qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Daniel Henrique Barboza <danielhb413@gmail.com>
To: qemu-devel@nongnu.org
Cc: qemu-ppc@nongnu.org, stefanha@redhat.com,
	"Lucas Mateus Castro (alqotel)" <lucas.araujo@eldorado.org.br>,
	Richard Henderson <richard.henderson@linaro.org>,
	Daniel Henrique Barboza <danielhb413@gmail.com>
Subject: [PULL 15/62] target/ppc: Use gvec to decode XVCPSGN[SD]P
Date: Fri, 28 Oct 2022 13:39:04 -0300	[thread overview]
Message-ID: <20221028163951.810456-16-danielhb413@gmail.com> (raw)
In-Reply-To: <20221028163951.810456-1-danielhb413@gmail.com>

From: "Lucas Mateus Castro (alqotel)" <lucas.araujo@eldorado.org.br>

Moved XVCPSGNSP and XVCPSGNDP to decodetree and used gvec to translate
them.

xvcpsgnsp:
rept    loop    master             patch
8       12500   0,00561400         0,00537900 (-4.2%)
25      4000    0,00562100         0,00400000 (-28.8%)
100     1000    0,00696900         0,00416300 (-40.3%)
500     200     0,02211900         0,00840700 (-62.0%)
2500    40      0,09328600         0,02728300 (-70.8%)
8000    12      0,27295300         0,06867800 (-74.8%)

xvcpsgndp:
rept    loop    master             patch
8       12500   0,00556300         0,00584200 (+5.0%)
25      4000    0,00482700         0,00431700 (-10.6%)
100     1000    0,00585800         0,00464400 (-20.7%)
500     200     0,01565300         0,00839700 (-46.4%)
2500    40      0,05766500         0,02430600 (-57.8%)
8000    12      0,19875300         0,07947100 (-60.0%)

Like the previous instructions there seemed to be a improvement on
translation time.

Signed-off-by: Lucas Mateus Castro (alqotel) <lucas.araujo@eldorado.org.br>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20221019125040.48028-10-lucas.araujo@eldorado.org.br>
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
---
 target/ppc/insn32.decode            |   2 +
 target/ppc/translate/vsx-impl.c.inc | 109 ++++++++++++++--------------
 target/ppc/translate/vsx-ops.c.inc  |   3 -
 3 files changed, 55 insertions(+), 59 deletions(-)

diff --git a/target/ppc/insn32.decode b/target/ppc/insn32.decode
index 395339ce40..3594c0c960 100644
--- a/target/ppc/insn32.decode
+++ b/target/ppc/insn32.decode
@@ -762,6 +762,8 @@ XVNABSDP        111100 ..... 00000 ..... 111101001 ..   @XX2
 XVNABSSP        111100 ..... 00000 ..... 110101001 ..   @XX2
 XVNEGDP         111100 ..... 00000 ..... 111111001 ..   @XX2
 XVNEGSP         111100 ..... 00000 ..... 110111001 ..   @XX2
+XVCPSGNDP       111100 ..... ..... ..... 11110000 ...   @XX3
+XVCPSGNSP       111100 ..... ..... ..... 11010000 ...   @XX3
 
 ## VSX Scalar Multiply-Add Instructions
 
diff --git a/target/ppc/translate/vsx-impl.c.inc b/target/ppc/translate/vsx-impl.c.inc
index 8717e20d08..1c289238ec 100644
--- a/target/ppc/translate/vsx-impl.c.inc
+++ b/target/ppc/translate/vsx-impl.c.inc
@@ -729,62 +729,6 @@ VSX_SCALAR_MOVE_QP(xsnabsqp, OP_NABS, SGN_MASK_DP)
 VSX_SCALAR_MOVE_QP(xsnegqp, OP_NEG, SGN_MASK_DP)
 VSX_SCALAR_MOVE_QP(xscpsgnqp, OP_CPSGN, SGN_MASK_DP)
 
