From: Adhemerval Zanella Netto <adhemerval.zanella@linaro.org>
To: caiyinyu <caiyinyu@loongson.cn>,
libc-alpha@sourceware.org, joseph_myers@mentor.com,
carlos@redhat.com, i.swmail@xen0n.name, xry111@xry111.site
Cc: xuchenghua@loongson.cn
Subject: Re: [PATCH 1/1] LoongArch: Add soft float support.
Date: Wed, 31 Aug 2022 09:28:29 -0300 [thread overview]
Message-ID: <8e153100-f89e-5934-921a-c48d219bc388@linaro.org> (raw)
In-Reply-To: <f019f6ca-e2ed-3b63-ebe5-797c3482157f@loongson.cn>
On 31/08/22 09:00, caiyinyu wrote:
> diff --git a/sysdeps/loongarch/Makefile b/sysdeps/loongarch/Makefile
> index 41c3449670..746d4c2c8f 100644
> --- a/sysdeps/loongarch/Makefile
> +++ b/sysdeps/loongarch/Makefile
> @@ -5,11 +5,3 @@ endif
> # LoongArch's assembler also needs to know about PIC as it changes the
> # definition of some assembler macros.
> ASFLAGS-.os += $(pic-ccflag)
> -
> -abi-variants := lp64
> -
> -ifeq (,$(filter $(default-abi),$(abi-variants)))
> -$(error Unknown ABI $(default-abi), must be one of $(abi-variants))
> -endif
> -
> -abi-lp64-condition := defined __loongarch_lp64
>
> Why move this to be Linux specific? It won't matter in the end though.
>
> The rest looks ok.
>
>
> I did this to follow the example of other architectures (like riscv arc
>
> aarch64 arm mips...) though there seems to be no difference.
Fair enough, the patch looks ok as it.
PS: Usually we reply with bottom-posting.
>
> I need help to include this patch since there are some changes in common
>
> file build-many-glibcs.py.
>
>
>
> 在 2022/8/30 下午9:00, Adhemerval Zanella Netto 写道:
>>
>> On 30/08/22 08:02, caiyinyu wrote:
>>> ---
>>> scripts/build-many-glibcs.py | 4 ++
>>> sysdeps/loongarch/Makefile | 8 ---
>>> sysdeps/loongarch/__longjmp.S | 2 +
>>> sysdeps/loongarch/bits/setjmp.h | 3 ++
>>> sysdeps/loongarch/dl-trampoline.S | 11 +++-
>>> sysdeps/loongarch/fpu_control.h | 13 +++++
>>> sysdeps/loongarch/nofpu/Implies | 1 +
>>> .../loongarch/nofpu/math-tests-exceptions.h | 28 ++++++++++
>>> sysdeps/loongarch/nofpu/math-tests-rounding.h | 27 ++++++++++
>>> sysdeps/loongarch/preconfigure | 1 -
>>> sysdeps/loongarch/preconfigure.ac | 1 -
>>> sysdeps/loongarch/setjmp.S | 2 +
>>> sysdeps/unix/sysv/linux/loongarch/Makefile | 9 ++++
>>> sysdeps/unix/sysv/linux/loongarch/configure | 51 +++++++++++++++++--
>>> .../unix/sysv/linux/loongarch/configure.ac | 22 ++++++--
>>> .../unix/sysv/linux/loongarch/ldd-rewrite.sed | 2 +-
>>> .../unix/sysv/linux/loongarch/shlib-versions | 2 +
>>> 17 files changed, 168 insertions(+), 19 deletions(-)
>>> create mode 100644 sysdeps/loongarch/nofpu/Implies
>>> create mode 100644 sysdeps/loongarch/nofpu/math-tests-exceptions.h
>>> create mode 100644 sysdeps/loongarch/nofpu/math-tests-rounding.h
>>>
>>> diff --git a/scripts/build-many-glibcs.py b/scripts/build-many-glibcs.py
>>> index 98f02104a2..db632b18a3 100755
>>> --- a/scripts/build-many-glibcs.py
>>> +++ b/scripts/build-many-glibcs.py
>>> @@ -222,6 +222,10 @@ class Context(object):
>>> os_name='linux-gnu',
>>> variant='lp64d',
>>> gcc_cfg=['--with-abi=lp64d','--disable-multilib'])
>>> + self.add_config(arch='loongarch64',
>>> + os_name='linux-gnu',
>>> + variant='lp64s',
>>> + gcc_cfg=['--with-abi=lp64s','--disable-multilib'])
>>> self.add_config(arch='m68k',
>>> os_name='linux-gnu',
>>> gcc_cfg=['--disable-multilib'])
>>> diff --git a/sysdeps/loongarch/Makefile b/sysdeps/loongarch/Makefile
>>> index 41c3449670..746d4c2c8f 100644
>>> --- a/sysdeps/loongarch/Makefile
>>> +++ b/sysdeps/loongarch/Makefile
>>> @@ -5,11 +5,3 @@ endif
>>> # LoongArch's assembler also needs to know about PIC as it changes the
>>> # definition of some assembler macros.
