From: "Alex Bennée" <alex.bennee@linaro.org>
To: Richard Henderson <richard.henderson@linaro.org>
Cc: qemu-devel@nongnu.org
Subject: Re: [PATCH v2 24/36] tcg/i386: Use tcg_constant_vec with tcg vec expanders
Date: Wed, 22 Apr 2020 20:43:14 +0100 [thread overview]
Message-ID: <87imhrb92l.fsf@linaro.org> (raw)
In-Reply-To: <20200422011722.13287-25-richard.henderson@linaro.org>
Richard Henderson <richard.henderson@linaro.org> writes:
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
> ---
> tcg/i386/tcg-target.inc.c | 26 +++++++++++++-------------
> 1 file changed, 13 insertions(+), 13 deletions(-)
>
> diff --git a/tcg/i386/tcg-target.inc.c b/tcg/i386/tcg-target.inc.c
> index 9cb627d6eb..deace219d2 100644
> --- a/tcg/i386/tcg-target.inc.c
> +++ b/tcg/i386/tcg-target.inc.c
> @@ -3452,7 +3452,7 @@ static void expand_vec_sari(TCGType type, unsigned vece,
> static void expand_vec_mul(TCGType type, unsigned vece,
> TCGv_vec v0, TCGv_vec v1, TCGv_vec v2)
> {
> - TCGv_vec t1, t2, t3, t4;
> + TCGv_vec t1, t2, t3, t4, zero;
>
> tcg_debug_assert(vece == MO_8);
>
> @@ -3470,11 +3470,11 @@ static void expand_vec_mul(TCGType type, unsigned vece,
> case TCG_TYPE_V64:
> t1 = tcg_temp_new_vec(TCG_TYPE_V128);
> t2 = tcg_temp_new_vec(TCG_TYPE_V128);
> - tcg_gen_dup16i_vec(t2, 0);
> + zero = tcg_constant_vec(TCG_TYPE_V128, MO_8, 0);
> vec_gen_3(INDEX_op_x86_punpckl_vec, TCG_TYPE_V128, MO_8,
> - tcgv_vec_arg(t1), tcgv_vec_arg(v1), tcgv_vec_arg(t2));
> + tcgv_vec_arg(t1), tcgv_vec_arg(v1), tcgv_vec_arg(zero));
> vec_gen_3(INDEX_op_x86_punpckl_vec, TCG_TYPE_V128, MO_8,
> - tcgv_vec_arg(t2), tcgv_vec_arg(t2), tcgv_vec_arg(v2));
> + tcgv_vec_arg(t2), tcgv_vec_arg(zero), tcgv_vec_arg(v2));
> tcg_gen_mul_vec(MO_16, t1, t1, t2);
> tcg_gen_shri_vec(MO_16, t1, t1, 8);
> vec_gen_3(INDEX_op_x86_packus_vec, TCG_TYPE_V128, MO_8,
> @@ -3489,15 +3489,15 @@ static void expand_vec_mul(TCGType type, unsigned vece,
> t2 = tcg_temp_new_vec(type);
> t3 = tcg_temp_new_vec(type);
> t4 = tcg_temp_new_vec(type);
> - tcg_gen_dup16i_vec(t4, 0);
> + zero = tcg_constant_vec(TCG_TYPE_V128, MO_8, 0);
> vec_gen_3(INDEX_op_x86_punpckl_vec, type, MO_8,
> - tcgv_vec_arg(t1), tcgv_vec_arg(v1), tcgv_vec_arg(t4));
> + tcgv_vec_arg(t1), tcgv_vec_arg(v1), tcgv_vec_arg(zero));
> vec_gen_3(INDEX_op_x86_punpckl_vec, type, MO_8,
> - tcgv_vec_arg(t2), tcgv_vec_arg(t4), tcgv_vec_arg(v2));
> + tcgv_vec_arg(t2), tcgv_vec_arg(zero), tcgv_vec_arg(v2));
> vec_gen_3(INDEX_op_x86_punpckh_vec, type, MO_8,
> - tcgv_vec_arg(t3), tcgv_vec_arg(v1), tcgv_vec_arg(t4));
> + tcgv_vec_arg(t3), tcgv_vec_arg(v1), tcgv_vec_arg(zero));
> vec_gen_3(INDEX_op_x86_punpckh_vec, type, MO_8,
> - tcgv_vec_arg(t4), tcgv_vec_arg(t4), tcgv_vec_arg(v2));
> + tcgv_vec_arg(t4), tcgv_vec_arg(zero), tcgv_vec_arg(v2));
> tcg_gen_mul_vec(MO_16, t1, t1, t2);
> tcg_gen_mul_vec(MO_16, t3, t3, t4);
> tcg_gen_shri_vec(MO_16, t1, t1, 8);
> @@ -3525,7 +3525,7 @@ static bool expand_vec_cmp_noinv(TCGType type, unsigned vece, TCGv_vec v0,
> NEED_UMIN = 8,
> NEED_UMAX = 16,
> };
> - TCGv_vec t1, t2;