-#define VSX_VECTOR_MOVE(name, op, sgn_mask)                      \
-static void glue(gen_, name)(DisasContext *ctx)                  \
-    {                                                            \
-        TCGv_i64 xbh, xbl, sgm;                                  \
-        if (unlikely(!ctx->vsx_enabled)) {                       \
-            gen_exception(ctx, POWERPC_EXCP_VSXU);               \
-            return;                                              \
-        }                                                        \
-        xbh = tcg_temp_new_i64();                                \
-        xbl = tcg_temp_new_i64();                                \
-        sgm = tcg_temp_new_i64();                                \
-        get_cpu_vsr(xbh, xB(ctx->opcode), true);                 \
-        get_cpu_vsr(xbl, xB(ctx->opcode), false);                \
-        tcg_gen_movi_i64(sgm, sgn_mask);                         \
-        switch (op) {                                            \
-            case OP_ABS: {                                       \
-                tcg_gen_andc_i64(xbh, xbh, sgm);                 \
-                tcg_gen_andc_i64(xbl, xbl, sgm);                 \
-                break;                                           \
-            }                                                    \
-            case OP_NABS: {                                      \
-                tcg_gen_or_i64(xbh, xbh, sgm);                   \
-                tcg_gen_or_i64(xbl, xbl, sgm);                   \
-                break;                                           \
-            }                                                    \
-            case OP_NEG: {                                       \
-                tcg_gen_xor_i64(xbh, xbh, sgm);                  \
-                tcg_gen_xor_i64(xbl, xbl, sgm);                  \
-                break;                                           \
-            }                                                    \
-            case OP_CPSGN: {                                     \
-                TCGv_i64 xah = tcg_temp_new_i64();               \
-                TCGv_i64 xal = tcg_temp_new_i64();               \
-                get_cpu_vsr(xah, xA(ctx->opcode), true);         \
-                get_cpu_vsr(xal, xA(ctx->opcode), false);        \
-                tcg_gen_and_i64(xah, xah, sgm);                  \
-                tcg_gen_and_i64(xal, xal, sgm);                  \
-                tcg_gen_andc_i64(xbh, xbh, sgm);                 \
-                tcg_gen_andc_i64(xbl, xbl, sgm);                 \
-                tcg_gen_or_i64(xbh, xbh, xah);                   \
-                tcg_gen_or_i64(xbl, xbl, xal);                   \
-                tcg_temp_free_i64(xah);                          \
-                tcg_temp_free_i64(xal);                          \
-                break;                                           \
-            }                                                    \
-        }                                                        \
-        set_cpu_vsr(xT(ctx->opcode), xbh, true);                 \
-        set_cpu_vsr(xT(ctx->opcode), xbl, false);                \
-        tcg_temp_free_i64(xbh);                                  \
-        tcg_temp_free_i64(xbl);                                  \
-        tcg_temp_free_i64(sgm);                                  \
-    }
-
-VSX_VECTOR_MOVE(xvcpsgndp, OP_CPSGN, SGN_MASK_DP)
-VSX_VECTOR_MOVE(xvcpsgnsp, OP_CPSGN, SGN_MASK_SP)
-
 #define TCG_OP_IMM_i64(FUNC, OP, IMM)                           \
     static void FUNC(TCGv_i64 t, TCGv_i64 b)                    \
     {                                                           \
@@ -852,6 +796,59 @@ TRANS(XVABSSP, do_vsx_msb_op, MO_32, do_xvabs_vec, do_xvabssp_i64)
 TRANS(XVNABSSP, do_vsx_msb_op, MO_32, do_xvnabs_vec, do_xvnabssp_i64)
 TRANS(XVNEGSP, do_vsx_msb_op, MO_32, do_xvneg_vec, do_xvnegsp_i64)
 
