[Bug binutils/29189] New: dlltool delaylibs corrupt float/double arguments
https://sourceware.org/bugzilla/show_bug.cgi?id=29189 Bug ID: 29189 Summary: dlltool delaylibs corrupt float/double arguments Product: binutils Version: 2.39 (HEAD) Status: UNCONFIRMED Severity: normal Priority: P2 Component: binutils Assignee: unassigned at sourceware dot org Reporter: strager.nds at gmail dot com Target Milestone: --- Created attachment 14120 --> https://sourceware.org/bugzilla/attachment.cgi?id=14120&action=edit semi-tested bug fix (This report was originally posted on the mailing list: https://lists.gnu.org/archive/html/bug-binutils/2022-05/msg00099.html) I am calling a function in another x64 DLL with the following C signature: int napi_create_double(void*, double, void*); The first time I call this function, the 'double' argument ends up as 1.20305e-307 inside napi_create_double, no matter what value the caller gives. The 'double' is corrupted. Calls after the first don't corrupt the 'double'. The cause is ntdll.dll, eventually called by MinGW's __delayLoadHelper2, modifying the xmm1 register: #0 0x7ffd26ce3006 in ntdll!RtlLookupFunctionEntry () from C:\WINDOWS\SYSTEM32\ntdll.dll #1 0x7ffd26ce05e8 in ntdll!LdrGetProcedureAddressForCaller () from C:\WINDOWS\SYSTEM32\ntdll.dll #2 0x7ffd26ce00a5 in ntdll!LdrGetProcedureAddressForCaller () from C:\WINDOWS\SYSTEM32\ntdll.dll #3 0x7ffd245b53dc in KERNELBASE!GetProcAddressForCaller () from C:\WINDOWS\System32\KernelBase.dll #4 0x7ffcd7b7ca6f in __delayLoadHelper2 (pidd=0x7ffcd7b8ba70 <__DELAY_IMPORT_DESCRIPTOR_node_napi_lib>, ppfnIATEntry=0x7ffcd7ecd134 <__imp_napi_create_double>) at C:/M/mingw-w64-crt-git/src/mingw-w64/mingw-w64-crt/misc/delayimp.c:209 #5 0x7ffcd7b717c9 in __tailMerge_node_napi_lib () from MYDLL.dll #6 0x02ad2fe84c50 in ?? () 0x7ffd26ce2ffb <+1051>: movups (%rdx),%xmm0 0x7ffd26ce2ffe <+1054>: movups %xmm0,(%rsi) 0x7ffd26ce3001 <+1057>: movsd 0x10(%rdx),%xmm1 => 0x7ffd26ce3006 <+1062>: movsd %xmm1,0x10(%rsi) 0x7ffd26ce300b <+1067>: mov(%rsi),%rbp 0x7ffd26ce300e <+1070>: mov%r11,%rax 0x7ffd26ce3011 <+1073>: lock cmpxchg %r12,0x1384d6(%rip)# 0x7ffd26e1b4f0 0x7ffd26ce301a <+1082>: jne0x7ffd26ce3102 According to Windows x64 documentation, xmm1 is a volatile register: https://docs.microsoft.com/en-us/cpp/build/x64-software-conventions?redirectedfrom=MSDN&view=msvc-170 I think the solution is for dll's delaylib trampoline to save xmm1 on the stack before calling __delayLoadHelper2. I made a patch which does this, and it fixes the bug for my code. See attached patch. I think my patch has two problems: 1. AVX/vmovupd/ymm might not be usable on the target machine, but saving just xmm isn't enough. Should we perform a CPUID check? 2. We store unaligned with vmovupd. Storing aligned with vmovapd would be better. I haven't looked into how to align ymm registers when storing on the stack. I'd love to get this bug fixed so others don't spend two days debugging assembly code! -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29189] dlltool delaylibs corrupt float/double arguments
https://sourceware.org/bugzilla/show_bug.cgi?id=29189 --- Comment #1 from strager --- My patch saves ymm4 and ymm5, but I think that's unnecessary, since they won't be used for parameters. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29189] dlltool delaylibs corrupt float/double arguments
https://sourceware.org/bugzilla/show_bug.cgi?id=29189 --- Comment #2 from strager --- Created attachment 14121 --> https://sourceware.org/bugzilla/attachment.cgi?id=14121&action=edit minimal repro Attached is a small program (DLL and EXE) which demonstrates the issue. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29189] dlltool delaylibs corrupt float/double arguments
