> -----Original Message----- > From: Tobias Burnus <[email protected]> > Sent: 18 March 2026 23:23 > To: Prathamesh Kulkarni <[email protected]>; Matthew Malcomson > <[email protected]>; [email protected] > Subject: Re: [PATCH 2/5] libgomp: Ensure memory sync after performing > tasks > > External email: Use caution opening links or attachments > > > Hi Prat, > > sorry that this simple patch fell through the cracks :-( > > Can you add "PR libgomp/124123" to the changelog as well → cf. > https://gcc.gnu.org/PR124123 > > On January 31, 2026 Prathamesh Kulkarni wrote: > > This happens because full_data is not initialized to 0 on second > call to test() on host, which results in call to abort. > > The attached patch modifies pr122356.c (and similarly pr122314.c) to > initialize full_data to 0 at beginning of test(). > > > > With patch, running make check-target-libgomp > RUNTESTFLAGS="c.exp=pr122356.c" shows: > > > > === libgomp Summary === > > # of expected passes 2 > > > > OK to commit ? > > OK. Thanks, I have committed the patch after adding PR libgomp/124123 to ChangeLog in 97682f93d7716dbd74e39846fff96cc5b3fb7107.
Thanks, Prathamesh > > Tobias > > PS: Due to overload of other things + being late in Stage 4 and still > hoping that Jakub looks at those patches as he knows that part better > than me, I put the review of patch 3+4 on the backburner, even though > I have partially reviewed it. – I obviously missed the 2/5 fix and I > should really pick up 3+4 soon, unless Jakub does it. :-/ > > Unfortunately, those patches aren't obvious and simple.