+static void do_xvcpsgndp_i64(TCGv_i64 t, TCGv_i64 a, TCGv_i64 b)
+{
+    tcg_gen_andi_i64(a, a, SGN_MASK_DP);
+    tcg_gen_andi_i64(b, b, ~SGN_MASK_DP);
+    tcg_gen_or_i64(t, a, b);
+}
+
+static void do_xvcpsgnsp_i64(TCGv_i64 t, TCGv_i64 a, TCGv_i64 b)
+{
+    tcg_gen_andi_i64(a, a, SGN_MASK_SP);
+    tcg_gen_andi_i64(b, b, ~SGN_MASK_SP);
+    tcg_gen_or_i64(t, a, b);
+}
+
+static void do_xvcpsgn_vec(unsigned vece, TCGv_vec t, TCGv_vec a, TCGv_vec b)
+{
+    uint64_t msb = (vece == MO_32) ? SGN_MASK_SP : SGN_MASK_DP;
+    tcg_gen_bitsel_vec(vece, t, tcg_constant_vec_matching(t, vece, msb), a, b);
+}
+
+static bool do_xvcpsgn(DisasContext *ctx, arg_XX3 *a, unsigned vece)
+{
+    static const TCGOpcode vecop_list[] = {
+        0
+    };
+
+    static const GVecGen3 op[] = {
+        {
+            .fni8 = do_xvcpsgnsp_i64,
+            .fniv = do_xvcpsgn_vec,
+            .opt_opc = vecop_list,
+            .vece = MO_32
+        },
+        {
+            .fni8 = do_xvcpsgndp_i64,
+            .fniv = do_xvcpsgn_vec,
+            .opt_opc = vecop_list,
+            .vece = MO_64
+        },
+    };
+
+    REQUIRE_INSNS_FLAGS2(ctx, VSX);
+    REQUIRE_VSX(ctx);
+
+    tcg_gen_gvec_3(vsr_full_offset(a->xt), vsr_full_offset(a->xa),
+                   vsr_full_offset(a->xb), 16, 16, &op[vece - MO_32]);
+
+    return true;
+}
+
+TRANS(XVCPSGNSP, do_xvcpsgn, MO_32)
+TRANS(XVCPSGNDP, do_xvcpsgn, MO_64)
+
 #define VSX_CMP(name, op1, op2, inval, type)                                  \
 static void gen_##name(DisasContext *ctx)                                     \
 {                                                                             \
diff --git a/target/ppc/translate/vsx-ops.c.inc b/target/ppc/translate/vsx-ops.c.inc
index b77324e0a8..f7d7377379 100644
--- a/target/ppc/translate/vsx-ops.c.inc
+++ b/target/ppc/translate/vsx-ops.c.inc
@@ -165,9 +165,6 @@ GEN_XX3FORM(name, opc2, opc3 | 1, fl2)
 GEN_XX2FORM_DCMX(xvtstdcdp, 0x14, 0x1E, PPC2_ISA300),
 GEN_XX2FORM_DCMX(xvtstdcsp, 0x14, 0x1A, PPC2_ISA300),
 
-GEN_XX3FORM(xvcpsgndp, 0x00, 0x1E, PPC2_VSX),
-GEN_XX3FORM(xvcpsgnsp, 0x00, 0x1A, PPC2_VSX),
-
 GEN_XX3FORM(xsadddp, 0x00, 0x04, PPC2_VSX),
 GEN_VSX_XFORM_300(xsaddqp, 0x04, 0x00, 0x0),
 GEN_XX3FORM(xssubdp, 0x00, 0x05, PPC2_VSX),
-- 
2.37.3



  parent reply	other threads:[~2022-10-28 16:42 UTC|newest]

