| Message ID | 000e01cfb224$7ba8caa0$72fa5fe0$@arm.com |
|---|---|
| State | New |
| Headers | show |
On Thu, Aug 7, 2014 at 11:46 AM, Bin Cheng <bin.cheng@arm.com> wrote: > Hi, > As analyzed in PR62032, this patch fixes the latent lto bug by switching > arguments of lto_define_builtins, otherwise vsnprintf-chk.c would fail on > arm/aarch64 with lto options. > > Is it ok if bootstrap and test pass? Ok (would be nice if you can check if LTO bootstrap still works). Thanks, Richard. > > 2014-08-07 Bin Cheng <bin.cheng@arm.com> > > PR lto/62032 > * lto/lto-lang.c (lto_init): Switch mis-matched arguments.
On Thu, Aug 7, 2014 at 8:06 PM, Richard Biener <richard.guenther@gmail.com> wrote: > On Thu, Aug 7, 2014 at 11:46 AM, Bin Cheng <bin.cheng@arm.com> wrote: >> Hi, >> As analyzed in PR62032, this patch fixes the latent lto bug by switching >> arguments of lto_define_builtins, otherwise vsnprintf-chk.c would fail on >> arm/aarch64 with lto options. >> >> Is it ok if bootstrap and test pass? > > Ok (would be nice if you can check if LTO bootstrap still works). Done. Bootstrap with "--with-build-config=bootstrap-lto" works fine. I will commit this to trunk and wait for couple of days before back-port to 4_9 branch. Is that OK? BTW, I found GCC trunk@213701 failed when building ada with below ICE message, so I have to do the bootstrap with ada disabled. ../../gcc/gcc/ada/exp_ch3.adb: In function 'exp_ch3__get_simple_init_val': ../../gcc/gcc/ada/exp_ch3.adb:7695:4: internal compiler error: in forward_edge_to_pdom, at tree-ssa-dce.c:1042 function Get_Simple_Init_Val ^ 0xc23a8d forward_edge_to_pdom ../../gcc/gcc/tree-ssa-dce.c:1042 0xc23c21 remove_dead_stmt ../../gcc/gcc/tree-ssa-dce.c:1091 0xc24120 eliminate_unnecessary_stmts ../../gcc/gcc/tree-ssa-dce.c:1218 0xc24b23 perform_tree_ssa_dce ../../gcc/gcc/tree-ssa-dce.c:1460 0xc24c1c tree_ssa_cd_dce ../../gcc/gcc/tree-ssa-dce.c:1501 0xc24d86 execute ../../gcc/gcc/tree-ssa-dce.c:1566 Please submit a full bug report, with preprocessed source if appropriate. Please include the complete backtrace with any bug report. See <http://gcc.gnu.org/bugs.html> for instructions. make[4]: *** [/tmp/cc4oj0WO.ltrans1.ltrans.o] Error 1 lto-wrapper: fatal error: make returned 2 exit status compilation terminated. /usr/bin/ld: lto-wrapper failed collect2: error: ld returned 1 exit status make[3]: *** [gnat1] Error 1 make[3]: *** Waiting for unfinished jobs.... Thanks, bin
On Fri, Aug 8, 2014 at 10:05 AM, Bin.Cheng <amker.cheng@gmail.com> wrote: > On Thu, Aug 7, 2014 at 8:06 PM, Richard Biener > <richard.guenther@gmail.com> wrote: >> On Thu, Aug 7, 2014 at 11:46 AM, Bin Cheng <bin.cheng@arm.com> wrote: >>> Hi, >>> As analyzed in PR62032, this patch fixes the latent lto bug by switching >>> arguments of lto_define_builtins, otherwise vsnprintf-chk.c would fail on >>> arm/aarch64 with lto options. >>> >>> Is it ok if bootstrap and test pass? >> >> Ok (would be nice if you can check if LTO bootstrap still works). > Done. Bootstrap with "--with-build-config=bootstrap-lto" works fine. > I will commit this to trunk and wait for couple of days before > back-port to 4_9 branch. Is that OK? Sure. > BTW, I found GCC trunk@213701 failed when building ada