https://sourceware.org/bugzilla/show_bug.cgi?id=29189 strager changed: What|Removed |Added Target||windows-amd64 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29189] dlltool delaylibs corrupt float/double arguments
https://sourceware.org/bugzilla/show_bug.cgi?id=29189 Evgeniy Dushistov changed: What|Removed |Added CC||dushistov at mail dot ru -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29189] dlltool delaylibs corrupt float/double arguments
https://sourceware.org/bugzilla/show_bug.cgi?id=29189 Egor Pugin changed: What|Removed |Added CC||egor.pugin at gmail dot com -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29191] New: gprofng ignores --sysconfdir configuration option
https://sourceware.org/bugzilla/show_bug.cgi?id=29191 Bug ID: 29191 Summary: gprofng ignores --sysconfdir configuration option Product: binutils Version: 2.39 (HEAD) Status: NEW Severity: normal Priority: P2 Component: gprofng Assignee: vladimir.mezentsev at oracle dot com Reporter: doko at debian dot org Target Milestone: --- gprofng.rc is installed into dbe_DATA = $(srcdir)/gprofng.rc while it probably should be installed into the sysconfdir, which can be specified by the --sysconfdir option. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29191] gprofng ignores --sysconfdir configuration option
https://sourceware.org/bugzilla/show_bug.cgi?id=29191 --- Comment #1 from Matthias Klose --- i.e. it ends up in /usr/etc, but should be /etc when configured with --sysconfdir=/etc -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29192] New: addr2line suspected incompatibility with DWARF5
https://sourceware.org/bugzilla/show_bug.cgi?id=29192
Bug ID: 29192
Summary: addr2line suspected incompatibility with DWARF5
Product: binutils
Version: 2.38
Status: UNCONFIRMED
Severity: normal
Priority: P2
Component: binutils
Assignee: unassigned at sourceware dot org
Reporter: ofekshilon at gmail dot com
Target Milestone: ---
==Example:
$ cat toy.cpp
#include
int main() {
std::cout << "hello world";
return 0;
}
$ clang-14 -g -stdlib=libc++ -lc++ -I/usr/lib/llvm-14/include/c++/v1 toy.cpp -o
toy14
$ addr2line -fe ./toy 0x1234
addr2line: DWARF error: invalid or unhandled FORM value: 0x25
main
??:?
=end example
According to the dwarf5 spec (https://dwarfstd.org/doc/DWARF5.pdf), section
7.5.6 page 221, 0x25 is a valid DW_FORM value:
DW_FORM_strx1 ‡ 0x25 string.
In other cases (not yet reduced to a small repro) we see:
addr2line: Dwarf Error: Could not find abbrev number 127
--
You are receiving this mail because:
You are on the CC list for the bug.
[Bug binutils/29192] addr2line suspected incompatibility with DWARF5
https://sourceware.org/bugzilla/show_bug.cgi?id=29192 --- Comment #1 from Ofek Shilon --- (1) Sorry, typo in the example - instead of $ addr2line -fe ./toy 0x1234 should be - $ addr2line -fe ./toy14 0x1234 Also, just verified that if I build with `-fdebug-default-version=4` the error disappears and lines are displayed properly. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29143] gprofng fails to build on x86_64-linux-gnux32
https://sourceware.org/bugzilla/show_bug.cgi?id=29143
--- Comment #3 from Matthias Klose ---
trunk 20220527 still fails to build, this time with:
../../../gprofng/src/Experiment.cc: In member function ‘void
Experiment::write_header()’:
../../../gprofng/src/Experiment.cc:4139:33: error: cannot convert ‘long int*’
to ‘const time_t*’
{aka ‘const long long int*’}
--
You are receiving this mail because:
You are on the CC list for the bug.
