qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: qemu-arm@nongnu.org, qemu-s390x@nongnu.org,
	qemu-riscv@nongnu.org, qemu-ppc@nongnu.org
Subject: [PATCH 18/42] tcg: Introduce tcg_out_movext2
Date: Fri,  7 Apr 2023 19:42:50 -0700	[thread overview]
Message-ID: <20230408024314.3357414-20-richard.henderson@linaro.org> (raw)
In-Reply-To: <20230408024314.3357414-1-richard.henderson@linaro.org>

This is common code in most qemu_{ld,st} slow paths, moving two
registers when there may be overlap between sources and destinations.
At present, this is only used by 32-bit hosts for 64-bit data,
but will shortly be used for more than that.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 tcg/tcg.c                 | 50 +++++++++++++++++++++++++++++++++++----
 tcg/arm/tcg-target.c.inc  | 34 +++++++-------------------
 tcg/i386/tcg-target.c.inc | 16 ++++---------
 3 files changed, 59 insertions(+), 41 deletions(-)

diff --git a/tcg/tcg.c b/tcg/tcg.c
index d82d99e1b0..1c11f15bce 100644
--- a/tcg/tcg.c
+++ b/tcg/tcg.c
@@ -115,8 +115,7 @@ static void tcg_out_exts_i32_i64(TCGContext *s, TCGReg ret, TCGReg arg);
 static void tcg_out_extu_i32_i64(TCGContext *s, TCGReg ret, TCGReg arg);
 static void tcg_out_extrl_i64_i32(TCGContext *s, TCGReg ret, TCGReg arg);
 static void tcg_out_addi_ptr(TCGContext *s, TCGReg, TCGReg, tcg_target_long);
-static bool tcg_out_xchg(TCGContext *s, TCGType type, TCGReg r1, TCGReg r2)
-    __attribute__((unused));
+static bool tcg_out_xchg(TCGContext *s, TCGType type, TCGReg r1, TCGReg r2);
 static void tcg_out_exit_tb(TCGContext *s, uintptr_t arg);
 static void tcg_out_goto_tb(TCGContext *s, int which);
 static void tcg_out_op(TCGContext *s, TCGOpcode opc,
@@ -365,9 +364,8 @@ void tcg_raise_tb_overflow(TCGContext *s)
  *
  * Move or extend @src into @dst, depending on @src_ext and the types.
  */
-static void __attribute__((unused))
-tcg_out_movext(TCGContext *s, TCGType dst_type, TCGReg dst,
-               TCGType src_type, MemOp src_ext, TCGReg src)
+static void tcg_out_movext(TCGContext *s, TCGType dst_type, TCGReg dst,
+                           TCGType src_type, MemOp src_ext, TCGReg src)
 {
     switch (src_ext) {
     case MO_UB:
@@ -413,6 +411,48 @@ tcg_out_movext(TCGContext *s, TCGType dst_type, TCGReg dst,
     }
 }
 
+/**
+ * tcg_out_movext2 -- move and extend two pair
+ * @s: tcg context
+ * @d1_type: integral type for destination
+ * @d1: destination register
+ * @s1_type: integral type for source
+ * @s1_ext: extension to apply to source
+ * @s1: source register
+ * @d2_type: integral type for destination
+ * @d2: destination register
+ * @s2_type: integral type for source
+ * @s2_ext: extension to apply to source
+ * @s2: source register
+ * @scratch: temporary register, or -1 for none
+ *
+ * As tcg_out_movext, for both s1->d1 and s2->d2, caring for overlap
+ * between the sources and destinations.
+ */
+static void __attribute__((unused))
+tcg_out_movext2(TCGContext *s, TCGType d1_type, TCGReg d1, TCGType s1_type,
+                MemOp s1_ext, TCGReg s1, TCGType d2_type, TCGReg d2,
+                TCGType s2_type, MemOp s2_ext, TCGReg s2, int scratch)
+{
+    if (d1 != s2) {
+        tcg_out_movext(s, d1_type, d1, s1_type, s1_ext, s1);
+        tcg_out_movext(s, d2_type, d2, s2_type, s2_ext, s2);
+        return;
+    }
+    if (d2 == s1) {
+        if (tcg_out_xchg(s, MAX(s1_type, s2_type), s1, s2)) {
+            /* The data is now in the correct registers, now extend. */
+            s1 = d1, s2 = d2;
+        } else {
+            tcg_debug_assert(scratch >= 0);
+            tcg_out_mov(s, s1_type, scratch, s1);
+            s1 = scratch;
+        }
+    }
+    tcg_out_movext(s, d2_type, d2, s2_type, s2_ext, s2);
+    tcg_out_movext(s, d1_type, d1, s1_type, s1_ext, s1);
+}
+
 #define C_PFX1(P, A)                    P##A
 #define C_PFX2(P, A, B)                 P##A##_##B
 #define C_PFX3(P, A, B, C)              P##A##_##B##_##C
diff --git a/tcg/arm/tcg-target.c.inc b/tcg/arm/tcg-target.c.inc
index 4a5d57a41c..bad1e6d399 100644
--- a/tcg/arm/tcg-target.c.inc
+++ b/tcg/arm/tcg-target.c.inc
@@ -1545,7 +1545,7 @@ static void add_qemu_ldst_label(TCGContext *s, bool is_ld, MemOpIdx oi,
 
