From: Richard Henderson <richard.henderson@linaro.org>
To: Jan Bobek <jan.bobek@gmail.com>, qemu-devel@nongnu.org
Cc: "Alex Bennée" <alex.bennee@linaro.org>
Subject: Re: [Qemu-devel] [RFC PATCH v1 08/22] target/i386: reimplement (V)PAND, (V)ANDPS, (V)ANDPD
Date: Wed, 31 Jul 2019 12:35:50 -0700 [thread overview]
Message-ID: <6d2c01c9-17a4-163c-349c-537bc54289fb@linaro.org> (raw)
In-Reply-To: <20190731175702.4916-9-jan.bobek@gmail.com>
On 7/31/19 10:56 AM, Jan Bobek wrote:
> +#define gen_pand_mm(env, s, modrm) gen_gvec_ld_modrm_mm ((env), (s), (modrm), MO_64, tcg_gen_gvec_and, 0112)
> +#define gen_pand_xmm(env, s, modrm) gen_gvec_ld_modrm_xmm ((env), (s), (modrm), MO_64, tcg_gen_gvec_and, 0112)
> +#define gen_vpand_xmm(env, s, modrm) gen_gvec_ld_modrm_vxmm((env), (s), (modrm), MO_64, tcg_gen_gvec_and, 0123)
> +#define gen_vpand_ymm(env, s, modrm) gen_gvec_ld_modrm_vymm((env), (s), (modrm), MO_64, tcg_gen_gvec_and, 0123)
> +#define gen_andps_xmm gen_pand_xmm
> +#define gen_vandps_xmm gen_vpand_xmm
> +#define gen_vandps_ymm gen_vpand_ymm
> +#define gen_andpd_xmm gen_pand_xmm
> +#define gen_vandpd_xmm gen_vpand_xmm
> +#define gen_vandpd_ymm gen_vpand_ymm
Why all of these extra defines?
> + enum {
> + M_0F = 0x01 << 8,
> + M_0F38 = 0x02 << 8,
> + M_0F3A = 0x04 << 8,
> + P_66 = 0x08 << 8,
> + P_F3 = 0x10 << 8,
> + P_F2 = 0x20 << 8,
> + VEX_128 = 0x40 << 8,
> + VEX_256 = 0x80 << 8,
> + };
> +
> + switch(b | M_0F
> + | (s->prefix & PREFIX_DATA ? P_66 : 0)
> + | (s->prefix & PREFIX_REPZ ? P_F3 : 0)
> + | (s->prefix & PREFIX_REPNZ ? P_F2 : 0)
> + | (s->prefix & PREFIX_VEX ? (s->vex_l ? VEX_256 : VEX_128) : 0)) {
I think you can move this above almost everything in this function, so that all
of the legacy bits follow this switch.
> + case 0xdb | M_0F: gen_pand_mm(env, s, modrm); return;
You'll want to put these on the next lines -- checkpatch.pl again.
> + case 0xdb | M_0F | P_66: gen_pand_xmm(env, s, modrm); return;
> + case 0xdb | M_0F | P_66 | VEX_128: gen_vpand_xmm(env, s, modrm); return;
> + case 0xdb | M_0F | P_66 | VEX_256: gen_vpand_ymm(env, s, modrm); return;
> + case 0x54 | M_0F: gen_andps_xmm(env, s, modrm); return;
> + case 0x54 | M_0F | VEX_128: gen_vandps_xmm(env, s, modrm); return;
> + case 0x54 | M_0F | VEX_256: gen_vandps_ymm(env, s, modrm); return;
> + case 0x54 | M_0F | P_66: gen_andpd_xmm(env, s, modrm); return;
> + case 0x54 | M_0F | P_66 | VEX_128: gen_vandpd_xmm(env, s, modrm); return;
> + case 0x54 | M_0F | P_66 | VEX_256: gen_vandpd_ymm(env, s, modrm); return;
> + default: break;
> + }
Perhaps group cases together?
