[libunwind] Use .irp directives. NFC

[MaskRay]

Summary

The repeated instructions make the file long and difficult to read.
Simplify them with .irp directives.

Skip PowerPC since AIX assembler doesn't support .irp

[MaskRay]

update more

[compnerd]

This is definitely a nice cleanup. There is a way to use .irp to have it count, but this is fine for now.

[MaskRay]

In D139368#3975014, @compnerd wrote:

This is definitely a nice cleanup. There is a way to use .irp to have it count, but this is fine for now.

Thanks. Unfortunately, AFAIK, MACRO-11 introduced .irp i,0,1,2,3 and .irpc i,0123 but no generalized iteration from X to Y... GNU as and LLVM MC inherited the directives from MACRO-11 but do not add a generalized iteration...

[daltenty]

This broke the AIX CI: https://buildkite.com/llvm-project/libcxx-ci/builds/15656#0184eac6-521b-4d1e-8502-c226673e0ece. Unfortunately it seems the .irp directive isn't supported by the AIX assembler.

[ldionne]

Did the libc++/libc++abi/libunwind CI run on this patch? I thought it should, but if not, we can get that setup. Otherwise, I think this is a gentle reminder for folks not to ignore CI failures :-)

[MaskRay]

In D139368#3981594, @ldionne wrote:

Did the libc++/libc++abi/libunwind CI run on this patch? I thought it should, but if not, we can get that setup. Otherwise, I think this is a gentle reminder for folks not to ignore CI failures :-)

I don't know a libunwind CI. I personally checked the output is the same with multiple architectures, x86_64, aarch64, arm, riscv64, ppc32 linux, ppc64 linux, etc, but I couldn't have known that this did with AIX.
Having a CI set up will provide some value but likely not very useful for many libunwind changes, but nevertheless, it is good if it can check what AIX and z/OS do since most contributors don't have access to the two.

[MaskRay]

In D139368#3978330, @daltenty wrote:

This broke the AIX CI: https://buildkite.com/llvm-project/libcxx-ci/builds/15656#0184eac6-521b-4d1e-8502-c226673e0ece. Unfortunately it seems the .irp directive isn't supported by the AIX assembler.

What chunks of code do AIX use? All __powerpc__ and __powerpc64__ with all variants including __VSX__?

[xingxue]

In D139368#3994092, @MaskRay wrote:

In D139368#3978330, @daltenty wrote:

This broke the AIX CI: https://buildkite.com/llvm-project/libcxx-ci/builds/15656#0184eac6-521b-4d1e-8502-c226673e0ece. Unfortunately it seems the .irp directive isn't supported by the AIX assembler.

What chunks of code do AIX use? All __powerpc__ and __powerpc64__ with all variants including __VSX__?

Both chunks under __powerpc__ (32-bit mode) and __powerpc64__ (64-bit mode) are used for AIX. For __powerpc__, code guarded by __ALTIVEC__ is used. For __powerpc64__, code guarded by __VSX__ is used.

libunwind/src/UnwindRegistersRestore.S

 //===----------------------------------------------------------------------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 #include "assembly.h"
 
+#define FROM_0_TO_15 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+#define FROM_16_TO_31 16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+
+#define FROM_0_TO_31 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+#define FROM_32_TO_63 32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63
+
 #if defined(_AIX)
   .toc
 #else
   .text
 #endif
 
 #if !defined(__USING_SJLJ_EXCEPTIONS__)
 
#define PPC64_LR(n) \
 
   // restore integral registers
   // skip r0 for now
   // skip r1 for now
   PPC64_LR(2)
   // skip r3 for now
   // skip r4 for now
   // skip r5 for now
-  PPC64_LR(6)
-  PPC64_LR(7)
-  PPC64_LR(8)
-  PPC64_LR(9)
-  PPC64_LR(10)
-  PPC64_LR(11)
-  PPC64_LR(12)
-  PPC64_LR(13)
-  PPC64_LR(14)
-  PPC64_LR(15)
-  PPC64_LR(16)
-  PPC64_LR(17)
-  PPC64_LR(18)
-  PPC64_LR(19)
-  PPC64_LR(20)
-  PPC64_LR(21)
-  PPC64_LR(22)
-  PPC64_LR(23)
-  PPC64_LR(24)
-  PPC64_LR(25)
-  PPC64_LR(26)
-  PPC64_LR(27)
-  PPC64_LR(28)
-  PPC64_LR(29)
-  PPC64_LR(30)
-  PPC64_LR(31)
+  .irp i,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+    PPC64_LR(\i)
+  .endr
 
 #if defined(__VSX__)
 
   // restore VS registers
   // (note that this also restores floating point registers and V registers,
   // because part of VS is mapped to these registers)
 
   addi  4, 3, PPC64_OFFS_FP
 
 // load VS register
 #define PPC64_LVS(n)         \
   lxvd2x  n, 0, 4           ;\
   addi    4, 4, 16
 
   // restore the first 32 VS regs (and also all floating point regs)
-  PPC64_LVS(0)
-  PPC64_LVS(1)
-  PPC64_LVS(2)
-  PPC64_LVS(3)
-  PPC64_LVS(4)
-  PPC64_LVS(5)
-  PPC64_LVS(6)
-  PPC64_LVS(7)
-  PPC64_LVS(8)
-  PPC64_LVS(9)
-  PPC64_LVS(10)
-  PPC64_LVS(11)
-  PPC64_LVS(12)
-  PPC64_LVS(13)
-  PPC64_LVS(14)
-  PPC64_LVS(15)
-  PPC64_LVS(16)
-  PPC64_LVS(17)
-  PPC64_LVS(18)
-  PPC64_LVS(19)
-  PPC64_LVS(20)
-  PPC64_LVS(21)
-  PPC64_LVS(22)
-  PPC64_LVS(23)
-  PPC64_LVS(24)
-  PPC64_LVS(25)
-  PPC64_LVS(26)
-  PPC64_LVS(27)
-  PPC64_LVS(28)
-  PPC64_LVS(29)
-  PPC64_LVS(30)
-  PPC64_LVS(31)
+  .irp i,FROM_0_TO_31
+    PPC64_LVS(\i)
+  .endr
 
 #define PPC64_CLVS_RESTORE(n)                    \
   addi   4, 3, PPC64_OFFS_FP + n * 16           ;\
   lxvd2x n, 0, 4
 
 #if !defined(_AIX)
   // use VRSAVE to conditionally restore the remaining VS regs, that are
   // where the V regs are mapped. In the AIX ABI, VRSAVE is not used.

 
 #else
 
 #define PPC64_CLVSl(n) PPC64_CLVS_RESTORE(n)
 #define PPC64_CLVSh(n) PPC64_CLVS_RESTORE(n)
 
 #endif // !defined(_AIX)
 
-  PPC64_CLVSl(32)
-  PPC64_CLVSl(33)
-  PPC64_CLVSl(34)
-  PPC64_CLVSl(35)
-  PPC64_CLVSl(36)
-  PPC64_CLVSl(37)
-  PPC64_CLVSl(38)
-  PPC64_CLVSl(39)
-  PPC64_CLVSl(40)
-  PPC64_CLVSl(41)
-  PPC64_CLVSl(42)
-  PPC64_CLVSl(43)
-  PPC64_CLVSl(44)
-  PPC64_CLVSl(45)
-  PPC64_CLVSl(46)
-  PPC64_CLVSl(47)
-  PPC64_CLVSh(48)
-  PPC64_CLVSh(49)
-  PPC64_CLVSh(50)
-  PPC64_CLVSh(51)
-  PPC64_CLVSh(52)
-  PPC64_CLVSh(53)
-  PPC64_CLVSh(54)
-  PPC64_CLVSh(55)
-  PPC64_CLVSh(56)
-  PPC64_CLVSh(57)
-  PPC64_CLVSh(58)
-  PPC64_CLVSh(59)
-  PPC64_CLVSh(60)
-  PPC64_CLVSh(61)
-  PPC64_CLVSh(62)
-  PPC64_CLVSh(63)
+  .irp i,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47
+    PPC64_CLVSl(\i)
+  .endr
+  .irp i,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63
+    PPC64_CLVSh(\i)
+  .endr
 
 #else
 
 // load FP register
 #define PPC64_LF(n) \
   lfd   n, (PPC64_OFFS_FP + n * 16)(3)
 
