* [Bug target/94576] Regression build newlib for Arm64
2020-04-13 9:09 [Bug target/94576] New: Regression build newlib for Arm64 trass3r at gmail dot com
@ 2020-04-13 9:11 ` trass3r at gmail dot com
2020-04-14 6:44 ` marxin at gcc dot gnu.org
` (4 subsequent siblings)
5 siblings, 0 replies; 7+ messages in thread
From: trass3r at gmail dot com @ 2020-04-13 9:11 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94576
--- Comment #1 from Trass3r <trass3r at gmail dot com> ---
Looks like it also happens with newlib-3.3:
https://github.com/Trass3r/bleeding-edge-toolchain/runs/582133302
^ permalink raw reply [flat|nested] 7+ messages in thread* [Bug target/94576] Regression build newlib for Arm64
2020-04-13 9:09 [Bug target/94576] New: Regression build newlib for Arm64 trass3r at gmail dot com
2020-04-13 9:11 ` [Bug target/94576] " trass3r at gmail dot com
@ 2020-04-14 6:44 ` marxin at gcc dot gnu.org
2020-04-14 9:44 ` clyon at gcc dot gnu.org
` (3 subsequent siblings)
5 siblings, 0 replies; 7+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-04-14 6:44 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94576
Martin Liška <marxin at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |WAITING
Last reconfirmed| |2020-04-14
CC| |marxin at gcc dot gnu.org
Ever confirmed|0 |1
--- Comment #2 from Martin Liška <marxin at gcc dot gnu.org> ---
Can you please attach pre-processed source file (-E option) and output of -v
option for the problematic object file?
^ permalink raw reply [flat|nested] 7+ messages in thread* [Bug target/94576] Regression build newlib for Arm64
2020-04-13 9:09 [Bug target/94576] New: Regression build newlib for Arm64 trass3r at gmail dot com
2020-04-13 9:11 ` [Bug target/94576] " trass3r at gmail dot com
2020-04-14 6:44 ` marxin at gcc dot gnu.org
@ 2020-04-14 9:44 ` clyon at gcc dot gnu.org
2020-04-14 11:39 ` trass3r at gmail dot com
` (2 subsequent siblings)
5 siblings, 0 replies; 7+ messages in thread
From: clyon at gcc dot gnu.org @ 2020-04-14 9:44 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94576
Christophe Lyon <clyon at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |clyon at gcc dot gnu.org
--- Comment #3 from Christophe Lyon <clyon at gcc dot gnu.org> ---
The title and target mention arm64/aarch64 but your description uses
arm-none-eabi-gcc, so it looks like the problem is when arm/aarch32 code.
Could you clarify?
There was a regression last week on aarch64 while building newlib, but it was a
different error it seems:
https://gcc.gnu.org/pipermail/gcc-patches/2020-April/543668.html
and it was fixed by:
https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=a615ea71bc8fbf31b9bc71cb373a7ca5b9cca44a
^ permalink raw reply [flat|nested] 7+ messages in thread* [Bug target/94576] Regression build newlib for Arm64
2020-04-13 9:09 [Bug target/94576] New: Regression build newlib for Arm64 trass3r at gmail dot com
` (2 preceding siblings ...)
2020-04-14 9:44 ` clyon at gcc dot gnu.org
@ 2020-04-14 11:39 ` trass3r at gmail dot com
2020-04-14 12:21 ` [Bug target/94576] Regression build newlib for Arm clyon at gcc dot gnu.org
2020-04-21 14:23 ` trass3r at gmail dot com
5 siblings, 0 replies; 7+ messages in thread
From: trass3r at gmail dot com @ 2020-04-14 11:39 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94576
--- Comment #4 from Trass3r <trass3r at gmail dot com> ---
Created attachment 48268
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=48268&action=edit
preprocessed source
It's -march=armv8.1-m.main+mve. arm-none-eabi is just the configured target
name which hasn't been adapted.
-v output:
Using built-in specs.
