https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
Jakub Jelinek changed:
What|Removed |Added
Status|WAITING |NEW
--- Comment #21 from Jakub Jelinek
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #20 from Zhendong Su ---
Created attachment 43695
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43695&action=edit
output from "gcctk -O3 -S -da small.c"
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #19 from Jakub Jelinek ---
Sorry, my fault, forgot about -da, can you repeat with that option as well?
Your optimized dump is identical to my, so if there is a difference (and there
must be, because the function in question isn't call
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #18 from Zhendong Su ---
> and attach small.tar.bz2, that will contain all the dumps.
Jakub, done; please take a look.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #17 from Zhendong Su ---
Created attachment 43694
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43694&action=edit
output from "gcctk -O3 -S -fdump-tree-all small.c"
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #16 from Jakub Jelinek ---
The first dump is some gimple dump (*.*t.*, so not cse dump, and the latter one
is without optimizations, I thought you get the ICE only with -O3.
Just do:
rm -rf small.c.*
gcctk -O3 -S -fdump-tree-all small
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #15 from Zhendong Su ---
(In reply to Jakub Jelinek from comment #12)
> Can you attach -fdump-tree-optimized and -fdump-rtl-cse2
> -fdump-rtl-pro_and_epilogue -fdump-rtl-dse ? I really wonder what is
> different in our vs. your compi
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #14 from Zhendong Su ---
Created attachment 43688
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43688&action=edit
output from "gcctk -fdump-rtl-cse2 -fdump-rtl-pro_and_epilogue -fdump-rtl-dse2
-c small.c"
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #13 from Zhendong Su ---
Created attachment 43687
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43687&action=edit
output from "gcctk -fdump-rtl-cse2 -c small.c"
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #12 from Jakub Jelinek ---
Can you attach -fdump-tree-optimized and -fdump-rtl-cse2
-fdump-rtl-pro_and_epilogue -fdump-rtl-dse ? I really wonder what is different
in our vs. your compiler.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #11 from Zhendong Su ---
(In reply to Martin Liška from comment #9)
> Sorry, I meant --verbose.
$ gcctk -O3 -c --verbose small.c
Using built-in specs.
COLLECT_GCC=gcctk
Target: x86_64-pc-linux-gnu
Configured with: ../gcc-source-trunk
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #10 from Jakub Jelinek ---
I've tried all the possible -march=/-mtune= options and -m32/-m64 and don't get
ICE with any of them.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #9 from Martin Liška ---
Sorry, I meant --verbose.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #8 from Zhendong Su ---
> Can you please attach content of --save-temps?
$ gcctk -O3 -c --save-temps small.c
during RTL pass: dse1
small.c: In function ‘fn1’:
small.c:12:1: internal compiler error: in smallest_mode_for_size, at
stor-
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #7 from Martin Liška ---
(In reply to Zhendong Su from comment #6)
> (In reply to Martin Liška from comment #5)
> > > gcc version 8.0.1 20180310 (experimental) [trunk revision 258413] (GCC)
> >
> > Just a nit, this revision mentioned
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #6 from Zhendong Su ---
(In reply to Martin Liška from comment #5)
> > gcc version 8.0.1 20180310 (experimental) [trunk revision 258413] (GCC)
>
> Just a nit, this revision mentioned above is actually from GCC 7 branch.
> Isn't that
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #5 from Martin Liška ---
> gcc version 8.0.1 20180310 (experimental) [trunk revision 258413] (GCC)
Just a nit, this revision mentioned above is actually from GCC 7 branch. Isn't
that the issue?
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #4 from Jakub Jelinek ---
get_stored_val isn't called at all for me. known_subrange_p is just never
true, there is no overlap between any of the stores.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
--- Comment #3 from Zhendong Su ---
I still see the ICE with rev. 258447.
$ gcctk -v
Using built-in specs.
COLLECT_GCC=gcctk
COLLECT_LTO_WRAPPER=/home/su/software/tmp/gcc/gcc-trunk/libexec/gcc/x86_64-pc-linux-gnu/8.0.1/lto-wrapper
Target: x86_6
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
Martin Liška changed:
What|Removed |Added
Status|UNCONFIRMED |WAITING
Last reconfirmed|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84811
Jakub Jelinek changed:
What|Removed |Added
CC||jakub at gcc dot gnu.org
--- Comment #1
21 matches
Mail list logo