   // restore float registers
-  PPC64_LF(0)
-  PPC64_LF(1)
-  PPC64_LF(2)
-  PPC64_LF(3)
-  PPC64_LF(4)
-  PPC64_LF(5)
-  PPC64_LF(6)
-  PPC64_LF(7)
-  PPC64_LF(8)
-  PPC64_LF(9)
-  PPC64_LF(10)
-  PPC64_LF(11)
-  PPC64_LF(12)
-  PPC64_LF(13)
-  PPC64_LF(14)
-  PPC64_LF(15)
-  PPC64_LF(16)
-  PPC64_LF(17)
-  PPC64_LF(18)
-  PPC64_LF(19)
-  PPC64_LF(20)
-  PPC64_LF(21)
-  PPC64_LF(22)
-  PPC64_LF(23)
-  PPC64_LF(24)
-  PPC64_LF(25)
-  PPC64_LF(26)
-  PPC64_LF(27)
-  PPC64_LF(28)
-  PPC64_LF(29)
-  PPC64_LF(30)
-  PPC64_LF(31)
+  .irp i,FROM_0_TO_31
+    PPC64_LF(\i)
+  .endr
 
 #if defined(__ALTIVEC__)
 
 #define PPC64_CLV_UNALIGNED_RESTORE(n)       \
   ld     0, (PPC64_OFFS_V + n * 16)(3)      ;\
   std    0, 0(4)                            ;\
   ld     0, (PPC64_OFFS_V + n * 16 + 8)(3)  ;\
   std    0, 8(4)                            ;\

 
 #endif // !defined(_AIX)
 
   subi  4, 1, 16
   // r4 is now a 16-byte aligned pointer into the red zone
   // the _vectorScalarRegisters may not be 16-byte aligned
   // so copy via red zone temp buffer
 
-  PPC64_CLV_UNALIGNEDl(0)
-  PPC64_CLV_UNALIGNEDl(1)
-  PPC64_CLV_UNALIGNEDl(2)
-  PPC64_CLV_UNALIGNEDl(3)
-  PPC64_CLV_UNALIGNEDl(4)
-  PPC64_CLV_UNALIGNEDl(5)
-  PPC64_CLV_UNALIGNEDl(6)
-  PPC64_CLV_UNALIGNEDl(7)
-  PPC64_CLV_UNALIGNEDl(8)
-  PPC64_CLV_UNALIGNEDl(9)
-  PPC64_CLV_UNALIGNEDl(10)
-  PPC64_CLV_UNALIGNEDl(11)
-  PPC64_CLV_UNALIGNEDl(12)
-  PPC64_CLV_UNALIGNEDl(13)
-  PPC64_CLV_UNALIGNEDl(14)
-  PPC64_CLV_UNALIGNEDl(15)
-  PPC64_CLV_UNALIGNEDh(16)
-  PPC64_CLV_UNALIGNEDh(17)
-  PPC64_CLV_UNALIGNEDh(18)
-  PPC64_CLV_UNALIGNEDh(19)
-  PPC64_CLV_UNALIGNEDh(20)
-  PPC64_CLV_UNALIGNEDh(21)
-  PPC64_CLV_UNALIGNEDh(22)
-  PPC64_CLV_UNALIGNEDh(23)
-  PPC64_CLV_UNALIGNEDh(24)
-  PPC64_CLV_UNALIGNEDh(25)
-  PPC64_CLV_UNALIGNEDh(26)
-  PPC64_CLV_UNALIGNEDh(27)
-  PPC64_CLV_UNALIGNEDh(28)
-  PPC64_CLV_UNALIGNEDh(29)
-  PPC64_CLV_UNALIGNEDh(30)
-  PPC64_CLV_UNALIGNEDh(31)
+  .irp i,FROM_0_TO_15
+    PPC64_CLV_UNALIGNEDl(\i)
+  .endr
+  .irp i,FROM_16_TO_31
+    PPC64_CLV_UNALIGNEDh(\i)
+  .endr
 
 #endif
 #endif
 
 Lnovec:
   ld    0, PPC64_OFFS_CR(3)
   mtcr  0
   ld    0, PPC64_OFFS_SRR0(3)
//
 
   // restore integral registers
   // skip r0 for now
   // skip r1 for now
   lwz     2,  16(3)
   // skip r3 for now
   // skip r4 for now
   // skip r5 for now
-  lwz     6,  32(3)
-  lwz     7,  36(3)
-  lwz     8,  40(3)
-  lwz     9,  44(3)
-  lwz     10, 48(3)
-  lwz     11, 52(3)
-  lwz     12, 56(3)
-  lwz     13, 60(3)
-  lwz     14, 64(3)
-  lwz     15, 68(3)
-  lwz     16, 72(3)
-  lwz     17, 76(3)
-  lwz     18, 80(3)
-  lwz     19, 84(3)
-  lwz     20, 88(3)
-  lwz     21, 92(3)
-  lwz     22, 96(3)
-  lwz     23,100(3)
-  lwz     24,104(3)
-  lwz     25,108(3)
-  lwz     26,112(3)
-  lwz     27,116(3)
-  lwz     28,120(3)
-  lwz     29,124(3)
-  lwz     30,128(3)
-  lwz     31,132(3)
+  .irp i,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+    lwz \i, (8+4*\i)(3)
+  .endr
 
 #ifndef __NO_FPRS__
   // restore float registers
-  lfd     0, 160(3)
-  lfd     1, 168(3)
-  lfd     2, 176(3)
-  lfd     3, 184(3)
-  lfd     4, 192(3)
-  lfd     5, 200(3)
-  lfd     6, 208(3)
-  lfd     7, 216(3)
-  lfd     8, 224(3)
-  lfd     9, 232(3)
-  lfd     10,240(3)
-  lfd     11,248(3)
-  lfd     12,256(3)
-  lfd     13,264(3)
-  lfd     14,272(3)
-  lfd     15,280(3)
-  lfd     16,288(3)
-  lfd     17,296(3)
-  lfd     18,304(3)
-  lfd     19,312(3)
-  lfd     20,320(3)
-  lfd     21,328(3)
-  lfd     22,336(3)
-  lfd     23,344(3)
-  lfd     24,352(3)
-  lfd     25,360(3)
-  lfd     26,368(3)
-  lfd     27,376(3)
-  lfd     28,384(3)
-  lfd     29,392(3)
-  lfd     30,400(3)
-  lfd     31,408(3)
+  .irp i,FROM_0_TO_31
+    lfd \i, (160+8*\i)(3)
+  .endr
 #endif
 
 #if defined(__ALTIVEC__)
 
 #define LOAD_VECTOR_RESTORE(_index)                 \
   lwz     0, 424+_index*16(3)             SEPARATOR \
   stw     0, 0(4)                         SEPARATOR \
   lwz     0, 424+_index*16+4(3)           SEPARATOR \

 
 #endif // !defined(_AIX)
 
   subi    4, 1, 16
   rlwinm  4, 4, 0, 0, 27  // mask low 4-bits
   // r4 is now a 16-byte aligned pointer into the red zone
   // the _vectorRegisters may not be 16-byte aligned so copy via red zone temp buffer
 
-  LOAD_VECTOR_UNALIGNEDl(0)
-  LOAD_VECTOR_UNALIGNEDl(1)
-  LOAD_VECTOR_UNALIGNEDl(2)
-  LOAD_VECTOR_UNALIGNEDl(3)
-  LOAD_VECTOR_UNALIGNEDl(4)
-  LOAD_VECTOR_UNALIGNEDl(5)
-  LOAD_VECTOR_UNALIGNEDl(6)
-  LOAD_VECTOR_UNALIGNEDl(7)
-  LOAD_VECTOR_UNALIGNEDl(8)
-  LOAD_VECTOR_UNALIGNEDl(9)
-  LOAD_VECTOR_UNALIGNEDl(10)
-  LOAD_VECTOR_UNALIGNEDl(11)
-  LOAD_VECTOR_UNALIGNEDl(12)
-  LOAD_VECTOR_UNALIGNEDl(13)
-  LOAD_VECTOR_UNALIGNEDl(14)
-  LOAD_VECTOR_UNALIGNEDl(15)
-  LOAD_VECTOR_UNALIGNEDh(16)
-  LOAD_VECTOR_UNALIGNEDh(17)
-  LOAD_VECTOR_UNALIGNEDh(18)
-  LOAD_VECTOR_UNALIGNEDh(19)
-  LOAD_VECTOR_UNALIGNEDh(20)
-  LOAD_VECTOR_UNALIGNEDh(21)
-  LOAD_VECTOR_UNALIGNEDh(22)
-  LOAD_VECTOR_UNALIGNEDh(23)
-  LOAD_VECTOR_UNALIGNEDh(24)
-  LOAD_VECTOR_UNALIGNEDh(25)
-  LOAD_VECTOR_UNALIGNEDh(26)
-  LOAD_VECTOR_UNALIGNEDh(27)
-  LOAD_VECTOR_UNALIGNEDh(28)
-  LOAD_VECTOR_UNALIGNEDh(29)
-  LOAD_VECTOR_UNALIGNEDh(30)
-  LOAD_VECTOR_UNALIGNEDh(31)
+  .irp i,FROM_0_TO_15
+    LOAD_VECTOR_UNALIGNEDl(\i)
+  .endr
+  .irp i,FROM_16_TO_31
+    LOAD_VECTOR_UNALIGNEDh(\i)
+  .endr
 #endif
 
