From 1aeefa5720a71e622e2f26bf10ec8e7ecbd76f4c Mon Sep 17 00:00:00 2001
From: Hongyu Wang <hongyu.wang@intel.com>
Date: Wed, 30 Jun 2021 14:38:31 +0800
Subject: [PATCH] Clear odata for aes(enc|dec)(wide)?kl intrinsics when ZF is
 set.

For Keylocker aesenc/aesdec intrinsics, current implementation
moves idata to odata unconditionally, which causes safety issue when
the instruction meets runtime error. So we add a branch to clear
odata when ZF is set after instruction exectution.

gcc/ChangeLog:

	* config/i386/i386-expand.c (ix86_expand_builtin):
	Add branch to clear odata when ZF is set for asedecenc_expand
	and wideaesdecenc_expand.

gcc/testsuite/ChangeLog:

	* gcc.target/i386/keylocker-aesdec128kl.c: Update test.
	* gcc.target/i386/keylocker-aesdec256kl.c: Likewise.
	* gcc.target/i386/keylocker-aesdecwide128kl.c: Likewise.
	* gcc.target/i386/keylocker-aesdecwide256kl.c: Likewise.
	* gcc.target/i386/keylocker-aesenc128kl.c: Likewise.
	* gcc.target/i386/keylocker-aesenc256kl.c: Likewise.
	* gcc.target/i386/keylocker-aesencwide128kl.c: Likewise.
	* gcc.target/i386/keylocker-aesencwide256kl.c: Likewise.
---
 gcc/config/i386/i386-expand.c                      | 33 ++++++++++++++++++----
 .../gcc.target/i386/keylocker-aesdec128kl.c        |  2 ++
 .../gcc.target/i386/keylocker-aesdec256kl.c        |  2 ++
 .../gcc.target/i386/keylocker-aesdecwide128kl.c    |  9 ++++++
 .../gcc.target/i386/keylocker-aesdecwide256kl.c    |  9 ++++++
 .../gcc.target/i386/keylocker-aesenc128kl.c        |  2 ++
 .../gcc.target/i386/keylocker-aesenc256kl.c        |  2 ++
 .../gcc.target/i386/keylocker-aesencwide128kl.c    |  9 ++++++
 .../gcc.target/i386/keylocker-aesencwide256kl.c    |  9 ++++++
 9 files changed, 72 insertions(+), 5 deletions(-)

diff --git a/gcc/config/i386/i386-expand.c b/gcc/config/i386/i386-expand.c
index 5c9170e..a63319a 100644
--- a/gcc/config/i386/i386-expand.c
+++ b/gcc/config/i386/i386-expand.c
@@ -11753,10 +11753,24 @@ ix86_expand_builtin (tree exp, rtx target, rtx subtarget,
       if (target == 0)
 	target = gen_reg_rtx (QImode);
 
-      pat = gen_rtx_EQ (QImode, gen_rtx_REG (CCZmode, FLAGS_REG),
-			const0_rtx);
-      emit_insn (gen_rtx_SET (target, pat));
+      /* NB: For aesenc/aesdec keylocker insn, ZF will be set when runtime
+	 error occurs. Then the output should be cleared for safety. */
+      rtx_code_label *ok_label;
+      rtx tmp;
+
+      tmp = gen_rtx_REG (CCZmode, FLAGS_REG);
+      pat = gen_rtx_EQ (QImode, tmp, const0_rtx);
+      ok_label = gen_label_rtx ();
+      emit_cmp_and_jump_insns (tmp, const0_rtx, NE, 0, GET_MODE (tmp),
+			       true, ok_label);
+      /* Usually the runtime error seldom occur, so predict OK path as
+	 hotspot to optimize it as fallthrough block. */
+      predict_jump (REG_BR_PROB_BASE * 90 / 100);
+
+      emit_insn (gen_rtx_SET (op1, const0_rtx));
 
