qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Peter Maydell <peter.maydell@linaro.org>
To: qemu-devel@nongnu.org
Subject: [PULL 24/27] target/arm: Optimize MVE VSHL, VSHR immediate forms
Date: Mon, 20 Sep 2021 15:19:44 +0100	[thread overview]
Message-ID: <20210920141947.5537-25-peter.maydell@linaro.org> (raw)
In-Reply-To: <20210920141947.5537-1-peter.maydell@linaro.org>

Optimize the MVE VSHL and VSHR immediate forms by using TCG vector
ops when possible.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20210913095440.13462-10-peter.maydell@linaro.org
---
 target/arm/translate-mve.c | 83 +++++++++++++++++++++++++++++---------
 1 file changed, 63 insertions(+), 20 deletions(-)

diff --git a/target/arm/translate-mve.c b/target/arm/translate-mve.c
index 4583e22f21c..00fa4379a74 100644
--- a/target/arm/translate-mve.c
+++ b/target/arm/translate-mve.c
@@ -1570,8 +1570,8 @@ static bool trans_Vimm_1r(DisasContext *s, arg_1imm *a)
     return do_1imm(s, a, fn);
 }
 
-static bool do_2shift(DisasContext *s, arg_2shift *a, MVEGenTwoOpShiftFn fn,
-                      bool negateshift)
+static bool do_2shift_vec(DisasContext *s, arg_2shift *a, MVEGenTwoOpShiftFn fn,
+                          bool negateshift, GVecGen2iFn vecfn)
 {
     TCGv_ptr qd, qm;
     int shift = a->shift;
@@ -1594,34 +1594,77 @@ static bool do_2shift(DisasContext *s, arg_2shift *a, MVEGenTwoOpShiftFn fn,
         shift = -shift;
     }
 
-    qd = mve_qreg_ptr(a->qd);
-    qm = mve_qreg_ptr(a->qm);
-    fn(cpu_env, qd, qm, tcg_constant_i32(shift));
-    tcg_temp_free_ptr(qd);
-    tcg_temp_free_ptr(qm);
+    if (vecfn && mve_no_predication(s)) {
+        vecfn(a->size, mve_qreg_offset(a->qd), mve_qreg_offset(a->qm),
+              shift, 16, 16);
+    } else {
+        qd = mve_qreg_ptr(a->qd);
+        qm = mve_qreg_ptr(a->qm);
+        fn(cpu_env, qd, qm, tcg_constant_i32(shift));
+        tcg_temp_free_ptr(qd);
+        tcg_temp_free_ptr(qm);
+    }
     mve_update_eci(s);
     return true;
 }
 
-#define DO_2SHIFT(INSN, FN, NEGATESHIFT)                         \
-    static bool trans_##INSN(DisasContext *s, arg_2shift *a)    \
-    {                                                           \
-        static MVEGenTwoOpShiftFn * const fns[] = {             \
-            gen_helper_mve_##FN##b,                             \
-            gen_helper_mve_##FN##h,                             \
-            gen_helper_mve_##FN##w,                             \
-            NULL,                                               \
-        };                                                      \
-        return do_2shift(s, a, fns[a->size], NEGATESHIFT);      \
+static bool do_2shift(DisasContext *s, arg_2shift *a, MVEGenTwoOpShiftFn fn,
+                      bool negateshift)
+{
+    return do_2shift_vec(s, a, fn, negateshift, NULL);
+}
+
+#define DO_2SHIFT_VEC(INSN, FN, NEGATESHIFT, VECFN)                     \
+    static bool trans_##INSN(DisasContext *s, arg_2shift *a)            \
+    {                                                                   \
+        static MVEGenTwoOpShiftFn * const fns[] = {                     \
+            gen_helper_mve_##FN##b,                                     \
+            gen_helper_mve_##FN##h,                                     \
+            gen_helper_mve_##FN##w,                                     \
+            NULL,                                                       \
+        };                                                              \
+        return do_2shift_vec(s, a, fns[a->size], NEGATESHIFT, VECFN);   \
     }
 
