Re: [PATCH v2 3/4] target/m68k: Implement packed decimal real stores

[Richard Henderson]

On 8/16/24 16:37, Philippe Mathieu-Daudé wrote:

Hi Richard,

On 12/8/24 02:44, Richard Henderson wrote:

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2488
Signed-off-by: Richard Henderson <richard.hender...@linaro.org>
---
  target/m68k/helper.h     |   1 +
  target/m68k/fpu_helper.c | 112 +++++++++++++++++++++++++++++++++++++++
  target/m68k/translate.c  |  27 ++++++++--
  3 files changed, 137 insertions(+), 3 deletions(-)

diff --git a/target/m68k/helper.h b/target/m68k/helper.h
index 2c71361451..21af6adb39 100644
--- a/target/m68k/helper.h
+++ b/target/m68k/helper.h
@@ -127,6 +127,7 @@ DEF_HELPER_3(chk, void, env, s32, s32)
  DEF_HELPER_4(chk2, void, env, s32, s32, s32)
  DEF_HELPER_FLAGS_3(load_pdr_to_fx80, TCG_CALL_NO_RWG, void, env, fp, tl)
+DEF_HELPER_FLAGS_4(store_fx80_to_pdr, TCG_CALL_NO_RWG, void, env, tl, fp, int)
  #if !defined(CONFIG_USER_ONLY)
  DEF_HELPER_3(ptest, void, env, i32, i32)
diff --git a/target/m68k/fpu_helper.c b/target/m68k/fpu_helper.c
index 50b0e36cbe..0c9c3b7b64 100644
--- a/target/m68k/fpu_helper.c
+++ b/target/m68k/fpu_helper.c
@@ -750,6 +750,29 @@ void HELPER(fcosh)(CPUM68KState *env, FPReg *res, FPReg 
*val)
      res->d = floatx80_cosh(val->d, &env->fp_status);
  }
+/* 10**0 through 10**18 */
+static const int64_t i64_pow10[] = {
+    1ll,
+    10ll,
+    100ll,
+    1000ll,
+    10000ll,
+    100000ll,
+    1000000ll,
+    10000000ll,
+    100000000ll,
+    1000000000ll,
+    10000000000ll,
+    100000000000ll,
+    1000000000000ll,
+    10000000000000ll,
+    100000000000000ll,
+    1000000000000000ll,
+    10000000000000000ll,
+    100000000000000000ll,
+    1000000000000000000ll,

10**18 is never used, is that expected?

A previous version did.

Could we define KFACTOR_MAX = 17 and use it as
array size?

Seems reasonable.

@@ -119,6 +119,8 @@ typedef struct DisasContext {
      int done_mac;
      int writeback_mask;
      TCGv writeback[8];
+    uint16_t insn;
+    uint16_t ext;

Maybe worth commenting here what this field is used for?

Ok.

r~