Mirrors/Ryujinx
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

ARMeilleure: Add initial support for AVX512(EVEX encoding) (#3663)

Browse source 
* ARMeilleure: Add AVX512{F,VL,DQ,BW} detection

Add `UseAvx512Ortho` and `UseAvx512OrthoFloat` optimization flags as
short-hands for `F+VL` and `F+VL+DQ`.

* ARMeilleure: Add initial support for EVEX instruction encoding

Does not implement rounding, or exception controls.

* ARMeilleure: Add `X86Vpternlogd`

Accelerates the vector-`Not` instruction.

* ARMeilleure: Add check for `OSXSAVE` for AVX{2,512}

* ARMeilleure: Add check for `XCR0` flags

Add XCR0 register checks for AVX and AVX512F, following the guidelines
from section 14.3 and 15.2 from the Intel Architecture Software
Developer's Manual.

* ARMeilleure: Increment InternalVersion

* ARMeilleure: Remove redundant `ReProtect` and `Dispose`, formatting

* ARMeilleure: Move XCR0 procedure to GetXcr0Eax

* ARMeilleure: Add `XCR0` to `FeatureInfo` structure

* ARMeilleure: Utilize `ReadOnlySpan` for Xcr0 assembly

Avoids an additional allocation

* ARMeilleure: Formatting fixes
This commit is contained in:
Wunk 2022-12-18 11:46:13 -08:00 • committed by GitHub
parent d7310d7a1c
commit 295fbd0542
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
10 changed files with 193 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

11
ARMeilleure/Optimizations.cs
View file

@ -17,6 +17,10 @@ namespace ARMeilleure
public static bool UseSse42IfAvailable { get; set; } = true;
public static bool UsePopCntIfAvailable { get; set; } = true;
public static bool UseAvxIfAvailable { get; set; } = true;
public static bool UseAvx512FIfAvailable { get; set; } = true;
public static bool UseAvx512VlIfAvailable { get; set; } = true;
public static bool UseAvx512BwIfAvailable { get; set; } = true;
public static bool UseAvx512DqIfAvailable { get; set; } = true;
public static bool UseF16cIfAvailable { get; set; } = true;
public static bool UseFmaIfAvailable { get; set; } = true;
public static bool UseAesniIfAvailable { get; set; } = true;
@ -38,11 +42,18 @@ namespace ARMeilleure
internal static bool UseSse42 => UseSse42IfAvailable && HardwareCapabilities.SupportsSse42;
internal static bool UsePopCnt => UsePopCntIfAvailable && HardwareCapabilities.SupportsPopcnt;
internal static bool UseAvx => UseAvxIfAvailable && HardwareCapabilities.SupportsAvx && !ForceLegacySse;
internal static bool UseAvx512F => UseAvx512FIfAvailable && HardwareCapabilities.SupportsAvx512F && !ForceLegacySse;
internal static bool UseAvx512Vl => UseAvx512VlIfAvailable && HardwareCapabilities.SupportsAvx512Vl && !ForceLegacySse;
internal static bool UseAvx512Bw => UseAvx512BwIfAvailable && HardwareCapabilities.SupportsAvx512Bw && !ForceLegacySse;
internal static bool UseAvx512Dq => UseAvx512DqIfAvailable && HardwareCapabilities.SupportsAvx512Dq && !ForceLegacySse;
internal static bool UseF16c => UseF16cIfAvailable && HardwareCapabilities.SupportsF16c;
internal static bool UseFma => UseFmaIfAvailable && HardwareCapabilities.SupportsFma;
internal static bool UseAesni => UseAesniIfAvailable && HardwareCapabilities.SupportsAesni;
internal static bool UsePclmulqdq => UsePclmulqdqIfAvailable && HardwareCapabilities.SupportsPclmulqdq;
internal static bool UseSha => UseShaIfAvailable && HardwareCapabilities.SupportsSha;
internal static bool UseGfni => UseGfniIfAvailable && HardwareCapabilities.SupportsGfni;
internal static bool UseAvx512Ortho => UseAvx512F && UseAvx512Vl;
internal static bool UseAvx512OrthoFloat => UseAvx512Ortho && UseAvx512Dq;
}
}