[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009 Martin Liška changed: What|Removed |Added Status|WAITING |RESOLVED Resolution|--- |INVALID --- Comment #15 from Martin Liška --- I'm closing this, please open a separate issue.
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009 Eric Gallager changed: What|Removed |Added CC||egallager at gcc dot gnu.org, ||marxin at gcc dot gnu.org, ||mpolacek at gcc dot gnu.org Status|NEW |WAITING --- Comment #14 from Eric Gallager --- Note that I can't reproduce my original bug anymore. Ryan, would you like to keep this bug open for your variation on it, or are you going to open a separate bug for it?
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009 --- Comment #13 from Eric Gallager --- (In reply to Ryan Schmidt from comment #12) > How would I do that? Add -save-temps to the compile command and then attach the lj_cconv.i file that is produced
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009 --- Comment #12 from Ryan Schmidt --- How would I do that?
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009 --- Comment #11 from Eric Gallager --- (In reply to Ryan Schmidt from comment #9) > I am also encountering this problem on i386-apple-darwin9.8.0 when compiling > texlive-bin 20170604 with gcc 6.4.0, though in my case it's -Os not -Ofast: > > libtool: compile: /opt/local/bin/gcc-mp-6 -DHAVE_CONFIG_H -I. > -I./LuaJIT-src/src -DLUAJIT_ENABLE_LUA52COMPAT -DLUAI_HASHLIMIT=6 > -U_FORTIFY_SOURCE -isystem/opt/local/include -fomit-frame-pointer > -march=i686 -msse -msse2 -mfpmath=sse -fno-stack-protector -Wall -pipe -Os > -m32 -MT LuaJIT-src/src/lj_cconv.lo -MD -MP -MF > LuaJIT-src/src/.deps/lj_cconv.Tpo -c LuaJIT-src/src/lj_cconv.c -fno-common > -DPIC -o LuaJIT-src/src/.libs/lj_cconv.o > LuaJIT-src/src/lj_cconv.c: In function 'lj_cconv_ct_ct': > LuaJIT-src/src/lj_cconv.c:368:1: internal compiler error: in gen_reg_rtx, at > emit-rtl.c:1025 > } > ^ > libbacktrace could not find executable to open Actually could you attach the preprocessed source for lj_cconv.c anyways? I want to compare to see if it really is the same bug as I was originally experiencing.
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009 Eric Gallager changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed||2017-12-30 CC||egallager at gcc dot gnu.org Ever confirmed|0 |1 --- Comment #10 from Eric Gallager --- (In reply to Ryan Schmidt from comment #9) > I am also encountering this problem on i386-apple-darwin9.8.0 when compiling > texlive-bin 20170604 with gcc 6.4.0, though in my case it's -Os not -Ofast: > > libtool: compile: /opt/local/bin/gcc-mp-6 -DHAVE_CONFIG_H -I. > -I./LuaJIT-src/src -DLUAJIT_ENABLE_LUA52COMPAT -DLUAI_HASHLIMIT=6 > -U_FORTIFY_SOURCE -isystem/opt/local/include -fomit-frame-pointer > -march=i686 -msse -msse2 -mfpmath=sse -fno-stack-protector -Wall -pipe -Os > -m32 -MT LuaJIT-src/src/lj_cconv.lo -MD -MP -MF > LuaJIT-src/src/.deps/lj_cconv.Tpo -c LuaJIT-src/src/lj_cconv.c -fno-common > -DPIC -o LuaJIT-src/src/.libs/lj_cconv.o > LuaJIT-src/src/lj_cconv.c: In function 'lj_cconv_ct_ct': > LuaJIT-src/src/lj_cconv.c:368:1: internal compiler error: in gen_reg_rtx, at > emit-rtl.c:1025 > } > ^ > libbacktrace could not find executable to open I'll take this as confirmation.
