qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Subject: [PATCH 29/37] target/i386: reimplement 0x0f 0xc2, 0xc4-0xc6, add AVX
Date: Mon, 12 Sep 2022 01:04:09 +0200	[thread overview]
Message-ID: <20220911230418.340941-30-pbonzini@redhat.com> (raw)
In-Reply-To: <20220911230418.340941-1-pbonzini@redhat.com>

Nothing special going on here, for once.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 target/i386/tcg/decode-new.c.inc |  5 +++
 target/i386/tcg/emit.c.inc       | 76 ++++++++++++++++++++++++++++++++
 target/i386/tcg/translate.c      |  1 +
 3 files changed, 82 insertions(+)

diff --git a/target/i386/tcg/decode-new.c.inc b/target/i386/tcg/decode-new.c.inc
index 7feb0eca4e..c51b59f721 100644
--- a/target/i386/tcg/decode-new.c.inc
+++ b/target/i386/tcg/decode-new.c.inc
@@ -579,6 +579,11 @@ static const X86OpEntry opcodes_0F[256] = {
     [0x7e] = X86_OP_GROUP0(0F7E),
     [0x7f] = X86_OP_GROUP3(0F6F,       W,x, None,None, V,x, vex5 mmx p_00_66_f3),
 
+    [0xc2] = X86_OP_ENTRY4(VCMP,       V,x, H,x, W,x,       vex2_rep3 p_00_66_f3_f2),
+    [0xc4] = X86_OP_ENTRY4(PINSRW,     V,dq,H,dq,E,w,       vex5 mmx p_00_66),
+    [0xc5] = X86_OP_ENTRY3(PEXTRW,     G,d, U,dq,I,b,       vex5 mmx p_00_66),
+    [0xc6] = X86_OP_ENTRY4(VSHUF,      V,x, H,x, W,x,       vex4 p_00_66),
+
     [0xd0] = X86_OP_ENTRY3(VADDSUB,   V,x, H,x, W,x,        vex2 cpuid(SSE3) p_66_f2),
     [0xd1] = X86_OP_ENTRY3(PSRLW_r,   V,x, H,x, W,x,        vex4 mmx avx2_256 p_00_66),
     [0xd2] = X86_OP_ENTRY3(PSRLD_r,   V,x, H,x, W,x,        vex4 mmx avx2_256 p_00_66),
diff --git a/target/i386/tcg/emit.c.inc b/target/i386/tcg/emit.c.inc
index 7084875af6..d1819f3581 100644
--- a/target/i386/tcg/emit.c.inc
+++ b/target/i386/tcg/emit.c.inc
@@ -1367,6 +1367,11 @@ static void gen_PINSRB(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode
     gen_pinsr(s, env, decode, MO_8);
 }
 
+static void gen_PINSRW(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode)
+{
+    gen_pinsr(s, env, decode, MO_16);
+}
+
 static void gen_PINSR(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode)
 {
     gen_pinsr(s, env, decode, decode->op[2].ot);
@@ -1779,6 +1784,66 @@ static inline void gen_VBROADCASTx128(DisasContext *s, CPUX86State *env, X86Deco
                      decode->op[2].offset, 16, 16);
 }
 