 static bool tcg_out_qemu_ld_slow_path(TCGContext *s, TCGLabelQemuLdst *lb)
 {
-    TCGReg argreg, datalo, datahi;
+    TCGReg argreg;
     MemOpIdx oi = lb->oi;
     MemOp opc = get_memop(oi);
 
@@ -1565,20 +1565,11 @@ static bool tcg_out_qemu_ld_slow_path(TCGContext *s, TCGLabelQemuLdst *lb)
     /* Use the canonical unsigned helpers and minimize icache usage. */
     tcg_out_call_int(s, qemu_ld_helpers[opc & MO_SIZE]);
 
-    datalo = lb->datalo_reg;
-    datahi = lb->datahi_reg;
     if ((opc & MO_SIZE) == MO_64) {
-        if (datalo != TCG_REG_R1) {
-            tcg_out_mov_reg(s, COND_AL, datalo, TCG_REG_R0);
-            tcg_out_mov_reg(s, COND_AL, datahi, TCG_REG_R1);
-        } else if (datahi != TCG_REG_R0) {
-            tcg_out_mov_reg(s, COND_AL, datahi, TCG_REG_R1);
-            tcg_out_mov_reg(s, COND_AL, datalo, TCG_REG_R0);
-        } else {
-            tcg_out_mov_reg(s, COND_AL, TCG_REG_TMP, TCG_REG_R0);
-            tcg_out_mov_reg(s, COND_AL, datahi, TCG_REG_R1);
-            tcg_out_mov_reg(s, COND_AL, datalo, TCG_REG_TMP);
-        }
+        tcg_out_movext2(s, TCG_TYPE_I32, lb->datalo_reg,
+                        TCG_TYPE_I32, MO_UL, TCG_REG_R0,
+                        TCG_TYPE_I32, lb->datahi_reg,
+                        TCG_TYPE_I32, MO_UL, TCG_REG_R1, TCG_REG_TMP);
     } else {
         tcg_out_movext(s, TCG_TYPE_I32, lb->datalo_reg,
                        TCG_TYPE_I32, opc & MO_SSIZE, TCG_REG_R0);
@@ -1663,17 +1654,10 @@ static bool tcg_out_fail_alignment(TCGContext *s, TCGLabelQemuLdst *l)
 
     if (TARGET_LONG_BITS == 64) {
         /* 64-bit target address is aligned into R2:R3. */
-        if (l->addrhi_reg != TCG_REG_R2) {
-            tcg_out_mov(s, TCG_TYPE_I32, TCG_REG_R2, l->addrlo_reg);
-            tcg_out_mov(s, TCG_TYPE_I32, TCG_REG_R3, l->addrhi_reg);
-        } else if (l->addrlo_reg != TCG_REG_R3) {
-            tcg_out_mov(s, TCG_TYPE_I32, TCG_REG_R3, l->addrhi_reg);
-            tcg_out_mov(s, TCG_TYPE_I32, TCG_REG_R2, l->addrlo_reg);
-        } else {
-            tcg_out_mov(s, TCG_TYPE_I32, TCG_REG_R1, TCG_REG_R2);
-            tcg_out_mov(s, TCG_TYPE_I32, TCG_REG_R2, TCG_REG_R3);
-            tcg_out_mov(s, TCG_TYPE_I32, TCG_REG_R3, TCG_REG_R1);
-        }
+        tcg_out_movext2(s, TCG_TYPE_I32, TCG_REG_R2,
+                        TCG_TYPE_I32, MO_UL, l->addrlo_reg,
+                        TCG_TYPE_I32, TCG_REG_R3,
+                        TCG_TYPE_I32, MO_UL, l->addrhi_reg, TCG_REG_TMP);
     } else {
         tcg_out_mov(s, TCG_TYPE_I32, TCG_REG_R1, l->addrlo_reg);
     }
diff --git a/tcg/i386/tcg-target.c.inc b/tcg/i386/tcg-target.c.inc
index 7d6bf30747..54465c7f46 100644
--- a/tcg/i386/tcg-target.c.inc
+++ b/tcg/i386/tcg-target.c.inc
@@ -1916,7 +1916,6 @@ static bool tcg_out_qemu_ld_slow_path(TCGContext *s, TCGLabelQemuLdst *l)
 {
     MemOpIdx oi = l->oi;
     MemOp opc = get_memop(oi);
-    TCGReg data_reg;
     tcg_insn_unit **label_ptr = &l->label_ptr[0];
 