>>> ASFLAGS-.os += $(pic-ccflag)
>>> -
>>> -abi-variants := lp64
>>> -
>>> -ifeq (,$(filter $(default-abi),$(abi-variants)))
>>> -$(error Unknown ABI $(default-abi), must be one of $(abi-variants))
>>> -endif
>>> -
>>> -abi-lp64-condition := defined __loongarch_lp64
>> Why move this to be Linux specific? It won't matter in the end though.
>>
>> The rest looks ok.
>>
>>> diff --git a/sysdeps/loongarch/__longjmp.S b/sysdeps/loongarch/__longjmp.S
>>> index 4207376f5e..d6a99fcbc8 100644
>>> --- a/sysdeps/loongarch/__longjmp.S
>>> +++ b/sysdeps/loongarch/__longjmp.S
>>> @@ -41,6 +41,7 @@ ENTRY (__longjmp)
>>> REG_L s7, a0, 11*SZREG
>>> REG_L s8, a0, 12*SZREG
>>> +#ifndef __loongarch_soft_float
>>> FREG_L $f24, a0, 13*SZREG + 0*SZFREG
>>> FREG_L $f25, a0, 13*SZREG + 1*SZFREG
>>> FREG_L $f26, a0, 13*SZREG + 2*SZFREG
>>> @@ -49,6 +50,7 @@ ENTRY (__longjmp)
>>> FREG_L $f29, a0, 13*SZREG + 5*SZFREG
>>> FREG_L $f30, a0, 13*SZREG + 6*SZFREG
>>> FREG_L $f31, a0, 13*SZREG + 7*SZFREG
>>> +#endif
>>> sltui a0,a1,1
>>> ADD a0, a0, a1 # a0 = (a1 == 0) ? 1 : a1
>>> diff --git a/sysdeps/loongarch/bits/setjmp.h b/sysdeps/loongarch/bits/setjmp.h
>>> index 42f8fa7657..8b323ad2b6 100644
>>> --- a/sysdeps/loongarch/bits/setjmp.h
>>> +++ b/sysdeps/loongarch/bits/setjmp.h
>>> @@ -31,8 +31,11 @@ typedef struct __jmp_buf_internal_tag
>>> long int __fp;
>>> /* Callee-saved registers. */
>>> long int __regs[9];
>>> +
>>> +#ifndef __loongarch_soft_float
>>> /* Callee-saved floating point registers. */
>>> double __fpregs[8];
>>> +#endif
>>> } __jmp_buf[1];
>>> diff --git a/sysdeps/loongarch/dl-trampoline.S b/sysdeps/loongarch/dl-trampoline.S
>>> index ad8ab0fda7..9a8f580971 100644
>>> --- a/sysdeps/loongarch/dl-trampoline.S
>>> +++ b/sysdeps/loongarch/dl-trampoline.S
>>> @@ -21,8 +21,11 @@
>>> /* Assembler veneer called from the PLT header code for lazy loading.
