diff mbox series

[PULL,2/3] target/ppc: Replicate double->int32 result for some vector insns

Message ID 20220321064458.1517999-3-clg@kaod.org
State New
Headers show
Series [PULL,1/3] ppc64: Avoid pt_regs struct definition | expand

Commit Message

Cédric Le Goater March 21, 2022, 6:44 a.m. UTC 
From: Richard Henderson <richard.henderson@linaro.org>

Power ISA v3.1 formalizes the previously undefined result in
words 1 and 3 to be a copy of the result in words 0 and 2.

This affects: xscvdpsxws, xscvdpuxws, xvcvdpsxws, xvcvdpuxws.

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/852
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
[ clg: checkpatch fixes ]
Message-Id: <20220315053934.377519-1-richard.henderson@linaro.org>
Signed-off-by: Cédric Le Goater <clg@kaod.org>
---
 target/ppc/fpu_helper.c | 45 +++++++++++++++++++++++++++++++++++------
 1 file changed, 39 insertions(+), 6 deletions(-)
diff mbox series

Patch

diff --git a/target/ppc/fpu_helper.c b/target/ppc/fpu_helper.c
index bd12db960af3..564f2ec35374 100644
--- a/target/ppc/fpu_helper.c
+++ b/target/ppc/fpu_helper.c
@@ -2891,22 +2891,55 @@  void helper_##op(CPUPPCState *env, ppc_vsr_t *xt, ppc_vsr_t *xb)             \
 
 VSX_CVT_FP_TO_INT(xscvdpsxds, 1, float64, int64, VsrD(0), VsrD(0), \
                   0x8000000000000000ULL)
-VSX_CVT_FP_TO_INT(xscvdpsxws, 1, float64, int32, VsrD(0), VsrW(1), \
-                  0x80000000U)
 VSX_CVT_FP_TO_INT(xscvdpuxds, 1, float64, uint64, VsrD(0), VsrD(0), 0ULL)
-VSX_CVT_FP_TO_INT(xscvdpuxws, 1, float64, uint32, VsrD(0), VsrW(1), 0U)
 VSX_CVT_FP_TO_INT(xvcvdpsxds, 2, float64, int64, VsrD(i), VsrD(i), \
                   0x8000000000000000ULL)
-VSX_CVT_FP_TO_INT(xvcvdpsxws, 2, float64, int32, VsrD(i), VsrW(2 * i), \
-                  0x80000000U)
 VSX_CVT_FP_TO_INT(xvcvdpuxds, 2, float64, uint64, VsrD(i), VsrD(i), 0ULL)
-VSX_CVT_FP_TO_INT(xvcvdpuxws, 2, float64, uint32, VsrD(i), VsrW(2 * i), 0U)
 VSX_CVT_FP_TO_INT(xvcvspsxds, 2, float32, int64, VsrW(2 * i), VsrD(i), \
                   0x8000000000000000ULL)
 VSX_CVT_FP_TO_INT(xvcvspsxws, 4, float32, int32, VsrW(i), VsrW(i), 0x80000000U)
 VSX_CVT_FP_TO_INT(xvcvspuxds, 2, float32, uint64, VsrW(2 * i), VsrD(i), 0ULL)
 VSX_CVT_FP_TO_INT(xvcvspuxws, 4, float32, uint32, VsrW(i), VsrW(i), 0U)
 
+/*
+ * Likewise, except that the result is duplicated into both subwords.
+ * Power ISA v3.1 has Programming Notes for these insns:
+ *     Previous versions of the architecture allowed the contents of
+ *     word 0 of the result register to be undefined. However, all
+ *     processors that support this instruction write the result into
+ *     words 0 and 1 (and words 2 and 3) of the result register, as
+ *     is required by this version of the architecture.
+ */
+#define VSX_CVT_FP_TO_INT2(op, nels, stp, ttp, rnan)                         \
+void helper_##op(CPUPPCState *env, ppc_vsr_t *xt, ppc_vsr_t *xb)             \
+{                                                                            \
+    int all_flags = env->fp_status.float_exception_flags, flags;             \
+    ppc_vsr_t t = { };                                                       \
+    int i;                                                                   \
+                                                                             \
+    for (i = 0; i < nels; i++) {                                             \
+        env->fp_status.float_exception_flags = 0;                            \
+        t.VsrW(2 * i) = stp##_to_##ttp##_round_to_zero(xb->VsrD(i),          \
+                                                       &env->fp_status);     \
+        flags = env->fp_status.float_exception_flags;                        \
+        if (unlikely(flags & float_flag_invalid)) {                          \
+            t.VsrW(2 * i) = float_invalid_cvt(env, flags, t.VsrW(2 * i),     \
+                                              rnan, 0, GETPC());             \
+        }                                                                    \
+        t.VsrW(2 * i + 1) = t.VsrW(2 * i);                                   \
+        all_flags |= flags;                                                  \
+    }                                                                        \
+                                                                             \
+    *xt = t;                                                                 \
+    env->fp_status.float_exception_flags = all_flags;                        \
+    do_float_check_status(env, GETPC());                                     \
+}
+
+VSX_CVT_FP_TO_INT2(xscvdpsxws, 1, float64, int32, 0x80000000U)
+VSX_CVT_FP_TO_INT2(xscvdpuxws, 1, float64, uint32, 0U)
+VSX_CVT_FP_TO_INT2(xvcvdpsxws, 2, float64, int32, 0x80000000U)
+VSX_CVT_FP_TO_INT2(xvcvdpuxws, 2, float64, uint32, 0U)
+
 /*
  * VSX_CVT_FP_TO_INT_VECTOR - VSX floating point to integer conversion
  *   op    - instruction mnemonic