All of lore.kernel.org
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: Peter Maydell <peter.maydell@linaro.org>,
	qemu-arm@nongnu.org, Stephen Long <steplong@quicinc.com>
Subject: [PATCH v7 46/92] target/arm: Implement SVE2 FMMLA
Date: Mon, 24 May 2021 18:03:12 -0700	[thread overview]
Message-ID: <20210525010358.152808-47-richard.henderson@linaro.org> (raw)
In-Reply-To: <20210525010358.152808-1-richard.henderson@linaro.org>

From: Stephen Long <steplong@quicinc.com>

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Stephen Long <steplong@quicinc.com>
Message-Id: <20200422165503.13511-1-steplong@quicinc.com>
[rth: Fix indexing in helpers, expand macro to straight functions.]
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 target/arm/cpu.h           | 10 ++++++
 target/arm/helper-sve.h    |  3 ++
 target/arm/sve.decode      |  4 +++
 target/arm/sve_helper.c    | 74 ++++++++++++++++++++++++++++++++++++++
 target/arm/translate-sve.c | 34 ++++++++++++++++++
 5 files changed, 125 insertions(+)

diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index ae787fac8a..595bc6349d 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -4246,6 +4246,16 @@ static inline bool isar_feature_aa64_sve2_bitperm(const ARMISARegisters *id)
     return FIELD_EX64(id->id_aa64zfr0, ID_AA64ZFR0, BITPERM) != 0;
 }
 
+static inline bool isar_feature_aa64_sve_f32mm(const ARMISARegisters *id)
+{
+    return FIELD_EX64(id->id_aa64zfr0, ID_AA64ZFR0, F32MM) != 0;
+}
+
+static inline bool isar_feature_aa64_sve_f64mm(const ARMISARegisters *id)
+{
+    return FIELD_EX64(id->id_aa64zfr0, ID_AA64ZFR0, F64MM) != 0;
+}
+
 /*
  * Feature tests for "does this exist in either 32-bit or 64-bit?"
  */
diff --git a/target/arm/helper-sve.h b/target/arm/helper-sve.h
index 28b8f00201..7e99dcd119 100644
--- a/target/arm/helper-sve.h
+++ b/target/arm/helper-sve.h
@@ -2662,3 +2662,6 @@ DEF_HELPER_FLAGS_5(sve2_sqrdcmlah_zzzz_s, TCG_CALL_NO_RWG,
                    void, ptr, ptr, ptr, ptr, i32)
 DEF_HELPER_FLAGS_5(sve2_sqrdcmlah_zzzz_d, TCG_CALL_NO_RWG,
                    void, ptr, ptr, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_6(fmmla_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_6(fmmla_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, ptr, i32)
diff --git a/target/arm/sve.decode b/target/arm/sve.decode
index c3958bed6a..cb2ee86228 100644
--- a/target/arm/sve.decode
+++ b/target/arm/sve.decode
@@ -1389,6 +1389,10 @@ UMLSLT_zzzw     01000100 .. 0 ..... 010 111 ..... .....  @rda_rn_rm
 CMLA_zzzz       01000100 esz:2 0 rm:5 0010 rot:2 rn:5 rd:5  ra=%reg_movprfx
 SQRDCMLAH_zzzz  01000100 esz:2 0 rm:5 0011 rot:2 rn:5 rd:5  ra=%reg_movprfx
 
+### SVE2 floating point matrix multiply accumulate
+
+FMMLA           01100100 .. 1 ..... 111001 ..... .....  @rda_rn_rm
+
 ### SVE2 Memory Gather Load Group
 