+      emit_label (ok_label);
+      emit_insn (gen_rtx_SET (target, pat));
       emit_insn (gen_rtx_SET (op0, op1));
 
       return target;
@@ -11811,8 +11825,17 @@ ix86_expand_builtin (tree exp, rtx target, rtx subtarget,
       if (target == 0)
 	target = gen_reg_rtx (QImode);
 
-      pat = gen_rtx_EQ (QImode, gen_rtx_REG (CCZmode, FLAGS_REG),
-			const0_rtx);
+      tmp = gen_rtx_REG (CCZmode, FLAGS_REG);
+      pat = gen_rtx_EQ (QImode, tmp, const0_rtx);
+      ok_label = gen_label_rtx ();
+      emit_cmp_and_jump_insns (tmp, const0_rtx, NE, 0, GET_MODE (tmp),
+			       true, ok_label);
+      predict_jump (REG_BR_PROB_BASE * 90 / 100);
+
+      for (i = 0; i < 8; i++)
+	emit_insn (gen_rtx_SET (xmm_regs[i], const0_rtx));
+
+      emit_label (ok_label);
       emit_insn (gen_rtx_SET (target, pat));
 
       for (i = 0; i < 8; i++)
diff --git a/gcc/testsuite/gcc.target/i386/keylocker-aesdec128kl.c b/gcc/testsuite/gcc.target/i386/keylocker-aesdec128kl.c
index d134612..71111c3 100644
--- a/gcc/testsuite/gcc.target/i386/keylocker-aesdec128kl.c
+++ b/gcc/testsuite/gcc.target/i386/keylocker-aesdec128kl.c
@@ -2,8 +2,10 @@
 /* { dg-options "-mkl -O2" } */
 /* { dg-final { scan-assembler "movdqa\[ \\t\]+\[^\\n\\r\]*, %xmm0" } } */
 /* { dg-final { scan-assembler "aesdec128kl\[ \\t\]+\[^\\n\\r\]*, %xmm0" } } */
+/* { dg-final { scan-assembler "j\[ez\]" } } */
 /* { dg-final { scan-assembler "sete" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm0,\[^\\n\\r\]*" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm0, %xmm0" } } */
 
 #include <immintrin.h>
 
diff --git a/gcc/testsuite/gcc.target/i386/keylocker-aesdec256kl.c b/gcc/testsuite/gcc.target/i386/keylocker-aesdec256kl.c
index 34736d2..30189d6 100644
--- a/gcc/testsuite/gcc.target/i386/keylocker-aesdec256kl.c
+++ b/gcc/testsuite/gcc.target/i386/keylocker-aesdec256kl.c
@@ -2,8 +2,10 @@
 /* { dg-options "-mkl -O2" } */
 /* { dg-final { scan-assembler "movdqa\[ \\t\]+\[^\\n\\r\]*, %xmm0" } } */
 /* { dg-final { scan-assembler "aesdec256kl\[ \\t\]+\[^\\n\\r\]*, %xmm0" } } */
+/* { dg-final { scan-assembler "j\[ez\]" } } */
 /* { dg-final { scan-assembler "sete" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm0,\[^\\n\\r\]*" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm0, %xmm0" } } */
 
 #include <immintrin.h>
 
diff --git a/gcc/testsuite/gcc.target/i386/keylocker-aesdecwide128kl.c b/gcc/testsuite/gcc.target/i386/keylocker-aesdecwide128kl.c
index d23cf4b..93806e5 100644
--- a/gcc/testsuite/gcc.target/i386/keylocker-aesdecwide128kl.c
+++ b/gcc/testsuite/gcc.target/i386/keylocker-aesdecwide128kl.c
@@ -9,6 +9,7 @@
 /* { dg-final { scan-assembler "movdqu\[ \\t\]+\[^\\n\\r\]*96\[^\\n\\r\]*, %xmm6" } } */
 /* { dg-final { scan-assembler "movdqu\[ \\t\]+\[^\\n\\r\]*112\[^\\n\\r\]*, %xmm7" } } */
 /* { dg-final { scan-assembler "aesdecwide128kl\[ \\t\]+\[^\\n\\r\]*" } } */
+/* { dg-final { scan-assembler "j\[ez\]" } } */
 /* { dg-final { scan-assembler "sete" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm0,\[^\\n\\r\]*" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm1,\[^\\n\\r\]*16\[^\\n\\r\]*" } } */
@@ -18,6 +19,14 @@
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm5,\[^\\n\\r\]*80\[^\\n\\r\]*" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm6,\[^\\n\\r\]*96\[^\\n\\r\]*" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm7,\[^\\n\\r\]*112\[^\\n\\r\]*" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm0, %xmm0" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm1, %xmm1" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm2, %xmm2" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm3, %xmm3" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm4, %xmm4" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm5, %xmm5" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm6, %xmm6" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm7, %xmm7" } } */
 
