qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Subject: [PATCH v5 23/23] tcg/aarch64: Use tcg_constant_vec with tcg vec expanders
Date: Thu, 17 Dec 2020 08:52:15 -0600	[thread overview]
Message-ID: <20201217145215.534637-24-richard.henderson@linaro.org> (raw)
In-Reply-To: <20201217145215.534637-1-richard.henderson@linaro.org>

Improve rotrv_vec to reduce "t1 = -v2, t2 = t1 + c" to
"t1 = -v, t2 = c - v2".  This avoids a serial dependency
between t1 and t2.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 tcg/aarch64/tcg-target.c.inc | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/tcg/aarch64/tcg-target.c.inc b/tcg/aarch64/tcg-target.c.inc
index a6c2d54c4d..c78e809de8 100644
--- a/tcg/aarch64/tcg-target.c.inc
+++ b/tcg/aarch64/tcg-target.c.inc
@@ -2511,7 +2511,7 @@ void tcg_expand_vec_op(TCGOpcode opc, TCGType type, unsigned vece,
                        TCGArg a0, ...)
 {
     va_list va;
-    TCGv_vec v0, v1, v2, t1, t2;
+    TCGv_vec v0, v1, v2, t1, t2, c1;
     TCGArg a2;
 
     va_start(va, a0);
@@ -2543,8 +2543,8 @@ void tcg_expand_vec_op(TCGOpcode opc, TCGType type, unsigned vece,
 
     case INDEX_op_rotlv_vec:
         t1 = tcg_temp_new_vec(type);
-        tcg_gen_dupi_vec(vece, t1, 8 << vece);
-        tcg_gen_sub_vec(vece, t1, v2, t1);
+        c1 = tcg_constant_vec(type, vece, 8 << vece);
+        tcg_gen_sub_vec(vece, t1, v2, c1);
         /* Right shifts are negative left shifts for AArch64.  */
         vec_gen_3(INDEX_op_shlv_vec, type, vece, tcgv_vec_arg(t1),
                   tcgv_vec_arg(v1), tcgv_vec_arg(t1));
@@ -2557,9 +2557,9 @@ void tcg_expand_vec_op(TCGOpcode opc, TCGType type, unsigned vece,
     case INDEX_op_rotrv_vec:
         t1 = tcg_temp_new_vec(type);
         t2 = tcg_temp_new_vec(type);
+        c1 = tcg_constant_vec(type, vece, 8 << vece);
         tcg_gen_neg_vec(vece, t1, v2);
-        tcg_gen_dupi_vec(vece, t2, 8 << vece);
-        tcg_gen_add_vec(vece, t2, t1, t2);
+        tcg_gen_sub_vec(vece, t2, c1, v2);
         /* Right shifts are negative left shifts for AArch64.  */
         vec_gen_3(INDEX_op_shlv_vec, type, vece, tcgv_vec_arg(t1),
                   tcgv_vec_arg(v1), tcgv_vec_arg(t1));
-- 
2.25.1



      parent reply	other threads:[~2020-12-17 15:20 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-12-17 14:51 [PATCH v5 00/23] tcg: Better handling of constants Richard Henderson
2020-12-17 14:51 ` [PATCH v5 01/23] tcg: Use tcg_out_dupi_vec from temp_load Richard Henderson
2020-12-17 14:51 ` [PATCH v5 02/23] tcg: Increase tcg_out_dupi_vec immediate to int64_t Richard Henderson
2020-12-17 14:51 ` [PATCH v5 03/23] tcg: Consolidate 3 bits into enum TCGTempKind Richard Henderson
2020-12-17 14:51 ` [PATCH v5 04/23] tcg: Add temp_readonly Richard Henderson
2020-12-17 14:51 ` [PATCH v5 05/23] tcg: Expand TCGTemp.val to 64-bits Richard Henderson
2020-12-17 14:51 ` [PATCH v5 06/23] tcg: Rename struct tcg_temp_info to TempOptInfo Richard Henderson
2020-12-17 14:51 ` [PATCH v5 07/23] tcg: Expand TempOptInfo to 64-bits Richard Henderson
2020-12-17 14:52 ` [PATCH v5 08/23] tcg: Introduce TYPE_CONST temporaries Richard Henderson
2020-12-17 14:52 ` [PATCH v5 09/23] tcg/optimize: Improve find_better_copy Richard Henderson
2020-12-17 14:52 ` [PATCH v5 10/23] tcg/optimize: Adjust TempOptInfo allocation Richard Henderson
2020-12-17 14:52 ` [PATCH v5 11/23] tcg/optimize: Use tcg_constant_internal with constant folding Richard Henderson
2020-12-17 14:52 ` [PATCH v5 12/23] tcg: Convert tcg_gen_dupi_vec to TCG_CONST Richard Henderson
2020-12-17 14:52 ` [PATCH v5 13/23] tcg: Use tcg_constant_i32 with icount expander Richard Henderson
2020-12-17 14:52 ` [PATCH v5 14/23] tcg: Use tcg_constant_{i32, i64} with tcg int expanders Richard Henderson
2020-12-17 14:52 ` [PATCH v5 15/23] tcg: Use tcg_constant_{i32,i64} with tcg plugins Richard Henderson
2020-12-17 14:52 ` [PATCH v5 16/23] tcg: Use tcg_constant_{i32, i64, vec} with gvec expanders Richard Henderson
2020-12-17 14:52 ` [PATCH v5 17/23] tcg/tci: Add special tci_movi_{i32,i64} opcodes Richard Henderson
2020-12-17 14:52 ` [PATCH v5 18/23] tcg: Remove movi and dupi opcodes Richard Henderson
2020-12-17 14:52 ` [PATCH v5 19/23] tcg: Add tcg_reg_alloc_dup2 Richard Henderson
2020-12-17 14:52 ` [PATCH v5 20/23] tcg/i386: Use tcg_constant_vec with tcg vec expanders Richard Henderson
2020-12-17 14:52 ` [PATCH v5 21/23] tcg: Remove tcg_gen_dup{8,16,32,64}i_vec Richard Henderson
2020-12-17 14:52 ` [PATCH v5 22/23] tcg/ppc: Use tcg_constant_vec with tcg vec expanders Richard Henderson
2020-12-17 14:52 ` Richard Henderson [this message]

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=20201217145215.534637-24-richard.henderson@linaro.org \
    --to=richard.henderson@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).