 # SVE2 64-bit gather non-temporal load
diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c
index 5b6292929e..fa96e28639 100644
--- a/target/arm/sve_helper.c
+++ b/target/arm/sve_helper.c
@@ -7241,3 +7241,77 @@ void HELPER(sve2_xar_s)(void *vd, void *vn, void *vm, uint32_t desc)
         d[i] = ror32(n[i] ^ m[i], shr);
     }
 }
+
+void HELPER(fmmla_s)(void *vd, void *vn, void *vm, void *va,
+                     void *status, uint32_t desc)
+{
+    intptr_t s, opr_sz = simd_oprsz(desc) / (sizeof(float32) * 4);
+
+    for (s = 0; s < opr_sz; ++s) {
+        float32 *n = vn + s * sizeof(float32) * 4;
+        float32 *m = vm + s * sizeof(float32) * 4;
+        float32 *a = va + s * sizeof(float32) * 4;
+        float32 *d = vd + s * sizeof(float32) * 4;
+        float32 n00 = n[H4(0)], n01 = n[H4(1)];
+        float32 n10 = n[H4(2)], n11 = n[H4(3)];
+        float32 m00 = m[H4(0)], m01 = m[H4(1)];
+        float32 m10 = m[H4(2)], m11 = m[H4(3)];
+        float32 p0, p1;
+
+        /* i = 0, j = 0 */
+        p0 = float32_mul(n00, m00, status);
+        p1 = float32_mul(n01, m01, status);
+        d[H4(0)] = float32_add(a[H4(0)], float32_add(p0, p1, status), status);
+
+        /* i = 0, j = 1 */
+        p0 = float32_mul(n00, m10, status);
+        p1 = float32_mul(n01, m11, status);
+        d[H4(1)] = float32_add(a[H4(1)], float32_add(p0, p1, status), status);
+
+        /* i = 1, j = 0 */
+        p0 = float32_mul(n10, m00, status);
+        p1 = float32_mul(n11, m01, status);
+        d[H4(2)] = float32_add(a[H4(2)], float32_add(p0, p1, status), status);
+
+        /* i = 1, j = 1 */
+        p0 = float32_mul(n10, m10, status);
+        p1 = float32_mul(n11, m11, status);
+        d[H4(3)] = float32_add(a[H4(3)], float32_add(p0, p1, status), status);
+    }
+}
+
+void HELPER(fmmla_d)(void *vd, void *vn, void *vm, void *va,
+                     void *status, uint32_t desc)
+{
+    intptr_t s, opr_sz = simd_oprsz(desc) / (sizeof(float64) * 4);
+
+    for (s = 0; s < opr_sz; ++s) {
+        float64 *n = vn + s * sizeof(float64) * 4;
+        float64 *m = vm + s * sizeof(float64) * 4;
+        float64 *a = va + s * sizeof(float64) * 4;
+        float64 *d = vd + s * sizeof(float64) * 4;
+        float64 n00 = n[0], n01 = n[1], n10 = n[2], n11 = n[3];
+        float64 m00 = m[0], m01 = m[1], m10 = m[2], m11 = m[3];
+        float64 p0, p1;
+
+        /* i = 0, j = 0 */
+        p0 = float64_mul(n00, m00, status);
+        p1 = float64_mul(n01, m01, status);
+        d[0] = float64_add(a[0], float64_add(p0, p1, status), status);
+
+        /* i = 0, j = 1 */
+        p0 = float64_mul(n00, m10, status);
+        p1 = float64_mul(n01, m11, status);
+        d[1] = float64_add(a[1], float64_add(p0, p1, status), status);
+
+        /* i = 1, j = 0 */
+        p0 = float64_mul(n10, m00, status);
+        p1 = float64_mul(n11, m01, status);
+        d[2] = float64_add(a[2], float64_add(p0, p1, status), status);
+
+        /* i = 1, j = 1 */
+        p0 = float64_mul(n10, m10, status);
+        p1 = float64_mul(n11, m11, status);
+        d[3] = float64_add(a[3], float64_add(p0, p1, status), status);
+    }
+}
diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c
index a64ad04c50..a94b399f67 100644
--- a/target/arm/translate-sve.c
+++ b/target/arm/translate-sve.c
@@ -7672,6 +7672,40 @@ DO_SVE2_ZPZZ_FP(FMINP, fminp)
  * SVE Integer Multiply-Add (unpredicated)
  */
 
+static bool trans_FMMLA(DisasContext *s, arg_rrrr_esz *a)
+{
+    gen_helper_gvec_4_ptr *fn;
+
+    switch (a->esz) {
+    case MO_32:
+        if (!dc_isar_feature(aa64_sve_f32mm, s)) {
+            return false;
+        }
+        fn = gen_helper_fmmla_s;
+        break;
+    case MO_64:
+        if (!dc_isar_feature(aa64_sve_f64mm, s)) {
+            return false;
+        }
+        fn = gen_helper_fmmla_d;
+        break;
+    default:
+        return false;
+    }
+
+    if (sve_access_check(s)) {
+        unsigned vsz = vec_full_reg_size(s);
+        TCGv_ptr status = fpstatus_ptr(FPST_FPCR);
+        tcg_gen_gvec_4_ptr(vec_full_reg_offset(s, a->rd),
+                           vec_full_reg_offset(s, a->rn),
+                           vec_full_reg_offset(s, a->rm),
+                           vec_full_reg_offset(s, a->ra),
+                           status, vsz, vsz, 0, fn);
+        tcg_temp_free_ptr(status);
+    }
+    return true;
+}
+
 static bool do_sqdmlal_zzzw(DisasContext *s, arg_rrrr_esz *a,
                             bool sel1, bool sel2)
 {
-- 
2.25.1



  parent reply	other threads:[~2021-05-25  2:12 UTC|newest]

