[Bug lto/88130] [9 Regression] ICE in copy_function_or_variable, at lto-streamer-out.c:2315 since r260963
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88130 Jan Hubicka changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #8 from Jan Hubicka --- Fixed.
[Bug lto/88130] [9 Regression] ICE in copy_function_or_variable, at lto-streamer-out.c:2315 since r260963
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88130 --- Comment #7 from Jan Hubicka --- Author: hubicka Date: Sun Jan 6 20:11:15 2019 New Revision: 267613 URL: https://gcc.gnu.org/viewcvs?rev=267613&root=gcc&view=rev Log: Backport from mainline 2019-01-02 Jan Hubicka PR lto/88130 * varpool.c (varpool_node::ctor_useable_for_folding_p): Also return false at WPA time when body was removed. Added: branches/gcc-8-branch/gcc/testsuite/g++.dg/torture/pr88130.C Modified: branches/gcc-8-branch/gcc/ChangeLog branches/gcc-8-branch/gcc/testsuite/ChangeLog branches/gcc-8-branch/gcc/varpool.c
[Bug lto/88130] [9 Regression] ICE in copy_function_or_variable, at lto-streamer-out.c:2315 since r260963
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88130 --- Comment #6 from Jan Hubicka --- This is fixed on mainline, but we should backport to gcc 7 and 8.
[Bug lto/88130] [9 Regression] ICE in copy_function_or_variable, at lto-streamer-out.c:2315 since r260963
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88130 --- Comment #5 from Jan Hubicka --- Author: hubicka Date: Wed Jan 2 15:23:27 2019 New Revision: 267515 URL: https://gcc.gnu.org/viewcvs?rev=267515&root=gcc&view=rev Log: PR lto/88130 * varpool.c (varpool_node::ctor_useable_for_folding_p): Also return false at WPA time when body was removed. * g++.dg/torture/pr88130.C: New testcase. Added: trunk/gcc/testsuite/g++.dg/torture/pr88130.C Modified: trunk/gcc/ChangeLog trunk/gcc/testsuite/ChangeLog trunk/gcc/varpool.c
[Bug lto/88130] [9 Regression] ICE in copy_function_or_variable, at lto-streamer-out.c:2315 since r260963
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88130 --- Comment #4 from Jan Hubicka --- Created attachment 45282 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=45282&action=edit patch I am testing This seems like ages old isse where at compile time we decide to not ship the vtable constructor (because we do not optimize) and at WPA time we think it may be good idea to stream it. This is done in lto-cgraph and we can't use optimization level there because variables have no optimize attribute and we do not know if there is code that is going to be devirtualized (well, we could probably track this info by tracking all symbols that go into partition but I doubt it is worth the effort) ctor_useable_for_folding is not considering case that the constructor is not in the stream which is fixed here.
[Bug lto/88130] [9 Regression] ICE in copy_function_or_variable, at lto-streamer-out.c:2315 since r260963
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88130 Richard Biener changed: What|Removed |Added Priority|P3 |P1
[Bug lto/88130] [9 Regression] ICE in copy_function_or_variable, at lto-streamer-out.c:2315 since r260963
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88130 Martin Liška changed: What|Removed |Added Keywords|needs-bisection | Summary|[9 Regression] ICE in |[9 Regression] ICE in |copy_function_or_variable, |copy_function_or_variable, |at lto-streamer-out.c:2315 |at lto-streamer-out.c:2315 |since r260970 |since r260963 --- Comment #3 from Martin Liška --- Revision updated.
