qsimd: add support for new x86 CPU features

This adds detection for: VAES, GFNI, AVX512VBMI2, AVX512VNNI,
AVX512BITALG, AVX512VPOPCNTDQ, AVX512_4NNIW, AVX512_4FMAPS. These
features were found in the "Intel® Architecture Instruction Set
Extensions and Future Features" manual, revision 30. This commit also
adds support for RDPID (already in the main manual) and the Control-flow
Enforcement Technology, which appears in a separate Intel paper.

This new support was done by adding a new generator script so we don't
have to maintain two tables in sync, one in qsimd.cpp with the feature
names, and the other in qsimd_p.h.

Since we now need a lot more bits, it's no longer worth keeping the two
halves of the qt_cpu_features variable mostly similar to the main two
CPUID results. This commit goes back to keeping things in order, like we
used to prior to commit 6a8251a89b6a61258498f4af1ba7b3d5b7f7096c (Qt 5.6)

At the time of this commit, GCC 8 has macros for AVX512VPOPCNTDQ,
AVX512_4NNIW, AVX512_4FMAPS, AVX512VBMI2 and GFNI.

Change-Id: I938b024e38bf4aac9154fffd14f7afae50faaa96
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>

1 files changed, 33 insertions, 1 deletions

diff --git a/config.tests/arch/arch.cpp b/config.tests/arch/arch.cpp
index bb3efec177..2be3630af8 100644
--- a/config.tests/arch/arch.cpp
+++ b/config.tests/arch/arch.cpp
@@ -115,6 +115,22 @@ const char msg2[] = "==Qt=magic=Qt== Sub-architecture:"
 // AVX512 Vector Byte Manipulation Instructions, Intel processor codename "Cannonlake"
 " avx512vbmi"
 #endif
+#ifdef __AVX512VBMI2__
+// AVX512 Vector Byte Manipulation Instructions #2, Intel processor codename "Ice Lake"
+" avx512vbmi2"
+#endif
+#ifdef __AVX512VPOPCNTDQ__
+// AVX512 Vector Population Count Double & Quad, Future Intel Xeon Phi processor codename "Knights Mill", Intel processor codename "Ice Lake"
+" avx512vpopcntdq"
+#endif
+#ifdef __AVX5124FMAPS__
+// AVX512 4-iteration Fused Multiply Accumulation Packed Single, Future Intel Xeon Phi processor codename "Knights Mill"
+" avx5124fmaps"
+#endif
+#ifdef __AVX5124VNNIW__
+// AVX512 4-iteration Vector Neural Network Instructions Word, Future Intel Xeon Phi processor codename "Knights Mill"
+" avx5124vnniw"
+#endif
 #ifdef __BMI__
 // Bit Manipulation Instructions 1, Intel Core 4th Generation ("Haswell"), AMD "Bulldozer 2"
 " bmi"
@@ -145,6 +161,14 @@ const char msg2[] = "==Qt=magic=Qt== Sub-architecture:"
 // rdfsgsbase, wrfsgsbase, Intel Core 3rd Generation ("Ivy Bridge")
 " fsgsbase"
 #endif
+#ifdef __GFNI__
+// Galois Field new instructions, Intel processor codename "Ice Lake"
+" gfni"
+#endif
+#ifdef __IBT__
+// Indirect Branch Tracking, Intel processor TBA
+" ibt"
+#endif
 #ifdef __LWP__
 // LWP instructions, AMD "Bulldozer"
 " lwp"
@@ -186,6 +210,10 @@ const char msg2[] = "==Qt=magic=Qt== Sub-architecture:"
 // Prefetch data for writing, Intel Core 5th Generation ("Broadwell")
 " prfchw"
 #endif
+#ifdef __RDPID__
+// Read Processor ID, Intel processors codename "Ice Lake" and "Goldmont Plus"
+" rdpid"
+#endif
 #ifdef __RDRND__
 // Random number generator, Intel Core 3rd Generation ("Ivy Bridge")
 " rdrnd"
@@ -199,9 +227,13 @@ const char msg2[] = "==Qt=magic=Qt== Sub-architecture:"
 " rtm"
 #endif
 #ifdef __SHA__
-// SHA-1 and SHA-256 instructions, Intel processor TBA
+// SHA-1 and SHA-256 instructions, Intel processors codename "Cannon Lake" and "Goldmont"
 " sha"
 #endif
+#ifdef __SHSTK__
+// Shadow stack, Intel processor TBA
+" shstk"
+#endif
 #if defined(__SSE__) || (defined(_M_IX86_FP) && _M_IX86_FP >= 1) || defined(_M_X64)
 // Streaming SIMD Extensions, Intel Pentium III, AMD Athlon
 " sse"