+/*
+ * 00 = v*ps Vps, Hps, Wpd
+ * 66 = v*pd Vpd, Hpd, Wps
+ * f3 = v*ss Vss, Hss, Wps
+ * f2 = v*sd Vsd, Hsd, Wps
+ */
+#define SSE_CMP(x) { \
+    gen_helper_ ## x ## ps ## _xmm, gen_helper_ ## x ## pd ## _xmm, \
+    gen_helper_ ## x ## ss, gen_helper_ ## x ## sd, \
+    gen_helper_ ## x ## ps ## _ymm, gen_helper_ ## x ## pd ## _ymm}
+static const SSEFunc_0_eppp gen_helper_cmp_funcs[32][6] = {
+    SSE_CMP(cmpeq),
+    SSE_CMP(cmplt),
+    SSE_CMP(cmple),
+    SSE_CMP(cmpunord),
+    SSE_CMP(cmpneq),
+    SSE_CMP(cmpnlt),
+    SSE_CMP(cmpnle),
+    SSE_CMP(cmpord),
+
+    SSE_CMP(cmpequ),
+    SSE_CMP(cmpnge),
+    SSE_CMP(cmpngt),
+    SSE_CMP(cmpfalse),
+    SSE_CMP(cmpnequ),
+    SSE_CMP(cmpge),
+    SSE_CMP(cmpgt),
+    SSE_CMP(cmptrue),
+
+    SSE_CMP(cmpeqs),
+    SSE_CMP(cmpltq),
+    SSE_CMP(cmpleq),
+    SSE_CMP(cmpunords),
+    SSE_CMP(cmpneqq),
+    SSE_CMP(cmpnltq),
+    SSE_CMP(cmpnleq),
+    SSE_CMP(cmpords),
+
+    SSE_CMP(cmpequs),
+    SSE_CMP(cmpngeq),
+    SSE_CMP(cmpngtq),
+    SSE_CMP(cmpfalses),
+    SSE_CMP(cmpnequs),
+    SSE_CMP(cmpgeq),
+    SSE_CMP(cmpgtq),
+    SSE_CMP(cmptrues),
+};
+#undef SSE_CMP
+
+static inline void gen_VCMP(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode)
+{
+    int index = decode->immediate & (s->prefix & PREFIX_VEX ? 31 : 7);
+    int b =
+        s->prefix & PREFIX_REPZ  ? 2 /* ss */ :
+        s->prefix & PREFIX_REPNZ ? 3 /* ss */ :
+        !!(s->prefix & PREFIX_DATA) + (s->vex_l << 2);
+
+    gen_helper_cmp_funcs[index][b](cpu_env, s->ptr0, s->ptr1, s->ptr2);
+}
+
 static void gen_VCVTfp2fp(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode)
 {
     gen_unary_fp_sse(s, env, decode,
@@ -1963,4 +2028,15 @@ static inline void gen_VROUNDSS(DisasContext *s, CPUX86State *env, X86DecodedIns
     tcg_temp_free_i32(imm);
 }
 
+static inline void gen_VSHUF(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode)
+{
+    TCGv_i32 imm = tcg_const_i32(decode->immediate);
+    SSEFunc_0_pppi ps, pd, fn;
+    ps = s->vex_l ? gen_helper_shufps_ymm : gen_helper_shufps_xmm;
+    pd = s->vex_l ? gen_helper_shufpd_ymm : gen_helper_shufpd_xmm;
+    fn = s->prefix & PREFIX_DATA ? pd : ps;
+    fn(s->ptr0, s->ptr1, s->ptr2, imm);
+    tcg_temp_free_i32(imm);
+}
+
 #define gen_VXOR  gen_PXOR
diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c
index e42cb275a1..468867afcf 100644
--- a/target/i386/tcg/translate.c
+++ b/target/i386/tcg/translate.c
@@ -4669,6 +4669,7 @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu)
         if (use_new &&
             (b == 0x138 || b == 0x13a ||
              (b >= 0x150 && b <= 0x17f) ||
+             b == 0x1c2 || (b >= 0x1c4 && b <= 0x1c6) ||
              (b >= 0x1d0 && b <= 0x1ff))) {
             return disas_insn_new(s, cpu, b);
         }
-- 
2.37.2




  parent reply	other threads:[~2022-09-11 23:45 UTC|newest]

