On 12.07.2018 07:14, Alexey Kardashevskiy wrote: > On Thu, 12 Jul 2018 07:01:34 +0200 > Thomas Huth wrote: > >> On 12.07.2018 03:01, David Gibson wrote: >>> On Wed, Jul 11, 2018 at 08:53:05PM +1000, Alexey Kardashevskiy wrote: >>>> On Wed, 11 Jul 2018 16:26:19 +1000 >>>> David Gibson wrote: >>>> >>>>> On Tue, Jul 10, 2018 at 05:14:53PM +1000, Alexey Kardashevskiy wrote: >>>>>> On Tue, 10 Jul 2018 16:42:48 +1000 >>>>>> David Gibson wrote: >>>>>> >>>>>>> On Tue, Jul 10, 2018 at 12:22:44PM +1000, Alexey Kardashevskiy wrote: >>>>>>>> On Mon, 2 Jul 2018 16:26:44 +1000 >>>>>>>> Alexey Kardashevskiy wrote: >>>>>>>> >>>>>>>>> The following changes since commit edb1fb337f65f82fe32b989c4f018efe85c1dddb: >>>>>>>>> >>>>>>>>> pseries: Update SLOF firmware image (2018-07-02 16:21:30 +1000) >>>>>>>>> >>>>>>>>> are available in the git repository at: >>>>>>>>> >>>>>>>>> git@github.com:aik/qemu.git tags/qemu-slof-20180702 >>>>>>>>> >>>>>>>>> for you to fetch changes up to edb1fb337f65f82fe32b989c4f018efe85c1dddb: >>>>>>>>> >>>>>>>>> pseries: Update SLOF firmware image (2018-07-02 16:21:30 +1000) >>>>>>>>> >>>>>>>>> ---------------------------------------------------------------- >>>>>>>>> >>>>>>>>> I just missed a couple of gcc 8.1 fixes and thought I can still squeeze >>>>>>>>> them before the 3.0 soft freeze. >>>>>>>>> >>>>>>>>> >>>>>>>>> *** Note: this is not for master, this is for pseries >>>>>>>> >>>>>>>> >>>>>>>> Not good for 3.1 either? >>>>>>> >>>>>>> Ugh.. somehow I missed this SLOF update request, and now we're in 3.0 >>>>>>> hard freeze. >>>>>>> >>>>>>> How bad are the bugs this SLOF update fixes? >>>>>> >>>>>> These only fix gcc 8.1 warnings and runtime errors => not really >>>>>> bugfixes so it is not a big deal if they miss 3.0. >>>>> >>>>> Ok, well, if you want to resend the update, I'll queue it for 3.1. >>>> >>>> >>>> It is still on github, why resend? >>> >>> Good point, done. >> >> Won't there be another update in the 3.1 time frame? If yes, this sounds >> like wasted space in the repository to me (the firmware blobs are rather >> big, and there are no new features in that binary, just fixes for >> compiling with GCC 8 ... and people who want to compile SLOF on their >> own can use the sources from aik's git tree, too). > > I compiled the blob with gcc 8.1 and if something goes wrong I'd rather > know this earlier than later - this my motivation, not the fixes > themselves. Ok, that's a fair argument. But in that case, I think it is even good that this missed the hard freeze - you likely don't want to hunt hard-to-debug compiler bugs in the freeze period, do you? Thomas