COLLECT_GCC=./installNative/bin/arm-none-eabi-gcc
Target: arm-none-eabi
Configured with:
/mnt/c/Code/bleeding-edge-toolchain/build/sources/gcc-master/configure
--enable-languages=c --without-headers --target=arm-none-eabi
--prefix=/mnt/c/Code/bleeding-edge-toolchain/build/installNative
--libexecdir=/mnt/c/Code/bleeding-edge-toolchain/build/installNative/lib
--disable-decimal-float --disable-libffi --disable-libgomp --disable-libmudflap
--disable-libquadmath --disable-libssp --disable-libstdcxx-pch --disable-nls
--disable-shared --disable-threads --disable-tls --with-newlib --with-gnu-as
--with-gnu-ld
--with-sysroot=/mnt/c/Code/bleeding-edge-toolchain/build/installNative/arm-none-eabi
--with-system-zlib
--with-gmp=/mnt/c/Code/bleeding-edge-toolchain/build/buildNative/prerequisites/gmp-6.2.0
--with-mpfr=/mnt/c/Code/bleeding-edge-toolchain/build/buildNative/prerequisites/mpfr-4.0.2
--with-mpc=/mnt/c/Code/bleeding-edge-toolchain/build/buildNative/prerequisites/mpc-1.1.0
--with-isl=/mnt/c/Code/bleeding-edge-toolchain/build/buildNative/prerequisites/isl-0.22.1
--with-pkgversion=heads/master-0-gdecb4a8 --with-multilib-list=rmprofile
Thread model: single
Supported LTO compression algorithms: zlib
gcc version 10.0.1 20200414 (experimental) (heads/master-0-gdecb4a8)
COLLECT_GCC_OPTIONS='-B'
'/mnt/c/Code/bleeding-edge-toolchain/build/buildNative/newlib-3.3.0-nano/arm-none-eabi/thumb/v8.1-m.main+mve/hard/newlib/'
'-isystem'
'/mnt/c/Code/bleeding-edge-toolchain/build/buildNative/newlib-3.3.0-nano/arm-none-eabi/thumb/v8.1-m.main+mve/hard/newlib/targ-include'
'-isystem'
'/mnt/c/Code/bleeding-edge-toolchain/build/sources/newlib-3.3.0/newlib/libc/include'
'-B'
'/mnt/c/Code/bleeding-edge-toolchain/build/buildNative/newlib-3.3.0-nano/arm-none-eabi/thumb/v8.1-m.main+mve/hard/libgloss/arm'
'-L/mnt/c/Code/bleeding-edge-toolchain/build/buildNative/newlib-3.3.0-nano/arm-none-eabi/thumb/v8.1-m.main+mve/hard/libgloss/libnosys'
'-L/mnt/c/Code/bleeding-edge-toolchain/build/sources/newlib-3.3.0/libgloss/arm'
'-mfloat-abi=hard' '-D' 'PACKAGE_NAME="newlib"' '-D' 'PACKAGE_TARNAME="newlib"'
'-D' 'PACKAGE_VERSION="3.3.0"' '-D' 'PACKAGE_STRING="newlib 3.3.0"' '-D'
'PACKAGE_BUGREPORT=""' '-D' 'PACKAGE_URL=""' '-I' '.' '-I'
'/mnt/c/Code/bleeding-edge-toolchain/build/sources/newlib-3.3.0/newlib/libc/stdlib'
'-fno-builtin' '-D' '__NO_SYSCALLS__' '-D' '_COMPILING_NEWLIB' '-D'
'HAVE_INIT_FINI' '-Os' '-pipe' '-ffunction-sections' '-fdata-sections'
'-mthumb' '-mfloat-abi=hard' '-c' '-o' 'lib_a-dtoa.o' '-v'
'-march=armv8.1-m.main+mve'
/mnt/c/Code/bleeding-edge-toolchain/build/installNative/lib/gcc/arm-none-eabi/10.0.1/cc1
-quiet -v -I . -I
/mnt/c/Code/bleeding-edge-toolchain/build/sources/newlib-3.3.0/newlib/libc/stdlib
-imultilib thumb/v8.1-m.main+mve/hard -D__USES_INITFINI__ -D
PACKAGE_NAME="newlib" -D PACKAGE_TARNAME="newlib" -D PACKAGE_VERSION="3.3.0" -D
PACKAGE_STRING="newlib 3.3.0" -D PACKAGE_BUGREPORT="" -D PACKAGE_URL="" -D
__NO_SYSCALLS__ -D _COMPILING_NEWLIB -D HAVE_INIT_FINI -isystem
/mnt/c/Code/bleeding-edge-toolchain/build/buildNative/newlib-3.3.0-nano/arm-none-eabi/thumb/v8.1-m.main+mve/hard/newlib/targ-include
-isystem
/mnt/c/Code/bleeding-edge-toolchain/build/sources/newlib-3.3.0/newlib/libc/include
/mnt/c/Code/bleeding-edge-toolchain/build/sources/newlib-3.3.0/newlib/libc/stdlib/dtoa.c
-quiet -dumpbase dtoa.c -mfloat-abi=hard -mthumb -mfloat-abi=hard
-march=armv8.1-m.main+mve -auxbase-strip lib_a-dtoa.o -Os -version -fno-builtin
-ffunction-sections -fdata-sections -o - |
/mnt/c/Code/bleeding-edge-toolchain/build/installNative/lib/gcc/arm-none-eabi/10.0.1/../../../../arm-none-eabi/bin/as
-v -I . -I
/mnt/c/Code/bleeding-edge-toolchain/build/sources/newlib-3.3.0/newlib/libc/stdlib
-march=armv8.1-m.main+mve -mfloat-abi=hard -mfloat-abi=hard -meabi=5 -o
lib_a-dtoa.o
GNU assembler version 2.34.50 (arm-none-eabi) using BFD version
(heads/master-0-gdecb4a8) 2.34.50.20200414
GNU C17 (heads/master-0-gdecb4a8) version 10.0.1 20200414 (experimental)
(arm-none-eabi)
compiled by GNU C version 9.2.1 20191008, GMP version 6.2.0, MPFR
version 4.0.2, MPC version 1.1.0, isl version isl-0.22.1-GMP
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096
ignoring nonexistent directory
"/mnt/c/Code/bleeding-edge-toolchain/build/installNative/arm-none-eabi/usr/local/include"
ignoring nonexistent directory
"/mnt/c/Code/bleeding-edge-toolchain/build/installNative/lib/gcc/arm-none-eabi/10.0.1/../../../../arm-none-eabi/include"
ignoring nonexistent directory
"/mnt/c/Code/bleeding-edge-toolchain/build/installNative/arm-none-eabi/usr/include"
#include "..." search starts here:
#include <...> search starts here:
.