>>> The PLT header passes its own args in t0-t2. */
>>> -
>>> -# define FRAME_SIZE (-((-10 * SZREG - 8 * SZFREG) & ALMASK))
>>> +#ifdef __loongarch_soft_float
>>> +#define FRAME_SIZE (-((-10 * SZREG) & ALMASK))
>>> +#else
>>> +#define FRAME_SIZE (-((-10 * SZREG - 8 * SZFREG) & ALMASK))
>>> +#endif
>>> ENTRY (_dl_runtime_resolve)
>>> @@ -39,6 +42,7 @@ ENTRY (_dl_runtime_resolve)
>>> REG_S a6, sp, 7*SZREG
>>> REG_S a7, sp, 8*SZREG
>>> +#ifndef __loongarch_soft_float
>>> FREG_S fa0, sp, 10*SZREG + 0*SZFREG
>>> FREG_S fa1, sp, 10*SZREG + 1*SZFREG
>>> FREG_S fa2, sp, 10*SZREG + 2*SZFREG
>>> @@ -47,6 +51,7 @@ ENTRY (_dl_runtime_resolve)
>>> FREG_S fa5, sp, 10*SZREG + 5*SZFREG
>>> FREG_S fa6, sp, 10*SZREG + 6*SZFREG
>>> FREG_S fa7, sp, 10*SZREG + 7*SZFREG
>>> +#endif
>>> /* Update .got.plt and obtain runtime address of callee */
>>> SLLI a1, t1, 1
>>> @@ -67,6 +72,7 @@ ENTRY (_dl_runtime_resolve)
>>> REG_L a6, sp, 7*SZREG
>>> REG_L a7, sp, 8*SZREG
>>> +#ifndef __loongarch_soft_float
>>> FREG_L fa0, sp, 10*SZREG + 0*SZFREG
>>> FREG_L fa1, sp, 10*SZREG + 1*SZFREG
>>> FREG_L fa2, sp, 10*SZREG + 2*SZFREG
>>> @@ -75,6 +81,7 @@ ENTRY (_dl_runtime_resolve)
>>> FREG_L fa5, sp, 10*SZREG + 5*SZFREG
>>> FREG_L fa6, sp, 10*SZREG + 6*SZFREG
>>> FREG_L fa7, sp, 10*SZREG + 7*SZFREG
>>> +#endif
>>> ADDI sp, sp, FRAME_SIZE
>>> diff --git a/sysdeps/loongarch/fpu_control.h b/sysdeps/loongarch/fpu_control.h
>>> index 1cccc933a5..f482395dba 100644
>>> --- a/sysdeps/loongarch/fpu_control.h
>>> +++ b/sysdeps/loongarch/fpu_control.h
>>> @@ -51,6 +51,17 @@
>>> #include <features.h>
>>> +#ifdef __loongarch_soft_float
>>> +
>>> +#define _FPU_RESERVED 0xffffffff
>>> +#define _FPU_DEFAULT 0x00000000
>>> +typedef unsigned int fpu_control_t;
>>> +#define _FPU_GETCW(cw) (cw) = 0
>>> +#define _FPU_SETCW(cw) (void) (cw)
>>> +extern fpu_control_t __fpu_control;
>>> +
>>> +#else /* __loongarch_soft_float */
>>> +
>>> /* Masks for interrupts. */
>>> #define _FPU_MASK_V 0x10 /* Invalid operation */
>>> #define _FPU_MASK_Z 0x08 /* Division by zero */
>>> @@ -86,4 +97,6 @@ extern void __loongarch_fpu_setcw (fpu_control_t) __THROW;
>>> /* Default control word set at startup. */
>>> extern fpu_control_t __fpu_control;
>>> +#endif /* __loongarch_soft_float */
>>> +
>>> #endif /* fpu_control.h */
>>> diff --git a/sysdeps/loongarch/nofpu/Implies b/sysdeps/loongarch/nofpu/Implies
>>> new file mode 100644
>>> index 0000000000..abcbadb25f
>>> --- /dev/null
>>> +++ b/sysdeps/loongarch/nofpu/Implies
>>> @@ -0,0 +1 @@
>>> +ieee754/soft-fp
>>> diff --git a/sysdeps/loongarch/nofpu/math-tests-exceptions.h b/sysdeps/loongarch/nofpu/math-tests-exceptions.h
>>> new file mode 100644
>>> index 0000000000..b22bb01d0e
>>> --- /dev/null
>>> +++ b/sysdeps/loongarch/nofpu/math-tests-exceptions.h
>>> @@ -0,0 +1,28 @@
>>> +/* Configuration for math tests: support for exceptions.