 #include <immintrin.h>
 
diff --git a/gcc/testsuite/gcc.target/i386/keylocker-aesdecwide256kl.c b/gcc/testsuite/gcc.target/i386/keylocker-aesdecwide256kl.c
index 44c3252..f9ccc82 100644
--- a/gcc/testsuite/gcc.target/i386/keylocker-aesdecwide256kl.c
+++ b/gcc/testsuite/gcc.target/i386/keylocker-aesdecwide256kl.c
@@ -9,6 +9,7 @@
 /* { dg-final { scan-assembler "movdqu\[ \\t\]+\[^\\n\\r\]*96\[^\\n\\r\]*, %xmm6" } } */
 /* { dg-final { scan-assembler "movdqu\[ \\t\]+\[^\\n\\r\]*112\[^\\n\\r\]*, %xmm7" } } */
 /* { dg-final { scan-assembler "aesdecwide256kl\[ \\t\]+\[^\\n\\r\]*" } } */
+/* { dg-final { scan-assembler "j\[ez\]" } } */
 /* { dg-final { scan-assembler "sete" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm0,\[^\\n\\r\]*" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm1,\[^\\n\\r\]*16\[^\\n\\r\]*" } } */
@@ -18,6 +19,14 @@
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm5,\[^\\n\\r\]*80\[^\\n\\r\]*" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm6,\[^\\n\\r\]*96\[^\\n\\r\]*" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm7,\[^\\n\\r\]*112\[^\\n\\r\]*" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm0, %xmm0" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm1, %xmm1" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm2, %xmm2" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm3, %xmm3" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm4, %xmm4" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm5, %xmm5" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm6, %xmm6" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm7, %xmm7" } } */
 
 #include <immintrin.h>
 
diff --git a/gcc/testsuite/gcc.target/i386/keylocker-aesenc128kl.c b/gcc/testsuite/gcc.target/i386/keylocker-aesenc128kl.c
index 9ff4836..61a9cc2 100644
--- a/gcc/testsuite/gcc.target/i386/keylocker-aesenc128kl.c
+++ b/gcc/testsuite/gcc.target/i386/keylocker-aesenc128kl.c
@@ -2,8 +2,10 @@
 /* { dg-options "-mkl -O2" } */
 /* { dg-final { scan-assembler "movdqa\[ \\t\]+\[^\\n\\r\]*, %xmm0" } } */
 /* { dg-final { scan-assembler "aesenc128kl\[ \\t\]+\[^\\n\\r\]*, %xmm0" } } */
+/* { dg-final { scan-assembler "j\[ez\]" } } */
 /* { dg-final { scan-assembler "sete" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm0,\[^\\n\\r\]*" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm0, %xmm0" } } */
 
 #include <immintrin.h>
 
diff --git a/gcc/testsuite/gcc.target/i386/keylocker-aesenc256kl.c b/gcc/testsuite/gcc.target/i386/keylocker-aesenc256kl.c
index 1c5e076..f8e6bb7 100644
--- a/gcc/testsuite/gcc.target/i386/keylocker-aesenc256kl.c
+++ b/gcc/testsuite/gcc.target/i386/keylocker-aesenc256kl.c
@@ -2,8 +2,10 @@
 /* { dg-options "-mkl -O2" } */
 /* { dg-final { scan-assembler "movdqa\[ \\t\]+\[^\\n\\r\]*, %xmm0" } } */
 /* { dg-final { scan-assembler "aesenc256kl\[ \\t\]+\[^\\n\\r\]*, %xmm0" } } */
+/* { dg-final { scan-assembler "j\[ez\]" } } */
 /* { dg-final { scan-assembler "sete" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm0,\[^\\n\\r\]*" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm0, %xmm0" } } */
 