     /* resolve label address */
@@ -1953,18 +1952,13 @@ static bool tcg_out_qemu_ld_slow_path(TCGContext *s, TCGLabelQemuLdst *l)
 
     tcg_out_branch(s, 1, qemu_ld_helpers[opc & (MO_BSWAP | MO_SIZE)]);
 
-    data_reg = l->datalo_reg;
     if (TCG_TARGET_REG_BITS == 32 && (opc & MO_SIZE) == MO_64) {
-        if (data_reg == TCG_REG_EDX) {
-            /* xchg %edx, %eax */
-            tcg_out_opc(s, OPC_XCHG_ax_r32 + TCG_REG_EDX, 0, 0, 0);
-            tcg_out_mov(s, TCG_TYPE_I32, l->datahi_reg, TCG_REG_EAX);
-        } else {
-            tcg_out_mov(s, TCG_TYPE_I32, data_reg, TCG_REG_EAX);
-            tcg_out_mov(s, TCG_TYPE_I32, l->datahi_reg, TCG_REG_EDX);
-        }
+        tcg_out_movext2(s, TCG_TYPE_I32, l->datalo_reg,
+                        TCG_TYPE_I32, MO_UL, TCG_REG_EAX,
+                        TCG_TYPE_I32, l->datahi_reg,
+                        TCG_TYPE_I32, MO_UL, TCG_REG_EDX, -1);
     } else {
-        tcg_out_movext(s, l->type, data_reg,
+        tcg_out_movext(s, l->type, l->datalo_reg,
                        TCG_TYPE_REG, opc & MO_SSIZE, TCG_REG_EAX);
     }
 