Thread overview: 86+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-11 23:03 [RFC PATCH 00/37] target/i386: new decoder + AVX implementation Paolo Bonzini
2022-09-11 23:03 ` [PATCH 01/37] target/i386: Define XMMReg and access macros, align ZMM registers Paolo Bonzini
2022-09-11 23:03 ` [PATCH 02/37] target/i386: make ldo/sto operations consistent with ldq Paolo Bonzini
2022-09-12  8:33   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 03/37] target/i386: REPZ and REPNZ are mutually exclusive Paolo Bonzini
2022-09-12  8:37   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 04/37] target/i386: introduce insn_get_addr Paolo Bonzini
2022-09-12  8:39   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 05/37] target/i386: add core of new i386 decoder Paolo Bonzini
2022-09-12  9:27   ` Richard Henderson
2022-09-12 10:54   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 06/37] target/i386: add ALU load/writeback core Paolo Bonzini
2022-09-12 10:02   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 07/37] target/i386: add CPUID[EAX=7, ECX=0].ECX to DisasContext Paolo Bonzini
2022-09-12 10:02   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 08/37] target/i386: add CPUID feature checks to new decoder Paolo Bonzini
2022-09-12 10:05   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 09/37] target/i386: add AVX_EN hflag Paolo Bonzini
2022-09-12 10:06   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 10/37] target/i386: validate VEX prefixes via the instructions' exception classes Paolo Bonzini
2022-09-12 10:39   ` Richard Henderson
2022-09-12 10:42   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 11/37] target/i386: validate SSE prefixes directly in the decoding table Paolo Bonzini
2022-09-12 10:51   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 12/37] target/i386: add scalar 0F 38 and 0F 3A instruction to new decoder Paolo Bonzini
2022-09-12 11:04   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 13/37] target/i386: remove scalar VEX instructions from old decoder Paolo Bonzini
2022-09-12 11:06   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 14/37] target/i386: Prepare ops_sse_header.h for 256 bit AVX Paolo Bonzini
2022-09-12 11:09   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 15/37] target/i386: extend helpers to support VEX.V 3- and 4- operand encodings Paolo Bonzini
2022-09-12 11:11   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 16/37] target/i386: support operand merging in binary scalar helpers Paolo Bonzini
2022-09-12 11:11   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 17/37] target/i386: provide 3-operand versions of unary " Paolo Bonzini
2022-09-12 11:14   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 18/37] target/i386: implement additional AVX comparison operators Paolo Bonzini
2022-09-12 11:19   ` Richard Henderson
2022-09-11 23:03 ` [PATCH 19/37] target/i386: Introduce 256-bit vector helpers Paolo Bonzini
2022-09-12 11:19   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 20/37] target/i386: reimplement 0x0f 0x60-0x6f, add AVX Paolo Bonzini
2022-09-12 11:41   ` Richard Henderson
2022-09-13 10:56     ` Paolo Bonzini
2022-09-13 11:35       ` Richard Henderson
2022-09-12 13:01   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 21/37] target/i386: reimplement 0x0f 0xd8-0xdf, 0xe8-0xef, 0xf8-0xff, " Paolo Bonzini
2022-09-12 13:19   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 22/37] target/i386: reimplement 0x0f 0x50-0x5f, " Paolo Bonzini
2022-09-12 13:46   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 23/37] target/i386: reimplement 0x0f 0x78-0x7f, " Paolo Bonzini
2022-09-12 13:56   ` Richard Henderson
2022-09-14 16:17     ` Paolo Bonzini
2022-09-11 23:04 ` [PATCH 24/37] target/i386: reimplement 0x0f 0x70-0x77, " Paolo Bonzini
2022-09-12 14:29   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 25/37] target/i386: reimplement 0x0f 0xd0-0xd7, 0xe0-0xe7, 0xf0-0xf7, " Paolo Bonzini
2022-09-12 15:06   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 26/37] target/i386: reimplement 0x0f 0x3a, " Paolo Bonzini
2022-09-12 15:33   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 27/37] target/i386: Use tcg gvec ops for pmovmskb Paolo Bonzini
2022-09-13  8:16   ` Richard Henderson
2022-09-14 22:59     ` Paolo Bonzini
2022-09-15  6:48       ` Richard Henderson
2022-09-11 23:04 ` [PATCH 28/37] target/i386: reimplement 0x0f 0x38, add AVX Paolo Bonzini
2022-09-13  9:31   ` Richard Henderson
2022-09-14 17:04     ` Paolo Bonzini
2022-09-15  6:50       ` Richard Henderson
2022-09-11 23:04 ` Paolo Bonzini [this message]
2022-09-13  9:44   ` [PATCH 29/37] target/i386: reimplement 0x0f 0xc2, 0xc4-0xc6, " Richard Henderson
2022-09-11 23:04 ` [PATCH 30/37] target/i386: reimplement 0x0f 0x10-0x17, " Paolo Bonzini
2022-09-13 10:14   ` Richard Henderson
2022-09-14 22:45     ` Paolo Bonzini
2022-09-15  6:51       ` Richard Henderson
2022-09-13 10:38   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 31/37] target/i386: reimplement 0x0f 0x28-0x2f, " Paolo Bonzini
2022-09-13 10:24   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 32/37] target/i386: implement XSAVE and XRSTOR of AVX registers Paolo Bonzini
2022-09-13 10:27   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 33/37] target/i386: Enable AVX cpuid bits when using TCG Paolo Bonzini
2022-09-13 10:28   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 34/37] target/i386: implement VLDMXCSR/VSTMXCSR Paolo Bonzini
2022-09-13 10:32   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 35/37] tests/tcg: extend SSE tests to AVX Paolo Bonzini
2022-09-13 10:33   ` Richard Henderson
2022-09-11 23:04 ` [PATCH 36/37] target/i386: move 3DNow completely out of gen_sse Paolo Bonzini
2022-09-13 10:34   ` Richard Henderson
2022-09-13 10:39 ` [RFC PATCH 00/37] target/i386: new decoder + AVX implementation 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=20220911230418.340941-30-pbonzini@redhat.com \
    --to=pbonzini@redhat.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).