[Bug gprofng/28984] Bad __collector_getpc implementation
https://sourceware.org/bugzilla/show_bug.cgi?id=28984 Vladimir Mezentsev changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #1 from Vladimir Mezentsev --- I fixed this along with fixes for 28983 gprofng: Don't build with -mx32 The master branch has been updated by Vladimir Mezentsev : https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=2a2cb7cf2c7f0d380ee9cb9ef361223b8770c508 commit 2a2cb7cf2c7f0d380ee9cb9ef361223b8770c508 Author: Vladimir Mezentsev Date: Thu May 26 12:59:19 2022 -0700 gprofng: fix build with -mx32 gprofng/ChangeLog 2022-05-26 Vladimir Mezentsev PR gprofng/28983 * libcollector/libcol_util.h (__collector_getsp, __collector_getfp, __collector_getpc): Adapt for build with -mx32 * libcollector/heaptrace.c: Fix -Wpointer-to-int-cast warnings. * libcollector/hwprofile.h: Likewise. * libcollector/mmaptrace.c: Likewise. * libcollector/synctrace.c: Likewise. * libcollector/unwind.c: Likewise. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29194] New: bfd/doc/chew.c error: ISO C forbids conversion of function pointer to object pointer type
https://sourceware.org/bugzilla/show_bug.cgi?id=29194 Bug ID: 29194 Summary: bfd/doc/chew.c error: ISO C forbids conversion of function pointer to object pointer type Product: binutils Version: 2.38 Status: UNCONFIRMED Severity: normal Priority: P2 Component: binutils Assignee: unassigned at sourceware dot org Reporter: dc at genunix dot com Target Milestone: --- Not entirely sure if this is a bug or perhaps the binutils sources are not C99 clean. Also not sure if this is a Big Flipping Deal[1] or maybe I can give up on C99? Regardless I have CFLAGS set for strict compliance : mimas$ echo $CC /opt/gcc/imed/gcc12/bin/gcc mimas$ echo $CFLAGS -std=iso9899:1999 -pedantic -pedantic-errors -fno-builtin -g -O0 -m64 -march=k8 -mtune=k8 -malign-double -mpc80 -mno-mmx -Wl,-rpath=/opt/bw/lib,--enable-new-dtags mimas$ echo $CPPFLAGS -I/opt/bw/include -D_POSIX_PTHREAD_SEMANTICS -D_LARGEFILE64_SOURCE mimas$ This is on an old Red Hat Enterprise Linux 6 machine wherein I did a bootstrap of GCC 9.4.0 and then used that result to bootstrap GCC 12.1.0 with great success. So I figure the compiler is reasonable. Configure was trivial : mimas$ ../binutils-2.38/configure --prefix=/opt/bw --enable-ld --with-gmp=/opt/bw --with-mpfr=/opt/bw --with-mpc=/opt/bw --with-isl=/opt/bw checking build system type... x86_64-pc-linux-gnu checking host system type... x86_64-pc-linux-gnu checking target system type... x86_64-pc-linux-gnu checking for a BSD-compatible install... /usr/bin/install -c checking whether ln works... yes checking whether ln -s works... yes checking for a sed that does not truncate output... /bin/sed checking for gawk... gawk checking for gcc... /opt/gcc/imed/gcc12/bin/gcc checking whether the C compiler works... yes checking for C compiler default output file name... a.out checking for suffix of executables... checking whether we are cross compiling... no checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether /opt/gcc/imed/gcc12/bin/gcc accepts -g... yes checking for /opt/gcc/imed/gcc12/bin/gcc option to accept ISO C89... none needed checking for /opt/gcc/imed/gcc12/bin/gcc option to accept ISO C99... none needed checking whether we are using the GNU C++ compiler... yes checking whether /opt/gcc/imed/gcc12/bin/g++ accepts -g... yes checking whether g++ accepts -static-libstdc++ -static-libgcc... yes checking for gnatbind... no checking for gnatmake... no checking whether compiler driver understands Ada... no checking how to compare bootstrapped objects... cmp --ignore-initial=16 $$f1 $$f2 checking for objdir... .libs checking for isl 0.15 or later... yes checking for default BUILD_CONFIG... checking for --enable-vtable-verify... no checking for bison... bison -y checking for bison... bison checking for gm4... /opt/bw/bin/m4 checking for flex... flex checking for flex... flex checking for makeinfo... makeinfo checking for expect... no checking for runtest... no checking for ar... (cached) /opt/bw/bin/gar checking for as... (cached) /opt/bw/bin/gas checking for dlltool... no checking for ld... (cached) /opt/bw/bin/gld checking for lipo... no checking for nm... (cached) /opt/bw/bin/gnm checking for ranlib... (cached) /opt/bw/bin/granlib checking for strip... (cached) /opt/bw/bin/gstrip checking for windres... no checking for windmc... no checking for objcopy... (cached) /opt/bw/bin/gobjcopy checking for objdump... (cached) /opt/bw/bin/gobjdump checking for readelf... (cached) /opt/bw/bin/greadelf checking for -plugin option... --plugin /opt/gcc/imed/gcc12/libexec/gcc/x86_64-pc-linux-gnu/12.1.0/liblto_plugin.so checking for cc... cc checking for c++... c++ checking for gcc... gcc checking for gfortran... gfortran checking for gccgo... no checking for ar... ar checking for as... as checking for dlltool... no checking for ld... ld checking for lipo... no