[Bug middle-end/61571] bad aliasing --> wrong FRE

2014-06-23 Thread xinliangli at gmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61571 --- Comment #5 from davidxl --- Thanks for the analysis. I agree it is invalid to use -fstrict-aliasing for the code. The implementation tries to save some space in __list_impl class by making the sentinel marker __end_ to be just __list_node_

[Bug middle-end/61571] bad aliasing --> wrong FRE

2014-06-23 Thread rguenth at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61571 Richard Biener changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|---

[Bug middle-end/61571] bad aliasing --> wrong FRE

2014-06-23 Thread rguenth at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61571 Richard Biener changed: What|Removed |Added Status|UNCONFIRMED |ASSIGNED Last reconfirmed|

[Bug middle-end/61571] bad aliasing --> wrong FRE

2014-06-20 Thread xinliangli at gmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61571 davidxl changed: What|Removed |Added CC||richard.guenther at gmail dot com --- Comment

[Bug middle-end/61571] bad aliasing --> wrong FRE

2014-06-20 Thread xinliangli at gmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61571 --- Comment #1 from davidxl --- Created attachment 32979 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=32979&action=edit source file