[RFC] Closing of all remaining Bugzilla PRs against powerpcspe
Hi, over the course of two years that had passed since the deprecation of the powerpcspe backend, and a year and a half since its removal from gcc, I've still been speaking out several times against immediate closing of Bugzilla PRs against that target[1,2]. IIRC, Andrew has been contemplating a revival of the backend during gcc 9 development cycle, and I was willing to wait for another full cycle. But it is clearly obvious that the powerpcspe backend won't be revived. After two years of development, rs6000 has diverged too much from the split point, including LRA adoption. There was not much interest in SPE support from the community even during its last years in gcc, let alone after its removal. Tellingly, my employer never had spare engineers to allocate to the subject; instead, we decided to migrate to different, modern and supported, hardware for our future products. Andrew is still listed as powerpcspe maintainer in MAINTAINERS even on the current master; and if I'm not mistaken, John Paul Adrian has been recently involved in successful bounty program that resulted in migration of m68k backend away from cc0, not to mention his unfailing interest in powerpcspe. Do you guys have something to say on the matter? Otherwise, I'm proposing to finally close all open PRs filed against powerpcspe. I've been able to identify the following ones: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=19490 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=30259 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=37759 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=37760 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=47856 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=47977 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=49854 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=51905 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57389 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57872 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71012 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86133 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79438 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79451 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80700 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81288 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81628 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82138 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84302 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85121 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85170 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87083 PRs from the second group were filed by me, so if there's consensus to close all of them, the ones from this second group I can close myself. I don't have the right permissions to modify PRs reported by someone else, so I'd like to ask a volunteer to step up and close the ones from the first group. WDYT? [1] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=37759#c9 [2] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=49854#c3 Arseny
Re: [RFC] Closing of all remaining Bugzilla PRs against powerpcspe
Hi! On Sat, May 09, 2020 at 02:58:11AM +0700, Arseny Solokha wrote: > over the course of two years that had passed since the deprecation of the > powerpcspe backend, and a year and a half since its removal from gcc, I've > still > been speaking out several times against immediate closing of Bugzilla PRs > against that target[1,2]. IIRC, Andrew has been contemplating a revival of the > backend during gcc 9 development cycle, and I was willing to wait for another > full cycle. > > But it is clearly obvious that the powerpcspe backend won't be revived. After > two years of development, rs6000 has diverged too much from the split point, > including LRA adoption. LRA has been supported by the rs6000 port since 2013 (01b1efaa1439), and made the default (and only!) option in 2017 (7a5cbf29beb2). Ah, the latter was slightly after the split, I see. Did powerpcspe never work with LRA? > There was not much interest in SPE support from the > community even during its last years in gcc, let alone after its > removal. Tellingly, my employer never had spare engineers to allocate to the > subject; instead, we decided to migrate to different, modern and supported, > hardware for our future products. > > Andrew is still listed as powerpcspe maintainer in MAINTAINERS even on the > current master; and if I'm not mistaken, John Paul Adrian has been recently > involved in successful bounty program that resulted in migration of m68k > backend > away from cc0, not to mention his unfailing interest in powerpcspe. Do you > guys > have something to say on the matter? > > Otherwise, I'm proposing to finally close all open PRs filed against > powerpcspe. I have had this on my list of things to do for over a year now, it just never was high enough priority. Thank you for volunteering! > I've been able to identify the following ones: Wow, thank you, that was most of the work already :-) > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=19490 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=30259 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=37759 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=37760 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=47856 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=47977 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=49854 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=51905 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57389 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57872 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71012 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86133 > > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79438 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79451 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80700 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81288 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81628 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82138 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84302 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85121 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85170 > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87083 > > PRs from the second group were filed by me, so if there's consensus to close > all > of them, the ones from this second group I can close myself. I don't have the > right permissions to modify PRs reported by someone else, so I'd like to ask a > volunteer to step up and close the ones from the first group. > > WDYT? I can do both, if you want, or just the first group? Your choice. But let's hear other opinions first. Thanks, Segher
gcc-9-20200508 is now available
Snapshot gcc-9-20200508 is now available on https://gcc.gnu.org/pub/gcc/snapshots/9-20200508/ and on various mirrors, see http://gcc.gnu.org/mirrors.html for details. This snapshot has been generated from the GCC 9 git branch with the following options: git://gcc.gnu.org/git/gcc.git branch releases/gcc-9 revision 9a3dde5fed64b412884d17b013cb105d4f5e452c You'll find: gcc-9-20200508.tar.xzComplete GCC SHA256=7e1d57e1aabe3453dd133d0e232f9e187a124bfec723c72b5d6e91bda442377d SHA1=bee8f708a56b78565c4f5045af6bcc37c34f9c4f Diffs from 9-20200501 are available in the diffs/ subdirectory. When a particular snapshot is ready for public consumption the LATEST-9 link is updated and a message is sent to the gcc list. Please do not use a snapshot before it has been announced that way.
