From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: luis.pires@eldorado.org.br, alex.bennee@linaro.org,
"Philippe Mathieu-Daudé" <f4bug@amsat.org>
Subject: [PATCH v3 40/48] tcg/optimize: Expand fold_addsub2_i32 to 64-bit ops
Date: Thu, 21 Oct 2021 14:05:31 -0700 [thread overview]
Message-ID: <20211021210539.825582-41-richard.henderson@linaro.org> (raw)
In-Reply-To: <20211021210539.825582-1-richard.henderson@linaro.org>
Rename to fold_addsub2.
Use Int128 to implement the wider operation.
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
tcg/optimize.c | 65 ++++++++++++++++++++++++++++++++++----------------
1 file changed, 44 insertions(+), 21 deletions(-)
diff --git a/tcg/optimize.c b/tcg/optimize.c
index efd5f58137..ba068e7d3e 100644
--- a/tcg/optimize.c
+++ b/tcg/optimize.c
@@ -24,6 +24,7 @@
*/
#include "qemu/osdep.h"
+#include "qemu/int128.h"
#include "tcg/tcg-op.h"
#include "tcg-internal.h"
@@ -838,37 +839,59 @@ static bool fold_add(OptContext *ctx, TCGOp *op)
return false;
}
-static bool fold_addsub2_i32(OptContext *ctx, TCGOp *op, bool add)
+static bool fold_addsub2(OptContext *ctx, TCGOp *op, bool add)
{
if (arg_is_const(op->args[2]) && arg_is_const(op->args[3]) &&
arg_is_const(op->args[4]) && arg_is_const(op->args[5])) {
- uint32_t al = arg_info(op->args[2])->val;
- uint32_t ah = arg_info(op->args[3])->val;
- uint32_t bl = arg_info(op->args[4])->val;
- uint32_t bh = arg_info(op->args[5])->val;
- uint64_t a = ((uint64_t)ah << 32) | al;
- uint64_t b = ((uint64_t)bh << 32) | bl;
+ uint64_t al = arg_info(op->args[2])->val;
+ uint64_t ah = arg_info(op->args[3])->val;
+ uint64_t bl = arg_info(op->args[4])->val;
+ uint64_t bh = arg_info(op->args[5])->val;
TCGArg rl, rh;
- TCGOp *op2 = tcg_op_insert_before(ctx->tcg, op, INDEX_op_mov_i32);
+ TCGOp *op2;
- if (add) {
- a += b;
+ if (ctx->type == TCG_TYPE_I32) {
+ uint64_t a = deposit64(al, 32, 32, ah);
+ uint64_t b = deposit64(bl, 32, 32, bh);
+
+ if (add) {
+ a += b;
+ } else {
+ a -= b;
+ }
+
+ al = sextract64(a, 0, 32);
+ ah = sextract64(a, 32, 32);
} else {
- a -= b;
+ Int128 a = int128_make128(al, ah);
+ Int128 b = int128_make128(bl, bh);
+
+ if (add) {
+ a = int128_add(a, b);
+ } else {
+ a = int128_sub(a, b);
+ }
+
+ al = int128_getlo(a);
+ ah = int128_gethi(a);
}
rl = op->args[0];
rh = op->args[1];
- tcg_opt_gen_movi(ctx, op, rl, (int32_t)a);
- tcg_opt_gen_movi(ctx, op2, rh, (int32_t)(a >> 32));
+
+ /* The proper opcode is supplied by tcg_opt_gen_mov. */
+ op2 = tcg_op_insert_before(ctx->tcg, op, 0);
+
+ tcg_opt_gen_movi(ctx, op, rl, al);
+ tcg_opt_gen_movi(ctx, op2, rh, ah);
return true;
}
return false;
}
-static bool fold_add2_i32(OptContext *ctx, TCGOp *op)
+static bool fold_add2(OptContext *ctx, TCGOp *op)
{
- return fold_addsub2_i32(ctx, op, true);
+ return fold_addsub2(ctx, op, true);
}
static bool fold_and(OptContext *ctx, TCGOp *op)
@@ -1705,9 +1728,9 @@ static bool fold_sub(OptContext *ctx, TCGOp *op)
return false;
}
-static bool fold_sub2_i32(OptContext *ctx, TCGOp *op)
+static bool fold_sub2(OptContext *ctx, TCGOp *op)
{
- return fold_addsub2_i32(ctx, op, false);
+ return fold_addsub2(ctx, op, false);
}
static bool fold_tcg_ld(OptContext *ctx, TCGOp *op)
@@ -1853,8 +1876,8 @@ void tcg_optimize(TCGContext *s)
CASE_OP_32_64_VEC(add):
done = fold_add(&ctx, op);
break;
- case INDEX_op_add2_i32:
- done = fold_add2_i32(&ctx, op);
+ CASE_OP_32_64(add2):
+ done = fold_add2(&ctx, op);
break;
CASE_OP_32_64_VEC(and):
done = fold_and(&ctx, op);
@@ -1989,8 +2012,8 @@ void tcg_optimize(TCGContext *s)
CASE_OP_32_64_VEC(sub):
done = fold_sub(&ctx, op);
break;
- case INDEX_op_sub2_i32:
- done = fold_sub2_i32(&ctx, op);
+ CASE_OP_32_64(sub2):
+ done = fold_sub2(&ctx, op);
break;
CASE_OP_32_64_VEC(xor):
done = fold_xor(&ctx, op);
--
2.25.1