>>> + Copyright (C) 2022 Free Software Foundation, Inc.
>>> + This file is part of the GNU C Library.
>>> +
>>> + The GNU C Library is free software; you can redistribute it and/or
>>> + modify it under the terms of the GNU Lesser General Public
>>> + License as published by the Free Software Foundation; either
>>> + version 2.1 of the License, or (at your option) any later version.
>>> +
>>> + The GNU C Library is distributed in the hope that it will be useful,
>>> + but WITHOUT ANY WARRANTY; without even the implied warranty of
>>> + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
>>> + Lesser General Public License for more details.
>>> +
>>> + You should have received a copy of the GNU Lesser General Public
>>> + License along with the GNU C Library; if not, see
>>> + <https://www.gnu.org/licenses/>. */
>>> +
>>> +#ifndef LOONGARCH_NOFPU_MATH_TESTS_EXCEPTIONS_H
>>> +#define LOONGARCH_NOFPU_MATH_TESTS_EXCEPTIONS_H 1
>>> +
>>> +/* We support setting floating-point exception flags on hard-float
>>> + targets. These are not supported on soft-float targets. */
>>> +#define EXCEPTION_TESTS_float 0
>>> +#define EXCEPTION_TESTS_double 0
>>> +#define EXCEPTION_TESTS_long_double 0
>>> +
>>> +#endif /* math-tests-exceptions.h. */
>>> diff --git a/sysdeps/loongarch/nofpu/math-tests-rounding.h b/sysdeps/loongarch/nofpu/math-tests-rounding.h
>>> new file mode 100644
>>> index 0000000000..5322e481b9
>>> --- /dev/null
>>> +++ b/sysdeps/loongarch/nofpu/math-tests-rounding.h
>>> @@ -0,0 +1,27 @@
>>> +/* Configuration for math tests: rounding mode support.
>>> + Copyright (C) 2022 Free Software Foundation, Inc.
>>> + This file is part of the GNU C Library.
>>> +
>>> + The GNU C Library is free software; you can redistribute it and/or
>>> + modify it under the terms of the GNU Lesser General Public
>>> + License as published by the Free Software Foundation; either
>>> + version 2.1 of the License, or (at your option) any later version.
>>> +
>>> + The GNU C Library is distributed in the hope that it will be useful,
>>> + but WITHOUT ANY WARRANTY; without even the implied warranty of
>>> + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
>>> + Lesser General Public License for more details.
