qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Philippe Mathieu-Daudé" <philmd@linaro.org>
To: Richard Henderson <richard.henderson@linaro.org>, qemu-devel@nongnu.org
Cc: pbonzini@redhat.com
Subject: Re: [PATCH v2 04/35] tcg/optimize: Do swap_commutative2 in do_constant_folding_cond2
Date: Mon, 6 Nov 2023 22:27:19 +0100	[thread overview]
Message-ID: <d0abd65b-e63a-4b9b-8ba7-2f642ed01804@linaro.org> (raw)
In-Reply-To: <20231028194522.245170-5-richard.henderson@linaro.org>

On 28/10/23 21:44, Richard Henderson wrote:
> Mirror the new do_constant_folding_cond1 by doing all
> argument and condition adjustment within one helper.
> 
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> ---
>   tcg/optimize.c | 107 ++++++++++++++++++++++++++-----------------------
>   1 file changed, 57 insertions(+), 50 deletions(-)

Easier to review with a preliminary code move patch:

-- >8 --
diff --git a/tcg/optimize.c b/tcg/optimize.c
index 89cc794d24..ae8debf6f2 100644
--- a/tcg/optimize.c
+++ b/tcg/optimize.c
@@ -589,43 +589,6 @@ static int do_constant_folding_cond(TCGType type, 
TCGArg x,
      return -1;
  }

-/*
- * Return -1 if the condition can't be simplified,
- * and the result of the condition (0 or 1) if it can.
- */
-static int do_constant_folding_cond2(TCGArg *p1, TCGArg *p2, TCGCond c)
-{
-    TCGArg al = p1[0], ah = p1[1];
-    TCGArg bl = p2[0], bh = p2[1];
-
-    if (arg_is_const(bl) && arg_is_const(bh)) {
-        tcg_target_ulong blv = arg_info(bl)->val;
-        tcg_target_ulong bhv = arg_info(bh)->val;
-        uint64_t b = deposit64(blv, 32, 32, bhv);
-
-        if (arg_is_const(al) && arg_is_const(ah)) {
-            tcg_target_ulong alv = arg_info(al)->val;
-            tcg_target_ulong ahv = arg_info(ah)->val;
-            uint64_t a = deposit64(alv, 32, 32, ahv);
-            return do_constant_folding_cond_64(a, b, c);
-        }
-        if (b == 0) {
-            switch (c) {
-            case TCG_COND_LTU:
-                return 0;
-            case TCG_COND_GEU:
-                return 1;
-            default:
-                break;
-            }
-        }
-    }
-    if (args_are_copies(al, bl) && args_are_copies(ah, bh)) {
-        return do_constant_folding_cond_eq(c);
-    }
-    return -1;
-}
-
  /**
   * swap_commutative:
   * @dest: TCGArg of the destination argument, or NO_DEST.
@@ -672,6 +635,10 @@ static bool swap_commutative2(TCGArg *p1, TCGArg *p2)
      return false;
  }

+/*
+ * Return -1 if the condition can't be simplified,
+ * and the result of the condition (0 or 1) if it can.
+ */
  static int do_constant_folding_cond1(OptContext *ctx, TCGArg dest,
                                       TCGArg *p1, TCGArg *p2, TCGArg 
*pcond)
  {
@@ -689,6 +656,39 @@ static int do_constant_folding_cond1(OptContext 
*ctx, TCGArg dest,
      return r;
  }

+static int do_constant_folding_cond2(TCGArg *p1, TCGArg *p2, TCGCond c)
+{
+    TCGArg al = p1[0], ah = p1[1];
+    TCGArg bl = p2[0], bh = p2[1];
+
+    if (arg_is_const(bl) && arg_is_const(bh)) {
+        tcg_target_ulong blv = arg_info(bl)->val;
+        tcg_target_ulong bhv = arg_info(bh)->val;
+        uint64_t b = deposit64(blv, 32, 32, bhv);
+
+        if (arg_is_const(al) && arg_is_const(ah)) {
+            tcg_target_ulong alv = arg_info(al)->val;
+            tcg_target_ulong ahv = arg_info(ah)->val;
+            uint64_t a = deposit64(alv, 32, 32, ahv);
+            return do_constant_folding_cond_64(a, b, c);
+        }
+        if (b == 0) {
+            switch (c) {
+            case TCG_COND_LTU:
+                return 0;
+            case TCG_COND_GEU:
+                return 1;
+            default:
+                break;
+            }
+        }
+    }
+    if (args_are_copies(al, bl) && args_are_copies(ah, bh)) {
+        return do_constant_folding_cond_eq(c);
+    }
+    return -1;
+}
+
  static void init_arguments(OptContext *ctx, TCGOp *op, int nb_args)
  {
      for (int i = 0; i < nb_args; i++) {
---