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009 Ryan Schmidt changed: What|Removed |Added CC||gcc at ryandesign dot com --- Comment #9 from Ryan Schmidt --- I am also encountering this problem on i386-apple-darwin9.8.0 when compiling texlive-bin 20170604 with gcc 6.4.0, though in my case it's -Os not -Ofast: libtool: compile: /opt/local/bin/gcc-mp-6 -DHAVE_CONFIG_H -I. -I./LuaJIT-src/src -DLUAJIT_ENABLE_LUA52COMPAT -DLUAI_HASHLIMIT=6 -U_FORTIFY_SOURCE -isystem/opt/local/include -fomit-frame-pointer -march=i686 -msse -msse2 -mfpmath=sse -fno-stack-protector -Wall -pipe -Os -m32 -MT LuaJIT-src/src/lj_cconv.lo -MD -MP -MF LuaJIT-src/src/.deps/lj_cconv.Tpo -c LuaJIT-src/src/lj_cconv.c -fno-common -DPIC -o LuaJIT-src/src/.libs/lj_cconv.o LuaJIT-src/src/lj_cconv.c: In function 'lj_cconv_ct_ct': LuaJIT-src/src/lj_cconv.c:368:1: internal compiler error: in gen_reg_rtx, at emit-rtl.c:1025 } ^ libbacktrace could not find executable to open
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009 --- Comment #8 from Eric Gallager --- I reduced the command line required to trigger the ICE. It depends on whether -fmath-errno is set or not. With: $ /usr/local/bin/gcc -c -Ofast -I. -I../include -DHAVE_CONFIG_H -I../bfd -Imacosx -Iconfig -Wno-deprecated-declarations -fno-math-errno valarith.c ...it ICEs as it did previously, but with: $ /usr/local/bin/gcc -c -Ofast -I. -I../include -DHAVE_CONFIG_H -I../bfd -Imacosx -Iconfig -Wno-deprecated-declarations -fmath-errno valarith.c ...it compiles successfully. That could explain why it wouldn't reproduce on x86_64-linux-gnu, as I believe the default for -fmath-errno is different between Darwin and GNU/Linux.
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009 --- Comment #7 from Eric Gallager --- Created attachment 39089 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=39089=edit some compressed compiler output Still happens with trunk from the end of July. I tried playing around with some of the developer options to generate some dumpfiles, but there's too many of them, and when I group them into folders, the archives are too big to upload even after compressing them, so I'm not really sure what to upload, besides the attached tempfiles. Current version info: $ /usr/local/bin/g++ -march=native -v Using built-in specs. COLLECT_GCC=/usr/local/bin/g++ COLLECT_LTO_WRAPPER=/usr/local/libexec/gcc/i386-apple-darwin9.8.0/7.0.0/lto-wrapper Target: i386-apple-darwin9.8.0 Configured with: ../configure --disable-werror --enable-languages=c,c++,lto,objc,obj-c++ --enable-stage1-checking=release -C --with-system-libunwind --enable-secureplt --enable-frame-pointer --enable-debug --with-isl --enable-objc-gc --enable-host-shared CC=/usr/local/bin/gcc CXX=/usr/local/bin/g++ Thread model: posix gcc version 7.0.0 20160730 (experimental) (GCC)
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009
--- Comment #6 from Eric Gallager ---
I investigated more. What's happening at ../../gcc/emit-rtl.c:1025 is that the
following assert is failing:
gcc_assert (can_create_pseudo_p ());
This fails if and only if gen_reg_rtx is passed V2DImode as its mode parameter.
I tried editing emit-rtl.c to make the assert conditional, like this:
if (mode != V2DImode)
gcc_assert (can_create_pseudo_p ());
...but that only resulted in a different error:
valarith.c: In function ‘value* value_binop(value*, value*, exp_opcode)’:
valarith.c:1408:1: error: insn does not satisfy its constraints:
(insn 3157 1585 3158 192 (set (reg:V2DI 1029)
(vec_concat:V2DI (mem/c:DI (reg/f:SI 7 sp) [50 %sfp+-64 S8 A128])
(const_int 0 [0]))) valarith.c:1180 3653 {vec_concatv2di}
(nil))
Breakpoint 1, internal_error (gmsgid=0x15ee807 "in %s, at %s:%d") at
../../gcc/diagnostic.c:1337
1337 va_start (ap, gmsgid);
Any ideas what's going on?