-- 
2.34.1



  parent reply	other threads:[~2023-04-08  2:50 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-04-08  2:42 [PATCH for-8.1 00/42] tcg: Simplify calls to load/store helpers Richard Henderson
2023-04-08  2:42 ` [PATCH for-8.0] tcg/i386: Adjust assert in tcg_out_addi_ptr Richard Henderson
2023-04-08  2:42 ` [PATCH 01/42] tcg: Replace if + tcg_abort with tcg_debug_assert Richard Henderson
2023-04-11 10:52   ` Philippe Mathieu-Daudé
2023-04-08  2:42 ` [PATCH 02/42] tcg: Replace tcg_abort with g_assert_not_reached Richard Henderson
2023-04-11 10:53   ` Philippe Mathieu-Daudé
2023-04-08  2:42 ` [PATCH 03/42] tcg: Split out tcg_out_ext8s Richard Henderson
2023-04-21 21:45   ` Philippe Mathieu-Daudé
2023-04-08  2:42 ` [PATCH 04/42] tcg: Split out tcg_out_ext8u Richard Henderson
2023-04-21 21:50   ` Philippe Mathieu-Daudé
2023-04-08  2:42 ` [PATCH 05/42] tcg: Split out tcg_out_ext16s Richard Henderson
2023-04-21 21:53   ` Philippe Mathieu-Daudé
2023-04-08  2:42 ` [PATCH 06/42] tcg: Split out tcg_out_ext16u Richard Henderson
2023-04-21 22:03   ` Philippe Mathieu-Daudé
2023-04-08  2:42 ` [PATCH 07/42] tcg: Split out tcg_out_ext32s Richard Henderson
2023-04-08  2:42 ` [PATCH 08/42] tcg: Split out tcg_out_ext32u Richard Henderson
2023-04-08  2:42 ` [PATCH 09/42] tcg: Split out tcg_out_exts_i32_i64 Richard Henderson
2023-04-08  2:42 ` [PATCH 10/42] tcg/loongarch64: Conditionalize tcg_out_exts_i32_i64 Richard Henderson
2023-04-08  2:42 ` [PATCH 11/42] tcg/mips: " Richard Henderson
2023-04-08  2:42 ` [PATCH 12/42] tcg/riscv: " Richard Henderson
2023-04-08  2:42 ` [PATCH 13/42] tcg: Split out tcg_out_extu_i32_i64 Richard Henderson
2023-04-08  2:42 ` [PATCH 14/42] tcg/i386: Conditionalize tcg_out_extu_i32_i64 Richard Henderson
2023-04-08  2:42 ` [PATCH 15/42] tcg: Split out tcg_out_extrl_i64_i32 Richard Henderson
2023-04-08  2:42 ` [PATCH 16/42] tcg: Introduce tcg_out_movext Richard Henderson
2023-04-08  2:42 ` [PATCH 17/42] tcg: Introduce tcg_out_xchg Richard Henderson
2023-04-08  2:42 ` Richard Henderson [this message]
2023-04-08  2:42 ` [PATCH 19/42] tcg: Clear TCGLabelQemuLdst on allocation Richard Henderson
2023-04-11 12:10   ` Philippe Mathieu-Daudé
2023-04-08  2:42 ` [PATCH 20/42] tcg/i386: Use TCGType not bool is_64 in tcg_out_qemu_{ld, st} Richard Henderson
2023-04-11 12:12   ` Philippe Mathieu-Daudé
2023-04-12 11:51     ` Richard Henderson
2023-04-08  2:42 ` [PATCH 21/42] tcg/aarch64: Rename ext to d_type in tcg_out_qemu_ld Richard Henderson
2023-04-11 12:14   ` Philippe Mathieu-Daudé
2023-04-08  2:42 ` [PATCH 22/42] tcg/aarch64: Pass TGType to tcg_out_qemu_st Richard Henderson
2023-04-08  2:42 ` [PATCH 23/42] tcg/arm: Use TCGType not bool is_64 in tcg_out_qemu_{ld, st} Richard Henderson
2023-04-08  2:42 ` [PATCH 24/42] tcg/i386: " Richard Henderson
2023-04-11 12:17   ` Philippe Mathieu-Daudé
2023-04-08  2:42 ` [PATCH 25/42] tcg/ppc: " Richard Henderson
2023-04-11 12:18   ` Philippe Mathieu-Daudé
2023-04-08  2:42 ` [PATCH 26/42] tcg/s390x: Pass TCGType to tcg_out_qemu_{ld,st} Richard Henderson
2023-04-11 12:20   ` Philippe Mathieu-Daudé
2023-04-08  2:42 ` [PATCH 27/42] tcg/riscv: Require TCG_TARGET_REG_BITS == 64 Richard Henderson
2023-04-08  2:43 ` [PATCH 28/42] tcg/riscv: Expand arguments to tcg_out_qemu_{ld,st} Richard Henderson
2023-04-08  2:43 ` [PATCH 29/42] tcg: Move TCGLabelQemuLdst to tcg.c Richard Henderson
2023-04-08  2:43 ` [PATCH 30/42] tcg: Introduce tcg_out_ld_helper_args Richard Henderson
2023-04-08  2:43 ` [PATCH 31/42] tcg: Introduce tcg_out_st_helper_args Richard Henderson
2023-04-08  2:43 ` [PATCH 32/42] tcg/loongarch64: Simplify constraints on qemu_ld/st Richard Henderson
2023-04-08  2:43 ` [PATCH 33/42] tcg/mips: Reorg tcg_out_tlb_load Richard Henderson
2023-04-08  2:43 ` [PATCH 34/42] tcg/mips: Simplify constraints on qemu_ld/st Richard Henderson
2023-04-08  2:43 ` [PATCH 35/42] tcg/ppc: Reorg tcg_out_tlb_read Richard Henderson
2023-04-08  2:43 ` [PATCH 36/42] tcg/ppc: Adjust constraints on qemu_ld/st Richard Henderson
2023-04-08  2:43 ` [PATCH 37/42] tcg/ppc: Remove unused constraints A, B, C, D Richard Henderson
2023-04-08  2:43 ` [PATCH 38/42] tcg/riscv: Simplify constraints on qemu_ld/st Richard Henderson
2023-04-08  2:43 ` [PATCH 39/42] tcg/s390x: Use ALGFR in constructing host address for qemu_ld/st Richard Henderson
2023-04-08  2:43 ` [PATCH 40/42] tcg/s390x: Simplify constraints on qemu_ld/st Richard Henderson
2023-04-08  2:43 ` [PATCH 41/42] tcg/sparc64: Drop is_64 test from tcg_out_qemu_ld data return Richard Henderson
2023-04-08  2:43 ` [PATCH 42/42] tcg/sparc64: Pass TCGType to tcg_out_qemu_{ld,st} 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=20230408024314.3357414-20-richard.henderson@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-ppc@nongnu.org \
    --cc=qemu-riscv@nongnu.org \
    --cc=qemu-s390x@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).