Thread overview: 68+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-28 16:38 [PULL 00/62] ppc queue Daniel Henrique Barboza
2022-10-28 16:38 ` [PULL 01/62] target/ppc: fix msgclr/msgsnd insns flags Daniel Henrique Barboza
2022-10-28 16:38 ` [PULL 02/62] target/ppc: fix msgsync " Daniel Henrique Barboza
2022-10-28 16:38 ` [PULL 03/62] target/ppc: fix REQUIRE_HV macro definition Daniel Henrique Barboza
2022-10-28 16:38 ` [PULL 04/62] target/ppc: move msgclr/msgsnd to decodetree Daniel Henrique Barboza
2022-10-28 16:38 ` [PULL 05/62] target/ppc: move msgclrp/msgsndp " Daniel Henrique Barboza
2022-10-28 16:38 ` [PULL 06/62] target/ppc: move msgsync " Daniel Henrique Barboza
2022-10-28 16:38 ` [PULL 07/62] target/ppc: Moved VMLADDUHM to decodetree and use gvec Daniel Henrique Barboza
2022-10-28 16:38 ` [PULL 08/62] target/ppc: Move VMH[R]ADDSHS instruction to decodetree Daniel Henrique Barboza
2022-10-28 16:38 ` [PULL 09/62] target/ppc: Move V(ADD|SUB)CUW to decodetree and use gvec Daniel Henrique Barboza
2022-10-28 16:38 ` [PULL 10/62] target/ppc: Move VNEG[WD] to decodtree " Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 11/62] target/ppc: Move VPRTYB[WDQ] to decodetree " Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 12/62] target/ppc: Move VAVG[SU][BHW] " Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 13/62] target/ppc: Move VABSDU[BHW] " Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 14/62] target/ppc: Use gvec to decode XV[N]ABS[DS]P/XVNEG[DS]P Daniel Henrique Barboza
2022-10-28 16:39 ` Daniel Henrique Barboza [this message]
2022-10-28 16:39 ` [PULL 16/62] target/ppc: Moved XVTSTDC[DS]P to decodetree Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 17/62] target/ppc: Moved XSTSTDC[QDS]P " Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 18/62] target/ppc: Use gvec to decode XVTSTDC[DS]P Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 19/62] target/ppc: define PPC_INTERRUPT_* values directly Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 20/62] target/ppc: always use ppc_set_irq to set env->pending_interrupts Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 21/62] target/ppc: split interrupt masking and delivery from ppc_hw_interrupt Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 22/62] target/ppc: prepare to split interrupt masking and delivery by excp_model Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 23/62] target/ppc: create an interrupt masking method for POWER9/POWER10 Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 24/62] target/ppc: remove unused interrupts from p9_next_unmasked_interrupt Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 25/62] target/ppc: create an interrupt deliver method for POWER9/POWER10 Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 26/62] target/ppc: remove unused interrupts from p9_deliver_interrupt Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 27/62] target/ppc: remove generic architecture checks " Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 28/62] target/ppc: move power-saving interrupt masking out of cpu_has_work_POWER9 Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 29/62] target/ppc: add power-saving interrupt masking logic to p9_next_unmasked_interrupt Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 30/62] target/ppc: create an interrupt masking method for POWER8 Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 31/62] target/ppc: remove unused interrupts from p8_next_unmasked_interrupt Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 32/62] target/ppc: create an interrupt deliver method for POWER8 Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 33/62] target/ppc: remove unused interrupts from p8_deliver_interrupt Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 34/62] target/ppc: remove generic architecture checks " Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 35/62] target/ppc: move power-saving interrupt masking out of cpu_has_work_POWER8 Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 36/62] target/ppc: add power-saving interrupt masking logic to p8_next_unmasked_interrupt Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 37/62] target/ppc: create an interrupt masking method for POWER7 Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 38/62] target/ppc: remove unused interrupts from p7_next_unmasked_interrupt Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 39/62] target/ppc: create an interrupt deliver method for POWER7 Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 40/62] target/ppc: remove unused interrupts from p7_deliver_interrupt Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 41/62] target/ppc: remove generic architecture checks " Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 42/62] target/ppc: move power-saving interrupt masking out of cpu_has_work_POWER7 Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 43/62] target/ppc: add power-saving interrupt masking logic to p7_next_unmasked_interrupt Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 44/62] target/ppc: remove ppc_store_lpcr from CONFIG_USER_ONLY builds Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 45/62] target/ppc: introduce ppc_maybe_interrupt Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 46/62] target/ppc: unify cpu->has_work based on cs->interrupt_request Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 47/62] target/ppc: move the p*_interrupt_powersave methods to excp_helper.c Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 48/62] ppc440_uc.c: Move DDR2 SDRAM controller model to ppc4xx_sdram.c Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 49/62] ppc4xx_devs.c: Move DDR " Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 50/62] ppc4xx_sdram: Move ppc4xx_sdram_banks() " Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 51/62] ppc4xx_sdram: Use hwaddr for memory bank size Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 52/62] ppc4xx_sdram: Rename local state variable for brevity Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 53/62] ppc4xx_sdram: Generalise bank setup Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 54/62] ppc4xx_sdram: Convert DDR SDRAM controller to new bank handling Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 55/62] ppc4xx_sdram: Add errp parameter to ppc4xx_sdram_banks() Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 56/62] target/ppc: Add new PMC HFLAGS Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 57/62] target/ppc: Increment PMC5 with inline insns Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 58/62] docs/system/ppc/ppce500: Use qemu-system-ppc64 across the board(s) Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 59/62] hw/block/pflash_cfi0{1, 2}: Error out if device length isn't a power of two Daniel Henrique Barboza
2022-11-01 22:23   ` Stefan Hajnoczi
2022-11-01 22:49     ` Philippe Mathieu-Daudé
2022-11-02 19:49       ` Daniel Henrique Barboza
2022-11-08 17:36       ` Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 60/62] hw/sd/sdhci-internal: Unexport ESDHC defines Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 61/62] hw/sd/sdhci: Rename ESDHC_* defines to USDHC_* Daniel Henrique Barboza
2022-10-28 16:39 ` [PULL 62/62] hw/ppc/e500: Implement pflash handling Daniel Henrique Barboza
2022-10-28 20:25 ` [PULL 00/62] ppc queue Daniel Henrique Barboza

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=20221028163951.810456-16-danielhb413@gmail.com \
    --to=danielhb413@gmail.com \
    --cc=lucas.araujo@eldorado.org.br \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-ppc@nongnu.org \
    --cc=richard.henderson@linaro.org \
    --cc=stefanha@redhat.com \
    /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).