> + TCGv_vec t1, t2, t3;
> uint8_t fixup;
>
> switch (cond) {
> @@ -3596,9 +3596,9 @@ static bool expand_vec_cmp_noinv(TCGType type, unsigned vece, TCGv_vec v0,
> } else if (fixup & NEED_BIAS) {
> t1 = tcg_temp_new_vec(type);
> t2 = tcg_temp_new_vec(type);
> - tcg_gen_dupi_vec(vece, t2, 1ull << ((8 << vece) - 1));
> - tcg_gen_sub_vec(vece, t1, v1, t2);
> - tcg_gen_sub_vec(vece, t2, v2, t2);
> + t3 = tcg_constant_vec(type, vece, 1ull << ((8 << vece) - 1));
> + tcg_gen_sub_vec(vece, t1, v1, t3);
> + tcg_gen_sub_vec(vece, t2, v2, t3);
> v1 = t1;
> v2 = t2;
> cond = tcg_signed_cond(cond);
--
Alex Bennée
next prev parent reply other threads:[~2020-04-22 19:44 UTC|newest]
Thread overview: 75+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-22 1:16 [PATCH v2 00/36] tcg 5.1 omnibus patch set Richard Henderson
2020-04-22 1:16 ` [PATCH v2 01/36] tcg: Add tcg_gen_gvec_dup_imm Richard Henderson
2020-04-22 1:16 ` [PATCH v2 02/36] target/s390x: Use tcg_gen_gvec_dup_imm Richard Henderson
2020-04-22 1:16 ` [PATCH v2 03/36] target/ppc: " Richard Henderson
2020-04-22 1:16 ` [PATCH v2 04/36] target/arm: " Richard Henderson
2020-04-22 1:16 ` [PATCH v2 05/36] tcg: Use tcg_gen_gvec_dup_imm in logical simplifications Richard Henderson
2020-04-22 1:16 ` [PATCH v2 06/36] tcg: Remove tcg_gen_gvec_dup{8,16,32,64}i Richard Henderson
2020-04-22 1:16 ` [PATCH v2 07/36] tcg: Add tcg_gen_gvec_dup_tl Richard Henderson
2020-04-22 1:16 ` [PATCH v2 08/36] tcg: Improve vector tail clearing Richard Henderson
2020-04-22 1:16 ` [PATCH v2 09/36] tcg: Consolidate 3 bits into enum TCGTempKind Richard Henderson
2020-04-22 11:25 ` Alex Bennée
2020-04-22 19:58 ` Aleksandar Markovic
2020-04-23 9:00 ` Philippe Mathieu-Daudé
2020-04-23 15:40 ` Richard Henderson
2020-04-23 17:24 ` Daniel P. Berrangé
2020-04-23 23:11 ` Richard Henderson
2020-04-24 9:08 ` Daniel P. Berrangé
2020-04-22 1:16 ` [PATCH v2 10/36] tcg: Add temp_readonly Richard Henderson
2020-04-22 11:26 ` Alex Bennée
2020-04-22 1:16 ` [PATCH v2 11/36] tcg: Introduce TYPE_CONST temporaries Richard Henderson
2020-04-22 15:17 ` Alex Bennée
2020-04-22 16:55 ` Richard Henderson
2020-04-22 1:16 ` [PATCH v2 12/36] tcg: Use tcg_constant_i32 with icount expander Richard Henderson
2020-04-22 15:40 ` Alex Bennée
2020-04-22 1:16 ` [PATCH v2 13/36] tcg: Use tcg_constant_{i32, i64} with tcg int expanders Richard Henderson
2020-04-22 16:18 ` [PATCH v2 13/36] tcg: Use tcg_constant_{i32,i64} " Alex Bennée
2020-04-22 17:02 ` Richard Henderson
2020-04-22 17:57 ` Alex Bennée
2020-04-22 20:04 ` Alex Bennée
2020-04-23 23:13 ` Richard Henderson
2020-04-24 13:23 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 14/36] tcg: Use tcg_constant_{i32, vec} with tcg vec expanders Richard Henderson
2020-04-22 17:00 ` [PATCH v2 14/36] tcg: Use tcg_constant_{i32,vec} " Alex Bennée
2020-04-22 1:17 ` [PATCH v2 15/36] tcg: Use tcg_constant_{i32,i64} with tcg plugins Richard Henderson
2020-04-22 17:18 ` [PATCH v2 15/36] tcg: Use tcg_constant_{i32, i64} " Alex Bennée
2020-04-22 1:17 ` [PATCH v2 16/36] tcg: Rename struct tcg_temp_info to TempOptInfo Richard Henderson
