On 9/14/22 11:50, Clément Chigot wrote:
When requested, the alignment for VLD4.32 is 8 and not 16.

See ARM documentation about VLD4 encoding:
     ebytes = 1 << UInt(size);
     if size == '10' then
         alignment = if a == '0' then 1 else 8;
     else
         alignment = if a == '0' then 1 else 4*ebytes;

Signed-off-by: Clément Chigot <[email protected]>

Reviewed-by: Richard Henderson <[email protected]>

r~

---
  target/arm/translate-neon.c | 6 +++++-
  1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/target/arm/translate-neon.c b/target/arm/translate-neon.c
index 321c17e2c7..4016339d46 100644
--- a/target/arm/translate-neon.c
+++ b/target/arm/translate-neon.c
@@ -584,7 +584,11 @@ static bool trans_VLD_all_lanes(DisasContext *s, 
arg_VLD_all_lanes *a)
          case 3:
              return false;
          case 4:
-            align = pow2_align(size + 2);
+            if (size == 2) {
+                align = pow2_align(3);
+            } else {
+                align = pow2_align(size + 2);
+            }
              break;
          default:
              g_assert_not_reached();


Reply via email to