Re: [patch] Fix reverse SSO issues in IPA-SRA

2022-01-14 Thread Martin Jambor
Hi Eric, On Tue, Jan 11 2022, Eric Botcazou via Gcc-patches wrote: > Hi, > > we recently received the report that the IPA-SRA pass introduced in GCC 10 > does not always play nice with the reverse scalar storage order that can be > used in structures/records/unions. Reading the code, the pass a

Re: [patch] Fix reverse SSO issues in IPA-SRA

2022-01-14 Thread Eric Botcazou via Gcc-patches
> Yes, it must still be copied. OK, revised patch attached, with testcases but they fail only on the 10 and 11 branches because of a change in the heuristics apparently. * ipa-param-manipulation.c (ipa_dump_adjusted_parameters): Dump reverse flag as "reverse" for the sake of con

Re: [patch] Fix reverse SSO issues in IPA-SRA

2022-01-12 Thread Martin Jambor
Hi, On Wed, Jan 12 2022, Eric Botcazou wrote: >> Thanks for the fixes, the forgotten duplication and streaming were quite >> embarrassing, but one reason is that there are few reverse SSO testcases >> in the suite. Would it be difficult to add some covering the issues you >> fixed? > > No, I thin

Re: [patch] Fix reverse SSO issues in IPA-SRA

2022-01-12 Thread Eric Botcazou via Gcc-patches
> Thanks for the fixes, the forgotten duplication and streaming were quite > embarrassing, but one reason is that there are few reverse SSO testcases > in the suite. Would it be difficult to add some covering the issues you > fixed? No, I think I should be able to cover 3 out of the 4 changes, se

Re: [patch] Fix reverse SSO issues in IPA-SRA

2022-01-11 Thread Martin Jambor
Hi Eric, On Tue, Jan 11 2022, Eric Botcazou via Gcc-patches wrote: > Hi, > > we recently received the report that the IPA-SRA pass introduced in GCC 10 > does not always play nice with the reverse scalar storage order that can be > used in structures/records/unions. Reading the code, the pass a

Re: [patch] Fix reverse SSO issues in IPA-SRA

2022-01-11 Thread Richard Biener via Gcc-patches
On Tue, Jan 11, 2022 at 11:45 AM Eric Botcazou via Gcc-patches wrote: > > Hi, > > we recently received the report that the IPA-SRA pass introduced in GCC 10 > does not always play nice with the reverse scalar storage order that can be > used in structures/records/unions. Reading the code, the pas

[patch] Fix reverse SSO issues in IPA-SRA

2022-01-11 Thread Eric Botcazou via Gcc-patches
Hi, we recently received the report that the IPA-SRA pass introduced in GCC 10 does not always play nice with the reverse scalar storage order that can be used in structures/records/unions. Reading the code, the pass apparently correctly detects it but fails to propagate the information to the