 Lnovec:
   lwz     0, 136(3)   // __cr
   mtcr    0
   lwz     0, 148(3)   // __ctr
   mtctr   0
   lwz     0,   0(3)   // __ssr0

 //  thread state pointer is in a0 ($4)
 //
 DEFINE_LIBUNWIND_FUNCTION(_ZN9libunwind21Registers_mips_newabi6jumptoEv)
   .set push
   .set noat
   .set noreorder
   .set nomacro
 #ifdef __mips_hard_float
-  ldc1  $f0, (8 * 35)($4)
-  ldc1  $f1, (8 * 36)($4)
-  ldc1  $f2, (8 * 37)($4)
-  ldc1  $f3, (8 * 38)($4)
-  ldc1  $f4, (8 * 39)($4)
-  ldc1  $f5, (8 * 40)($4)
-  ldc1  $f6, (8 * 41)($4)
-  ldc1  $f7, (8 * 42)($4)
-  ldc1  $f8, (8 * 43)($4)
-  ldc1  $f9, (8 * 44)($4)
-  ldc1  $f10, (8 * 45)($4)
-  ldc1  $f11, (8 * 46)($4)
-  ldc1  $f12, (8 * 47)($4)
-  ldc1  $f13, (8 * 48)($4)
-  ldc1  $f14, (8 * 49)($4)
-  ldc1  $f15, (8 * 50)($4)
-  ldc1  $f16, (8 * 51)($4)
-  ldc1  $f17, (8 * 52)($4)
-  ldc1  $f18, (8 * 53)($4)
-  ldc1  $f19, (8 * 54)($4)
-  ldc1  $f20, (8 * 55)($4)
-  ldc1  $f21, (8 * 56)($4)
-  ldc1  $f22, (8 * 57)($4)
-  ldc1  $f23, (8 * 58)($4)
-  ldc1  $f24, (8 * 59)($4)
-  ldc1  $f25, (8 * 60)($4)
-  ldc1  $f26, (8 * 61)($4)
-  ldc1  $f27, (8 * 62)($4)
-  ldc1  $f28, (8 * 63)($4)
-  ldc1  $f29, (8 * 64)($4)
-  ldc1  $f30, (8 * 65)($4)
-  ldc1  $f31, (8 * 66)($4)
+  .irp i,FROM_0_TO_31
+    ldc1 $f\i, (280+8*\i)($4)
+  .endr
 #endif
   // restore hi and lo
   ld    $8, (8 * 33)($4)
   mthi  $8
   ld    $8, (8 * 34)($4)
   mtlo  $8
   // r0 is zero
   ld    $1, (8 * 1)($4)
   ld    $2, (8 * 2)($4)
   ld    $3, (8 * 3)($4)
   // skip a0 for now
-  ld    $5, (8 * 5)($4)
-  ld    $6, (8 * 6)($4)
-  ld    $7, (8 * 7)($4)
-  ld    $8, (8 * 8)($4)
-  ld    $9, (8 * 9)($4)
-  ld    $10, (8 * 10)($4)
-  ld    $11, (8 * 11)($4)
-  ld    $12, (8 * 12)($4)
-  ld    $13, (8 * 13)($4)
-  ld    $14, (8 * 14)($4)
-  ld    $15, (8 * 15)($4)
-  ld    $16, (8 * 16)($4)
-  ld    $17, (8 * 17)($4)
-  ld    $18, (8 * 18)($4)
-  ld    $19, (8 * 19)($4)
-  ld    $20, (8 * 20)($4)
-  ld    $21, (8 * 21)($4)
-  ld    $22, (8 * 22)($4)
-  ld    $23, (8 * 23)($4)
-  ld    $24, (8 * 24)($4)
-  ld    $25, (8 * 25)($4)
-  ld    $26, (8 * 26)($4)
-  ld    $27, (8 * 27)($4)
-  ld    $28, (8 * 28)($4)
-  ld    $29, (8 * 29)($4)
-  ld    $30, (8 * 30)($4)
+  .irp i,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30
+    ld $\i, (8 * \i)($4)
+  .endr
   // load new pc into ra
   ld    $31, (8 * 32)($4)
   // jump to ra, load a0 in the delay slot
   jr    $31
   ld    $4, (8 * 4)($4)
   .set pop
 
 #elif defined(__sparc__) && defined(__arch64__)

 // void libunwind::Registers_riscv::jumpto()
 //
 // On entry:
 //  thread_state pointer is in a0
 //
   .p2align 2
 DEFINE_LIBUNWIND_FUNCTION(_ZN9libunwind15Registers_riscv6jumptoEv)
 # if defined(__riscv_flen)
-  FLOAD    f0, (RISCV_FOFFSET + RISCV_FSIZE * 0)(a0)
-  FLOAD    f1, (RISCV_FOFFSET + RISCV_FSIZE * 1)(a0)
-  FLOAD    f2, (RISCV_FOFFSET + RISCV_FSIZE * 2)(a0)
-  FLOAD    f3, (RISCV_FOFFSET + RISCV_FSIZE * 3)(a0)
-  FLOAD    f4, (RISCV_FOFFSET + RISCV_FSIZE * 4)(a0)
-  FLOAD    f5, (RISCV_FOFFSET + RISCV_FSIZE * 5)(a0)
-  FLOAD    f6, (RISCV_FOFFSET + RISCV_FSIZE * 6)(a0)
-  FLOAD    f7, (RISCV_FOFFSET + RISCV_FSIZE * 7)(a0)
-  FLOAD    f8, (RISCV_FOFFSET + RISCV_FSIZE * 8)(a0)
-  FLOAD    f9, (RISCV_FOFFSET + RISCV_FSIZE * 9)(a0)
-  FLOAD    f10, (RISCV_FOFFSET + RISCV_FSIZE * 10)(a0)
-  FLOAD    f11, (RISCV_FOFFSET + RISCV_FSIZE * 11)(a0)
-  FLOAD    f12, (RISCV_FOFFSET + RISCV_FSIZE * 12)(a0)
-  FLOAD    f13, (RISCV_FOFFSET + RISCV_FSIZE * 13)(a0)
-  FLOAD    f14, (RISCV_FOFFSET + RISCV_FSIZE * 14)(a0)
-  FLOAD    f15, (RISCV_FOFFSET + RISCV_FSIZE * 15)(a0)
-  FLOAD    f16, (RISCV_FOFFSET + RISCV_FSIZE * 16)(a0)
-  FLOAD    f17, (RISCV_FOFFSET + RISCV_FSIZE * 17)(a0)
-  FLOAD    f18, (RISCV_FOFFSET + RISCV_FSIZE * 18)(a0)
-  FLOAD    f19, (RISCV_FOFFSET + RISCV_FSIZE * 19)(a0)
-  FLOAD    f20, (RISCV_FOFFSET + RISCV_FSIZE * 20)(a0)
-  FLOAD    f21, (RISCV_FOFFSET + RISCV_FSIZE * 21)(a0)
-  FLOAD    f22, (RISCV_FOFFSET + RISCV_FSIZE * 22)(a0)
-  FLOAD    f23, (RISCV_FOFFSET + RISCV_FSIZE * 23)(a0)
-  FLOAD    f24, (RISCV_FOFFSET + RISCV_FSIZE * 24)(a0)
-  FLOAD    f25, (RISCV_FOFFSET + RISCV_FSIZE * 25)(a0)
-  FLOAD    f26, (RISCV_FOFFSET + RISCV_FSIZE * 26)(a0)
-  FLOAD    f27, (RISCV_FOFFSET + RISCV_FSIZE * 27)(a0)
-  FLOAD    f28, (RISCV_FOFFSET + RISCV_FSIZE * 28)(a0)
-  FLOAD    f29, (RISCV_FOFFSET + RISCV_FSIZE * 29)(a0)
-  FLOAD    f30, (RISCV_FOFFSET + RISCV_FSIZE * 30)(a0)
-  FLOAD    f31, (RISCV_FOFFSET + RISCV_FSIZE * 31)(a0)
+  .irp i,FROM_0_TO_31
+    FLOAD f\i, (RISCV_FOFFSET + RISCV_FSIZE * \i)(a0)
+  .endr
 # endif
 
