This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Fix PRs 66502 and 67167
- From: Richard Biener <rguenther at suse dot de>
- To: Jiong Wang <jiong dot wang at arm dot com>
- Cc: "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>
- Date: Fri, 21 Aug 2015 11:45:12 +0200 (CEST)
- Subject: Re: [PATCH] Fix PRs 66502 and 67167
- Authentication-results: sourceware.org; auth=none
- References: <alpine dot LSU dot 2 dot 11 dot 1508130901080 dot 19998 at zhemvz dot fhfr dot qr> <n99h9ntgfbf dot fsf at arm dot com>
On Fri, 21 Aug 2015, Jiong Wang wrote:
>
> Richard Biener writes:
>
> > Given there is now PR67167 I am going forward with the earlier posted
> > patch to switch SCCVN to PHI elimination in favor of another PHI
> > (to remove IVs) rather than in favor of its only executable edge value.
> >
> > I still see no way to capture both cases without detecting the choice
> > and re-numbering the SCC twice, eventually choosing the "better" outcome.
> > And then the situation where both cases happen in the same SCC is not
> > handled either.
> >
> > Bootstrapped and tested on x86_64-unknown-linux-gnu, applied to trunk.
> >
> > Richard.
> >
> > 2015-08-13 Richard Biener <rguenther@suse.de>
> >
> > PR tree-optimization/66502
> > PR tree-optimization/67167
> > * tree-ssa-sccvn.c (vn_phi_compute_hash): Do not include
> > backedge arguments.
> > (vn_phi_lookup): Adjust.
> > (vn_phi_insert): Likewise.
> > (visit_phi): Prefer to value-number to another PHI node
> > over value-numbering to a PHI argument.
> > (init_scc_vn): Mark DFS back edges.
>
> Richard,
>
> I suspect this patch r226850 caused internal compiler error on
> ./gcc/testsuite/gcc.c-torture/compile/20121027-1.c on arm-non-eabi.
> The ICE gone away if I revert this patch.
>
> it can be easily reproduced by the following command. -mfpu and
> -mfloat are necessary.
>
> ./cc1 -O3 -nostdinc 20121027-1.c -march=armv8-a -mthumb
> -mfpu=crypto-neon-fp-armv8 -mfloat-abi=hard
>
> cc1 is generated from
>
> ../gcc/configure --target=arm-none-eabi --enable-languages=c,c++
>
> do you mind to have a look?
I see the following ICE:
t.c:13:1: internal compiler error: in decompose_normal_address, at
rtlanal.c:6090
}
^
0xc94a37 decompose_normal_address
/space/rguenther/tramp3d/trunk/gcc/rtlanal.c:6090
0xc94d25 decompose_address(address_info*, rtx_def**, machine_mode,
unsigned char, rtx_code)
/space/rguenther/tramp3d/trunk/gcc/rtlanal.c:6167
0xc94dc3 decompose_mem_address(address_info*, rtx_def*)
/space/rguenther/tramp3d/trunk/gcc/rtlanal.c:6187
0xb61149 process_address_1
/space/rguenther/tramp3d/trunk/gcc/lra-constraints.c:2867
0xb61c4e process_address
/space/rguenther/tramp3d/trunk/gcc/lra-constraints.c:3124
0xb62607 curr_insn_transform
/space/rguenther/tramp3d/trunk/gcc/lra-constraints.c:3419
0xb65250 lra_constraints(bool)
/space/rguenther/tramp3d/trunk/gcc/lra-constraints.c:4421
that looks like a latent issue to me in an area of GCC I am not
familiar with. I suggest to open a bugreport and CC Vladimir.
The r226850 change caused us to eliminate an induction variable
early (I suspect IVOPTs would have done this later anyway, but
I did not verify that):
Replaced redundant PHI node defining bl_2 with c_1
Replaced c_1 + 1 with bl_15 in all uses of c_16 = c_1 + 1;
Removing dead stmt c_16 = c_1 + 1;
Removing dead stmt bl_2 = PHI <0(2), bl_15(3)>
Thanks,
Richard.
> Thanks.
>
--
Richard Biener <rguenther@suse.de>
SUSE LINUX GmbH, GF: Felix Imendoerffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nuernberg)