Thread overview: 95+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-25  1:02 [PATCH v7 00/92] target/arm: Implement SVE2 Richard Henderson
2021-05-25  1:02 ` [PATCH v7 01/92] target/arm: Add ID_AA64ZFR0 fields and isar_feature_aa64_sve2 Richard Henderson
2021-05-25  1:02 ` [PATCH v7 02/92] target/arm: Implement SVE2 Integer Multiply - Unpredicated Richard Henderson
2021-05-25  1:02 ` [PATCH v7 03/92] target/arm: Implement SVE2 integer pairwise add and accumulate long Richard Henderson
2021-05-25  1:02 ` [PATCH v7 04/92] target/arm: Implement SVE2 integer unary operations (predicated) Richard Henderson
2021-05-25  1:02 ` [PATCH v7 05/92] target/arm: Split out saturating/rounding shifts from neon Richard Henderson
2021-05-25  1:02 ` [PATCH v7 06/92] target/arm: Implement SVE2 saturating/rounding bitwise shift left (predicated) Richard Henderson
2021-05-25  1:02 ` [PATCH v7 07/92] target/arm: Implement SVE2 integer halving add/subtract (predicated) Richard Henderson
2021-05-25  1:02 ` [PATCH v7 08/92] target/arm: Implement SVE2 integer pairwise arithmetic Richard Henderson
2021-05-25  1:02 ` [PATCH v7 09/92] target/arm: Implement SVE2 saturating add/subtract (predicated) Richard Henderson
2021-05-25  1:02 ` [PATCH v7 10/92] target/arm: Implement SVE2 integer add/subtract long Richard Henderson
2021-05-25  1:02 ` [PATCH v7 11/92] target/arm: Implement SVE2 integer add/subtract interleaved long Richard Henderson
2021-05-25  1:02 ` [PATCH v7 12/92] target/arm: Implement SVE2 integer add/subtract wide Richard Henderson
2021-05-25  1:02 ` [PATCH v7 13/92] target/arm: Implement SVE2 integer multiply long Richard Henderson
2021-05-25  1:02 ` [PATCH v7 14/92] target/arm: Implement SVE2 PMULLB, PMULLT Richard Henderson
2021-05-25  1:02 ` [PATCH v7 15/92] target/arm: Implement SVE2 bitwise shift left long Richard Henderson
2021-05-25  1:02 ` [PATCH v7 16/92] target/arm: Implement SVE2 bitwise exclusive-or interleaved Richard Henderson
2021-05-25  1:02 ` [PATCH v7 17/92] target/arm: Implement SVE2 bitwise permute Richard Henderson
2021-05-25  1:02 ` [PATCH v7 18/92] target/arm: Implement SVE2 complex integer add Richard Henderson
2021-05-25  1:02 ` [PATCH v7 19/92] target/arm: Implement SVE2 integer absolute difference and accumulate long Richard Henderson
2021-05-25  1:02 ` [PATCH v7 20/92] target/arm: Implement SVE2 integer add/subtract long with carry Richard Henderson
2021-05-25  1:02 ` [PATCH v7 21/92] target/arm: Implement SVE2 bitwise shift right and accumulate Richard Henderson
2021-05-25  1:02 ` [PATCH v7 22/92] target/arm: Implement SVE2 bitwise shift and insert Richard Henderson
2021-05-25  1:02 ` [PATCH v7 23/92] target/arm: Implement SVE2 integer absolute difference and accumulate Richard Henderson
2021-05-25  1:02 ` [PATCH v7 24/92] target/arm: Implement SVE2 saturating extract narrow Richard Henderson
2021-05-25  1:02 ` [PATCH v7 25/92] target/arm: Implement SVE2 floating-point pairwise Richard Henderson
2021-05-25  1:02 ` [PATCH v7 26/92] target/arm: Implement SVE2 SHRN, RSHRN Richard Henderson
2021-05-25  1:02 ` [PATCH v7 27/92] target/arm: Implement SVE2 SQSHRUN, SQRSHRUN Richard Henderson
2021-05-25  1:02 ` [PATCH v7 28/92] target/arm: Implement SVE2 UQSHRN, UQRSHRN Richard Henderson
2021-05-25  1:02 ` [PATCH v7 29/92] target/arm: Implement SVE2 SQSHRN, SQRSHRN Richard Henderson
