bug#70474: [PATCH 2/2] Replace aarch64 CAS and atomic-swap generated JIT code with CASAL and SWPAL instructions

[Tony Garnock-Jones]

This appears to make the problem go away. I'm new at working with 
`lightening` so I'm not confident I've covered everything that needs 
covering, particularly wrt the implementation of cas_atomic. But perhaps 
this can be a foundation for someone who knows more than I do to work from.

---
 libguile/lightening/lightening/aarch64-cpu.c | 41 ++++++--------------
 1 file changed, 11 insertions(+), 30 deletions(-)

diff --git a/libguile/lightening/lightening/aarch64-cpu.c 
b/libguile/lightening/lightening/aarch64-cpu.c
index 13aa351e9..30766652f 100644
--- a/libguile/lightening/lightening/aarch64-cpu.c
+++ b/libguile/lightening/lightening/aarch64-cpu.c
@@ -223,8 +223,8 @@ oxxrs(jit_state_t *_jit, int32_t Op,
 #define A64_UMULH                     0x9bc07c00
 #define A64_LDAR                      0xc8dffc00
 #define A64_STLR                      0xc89ffc00
-#define A64_LDAXR                     0xc85ffc00
-#define A64_STLXR                     0xc800fc00
+#define A64_SWPAL                     0xf8e08000
+#define A64_CASAL                     0xc8e0fc00
 #define A64_STRBI                     0x39000000
 #define A64_LDRBI                     0x39400000
 #define A64_LDRSBI                    0x39800000
@@ -664,15 +664,15 @@ STLR(jit_state_t *_jit, int32_t Rt, int32_t Rn)
 }
  static void
-LDAXR(jit_state_t *_jit, int32_t Rt, int32_t Rn) +SWPAL(jit_state_t 
*_jit, int32_t Rt, int32_t Rn, int32_t Rs)
 {
-  return o_xx(_jit, A64_LDAXR, Rt, Rn);
+  return oxxx(_jit, A64_SWPAL, Rt, Rn, Rs);
 }
  static void
-STLXR(jit_state_t *_jit, int32_t Rt, int32_t Rn, int32_t Rm)
+CASAL(jit_state_t *_jit, int32_t Rt, int32_t Rn, int32_t Rs)
 {
-  return oxxx(_jit, A64_STLXR, Rt, Rn, Rm);
+  return oxxx(_jit, A64_CASAL, Rt, Rn, Rs);
 }
  static void
@@ -2532,36 +2532,17 @@ str_atomic(jit_state_t *_jit, int32_t loc, 
int32_t val)
 static void
 swap_atomic(jit_state_t *_jit, int32_t dst, int32_t loc, int32_t val)
 {
-  void *retry = jit_address(_jit);
-  int32_t result = jit_gpr_regno(get_temp_gpr(_jit));
-  int32_t val_or_tmp = dst == val ? jit_gpr_regno(get_temp_gpr(_jit)) : 
val;
-  movr(_jit, val_or_tmp, val);
-  LDAXR(_jit, dst, loc);
-  STLXR(_jit, val_or_tmp, loc, result);
-  jit_patch_there(_jit, bnei(_jit, result, 0), retry);
-  if (dst == val) unget_temp_gpr(_jit);
-  unget_temp_gpr(_jit);
+  SWPAL(_jit, dst, loc, val);
 }
  static void
 cas_atomic(jit_state_t *_jit, int32_t dst, int32_t loc, int32_t expected,
            int32_t desired)
 {
-  int32_t dst_or_tmp;
-  if (dst == loc || dst == expected || dst == expected)
-    dst_or_tmp = jit_gpr_regno(get_temp_gpr(_jit));
-  else
-    dst_or_tmp = dst;
-  void *retry = jit_address(_jit);
-  LDAXR(_jit, dst_or_tmp, loc);
-  jit_reloc_t bad = bner(_jit, dst_or_tmp, expected);
-  int result = jit_gpr_regno(get_temp_gpr(_jit));
-  STLXR(_jit, desired, loc, result);
-  jit_patch_there(_jit, bnei(_jit, result, 0), retry);
-  unget_temp_gpr(_jit);
-  jit_patch_here(_jit, bad);
-  movr(_jit, dst, dst_or_tmp);
-  unget_temp_gpr(_jit);
+  if (dst != expected) {
+    movr(_jit, dst, expected);
+  }
+  CASAL(_jit, desired, loc, dst);
 }
  static void
--
2.44.0