qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: shorne@gmail.com
Subject: [PATCH 2/4] target/openrisc: Use tcg_constant_tl for dc->R0
Date: Thu,  8 Jul 2021 14:37:52 -0700	[thread overview]
Message-ID: <20210708213754.830485-3-richard.henderson@linaro.org> (raw)
In-Reply-To: <20210708213754.830485-1-richard.henderson@linaro.org>

The temp allocated for tcg_const_tl is auto-freed at branches,
but pure constants are not.  So we can remove the extra hoop
jumping in trans_l_swa.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 target/openrisc/translate.c | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)

diff --git a/target/openrisc/translate.c b/target/openrisc/translate.c
index 1e437d2f9d..96df513fd3 100644
--- a/target/openrisc/translate.c
+++ b/target/openrisc/translate.c
@@ -733,12 +733,6 @@ static bool trans_l_swa(DisasContext *dc, arg_store *a)
     ea = tcg_temp_new();
     tcg_gen_addi_tl(ea, cpu_R(dc, a->a), a->i);
 
-    /* For TB_FLAGS_R0_0, the branch below invalidates the temporary assigned
-       to cpu_regs[0].  Since l.swa is quite often immediately followed by a
-       branch, don't bother reallocating; finish the TB using the "real" R0.
-       This also takes care of RB input across the branch.  */
-    dc->R0 = cpu_regs[0];
-
     lab_fail = gen_new_label();
     lab_done = gen_new_label();
     tcg_gen_brcond_tl(TCG_COND_NE, ea, cpu_lock_addr, lab_fail);
@@ -746,7 +740,7 @@ static bool trans_l_swa(DisasContext *dc, arg_store *a)
 
     val = tcg_temp_new();
     tcg_gen_atomic_cmpxchg_tl(val, cpu_lock_addr, cpu_lock_value,
-                              cpu_regs[a->b], dc->mem_idx, MO_TEUL);
+                              cpu_R(dc, a->b), dc->mem_idx, MO_TEUL);
     tcg_gen_setcond_tl(TCG_COND_EQ, cpu_sr_f, val, cpu_lock_value);
     tcg_temp_free(val);
 
@@ -1602,7 +1596,7 @@ static void openrisc_tr_tb_start(DisasContextBase *db, CPUState *cs)
     /* Allow the TCG optimizer to see that R0 == 0,
        when it's true, which is the common case.  */
     if (dc->tb_flags & TB_FLAGS_R0_0) {
-        dc->R0 = tcg_const_tl(0);
+        dc->R0 = tcg_constant_tl(0);
     } else {
         dc->R0 = cpu_regs[0];
     }
-- 
2.25.1



  parent reply	other threads:[~2021-07-08 21:39 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-08 21:37 [PATCH 0/4] target/openrisc: Use tcg_constant_* Richard Henderson
2021-07-08 21:37 ` [PATCH 1/4] " Richard Henderson
2021-07-09 16:21   ` Philippe Mathieu-Daudé
2021-07-09 21:02   ` Stafford Horne
2021-07-08 21:37 ` Richard Henderson [this message]
2021-07-10 12:11   ` [PATCH 2/4] target/openrisc: Use tcg_constant_tl for dc->R0 Stafford Horne
2021-07-08 21:37 ` [PATCH 3/4] target/openrisc: Cache constant 0 in DisasContext Richard Henderson
2021-07-09 16:22   ` Philippe Mathieu-Daudé
2021-07-09 20:58   ` Stafford Horne
2021-07-08 21:37 ` [PATCH 4/4] target/openrisc: Use dc->zero in gen_add, gen_addc Richard Henderson
2021-07-09 21:03   ` Stafford Horne

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=20210708213754.830485-3-richard.henderson@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=shorne@gmail.com \
    /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).