with below ICE > message, so I have to do the bootstrap with ada disabled. Heh - I usually do LTO bootstrap only for c,c++,fortran as otherwise it takes ages... ;) Richard. > ../../gcc/gcc/ada/exp_ch3.adb: In function 'exp_ch3__get_simple_init_val': > ../../gcc/gcc/ada/exp_ch3.adb:7695:4: internal compiler error: in > forward_edge_to_pdom, at tree-ssa-dce.c:1042 > function Get_Simple_Init_Val > ^ > 0xc23a8d forward_edge_to_pdom > ../../gcc/gcc/tree-ssa-dce.c:1042 > 0xc23c21 remove_dead_stmt > ../../gcc/gcc/tree-ssa-dce.c:1091 > 0xc24120 eliminate_unnecessary_stmts > ../../gcc/gcc/tree-ssa-dce.c:1218 > 0xc24b23 perform_tree_ssa_dce > ../../gcc/gcc/tree-ssa-dce.c:1460 > 0xc24c1c tree_ssa_cd_dce > ../../gcc/gcc/tree-ssa-dce.c:1501 > 0xc24d86 execute > ../../gcc/gcc/tree-ssa-dce.c:1566 > Please submit a full bug report, > with preprocessed source if appropriate. > Please include the complete backtrace with any bug report. > See <http://gcc.gnu.org/bugs.html> for instructions. > make[4]: *** [/tmp/cc4oj0WO.ltrans1.ltrans.o] Error 1 > lto-wrapper: fatal error: make returned 2 exit status > compilation terminated. > /usr/bin/ld: lto-wrapper failed > collect2: error: ld returned 1 exit status > make[3]: *** [gnat1] Error 1 > make[3]: *** Waiting for unfinished jobs.... > > Thanks, > bin
On Fri, Aug 8, 2014 at 4:54 PM, Richard Biener <richard.guenther@gmail.com> wrote: > On Fri, Aug 8, 2014 at 10:05 AM, Bin.Cheng <amker.cheng@gmail.com> wrote: >> On Thu, Aug 7, 2014 at 8:06 PM, Richard Biener >> <richard.guenther@gmail.com> wrote: >>> On Thu, Aug 7, 2014 at 11:46 AM, Bin Cheng <bin.cheng@arm.com> wrote: >>>> Hi, >>>> As analyzed in PR62032, this patch fixes the latent lto bug by switching >>>> arguments of lto_define_builtins, otherwise vsnprintf-chk.c would fail on >>>> arm/aarch64 with lto options. >>>> >>>> Is it ok if bootstrap and test pass? >>> >>> Ok (would be nice if you can check if LTO bootstrap still works). >> Done. Bootstrap with "--with-build-config=bootstrap-lto" works fine. >> I will commit this to trunk and wait for couple of days before >> back-port to 4_9 branch. Is that OK? > > Sure. > >> BTW, I found GCC trunk@213701 failed when building ada with below ICE >> message, so I have to do the bootstrap with ada disabled. > > Heh - I usually do LTO bootstrap only for c,c++,fortran as otherwise it > takes ages... ;) Yes,it is slow, good news is lto works for java,objc,objc++,go on my machine. Thanks, bin > > Richard. > >> ../../gcc/gcc/ada/exp_ch3.adb: In function 'exp_ch3__get_simple_init_val': >> ../../gcc/gcc/ada/exp_ch3.adb:7695:4: internal compiler error: in >> forward_edge_to_pdom, at tree-ssa-dce.c:1042 >> function Get_Simple_Init_Val >> ^ >> 0xc23a8d forward_edge_to_pdom >> ../../gcc/gcc/tree-ssa-dce.c:1042 >> 0xc23c21 remove_dead_stmt >> ../../gcc/gcc/tree-ssa-dce.c:1091 >> 0xc24120 eliminate_unnecessary_stmts >> ../../gcc/gcc/tree-ssa-dce.c:1218 >> 0xc24b23 perform_tree_ssa_dce >> ../../gcc/gcc/tree-ssa-dce.c:1460 >> 0xc24c1c tree_ssa_cd_dce >> ../../gcc/gcc/tree-ssa-dce.c:1501 >> 0xc24d86 execute >> ../../gcc/gcc/tree-ssa-dce.c:1566 >> Please submit a full bug report, >> with preprocessed source if appropriate. >> Please include the complete backtrace with any bug report. >> See <http://gcc.gnu.org/bugs.html> for instructions. >> make[4]: *** [/tmp/cc4oj0WO.ltrans1.ltrans.o] Error 1 >> lto-wrapper: fatal error: make returned 2 exit status >> compilation terminated. >> /usr/bin/ld: lto-wrapper failed >> collect2: error: ld returned 1 exit status >> make[3]: *** [gnat1] Error 1 >> make[3]: *** Waiting for unfinished jobs.... >> >> Thanks, >> bin