-DO_2SHIFT(VSHLI, vshli_u, false)
+#define DO_2SHIFT(INSN, FN, NEGATESHIFT)        \
+    DO_2SHIFT_VEC(INSN, FN, NEGATESHIFT, NULL)
+
+static void do_gvec_shri_s(unsigned vece, uint32_t dofs, uint32_t aofs,
+                           int64_t shift, uint32_t oprsz, uint32_t maxsz)
+{
+    /*
+     * We get here with a negated shift count, and we must handle
+     * shifts by the element size, which tcg_gen_gvec_sari() does not do.
+     */
+    shift = -shift;
+    if (shift == (8 << vece)) {
+        shift--;
+    }
+    tcg_gen_gvec_sari(vece, dofs, aofs, shift, oprsz, maxsz);
+}
+
+static void do_gvec_shri_u(unsigned vece, uint32_t dofs, uint32_t aofs,
+                           int64_t shift, uint32_t oprsz, uint32_t maxsz)
+{
+    /*
+     * We get here with a negated shift count, and we must handle
+     * shifts by the element size, which tcg_gen_gvec_shri() does not do.
+     */
+    shift = -shift;
+    if (shift == (8 << vece)) {
+        tcg_gen_gvec_dup_imm(vece, dofs, oprsz, maxsz, 0);
+    } else {
+        tcg_gen_gvec_shri(vece, dofs, aofs, shift, oprsz, maxsz);
+    }
+}
+
+DO_2SHIFT_VEC(VSHLI, vshli_u, false, tcg_gen_gvec_shli)
 DO_2SHIFT(VQSHLI_S, vqshli_s, false)
 DO_2SHIFT(VQSHLI_U, vqshli_u, false)
 DO_2SHIFT(VQSHLUI, vqshlui_s, false)
 /* These right shifts use a left-shift helper with negated shift count */
-DO_2SHIFT(VSHRI_S, vshli_s, true)
-DO_2SHIFT(VSHRI_U, vshli_u, true)
+DO_2SHIFT_VEC(VSHRI_S, vshli_s, true, do_gvec_shri_s)
+DO_2SHIFT_VEC(VSHRI_U, vshli_u, true, do_gvec_shri_u)
 DO_2SHIFT(VRSHRI_S, vrshli_s, true)
 DO_2SHIFT(VRSHRI_U, vrshli_u, true)
 
-- 
2.20.1



  parent reply	other threads:[~2021-09-20 14:27 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-20 14:19 [PULL 00/27] target-arm queue Peter Maydell
2021-09-20 14:19 ` [PULL 01/27] elf2dmp: Check curl_easy_setopt() return value Peter Maydell
2021-09-20 14:19 ` [PULL 02/27] elf2dmp: Fail cleanly if PDB file specifies zero block_size Peter Maydell
2021-09-20 14:19 ` [PULL 03/27] target/arm: Don't skip M-profile reset entirely in user mode Peter Maydell
2021-09-20 14:19 ` [PULL 04/27] target/arm: Always clear exclusive monitor on reset Peter Maydell
2021-09-20 14:19 ` [PULL 05/27] target/arm: Consolidate ifdef blocks in reset Peter Maydell
2021-09-20 14:19 ` [PULL 06/27] hw/intc: Set GIC maintenance interrupt level to only 0 or 1 Peter Maydell
2021-09-20 14:19 ` [PULL 07/27] arm: Move PMC register definitions to internals.h Peter Maydell
2021-09-20 14:19 ` [PULL 08/27] hvf: Add execute to dirty log permission bitmap Peter Maydell
2021-09-20 14:19 ` [PULL 09/27] hvf: Introduce hvf_arch_init() callback Peter Maydell
2021-09-20 14:19 ` [PULL 10/27] hvf: Add Apple Silicon support Peter Maydell
2021-09-20 14:19 ` [PULL 11/27] arm/hvf: Add a WFI handler Peter Maydell
2021-09-20 14:19 ` [PULL 12/27] hvf: arm: Implement -cpu host Peter Maydell
2021-09-20 14:19 ` [PULL 13/27] hvf: arm: Implement PSCI handling Peter Maydell
2021-09-20 14:19 ` [PULL 14/27] arm: Add Hypervisor.framework build target Peter Maydell
2021-09-20 14:19 ` [PULL 15/27] hvf: arm: Add rudimentary PMC support Peter Maydell
2021-09-20 14:19 ` [PULL 16/27] target/arm: Avoid goto_tb if we're trying to exit to the main loop Peter Maydell
2021-09-20 14:19 ` [PULL 17/27] target/arm: Enforce that FPDSCR.LTPSIZE is 4 on inbound migration Peter Maydell
2021-09-20 14:19 ` [PULL 18/27] target/arm: Add TB flag for "MVE insns not predicated" Peter Maydell
2021-09-20 14:19 ` [PULL 19/27] target/arm: Optimize MVE logic ops Peter Maydell
2021-09-20 14:19 ` [PULL 20/27] target/arm: Optimize MVE arithmetic ops Peter Maydell
2021-09-20 14:19 ` [PULL 21/27] target/arm: Optimize MVE VNEG, VABS Peter Maydell
2021-09-20 14:19 ` [PULL 22/27] target/arm: Optimize MVE VDUP Peter Maydell
2021-09-20 14:19 ` [PULL 23/27] target/arm: Optimize MVE VMVN Peter Maydell
2021-09-20 14:19 ` Peter Maydell [this message]
2021-09-20 14:19 ` [PULL 25/27] target/arm: Optimize MVE VSHLL and VMOVL Peter Maydell
2021-09-20 14:19 ` [PULL 26/27] target/arm: Optimize MVE VSLI and VSRI Peter Maydell
2021-09-20 14:19 ` [PULL 27/27] target/arm: Optimize MVE 1op-immediate insns 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=20210920141947.5537-25-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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).