>>> +
>>> + You should have received a copy of the GNU Lesser General Public
>>> + License along with the GNU C Library; if not, see
>>> + <https://www.gnu.org/licenses/>. */
>>> +
>>> +#ifndef LOONGARCH_NOFPU_MATH_TESTS_ROUNDING_H
>>> +#define LOONGARCH_NOFPU_MATH_TESTS_ROUNDING_H 1
>>> +
>>> +/* On soft-float targets we only support the "to nearest" rounding mode. */
>>> +#define ROUNDING_TESTS_float(MODE) ((MODE) == FE_TONEAREST)
>>> +#define ROUNDING_TESTS_double(MODE) ((MODE) == FE_TONEAREST)
>>> +#define ROUNDING_TESTS_long_double(MODE) ((MODE) == FE_TONEAREST)
>>> +
>>> +#endif /* math-tests-rounding.h. */
>>> diff --git a/sysdeps/loongarch/preconfigure b/sysdeps/loongarch/preconfigure
>>> index 118963cda6..f2d1a0d87b 100644
>>> --- a/sysdeps/loongarch/preconfigure
>>> +++ b/sysdeps/loongarch/preconfigure
>>> @@ -12,7 +12,6 @@ loongarch*)
>>> case "$float_abi" in
>>> soft)
>>> abi_flen=0
>>> - as_fn_error 1 "loongarch does not yet support soft floating-point ABI!!" "$LINENO" 5
>>> ;;
>>> single)
>>> as_fn_error 1 "glibc does not yet support the single floating-point ABI!!" "$LINENO" 5
>>> diff --git a/sysdeps/loongarch/preconfigure.ac b/sysdeps/loongarch/preconfigure.ac
>>> index 1aba743c15..67e4357013 100644
>>> --- a/sysdeps/loongarch/preconfigure.ac
>>> +++ b/sysdeps/loongarch/preconfigure.ac
>>> @@ -12,7 +12,6 @@ loongarch*)
>>> case "$float_abi" in
>>> soft)
>>> abi_flen=0
>>> - AC_MSG_ERROR([loongarch does not yet support soft floating-point ABI!!], 1)
>>> ;;
>>> single)
>>> AC_MSG_ERROR([glibc does not yet support the single floating-point ABI!!], 1)
>>> diff --git a/sysdeps/loongarch/setjmp.S b/sysdeps/loongarch/setjmp.S
>>> index 298bb02a82..9b1cdea48c 100644
>>> --- a/sysdeps/loongarch/setjmp.S
>>> +++ b/sysdeps/loongarch/setjmp.S
>>> @@ -50,6 +50,7 @@ ENTRY (__sigsetjmp)
>>> REG_S s7, a0, 11*SZREG
>>> REG_S s8, a0, 12*SZREG
>>> +#ifndef __loongarch_soft_float
>>> FREG_S $f24, a0, 13*SZREG + 0*SZFREG
>>> FREG_S $f25, a0, 13*SZREG + 1*SZFREG
>>> FREG_S $f26, a0, 13*SZREG + 2*SZFREG
>>> @@ -58,6 +59,7 @@ ENTRY (__sigsetjmp)
>>> FREG_S $f29, a0, 13*SZREG + 5*SZFREG
>>> FREG_S $f30, a0, 13*SZREG + 6*SZFREG
>>> FREG_S $f31, a0, 13*SZREG + 7*SZFREG
>>> +#endif
>>> #if !IS_IN (libc) && IS_IN(rtld)
>>> li.w v0, 0
>>> diff --git a/sysdeps/unix/sysv/linux/loongarch/Makefile b/sysdeps/unix/sysv/linux/loongarch/Makefile
>>> index 91bd35800a..c84a1762ed 100644
>>> --- a/sysdeps/unix/sysv/linux/loongarch/Makefile
>>> +++ b/sysdeps/unix/sysv/linux/loongarch/Makefile
>>> @@ -1,3 +1,12 @@
>>> ifeq ($(subdir),stdlib)
>>> gen-as-const-headers += ucontext_i.sym
>>> endif
>>> +
>>> +abi-variants := lp64s lp64d
>>> +