/mnt/c/Code/bleeding-edge-toolchain/build/sources/newlib-3.3.0/newlib/libc/stdlib
/mnt/c/Code/bleeding-edge-toolchain/build/buildNative/newlib-3.3.0-nano/arm-none-eabi/thumb/v8.1-m.main+mve/hard/newlib/targ-include
/mnt/c/Code/bleeding-edge-toolchain/build/sources/newlib-3.3.0/newlib/libc/include
/mnt/c/Code/bleeding-edge-toolchain/build/installNative/lib/gcc/arm-none-eabi/10.0.1/include
/mnt/c/Code/bleeding-edge-toolchain/build/installNative/lib/gcc/arm-none-eabi/10.0.1/include-fixed
End of search list.
GNU C17 (heads/master-0-gdecb4a8) version 10.0.1 20200414 (experimental)
(arm-none-eabi)
compiled by GNU C version 9.2.1 20191008, GMP version 6.2.0, MPFR
version 4.0.2, MPC version 1.1.0, isl version isl-0.22.1-GMP
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096
Compiler executable checksum: 05c35768bf3e95c0bfa673dce95ff527
during RTL pass: reload
/mnt/c/Code/bleeding-edge-toolchain/build/sources/newlib-3.3.0/newlib/libc/stdlib/dtoa.c:
In function '_dtoa_r':
/mnt/c/Code/bleeding-edge-toolchain/build/sources/newlib-3.3.0/newlib/libc/stdlib/dtoa.c:858:1:
internal compiler error: maximum number of generated reload insns per insn
achieved (90)
^ permalink raw reply [flat|nested] 7+ messages in thread* [Bug target/94576] Regression build newlib for Arm
2020-04-13 9:09 [Bug target/94576] New: Regression build newlib for Arm64 trass3r at gmail dot com
` (3 preceding siblings ...)
2020-04-14 11:39 ` trass3r at gmail dot com
@ 2020-04-14 12:21 ` clyon at gcc dot gnu.org
2020-04-21 14:23 ` trass3r at gmail dot com
5 siblings, 0 replies; 7+ messages in thread
From: clyon at gcc dot gnu.org @ 2020-04-14 12:21 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94576
Christophe Lyon <clyon at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Target|aarch64 |arm
Summary|Regression build newlib for |Regression build newlib for
|Arm64 |Arm
--- Comment #5 from Christophe Lyon <clyon at gcc dot gnu.org> ---
(In reply to Trass3r from comment #4)
> It's -march=armv8.1-m.main+mve. arm-none-eabi is just the configured target
> name which hasn't been adapted.
>
Yes, I saw that, so it's definitely arm/aarch32 mode, *not* arm64/aarch64.
You are using armv8.1-m architecture in aarch32 mode.
^ permalink raw reply [flat|nested] 7+ messages in thread* [Bug target/94576] Regression build newlib for Arm
2020-04-13 9:09 [Bug target/94576] New: Regression build newlib for Arm64 trass3r at gmail dot com
` (4 preceding siblings ...)
2020-04-14 12:21 ` [Bug target/94576] Regression build newlib for Arm clyon at gcc dot gnu.org
@ 2020-04-21 14:23 ` trass3r at gmail dot com
5 siblings, 0 replies; 7+ messages in thread
From: trass3r at gmail dot com @ 2020-04-21 14:23 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94576
Trass3r <trass3r at gmail dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|WAITING |RESOLVED
Resolution|--- |FIXED
--- Comment #6 from Trass3r <trass3r at gmail dot com> ---
Looks like it's been fixed already?
https://github.com/Trass3r/bleeding-edge-toolchain/actions/runs/82552193
^ permalink raw reply [flat|nested] 7+ messages in thread