This is the mail archive of the
mailing list for the GCC project.
Re: [Patch, Darwin] Fix PR46904/PR46916
- From: Mike Stump <mikestump at comcast dot net>
- To: IainS <developer at sandoe-acoustics dot co dot uk>
- Cc: GCC Patches <gcc-patches at gcc dot gnu dot org>, Mike Stump <mrs at gcc dot gnu dot org>, Jan Hubicka <hubicka at ucw dot cz>
- Date: Sat, 18 Dec 2010 10:58:01 -0800
- Subject: Re: [Patch, Darwin] Fix PR46904/PR46916
- References: <D67DFCA2-7487-4B35-898C-C861C6BFF9E8@sandoe-acoustics.co.uk>
On Dec 17, 2010, at 1:37 AM, IainS wrote:
> This has now been regstrapped on *-darwin9 and x86_64-darwin10.
> There are outstanding issues with -freorder-blocks-and-partition (on Darwin, at least) which results in the fail of partition2.C on Darwin10.
:-( I kinda hate regressions.... and the opt.c change I kinda wish were broken out, either, it goes in first, or it goes in second.
Normally, I'd say since I think the changes for the two remaining problems are well underway, I'd propose to stage in the fixes for those first, before the below goes in, so that when it does go in, there are no regressions, but, the problem with that is this patch is to `fix' regressions, and as I said, I hate regressions....
> OK for trunk?
Though, that said, I'd rather have the patch that originally broke, backed out to fix the immediate regression, then the sub problems found fixed first, as prep work for this patch, then this patch and the other fix that was reverted could go in together, all, without regressions. We are late enough in the cycle, that having regressions is, well, let me put it this way, I'd still like to see a bot that just reverts any patch that causes a regression, no exceptions. I still think it would be a nice property to have.