case 0xdb | M_0F | P_66: /* PAND */
case 0x54 | M_0F: /* ANDPS */
case 0x54 | M_0F | P_66: /* ANDPD */
gen_gvec_ld_modrm_xmm(env, s, modrm, MO_64, tcg_gen_gvec_and, 0112);
return;
How are you planning to handle CPUID checks? I know the currently handling is
quite spotty, but with a reorg we might as well fix that too.
r~
next prev parent reply other threads:[~2019-07-31 19:36 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-31 17:56 [Qemu-devel] [RFC PATCH v1 00/22] reimplement (some) x86 vector instructions using tcg-gvec Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 01/22] target/i386: Push rex_r into DisasContext Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 02/22] target/i386: Push rex_w " Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 03/22] target/i386: Use prefix, aflag and dflag from DisasContext Jan Bobek
2019-07-31 19:41 ` Aleksandar Markovic
2019-07-31 20:04 ` Aleksandar Markovic
2019-08-02 13:20 ` Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 04/22] target/i386: Simplify gen_exception arguments Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 05/22] target/i386: introduce gen_ld_modrm_* helpers Jan Bobek
2019-07-31 19:08 ` Richard Henderson
2019-08-02 13:26 ` Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 06/22] target/i386: introduce gen_gvec_ld_modrm_* helpers Jan Bobek
2019-07-31 22:47 ` Richard Henderson
2019-08-02 13:34 ` Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 07/22] target/i386: add vector register file alignment constraints Jan Bobek
2019-07-31 19:14 ` Richard Henderson
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 08/22] target/i386: reimplement (V)PAND, (V)ANDPS, (V)ANDPD Jan Bobek
2019-07-31 19:35 ` Richard Henderson [this message]
2019-07-31 20:27 ` Aleksandar Markovic
2019-07-31 21:21 ` Richard Henderson
2019-08-02 13:53 ` Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 09/22] target/i386: reimplement (V)POR, (V)ORPS, (V)ORPD Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 10/22] target/i386: reimplement (V)PXOR, (V)XORPS, (V)XORPD Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 11/22] target/i386: reimplement (V)PANDN, (V)ANDNPS, (V)ANDNPD Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 12/22] target/i386: reimplement (V)PADD(B, W, D, Q) Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 13/22] target/i386: reimplement (V)PSUB(B, " Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 14/22] target/i386: reimplement (V)PADDS(B, W) Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 15/22] target/i386: reimplement (V)PADDUS(B, W) Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 16/22] target/i386: reimplement (V)PSUBS(B, W) Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 17/22] target/i386: reimplement (V)PSUBUS(B, W) Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 18/22] target/i386: reimplement (V)PMINSW Jan Bobek
2019-07-31 17:56 ` [Qemu-devel] [RFC PATCH v1 19/22] target/i386: reimplement (V)PMINUB Jan Bobek
2019-07-31 17:57 ` [Qemu-devel] [RFC PATCH v1 20/22] target/i386: reimplement (V)PMAXSW Jan Bobek
2019-07-31 17:57 ` [Qemu-devel] [RFC PATCH v1 21/22] target/i386: reimplement (V)PMAXUB Jan Bobek
2019-07-31 17:57 ` [Qemu-devel] [RFC PATCH v1 22/22] target/i386: reimplement (V)P(EQ, CMP)(B, W, D) Jan Bobek
2019-07-31 19:50 ` Richard Henderson
2019-07-31 20:09 ` Aleksandar Markovic
2019-07-31 21:31 ` Richard Henderson
2019-08-02 14:07 ` Jan Bobek
2019-08-02 14:18 ` Aleksandar Markovic
2019-07-31 18:20 ` [Qemu-devel] [RFC PATCH v1 00/22] reimplement (some) x86 vector instructions using tcg-gvec no-reply
2019-07-31 19:21 ` no-reply
2019-07-31 19:21 ` no-reply
2019-08-01 15:46 ` no-reply
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=6d2c01c9-17a4-163c-349c-537bc54289fb@linaro.org \
--to=richard.henderson@linaro.org \
--cc=alex.bennee@linaro.org \
--cc=jan.bobek@gmail.com \
--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).