To the best of my knowledge:

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>



  reply	other threads:[~2023-11-06 21:28 UTC|newest]

Thread overview: 78+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-10-28 19:44 [PATCH v2 00/35] tcg: Introduce TCG_COND_TST{EQ,NE} Richard Henderson
2023-10-28 19:44 ` [PATCH v2 01/35] " Richard Henderson
2023-11-06 15:26   ` Philippe Mathieu-Daudé
2023-10-28 19:44 ` [PATCH v2 02/35] tcg/optimize: Split out arg_is_const_val Richard Henderson
2023-11-06 15:28   ` Philippe Mathieu-Daudé
2023-10-28 19:44 ` [PATCH v2 03/35] tcg/optimize: Split out do_constant_folding_cond1 Richard Henderson
2023-11-06 15:33   ` Philippe Mathieu-Daudé
2023-10-28 19:44 ` [PATCH v2 04/35] tcg/optimize: Do swap_commutative2 in do_constant_folding_cond2 Richard Henderson
2023-11-06 21:27   ` Philippe Mathieu-Daudé [this message]
2023-10-28 19:44 ` [PATCH v2 05/35] tcg/optimize: Split out arg_new_constant Richard Henderson
2023-11-06 15:34   ` Philippe Mathieu-Daudé
2023-10-28 19:44 ` [PATCH v2 06/35] tcg/optimize: Handle TCG_COND_TST{EQ,NE} Richard Henderson
2023-11-06 21:20   ` Philippe Mathieu-Daudé
2023-10-28 19:44 ` [PATCH v2 07/35] tcg: Add TCGConst argument to tcg_target_const_match Richard Henderson
2023-11-06 18:47   ` Philippe Mathieu-Daudé
2023-10-28 19:44 ` [PATCH v2 08/35] tcg/aarch64: Support TCG_COND_TST{EQ,NE} Richard Henderson
2023-10-28 19:44 ` [PATCH v2 09/35] tcg/aarch64: Generate TBZ, TBNZ Richard Henderson
2023-10-28 19:44 ` [PATCH v2 10/35] tcg/aarch64: Generate CBNZ for TSTNE of UINT32_MAX Richard Henderson
2023-10-28 19:44 ` [PATCH v2 11/35] tcg/arm: Support TCG_COND_TST{EQ,NE} Richard Henderson
2023-11-08 14:52   ` [PATCH v2 11/35 1/2] tcg/arm: Factor tcg_out_cmp() out Philippe Mathieu-Daudé
2023-11-08 14:52   ` [PATCH v2 11/35 2/2] tcg/arm: Support TCG_COND_TST{EQ,NE} Philippe Mathieu-Daudé
2023-11-08 17:59     ` Philippe Mathieu-Daudé
2023-10-28 19:44 ` [PATCH v2 12/35] tcg/i386: Pass x86 condition codes to tcg_out_cmov Richard Henderson
2023-11-06 20:55   ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 13/35] tcg/i386: Move tcg_cond_to_jcc[] into tcg_out_cmp Richard Henderson
2023-11-06 19:46   ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 14/35] tcg/i386: Support TCG_COND_TST{EQ,NE} Richard Henderson
2023-11-08 18:16   ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 15/35] tcg/i386: Improve TSTNE/TESTEQ vs powers of two Richard Henderson
2023-10-28 19:45 ` [PATCH v2 16/35] tcg/loongarch64: Support TCG_COND_TST{EQ,NE} Richard Henderson
2023-11-17  7:48   ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 17/35] tcg/mips: " Richard Henderson
2023-11-17  7:46   ` Philippe Mathieu-Daudé
2023-11-17 16:36     ` Richard Henderson
2023-12-13 14:06       ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 18/35] tcg/riscv: " Richard Henderson
2023-11-06 20:59   ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 19/35] tcg/sparc64: Implement tcg_out_extrl_i64_i32 Richard Henderson
2023-11-06 15:05   ` Philippe Mathieu-Daudé
2023-11-06 18:07     ` Richard Henderson
