All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Maydell <peter.maydell@linaro.org>
To: qemu-devel@nongnu.org
Subject: [PULL 16/51] target/arm: Implement MVE fp vector comparisons
Date: Wed,  1 Sep 2021 11:36:18 +0100	[thread overview]
Message-ID: <20210901103653.13435-17-peter.maydell@linaro.org> (raw)
In-Reply-To: <20210901103653.13435-1-peter.maydell@linaro.org>

Implement the MVE fp vector comparisons VCMP and VPT.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
---
 target/arm/helper-mve.h    | 18 +++++++++++
 target/arm/mve.decode      | 39 +++++++++++++++++++----
 target/arm/mve_helper.c    | 64 ++++++++++++++++++++++++++++++++++++++
 target/arm/translate-mve.c | 22 +++++++++++++
 4 files changed, 137 insertions(+), 6 deletions(-)

diff --git a/target/arm/helper-mve.h b/target/arm/helper-mve.h
index 47fd18dddbf..0c15c531641 100644
--- a/target/arm/helper-mve.h
+++ b/target/arm/helper-mve.h
@@ -813,6 +813,24 @@ DEF_HELPER_FLAGS_3(mve_vcmple_scalarb, TCG_CALL_NO_WG, void, env, ptr, i32)
 DEF_HELPER_FLAGS_3(mve_vcmple_scalarh, TCG_CALL_NO_WG, void, env, ptr, i32)
 DEF_HELPER_FLAGS_3(mve_vcmple_scalarw, TCG_CALL_NO_WG, void, env, ptr, i32)
 
+DEF_HELPER_FLAGS_3(mve_vfcmpeqh, TCG_CALL_NO_WG, void, env, ptr, ptr)
+DEF_HELPER_FLAGS_3(mve_vfcmpeqs, TCG_CALL_NO_WG, void, env, ptr, ptr)
+
+DEF_HELPER_FLAGS_3(mve_vfcmpneh, TCG_CALL_NO_WG, void, env, ptr, ptr)
+DEF_HELPER_FLAGS_3(mve_vfcmpnes, TCG_CALL_NO_WG, void, env, ptr, ptr)
+
+DEF_HELPER_FLAGS_3(mve_vfcmpgeh, TCG_CALL_NO_WG, void, env, ptr, ptr)
+DEF_HELPER_FLAGS_3(mve_vfcmpges, TCG_CALL_NO_WG, void, env, ptr, ptr)
+
+DEF_HELPER_FLAGS_3(mve_vfcmplth, TCG_CALL_NO_WG, void, env, ptr, ptr)
+DEF_HELPER_FLAGS_3(mve_vfcmplts, TCG_CALL_NO_WG, void, env, ptr, ptr)
+
+DEF_HELPER_FLAGS_3(mve_vfcmpgth, TCG_CALL_NO_WG, void, env, ptr, ptr)
+DEF_HELPER_FLAGS_3(mve_vfcmpgts, TCG_CALL_NO_WG, void, env, ptr, ptr)
+
+DEF_HELPER_FLAGS_3(mve_vfcmpleh, TCG_CALL_NO_WG, void, env, ptr, ptr)
+DEF_HELPER_FLAGS_3(mve_vfcmples, TCG_CALL_NO_WG, void, env, ptr, ptr)
+
 DEF_HELPER_FLAGS_4(mve_vfadd_scalarh, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
 DEF_HELPER_FLAGS_4(mve_vfadd_scalars, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
 
diff --git a/target/arm/mve.decode b/target/arm/mve.decode
index a46372f8c77..49b7ef35937 100644
--- a/target/arm/mve.decode
+++ b/target/arm/mve.decode
@@ -124,6 +124,9 @@
 @vcmp_scalar .... .... .. size:2 qn:3 . .... .... .... rm:4 &vcmp_scalar \
              mask=%mask_22_13
 
+@vcmp_fp .... .... .... qn:3 . .... .... .... .... &vcmp \
+         qm=%qm size=%2op_fp_scalar_size mask=%mask_22_13
+
 @vmaxv .... .... .... size:2 .. rda:4 .... .... .... &vmaxv qm=%qm
 