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009
--- Comment #5 from Eric Gallager ---
I rebuilt with debugging info and got a better backtrace:
Breakpoint 1, internal_error (gmsgid=0x15ee807 "in %s, at %s:%d") at
../../gcc/diagnostic.c:1337
1337 va_start (ap, gmsgid);
(gdb) bt
#0 internal_error (gmsgid=0x15ee807 "in %s, at %s:%d") at
../../gcc/diagnostic.c:1337
#1 0x01266866 in fancy_abort (file=0x157ccd5 "../../gcc/emit-rtl.c",
line=1025, function=0x190cd98 "gen_reg_rtx") at ../../gcc/diagnostic.c:1405
#2 0x00945479 in gen_reg_rtx (mode=V2DImode) at ../../gcc/emit-rtl.c:1025
#3 0x00579f86 in gen_split_349 (curr_insn=0x325acf0, operands=0x1ff95c0) at
../../gcc/config/i386/sse.md:1129
#4 0x0076bb58 in split_8 (x1=0x3255ce4, insn=0x325acf0) at
../../gcc/config/i386/sse.md:1115
#5 0x00778c75 in split_insns (x1=0x3255ce4, insn=0x325acf0) at
../../gcc/config/i386/sse.md:13387
#6 0x0094b3e4 in try_split (pat=0x3255ce4, trial=0x325acf0, last=1) at
../../gcc/emit-rtl.c:3658
#7 0x00d4a502 in split_insn (insn=0x325acf0) at ../../gcc/recog.c:2887
#8 0x00d4a7c5 in split_all_insns () at ../../gcc/recog.c:2977
#9 0x00d4c34b in rest_of_handle_split_after_reload () at
../../gcc/recog.c:3913
#10 0x00d4c389 in (anonymous namespace)::pass_split_after_reload::execute
(this=0x50d0e190) at ../../gcc/recog.c:3942
#11 0x00d12144 in execute_one_pass (pass=0x50d0e190) at ../../gcc/passes.c:2344
#12 0x00d12495 in execute_pass_list_1 (pass=0x50d0e190) at
../../gcc/passes.c:2428
#13 0x00d124c5 in execute_pass_list_1 (pass=0x50d0e0d0) at
../../gcc/passes.c:2429
#14 0x00d124c5 in execute_pass_list_1 (pass=0x50d0d590) at
../../gcc/passes.c:2429
#15 0x00d12515 in execute_pass_list (fn=0x2b35bb8, pass=0x50d0b190) at
../../gcc/passes.c:2439
#16 0x008539ee in cgraph_node::expand (this=0x2d4b424) at
../../gcc/cgraphunit.c:1983
#17 0x00853fff in expand_all_functions () at ../../gcc/cgraphunit.c:2119
#18 0x00854c8b in symbol_table::compile (this=0x50e0b000) at
../../gcc/cgraphunit.c:2475
#19 0x00854ed1 in symbol_table::finalize_compilation_unit (this=0x50e0b000) at
../../gcc/cgraphunit.c:2565
#20 0x00e246cd in compile_file () at ../../gcc/toplev.c:490
#21 0x00e26f70 in do_compile () at ../../gcc/toplev.c:1998
#22 0x00e27332 in toplev::main (this=0xbfff94de, argc=140, argv=0xbfff9510) at
../../gcc/toplev.c:2132
#23 0x0124f80f in main (argc=140, argv=0xbfff9510) at ../../gcc/main.c:39
Current language: auto; currently c++
(gdb) bt full
#0 internal_error (gmsgid=0x15ee807 "in %s, at %s:%d") at
../../gcc/diagnostic.c:1337
ap = 0x0
richloc = {
static MAX_RANGES = ,
static MAX_FIXIT_HINTS = ,
m_num_ranges = 0,
m_ranges = {{
m_loc = 33527040,
m_show_caret_p = 204
}, {
m_loc = 0,
m_show_caret_p = false
}, {
m_loc = 0,
m_show_caret_p = true
}},
m_column_override = 0,
m_have_expanded_location = 104,
m_expanded_location = {
file = 0xd47953 "??\020\017U???\030?x?\001\002??u2?E\b\017?",
line = 18,
column = 1356874188,
data = 0x0,
sysp = false
},
m_num_fixit_hints = 18,
m_fixit_hints = {0x50e041cc, 0x0}
}
__FUNCTION__ = "internal_error"
#1 0x01266866 in fancy_abort (file=0x157ccd5 "../../gcc/emit-rtl.c",
line=1025, function=0x190cd98 "gen_reg_rtx") at ../../gcc/diagnostic.c:1405
No locals.
