> On 2 Jul 2026, at 09:12, Richard Biener <[email protected]> wrote: > > On Thu, 2 Jul 2026, Kyrylo Tkachov wrote: > >> >> >>> On 2 Jul 2026, at 09:00, Richard Biener <[email protected]> wrote: >>> >>> On Wed, 1 Jul 2026, Jeffrey Law wrote: >>> >>>> >>>> >>>> On 7/1/2026 9:43 AM, Kyrylo Tkachov wrote: >>>>>> What I would suggest would be to look at the code we generate for the >>>>>> split-paths tests before/after removal and make sure we're not missing >>>>>> something. The tests, more likely than not, test that split-paths >>>>>> triggered rather than looking at the final assembly codes. >>>>> So out of all split-path tests only one shows a difference with >>>>> -fsplit-paths vs -fno-split-paths: split-paths-6.c where -fno-split-paths >>>>> gives +3 aarch64 instructions, but this is a downstream artifact. >>>>> In the source: >>>>> void givehelp (int interactive) >>>>> { >>>>> if (interactive) >>>>> while ((--((_impure_ptr->_stdin))->_r < 0 >>>>> ? __srget_r (_impure_ptr, _impure_ptr->_stdin) >>>>> : (int) (*((_impure_ptr->_stdin))->_p++)) != ' ‘); >>>>> } >>>>> >>>>> -fsplit-paths triggers once but the duplication is later remerged but the >>>>> edge probabilities end up different and then loop rotation ends up making >>>>> different layout decisions. >>>>> So I don’t think it’s a very deliberate effect. >>>> Yea, that's sounds more like a random downstream effect than something >>>> intentional. I don't really recall split-path-6, though a reasonable >>>> assumption would be that I was playing around with the original patch and >>>> included some tests I found when experimenting with the patch's behavior. >>> >>> Can we preserve testcases as compile-only btw? Some might have been >>> introduced for ICEs, some for overall expected optimization (with >>> no path splitting). >>> >> >> You mean put them in the torture suite without the scans for expected >> dump/assembly? > > Yes, or just remove the scans for expected dump/assembly (if those are > related to path splitting).
Ok, I’ve sent an updated patch with the tests moved/updated. Kyrill > >> Thanks, >> Kyrill >> >>> Richard. >>> >>>> >>>> Jeff >>>> >>>> >>> >>> -- >>> Richard Biener <[email protected]> >>> SUSE Software Solutions Germany GmbH, >>> Frankenstrasse 146, 90461 Nuernberg, Germany; >>> GF: Jochen Jaser, Andrew McDonald; (HRB 36809, AG Nuernberg) >> >> >> > > -- > Richard Biener <[email protected]> > SUSE Software Solutions Germany GmbH, > Frankenstrasse 146, 90461 Nuernberg, Germany; > GF: Jochen Jaser, Andrew McDonald; (HRB 36809, AG Nuernberg)