 @2op_fp .... .... .... .... .... .... .... .... &2op \
@@ -671,17 +674,41 @@ VSHLC             111 0 1110 1 . 1 imm:5 ... 0 1111 1100 rdm:4 qd=%qd
 # Comparisons. We expand out the conditions which are split across
 # encodings T1, T2, T3 and the fc bits. These include VPT, which is
 # effectively "VCMP then VPST". A plain "VCMP" has a mask field of zero.
-VCMPEQ            1111 1110 0 . .. ... 1 ... 0 1111 0 0 . 0 ... 0 @vcmp
-VCMPNE            1111 1110 0 . .. ... 1 ... 0 1111 1 0 . 0 ... 0 @vcmp
+{
+  VCMPEQ_fp       111 . 1110 0 . 11 ... 1 ... 0 1111 0 0 . 0 ... 0 @vcmp_fp
+  VCMPEQ          111 1 1110 0 . .. ... 1 ... 0 1111 0 0 . 0 ... 0 @vcmp
+}
+
+{
+  VCMPNE_fp       111 . 1110 0 . 11 ... 1 ... 0 1111 1 0 . 0 ... 0 @vcmp_fp
+  VCMPNE          111 1 1110 0 . .. ... 1 ... 0 1111 1 0 . 0 ... 0 @vcmp
+}
+
+{
+  VCMPGE_fp       111 . 1110 0 . 11 ... 1 ... 1 1111 0 0 . 0 ... 0 @vcmp_fp
+  VCMPGE          111 1 1110 0 . .. ... 1 ... 1 1111 0 0 . 0 ... 0 @vcmp
+}
+
+{
+  VCMPLT_fp       111 . 1110 0 . 11 ... 1 ... 1 1111 1 0 . 0 ... 0 @vcmp_fp
+  VCMPLT          111 1 1110 0 . .. ... 1 ... 1 1111 1 0 . 0 ... 0 @vcmp
+}
+
+{
+  VCMPGT_fp       111 . 1110 0 . 11 ... 1 ... 1 1111 0 0 . 0 ... 1 @vcmp_fp
+  VCMPGT          111 1 1110 0 . .. ... 1 ... 1 1111 0 0 . 0 ... 1 @vcmp
+}
+
+{
+  VCMPLE_fp         111 . 1110 0 . 11 ... 1 ... 1 1111 1 0 . 0 ... 1 @vcmp_fp
+  VCMPLE            1111 1110 0 . .. ... 1 ... 1 1111 1 0 . 0 ... 1 @vcmp
+}
+
 {
   VPSEL           1111 1110 0 . 11 ... 1 ... 0 1111 . 0 . 0 ... 1 @2op_nosz
   VCMPCS          1111 1110 0 . .. ... 1 ... 0 1111 0 0 . 0 ... 1 @vcmp
   VCMPHI          1111 1110 0 . .. ... 1 ... 0 1111 1 0 . 0 ... 1 @vcmp
 }
-VCMPGE            1111 1110 0 . .. ... 1 ... 1 1111 0 0 . 0 ... 0 @vcmp
-VCMPLT            1111 1110 0 . .. ... 1 ... 1 1111 1 0 . 0 ... 0 @vcmp
-VCMPGT            1111 1110 0 . .. ... 1 ... 1 1111 0 0 . 0 ... 1 @vcmp
-VCMPLE            1111 1110 0 . .. ... 1 ... 1 1111 1 0 . 0 ... 1 @vcmp
 