 #include <immintrin.h>
 
diff --git a/gcc/testsuite/gcc.target/i386/keylocker-aesencwide128kl.c b/gcc/testsuite/gcc.target/i386/keylocker-aesencwide128kl.c
index 9fb9c49..c0fcd28 100644
--- a/gcc/testsuite/gcc.target/i386/keylocker-aesencwide128kl.c
+++ b/gcc/testsuite/gcc.target/i386/keylocker-aesencwide128kl.c
@@ -9,6 +9,7 @@
 /* { dg-final { scan-assembler "movdqu\[ \\t\]+\[^\\n\\r\]*96\[^\\n\\r\]*, %xmm6" } } */
 /* { dg-final { scan-assembler "movdqu\[ \\t\]+\[^\\n\\r\]*112\[^\\n\\r\]*, %xmm7" } } */
 /* { dg-final { scan-assembler "aesencwide128kl\[ \\t\]+\[^\\n\\r\]*" } } */
+/* { dg-final { scan-assembler "j\[ez\]" } } */
 /* { dg-final { scan-assembler "sete" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm0,\[^\\n\\r\]*" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm1,\[^\\n\\r\]*16\[^\\n\\r\]*" } } */
@@ -18,6 +19,14 @@
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm5,\[^\\n\\r\]*80\[^\\n\\r\]*" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm6,\[^\\n\\r\]*96\[^\\n\\r\]*" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm7,\[^\\n\\r\]*112\[^\\n\\r\]*" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm0, %xmm0" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm1, %xmm1" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm2, %xmm2" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm3, %xmm3" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm4, %xmm4" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm5, %xmm5" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm6, %xmm6" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm7, %xmm7" } } */
 
 #include <immintrin.h>
 
diff --git a/gcc/testsuite/gcc.target/i386/keylocker-aesencwide256kl.c b/gcc/testsuite/gcc.target/i386/keylocker-aesencwide256kl.c
index 125a787..31463a8 100644
--- a/gcc/testsuite/gcc.target/i386/keylocker-aesencwide256kl.c
+++ b/gcc/testsuite/gcc.target/i386/keylocker-aesencwide256kl.c
@@ -9,6 +9,7 @@
 /* { dg-final { scan-assembler "movdqu\[ \\t\]+\[^\\n\\r\]*96\[^\\n\\r\]*, %xmm6" } } */
 /* { dg-final { scan-assembler "movdqu\[ \\t\]+\[^\\n\\r\]*112\[^\\n\\r\]*, %xmm7" } } */
 /* { dg-final { scan-assembler "aesencwide256kl\[ \\t\]+\[^\\n\\r\]*" } } */
+/* { dg-final { scan-assembler "j\[ez\]" } } */
 /* { dg-final { scan-assembler "sete" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm0,\[^\\n\\r\]*" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm1,\[^\\n\\r\]*16\[^\\n\\r\]*" } } */
@@ -18,6 +19,14 @@
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm5,\[^\\n\\r\]*80\[^\\n\\r\]*" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm6,\[^\\n\\r\]*96\[^\\n\\r\]*" } } */
 /* { dg-final { scan-assembler "(?:movdqu|movups)\[ \\t\]+\[^\\n\\r\]*%xmm7,\[^\\n\\r\]*112\[^\\n\\r\]*" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm0, %xmm0" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm1, %xmm1" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm2, %xmm2" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm3, %xmm3" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm4, %xmm4" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm5, %xmm5" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm6, %xmm6" } } */
+/* { dg-final { scan-assembler "pxor\[ \t\]+%xmm7, %xmm7" } } */
 
 #include <immintrin.h>
 
-- 
2.7.4