2023-10-28 19:45 ` [PATCH v2 20/35] tcg/sparc64: Hoist read of tcg_cond_to_rcond Richard Henderson
2023-11-06 21:02   ` Philippe Mathieu-Daudé
2023-11-08 20:57     ` Richard Henderson
2023-11-09  8:35       ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 21/35] tcg/sparc64: Pass TCGCond to tcg_out_cmp Richard Henderson
2023-11-09 11:29   ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 22/35] tcg/sparc64: Support TCG_COND_TST{EQ,NE} Richard Henderson
2023-11-06 21:07   ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 23/35] tcg/ppc: Sink tcg_to_bc usage into tcg_out_bc Richard Henderson
2023-11-06 18:54   ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 24/35] tcg/ppc: Use cr0 in tcg_to_bc and tcg_to_isel Richard Henderson
2023-10-28 19:45 ` [PATCH v2 25/35] tcg/ppc: Tidy up tcg_target_const_match Richard Henderson
2023-11-06 21:08   ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 26/35] tcg/ppc: Add TCG_CT_CONST_CMP Richard Henderson
2023-10-28 19:45 ` [PATCH v2 27/35] tcg/ppc: Support TCG_COND_TST{EQ,NE} Richard Henderson
2023-11-08 20:40   ` Philippe Mathieu-Daudé
2023-11-08 21:27     ` Richard Henderson
2023-11-09  8:42   ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 28/35] tcg/s390x: Split constraint A into J+U Richard Henderson
2023-10-28 19:45 ` [PATCH v2 29/35] tcg/s390x: Add TCG_CT_CONST_CMP Richard Henderson
2023-10-28 19:45 ` [PATCH v2 30/35] tcg/s390x: Support TCG_COND_TST{EQ,NE} Richard Henderson
2023-10-28 19:45 ` [PATCH v2 31/35] tcg/tci: " Richard Henderson
2023-11-06 18:48   ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 32/35] target/alpha: Use TCG_COND_TST{EQ,NE} for BLB{C,S} Richard Henderson
2023-11-08 20:52   ` [PATCH v2 32/35 1/2] target/alpha: Pass immediate value to gen_bcond_internal() Philippe Mathieu-Daudé
2023-11-08 20:55     ` Philippe Mathieu-Daudé
2023-11-08 20:52   ` [PATCH v2 32/35 2/2] target/alpha: Use TCG_COND_TST{EQ, NE} for BLB{C, S} Philippe Mathieu-Daudé
2023-11-08 20:56   ` [PATCH v2 32/35] target/alpha: Use TCG_COND_TST{EQ,NE} for BLB{C,S} Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 33/35] target/alpha: Use TCG_COND_TST{EQ, NE} for CMOVLB{C, S} Richard Henderson
2023-11-08 20:54   ` [PATCH v2 33/35] target/alpha: Use TCG_COND_TST{EQ,NE} for CMOVLB{C,S} Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 34/35] target/alpha: Use TCG_COND_TSTNE for gen_fold_mzero Richard Henderson
2023-11-08 21:03   ` Philippe Mathieu-Daudé
2023-10-28 19:45 ` [PATCH v2 35/35] target/m68k: Use TCG_COND_TST{EQ, NE} in gen_fcc_cond Richard Henderson
2023-11-02 22:17 ` [PATCH v2 00/35] tcg: Introduce TCG_COND_TST{EQ,NE} Richard Henderson
2024-01-06 17:43 ` Paolo Bonzini
2024-01-08 21:45   ` Richard Henderson
2024-01-08 22:55     ` Paolo Bonzini
2024-01-09  8:36       ` Richard Henderson

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=d0abd65b-e63a-4b9b-8ba7-2f642ed01804@linaro.org \
    --to=philmd@linaro.org \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=richard.henderson@linaro.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).