>>> +ifeq (,$(filter $(default-abi),$(abi-variants)))
>>> +$(error Unknown ABI $(default-abi), must be one of $(abi-variants))
>>> +endif
>>> +
>>> +abi-lp64s-condition := __WORDSIZE == 64 && defined __loongarch_soft_float
>>> +abi-lp64d-condition := __WORDSIZE == 64 && defined __loongarch_double_float
>>> diff --git a/sysdeps/unix/sysv/linux/loongarch/configure b/sysdeps/unix/sysv/linux/loongarch/configure
>>> index 60a410303e..28216c165e 100644
>>> --- a/sysdeps/unix/sysv/linux/loongarch/configure
>>> +++ b/sysdeps/unix/sysv/linux/loongarch/configure
>>> @@ -151,11 +151,56 @@ if test $libc_cv_loongarch_int_abi = no; then
>>> as_fn_error $? "Unable to determine integer ABI" "$LINENO" 5
>>> fi
>>> +libc_cv_loongarch_float_abi=no
>>> +cat confdefs.h - <<_ACEOF >conftest.$ac_ext
>>> +/* end confdefs.h. */
>>> +#ifdef __loongarch_double_float
>>> + yes
>>> + #endif
>>> +
>>> +_ACEOF
>>> +if (eval "$ac_cpp conftest.$ac_ext") 2>&5 |
>>> + $EGREP "yes" >/dev/null 2>&1; then :
>>> + libc_cv_loongarch_float_abi=d
>>> +fi
>>> +rm -f conftest*
>>> +
>>> +cat confdefs.h - <<_ACEOF >conftest.$ac_ext
>>> +/* end confdefs.h. */
>>> +#ifdef __loongarch_soft_float
>>> + yes
>>> + #endif
>>> +
>>> +_ACEOF
>>> +if (eval "$ac_cpp conftest.$ac_ext") 2>&5 |
>>> + $EGREP "yes" >/dev/null 2>&1; then :
>>> + libc_cv_loongarch_float_abi=s
>>> +fi
>>> +rm -f conftest*
>>> +
>>> +if test "$libc_cv_loongarch_float_abi" = no; then
>>> + as_fn_error $? "Unable to determine floating-point ABI" "$LINENO" 5
>>> +fi
>>> +
>>> config_vars="$config_vars
>>> -default-abi = $libc_cv_loongarch_int_abi"
>>> +default-abi = $libc_cv_loongarch_int_abi$libc_cv_loongarch_float_abi"
>>> -case $libc_cv_loongarch_int_abi in
>>> -lp64)
>>> +case $libc_cv_loongarch_int_abi$libc_cv_loongarch_float_abi in
>>> +lp64s)
>>> + test -n "$libc_cv_slibdir" ||
>>> +case "$prefix" in
>>> +/usr | /usr/)
>>> + libc_cv_slibdir='/lib64/sf'
>>> + libc_cv_rtlddir='/lib64'
>>> + if test "$libdir" = '${exec_prefix}/lib'; then
>>> + libdir='${exec_prefix}/lib64/sf';
>>> + # Locale data can be shared between 32-bit and 64-bit libraries.
>>> + libc_cv_complocaledir='${exec_prefix}/lib/locale'
>>> + fi
>>> + ;;
>>> +esac
>>> + ;;
>>> +lp64d)
>>> test -n "$libc_cv_slibdir" ||
>>> case "$prefix" in
>>> /usr | /usr/)
>>> diff --git a/sysdeps/unix/sysv/linux/loongarch/configure.ac b/sysdeps/unix/sysv/linux/loongarch/configure.ac
>>> index 7de1e95ff6..04e9150a9b 100644
>>> --- a/sysdeps/unix/sysv/linux/loongarch/configure.ac
>>> +++ b/sysdeps/unix/sysv/linux/loongarch/configure.ac
>>> @@ -11,10 +11,26 @@ if test $libc_cv_loongarch_int_abi = no; then
>>> AC_MSG_ERROR([Unable to determine integer ABI])
>>> fi