 {
   VPNOT           1111 1110 0 0 11 000 1 000 0 1111 0100 1101
diff --git a/target/arm/mve_helper.c b/target/arm/mve_helper.c
index 52e5a8f2a8b..07a1ab88814 100644
--- a/target/arm/mve_helper.c
+++ b/target/arm/mve_helper.c
@@ -3156,3 +3156,67 @@ DO_FP_VMAXMINV(vmaxnmavh, 2, float16, true, float16_maxnum)
 DO_FP_VMAXMINV(vmaxnmavs, 4, float32, true, float32_maxnum)
 DO_FP_VMAXMINV(vminnmavh, 2, float16, true, float16_minnum)
 DO_FP_VMAXMINV(vminnmavs, 4, float32, true, float32_minnum)
+
+/* FP compares; note that all comparisons signal InvalidOp for QNaNs */
+#define DO_VCMP_FP(OP, ESIZE, TYPE, FN)                                 \
+    void HELPER(glue(mve_, OP))(CPUARMState *env, void *vn, void *vm)   \
+    {                                                                   \
+        TYPE *n = vn, *m = vm;                                          \
+        uint16_t mask = mve_element_mask(env);                          \
+        uint16_t eci_mask = mve_eci_mask(env);                          \
+        uint16_t beatpred = 0;                                          \
+        uint16_t emask = MAKE_64BIT_MASK(0, ESIZE);                     \
+        unsigned e;                                                     \
+        float_status *fpst;                                             \
+        float_status scratch_fpst;                                      \
+        bool r;                                                         \
+        for (e = 0; e < 16 / ESIZE; e++, emask <<= ESIZE) {             \
+            if ((mask & emask) == 0) {                                  \
+                continue;                                               \
+            }                                                           \
+            fpst = (ESIZE == 2) ? &env->vfp.standard_fp_status_f16 :    \
+                &env->vfp.standard_fp_status;                           \
+            if (!(mask & (1 << (e * ESIZE)))) {                         \
+                /* We need the result but without updating flags */     \
+                scratch_fpst = *fpst;                                   \
+                fpst = &scratch_fpst;                                   \
+            }                                                           \
+            r = FN(n[H##ESIZE(e)], m[H##ESIZE(e)], fpst);               \
+            /* Comparison sets 0/1 bits for each byte in the element */ \
+            beatpred |= r * emask;                                      \
+        }                                                               \
+        beatpred &= mask;                                               \
+        env->v7m.vpr = (env->v7m.vpr & ~(uint32_t)eci_mask) |           \
+            (beatpred & eci_mask);                                      \
+        mve_advance_vpt(env);                                           \
+    }
+
+/*
+ * Some care is needed here to get the correct result for the unordered case.
+ * Architecturally EQ, GE and GT are defined to be false for unordered, but
+ * the NE, LT and LE comparisons are defined as simple logical inverses of
+ * EQ, GE and GT and so they must return true for unordered. The softfloat
+ * comparison functions float*_{eq,le,lt} all return false for unordered.
+ */
+#define DO_GE16(X, Y, S) float16_le(Y, X, S)
+#define DO_GE32(X, Y, S) float32_le(Y, X, S)
+#define DO_GT16(X, Y, S) float16_lt(Y, X, S)
+#define DO_GT32(X, Y, S) float32_lt(Y, X, S)
+
+DO_VCMP_FP(vfcmpeqh, 2, float16, float16_eq)
+DO_VCMP_FP(vfcmpeqs, 4, float32, float32_eq)
+
+DO_VCMP_FP(vfcmpneh, 2, float16, !float16_eq)
+DO_VCMP_FP(vfcmpnes, 4, float32, !float32_eq)
+
+DO_VCMP_FP(vfcmpgeh, 2, float16, DO_GE16)
+DO_VCMP_FP(vfcmpges, 4, float32, DO_GE32)
+
+DO_VCMP_FP(vfcmplth, 2, float16, !DO_GE16)
+DO_VCMP_FP(vfcmplts, 4, float32, !DO_GE32)
+
+DO_VCMP_FP(vfcmpgth, 2, float16, DO_GT16)
+DO_VCMP_FP(vfcmpgts, 4, float32, DO_GT32)
+
+DO_VCMP_FP(vfcmpleh, 2, float16, !DO_GT16)
+DO_VCMP_FP(vfcmples, 4, float32, !DO_GT32)
diff --git a/target/arm/translate-mve.c b/target/arm/translate-mve.c
index 4e2aa2cae2d..da14a6f790e 100644
--- a/target/arm/translate-mve.c
+++ b/target/arm/translate-mve.c
@@ -1758,6 +1758,28 @@ DO_VCMP(VCMPLT, vcmplt)
 DO_VCMP(VCMPGT, vcmpgt)
 DO_VCMP(VCMPLE, vcmple)
 