2021-05-25  1:02 ` [PATCH v7 30/92] target/arm: Implement SVE2 WHILEGT, WHILEGE, WHILEHI, WHILEHS Richard Henderson
2021-05-25  1:02 ` [PATCH v7 31/92] target/arm: Implement SVE2 WHILERW, WHILEWR Richard Henderson
2021-05-25  1:02 ` [PATCH v7 32/92] target/arm: Implement SVE2 bitwise ternary operations Richard Henderson
2021-05-25  1:02 ` [PATCH v7 33/92] target/arm: Implement SVE2 MATCH, NMATCH Richard Henderson
2021-05-25  1:03 ` [PATCH v7 34/92] target/arm: Implement SVE2 saturating multiply-add long Richard Henderson
2021-05-25  1:03 ` [PATCH v7 35/92] target/arm: Implement SVE2 saturating multiply-add high Richard Henderson
2021-05-25  1:03 ` [PATCH v7 36/92] target/arm: Implement SVE2 integer multiply-add long Richard Henderson
2021-05-25  1:03 ` [PATCH v7 37/92] target/arm: Implement SVE2 complex integer multiply-add Richard Henderson
2021-05-25  1:03 ` [PATCH v7 38/92] target/arm: Implement SVE2 ADDHNB, ADDHNT Richard Henderson
2021-05-25  1:03 ` [PATCH v7 39/92] target/arm: Implement SVE2 RADDHNB, RADDHNT Richard Henderson
2021-05-25  1:03 ` [PATCH v7 40/92] target/arm: Implement SVE2 SUBHNB, SUBHNT Richard Henderson
2021-05-25  1:03 ` [PATCH v7 41/92] target/arm: Implement SVE2 RSUBHNB, RSUBHNT Richard Henderson
2021-05-25  1:03 ` [PATCH v7 42/92] target/arm: Implement SVE2 HISTCNT, HISTSEG Richard Henderson
2021-05-25  1:03 ` [PATCH v7 43/92] target/arm: Implement SVE2 XAR Richard Henderson
2021-05-25  1:03 ` [PATCH v7 44/92] target/arm: Implement SVE2 scatter store insns Richard Henderson
2021-05-25  1:03 ` [PATCH v7 45/92] target/arm: Implement SVE2 gather load insns Richard Henderson
2021-05-25  1:03 ` Richard Henderson [this message]
2021-05-25  1:03 ` [PATCH v7 47/92] target/arm: Implement SVE2 SPLICE, EXT Richard Henderson
2021-05-25  1:03 ` [PATCH v7 48/92] target/arm: Use correct output type for gvec_sdot_*_b Richard Henderson
2021-05-25  1:03 ` [PATCH v7 49/92] target/arm: Pass separate addend to {U, S}DOT helpers Richard Henderson
2021-05-25  1:03 ` [PATCH v7 50/92] target/arm: Pass separate addend to FCMLA helpers Richard Henderson
2021-05-25  1:03 ` [PATCH v7 51/92] target/arm: Split out formats for 2 vectors + 1 index Richard Henderson
2021-05-25  1:03 ` [PATCH v7 52/92] target/arm: Split out formats for 3 " Richard Henderson
2021-05-25  1:03 ` [PATCH v7 53/92] target/arm: Implement SVE2 integer multiply (indexed) Richard Henderson
2021-05-25  1:03 ` [PATCH v7 54/92] target/arm: Implement SVE2 integer multiply-add (indexed) Richard Henderson
2021-05-25  1:03 ` [PATCH v7 55/92] target/arm: Implement SVE2 saturating multiply-add high (indexed) Richard Henderson
2021-05-25  1:03 ` [PATCH v7 56/92] target/arm: Implement SVE2 saturating multiply-add (indexed) Richard Henderson
2021-05-25  1:03 ` [PATCH v7 57/92] target/arm: Implement SVE2 saturating multiply (indexed) Richard Henderson
2021-05-25  1:03 ` [PATCH v7 58/92] target/arm: Implement SVE2 signed saturating doubling multiply high Richard Henderson
2021-05-25  1:03 ` [PATCH v7 59/92] target/arm: Implement SVE2 saturating multiply high (indexed) Richard Henderson
2021-05-25  1:03 ` [PATCH v7 60/92] target/arm: Implement SVE2 multiply-add long (indexed) Richard Henderson
2021-05-25  1:03 ` [PATCH v7 61/92] target/arm: Implement SVE2 integer multiply " Richard Henderson
2021-05-25  1:03 ` [PATCH v7 62/92] target/arm: Implement SVE2 complex integer multiply-add (indexed) Richard Henderson