#2 0x00945479 in gen_reg_rtx (mode=V2DImode) at ../../gcc/emit-rtl.c:1025
val = (rtx) 0xd479ef
align = 128
__FUNCTION__ = "gen_reg_rtx"
#3 0x00579f86 in gen_split_349 (curr_insn=0x325acf0, operands=0x1ff95c0) at
../../gcc/config/i386/sse.md:1129
tmp = (rtx) 0x50e0e430
_val = (rtx_insn *) 0x0
__FUNCTION__ = "gen_split_349"
#4 0x0076bb58 in split_8 (x1=0x3255ce4, insn=0x325acf0) at
../../gcc/config/i386/sse.md:1115
operands = (rtx * const) 0x1ff95c0
x2 = (rtx) 0x3259dd0
x3 = (rtx) 0x3259dc0
x4 = (rtx) 0x32c22f0
x5 = (rtx) 0x27c6018
x6 = (rtx) 0x0
x7 = (rtx) 0x3255ccc
x8 = (rtx) 0x3254eb0
x9 = (rtx) 0x32797e0
x10 = (rtx) 0x0
#5 0x00778c75 in split_insns (x1=0x3255ce4, insn=0x325acf0) at
../../gcc/config/i386/sse.md:13387
operands = (rtx * const) 0x1ff95c0
x2 = (rtx) 0xbfff91d8
x3 = (rtx) 0xbfff922b
x4 = (rtx) 0x2d4b424
x5 = (rtx) 0x3251a28
x6 = (rtx) 0xbfff91c8
x7 = (rtx) 0x27c0cdc
x8 = (rtx) 0x325acf0
x9 = (rtx) 0xc07a4e
x10 = (rtx) 0xbfff91c8
x11 = (rtx) 0x0
x12 = (rtx) 0x3251a28
x13 = (rtx) 0x95215c
x14 = (rtx) 0xbfff91b8
x15 = (rtx) 0x27c0cdc
x16 = (rtx) 0x325ad14
x17 = (rtx) 0x9521eb
#6 0x0094b3e4 in try_split (pat=0x3255ce4, trial=0x325acf0, last=1) at
../../gcc/emit-rtl.c:3658
before = (rtx_insn *) 0x325aca8
after = (rtx_insn *) 0x325ad14
note = (rtx) 0x325aca8
seq = (rtx_insn *) 0x0
tem = (rtx_insn *) 0xbfff9248
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009 Martin Liška changed: What|Removed |Added CC||marxin at gcc dot gnu.org --- Comment #4 from Martin Liška --- I also can't reproduce the ICE on x86_64-linux-gnu.
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009 --- Comment #3 from Eric Gallager --- Still happens with: $ /usr/local/bin/g++ -v Using built-in specs. COLLECT_GCC=/usr/local/bin/g++ COLLECT_LTO_WRAPPER=/usr/local/libexec/gcc/i386-apple-darwin9.8.0/7.0.0/lto-wrapper Target: i386-apple-darwin9.8.0 Configured with: ../configure --disable-werror --enable-languages=c,c++,lto,objc,obj-c++ --enable-stage1-checking=release -C --with-system-libunwind --enable-secureplt --enable-frame-pointer --enable-debug --without-isl --enable-objc-gc --enable-host-shared CC=/usr/local/bin/gcc CXX=/usr/local/bin/g++ Thread model: posix gcc version 7.0.0 20160526 (experimental) (GCC) This time I attached gdb and got a backtrace: (gdb) bt #0 0x00ed6004 in internal_error () #1 0x00ed5040 in fancy_abort () #2 0x00719077 in gen_reg_rtx () #3 0x0048fc28 in gen_split_343 () #4 0x00569b62 in split_8 () #5 0x00720956 in try_split () #6 0x00a6a7be in split_insn () #7 0x00a6ec9a in split_all_insns () #8 0x00a6ecfb in (anonymous namespace)::pass_split_after_reload::execute () #9 0x00a3c567 in execute_one_pass () #10 0x00a3cb59 in execute_pass_list_1 () #11 0x00a3cb6c in execute_pass_list_1 () #12 0x00a3cb6c in execute_pass_list_1 () #13 0x00a3cbba in execute_pass_list () #14 0x0063b7d9 in cgraph_node::expand () #15 0x0063d244 in symbol_table::compile () #16 0x0063f74d in symbol_table::finalize_compilation_unit () #17 0x00b19b51 in compile_file () #18 0x0177f168 in toplev::main () #19 0x01780a24 in main () I'll have to rebuild gcc for debugging to get further info.
[Bug target/71009] g++: ICE on modified gdb/valarith.c with -Ofast
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71009 Marek Polacek changed: What|Removed |Added CC||mpolacek at gcc dot gnu.org Component|c++ |target --- Comment #2 from Marek Polacek --- Must be target-specific.