next prev parent reply other threads:[~2021-10-21 21:41 UTC|newest]
Thread overview: 108+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-21 21:04 [PATCH v3 00/48] tcg: optimize redundant sign extensions Richard Henderson
2021-10-21 21:04 ` [PATCH v3 01/48] tcg/optimize: Rename "mask" to "z_mask" Richard Henderson
2021-10-21 21:04 ` [PATCH v3 02/48] tcg/optimize: Split out OptContext Richard Henderson
2021-10-22 14:53 ` Philippe Mathieu-Daudé
2021-10-21 21:04 ` [PATCH v3 03/48] tcg/optimize: Remove do_default label Richard Henderson
2021-10-22 15:12 ` Philippe Mathieu-Daudé
2021-10-21 21:04 ` [PATCH v3 04/48] tcg/optimize: Change tcg_opt_gen_{mov, movi} interface Richard Henderson
2021-10-21 21:04 ` [PATCH v3 05/48] tcg/optimize: Move prev_mb into OptContext Richard Henderson
2021-10-22 15:13 ` Philippe Mathieu-Daudé
2021-10-21 21:04 ` [PATCH v3 06/48] tcg/optimize: Split out init_arguments Richard Henderson
2021-10-21 21:04 ` [PATCH v3 07/48] tcg/optimize: Split out copy_propagate Richard Henderson
2021-10-21 21:04 ` [PATCH v3 08/48] tcg/optimize: Split out fold_call Richard Henderson
2021-10-21 21:05 ` [PATCH v3 09/48] tcg/optimize: Drop nb_oargs, nb_iargs locals Richard Henderson
2021-10-21 21:05 ` [PATCH v3 10/48] tcg/optimize: Change fail return for do_constant_folding_cond* Richard Henderson
2021-10-22 13:53 ` Philippe Mathieu-Daudé
2021-10-21 21:05 ` [PATCH v3 11/48] tcg/optimize: Return true from tcg_opt_gen_{mov, movi} Richard Henderson
2021-10-26 15:13 ` [PATCH v3 11/48] tcg/optimize: Return true from tcg_opt_gen_{mov,movi} Alex Bennée
2021-10-21 21:05 ` [PATCH v3 12/48] tcg/optimize: Split out finish_folding Richard Henderson
2021-10-26 15:16 ` Alex Bennée
2021-10-21 21:05 ` [PATCH v3 13/48] tcg/optimize: Use a boolean to avoid a mass of continues Richard Henderson
2021-10-26 15:17 ` Alex Bennée
2021-10-21 21:05 ` [PATCH v3 14/48] tcg/optimize: Split out fold_mb, fold_qemu_{ld,st} Richard Henderson
2021-10-22 13:56 ` Philippe Mathieu-Daudé
2021-10-21 21:05 ` [PATCH v3 15/48] tcg/optimize: Split out fold_const{1,2} Richard Henderson
2021-10-26 15:19 ` Alex Bennée
2021-10-21 21:05 ` [PATCH v3 16/48] tcg/optimize: Split out fold_setcond2 Richard Henderson
2021-10-26 15:31 ` Alex Bennée
2021-10-21 21:05 ` [PATCH v3 17/48] tcg/optimize: Split out fold_brcond2 Richard Henderson
2021-10-22 17:26 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 18/48] tcg/optimize: Split out fold_brcond Richard Henderson
2021-10-22 13:58 ` Philippe Mathieu-Daudé
2021-10-21 21:05 ` [PATCH v3 19/48] tcg/optimize: Split out fold_setcond Richard Henderson
2021-10-22 13:59 ` Philippe Mathieu-Daudé
2021-10-21 21:05 ` [PATCH v3 20/48] tcg/optimize: Split out fold_mulu2_i32 Richard Henderson
2021-10-22 14:00 ` Philippe Mathieu-Daudé
2021-10-22 17:32 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 21/48] tcg/optimize: Split out fold_addsub2_i32 Richard Henderson
2021-10-22 14:01 ` Philippe Mathieu-Daudé
2021-10-22 17:33 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 22/48] tcg/optimize: Split out fold_movcond Richard Henderson
2021-10-22 14:05 ` Philippe Mathieu-Daudé
2021-10-22 17:34 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 23/48] tcg/optimize: Split out fold_extract2 Richard Henderson
2021-10-22 14:07 ` Philippe Mathieu-Daudé
2021-10-22 17:34 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 24/48] tcg/optimize: Split out fold_extract, fold_sextract Richard Henderson
2021-10-22 14:08 ` Philippe Mathieu-Daudé
2021-10-22 17:34 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 25/48] tcg/optimize: Split out fold_deposit Richard Henderson
2021-10-22 14:09 ` Philippe Mathieu-Daudé
2021-10-22 17:35 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 26/48] tcg/optimize: Split out fold_count_zeros Richard Henderson