On Fri, Aug 8, 2014 at 4:54 PM, Richard Biener <richard.guenther@gmail.com> wrote: > On Fri, Aug 8, 2014 at 10:05 AM, Bin.Cheng <amker.cheng@gmail.com> wrote: >> On Thu, Aug 7, 2014 at 8:06 PM, Richard Biener >> <richard.guenther@gmail.com> wrote: >>> On Thu, Aug 7, 2014 at 11:46 AM, Bin Cheng <bin.cheng@arm.com> wrote: >>>> Hi, >>>> As analyzed in PR62032, this patch fixes the latent lto bug by switching >>>> arguments of lto_define_builtins, otherwise vsnprintf-chk.c would fail on >>>> arm/aarch64 with lto options. >>>> >>>> Is it ok if bootstrap and test pass? >>> >>> Ok (would be nice if you can check if LTO bootstrap still works). >> Done. Bootstrap with "--with-build-config=bootstrap-lto" works fine. >> I will commit this to trunk and wait for couple of days before >> back-port to 4_9 branch. Is that OK? > > Sure. Since the patch is on trunk for one week with no issue reported, I backported it to gcc-4_9 branch as revision 214000. Thanks, bin > >> BTW, I found GCC trunk@213701 failed when building ada with below ICE >> message, so I have to do the bootstrap with ada disabled. > > Heh - I usually do LTO bootstrap only for c,c++,fortran as otherwise it > takes ages... ;) > > Richard. > >> ../../gcc/gcc/ada/exp_ch3.adb: In function 'exp_ch3__get_simple_init_val': >> ../../gcc/gcc/ada/exp_ch3.adb:7695:4: internal compiler error: in >> forward_edge_to_pdom, at tree-ssa-dce.c:1042 >> function Get_Simple_Init_Val >> ^ >> 0xc23a8d forward_edge_to_pdom >> ../../gcc/gcc/tree-ssa-dce.c:1042 >> 0xc23c21 remove_dead_stmt >> ../../gcc/gcc/tree-ssa-dce.c:1091 >> 0xc24120 eliminate_unnecessary_stmts >> ../../gcc/gcc/tree-ssa-dce.c:1218 >> 0xc24b23 perform_tree_ssa_dce >> ../../gcc/gcc/tree-ssa-dce.c:1460 >> 0xc24c1c tree_ssa_cd_dce >> ../../gcc/gcc/tree-ssa-dce.c:1501 >> 0xc24d86 execute >> ../../gcc/gcc/tree-ssa-dce.c:1566 >> Please submit a full bug report, >> with preprocessed source if appropriate. >> Please include the complete backtrace with any bug report. >> See <http://gcc.gnu.org/bugs.html> for instructions. >> make[4]: *** [/tmp/cc4oj0WO.ltrans1.ltrans.o] Error 1 >> lto-wrapper: fatal error: make returned 2 exit status >> compilation terminated. >> /usr/bin/ld: lto-wrapper failed >> collect2: error: ld returned 1 exit status >> make[3]: *** [gnat1] Error 1 >> make[3]: *** Waiting for unfinished jobs.... >> >> Thanks, >> bin
Index: gcc/lto/lto-lang.c =================================================================== --- gcc/lto/lto-lang.c (revision 213698) +++ gcc/lto/lto-lang.c (working copy) @@ -1192,10 +1192,10 @@ lto_init (void) } else { - lto_define_builtins (va_list_type_node, - build_reference_type (va_list_type_node)); + lto_define_builtins (build_reference_type (va_list_type_node), + va_list_type_node); } - + if (flag_cilkplus) cilk_init_builtins ();