>>> -LIBC_CONFIG_VAR([default-abi], [$libc_cv_loongarch_int_abi])
>>> +libc_cv_loongarch_float_abi=no
>>> +AC_EGREP_CPP(yes, [#ifdef __loongarch_double_float
>>> + yes
>>> + #endif
>>> + ],libc_cv_loongarch_float_abi=d)
>>> +AC_EGREP_CPP(yes, [#ifdef __loongarch_soft_float
>>> + yes
>>> + #endif
>>> + ],libc_cv_loongarch_float_abi=s)
>>> +if test "$libc_cv_loongarch_float_abi" = no; then
>>> + AC_MSG_ERROR([Unable to determine floating-point ABI])
>>> +fi
>>> +
>>> +LIBC_CONFIG_VAR([default-abi], [$libc_cv_loongarch_int_abi$libc_cv_loongarch_float_abi])
>>> -case $libc_cv_loongarch_int_abi in
>>> -lp64)
>>> +case $libc_cv_loongarch_int_abi$libc_cv_loongarch_float_abi in
>>> +lp64s)
>>> + LIBC_SLIBDIR_RTLDDIR([lib64/sf], [lib64])
>>> + ;;
>>> +lp64d)
>>> LIBC_SLIBDIR_RTLDDIR([lib64], [lib64])
>>> ;;
>>> esac
>>> diff --git a/sysdeps/unix/sysv/linux/loongarch/ldd-rewrite.sed b/sysdeps/unix/sysv/linux/loongarch/ldd-rewrite.sed
>>> index f8976fd239..cdbe5c3dc5 100644
>>> --- a/sysdeps/unix/sysv/linux/loongarch/ldd-rewrite.sed
>>> +++ b/sysdeps/unix/sysv/linux/loongarch/ldd-rewrite.sed
>>> @@ -1 +1 @@
>>> -s_^\(RTLDLIST=\)\(.*lib\)\(\|64\|32\)\(/[^/]*\)\(-loongarch\)\(64\|32\)\(\.so\.[0-9.]*\)[[:blank:]]*$_\1"\2\4\7 \264\4-loongarch64\7 \232\4-loongarch32\7"_
>>> +s_^\(RTLDLIST=\)\(.*lib\)\(\|64\|32\)\(/ld-linux-loongarch-\)\(lp\|ilp\)\(64\|32\)\(d\|s\)\(\.so\.[0-9.]*\)[[:blank:]]*$_\1"\264\4lp64d\8 \264\4lp64s\8"_
>>> diff --git a/sysdeps/unix/sysv/linux/loongarch/shlib-versions b/sysdeps/unix/sysv/linux/loongarch/shlib-versions
>>> index dc2220b4be..5f40e7f5ec 100644
>>> --- a/sysdeps/unix/sysv/linux/loongarch/shlib-versions
>>> +++ b/sysdeps/unix/sysv/linux/loongarch/shlib-versions
>>> @@ -2,6 +2,8 @@ DEFAULT GLIBC_2.36
>>> %if LOONGARCH_ABI_GRLEN == 64 && LOONGARCH_ABI_FRLEN == 64
>>> ld=ld-linux-loongarch-lp64d.so.1
>>> +%elif LOONGARCH_ABI_GRLEN == 64 && LOONGARCH_ABI_FRLEN == 0
>>> +ld=ld-linux-loongarch-lp64s.so.1
>>> %else
>>> %error cannot determine ABI
>>> %endif
>
prev parent reply other threads:[~2022-08-31 12:28 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-30 11:02 [PATCH 0/1] " caiyinyu
2022-08-30 11:02 ` [PATCH 1/1] " caiyinyu
2022-08-30 13:00 ` Adhemerval Zanella Netto
2022-08-31 12:00 ` caiyinyu
2022-08-31 12:28 ` Adhemerval Zanella Netto [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=8e153100-f89e-5934-921a-c48d219bc388@linaro.org \
--to=adhemerval.zanella@linaro.org \
--cc=caiyinyu@loongson.cn \
--cc=carlos@redhat.com \
--cc=i.swmail@xen0n.name \
--cc=joseph_myers@mentor.com \
--cc=libc-alpha@sourceware.org \
--cc=xry111@xry111.site \
--cc=xuchenghua@loongson.cn \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).