86 lines
3.0 KiB
Diff
86 lines
3.0 KiB
Diff
From cd17e816bbac907f0fd704828230bb04db6921dd Mon Sep 17 00:00:00 2001
|
|
From: Tom Rix <Tom.Rix@amd.com>
|
|
Date: Mon, 28 Oct 2024 05:42:27 -0700
|
|
Subject: [PATCH] Add gfx1151 support
|
|
|
|
Cherry-picked from upstream 642974ee2f3f9d8a8280bf87c462645fb07a2897
|
|
|
|
Signed-off-by: Tom Rix <Tom.Rix@amd.com>
|
|
---
|
|
Tensile/AsmCaps.py | 47 +++++++++++++++++++++++++++++++++++++++++++++-
|
|
Tensile/Common.py | 3 ++-
|
|
2 files changed, 48 insertions(+), 2 deletions(-)
|
|
|
|
diff --git a/Tensile/AsmCaps.py b/Tensile/AsmCaps.py
|
|
index 22c67e977aa0..b4899da6284c 100644
|
|
--- a/Tensile/AsmCaps.py
|
|
+++ b/Tensile/AsmCaps.py
|
|
@@ -770,4 +770,49 @@ CACHED_ASM_CAPS = \
|
|
'v_mad_mix_f32': False,
|
|
'v_mov_b64': False,
|
|
'v_pk_fma_f16': True,
|
|
- 'v_pk_fmac_f16': False}}
|
|
+ 'v_pk_fmac_f16': False},
|
|
+ (11, 5, 1): {'HasAddLshl': True,
|
|
+ 'HasAtomicAdd': True,
|
|
+ 'HasDirectToLdsDest': False,
|
|
+ 'HasDirectToLdsNoDest': False,
|
|
+ 'HasExplicitCO': True,
|
|
+ 'HasExplicitNC': True,
|
|
+ 'HasGLCModifier': True,
|
|
+ 'HasNTModifier': False,
|
|
+ 'HasLshlOr': True,
|
|
+ 'HasMFMA': False,
|
|
+ 'HasMFMA_b8': False,
|
|
+ 'HasMFMA_bf16_1k': False,
|
|
+ 'HasMFMA_bf16_original': False,
|
|
+ 'HasMFMA_constSrc': False,
|
|
+ 'HasMFMA_f64': False,
|
|
+ 'HasMFMA_f8': False,
|
|
+ 'HasMFMA_i8_908': False,
|
|
+ 'HasMFMA_i8_940': False,
|
|
+ 'HasMFMA_vgpr': False,
|
|
+ 'HasMFMA_xf32': False,
|
|
+ 'HasSMulHi': True,
|
|
+ 'HasWMMA': True,
|
|
+ 'KernargPreloading': False,
|
|
+ 'MaxLgkmcnt': 15,
|
|
+ 'MaxVmcnt': 63,
|
|
+ 'SupportedISA': True,
|
|
+ 'SupportedSource': True,
|
|
+ 'VOP3v_dot4_i32_i8': False,
|
|
+ 'v_dot2_f32_f16': True,
|
|
+ 'v_dot2c_f32_f16': True,
|
|
+ 'v_dot4_i32_i8': False,
|
|
+ 'v_dot4c_i32_i8': False,
|
|
+ 'v_fma_f16': True,
|
|
+ 'v_fma_f32': True,
|
|
+ 'v_fma_f64': True,
|
|
+ 'v_fma_mix_f32': True,
|
|
+ 'v_fmac_f16': False,
|
|
+ 'v_fmac_f32': True,
|
|
+ 'v_mac_f16': False,
|
|
+ 'v_mac_f32': False,
|
|
+ 'v_mad_mix_f32': False,
|
|
+ 'v_mov_b64': False,
|
|
+ 'v_pk_fma_f16': True,
|
|
+ 'v_pk_fmac_f16': False},
|
|
+}
|
|
diff --git a/Tensile/Common.py b/Tensile/Common.py
|
|
index 07abbf59397f..8c27486338fc 100644
|
|
--- a/Tensile/Common.py
|
|
+++ b/Tensile/Common.py
|
|
@@ -306,7 +306,8 @@ architectureMap = {
|
|
'gfx942':'aquavanjaram942', 'gfx942:xnack+':'aquavanjaram942', 'gfx942:xnack-':'aquavanjaram942',
|
|
'gfx1010':'navi10', 'gfx1011':'navi12', 'gfx1012':'navi14',
|
|
'gfx1030':'navi21', 'gfx1031':'navi22', 'gfx1032':'navi23', 'gfx1034':'navi24', 'gfx1035':'rembrandt',
|
|
- 'gfx1100':'navi31', 'gfx1101':'navi32', 'gfx1102':'navi33'
|
|
+ 'gfx1100':'navi31', 'gfx1101':'navi32', 'gfx1102':'navi33',
|
|
+ 'gfx1151':'gfx1151'
|
|
}
|
|
|
|
def getArchitectureName(gfxName):
|
|
--
|
|
2.47.0
|
|
|