+#define DO_VCMP_FP(INSN, FN)                                    \
+    static bool trans_##INSN(DisasContext *s, arg_vcmp *a)      \
+    {                                                           \
+        static MVEGenCmpFn * const fns[] = {                    \
+            NULL,                                               \
+            gen_helper_mve_##FN##h,                             \
+            gen_helper_mve_##FN##s,                             \
+            NULL,                                               \
+        };                                                      \
+        if (!dc_isar_feature(aa32_mve_fp, s)) {                 \
+            return false;                                       \
+        }                                                       \
+        return do_vcmp(s, a, fns[a->size]);                     \
+    }
+
+DO_VCMP_FP(VCMPEQ_fp, vfcmpeq)
+DO_VCMP_FP(VCMPNE_fp, vfcmpne)
+DO_VCMP_FP(VCMPGE_fp, vfcmpge)
+DO_VCMP_FP(VCMPLT_fp, vfcmplt)
+DO_VCMP_FP(VCMPGT_fp, vfcmpgt)
+DO_VCMP_FP(VCMPLE_fp, vfcmple)
+
 static bool do_vmaxv(DisasContext *s, arg_vmaxv *a, MVEGenVADDVFn fn)
 {
     /*
-- 
2.20.1



  parent reply	other threads:[~2021-09-01 11:03 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-01 10:36 [PULL 00/51] target-arm queue Peter Maydell
2021-09-01 10:36 ` [PULL 01/51] tests: Remove uses of deprecated raspi2/raspi3 machine names Peter Maydell
2021-09-01 10:36 ` [PULL 02/51] hw/arm/raspi: Remove deprecated raspi2/raspi3 aliases Peter Maydell
2021-09-01 10:36 ` [PULL 03/51] hw/intc/arm_gicv3_dist: Rename 64-bit accessors with 'q' suffix Peter Maydell
2021-09-01 10:36 ` [PULL 04/51] hw/intc/arm_gicv3: Replace mis-used MEMTX_* constants by booleans Peter Maydell
2021-09-01 10:36 ` [PULL 05/51] hw: Add compat machines for 6.2 Peter Maydell
2021-09-01 10:36 ` [PULL 06/51] target/arm: Implement MVE VADD (floating-point) Peter Maydell
2021-09-01 10:36 ` [PULL 07/51] target/arm: Implement MVE VSUB, VMUL, VABD, VMAXNM, VMINNM Peter Maydell
2021-09-01 10:36 ` [PULL 08/51] target/arm: Implement MVE VCADD Peter Maydell
2021-09-01 10:36 ` [PULL 09/51] target/arm: Implement MVE VFMA and VFMS Peter Maydell
2021-09-01 10:36 ` [PULL 10/51] target/arm: Implement MVE VCMUL and VCMLA Peter Maydell
2021-09-01 10:36 ` [PULL 11/51] target/arm: Implement MVE VMAXNMA and VMINNMA Peter Maydell
2021-09-01 10:36 ` [PULL 12/51] target/arm: Implement MVE scalar fp insns Peter Maydell
2021-09-01 10:36 ` [PULL 13/51] target/arm: Implement MVE fp-with-scalar VFMA, VFMAS Peter Maydell
2021-09-01 10:36 ` [PULL 14/51] softfloat: Remove assertion preventing silencing of NaN in default-NaN mode Peter Maydell
2021-09-01 10:36 ` [PULL 15/51] target/arm: Implement MVE FP max/min across vector Peter Maydell
2021-09-01 10:36 ` Peter Maydell [this message]
2021-09-01 10:36 ` [PULL 17/51] target/arm: Implement MVE fp scalar comparisons Peter Maydell
2021-09-01 10:36 ` [PULL 18/51] target/arm: Implement MVE VCVT between floating and fixed point Peter Maydell
2021-09-01 10:36 ` [PULL 19/51] target/arm: Implement MVE VCVT between fp and integer Peter Maydell
2021-09-01 10:36 ` [PULL 20/51] target/arm: Implement MVE VCVT with specified rounding mode Peter Maydell