   // x0 is zero
   ILOAD    x1, (RISCV_ISIZE * 0)(a0) // restore pc into ra
-  ILOAD    x2, (RISCV_ISIZE * 2)(a0)
-  ILOAD    x3, (RISCV_ISIZE * 3)(a0)
-  ILOAD    x4, (RISCV_ISIZE * 4)(a0)
-  ILOAD    x5, (RISCV_ISIZE * 5)(a0)
-  ILOAD    x6, (RISCV_ISIZE * 6)(a0)
-  ILOAD    x7, (RISCV_ISIZE * 7)(a0)
-  ILOAD    x8, (RISCV_ISIZE * 8)(a0)
-  ILOAD    x9, (RISCV_ISIZE * 9)(a0)
+  .irp i,2,3,4,5,6,7,8,9
+    ILOAD x\i, (RISCV_ISIZE * \i)(a0)
+  .endr
   // skip a0 for now
-  ILOAD    x11, (RISCV_ISIZE * 11)(a0)
-  ILOAD    x12, (RISCV_ISIZE * 12)(a0)
-  ILOAD    x13, (RISCV_ISIZE * 13)(a0)
-  ILOAD    x14, (RISCV_ISIZE * 14)(a0)
-  ILOAD    x15, (RISCV_ISIZE * 15)(a0)
-  ILOAD    x16, (RISCV_ISIZE * 16)(a0)
-  ILOAD    x17, (RISCV_ISIZE * 17)(a0)
-  ILOAD    x18, (RISCV_ISIZE * 18)(a0)
-  ILOAD    x19, (RISCV_ISIZE * 19)(a0)
-  ILOAD    x20, (RISCV_ISIZE * 20)(a0)
-  ILOAD    x21, (RISCV_ISIZE * 21)(a0)
-  ILOAD    x22, (RISCV_ISIZE * 22)(a0)
-  ILOAD    x23, (RISCV_ISIZE * 23)(a0)
-  ILOAD    x24, (RISCV_ISIZE * 24)(a0)
-  ILOAD    x25, (RISCV_ISIZE * 25)(a0)
-  ILOAD    x26, (RISCV_ISIZE * 26)(a0)
-  ILOAD    x27, (RISCV_ISIZE * 27)(a0)
-  ILOAD    x28, (RISCV_ISIZE * 28)(a0)
-  ILOAD    x29, (RISCV_ISIZE * 29)(a0)
-  ILOAD    x30, (RISCV_ISIZE * 30)(a0)
-  ILOAD    x31, (RISCV_ISIZE * 31)(a0)
+  .irp i,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+    ILOAD x\i, (RISCV_ISIZE * \i)(a0)
+  .endr
   ILOAD    x10, (RISCV_ISIZE * 10)(a0)   // restore a0
 
   ret                       // jump to ra
 
 #elif defined(__s390x__)
 
 DEFINE_LIBUNWIND_FUNCTION(_ZN9libunwind15Registers_s390x6jumptoEv)
 //
 // void libunwind::Registers_s390x::jumpto()
 //
 // On entry:
 //  thread_state pointer is in r2
 //
 
   // Skip PSWM, but load PSWA into r1
   lg %r1, 8(%r2)
 
   // Restore FPRs
-  ld %f0, 144(%r2)
-  ld %f1, 152(%r2)
-  ld %f2, 160(%r2)
-  ld %f3, 168(%r2)
-  ld %f4, 176(%r2)
-  ld %f5, 184(%r2)
-  ld %f6, 192(%r2)
-  ld %f7, 200(%r2)
-  ld %f8, 208(%r2)
-  ld %f9, 216(%r2)
-  ld %f10, 224(%r2)
-  ld %f11, 232(%r2)
-  ld %f12, 240(%r2)
-  ld %f13, 248(%r2)
-  ld %f14, 256(%r2)
-  ld %f15, 264(%r2)
+  .irp i,FROM_0_TO_15
+    ld %f\i, (144+8*\i)(%r2)
+  .endr
 
   // Restore GPRs - skipping %r0 and %r1
   lmg  %r2, %r15, 32(%r2)
 
   // Return to PSWA (was loaded into %r1 above)
   br %r1
 
 #elif defined(__loongarch__) && __loongarch_grlen == 64
 
 //
 // void libunwind::Registers_loongarch::jumpto()
 //
 // On entry:
 //  thread_state pointer is in $a0($r4)
 //
   .p2align 2
 DEFINE_LIBUNWIND_FUNCTION(_ZN9libunwind19Registers_loongarch6jumptoEv)
 # if __loongarch_frlen == 64
-  fld.d    $f0, $a0, (8 * 33 + 8 * 0)
-  fld.d    $f1, $a0, (8 * 33 + 8 * 1)
-  fld.d    $f2, $a0, (8 * 33 + 8 * 2)
-  fld.d    $f3, $a0, (8 * 33 + 8 * 3)
-  fld.d    $f4, $a0, (8 * 33 + 8 * 4)
-  fld.d    $f5, $a0, (8 * 33 + 8 * 5)
-  fld.d    $f6, $a0, (8 * 33 + 8 * 6)
-  fld.d    $f7, $a0, (8 * 33 + 8 * 7)
-  fld.d    $f8, $a0, (8 * 33 + 8 * 8)
-  fld.d    $f9, $a0, (8 * 33 + 8 * 9)
-  fld.d    $f10, $a0, (8 * 33 + 8 * 10)
-  fld.d    $f11, $a0, (8 * 33 + 8 * 11)
-  fld.d    $f12, $a0, (8 * 33 + 8 * 12)
-  fld.d    $f13, $a0, (8 * 33 + 8 * 13)
-  fld.d    $f14, $a0, (8 * 33 + 8 * 14)
-  fld.d    $f15, $a0, (8 * 33 + 8 * 15)
-  fld.d    $f16, $a0, (8 * 33 + 8 * 16)
-  fld.d    $f17, $a0, (8 * 33 + 8 * 17)
-  fld.d    $f18, $a0, (8 * 33 + 8 * 18)
-  fld.d    $f19, $a0, (8 * 33 + 8 * 19)
-  fld.d    $f20, $a0, (8 * 33 + 8 * 20)
-  fld.d    $f21, $a0, (8 * 33 + 8 * 21)
-  fld.d    $f22, $a0, (8 * 33 + 8 * 22)
-  fld.d    $f23, $a0, (8 * 33 + 8 * 23)
-  fld.d    $f24, $a0, (8 * 33 + 8 * 24)
-  fld.d    $f25, $a0, (8 * 33 + 8 * 25)
-  fld.d    $f26, $a0, (8 * 33 + 8 * 26)
-  fld.d    $f27, $a0, (8 * 33 + 8 * 27)
-  fld.d    $f28, $a0, (8 * 33 + 8 * 28)
-  fld.d    $f29, $a0, (8 * 33 + 8 * 29)
-  fld.d    $f30, $a0, (8 * 33 + 8 * 30)
-  fld.d    $f31, $a0, (8 * 33 + 8 * 31)
+  .irp i,FROM_0_TO_31
+    fld.d $f\i, $a0, (8 * 33 + 8 * \i)
+  .endr
 # endif
 
