Re: redundant bitmap_bit_p followed by bitmap_clear_bit [was: Re: [COMMITTED] Kill second order relations in the path solver.]

On Tue, Nov 2, 2021 at 10:00 PM Bernhard Reutner-Fischer wrote: > > On 2 November 2021 14:43:38 CET, Richard Biener > wrote: > >On Mon, Nov 1, 2021 at 10:02 PM Bernhard Reutner-Fischer via > >Gcc-patches wrote: > >> > >> On Mon, 1 Nov 2021 15:21:03 +0100 > >> Aldy Hernandez wrote: > >> > >> >

Re: redundant bitmap_bit_p followed by bitmap_clear_bit [was: Re: [COMMITTED] Kill second order relations in the path solver.]

On 2 November 2021 14:43:38 CET, Richard Biener wrote: >On Mon, Nov 1, 2021 at 10:02 PM Bernhard Reutner-Fischer via >Gcc-patches wrote: >> >> On Mon, 1 Nov 2021 15:21:03 +0100 >> Aldy Hernandez wrote: >> >> > I'm not convinced this makes the code clearer to read, especially if >> > it's not on

Re: redundant bitmap_bit_p followed by bitmap_clear_bit [was: Re: [COMMITTED] Kill second order relations in the path solver.]

Richard Biener via Gcc-patches writes: > On Mon, Nov 1, 2021 at 10:02 PM Bernhard Reutner-Fischer via > Gcc-patches wrote: >> >> On Mon, 1 Nov 2021 15:21:03 +0100 >> Aldy Hernandez wrote: >> >> > I'm not convinced this makes the code clearer to read, especially if >> > it's not on a critical pat

Re: redundant bitmap_bit_p followed by bitmap_clear_bit [was: Re: [COMMITTED] Kill second order relations in the path solver.]

On Mon, Nov 1, 2021 at 10:02 PM Bernhard Reutner-Fischer via Gcc-patches wrote: > > On Mon, 1 Nov 2021 15:21:03 +0100 > Aldy Hernandez wrote: > > > I'm not convinced this makes the code clearer to read, especially if > > it's not on a critical path. But if you feel strongly, please submit > > a

Re: redundant bitmap_bit_p followed by bitmap_clear_bit [was: Re: [COMMITTED] Kill second order relations in the path solver.]

On Mon, 1 Nov 2021 15:21:03 +0100 Aldy Hernandez wrote: > I'm not convinced this makes the code clearer to read, especially if > it's not on a critical path. But if you feel strongly, please submit > a patch ;-). No i don't feel strongly about it. Compiling e.g. -O2 ira.o # Overhead Sampl

Re: redundant bitmap_bit_p followed by bitmap_clear_bit [was: Re: [COMMITTED] Kill second order relations in the path solver.]

I'm not convinced this makes the code clearer to read, especially if it's not on a critical path. But if you feel strongly, please submit a patch ;-). Aldy On Mon, Nov 1, 2021 at 3:10 PM Bernhard Reutner-Fischer wrote: > > On Thu, 28 Oct 2021 01:55:30 +0200 > Bernhard Reutner-Fischer wrote: >

redundant bitmap_bit_p followed by bitmap_clear_bit [was: Re: [COMMITTED] Kill second order relations in the path solver.]

On Thu, 28 Oct 2021 01:55:30 +0200 Bernhard Reutner-Fischer wrote: > On Wed, 27 Oct 2021 20:13:21 +0200 > Aldy Hernandez via Gcc-patches wrote: > > @@ -1306,6 +1307,24 @@ path_oracle::killing_def (tree ssa) > >ptr->m_next = m_equiv.m_next; > >m_equiv.m_next = ptr; > >bitmap_ior_into

Re: [COMMITTED] Kill second order relations in the path solver.

On Wed, 27 Oct 2021 20:13:21 +0200 Aldy Hernandez via Gcc-patches wrote: [would have to think about this some more but it's late here. Nits:] > diff --git a/gcc/value-relation.cc b/gcc/value-relation.cc > index 2acf375ca9a..0ad4f7a9495 100644 > --- a/gcc/value-relation.cc > +++ b/gcc/value-relat

[COMMITTED] Kill second order relations in the path solver.

My upcoming work replacing the VRP threaders with a fully resolving backward threader has tripped over various corner cases in the path sensitive relation oracle. This patch kills second order relations when we kill a relation. Tested on x86-64 and ppc64le Linux. Co-authored-by: Andrew MacLeod