testsuite, X86, Darwin: Fix bf16 ABI tests for Mach-O/macOS ABI.
Commit Message
This is almost a completely Darwin-local patch, but there is one (repeated)
place where a general change is needed - which is in making xmm_regs and
x87_regs extern in the three copies of args.h (this is consistent with the
other saved vars). These fails represent most of the current testsuite noise
on x86 Darwin.
tested on x86-64 Darwin and Linux.
OK for master?
Iain
-- >8 --
These tests have failed since introduction since they assume that the
assembler output is ELF and that the ABI targeted supports the addressing.
For Darwin, Mach-O and ABI we need to make several changes:
1. Use the __USER_LABEL__PREFIX__
2. Remove the use of ELF-specific constructs (.size, .type etc.)
3. We cannot make direct access to common variables in the ABI, so that we
must move these to BSS.
Since that set is quite significant, I elected to make a separate source
section for Darwin. This is introduced by #elif defined(__APPLE__) because
__MACH__ is also used by HURD.
There are potentially other X86 targets (e.g. XCOFF) that could have yet
more changes, so I added a catchall section that #errors if the object format
is neither ELF or Mach-O.
Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
gcc/testsuite/ChangeLog:
* gcc.target/x86_64/abi/bf16/args.h:
* gcc.target/x86_64/abi/bf16/asm-support.S:
* gcc.target/x86_64/abi/bf16/m256bf16/args.h:
* gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S:
* gcc.target/x86_64/abi/bf16/m512bf16/args.h:
* gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S:
---
.../gcc.target/x86_64/abi/bf16/args.h | 4 +-
.../gcc.target/x86_64/abi/bf16/asm-support.S | 105 +++++++++++++++
.../x86_64/abi/bf16/m256bf16/args.h | 4 +-
.../x86_64/abi/bf16/m256bf16/asm-support.S | 107 +++++++++++++++
.../x86_64/abi/bf16/m512bf16/args.h | 4 +-
.../x86_64/abi/bf16/m512bf16/asm-support.S | 123 ++++++++++++++++++
6 files changed, 341 insertions(+), 6 deletions(-)
Comments
On Sun, Dec 4, 2022 at 12:51 PM Iain Sandoe <iains.gcc@gmail.com> wrote:
>
> This is almost a completely Darwin-local patch, but there is one (repeated)
> place where a general change is needed - which is in making xmm_regs and
> x87_regs extern in the three copies of args.h (this is consistent with the
> other saved vars). These fails represent most of the current testsuite noise
> on x86 Darwin.
>
> tested on x86-64 Darwin and Linux.
>
> OK for master?
> Iain
>
> -- >8 --
>
> These tests have failed since introduction since they assume that the
> assembler output is ELF and that the ABI targeted supports the addressing.
>
> For Darwin, Mach-O and ABI we need to make several changes:
> 1. Use the __USER_LABEL__PREFIX__
> 2. Remove the use of ELF-specific constructs (.size, .type etc.)
> 3. We cannot make direct access to common variables in the ABI, so that we
> must move these to BSS.
>
> Since that set is quite significant, I elected to make a separate source
> section for Darwin. This is introduced by #elif defined(__APPLE__) because
> __MACH__ is also used by HURD.
>
> There are potentially other X86 targets (e.g. XCOFF) that could have yet
> more changes, so I added a catchall section that #errors if the object format
> is neither ELF or Mach-O.
>
> Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
>
> gcc/testsuite/ChangeLog:
>
> * gcc.target/x86_64/abi/bf16/args.h:
> * gcc.target/x86_64/abi/bf16/asm-support.S:
> * gcc.target/x86_64/abi/bf16/m256bf16/args.h:
> * gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S:
> * gcc.target/x86_64/abi/bf16/m512bf16/args.h:
> * gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S:
Missing descriptions in ChangeLog entry.
Uros.
> ---
> .../gcc.target/x86_64/abi/bf16/args.h | 4 +-
> .../gcc.target/x86_64/abi/bf16/asm-support.S | 105 +++++++++++++++
> .../x86_64/abi/bf16/m256bf16/args.h | 4 +-
> .../x86_64/abi/bf16/m256bf16/asm-support.S | 107 +++++++++++++++
> .../x86_64/abi/bf16/m512bf16/args.h | 4 +-
> .../x86_64/abi/bf16/m512bf16/asm-support.S | 123 ++++++++++++++++++
> 6 files changed, 341 insertions(+), 6 deletions(-)
>
> diff --git a/gcc/testsuite/gcc.target/x86_64/abi/bf16/args.h b/gcc/testsuite/gcc.target/x86_64/abi/bf16/args.h
> index 11d7e2b3a1c..95f9a394f2c 100644
> --- a/gcc/testsuite/gcc.target/x86_64/abi/bf16/args.h
> +++ b/gcc/testsuite/gcc.target/x86_64/abi/bf16/args.h
> @@ -42,8 +42,8 @@ typedef union {
> } X87_T;
> extern void (*callthis)(void);
> extern unsigned long long rax,rbx,rcx,rdx,rsi,rdi,rsp,rbp,r8,r9,r10,r11,r12,r13,r14,r15;
> -XMM_T xmm_regs[16];
> -X87_T x87_regs[8];
> +extern XMM_T xmm_regs[16];
> +extern X87_T x87_regs[8];
> extern volatile unsigned long long volatile_var;
> extern void snapshot (void);
> extern void snapshot_ret (void);
> diff --git a/gcc/testsuite/gcc.target/x86_64/abi/bf16/asm-support.S b/gcc/testsuite/gcc.target/x86_64/abi/bf16/asm-support.S
> index 7559aa910c4..331bf92d761 100644
> --- a/gcc/testsuite/gcc.target/x86_64/abi/bf16/asm-support.S
> +++ b/gcc/testsuite/gcc.target/x86_64/abi/bf16/asm-support.S
> @@ -1,3 +1,5 @@
> +
> +#ifdef __ELF__
> .text
> .p2align 4,,15
> .globl snapshot
> @@ -82,3 +84,106 @@ snapshot_ret:
> #ifdef __linux__
> .section .note.GNU-stack,"",@progbits
> #endif
> +
> +#elif defined(__APPLE__)
> + .text
> + .p2align 4,,15
> + .globl _snapshot
> +_snapshot:
> +LFB3:
> + movq %rax, _rax(%rip)
> + movq %rbx, _rbx(%rip)
> + movq %rcx, _rcx(%rip)
> + movq %rdx, _rdx(%rip)
> + movq %rdi, _rdi(%rip)
> + movq %rsi, _rsi(%rip)
> + movq %rbp, _rbp(%rip)
> + movq %rsp, _rsp(%rip)
> + movq %r8, _r8(%rip)
> + movq %r9, _r9(%rip)
> + movq %r10, _r10(%rip)
> + movq %r11, _r11(%rip)
> + movq %r12, _r12(%rip)
> + movq %r13, _r13(%rip)
> + movq %r14, _r14(%rip)
> + movq %r15, _r15(%rip)
> + movdqu %xmm0, _xmm_regs+0(%rip)
> + movdqu %xmm1, _xmm_regs+16(%rip)
> + movdqu %xmm2, _xmm_regs+32(%rip)
> + movdqu %xmm3, _xmm_regs+48(%rip)
> + movdqu %xmm4, _xmm_regs+64(%rip)
> + movdqu %xmm5, _xmm_regs+80(%rip)
> + movdqu %xmm6, _xmm_regs+96(%rip)
> + movdqu %xmm7, _xmm_regs+112(%rip)
> + movdqu %xmm8, _xmm_regs+128(%rip)
> + movdqu %xmm9, _xmm_regs+144(%rip)
> + movdqu %xmm10, _xmm_regs+160(%rip)
> + movdqu %xmm11, _xmm_regs+176(%rip)
> + movdqu %xmm12, _xmm_regs+192(%rip)
> + movdqu %xmm13, _xmm_regs+208(%rip)
> + movdqu %xmm14, _xmm_regs+224(%rip)
> + movdqu %xmm15, _xmm_regs+240(%rip)
> + jmp *_callthis(%rip)
> +LFE3:
> +
> + .p2align 4,,15
> + .globl _snapshot_ret
> +_snapshot_ret:
> + movq %rdi, _rdi(%rip)
> + subq $8, %rsp
> + call *_callthis(%rip)
> + addq $8, %rsp
> + movq %rax, _rax(%rip)
> + movq %rdx, _rdx(%rip)
> + movdqu %xmm0, _xmm_regs+0(%rip)
> + movdqu %xmm1, _xmm_regs+16(%rip)
> + fstpt _x87_regs(%rip)
> + fstpt _x87_regs+16(%rip)
> + fldt _x87_regs+16(%rip)
> + fldt _x87_regs(%rip)
> + ret
> +
> + .globl _callthis
> + .zerofill __DATA,__bss,_callthis,8,3
> + .globl _rax
> + .zerofill __DATA,__bss,_rax,8,3
> + .globl _rbx
> + .zerofill __DATA,__bss,_rbx,8,3
> + .globl _rcx
> + .zerofill __DATA,__bss,_rcx,8,3
> + .globl _rdx
> + .zerofill __DATA,__bss,_rdx,8,3
> + .globl _rsi
> + .zerofill __DATA,__bss,_rsi,8,3
> + .globl _rdi
> + .zerofill __DATA,__bss,_rdi,8,3
> + .globl _rsp
> + .zerofill __DATA,__bss,_rsp,8,3
> + .globl _rbp
> + .zerofill __DATA,__bss,_rbp,8,3
> + .globl _r8
> + .zerofill __DATA,__bss,_r8,8,3
> + .globl _r9
> + .zerofill __DATA,__bss,_r9,8,3
> + .globl _r10
> + .zerofill __DATA,__bss,_r10,8,3
> + .globl _r11
> + .zerofill __DATA,__bss,_r11,8,3
> + .globl _r12
> + .zerofill __DATA,__bss,_r12,8,3
> + .globl _r13
> + .zerofill __DATA,__bss,_r13,8,3
> + .globl _r14
> + .zerofill __DATA,__bss,_r14,8,3
> + .globl _r15
> + .zerofill __DATA,__bss,_r15,8,3
> + .globl _xmm_regs
> + .zerofill __DATA,__bss,_xmm_regs,256,5
> + .globl _x87_regs
> + .zerofill __DATA,__bss,_x87_regs,128,5
> + .globl _volatile_var
> + .zerofill __DATA,__bss,_volatile_var,8,3
> +
> +#else
> +#error unknown object format
> +#endif
> diff --git a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/args.h b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/args.h
> index 94627ffbd44..1027742cbb2 100644
> --- a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/args.h
> +++ b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/args.h
> @@ -50,8 +50,8 @@ typedef union {
> } X87_T;
> extern void (*callthis)(void);
> extern unsigned long long rax,rbx,rcx,rdx,rsi,rdi,rsp,rbp,r8,r9,r10,r11,r12,r13,r14,r15;
> -YMM_T ymm_regs[16];
> -X87_T x87_regs[8];
> +extern YMM_T ymm_regs[16];
> +extern X87_T x87_regs[8];
> extern volatile unsigned long long volatile_var;
> extern void snapshot (void);
> extern void snapshot_ret (void);
> diff --git a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S
> index 24c8b3c9023..171654aa4db 100644
> --- a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S
> +++ b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S
> @@ -1,3 +1,5 @@
> +
> +#ifdef __ELF__
> .text
> .p2align 4,,15
> .globl snapshot
> @@ -82,3 +84,108 @@ snapshot_ret:
> #ifdef __linux__
> .section .note.GNU-stack,"",@progbits
> #endif
> +
> +#elif defined(__APPLE__)
> +
> + .text
> + .p2align 4,,15
> + .globl _snapshot
> +_snapshot:
> +.LFB3:
> + movq %rax, _rax(%rip)
> + movq %rbx, _rbx(%rip)
> + movq %rcx, _rcx(%rip)
> + movq %rdx, _rdx(%rip)
> + movq %rdi, _rdi(%rip)
> + movq %rsi, _rsi(%rip)
> + movq %rbp, _rbp(%rip)
> + movq %rsp, _rsp(%rip)
> + movq %r8, _r8(%rip)
> + movq %r9, _r9(%rip)
> + movq %r10, _r10(%rip)
> + movq %r11, _r11(%rip)
> + movq %r12, _r12(%rip)
> + movq %r13, _r13(%rip)
> + movq %r14, _r14(%rip)
> + movq %r15, _r15(%rip)
> + vmovdqu %ymm0, _ymm_regs+0(%rip)
> + vmovdqu %ymm1, _ymm_regs+32(%rip)
> + vmovdqu %ymm2, _ymm_regs+64(%rip)
> + vmovdqu %ymm3, _ymm_regs+96(%rip)
> + vmovdqu %ymm4, _ymm_regs+128(%rip)
> + vmovdqu %ymm5, _ymm_regs+160(%rip)
> + vmovdqu %ymm6, _ymm_regs+192(%rip)
> + vmovdqu %ymm7, _ymm_regs+224(%rip)
> + vmovdqu %ymm8, _ymm_regs+256(%rip)
> + vmovdqu %ymm9, _ymm_regs+288(%rip)
> + vmovdqu %ymm10, _ymm_regs+320(%rip)
> + vmovdqu %ymm11, _ymm_regs+352(%rip)
> + vmovdqu %ymm12, _ymm_regs+384(%rip)
> + vmovdqu %ymm13, _ymm_regs+416(%rip)
> + vmovdqu %ymm14, _ymm_regs+448(%rip)
> + vmovdqu %ymm15, _ymm_regs+480(%rip)
> + jmp *_callthis(%rip)
> +.LFE3:
> +
> + .p2align 4,,15
> + .globl _snapshot_ret
> +_snapshot_ret:
> + movq %rdi, _rdi(%rip)
> + subq $8, %rsp
> + call *_callthis(%rip)
> + addq $8, %rsp
> + movq %rax, _rax(%rip)
> + movq %rdx, _rdx(%rip)
> + vmovdqu %ymm0, _ymm_regs+0(%rip)
> + vmovdqu %ymm1, _ymm_regs+32(%rip)
> + fstpt _x87_regs(%rip)
> + fstpt _x87_regs+16(%rip)
> + fldt _x87_regs+16(%rip)
> + fldt _x87_regs(%rip)
> + ret
> +
> + .globl _callthis
> + .zerofill __DATA,__bss,_callthis,8,3
> + .globl _rax
> + .zerofill __DATA,__bss,_rax,8,3
> + .globl _rbx
> + .zerofill __DATA,__bss,_rbx,8,3
> + .globl _rcx
> + .zerofill __DATA,__bss,_rcx,8,3
> + .globl _rdx
> + .zerofill __DATA,__bss,_rdx,8,3
> + .globl _rsi
> + .zerofill __DATA,__bss,_rsi,8,3
> + .globl _rdi
> + .zerofill __DATA,__bss,_rdi,8,3
> + .globl _rsp
> + .zerofill __DATA,__bss,_rsp,8,3
> + .globl _rbp
> + .zerofill __DATA,__bss,_rbp,8,3
> + .globl _r8
> + .zerofill __DATA,__bss,_r8,8,3
> + .globl _r9
> + .zerofill __DATA,__bss,_r9,8,3
> + .globl _r10
> + .zerofill __DATA,__bss,_r10,8,3
> + .globl _r11
> + .zerofill __DATA,__bss,_r11,8,3
> + .globl _r12
> + .zerofill __DATA,__bss,_r12,8,3
> + .globl _r13
> + .zerofill __DATA,__bss,_r13,8,3
> + .globl _r14
> + .zerofill __DATA,__bss,_r14,8,3
> + .globl _r15
> + .zerofill __DATA,__bss,_r15,8,3
> + .globl _ymm_regs
> + .zerofill __DATA,__bss,_ymm_regs,512,5
> + .globl _x87_regs
> + .zerofill __DATA,__bss,_x87_regs,128,5
> + .globl _volatile_var
> + .zerofill __DATA,__bss,_volatile_var,8,3
> +
> +#else
> +#error unknown object format
> +#endif
> +
> diff --git a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/args.h b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/args.h
> index 64b24783833..f9710bae347 100644
> --- a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/args.h
> +++ b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/args.h
> @@ -51,8 +51,8 @@ typedef union {
> } X87_T;
> extern void (*callthis)(void);
> extern unsigned long long rax,rbx,rcx,rdx,rsi,rdi,rsp,rbp,r8,r9,r10,r11,r12,r13,r14,r15;
> -ZMM_T zmm_regs[32];
> -X87_T x87_regs[8];
> +extern ZMM_T zmm_regs[32];
> +extern X87_T x87_regs[8];
> extern volatile unsigned long long volatile_var;
> extern void snapshot (void);
> extern void snapshot_ret (void);
> diff --git a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S
> index 86d54d11c58..9dc6d173a61 100644
> --- a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S
> +++ b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S
> @@ -1,3 +1,5 @@
> +
> +#ifdef __ELF__
> .text
> .p2align 4,,15
> .globl snapshot
> @@ -98,3 +100,124 @@ snapshot_ret:
> #ifdef __linux__
> .section .note.GNU-stack,"",@progbits
> #endif
> +
> +#elif defined(__APPLE__)
> +
> + .text
> + .p2align 4,,15
> + .globl _snapshot
> +_snapshot:
> +.LFB3:
> + movq %rax, _rax(%rip)
> + movq %rbx, _rbx(%rip)
> + movq %rcx, _rcx(%rip)
> + movq %rdx, _rdx(%rip)
> + movq %rdi, _rdi(%rip)
> + movq %rsi, _rsi(%rip)
> + movq %rbp, _rbp(%rip)
> + movq %rsp, _rsp(%rip)
> + movq %r8, _r8(%rip)
> + movq %r9, _r9(%rip)
> + movq %r10, _r10(%rip)
> + movq %r11, _r11(%rip)
> + movq %r12, _r12(%rip)
> + movq %r13, _r13(%rip)
> + movq %r14, _r14(%rip)
> + movq %r15, _r15(%rip)
> + vmovdqu32 %zmm0, _zmm_regs+0(%rip)
> + vmovdqu32 %zmm1, _zmm_regs+64(%rip)
> + vmovdqu32 %zmm2, _zmm_regs+128(%rip)
> + vmovdqu32 %zmm3, _zmm_regs+192(%rip)
> + vmovdqu32 %zmm4, _zmm_regs+256(%rip)
> + vmovdqu32 %zmm5, _zmm_regs+320(%rip)
> + vmovdqu32 %zmm6, _zmm_regs+384(%rip)
> + vmovdqu32 %zmm7, _zmm_regs+448(%rip)
> + vmovdqu32 %zmm8, _zmm_regs+512(%rip)
> + vmovdqu32 %zmm9, _zmm_regs+576(%rip)
> + vmovdqu32 %zmm10, _zmm_regs+640(%rip)
> + vmovdqu32 %zmm11, _zmm_regs+704(%rip)
> + vmovdqu32 %zmm12, _zmm_regs+768(%rip)
> + vmovdqu32 %zmm13, _zmm_regs+832(%rip)
> + vmovdqu32 %zmm14, _zmm_regs+896(%rip)
> + vmovdqu32 %zmm15, _zmm_regs+960(%rip)
> + vmovdqu32 %zmm16, _zmm_regs+1024(%rip)
> + vmovdqu32 %zmm17, _zmm_regs+1088(%rip)
> + vmovdqu32 %zmm18, _zmm_regs+1152(%rip)
> + vmovdqu32 %zmm19, _zmm_regs+1216(%rip)
> + vmovdqu32 %zmm20, _zmm_regs+1280(%rip)
> + vmovdqu32 %zmm21, _zmm_regs+1344(%rip)
> + vmovdqu32 %zmm22, _zmm_regs+1408(%rip)
> + vmovdqu32 %zmm23, _zmm_regs+1472(%rip)
> + vmovdqu32 %zmm24, _zmm_regs+1536(%rip)
> + vmovdqu32 %zmm25, _zmm_regs+1600(%rip)
> + vmovdqu32 %zmm26, _zmm_regs+1664(%rip)
> + vmovdqu32 %zmm27, _zmm_regs+1728(%rip)
> + vmovdqu32 %zmm28, _zmm_regs+1792(%rip)
> + vmovdqu32 %zmm29, _zmm_regs+1856(%rip)
> + vmovdqu32 %zmm30, _zmm_regs+1920(%rip)
> + vmovdqu32 %zmm31, _zmm_regs+1984(%rip)
> + jmp *_callthis(%rip)
> +.LFE3:
> +
> + .p2align 4,,15
> + .globl _snapshot_ret
> +_snapshot_ret:
> + movq %rdi, _rdi(%rip)
> + subq $8, %rsp
> + call *_callthis(%rip)
> + addq $8, %rsp
> + movq %rax, _rax(%rip)
> + movq %rdx, _rdx(%rip)
> + vmovdqu32 %zmm0, _zmm_regs+0(%rip)
> + vmovdqu32 %zmm1, _zmm_regs+64(%rip)
> + fstpt _x87_regs(%rip)
> + fstpt _x87_regs+16(%rip)
> + fldt _x87_regs+16(%rip)
> + fldt _x87_regs(%rip)
> + ret
> +
> + .globl _callthis
> + .zerofill __DATA,__bss,_callthis,8,3
> + .globl _rax
> + .zerofill __DATA,__bss,_rax,8,3
> + .globl _rbx
> + .zerofill __DATA,__bss,_rbx,8,3
> + .globl _rcx
> + .zerofill __DATA,__bss,_rcx,8,3
> + .globl _rdx
> + .zerofill __DATA,__bss,_rdx,8,3
> + .globl _rsi
> + .zerofill __DATA,__bss,_rsi,8,3
> + .globl _rdi
> + .zerofill __DATA,__bss,_rdi,8,3
> + .globl _rsp
> + .zerofill __DATA,__bss,_rsp,8,3
> + .globl _rbp
> + .zerofill __DATA,__bss,_rbp,8,3
> + .globl _r8
> + .zerofill __DATA,__bss,_r8,8,3
> + .globl _r9
> + .zerofill __DATA,__bss,_r9,8,3
> + .globl _r10
> + .zerofill __DATA,__bss,_r10,8,3
> + .globl _r11
> + .zerofill __DATA,__bss,_r11,8,3
> + .globl _r12
> + .zerofill __DATA,__bss,_r12,8,3
> + .globl _r13
> + .zerofill __DATA,__bss,_r13,8,3
> + .globl _r14
> + .zerofill __DATA,__bss,_r14,8,3
> + .globl _r15
> + .zerofill __DATA,__bss,_r15,8,3
> + .globl _zmm_regs
> + .zerofill __DATA,__bss,_zmm_regs,2048,6
> + .globl _x87_regs
> + .zerofill __DATA,__bss,_x87_regs,128,5
> + .globl _volatile_var
> + .zerofill __DATA,__bss,_volatile_var,8,3
> +
> +#else
> +#error unknown object format
> +#endif
> +
> --
> 2.37.1 (Apple Git-137.1)
>
> On 4 Dec 2022, at 20:20, Uros Bizjak via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
>
> On Sun, Dec 4, 2022 at 12:51 PM Iain Sandoe <iains.gcc@gmail.com> wrote:
>>
>> This is almost a completely Darwin-local patch, but there is one (repeated)
>> place where a general change is needed - which is in making xmm_regs and
>> x87_regs extern in the three copies of args.h (this is consistent with the
>> other saved vars). These fails represent most of the current testsuite noise
>> on x86 Darwin.
>>
>> tested on x86-64 Darwin and Linux.
>>
>> OK for master?
>> Iain
>>
>> -- >8 --
>>
>> These tests have failed since introduction since they assume that the
>> assembler output is ELF and that the ABI targeted supports the addressing.
>>
>> For Darwin, Mach-O and ABI we need to make several changes:
>> 1. Use the __USER_LABEL__PREFIX__
>> 2. Remove the use of ELF-specific constructs (.size, .type etc.)
>> 3. We cannot make direct access to common variables in the ABI, so that we
>> must move these to BSS.
>>
>> Since that set is quite significant, I elected to make a separate source
>> section for Darwin. This is introduced by #elif defined(__APPLE__) because
>> __MACH__ is also used by HURD.
>>
>> There are potentially other X86 targets (e.g. XCOFF) that could have yet
>> more changes, so I added a catchall section that #errors if the object format
>> is neither ELF or Mach-O.
>>
>> Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
>>
>> gcc/testsuite/ChangeLog:
>>
>> * gcc.target/x86_64/abi/bf16/args.h:
>> * gcc.target/x86_64/abi/bf16/asm-support.S:
>> * gcc.target/x86_64/abi/bf16/m256bf16/args.h:
>> * gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S:
>> * gcc.target/x86_64/abi/bf16/m512bf16/args.h:
>> * gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S:
>
> Missing descriptions in ChangeLog entry.
oops, here:
gcc/testsuite/ChangeLog:
* gcc.target/x86_64/abi/bf16/args.h: Make xmm_regs, x87_regs extern.
* gcc.target/x86_64/abi/bf16/m256bf16/args.h: Likewise.
* gcc.target/x86_64/abi/bf16/m512bf16/args.h: Likewise.
* gcc.target/x86_64/abi/bf16/asm-support.S: Add Mach-O variant.
* gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S: Likewise.
* gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S: Likewise.
Iain.
>
> Uros.
>
>> ---
>> .../gcc.target/x86_64/abi/bf16/args.h | 4 +-
>> .../gcc.target/x86_64/abi/bf16/asm-support.S | 105 +++++++++++++++
>> .../x86_64/abi/bf16/m256bf16/args.h | 4 +-
>> .../x86_64/abi/bf16/m256bf16/asm-support.S | 107 +++++++++++++++
>> .../x86_64/abi/bf16/m512bf16/args.h | 4 +-
>> .../x86_64/abi/bf16/m512bf16/asm-support.S | 123 ++++++++++++++++++
>> 6 files changed, 341 insertions(+), 6 deletions(-)
>>
>> diff --git a/gcc/testsuite/gcc.target/x86_64/abi/bf16/args.h b/gcc/testsuite/gcc.target/x86_64/abi/bf16/args.h
>> index 11d7e2b3a1c..95f9a394f2c 100644
>> --- a/gcc/testsuite/gcc.target/x86_64/abi/bf16/args.h
>> +++ b/gcc/testsuite/gcc.target/x86_64/abi/bf16/args.h
>> @@ -42,8 +42,8 @@ typedef union {
>> } X87_T;
>> extern void (*callthis)(void);
>> extern unsigned long long rax,rbx,rcx,rdx,rsi,rdi,rsp,rbp,r8,r9,r10,r11,r12,r13,r14,r15;
>> -XMM_T xmm_regs[16];
>> -X87_T x87_regs[8];
>> +extern XMM_T xmm_regs[16];
>> +extern X87_T x87_regs[8];
>> extern volatile unsigned long long volatile_var;
>> extern void snapshot (void);
>> extern void snapshot_ret (void);
>> diff --git a/gcc/testsuite/gcc.target/x86_64/abi/bf16/asm-support.S b/gcc/testsuite/gcc.target/x86_64/abi/bf16/asm-support.S
>> index 7559aa910c4..331bf92d761 100644
>> --- a/gcc/testsuite/gcc.target/x86_64/abi/bf16/asm-support.S
>> +++ b/gcc/testsuite/gcc.target/x86_64/abi/bf16/asm-support.S
>> @@ -1,3 +1,5 @@
>> +
>> +#ifdef __ELF__
>> .text
>> .p2align 4,,15
>> .globl snapshot
>> @@ -82,3 +84,106 @@ snapshot_ret:
>> #ifdef __linux__
>> .section .note.GNU-stack,"",@progbits
>> #endif
>> +
>> +#elif defined(__APPLE__)
>> + .text
>> + .p2align 4,,15
>> + .globl _snapshot
>> +_snapshot:
>> +LFB3:
>> + movq %rax, _rax(%rip)
>> + movq %rbx, _rbx(%rip)
>> + movq %rcx, _rcx(%rip)
>> + movq %rdx, _rdx(%rip)
>> + movq %rdi, _rdi(%rip)
>> + movq %rsi, _rsi(%rip)
>> + movq %rbp, _rbp(%rip)
>> + movq %rsp, _rsp(%rip)
>> + movq %r8, _r8(%rip)
>> + movq %r9, _r9(%rip)
>> + movq %r10, _r10(%rip)
>> + movq %r11, _r11(%rip)
>> + movq %r12, _r12(%rip)
>> + movq %r13, _r13(%rip)
>> + movq %r14, _r14(%rip)
>> + movq %r15, _r15(%rip)
>> + movdqu %xmm0, _xmm_regs+0(%rip)
>> + movdqu %xmm1, _xmm_regs+16(%rip)
>> + movdqu %xmm2, _xmm_regs+32(%rip)
>> + movdqu %xmm3, _xmm_regs+48(%rip)
>> + movdqu %xmm4, _xmm_regs+64(%rip)
>> + movdqu %xmm5, _xmm_regs+80(%rip)
>> + movdqu %xmm6, _xmm_regs+96(%rip)
>> + movdqu %xmm7, _xmm_regs+112(%rip)
>> + movdqu %xmm8, _xmm_regs+128(%rip)
>> + movdqu %xmm9, _xmm_regs+144(%rip)
>> + movdqu %xmm10, _xmm_regs+160(%rip)
>> + movdqu %xmm11, _xmm_regs+176(%rip)
>> + movdqu %xmm12, _xmm_regs+192(%rip)
>> + movdqu %xmm13, _xmm_regs+208(%rip)
>> + movdqu %xmm14, _xmm_regs+224(%rip)
>> + movdqu %xmm15, _xmm_regs+240(%rip)
>> + jmp *_callthis(%rip)
>> +LFE3:
>> +
>> + .p2align 4,,15
>> + .globl _snapshot_ret
>> +_snapshot_ret:
>> + movq %rdi, _rdi(%rip)
>> + subq $8, %rsp
>> + call *_callthis(%rip)
>> + addq $8, %rsp
>> + movq %rax, _rax(%rip)
>> + movq %rdx, _rdx(%rip)
>> + movdqu %xmm0, _xmm_regs+0(%rip)
>> + movdqu %xmm1, _xmm_regs+16(%rip)
>> + fstpt _x87_regs(%rip)
>> + fstpt _x87_regs+16(%rip)
>> + fldt _x87_regs+16(%rip)
>> + fldt _x87_regs(%rip)
>> + ret
>> +
>> + .globl _callthis
>> + .zerofill __DATA,__bss,_callthis,8,3
>> + .globl _rax
>> + .zerofill __DATA,__bss,_rax,8,3
>> + .globl _rbx
>> + .zerofill __DATA,__bss,_rbx,8,3
>> + .globl _rcx
>> + .zerofill __DATA,__bss,_rcx,8,3
>> + .globl _rdx
>> + .zerofill __DATA,__bss,_rdx,8,3
>> + .globl _rsi
>> + .zerofill __DATA,__bss,_rsi,8,3
>> + .globl _rdi
>> + .zerofill __DATA,__bss,_rdi,8,3
>> + .globl _rsp
>> + .zerofill __DATA,__bss,_rsp,8,3
>> + .globl _rbp
>> + .zerofill __DATA,__bss,_rbp,8,3
>> + .globl _r8
>> + .zerofill __DATA,__bss,_r8,8,3
>> + .globl _r9
>> + .zerofill __DATA,__bss,_r9,8,3
>> + .globl _r10
>> + .zerofill __DATA,__bss,_r10,8,3
>> + .globl _r11
>> + .zerofill __DATA,__bss,_r11,8,3
>> + .globl _r12
>> + .zerofill __DATA,__bss,_r12,8,3
>> + .globl _r13
>> + .zerofill __DATA,__bss,_r13,8,3
>> + .globl _r14
>> + .zerofill __DATA,__bss,_r14,8,3
>> + .globl _r15
>> + .zerofill __DATA,__bss,_r15,8,3
>> + .globl _xmm_regs
>> + .zerofill __DATA,__bss,_xmm_regs,256,5
>> + .globl _x87_regs
>> + .zerofill __DATA,__bss,_x87_regs,128,5
>> + .globl _volatile_var
>> + .zerofill __DATA,__bss,_volatile_var,8,3
>> +
>> +#else
>> +#error unknown object format
>> +#endif
>> diff --git a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/args.h b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/args.h
>> index 94627ffbd44..1027742cbb2 100644
>> --- a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/args.h
>> +++ b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/args.h
>> @@ -50,8 +50,8 @@ typedef union {
>> } X87_T;
>> extern void (*callthis)(void);
>> extern unsigned long long rax,rbx,rcx,rdx,rsi,rdi,rsp,rbp,r8,r9,r10,r11,r12,r13,r14,r15;
>> -YMM_T ymm_regs[16];
>> -X87_T x87_regs[8];
>> +extern YMM_T ymm_regs[16];
>> +extern X87_T x87_regs[8];
>> extern volatile unsigned long long volatile_var;
>> extern void snapshot (void);
>> extern void snapshot_ret (void);
>> diff --git a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S
>> index 24c8b3c9023..171654aa4db 100644
>> --- a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S
>> +++ b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S
>> @@ -1,3 +1,5 @@
>> +
>> +#ifdef __ELF__
>> .text
>> .p2align 4,,15
>> .globl snapshot
>> @@ -82,3 +84,108 @@ snapshot_ret:
>> #ifdef __linux__
>> .section .note.GNU-stack,"",@progbits
>> #endif
>> +
>> +#elif defined(__APPLE__)
>> +
>> + .text
>> + .p2align 4,,15
>> + .globl _snapshot
>> +_snapshot:
>> +.LFB3:
>> + movq %rax, _rax(%rip)
>> + movq %rbx, _rbx(%rip)
>> + movq %rcx, _rcx(%rip)
>> + movq %rdx, _rdx(%rip)
>> + movq %rdi, _rdi(%rip)
>> + movq %rsi, _rsi(%rip)
>> + movq %rbp, _rbp(%rip)
>> + movq %rsp, _rsp(%rip)
>> + movq %r8, _r8(%rip)
>> + movq %r9, _r9(%rip)
>> + movq %r10, _r10(%rip)
>> + movq %r11, _r11(%rip)
>> + movq %r12, _r12(%rip)
>> + movq %r13, _r13(%rip)
>> + movq %r14, _r14(%rip)
>> + movq %r15, _r15(%rip)
>> + vmovdqu %ymm0, _ymm_regs+0(%rip)
>> + vmovdqu %ymm1, _ymm_regs+32(%rip)
>> + vmovdqu %ymm2, _ymm_regs+64(%rip)
>> + vmovdqu %ymm3, _ymm_regs+96(%rip)
>> + vmovdqu %ymm4, _ymm_regs+128(%rip)
>> + vmovdqu %ymm5, _ymm_regs+160(%rip)
>> + vmovdqu %ymm6, _ymm_regs+192(%rip)
>> + vmovdqu %ymm7, _ymm_regs+224(%rip)
>> + vmovdqu %ymm8, _ymm_regs+256(%rip)
>> + vmovdqu %ymm9, _ymm_regs+288(%rip)
>> + vmovdqu %ymm10, _ymm_regs+320(%rip)
>> + vmovdqu %ymm11, _ymm_regs+352(%rip)
>> + vmovdqu %ymm12, _ymm_regs+384(%rip)
>> + vmovdqu %ymm13, _ymm_regs+416(%rip)
>> + vmovdqu %ymm14, _ymm_regs+448(%rip)
>> + vmovdqu %ymm15, _ymm_regs+480(%rip)
>> + jmp *_callthis(%rip)
>> +.LFE3:
>> +
>> + .p2align 4,,15
>> + .globl _snapshot_ret
>> +_snapshot_ret:
>> + movq %rdi, _rdi(%rip)
>> + subq $8, %rsp
>> + call *_callthis(%rip)
>> + addq $8, %rsp
>> + movq %rax, _rax(%rip)
>> + movq %rdx, _rdx(%rip)
>> + vmovdqu %ymm0, _ymm_regs+0(%rip)
>> + vmovdqu %ymm1, _ymm_regs+32(%rip)
>> + fstpt _x87_regs(%rip)
>> + fstpt _x87_regs+16(%rip)
>> + fldt _x87_regs+16(%rip)
>> + fldt _x87_regs(%rip)
>> + ret
>> +
>> + .globl _callthis
>> + .zerofill __DATA,__bss,_callthis,8,3
>> + .globl _rax
>> + .zerofill __DATA,__bss,_rax,8,3
>> + .globl _rbx
>> + .zerofill __DATA,__bss,_rbx,8,3
>> + .globl _rcx
>> + .zerofill __DATA,__bss,_rcx,8,3
>> + .globl _rdx
>> + .zerofill __DATA,__bss,_rdx,8,3
>> + .globl _rsi
>> + .zerofill __DATA,__bss,_rsi,8,3
>> + .globl _rdi
>> + .zerofill __DATA,__bss,_rdi,8,3
>> + .globl _rsp
>> + .zerofill __DATA,__bss,_rsp,8,3
>> + .globl _rbp
>> + .zerofill __DATA,__bss,_rbp,8,3
>> + .globl _r8
>> + .zerofill __DATA,__bss,_r8,8,3
>> + .globl _r9
>> + .zerofill __DATA,__bss,_r9,8,3
>> + .globl _r10
>> + .zerofill __DATA,__bss,_r10,8,3
>> + .globl _r11
>> + .zerofill __DATA,__bss,_r11,8,3
>> + .globl _r12
>> + .zerofill __DATA,__bss,_r12,8,3
>> + .globl _r13
>> + .zerofill __DATA,__bss,_r13,8,3
>> + .globl _r14
>> + .zerofill __DATA,__bss,_r14,8,3
>> + .globl _r15
>> + .zerofill __DATA,__bss,_r15,8,3
>> + .globl _ymm_regs
>> + .zerofill __DATA,__bss,_ymm_regs,512,5
>> + .globl _x87_regs
>> + .zerofill __DATA,__bss,_x87_regs,128,5
>> + .globl _volatile_var
>> + .zerofill __DATA,__bss,_volatile_var,8,3
>> +
>> +#else
>> +#error unknown object format
>> +#endif
>> +
>> diff --git a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/args.h b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/args.h
>> index 64b24783833..f9710bae347 100644
>> --- a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/args.h
>> +++ b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/args.h
>> @@ -51,8 +51,8 @@ typedef union {
>> } X87_T;
>> extern void (*callthis)(void);
>> extern unsigned long long rax,rbx,rcx,rdx,rsi,rdi,rsp,rbp,r8,r9,r10,r11,r12,r13,r14,r15;
>> -ZMM_T zmm_regs[32];
>> -X87_T x87_regs[8];
>> +extern ZMM_T zmm_regs[32];
>> +extern X87_T x87_regs[8];
>> extern volatile unsigned long long volatile_var;
>> extern void snapshot (void);
>> extern void snapshot_ret (void);
>> diff --git a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S
>> index 86d54d11c58..9dc6d173a61 100644
>> --- a/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S
>> +++ b/gcc/testsuite/gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S
>> @@ -1,3 +1,5 @@
>> +
>> +#ifdef __ELF__
>> .text
>> .p2align 4,,15
>> .globl snapshot
>> @@ -98,3 +100,124 @@ snapshot_ret:
>> #ifdef __linux__
>> .section .note.GNU-stack,"",@progbits
>> #endif
>> +
>> +#elif defined(__APPLE__)
>> +
>> + .text
>> + .p2align 4,,15
>> + .globl _snapshot
>> +_snapshot:
>> +.LFB3:
>> + movq %rax, _rax(%rip)
>> + movq %rbx, _rbx(%rip)
>> + movq %rcx, _rcx(%rip)
>> + movq %rdx, _rdx(%rip)
>> + movq %rdi, _rdi(%rip)
>> + movq %rsi, _rsi(%rip)
>> + movq %rbp, _rbp(%rip)
>> + movq %rsp, _rsp(%rip)
>> + movq %r8, _r8(%rip)
>> + movq %r9, _r9(%rip)
>> + movq %r10, _r10(%rip)
>> + movq %r11, _r11(%rip)
>> + movq %r12, _r12(%rip)
>> + movq %r13, _r13(%rip)
>> + movq %r14, _r14(%rip)
>> + movq %r15, _r15(%rip)
>> + vmovdqu32 %zmm0, _zmm_regs+0(%rip)
>> + vmovdqu32 %zmm1, _zmm_regs+64(%rip)
>> + vmovdqu32 %zmm2, _zmm_regs+128(%rip)
>> + vmovdqu32 %zmm3, _zmm_regs+192(%rip)
>> + vmovdqu32 %zmm4, _zmm_regs+256(%rip)
>> + vmovdqu32 %zmm5, _zmm_regs+320(%rip)
>> + vmovdqu32 %zmm6, _zmm_regs+384(%rip)
>> + vmovdqu32 %zmm7, _zmm_regs+448(%rip)
>> + vmovdqu32 %zmm8, _zmm_regs+512(%rip)
>> + vmovdqu32 %zmm9, _zmm_regs+576(%rip)
>> + vmovdqu32 %zmm10, _zmm_regs+640(%rip)
>> + vmovdqu32 %zmm11, _zmm_regs+704(%rip)
>> + vmovdqu32 %zmm12, _zmm_regs+768(%rip)
>> + vmovdqu32 %zmm13, _zmm_regs+832(%rip)
>> + vmovdqu32 %zmm14, _zmm_regs+896(%rip)
>> + vmovdqu32 %zmm15, _zmm_regs+960(%rip)
>> + vmovdqu32 %zmm16, _zmm_regs+1024(%rip)
>> + vmovdqu32 %zmm17, _zmm_regs+1088(%rip)
>> + vmovdqu32 %zmm18, _zmm_regs+1152(%rip)
>> + vmovdqu32 %zmm19, _zmm_regs+1216(%rip)
>> + vmovdqu32 %zmm20, _zmm_regs+1280(%rip)
>> + vmovdqu32 %zmm21, _zmm_regs+1344(%rip)
>> + vmovdqu32 %zmm22, _zmm_regs+1408(%rip)
>> + vmovdqu32 %zmm23, _zmm_regs+1472(%rip)
>> + vmovdqu32 %zmm24, _zmm_regs+1536(%rip)
>> + vmovdqu32 %zmm25, _zmm_regs+1600(%rip)
>> + vmovdqu32 %zmm26, _zmm_regs+1664(%rip)
>> + vmovdqu32 %zmm27, _zmm_regs+1728(%rip)
>> + vmovdqu32 %zmm28, _zmm_regs+1792(%rip)
>> + vmovdqu32 %zmm29, _zmm_regs+1856(%rip)
>> + vmovdqu32 %zmm30, _zmm_regs+1920(%rip)
>> + vmovdqu32 %zmm31, _zmm_regs+1984(%rip)
>> + jmp *_callthis(%rip)
>> +.LFE3:
>> +
>> + .p2align 4,,15
>> + .globl _snapshot_ret
>> +_snapshot_ret:
>> + movq %rdi, _rdi(%rip)
>> + subq $8, %rsp
>> + call *_callthis(%rip)
>> + addq $8, %rsp
>> + movq %rax, _rax(%rip)
>> + movq %rdx, _rdx(%rip)
>> + vmovdqu32 %zmm0, _zmm_regs+0(%rip)
>> + vmovdqu32 %zmm1, _zmm_regs+64(%rip)
>> + fstpt _x87_regs(%rip)
>> + fstpt _x87_regs+16(%rip)
>> + fldt _x87_regs+16(%rip)
>> + fldt _x87_regs(%rip)
>> + ret
>> +
>> + .globl _callthis
>> + .zerofill __DATA,__bss,_callthis,8,3
>> + .globl _rax
>> + .zerofill __DATA,__bss,_rax,8,3
>> + .globl _rbx
>> + .zerofill __DATA,__bss,_rbx,8,3
>> + .globl _rcx
>> + .zerofill __DATA,__bss,_rcx,8,3
>> + .globl _rdx
>> + .zerofill __DATA,__bss,_rdx,8,3
>> + .globl _rsi
>> + .zerofill __DATA,__bss,_rsi,8,3
>> + .globl _rdi
>> + .zerofill __DATA,__bss,_rdi,8,3
>> + .globl _rsp
>> + .zerofill __DATA,__bss,_rsp,8,3
>> + .globl _rbp
>> + .zerofill __DATA,__bss,_rbp,8,3
>> + .globl _r8
>> + .zerofill __DATA,__bss,_r8,8,3
>> + .globl _r9
>> + .zerofill __DATA,__bss,_r9,8,3
>> + .globl _r10
>> + .zerofill __DATA,__bss,_r10,8,3
>> + .globl _r11
>> + .zerofill __DATA,__bss,_r11,8,3
>> + .globl _r12
>> + .zerofill __DATA,__bss,_r12,8,3
>> + .globl _r13
>> + .zerofill __DATA,__bss,_r13,8,3
>> + .globl _r14
>> + .zerofill __DATA,__bss,_r14,8,3
>> + .globl _r15
>> + .zerofill __DATA,__bss,_r15,8,3
>> + .globl _zmm_regs
>> + .zerofill __DATA,__bss,_zmm_regs,2048,6
>> + .globl _x87_regs
>> + .zerofill __DATA,__bss,_x87_regs,128,5
>> + .globl _volatile_var
>> + .zerofill __DATA,__bss,_volatile_var,8,3
>> +
>> +#else
>> +#error unknown object format
>> +#endif
>> +
>> --
>> 2.37.1 (Apple Git-137.1)
On Sun, Dec 4, 2022 at 9:30 PM Iain Sandoe <iain@sandoe.co.uk> wrote:
>
>
>
> > On 4 Dec 2022, at 20:20, Uros Bizjak via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
> >
> > On Sun, Dec 4, 2022 at 12:51 PM Iain Sandoe <iains.gcc@gmail.com> wrote:
> >>
> >> This is almost a completely Darwin-local patch, but there is one (repeated)
> >> place where a general change is needed - which is in making xmm_regs and
> >> x87_regs extern in the three copies of args.h (this is consistent with the
> >> other saved vars). These fails represent most of the current testsuite noise
> >> on x86 Darwin.
> >>
> >> tested on x86-64 Darwin and Linux.
> >>
> >> OK for master?
> >> Iain
> >>
> >> -- >8 --
> >>
> >> These tests have failed since introduction since they assume that the
> >> assembler output is ELF and that the ABI targeted supports the addressing.
> >>
> >> For Darwin, Mach-O and ABI we need to make several changes:
> >> 1. Use the __USER_LABEL__PREFIX__
> >> 2. Remove the use of ELF-specific constructs (.size, .type etc.)
> >> 3. We cannot make direct access to common variables in the ABI, so that we
> >> must move these to BSS.
> >>
> >> Since that set is quite significant, I elected to make a separate source
> >> section for Darwin. This is introduced by #elif defined(__APPLE__) because
> >> __MACH__ is also used by HURD.
> >>
> >> There are potentially other X86 targets (e.g. XCOFF) that could have yet
> >> more changes, so I added a catchall section that #errors if the object format
> >> is neither ELF or Mach-O.
> >>
> >> Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>
> >>
> >> gcc/testsuite/ChangeLog:
> >>
> >> * gcc.target/x86_64/abi/bf16/args.h:
> >> * gcc.target/x86_64/abi/bf16/asm-support.S:
> >> * gcc.target/x86_64/abi/bf16/m256bf16/args.h:
> >> * gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S:
> >> * gcc.target/x86_64/abi/bf16/m512bf16/args.h:
> >> * gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S:
> >
> > Missing descriptions in ChangeLog entry.
>
> oops, here:
>
> gcc/testsuite/ChangeLog:
>
> * gcc.target/x86_64/abi/bf16/args.h: Make xmm_regs, x87_regs extern.
> * gcc.target/x86_64/abi/bf16/m256bf16/args.h: Likewise.
> * gcc.target/x86_64/abi/bf16/m512bf16/args.h: Likewise.
> * gcc.target/x86_64/abi/bf16/asm-support.S: Add Mach-O variant.
> * gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S: Likewise.
> * gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S: Likewise.
Please note that in other directories asm-support-darwin.s is
introduced and included via .exp file. Is there a reason a different
approach is introduced here?
Uros.
Hi Uros,
> On 5 Dec 2022, at 10:37, Uros Bizjak via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
>
> On Sun, Dec 4, 2022 at 9:30 PM Iain Sandoe <iain@sandoe.co.uk> wrote:
>>
>> gcc/testsuite/ChangeLog:
>>
>> * gcc.target/x86_64/abi/bf16/args.h: Make xmm_regs, x87_regs extern.
>> * gcc.target/x86_64/abi/bf16/m256bf16/args.h: Likewise.
>> * gcc.target/x86_64/abi/bf16/m512bf16/args.h: Likewise.
>> * gcc.target/x86_64/abi/bf16/asm-support.S: Add Mach-O variant.
>> * gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S: Likewise.
>> * gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S: Likewise.
>
> Please note that in other directories asm-support-darwin.s is
> introduced and included via .exp file. Is there a reason a different
> approach is introduced here?
Since it seems that testcases get added and amended without considering any
sub-target apart from x86_64-linux-gnu (even by very experienced contributors),
I was hoping that the Darwin section might prompt folks to remember that there
are several other sub-targets.
However, the main thing is to fix the tests .. so here’s a version using separate
files.
OK?
thanks,
Iain
On Mon, Dec 5, 2022 at 3:54 PM Iain Sandoe <iain@sandoe.co.uk> wrote:
>
> Hi Uros,
>
> > On 5 Dec 2022, at 10:37, Uros Bizjak via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
> >
> > On Sun, Dec 4, 2022 at 9:30 PM Iain Sandoe <iain@sandoe.co.uk> wrote:
> >>
>
> >> gcc/testsuite/ChangeLog:
> >>
> >> * gcc.target/x86_64/abi/bf16/args.h: Make xmm_regs, x87_regs extern.
> >> * gcc.target/x86_64/abi/bf16/m256bf16/args.h: Likewise.
> >> * gcc.target/x86_64/abi/bf16/m512bf16/args.h: Likewise.
> >> * gcc.target/x86_64/abi/bf16/asm-support.S: Add Mach-O variant.
> >> * gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S: Likewise.
> >> * gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S: Likewise.
> >
> > Please note that in other directories asm-support-darwin.s is
> > introduced and included via .exp file. Is there a reason a different
> > approach is introduced here?
>
> Since it seems that testcases get added and amended without considering any
> sub-target apart from x86_64-linux-gnu (even by very experienced contributors),
> I was hoping that the Darwin section might prompt folks to remember that there
> are several other sub-targets.
>
> However, the main thing is to fix the tests .. so here’s a version using separate
> files.
extern void (*callthis)(void);
extern unsigned long long
rax,rbx,rcx,rdx,rsi,rdi,rsp,rbp,r8,r9,r10,r11,r12,r13,r14,r15;
-XMM_T xmm_regs[16];
-X87_T x87_regs[8];
+extern XMM_T xmm_regs[16];
+extern X87_T x87_regs[8];
Do you still need this change? Existing test files are compiled without extern.
+ .globl _callthis
+ .zerofill __DATA,__bss,_callthis,8,3
+ .globl _rax
+ .zerofill __DATA,__bss,_rax,8,3
+ .globl _rbx
+ .zerofill __DATA,__bss,_rbx,8,3
+ .globl _rcx
+ .zerofill __DATA,__bss,_rcx,8,3
+ .globl _rdx
+ .zerofill __DATA,__bss,_rdx,8,3
...
I wonder if the above approach is better than existing:
.comm _callthis,8
.comm _rax,8
.comm _rbx,8
.comm _rcx,8
.comm _rdx,8
...
It is strange to have two different approaches for similar tests. If
the new approach is better, we should also change existing asm-support
files.
Uros.
Hi Uros,
> On 5 Dec 2022, at 21:07, Uros Bizjak via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
>
> On Mon, Dec 5, 2022 at 3:54 PM Iain Sandoe <iain@sandoe.co.uk> wrote:
>>
>> Hi Uros,
>>
>>> On 5 Dec 2022, at 10:37, Uros Bizjak via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
>>>
>>> On Sun, Dec 4, 2022 at 9:30 PM Iain Sandoe <iain@sandoe.co.uk> wrote:
>>>>
>>
>>>> gcc/testsuite/ChangeLog:
>>>>
>>>> * gcc.target/x86_64/abi/bf16/args.h: Make xmm_regs, x87_regs extern.
>>>> * gcc.target/x86_64/abi/bf16/m256bf16/args.h: Likewise.
>>>> * gcc.target/x86_64/abi/bf16/m512bf16/args.h: Likewise.
>>>> * gcc.target/x86_64/abi/bf16/asm-support.S: Add Mach-O variant.
>>>> * gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S: Likewise.
>>>> * gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S: Likewise.
>>>
>>> Please note that in other directories asm-support-darwin.s is
>>> introduced and included via .exp file. Is there a reason a different
>>> approach is introduced here?
>>
>> Since it seems that testcases get added and amended without considering any
>> sub-target apart from x86_64-linux-gnu (even by very experienced contributors),
>> I was hoping that the Darwin section might prompt folks to remember that there
>> are several other sub-targets.
>>
>> However, the main thing is to fix the tests .. so here’s a version using separate
>> files.
>
> extern void (*callthis)(void);
> extern unsigned long long
> rax,rbx,rcx,rdx,rsi,rdi,rsp,rbp,r8,r9,r10,r11,r12,r13,r14,r15;
> -XMM_T xmm_regs[16];
> -X87_T x87_regs[8];
> +extern XMM_T xmm_regs[16];
> +extern X87_T x87_regs[8];
>
> Do you still need this change? Existing test files are compiled without extern.
>
> + .globl _callthis
> + .zerofill __DATA,__bss,_callthis,8,3
> + .globl _rax
> + .zerofill __DATA,__bss,_rax,8,3
> + .globl _rbx
> + .zerofill __DATA,__bss,_rbx,8,3
> + .globl _rcx
> + .zerofill __DATA,__bss,_rcx,8,3
> + .globl _rdx
> + .zerofill __DATA,__bss,_rdx,8,3
> ...
>
> I wonder if the above approach is better than existing:
>
> .comm _callthis,8
> .comm _rax,8
> .comm _rbx,8
> .comm _rcx,8
> .comm _rdx,8
> ...
As noted in the changelog, direct access to common data is not permitted in the Darwin
ABI [for x86_64, it would need to be _xxx@GOTPCREL(%rip)..] that’s why these have
been moved to bss.
> It is strange to have two different approaches for similar tests. If
> the new approach is better, we should also change existing asm-support
> files.
could be, I have not checked other case so far (extremely limited time at the moment)
Quite likely, the accesses work in the testcases, despite violating the ABI.
Iain
On Mon, Dec 5, 2022 at 10:17 PM Iain Sandoe <iain@sandoe.co.uk> wrote:
>
> Hi Uros,
>
> > On 5 Dec 2022, at 21:07, Uros Bizjak via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
> >
> > On Mon, Dec 5, 2022 at 3:54 PM Iain Sandoe <iain@sandoe.co.uk> wrote:
> >>
> >> Hi Uros,
> >>
> >>> On 5 Dec 2022, at 10:37, Uros Bizjak via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
> >>>
> >>> On Sun, Dec 4, 2022 at 9:30 PM Iain Sandoe <iain@sandoe.co.uk> wrote:
> >>>>
> >>
> >>>> gcc/testsuite/ChangeLog:
> >>>>
> >>>> * gcc.target/x86_64/abi/bf16/args.h: Make xmm_regs, x87_regs extern.
> >>>> * gcc.target/x86_64/abi/bf16/m256bf16/args.h: Likewise.
> >>>> * gcc.target/x86_64/abi/bf16/m512bf16/args.h: Likewise.
> >>>> * gcc.target/x86_64/abi/bf16/asm-support.S: Add Mach-O variant.
> >>>> * gcc.target/x86_64/abi/bf16/m256bf16/asm-support.S: Likewise.
> >>>> * gcc.target/x86_64/abi/bf16/m512bf16/asm-support.S: Likewise.
> >>>
> >>> Please note that in other directories asm-support-darwin.s is
> >>> introduced and included via .exp file. Is there a reason a different
> >>> approach is introduced here?
> >>
> >> Since it seems that testcases get added and amended without considering any
> >> sub-target apart from x86_64-linux-gnu (even by very experienced contributors),
> >> I was hoping that the Darwin section might prompt folks to remember that there
> >> are several other sub-targets.
> >>
> >> However, the main thing is to fix the tests .. so here’s a version using separate
> >> files.
> >
> > extern void (*callthis)(void);
> > extern unsigned long long
> > rax,rbx,rcx,rdx,rsi,rdi,rsp,rbp,r8,r9,r10,r11,r12,r13,r14,r15;
> > -XMM_T xmm_regs[16];
> > -X87_T x87_regs[8];
> > +extern XMM_T xmm_regs[16];
> > +extern X87_T x87_regs[8];
> >
> > Do you still need this change? Existing test files are compiled without extern.
> >
> > + .globl _callthis
> > + .zerofill __DATA,__bss,_callthis,8,3
> > + .globl _rax
> > + .zerofill __DATA,__bss,_rax,8,3
> > + .globl _rbx
> > + .zerofill __DATA,__bss,_rbx,8,3
> > + .globl _rcx
> > + .zerofill __DATA,__bss,_rcx,8,3
> > + .globl _rdx
> > + .zerofill __DATA,__bss,_rdx,8,3
> > ...
> >
> > I wonder if the above approach is better than existing:
> >
> > .comm _callthis,8
> > .comm _rax,8
> > .comm _rbx,8
> > .comm _rcx,8
> > .comm _rdx,8
> > ...
>
> As noted in the changelog, direct access to common data is not permitted in the Darwin
> ABI [for x86_64, it would need to be _xxx@GOTPCREL(%rip)..] that’s why these have
> been moved to bss.
Thanks for the explanation!
The patch is OK.
> > It is strange to have two different approaches for similar tests. If
> > the new approach is better, we should also change existing asm-support
> > files.
>
> could be, I have not checked other case so far (extremely limited time at the moment)
>
> Quite likely, the accesses work in the testcases, despite violating the ABI.
This is never a good sign, it will break sooner or later...
Thanks,
Uros.
@@ -42,8 +42,8 @@ typedef union {
} X87_T;
extern void (*callthis)(void);
extern unsigned long long rax,rbx,rcx,rdx,rsi,rdi,rsp,rbp,r8,r9,r10,r11,r12,r13,r14,r15;
-XMM_T xmm_regs[16];
-X87_T x87_regs[8];
+extern XMM_T xmm_regs[16];
+extern X87_T x87_regs[8];
extern volatile unsigned long long volatile_var;
extern void snapshot (void);
extern void snapshot_ret (void);
@@ -1,3 +1,5 @@
+
+#ifdef __ELF__
.text
.p2align 4,,15
.globl snapshot
@@ -82,3 +84,106 @@ snapshot_ret:
#ifdef __linux__
.section .note.GNU-stack,"",@progbits
#endif
+
+#elif defined(__APPLE__)
+ .text
+ .p2align 4,,15
+ .globl _snapshot
+_snapshot:
+LFB3:
+ movq %rax, _rax(%rip)
+ movq %rbx, _rbx(%rip)
+ movq %rcx, _rcx(%rip)
+ movq %rdx, _rdx(%rip)
+ movq %rdi, _rdi(%rip)
+ movq %rsi, _rsi(%rip)
+ movq %rbp, _rbp(%rip)
+ movq %rsp, _rsp(%rip)
+ movq %r8, _r8(%rip)
+ movq %r9, _r9(%rip)
+ movq %r10, _r10(%rip)
+ movq %r11, _r11(%rip)
+ movq %r12, _r12(%rip)
+ movq %r13, _r13(%rip)
+ movq %r14, _r14(%rip)
+ movq %r15, _r15(%rip)
+ movdqu %xmm0, _xmm_regs+0(%rip)
+ movdqu %xmm1, _xmm_regs+16(%rip)
+ movdqu %xmm2, _xmm_regs+32(%rip)
+ movdqu %xmm3, _xmm_regs+48(%rip)
+ movdqu %xmm4, _xmm_regs+64(%rip)
+ movdqu %xmm5, _xmm_regs+80(%rip)
+ movdqu %xmm6, _xmm_regs+96(%rip)
+ movdqu %xmm7, _xmm_regs+112(%rip)
+ movdqu %xmm8, _xmm_regs+128(%rip)
+ movdqu %xmm9, _xmm_regs+144(%rip)
+ movdqu %xmm10, _xmm_regs+160(%rip)
+ movdqu %xmm11, _xmm_regs+176(%rip)
+ movdqu %xmm12, _xmm_regs+192(%rip)
+ movdqu %xmm13, _xmm_regs+208(%rip)
+ movdqu %xmm14, _xmm_regs+224(%rip)
+ movdqu %xmm15, _xmm_regs+240(%rip)
+ jmp *_callthis(%rip)
+LFE3:
+
+ .p2align 4,,15
+ .globl _snapshot_ret
+_snapshot_ret:
+ movq %rdi, _rdi(%rip)
+ subq $8, %rsp
+ call *_callthis(%rip)
+ addq $8, %rsp
+ movq %rax, _rax(%rip)
+ movq %rdx, _rdx(%rip)
+ movdqu %xmm0, _xmm_regs+0(%rip)
+ movdqu %xmm1, _xmm_regs+16(%rip)
+ fstpt _x87_regs(%rip)
+ fstpt _x87_regs+16(%rip)
+ fldt _x87_regs+16(%rip)
+ fldt _x87_regs(%rip)
+ ret
+
+ .globl _callthis
+ .zerofill __DATA,__bss,_callthis,8,3
+ .globl _rax
+ .zerofill __DATA,__bss,_rax,8,3
+ .globl _rbx
+ .zerofill __DATA,__bss,_rbx,8,3
+ .globl _rcx
+ .zerofill __DATA,__bss,_rcx,8,3
+ .globl _rdx
+ .zerofill __DATA,__bss,_rdx,8,3
+ .globl _rsi
+ .zerofill __DATA,__bss,_rsi,8,3
+ .globl _rdi
+ .zerofill __DATA,__bss,_rdi,8,3
+ .globl _rsp
+ .zerofill __DATA,__bss,_rsp,8,3
+ .globl _rbp
+ .zerofill __DATA,__bss,_rbp,8,3
+ .globl _r8
+ .zerofill __DATA,__bss,_r8,8,3
+ .globl _r9
+ .zerofill __DATA,__bss,_r9,8,3
+ .globl _r10
+ .zerofill __DATA,__bss,_r10,8,3
+ .globl _r11
+ .zerofill __DATA,__bss,_r11,8,3
+ .globl _r12
+ .zerofill __DATA,__bss,_r12,8,3
+ .globl _r13
+ .zerofill __DATA,__bss,_r13,8,3
+ .globl _r14
+ .zerofill __DATA,__bss,_r14,8,3
+ .globl _r15
+ .zerofill __DATA,__bss,_r15,8,3
+ .globl _xmm_regs
+ .zerofill __DATA,__bss,_xmm_regs,256,5
+ .globl _x87_regs
+ .zerofill __DATA,__bss,_x87_regs,128,5
+ .globl _volatile_var
+ .zerofill __DATA,__bss,_volatile_var,8,3
+
+#else
+#error unknown object format
+#endif
@@ -50,8 +50,8 @@ typedef union {
} X87_T;
extern void (*callthis)(void);
extern unsigned long long rax,rbx,rcx,rdx,rsi,rdi,rsp,rbp,r8,r9,r10,r11,r12,r13,r14,r15;
-YMM_T ymm_regs[16];
-X87_T x87_regs[8];
+extern YMM_T ymm_regs[16];
+extern X87_T x87_regs[8];
extern volatile unsigned long long volatile_var;
extern void snapshot (void);
extern void snapshot_ret (void);
@@ -1,3 +1,5 @@
+
+#ifdef __ELF__
.text
.p2align 4,,15
.globl snapshot
@@ -82,3 +84,108 @@ snapshot_ret:
#ifdef __linux__
.section .note.GNU-stack,"",@progbits
#endif
+
+#elif defined(__APPLE__)
+
+ .text
+ .p2align 4,,15
+ .globl _snapshot
+_snapshot:
+.LFB3:
+ movq %rax, _rax(%rip)
+ movq %rbx, _rbx(%rip)
+ movq %rcx, _rcx(%rip)
+ movq %rdx, _rdx(%rip)
+ movq %rdi, _rdi(%rip)
+ movq %rsi, _rsi(%rip)
+ movq %rbp, _rbp(%rip)
+ movq %rsp, _rsp(%rip)
+ movq %r8, _r8(%rip)
+ movq %r9, _r9(%rip)
+ movq %r10, _r10(%rip)
+ movq %r11, _r11(%rip)
+ movq %r12, _r12(%rip)
+ movq %r13, _r13(%rip)
+ movq %r14, _r14(%rip)
+ movq %r15, _r15(%rip)
+ vmovdqu %ymm0, _ymm_regs+0(%rip)
+ vmovdqu %ymm1, _ymm_regs+32(%rip)
+ vmovdqu %ymm2, _ymm_regs+64(%rip)
+ vmovdqu %ymm3, _ymm_regs+96(%rip)
+ vmovdqu %ymm4, _ymm_regs+128(%rip)
+ vmovdqu %ymm5, _ymm_regs+160(%rip)
+ vmovdqu %ymm6, _ymm_regs+192(%rip)
+ vmovdqu %ymm7, _ymm_regs+224(%rip)
+ vmovdqu %ymm8, _ymm_regs+256(%rip)
+ vmovdqu %ymm9, _ymm_regs+288(%rip)
+ vmovdqu %ymm10, _ymm_regs+320(%rip)
+ vmovdqu %ymm11, _ymm_regs+352(%rip)
+ vmovdqu %ymm12, _ymm_regs+384(%rip)
+ vmovdqu %ymm13, _ymm_regs+416(%rip)
+ vmovdqu %ymm14, _ymm_regs+448(%rip)
+ vmovdqu %ymm15, _ymm_regs+480(%rip)
+ jmp *_callthis(%rip)
+.LFE3:
+
+ .p2align 4,,15
+ .globl _snapshot_ret
+_snapshot_ret:
+ movq %rdi, _rdi(%rip)
+ subq $8, %rsp
+ call *_callthis(%rip)
+ addq $8, %rsp
+ movq %rax, _rax(%rip)
+ movq %rdx, _rdx(%rip)
+ vmovdqu %ymm0, _ymm_regs+0(%rip)
+ vmovdqu %ymm1, _ymm_regs+32(%rip)
+ fstpt _x87_regs(%rip)
+ fstpt _x87_regs+16(%rip)
+ fldt _x87_regs+16(%rip)
+ fldt _x87_regs(%rip)
+ ret
+
+ .globl _callthis
+ .zerofill __DATA,__bss,_callthis,8,3
+ .globl _rax
+ .zerofill __DATA,__bss,_rax,8,3
+ .globl _rbx
+ .zerofill __DATA,__bss,_rbx,8,3
+ .globl _rcx
+ .zerofill __DATA,__bss,_rcx,8,3
+ .globl _rdx
+ .zerofill __DATA,__bss,_rdx,8,3
+ .globl _rsi
+ .zerofill __DATA,__bss,_rsi,8,3
+ .globl _rdi
+ .zerofill __DATA,__bss,_rdi,8,3
+ .globl _rsp
+ .zerofill __DATA,__bss,_rsp,8,3
+ .globl _rbp
+ .zerofill __DATA,__bss,_rbp,8,3
+ .globl _r8
+ .zerofill __DATA,__bss,_r8,8,3
+ .globl _r9
+ .zerofill __DATA,__bss,_r9,8,3
+ .globl _r10
+ .zerofill __DATA,__bss,_r10,8,3
+ .globl _r11
+ .zerofill __DATA,__bss,_r11,8,3
+ .globl _r12
+ .zerofill __DATA,__bss,_r12,8,3
+ .globl _r13
+ .zerofill __DATA,__bss,_r13,8,3
+ .globl _r14
+ .zerofill __DATA,__bss,_r14,8,3
+ .globl _r15
+ .zerofill __DATA,__bss,_r15,8,3
+ .globl _ymm_regs
+ .zerofill __DATA,__bss,_ymm_regs,512,5
+ .globl _x87_regs
+ .zerofill __DATA,__bss,_x87_regs,128,5
+ .globl _volatile_var
+ .zerofill __DATA,__bss,_volatile_var,8,3
+
+#else
+#error unknown object format
+#endif
+
@@ -51,8 +51,8 @@ typedef union {
} X87_T;
extern void (*callthis)(void);
extern unsigned long long rax,rbx,rcx,rdx,rsi,rdi,rsp,rbp,r8,r9,r10,r11,r12,r13,r14,r15;
-ZMM_T zmm_regs[32];
-X87_T x87_regs[8];
+extern ZMM_T zmm_regs[32];
+extern X87_T x87_regs[8];
extern volatile unsigned long long volatile_var;
extern void snapshot (void);
extern void snapshot_ret (void);
@@ -1,3 +1,5 @@
+
+#ifdef __ELF__
.text
.p2align 4,,15
.globl snapshot
@@ -98,3 +100,124 @@ snapshot_ret:
#ifdef __linux__
.section .note.GNU-stack,"",@progbits
#endif
+
+#elif defined(__APPLE__)
+
+ .text
+ .p2align 4,,15
+ .globl _snapshot
+_snapshot:
+.LFB3:
+ movq %rax, _rax(%rip)
+ movq %rbx, _rbx(%rip)
+ movq %rcx, _rcx(%rip)
+ movq %rdx, _rdx(%rip)
+ movq %rdi, _rdi(%rip)
+ movq %rsi, _rsi(%rip)
+ movq %rbp, _rbp(%rip)
+ movq %rsp, _rsp(%rip)
+ movq %r8, _r8(%rip)
+ movq %r9, _r9(%rip)
+ movq %r10, _r10(%rip)
+ movq %r11, _r11(%rip)
+ movq %r12, _r12(%rip)
+ movq %r13, _r13(%rip)
+ movq %r14, _r14(%rip)
+ movq %r15, _r15(%rip)
+ vmovdqu32 %zmm0, _zmm_regs+0(%rip)
+ vmovdqu32 %zmm1, _zmm_regs+64(%rip)
+ vmovdqu32 %zmm2, _zmm_regs+128(%rip)
+ vmovdqu32 %zmm3, _zmm_regs+192(%rip)
+ vmovdqu32 %zmm4, _zmm_regs+256(%rip)
+ vmovdqu32 %zmm5, _zmm_regs+320(%rip)
+ vmovdqu32 %zmm6, _zmm_regs+384(%rip)
+ vmovdqu32 %zmm7, _zmm_regs+448(%rip)
+ vmovdqu32 %zmm8, _zmm_regs+512(%rip)
+ vmovdqu32 %zmm9, _zmm_regs+576(%rip)
+ vmovdqu32 %zmm10, _zmm_regs+640(%rip)
+ vmovdqu32 %zmm11, _zmm_regs+704(%rip)
+ vmovdqu32 %zmm12, _zmm_regs+768(%rip)
+ vmovdqu32 %zmm13, _zmm_regs+832(%rip)
+ vmovdqu32 %zmm14, _zmm_regs+896(%rip)
+ vmovdqu32 %zmm15, _zmm_regs+960(%rip)
+ vmovdqu32 %zmm16, _zmm_regs+1024(%rip)
+ vmovdqu32 %zmm17, _zmm_regs+1088(%rip)
+ vmovdqu32 %zmm18, _zmm_regs+1152(%rip)
+ vmovdqu32 %zmm19, _zmm_regs+1216(%rip)
+ vmovdqu32 %zmm20, _zmm_regs+1280(%rip)
+ vmovdqu32 %zmm21, _zmm_regs+1344(%rip)
+ vmovdqu32 %zmm22, _zmm_regs+1408(%rip)
+ vmovdqu32 %zmm23, _zmm_regs+1472(%rip)
+ vmovdqu32 %zmm24, _zmm_regs+1536(%rip)
+ vmovdqu32 %zmm25, _zmm_regs+1600(%rip)
+ vmovdqu32 %zmm26, _zmm_regs+1664(%rip)
+ vmovdqu32 %zmm27, _zmm_regs+1728(%rip)
+ vmovdqu32 %zmm28, _zmm_regs+1792(%rip)
+ vmovdqu32 %zmm29, _zmm_regs+1856(%rip)
+ vmovdqu32 %zmm30, _zmm_regs+1920(%rip)
+ vmovdqu32 %zmm31, _zmm_regs+1984(%rip)
+ jmp *_callthis(%rip)
+.LFE3:
+
+ .p2align 4,,15
+ .globl _snapshot_ret
+_snapshot_ret:
+ movq %rdi, _rdi(%rip)
+ subq $8, %rsp
+ call *_callthis(%rip)
+ addq $8, %rsp
+ movq %rax, _rax(%rip)
+ movq %rdx, _rdx(%rip)
+ vmovdqu32 %zmm0, _zmm_regs+0(%rip)
+ vmovdqu32 %zmm1, _zmm_regs+64(%rip)
+ fstpt _x87_regs(%rip)
+ fstpt _x87_regs+16(%rip)
+ fldt _x87_regs+16(%rip)
+ fldt _x87_regs(%rip)
+ ret
+
+ .globl _callthis
+ .zerofill __DATA,__bss,_callthis,8,3
+ .globl _rax
+ .zerofill __DATA,__bss,_rax,8,3
+ .globl _rbx
+ .zerofill __DATA,__bss,_rbx,8,3
+ .globl _rcx
+ .zerofill __DATA,__bss,_rcx,8,3
+ .globl _rdx
+ .zerofill __DATA,__bss,_rdx,8,3
+ .globl _rsi
+ .zerofill __DATA,__bss,_rsi,8,3
+ .globl _rdi
+ .zerofill __DATA,__bss,_rdi,8,3
+ .globl _rsp
+ .zerofill __DATA,__bss,_rsp,8,3
+ .globl _rbp
+ .zerofill __DATA,__bss,_rbp,8,3
+ .globl _r8
+ .zerofill __DATA,__bss,_r8,8,3
+ .globl _r9
+ .zerofill __DATA,__bss,_r9,8,3
+ .globl _r10
+ .zerofill __DATA,__bss,_r10,8,3
+ .globl _r11
+ .zerofill __DATA,__bss,_r11,8,3
+ .globl _r12
+ .zerofill __DATA,__bss,_r12,8,3
+ .globl _r13
+ .zerofill __DATA,__bss,_r13,8,3
+ .globl _r14
+ .zerofill __DATA,__bss,_r14,8,3
+ .globl _r15
+ .zerofill __DATA,__bss,_r15,8,3
+ .globl _zmm_regs
+ .zerofill __DATA,__bss,_zmm_regs,2048,6
+ .globl _x87_regs
+ .zerofill __DATA,__bss,_x87_regs,128,5
+ .globl _volatile_var
+ .zerofill __DATA,__bss,_volatile_var,8,3
+
+#else
+#error unknown object format
+#endif
+