   // $r0 is zero
-  ld.d    $r1, $a0, (8 * 1)
-  ld.d    $r2, $a0, (8 * 2)
-  ld.d    $r3, $a0, (8 * 3)
+  .irp i,1,2,3
+    ld.d $r\i, $a0, (8 * \i)
+  .endr
   // skip $a0 for now
-  ld.d    $r5, $a0, (8 * 5)
-  ld.d    $r6, $a0, (8 * 6)
-  ld.d    $r7, $a0, (8 * 7)
-  ld.d    $r8, $a0, (8 * 8)
-  ld.d    $r9, $a0, (8 * 9)
-  ld.d    $r10, $a0, (8 * 10)
-  ld.d    $r11, $a0, (8 * 11)
-  ld.d    $r12, $a0, (8 * 12)
-  ld.d    $r13, $a0, (8 * 13)
-  ld.d    $r14, $a0, (8 * 14)
-  ld.d    $r15, $a0, (8 * 15)
-  ld.d    $r16, $a0, (8 * 16)
-  ld.d    $r17, $a0, (8 * 17)
-  ld.d    $r18, $a0, (8 * 18)
-  ld.d    $r19, $a0, (8 * 19)
-  ld.d    $r20, $a0, (8 * 20)
-  ld.d    $r21, $a0, (8 * 21)
-  ld.d    $r22, $a0, (8 * 22)
-  ld.d    $r23, $a0, (8 * 23)
-  ld.d    $r24, $a0, (8 * 24)
-  ld.d    $r25, $a0, (8 * 25)
-  ld.d    $r26, $a0, (8 * 26)
-  ld.d    $r27, $a0, (8 * 27)
-  ld.d    $r28, $a0, (8 * 28)
-  ld.d    $r29, $a0, (8 * 29)
-  ld.d    $r30, $a0, (8 * 30)
-  ld.d    $r31, $a0, (8 * 31)
+  .irp i,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+    ld.d $r\i, $a0, (8 * \i)
+  .endr
+
   ld.d    $r4,  $a0, (8 * 4)   // restore $a0 last
   ld.d    $r1,  $a0, (8 * 32)  // load new pc into $ra
 
   jr      $ra
 
 #endif
 
 #endif /* !defined(__USING_SJLJ_EXCEPTIONS__) */
 
 NO_EXEC_STACK_DIRECTIVE
 

libunwind/src/UnwindRegistersSave.S

 //===----------------------------------------------------------------------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 #include "assembly.h"
 
+#define FROM_0_TO_15 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+#define FROM_16_TO_31 16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+
+#define FROM_0_TO_31 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+#define FROM_32_TO_63 32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63
+
 #if defined(_AIX)
     .toc
 #else
     .text
 #endif
 
 #if !defined(__USING_SJLJ_EXCEPTIONS__)
 

 # On entry:
 #  thread_state pointer is in a0 ($4)
 #
 DEFINE_LIBUNWIND_FUNCTION(__unw_getcontext)
   .set push
   .set noat
   .set noreorder
   .set nomacro
-  sd    $1, (8 * 1)($4)
-  sd    $2, (8 * 2)($4)
-  sd    $3, (8 * 3)($4)
-  sd    $4, (8 * 4)($4)
-  sd    $5, (8 * 5)($4)
-  sd    $6, (8 * 6)($4)
-  sd    $7, (8 * 7)($4)
-  sd    $8, (8 * 8)($4)
-  sd    $9, (8 * 9)($4)
-  sd    $10, (8 * 10)($4)
-  sd    $11, (8 * 11)($4)
-  sd    $12, (8 * 12)($4)
-  sd    $13, (8 * 13)($4)
-  sd    $14, (8 * 14)($4)
-  sd    $15, (8 * 15)($4)
-  sd    $16, (8 * 16)($4)
-  sd    $17, (8 * 17)($4)
-  sd    $18, (8 * 18)($4)
-  sd    $19, (8 * 19)($4)
-  sd    $20, (8 * 20)($4)
-  sd    $21, (8 * 21)($4)
-  sd    $22, (8 * 22)($4)
-  sd    $23, (8 * 23)($4)
-  sd    $24, (8 * 24)($4)
-  sd    $25, (8 * 25)($4)
-  sd    $26, (8 * 26)($4)
-  sd    $27, (8 * 27)($4)
-  sd    $28, (8 * 28)($4)
-  sd    $29, (8 * 29)($4)
-  sd    $30, (8 * 30)($4)
-  sd    $31, (8 * 31)($4)
+  .irp i,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+    sd $\i, (8 * \i)($4)
+  .endr
   # Store return address to pc
   sd    $31, (8 * 32)($4)
   # hi and lo
   mfhi  $8
   sd    $8,  (8 * 33)($4)
   mflo  $8
   sd    $8,  (8 * 34)($4)
 #ifdef __mips_hard_float
-  sdc1  $f0, (8 * 35)($4)
-  sdc1  $f1, (8 * 36)($4)
-  sdc1  $f2, (8 * 37)($4)
-  sdc1  $f3, (8 * 38)($4)
-  sdc1  $f4, (8 * 39)($4)
-  sdc1  $f5, (8 * 40)($4)
-  sdc1  $f6, (8 * 41)($4)
-  sdc1  $f7, (8 * 42)($4)
-  sdc1  $f8, (8 * 43)($4)
-  sdc1  $f9, (8 * 44)($4)
-  sdc1  $f10, (8 * 45)($4)
-  sdc1  $f11, (8 * 46)($4)
-  sdc1  $f12, (8 * 47)($4)
-  sdc1  $f13, (8 * 48)($4)
-  sdc1  $f14, (8 * 49)($4)
-  sdc1  $f15, (8 * 50)($4)
-  sdc1  $f16, (8 * 51)($4)
-  sdc1  $f17, (8 * 52)($4)
-  sdc1  $f18, (8 * 53)($4)
-  sdc1  $f19, (8 * 54)($4)
-  sdc1  $f20, (8 * 55)($4)
-  sdc1  $f21, (8 * 56)($4)
-  sdc1  $f22, (8 * 57)($4)
-  sdc1  $f23, (8 * 58)($4)
-  sdc1  $f24, (8 * 59)($4)
-  sdc1  $f25, (8 * 60)($4)
-  sdc1  $f26, (8 * 61)($4)
-  sdc1  $f27, (8 * 62)($4)
-  sdc1  $f28, (8 * 63)($4)
-  sdc1  $f29, (8 * 64)($4)
-  sdc1  $f30, (8 * 65)($4)
-  sdc1  $f31, (8 * 66)($4)
+  .irp i,FROM_0_TO_31
+    sdc1 $f\i, (280+8*\i)($4)
+  .endr
 #endif
   jr  $31
   # return UNW_ESUCCESS
   or    $2, $0, $0
   .set pop
 
 # elif defined(__mips__)
 

 // store register (GPR)
 #define PPC64_STR(n) \
   std   n, (8 * (n + 2))(3)
 
   // save GPRs
   PPC64_STR(0)
   mflr  0
   std   0, PPC64_OFFS_SRR0(3) // store lr as ssr0
-  PPC64_STR(1)
-  PPC64_STR(2)
-  PPC64_STR(3)
-  PPC64_STR(4)
-  PPC64_STR(5)
-  PPC64_STR(6)
-  PPC64_STR(7)
-  PPC64_STR(8)
-  PPC64_STR(9)
-  PPC64_STR(10)
-  PPC64_STR(11)
-  PPC64_STR(12)
-  PPC64_STR(13)
-  PPC64_STR(14)
-  PPC64_STR(15)
-  PPC64_STR(16)
-  PPC64_STR(17)
-  PPC64_STR(18)
-  PPC64_STR(19)
-  PPC64_STR(20)
-  PPC64_STR(21)
-  PPC64_STR(22)
-  PPC64_STR(23)
-  PPC64_STR(24)
-  PPC64_STR(25)
-  PPC64_STR(26)
-  PPC64_STR(27)
-  PPC64_STR(28)
-  PPC64_STR(29)
-  PPC64_STR(30)
-  PPC64_STR(31)
+  .irp i,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+    PPC64_STR(\i)
+  .endr
 
   mfcr  0
   std   0,  PPC64_OFFS_CR(3)
   mfxer 0
   std   0,  PPC64_OFFS_XER(3)
   mflr  0
   std   0,  PPC64_OFFS_LR(3)
   mfctr 0
   std   0,  PPC64_OFFS_CTR(3)
   mfvrsave    0
   std   0,  PPC64_OFFS_VRSAVE(3)
 
 #if defined(__VSX__)
   // save VS registers
   // (note that this also saves floating point registers and V registers,
   // because part of VS is mapped to these registers)
 
   addi  4, 3, PPC64_OFFS_FP
 
 // store VS register
 #define PPC64_STVS(n)      \
   stxvd2x n, 0, 4         ;\
   addi    4, 4, 16
 
