[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-13 Thread krebbel at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 Andreas Krebbel changed: What|Removed |Added CC||doko at gcc dot gnu.org --- Comment

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread jakub at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 Jakub Jelinek changed: What|Removed |Added Status|NEW |RESOLVED Resolution|---

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread jakub at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 --- Comment #11 from Jakub Jelinek --- Author: jakub Date: Thu Jan 12 21:29:43 2017 New Revision: 244383 URL: https://gcc.gnu.org/viewcvs?rev=244383=gcc=rev Log: PR bootstrap/79069 * cfgrtl.c (rtl_tidy_fallthru_edge): For

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread jakub at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 --- Comment #10 from Jakub Jelinek --- Seems the barrier after the unconditional __morestack call is removed by rtl_tidy_fallthru_edge. I think: --- gcc/cfgrtl.c.jj 2017-01-01 12:45:35.0 +0100 +++ gcc/cfgrtl.c2017-01-12

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread jakub at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 Jakub Jelinek changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed|

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread jakub at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 --- Comment #8 from Jakub Jelinek --- Maybe the easiest fix would be to emit the __morestack magic call patterns as ASM_OPERANDS like one that is emitted for asm goto. The cfgrtl.c etc. code has already years of fixes to handle those calls and

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread jakub at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 --- Comment #7 from Jakub Jelinek --- The ICE is during the jump2 pass, there is EDGE_FALLTHRU in between (note 21 1 942 2 [bb 2] NOTE_INSN_BASIC_BLOCK) (insn 942 21 943 2 (unspec_volatile [ (code_label 0 0 0 10311 (nil) [2 uses])

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread vogt at linux dot vnet.ibm.com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 --- Comment #6 from Dominik Vogt --- Confirmed; bisecting now.

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread jakub at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 --- Comment #5 from Jakub Jelinek --- (In reply to Dominik Vogt from comment #3) > > --disable-bootstrap > > ? Originally (the #c0 error) it has been of course --enable-bootstrap). But when trying to reproduce it, I didn't want to wait so

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread jakub at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 --- Comment #4 from Jakub Jelinek --- And with r244350 ../configure --disable-bootstrap --enable-languages=c,c++,go,lto --with-long-double-128 --with-arch=z9-109 --with-tune=z10 reproduced too. Note that without --enable-checking=release, the

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread vogt at linux dot vnet.ibm.com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 --- Comment #3 from Dominik Vogt --- > --disable-bootstrap ?

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread jakub at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 --- Comment #2 from Jakub Jelinek --- Vanilla r244305, and ../configure --disable-bootstrap --enable-languages=c,c++,go,lto \ --with-long-double-128 --with-arch=z9-109 --with-tune=z10 make -j4 on s390x. Let me retry with latest trunk.

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread vogt at linux dot vnet.ibm.com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 --- Comment #1 from Dominik Vogt --- What are the revision and the configure flags that trigger this, please? r244350 bootstraps without problem here.

[Bug bootstrap/79069] [7 Regression] Bootstrap failure on s390x-linux while building libgo

2017-01-12 Thread jakub at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79069 Jakub Jelinek changed: What|Removed |Added CC||ian at gcc dot gnu.org,