2020-04-22 17:19 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 17/36] tcg/optimize: Adjust TempOptInfo allocation Richard Henderson
2020-04-22 17:53 ` Alex Bennée
2020-04-22 18:28 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 18/36] tcg/optimize: Use tcg_constant_internal with constant folding Richard Henderson
2020-04-22 18:28 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 19/36] tcg/tci: Add special tci_movi_{i32,i64} opcodes Richard Henderson
2020-04-22 19:02 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 20/36] tcg: Remove movi and dupi opcodes Richard Henderson
2020-04-22 9:12 ` Aleksandar Markovic
2020-04-22 19:03 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 21/36] tcg: Use tcg_out_dupi_vec from temp_load Richard Henderson
2020-04-22 19:28 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 22/36] tcg: Increase tcg_out_dupi_vec immediate to int64_t Richard Henderson
2020-04-22 19:33 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 23/36] tcg: Add tcg_reg_alloc_dup2 Richard Henderson
2020-04-22 19:40 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 24/36] tcg/i386: Use tcg_constant_vec with tcg vec expanders Richard Henderson
2020-04-22 19:43 ` Alex Bennée [this message]
2020-04-22 1:17 ` [PATCH v2 25/36] tcg: Remove tcg_gen_dup{8,16,32,64}i_vec Richard Henderson
2020-04-23 9:11 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 26/36] tcg: Add load_dest parameter to GVecGen2 Richard Henderson
2020-04-23 9:37 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 27/36] tcg: Fix integral argument type to tcg_gen_rot[rl]i_i{32, 64} Richard Henderson
2020-04-22 10:19 ` Philippe Mathieu-Daudé
2020-04-23 9:38 ` [PATCH v2 27/36] tcg: Fix integral argument type to tcg_gen_rot[rl]i_i{32,64} Alex Bennée
2020-04-22 1:17 ` [PATCH v2 28/36] tcg: Implement gvec support for rotate by immediate Richard Henderson
2020-04-23 13:28 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 29/36] tcg: Implement gvec support for rotate by vector Richard Henderson
2020-04-23 13:41 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 30/36] tcg: Remove expansion to shift by vector from do_shifts Richard Henderson
2020-04-22 1:17 ` [PATCH v2 31/36] tcg: Implement gvec support for rotate by scalar Richard Henderson
2020-04-23 13:46 ` Alex Bennée
2020-04-22 1:17 ` [PATCH v2 32/36] tcg/i386: Implement INDEX_op_rotl[is]_vec Richard Henderson
2020-04-22 1:17 ` [PATCH v2 33/36] tcg/aarch64: Implement INDEX_op_rotli_vec Richard Henderson
2020-04-22 1:17 ` [PATCH v2 34/36] tcg/ppc: Implement INDEX_op_rot[lr]v_vec Richard Henderson
2020-04-22 1:17 ` [PATCH v2 35/36] target/ppc: Use tcg_gen_gvec_rotlv Richard Henderson
2020-04-22 1:17 ` [PATCH v2 36/36] target/s390x: Use tcg_gen_gvec_rotl{i,s,v} Richard Henderson
2020-04-23 13:50 ` [PATCH v2 00/36] tcg 5.1 omnibus patch set 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=87imhrb92l.fsf@linaro.org \
--to=alex.bennee@linaro.org \
--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).