2021-10-22 14:10 ` Philippe Mathieu-Daudé
2021-10-22 17:35 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 27/48] tcg/optimize: Split out fold_bswap Richard Henderson
2021-10-22 17:36 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 28/48] tcg/optimize: Split out fold_dup, fold_dup2 Richard Henderson
2021-10-22 14:12 ` Philippe Mathieu-Daudé
2021-10-22 17:36 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 29/48] tcg/optimize: Split out fold_mov Richard Henderson
2021-10-22 17:39 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 30/48] tcg/optimize: Split out fold_xx_to_i Richard Henderson
2021-10-22 21:04 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 31/48] tcg/optimize: Split out fold_xx_to_x Richard Henderson
2021-10-22 21:04 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 32/48] tcg/optimize: Split out fold_xi_to_i Richard Henderson
2021-10-22 21:17 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 33/48] tcg/optimize: Add type to OptContext Richard Henderson
2021-10-22 22:11 ` Luis Fernando Fujita Pires
2021-10-23 18:25 ` Richard Henderson
2021-10-21 21:05 ` [PATCH v3 34/48] tcg/optimize: Split out fold_to_not Richard Henderson
2021-10-25 14:17 ` Luis Fernando Fujita Pires
2021-10-25 17:31 ` Richard Henderson
2021-10-25 18:13 ` Luis Fernando Fujita Pires
2021-10-25 18:35 ` Richard Henderson
2021-10-21 21:05 ` [PATCH v3 35/48] tcg/optimize: Split out fold_sub_to_neg Richard Henderson
2021-10-25 14:17 ` Luis Fernando Fujita Pires
2021-10-21 21:05 ` [PATCH v3 36/48] tcg/optimize: Split out fold_xi_to_x Richard Henderson
2021-10-25 14:26 ` Luis Fernando Fujita Pires
2021-10-25 17:46 ` Richard Henderson
2021-10-21 21:05 ` [PATCH v3 37/48] tcg/optimize: Split out fold_ix_to_i Richard Henderson
2021-10-25 14:26 ` Luis Fernando Fujita Pires
2021-10-26 19:11 ` Philippe Mathieu-Daudé
2021-10-21 21:05 ` [PATCH v3 38/48] tcg/optimize: Split out fold_masks Richard Henderson
2021-10-26 15:32 ` Alex Bennée
2021-10-26 18:50 ` Richard Henderson
2021-10-21 21:05 ` [PATCH v3 39/48] tcg/optimize: Expand fold_mulu2_i32 to all 4-arg multiplies Richard Henderson
2021-10-22 14:16 ` Philippe Mathieu-Daudé
2021-10-21 21:05 ` Richard Henderson [this message]
2021-10-26 16:21 ` [PATCH v3 40/48] tcg/optimize: Expand fold_addsub2_i32 to 64-bit ops Alex Bennée
2021-10-21 21:05 ` [PATCH v3 41/48] tcg/optimize: Sink commutative operand swapping into fold functions Richard Henderson
2021-10-26 16:27 ` Alex Bennée
2021-10-26 19:33 ` Richard Henderson
2021-10-27 13:22 ` Alex Bennée
2021-10-21 21:05 ` [PATCH v3 42/48] tcg/optimize: Add more simplifications for orc Richard Henderson
2021-10-26 16:22 ` Alex Bennée
2021-10-21 21:05 ` [PATCH v3 43/48] tcg/optimize: Stop forcing z_mask to "garbage" for 32-bit values Richard Henderson
2021-10-26 16:35 ` Alex Bennée
2021-10-21 21:05 ` [PATCH v3 44/48] tcg/optimize: Optimize sign extensions Richard Henderson
2021-10-26 16:34 ` Alex Bennée
2021-10-21 21:05 ` [PATCH v3 45/48] tcg/optimize: Propagate sign info for logical operations Richard Henderson
2021-10-26 16:35 ` Alex Bennée
2021-10-21 21:05 ` [PATCH v3 46/48] tcg/optimize: Propagate sign info for setcond Richard Henderson
2021-10-26 16:36 ` Alex Bennée
2021-10-21 21:05 ` [PATCH v3 47/48] tcg/optimize: Propagate sign info for bit counting Richard Henderson
2021-10-26 16:36 ` Alex Bennée
2021-10-21 21:05 ` [PATCH v3 48/48] tcg/optimize: Propagate sign info for shifting Richard Henderson
2021-10-26 16:37 ` Alex Bennée
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=20211021210539.825582-41-richard.henderson@linaro.org \
--to=richard.henderson@linaro.org \
--cc=alex.bennee@linaro.org \
--cc=f4bug@amsat.org \
--cc=luis.pires@eldorado.org.br \
--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).