-  PPC64_STVS(0)
-  PPC64_STVS(1)
-  PPC64_STVS(2)
-  PPC64_STVS(3)
-  PPC64_STVS(4)
-  PPC64_STVS(5)
-  PPC64_STVS(6)
-  PPC64_STVS(7)
-  PPC64_STVS(8)
-  PPC64_STVS(9)
-  PPC64_STVS(10)
-  PPC64_STVS(11)
-  PPC64_STVS(12)
-  PPC64_STVS(13)
-  PPC64_STVS(14)
-  PPC64_STVS(15)
-  PPC64_STVS(16)
-  PPC64_STVS(17)
-  PPC64_STVS(18)
-  PPC64_STVS(19)
-  PPC64_STVS(20)
-  PPC64_STVS(21)
-  PPC64_STVS(22)
-  PPC64_STVS(23)
-  PPC64_STVS(24)
-  PPC64_STVS(25)
-  PPC64_STVS(26)
-  PPC64_STVS(27)
-  PPC64_STVS(28)
-  PPC64_STVS(29)
-  PPC64_STVS(30)
-  PPC64_STVS(31)
-  PPC64_STVS(32)
-  PPC64_STVS(33)
-  PPC64_STVS(34)
-  PPC64_STVS(35)
-  PPC64_STVS(36)
-  PPC64_STVS(37)
-  PPC64_STVS(38)
-  PPC64_STVS(39)
-  PPC64_STVS(40)
-  PPC64_STVS(41)
-  PPC64_STVS(42)
-  PPC64_STVS(43)
-  PPC64_STVS(44)
-  PPC64_STVS(45)
-  PPC64_STVS(46)
-  PPC64_STVS(47)
-  PPC64_STVS(48)
-  PPC64_STVS(49)
-  PPC64_STVS(50)
-  PPC64_STVS(51)
-  PPC64_STVS(52)
-  PPC64_STVS(53)
-  PPC64_STVS(54)
-  PPC64_STVS(55)
-  PPC64_STVS(56)
-  PPC64_STVS(57)
-  PPC64_STVS(58)
-  PPC64_STVS(59)
-  PPC64_STVS(60)
-  PPC64_STVS(61)
-  PPC64_STVS(62)
-  PPC64_STVS(63)
+  .irp i,FROM_0_TO_31
+    PPC64_STVS(\i)
+  .endr
+  .irp i,FROM_32_TO_63
+    PPC64_STVS(\i)
+  .endr
 
 #else
 
 // store FP register
 #define PPC64_STF(n) \
   stfd  n, (PPC64_OFFS_FP + n * 16)(3)
 
   // save float registers
-  PPC64_STF(0)
-  PPC64_STF(1)
-  PPC64_STF(2)
-  PPC64_STF(3)
-  PPC64_STF(4)
-  PPC64_STF(5)
-  PPC64_STF(6)
-  PPC64_STF(7)
-  PPC64_STF(8)
-  PPC64_STF(9)
-  PPC64_STF(10)
-  PPC64_STF(11)
-  PPC64_STF(12)
-  PPC64_STF(13)
-  PPC64_STF(14)
-  PPC64_STF(15)
-  PPC64_STF(16)
-  PPC64_STF(17)
-  PPC64_STF(18)
-  PPC64_STF(19)
-  PPC64_STF(20)
-  PPC64_STF(21)
-  PPC64_STF(22)
-  PPC64_STF(23)
-  PPC64_STF(24)
-  PPC64_STF(25)
-  PPC64_STF(26)
-  PPC64_STF(27)
-  PPC64_STF(28)
-  PPC64_STF(29)
-  PPC64_STF(30)
-  PPC64_STF(31)
+  .irp i,FROM_0_TO_31
+    PPC64_STF(\i)
+  .endr
 
 #if defined(__ALTIVEC__)
   // save vector registers
 
   // Use 16-bytes below the stack pointer as an
   // aligned buffer to save each vector register.
   // Note that the stack pointer is always 16-byte aligned.
   subi  4, 1, 16
 
 #define PPC64_STV_UNALIGNED(n)             \
   stvx  n, 0, 4                           ;\
   ld    5, 0(4)                           ;\
   std   5, (PPC64_OFFS_V + n * 16)(3)     ;\
   ld    5, 8(4)                           ;\
   std   5, (PPC64_OFFS_V + n * 16 + 8)(3)
 
-  PPC64_STV_UNALIGNED(0)
-  PPC64_STV_UNALIGNED(1)
-  PPC64_STV_UNALIGNED(2)
-  PPC64_STV_UNALIGNED(3)
-  PPC64_STV_UNALIGNED(4)
-  PPC64_STV_UNALIGNED(5)
-  PPC64_STV_UNALIGNED(6)
-  PPC64_STV_UNALIGNED(7)
-  PPC64_STV_UNALIGNED(8)
-  PPC64_STV_UNALIGNED(9)
-  PPC64_STV_UNALIGNED(10)
-  PPC64_STV_UNALIGNED(11)
-  PPC64_STV_UNALIGNED(12)
-  PPC64_STV_UNALIGNED(13)
-  PPC64_STV_UNALIGNED(14)
-  PPC64_STV_UNALIGNED(15)
-  PPC64_STV_UNALIGNED(16)
-  PPC64_STV_UNALIGNED(17)
-  PPC64_STV_UNALIGNED(18)
-  PPC64_STV_UNALIGNED(19)
-  PPC64_STV_UNALIGNED(20)
-  PPC64_STV_UNALIGNED(21)
-  PPC64_STV_UNALIGNED(22)
-  PPC64_STV_UNALIGNED(23)
-  PPC64_STV_UNALIGNED(24)
-  PPC64_STV_UNALIGNED(25)
-  PPC64_STV_UNALIGNED(26)
-  PPC64_STV_UNALIGNED(27)
-  PPC64_STV_UNALIGNED(28)
-  PPC64_STV_UNALIGNED(29)
-  PPC64_STV_UNALIGNED(30)
-  PPC64_STV_UNALIGNED(31)
+  .irp i,FROM_0_TO_31
+    PPC64_STV_UNALIGNED(\i)
+  .endr
 
 #endif
 #endif
 
   li    3,  0   // return UNW_ESUCCESS
   blr
 
 
 #elif defined(__powerpc__)
 
 //
 // extern int unw_getcontext(unw_context_t* thread_state)
 //
 // On entry:
 //  thread_state pointer is in r3
 //
 #if defined(_AIX)
 DEFINE_LIBUNWIND_FUNCTION_AND_WEAK_ALIAS(__unw_getcontext, unw_getcontext)
 #else
 DEFINE_LIBUNWIND_FUNCTION(__unw_getcontext)
 #endif
   stw     0,   8(3)
   mflr    0
   stw     0,   0(3) // store lr as ssr0
-  stw     1,  12(3)
-  stw     2,  16(3)
-  stw     3,  20(3)
-  stw     4,  24(3)
-  stw     5,  28(3)
-  stw     6,  32(3)
-  stw     7,  36(3)
-  stw     8,  40(3)
-  stw     9,  44(3)
-  stw     10, 48(3)
-  stw     11, 52(3)
-  stw     12, 56(3)
-  stw     13, 60(3)
-  stw     14, 64(3)
-  stw     15, 68(3)
-  stw     16, 72(3)
-  stw     17, 76(3)
-  stw     18, 80(3)
-  stw     19, 84(3)
-  stw     20, 88(3)
-  stw     21, 92(3)
-  stw     22, 96(3)
-  stw     23,100(3)
-  stw     24,104(3)
-  stw     25,108(3)
-  stw     26,112(3)
-  stw     27,116(3)
-  stw     28,120(3)
-  stw     29,124(3)
-  stw     30,128(3)
-  stw     31,132(3)
+  .irp i,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+    stw \i, (8+4*\i)(3)
+  .endr
 
 #if defined(__ALTIVEC__)
   // save VRSave register
   mfspr   0, 256
   stw     0, 156(3)
 #endif
   // save CR registers
   mfcr    0
   stw     0, 136(3)
   // save CTR register
   mfctr   0
   stw     0, 148(3)
 
 #if !defined(__NO_FPRS__)
   // save float registers
-  stfd    0, 160(3)
-  stfd    1, 168(3)
-  stfd    2, 176(3)
-  stfd    3, 184(3)
-  stfd    4, 192(3)
-  stfd    5, 200(3)
-  stfd    6, 208(3)
-  stfd    7, 216(3)
-  stfd    8, 224(3)
-  stfd    9, 232(3)
-  stfd    10,240(3)
-  stfd    11,248(3)
-  stfd    12,256(3)
-  stfd    13,264(3)
-  stfd    14,272(3)
-  stfd    15,280(3)
-  stfd    16,288(3)
-  stfd    17,296(3)
-  stfd    18,304(3)
-  stfd    19,312(3)
-  stfd    20,320(3)
-  stfd    21,328(3)
-  stfd    22,336(3)
-  stfd    23,344(3)
-  stfd    24,352(3)
-  stfd    25,360(3)
-  stfd    26,368(3)
-  stfd    27,376(3)
-  stfd    28,384(3)
-  stfd    29,392(3)
-  stfd    30,400(3)
-  stfd    31,408(3)
+  .irp i,FROM_0_TO_31
+    stfd \i, (160+8*\i)(3)
+  .endr
 #endif
 
 #if defined(__ALTIVEC__)
   // save vector registers
 
   subi    4, 1, 16
   rlwinm  4, 4, 0, 0, 27  // mask low 4-bits
   // r4 is now a 16-byte aligned pointer into the red zone
 
 #define SAVE_VECTOR_UNALIGNED(_vec, _offset) \
   stvx    _vec, 0, 4               SEPARATOR \
   lwz     5, 0(4)                  SEPARATOR \
   stw     5, _offset(3)            SEPARATOR \
   lwz     5, 4(4)                  SEPARATOR \
   stw     5, _offset+4(3)          SEPARATOR \
   lwz     5, 8(4)                  SEPARATOR \
   stw     5, _offset+8(3)          SEPARATOR \
   lwz     5, 12(4)                 SEPARATOR \
   stw     5, _offset+12(3)
 
-  SAVE_VECTOR_UNALIGNED( 0, 424+0x000)
-  SAVE_VECTOR_UNALIGNED( 1, 424+0x010)
-  SAVE_VECTOR_UNALIGNED( 2, 424+0x020)
-  SAVE_VECTOR_UNALIGNED( 3, 424+0x030)
-  SAVE_VECTOR_UNALIGNED( 4, 424+0x040)
-  SAVE_VECTOR_UNALIGNED( 5, 424+0x050)
-  SAVE_VECTOR_UNALIGNED( 6, 424+0x060)
-  SAVE_VECTOR_UNALIGNED( 7, 424+0x070)
-  SAVE_VECTOR_UNALIGNED( 8, 424+0x080)
-  SAVE_VECTOR_UNALIGNED( 9, 424+0x090)
-  SAVE_VECTOR_UNALIGNED(10, 424+0x0A0)
-  SAVE_VECTOR_UNALIGNED(11, 424+0x0B0)
-  SAVE_VECTOR_UNALIGNED(12, 424+0x0C0)
-  SAVE_VECTOR_UNALIGNED(13, 424+0x0D0)
-  SAVE_VECTOR_UNALIGNED(14, 424+0x0E0)
-  SAVE_VECTOR_UNALIGNED(15, 424+0x0F0)
-  SAVE_VECTOR_UNALIGNED(16, 424+0x100)
-  SAVE_VECTOR_UNALIGNED(17, 424+0x110)
-  SAVE_VECTOR_UNALIGNED(18, 424+0x120)
-  SAVE_VECTOR_UNALIGNED(19, 424+0x130)
-  SAVE_VECTOR_UNALIGNED(20, 424+0x140)
-  SAVE_VECTOR_UNALIGNED(21, 424+0x150)
-  SAVE_VECTOR_UNALIGNED(22, 424+0x160)
-  SAVE_VECTOR_UNALIGNED(23, 424+0x170)
-  SAVE_VECTOR_UNALIGNED(24, 424+0x180)
-  SAVE_VECTOR_UNALIGNED(25, 424+0x190)
-  SAVE_VECTOR_UNALIGNED(26, 424+0x1A0)
-  SAVE_VECTOR_UNALIGNED(27, 424+0x1B0)
-  SAVE_VECTOR_UNALIGNED(28, 424+0x1C0)
-  SAVE_VECTOR_UNALIGNED(29, 424+0x1D0)
-  SAVE_VECTOR_UNALIGNED(30, 424+0x1E0)
-  SAVE_VECTOR_UNALIGNED(31, 424+0x1F0)
+  .irp i,FROM_0_TO_31
+    SAVE_VECTOR_UNALIGNED(\i, \i*16+424)
+  .endr
 #endif
 
   li      3, 0  // return UNW_ESUCCESS
   blr
 
 
 #elif defined(__aarch64__)
 

 #
 # extern int __unw_getcontext(unw_context_t* thread_state)
 #
 # On entry:
 #  thread_state pointer is in a0
 #
 DEFINE_LIBUNWIND_FUNCTION(__unw_getcontext)
   ISTORE    x1, (RISCV_ISIZE * 0)(a0) // store ra as pc
-  ISTORE    x1, (RISCV_ISIZE * 1)(a0)
-  ISTORE    x2, (RISCV_ISIZE * 2)(a0)
-  ISTORE    x3, (RISCV_ISIZE * 3)(a0)
-  ISTORE    x4, (RISCV_ISIZE * 4)(a0)
-  ISTORE    x5, (RISCV_ISIZE * 5)(a0)
-  ISTORE    x6, (RISCV_ISIZE * 6)(a0)
-  ISTORE    x7, (RISCV_ISIZE * 7)(a0)
-  ISTORE    x8, (RISCV_ISIZE * 8)(a0)
-  ISTORE    x9, (RISCV_ISIZE * 9)(a0)
-  ISTORE    x10, (RISCV_ISIZE * 10)(a0)
-  ISTORE    x11, (RISCV_ISIZE * 11)(a0)
-  ISTORE    x12, (RISCV_ISIZE * 12)(a0)
-  ISTORE    x13, (RISCV_ISIZE * 13)(a0)
-  ISTORE    x14, (RISCV_ISIZE * 14)(a0)
-  ISTORE    x15, (RISCV_ISIZE * 15)(a0)
-  ISTORE    x16, (RISCV_ISIZE * 16)(a0)
-  ISTORE    x17, (RISCV_ISIZE * 17)(a0)
-  ISTORE    x18, (RISCV_ISIZE * 18)(a0)
-  ISTORE    x19, (RISCV_ISIZE * 19)(a0)
-  ISTORE    x20, (RISCV_ISIZE * 20)(a0)
-  ISTORE    x21, (RISCV_ISIZE * 21)(a0)
-  ISTORE    x22, (RISCV_ISIZE * 22)(a0)
-  ISTORE    x23, (RISCV_ISIZE * 23)(a0)
-  ISTORE    x24, (RISCV_ISIZE * 24)(a0)
-  ISTORE    x25, (RISCV_ISIZE * 25)(a0)
-  ISTORE    x26, (RISCV_ISIZE * 26)(a0)
-  ISTORE    x27, (RISCV_ISIZE * 27)(a0)
-  ISTORE    x28, (RISCV_ISIZE * 28)(a0)
-  ISTORE    x29, (RISCV_ISIZE * 29)(a0)
-  ISTORE    x30, (RISCV_ISIZE * 30)(a0)
-  ISTORE    x31, (RISCV_ISIZE * 31)(a0)
+  .irp i,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+    ISTORE x\i, (RISCV_ISIZE * \i)(a0)
+  .endr
 
 # if defined(__riscv_flen)
-  FSTORE    f0, (RISCV_FOFFSET + RISCV_FSIZE * 0)(a0)
-  FSTORE    f1, (RISCV_FOFFSET + RISCV_FSIZE * 1)(a0)
-  FSTORE    f2, (RISCV_FOFFSET + RISCV_FSIZE * 2)(a0)
-  FSTORE    f3, (RISCV_FOFFSET + RISCV_FSIZE * 3)(a0)
-  FSTORE    f4, (RISCV_FOFFSET + RISCV_FSIZE * 4)(a0)
-  FSTORE    f5, (RISCV_FOFFSET + RISCV_FSIZE * 5)(a0)
-  FSTORE    f6, (RISCV_FOFFSET + RISCV_FSIZE * 6)(a0)
-  FSTORE    f7, (RISCV_FOFFSET + RISCV_FSIZE * 7)(a0)
-  FSTORE    f8, (RISCV_FOFFSET + RISCV_FSIZE * 8)(a0)
-  FSTORE    f9, (RISCV_FOFFSET + RISCV_FSIZE * 9)(a0)
-  FSTORE    f10, (RISCV_FOFFSET + RISCV_FSIZE * 10)(a0)
-  FSTORE    f11, (RISCV_FOFFSET + RISCV_FSIZE * 11)(a0)
-  FSTORE    f12, (RISCV_FOFFSET + RISCV_FSIZE * 12)(a0)
-  FSTORE    f13, (RISCV_FOFFSET + RISCV_FSIZE * 13)(a0)
-  FSTORE    f14, (RISCV_FOFFSET + RISCV_FSIZE * 14)(a0)
-  FSTORE    f15, (RISCV_FOFFSET + RISCV_FSIZE * 15)(a0)
-  FSTORE    f16, (RISCV_FOFFSET + RISCV_FSIZE * 16)(a0)
-  FSTORE    f17, (RISCV_FOFFSET + RISCV_FSIZE * 17)(a0)
-  FSTORE    f18, (RISCV_FOFFSET + RISCV_FSIZE * 18)(a0)
-  FSTORE    f19, (RISCV_FOFFSET + RISCV_FSIZE * 19)(a0)
-  FSTORE    f20, (RISCV_FOFFSET + RISCV_FSIZE * 20)(a0)
-  FSTORE    f21, (RISCV_FOFFSET + RISCV_FSIZE * 21)(a0)
-  FSTORE    f22, (RISCV_FOFFSET + RISCV_FSIZE * 22)(a0)
-  FSTORE    f23, (RISCV_FOFFSET + RISCV_FSIZE * 23)(a0)
-  FSTORE    f24, (RISCV_FOFFSET + RISCV_FSIZE * 24)(a0)
-  FSTORE    f25, (RISCV_FOFFSET + RISCV_FSIZE * 25)(a0)
-  FSTORE    f26, (RISCV_FOFFSET + RISCV_FSIZE * 26)(a0)
-  FSTORE    f27, (RISCV_FOFFSET + RISCV_FSIZE * 27)(a0)
-  FSTORE    f28, (RISCV_FOFFSET + RISCV_FSIZE * 28)(a0)
-  FSTORE    f29, (RISCV_FOFFSET + RISCV_FSIZE * 29)(a0)
-  FSTORE    f30, (RISCV_FOFFSET + RISCV_FSIZE * 30)(a0)
-  FSTORE    f31, (RISCV_FOFFSET + RISCV_FSIZE * 31)(a0)
+  .irp i,FROM_0_TO_31
+    FSTORE f\i, (RISCV_FOFFSET + RISCV_FSIZE * \i)(a0)
+  .endr
 # endif
 
   li     a0, 0  // return UNW_ESUCCESS
   ret           // jump to ra
 
 #elif defined(__s390x__)
 
 //
DEFINE_LIBUNWIND_FUNCTION(__unw_getcontext)
   // Save PSWM
   epsw %r0, %r1
   stm %r0, %r1, 0(%r2)
 
   // Store return address as PSWA
   stg %r14, 8(%r2)
 
   // Save FPRs
-  std %f0, 144(%r2)
-  std %f1, 152(%r2)
-  std %f2, 160(%r2)
-  std %f3, 168(%r2)
-  std %f4, 176(%r2)
-  std %f5, 184(%r2)
-  std %f6, 192(%r2)
-  std %f7, 200(%r2)
-  std %f8, 208(%r2)
-  std %f9, 216(%r2)
-  std %f10, 224(%r2)
-  std %f11, 232(%r2)
-  std %f12, 240(%r2)
-  std %f13, 248(%r2)
-  std %f14, 256(%r2)
-  std %f15, 264(%r2)
+  .irp i,FROM_0_TO_15
+    std %f\i, (144+8*\i)(%r2)
+  .endr
 
   // Return UNW_ESUCCESS
   lghi %r2, 0
   br %r14
 
 #elif defined(__loongarch__) && __loongarch_grlen == 64
 
 #
 # extern int __unw_getcontext(unw_context_t* thread_state)
 #
 # On entry:
 #  thread_state pointer is in $a0($r4)
 #
 DEFINE_LIBUNWIND_FUNCTION(__unw_getcontext)
-  st.d    $r1, $a0, (8 * 1)
-  st.d    $r2, $a0, (8 * 2)
-  st.d    $r3, $a0, (8 * 3)
-  st.d    $r4, $a0, (8 * 4)
-  st.d    $r5, $a0, (8 * 5)
-  st.d    $r6, $a0, (8 * 6)
-  st.d    $r7, $a0, (8 * 7)
-  st.d    $r8, $a0, (8 * 8)
-  st.d    $r9, $a0, (8 * 9)
-  st.d    $r10, $a0, (8 * 10)
-  st.d    $r11, $a0, (8 * 11)
-  st.d    $r12, $a0, (8 * 12)
-  st.d    $r13, $a0, (8 * 13)
-  st.d    $r14, $a0, (8 * 14)
-  st.d    $r15, $a0, (8 * 15)
-  st.d    $r16, $a0, (8 * 16)
-  st.d    $r17, $a0, (8 * 17)
-  st.d    $r18, $a0, (8 * 18)
-  st.d    $r19, $a0, (8 * 19)
-  st.d    $r20, $a0, (8 * 20)
-  st.d    $r21, $a0, (8 * 21)
-  st.d    $r22, $a0, (8 * 22)
-  st.d    $r23, $a0, (8 * 23)
-  st.d    $r24, $a0, (8 * 24)
-  st.d    $r25, $a0, (8 * 25)
-  st.d    $r26, $a0, (8 * 26)
-  st.d    $r27, $a0, (8 * 27)
-  st.d    $r28, $a0, (8 * 28)
-  st.d    $r29, $a0, (8 * 29)
-  st.d    $r30, $a0, (8 * 30)
-  st.d    $r31, $a0, (8 * 31)
+  .irp i,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
+    std.d $r\i, $a0, (8*\i)
+  .endr
   st.d    $r1,  $a0, (8 * 32) // store $ra to pc
 
 # if __loongarch_frlen == 64
-  fst.d    $f0, $a0, (8 * 33 + 8 * 0)
-  fst.d    $f1, $a0, (8 * 33 + 8 * 1)
-  fst.d    $f2, $a0, (8 * 33 + 8 * 2)
-  fst.d    $f3, $a0, (8 * 33 + 8 * 3)
-  fst.d    $f4, $a0, (8 * 33 + 8 * 4)
-  fst.d    $f5, $a0, (8 * 33 + 8 * 5)
-  fst.d    $f6, $a0, (8 * 33 + 8 * 6)
-  fst.d    $f7, $a0, (8 * 33 + 8 * 7)
-  fst.d    $f8, $a0, (8 * 33 + 8 * 8)
-  fst.d    $f9, $a0, (8 * 33 + 8 * 9)
-  fst.d    $f10, $a0, (8 * 33 + 8 * 10)
-  fst.d    $f11, $a0, (8 * 33 + 8 * 11)
-  fst.d    $f12, $a0, (8 * 33 + 8 * 12)
-  fst.d    $f13, $a0, (8 * 33 + 8 * 13)
-  fst.d    $f14, $a0, (8 * 33 + 8 * 14)
-  fst.d    $f15, $a0, (8 * 33 + 8 * 15)
-  fst.d    $f16, $a0, (8 * 33 + 8 * 16)
-  fst.d    $f17, $a0, (8 * 33 + 8 * 17)
-  fst.d    $f18, $a0, (8 * 33 + 8 * 18)
-  fst.d    $f19, $a0, (8 * 33 + 8 * 19)
-  fst.d    $f20, $a0, (8 * 33 + 8 * 20)
-  fst.d    $f21, $a0, (8 * 33 + 8 * 21)
-  fst.d    $f22, $a0, (8 * 33 + 8 * 22)
-  fst.d    $f23, $a0, (8 * 33 + 8 * 23)
-  fst.d    $f24, $a0, (8 * 33 + 8 * 24)
-  fst.d    $f25, $a0, (8 * 33 + 8 * 25)
-  fst.d    $f26, $a0, (8 * 33 + 8 * 26)
-  fst.d    $f27, $a0, (8 * 33 + 8 * 27)
-  fst.d    $f28, $a0, (8 * 33 + 8 * 28)
-  fst.d    $f29, $a0, (8 * 33 + 8 * 29)
-  fst.d    $f30, $a0, (8 * 33 + 8 * 30)
-  fst.d    $f31, $a0, (8 * 33 + 8 * 31)
+  .irp i,FROM_0_TO_31
+    fst.d $f\i, $a0, (8 * 33 + 8 * \i)
+  .endr
 # endif
 
   move     $a0, $zero  // UNW_ESUCCESS
   jr       $ra
 
 #endif
 
   WEAK_ALIAS(__unw_getcontext, unw_getcontext)
 
 #endif /* !defined(__USING_SJLJ_EXCEPTIONS__) */
 
 NO_EXEC_STACK_DIRECTIVE