:19:19 +0100, wrote: > > > > > > > > > I have now rebuilt gcc-2.24-6 without hurd-i386/tg-libpthread-gsync- > > > spin.diff > > > and do not experience any filesystem corruptions when sync is issued for > > > the > > > patch target in gcc-6. > > Ok, so that confirms that it's the culprit. > > > > > > > > > > > I'll rebuild with your updated patch ASAP. BBL. > > Thanks! > > Samuel > With the revised patch the filesystem corruption is no longer, yay! > Back to porting and debugging gccgo for GNU/Hurd...
Sorry to bring bad news. The filesystem corruption is back, but not as bad as before :( And when disks are checked the date is again set to: Tue Apr 13 23:07:40 2032. Going back to the version without the tg-libpthread-gsync- spin.diff patch.