Re: [RFC] Closing of all remaining Bugzilla PRs against powerpcspe
> Hi! >> But it is clearly obvious that the powerpcspe backend won't be revived. After >> two years of development, rs6000 has diverged too much from the split point, >> including LRA adoption. > > LRA has been supported by the rs6000 port since 2013 (01b1efaa1439), > and made the default (and only!) option in 2017 (7a5cbf29beb2). Ah, the > latter was slightly after the split, I see. Did powerpcspe never work > with LRA? It worked, but of course had its own issues: take PR79438 as one example (but perhaps the problem there had not stemmed from RA at all), and I would undoubtedly find more if I had enough time and knew they would be fixed. I never enabled LRA for our production code, though. So what I've been originally trying to say was that, while switching to LRA by default would expose more problems and maybe wrong-code issues, removing reload as a fallback would put extra pressure on finding and fixing them. If the switch happened when SPE was just a part of rs6000 backend, some of the necessary work would be likely offset by people (and corporations!) behind the PowerPC (sorry, Segher), but reviving powerpcspe now means that all of that has to be done by relatively small group of maintainers (or even by a single person) of a quite niche target. There was also some serious work during migration to LRA by default that resulted in filing and then presumably fixing PR69847 (and others like that one), and, I believe, nothing like that were possible for powerpcspe. >> I've been able to identify the following ones: > > Wow, thank you, that was most of the work already :-) Strangely, I failed to find any PR for e200, so maybe some unnoticed ones are still lying around. PR19490 probably needs some consideration. Is there something left in rs6000 backend that could be done prior to closing this one? rs6000 backend still has some remnants of support of 854[08], namely, instruction costs and some machine descriptions; is it on purpose? The idea is that Power ISA on these cores is still supported, right? >> PRs from the second group were filed by me, so if there's consensus to close >> all >> of them, the ones from this second group I can close myself. I don't have the >> right permissions to modify PRs reported by someone else, so I'd like to ask >> a >> volunteer to step up and close the ones from the first group. >> >> WDYT? > > I can do both, if you want, or just the first group? Your choice. > > But let's hear other opinions first. Thanks. I think I'll close the second group myself, as they constitute almost half of the total amount. Arseny > Thanks, > > > Segher
Re: [RFC] Closing of all remaining Bugzilla PRs against powerpcspe
Hi! On 5/9/20 12:15 AM, Segher Boessenkool wrote: > I can do both, if you want, or just the first group? Your choice. > > But let's hear other opinions first. These bugs document the current issues with the backend as it existed in gcc-8 (or was it -9)? The bugs are still in the removed code, so I don't really understand what you gain by closing bugs? Is it important to keep the number of open issues low? I don't consider bug reports a bad thing, they document the code quality and are a useful resource to anyone working on the code or using these versions. Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer - glaub...@debian.org `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