2021-05-25  1:03 ` [PATCH v7 63/92] target/arm: Implement SVE2 complex integer dot product Richard Henderson
2021-05-25  1:03 ` [PATCH v7 64/92] target/arm: Macroize helper_gvec_{s,u}dot_{b,h} Richard Henderson
2021-05-25  1:03 ` [PATCH v7 65/92] target/arm: Macroize helper_gvec_{s,u}dot_idx_{b,h} Richard Henderson
2021-05-25  1:03 ` [PATCH v7 66/92] target/arm: Implement SVE mixed sign dot product (indexed) Richard Henderson
2021-05-25  1:03 ` [PATCH v7 67/92] target/arm: Implement SVE mixed sign dot product Richard Henderson
2021-05-25  1:03 ` [PATCH v7 68/92] target/arm: Implement SVE2 crypto unary operations Richard Henderson
2021-05-25  1:03 ` [PATCH v7 69/92] target/arm: Implement SVE2 crypto destructive binary operations Richard Henderson
2021-05-25  1:03 ` [PATCH v7 70/92] target/arm: Implement SVE2 crypto constructive " Richard Henderson
2021-05-25  1:03 ` [PATCH v7 71/92] target/arm: Implement SVE2 TBL, TBX Richard Henderson
2021-05-25  1:03 ` [PATCH v7 72/92] target/arm: Implement SVE2 FCVTNT Richard Henderson
2021-05-25  1:03 ` [PATCH v7 73/92] target/arm: Implement SVE2 FCVTLT Richard Henderson
2021-05-25  1:03 ` [PATCH v7 74/92] target/arm: Implement SVE2 FCVTXNT, FCVTX Richard Henderson
2021-05-25  1:03 ` [PATCH v7 75/92] target/arm: Implement SVE2 FLOGB Richard Henderson
2021-05-25  1:03 ` [PATCH v7 76/92] target/arm: Share table of sve load functions Richard Henderson
2021-05-25  1:03 ` [PATCH v7 77/92] target/arm: Tidy do_ldrq Richard Henderson
2021-05-25  1:03 ` [PATCH v7 78/92] target/arm: Implement SVE2 LD1RO Richard Henderson
2021-05-25  1:03 ` [PATCH v7 79/92] target/arm: Implement 128-bit ZIP, UZP, TRN Richard Henderson
2021-05-25  1:03 ` [PATCH v7 80/92] target/arm: Implement SVE2 bitwise shift immediate Richard Henderson
2021-05-25  1:03 ` [PATCH v7 81/92] target/arm: Move endian adjustment macros to vec_internal.h Richard Henderson
2021-05-25  1:03 ` [PATCH v7 82/92] target/arm: Implement SVE2 fp multiply-add long Richard Henderson
2021-05-25  1:03 ` [PATCH v7 83/92] target/arm: Implement aarch64 SUDOT, USDOT Richard Henderson
2021-05-25  1:03 ` [PATCH v7 84/92] target/arm: Split out do_neon_ddda_fpst Richard Henderson
2021-05-25  1:03 ` [PATCH v7 85/92] target/arm: Remove unused fpst from VDOT_scalar Richard Henderson
2021-05-25  1:03 ` [PATCH v7 86/92] target/arm: Fix decode for VDOT (indexed) Richard Henderson
2021-05-25  1:03 ` [PATCH v7 87/92] target/arm: Split out do_neon_ddda Richard Henderson
2021-05-25  1:03 ` [PATCH v7 88/92] target/arm: Split decode of VSDOT and VUDOT Richard Henderson
2021-05-25  1:03 ` [PATCH v7 89/92] target/arm: Implement aarch32 VSUDOT, VUSDOT Richard Henderson
2021-05-25  1:03 ` [PATCH v7 90/92] target/arm: Implement integer matrix multiply accumulate Richard Henderson
2021-05-25  1:03 ` [PATCH v7 91/92] linux-user/aarch64: Enable hwcap bits for sve2 and related extensions Richard Henderson
2021-05-25  1:03 ` [PATCH v7 92/92] target/arm: Enable SVE2 " Richard Henderson
2021-05-25  2:37 ` [PATCH v7 00/92] target/arm: Implement SVE2 no-reply
2021-05-25 12:33 ` 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=20210525010358.152808-47-richard.henderson@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=steplong@quicinc.com \
    --subject='Re: [PATCH v7 46/92] target/arm: Implement SVE2 FMMLA' \
    /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

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.