2021-09-01 10:36 ` [PULL 21/51] target/arm: Implement MVE VCVT between single and half precision Peter Maydell
2021-09-01 10:36 ` [PULL 22/51] target/arm: Implement MVE VRINT insns Peter Maydell
2021-09-01 10:36 ` [PULL 23/51] target/arm: Enable MVE in Cortex-M55 Peter Maydell
2021-09-01 10:36 ` [PULL 24/51] target-arm: Add support for Fujitsu A64FX Peter Maydell
2021-09-01 10:36 ` [PULL 25/51] hw/arm/virt: target-arm: Add A64FX processor support to virt machine Peter Maydell
2021-09-01 10:36 ` [PULL 26/51] tests/arm-cpu-features: Add A64FX processor related tests Peter Maydell
2021-09-01 10:36 ` [PULL 27/51] arm: Move M-profile RAS register block into its own device Peter Maydell
2021-09-01 10:36 ` [PULL 28/51] arm: Move systick device creation from NVIC to ARMv7M object Peter Maydell
2021-09-01 10:36 ` [PULL 29/51] arm: Move system PPB container handling to armv7m Peter Maydell
2021-09-01 10:36 ` [PULL 30/51] hw/timer/armv7m_systick: Add usual QEMU interface comment Peter Maydell
2021-09-01 10:36 ` [PULL 31/51] hw/timer/armv7m_systick: Add input clocks Peter Maydell
2021-09-01 10:36 ` [PULL 32/51] hw/arm/armv7m: Create " Peter Maydell
2021-09-01 10:36 ` [PULL 33/51] armsse: Wire up systick cpuclk clock Peter Maydell
2021-09-01 10:36 ` [PULL 34/51] hw/arm/mps2.c: Connect up armv7m clocks Peter Maydell
2021-09-01 10:36 ` [PULL 35/51] clock: Provide builtin multiplier/divider Peter Maydell
2021-09-01 10:36 ` [PULL 36/51] hw/arm: Don't allocate separate MemoryRegions in stm32 SoC realize Peter Maydell
2021-09-01 10:36 ` [PULL 37/51] hw/arm/stm32f100: Wire up sysclk and refclk Peter Maydell
2021-09-01 10:36 ` [PULL 38/51] hw/arm/stm32f205: " Peter Maydell
2021-09-01 10:36 ` [PULL 39/51] hw/arm/stm32f405: " Peter Maydell
2021-09-01 10:36 ` [PULL 40/51] hw/arm/stm32vldiscovery: Delete trailing blank line Peter Maydell
2021-09-01 10:36 ` [PULL 41/51] hw/arm/nrf51: Wire up sysclk Peter Maydell
2021-09-01 10:36 ` [PULL 42/51] hw/arm/stellaris: split stellaris_sys_init() Peter Maydell
2021-09-01 10:36 ` [PULL 43/51] hw/arm/stellaris: Wire sysclk up to armv7m Peter Maydell
2021-09-01 10:36 ` [PULL 44/51] hw/arm/msf2_soc: Don't allocate separate MemoryRegions Peter Maydell
2021-09-01 10:36 ` [PULL 45/51] hw/arm/msf2: Use Clock input to MSF2_SOC instead of m3clk property Peter Maydell
2021-09-01 10:36 ` [PULL 46/51] hw/arm/msf2-soc: Wire up refclk Peter Maydell
2021-09-01 10:36 ` [PULL 47/51] hw/timer/armv7m_systick: Use clock inputs instead of system_clock_scale Peter Maydell
2021-09-01 10:36 ` [PULL 48/51] hw/arm/stellaris: Fix code style issues in GPTM code Peter Maydell
2021-09-01 10:36 ` [PULL 49/51] hw/arm/stellaris: Split stellaris-gptm into its own file Peter Maydell
2021-09-01 10:36 ` [PULL 50/51] hw/timer/stellaris-gptm: Use Clock input instead of system_clock_scale Peter Maydell
2021-09-01 10:36 ` [PULL 51/51] arm: Remove system_clock_scale global Peter Maydell
2021-09-02  7:48 ` [PULL 00/51] target-arm queue Peter Maydell

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210901103653.13435-17-peter.maydell@linaro.org \
    --to=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.