checking for nm... nm checking for objcopy... objcopy checking for objdump... objdump checking for ranlib... ranlib checking for readelf... readelf checking for strip... strip checking for windres... no checking for windmc... no checking where to find the target ar... just compiled checking where to find the target as... just compiled checking where to find the target cc... host tool checking where to find the target c++... host tool checking where to find the target c++ for libstdc++... host tool checking where to find the target dlltool... just compiled checking where to find the target gcc... host tool checking where to find the target gfortran... host tool checking where to find the target gccgo... host tool checking where to find the target ld... just compiled checking where to find the target lipo... host tool checking where to find the target nm... just compiled checking where to find the target objcopy... just compiled checking where to find the target objdump... j
[Bug binutils/29194] bfd/doc/chew.c error: ISO C forbids conversion of function pointer to object pointer type
https://sourceware.org/bugzilla/show_bug.cgi?id=29194 --- Comment #1 from Dennis Clarke --- As a further comment where I did also try C90 but that gets me only a little further : . . . /opt/gcc/imed/gcc12/bin/gcc -c -DHAVE_CONFIG_H -std=iso9899:1990 -pedantic -pedantic-errors -fno-builtin -g -O0 -m64 -march=k8 -mtune=k8 -malign-double -mpc80 -mno-mmx -Wl,-rpath=/opt/bw/lib,--enable-new-dtags -I/opt/bw/include -D_POSIX_PTHREAD_SEMANTICS -D_LARGEFILE64_SOURCE -I. -I../../binutils-2.38/libiberty/../include -W -Wall -Wwrite-strings -Wc++-compat -Wstrict-prototypes -Wshadow=local -pedantic -D_GNU_SOURCE -fcf-protection ../../binutils-2.38/libiberty/cp-demangle.c -o cp-demangle.o ../../binutils-2.38/libiberty/cp-demangle.c: In function 'is_designated_init': ../../binutils-2.38/libiberty/cp-demangle.c:4763:3: error: ISO C90 forbids mixed declarations and code [-Wdeclaration-after-statement] 4763 | struct demangle_component *op = d_left (dc); | ^~ ../../binutils-2.38/libiberty/cp-demangle.c: In function 'd_maybe_print_designated_init': ../../binutils-2.38/libiberty/cp-demangle.c:4779:3: error: ISO C90 forbids mixed declarations and code [-Wdeclaration-after-statement] 4779 | const char *code = d_left (dc)->u.s_operator.op->code; | ^ make[2]: *** [cp-demangle.o] Error 1 make[2]: Leaving directory `/opt/bw/build/binutils-2.38_rhel6_amd64.003/libiberty' make[1]: *** [all-libiberty] Error 2 make[1]: Leaving directory `/opt/bw/build/binutils-2.38_rhel6_amd64.003' make: *** [all] Error 2 So I am not sure what C standard the binutils comply with. If at all. Dennis -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/28983] gprofng: Don't build with -mx32
https://sourceware.org/bugzilla/show_bug.cgi?id=28983 --- Comment #2 from cvs-commit at gcc dot gnu.org --- The master branch has been updated by Vladimir Mezentsev : https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=6094a48ec821a52731172f3471dce4923a482cd1 commit 6094a48ec821a52731172f3471dce4923a482cd1 Author: Vladimir Mezentsev Date: Fri May 27 19:54:26 2022 -0700 gprofng: fix build with -mx32 gprofng/ChangeLog 2022-05-27 Vladimir Mezentsev PR gprofng/28983 PR gprofng/29143 * src/Experiment.cc (write_header): Fix argument for ctime. Fix -Wformat= warnings. * src/Dbe.cc: Likewise. * src/DwarfLib.h: Fix [-Wsign-compare] warnings. * src/Experiment.h: Likewise. * src/ipc.cc: Fix -Wformat= warnings. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug gprofng/29143] gprofng fails to build on x86_64-linux-gnux32
https://sourceware.org/bugzilla/show_bug.cgi?id=29143 --- Comment #4 from cvs-commit at gcc dot gnu.org --- The master branch has been updated by Vladimir Mezentsev : https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=6094a48ec821a52731172f3471dce4923a482cd1 commit 6094a48ec821a52731172f3471dce4923a482cd1 Author: Vladimir Mezentsev Date: Fri May 27 19:54:26 2022 -0700 gprofng: fix build with -mx32 gprofng/ChangeLog 2022-05-27 Vladimir Mezentsev PR gprofng/28983 PR gprofng/29143 * src/Experiment.cc (write_header): Fix argument for ctime. Fix -Wformat= warnings. * src/Dbe.cc: Likewise. * src/DwarfLib.h: Fix [-Wsign-compare] warnings. * src/Experiment.h: Likewise. * src/ipc.cc: Fix -Wformat= warnings. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug binutils/29194] bfd/doc/chew.c error: ISO C forbids conversion of function pointer to object pointer type
https://sourceware.org/bugzilla/show_bug.cgi?id=29194 Alan Modra changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|--- |WONTFIX --- Comment #2 from Alan Modra --- If -Wpedantic warnings in part of the binutils that is used by developers (and others building from git source) to regenerate sources is of concern to you, I suggest posting patches to fix those warnings to binut...@sourceware.org. -- You are receiving this mail because: You are on the CC list for the bug.
