[glibc/ibm/2.32/master] x86: Set Prefer_No_VZEROUPPER and add Prefer_AVX2_STRCMP

[glibc/ibm/2.32/master] x86: Set Prefer_No_VZEROUPPER and add Prefer_AVX2_STRCMP

[Raoni Fassina Firmino]

https://sourceware.org/git/gitweb.cgi?p=glibc.git;h=3bb13e91495c763c1ce7759e3b0d2cb32d4095b6

commit 3bb13e91495c763c1ce7759e3b0d2cb32d4095b6
Author: H.J. Lu <hjl.tools@gmail.com>
Date:   Fri Feb 26 05:36:59 2021 -0800

    x86: Set Prefer_No_VZEROUPPER and add Prefer_AVX2_STRCMP
    
    1. Set Prefer_No_VZEROUPPER if RTM is usable to avoid RTM abort triggered
    by VZEROUPPER inside a transactionally executing RTM region.
    2. Since to compare 2 32-byte strings, 256-bit EVEX strcmp requires 2
    loads, 3 VPCMPs and 2 KORDs while AVX2 strcmp requires 1 load, 2 VPCMPEQs,
    1 VPMINU and 1 VPMOVMSKB, AVX2 strcmp is faster than EVEX strcmp.  Add
    Prefer_AVX2_STRCMP to prefer AVX2 strcmp family functions.
    
    (cherry picked from commit 1da50d4bda07f04135dca39f40e79fc9eabed1f8)

Diff:
---
 sysdeps/x86/cpu-features.c | 20 ++++++++++++++++++--
 sysdeps/x86/cpu-features.h |  2 ++
 sysdeps/x86/cpu-tunables.c |  2 ++
 3 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/sysdeps/x86/cpu-features.c b/sysdeps/x86/cpu-features.c
index 4c24ba7c31..eb554b6d14 100644
--- a/sysdeps/x86/cpu-features.c
+++ b/sysdeps/x86/cpu-features.c
@@ -546,8 +546,24 @@ init_cpu_features (struct cpu_features *cpu_features)
 	cpu_features->preferred[index_arch_Prefer_No_VZEROUPPER]
 	  |= bit_arch_Prefer_No_VZEROUPPER;
       else
-	cpu_features->preferred[index_arch_Prefer_No_AVX512]
-	  |= bit_arch_Prefer_No_AVX512;
+	{
+	  cpu_features->preferred[index_arch_Prefer_No_AVX512]
+	    |= bit_arch_Prefer_No_AVX512;
+
+	  /* Avoid RTM abort triggered by VZEROUPPER inside a
+	     transactionally executing RTM region.  */
+	  if (CPU_FEATURE_USABLE_P (cpu_features, RTM))
+	    cpu_features->preferred[index_arch_Prefer_No_VZEROUPPER]
+	      |= bit_arch_Prefer_No_VZEROUPPER;
+
+	  /* Since to compare 2 32-byte strings, 256-bit EVEX strcmp
+	     requires 2 loads, 3 VPCMPs and 2 KORDs while AVX2 strcmp
+	     requires 1 load, 2 VPCMPEQs, 1 VPMINU and 1 VPMOVMSKB,
+	     AVX2 strcmp is faster than EVEX strcmp.  */
+	  if (CPU_FEATURE_USABLE_P (cpu_features, AVX2))
+	    cpu_features->preferred[index_arch_Prefer_AVX2_STRCMP]
+	      |= bit_arch_Prefer_AVX2_STRCMP;
+	}
     }
   /* This spells out "AuthenticAMD" or "HygonGenuine".  */
   else if ((ebx == 0x68747541 && ecx == 0x444d4163 && edx == 0x69746e65)
diff --git a/sysdeps/x86/cpu-features.h b/sysdeps/x86/cpu-features.h
index a0b9b9177c..72ea74d083 100644
--- a/sysdeps/x86/cpu-features.h
+++ b/sysdeps/x86/cpu-features.h
@@ -804,6 +804,7 @@ extern const struct cpu_features *__get_cpu_features (void)
 #define bit_arch_Prefer_FSRM			(1u << 13)
 #define bit_arch_Prefer_No_AVX512		(1u << 14)
 #define bit_arch_MathVec_Prefer_No_AVX512	(1u << 15)
+#define bit_arch_Prefer_AVX2_STRCMP		(1u << 16)
 
 #define index_arch_Fast_Rep_String		PREFERRED_FEATURE_INDEX_1
 #define index_arch_Fast_Copy_Backward		PREFERRED_FEATURE_INDEX_1
@@ -821,6 +822,7 @@ extern const struct cpu_features *__get_cpu_features (void)
 #define index_arch_Prefer_No_AVX512		PREFERRED_FEATURE_INDEX_1
 #define index_arch_MathVec_Prefer_No_AVX512	PREFERRED_FEATURE_INDEX_1
 #define index_arch_Prefer_FSRM			PREFERRED_FEATURE_INDEX_1
+#define index_arch_Prefer_AVX2_STRCMP		PREFERRED_FEATURE_INDEX_1
 
 /* XCR0 Feature flags.  */
 #define bit_XMM_state		(1u << 1)
diff --git a/sysdeps/x86/cpu-tunables.c b/sysdeps/x86/cpu-tunables.c
index 588bbf9448..b251a91af3 100644
--- a/sysdeps/x86/cpu-tunables.c
+++ b/sysdeps/x86/cpu-tunables.c
@@ -238,6 +238,8 @@ TUNABLE_CALLBACK (set_hwcaps) (tunable_val_t *valp)
 	      CHECK_GLIBC_IFUNC_PREFERRED_BOTH (n, cpu_features,
 						Fast_Copy_Backward,
 						disable, 18);
+	      CHECK_GLIBC_IFUNC_PREFERRED_NEED_BOTH
+		(n, cpu_features, Prefer_AVX2_STRCMP, AVX2, disable, 18);
 	    }
 	  break;
 	case 19: