1580158
/
automatic
mirror of https://github.com/vladmandic/automatic
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

IPEX set UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS to 1

pull/4195/head
Disty0 2025-09-08 15:47:42 +03:00
parent 78ab98b92e
commit f5109657b1
2 changed files with 3 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
installer.py
View File

@ -803,6 +803,9 @@ def install_ipex():
if os.environ.get("PYTORCH_ENABLE_XPU_FALLBACK", None) is None:
os.environ.setdefault('PYTORCH_ENABLE_XPU_FALLBACK', '1') # CPU fallback for unsupported ops
if os.environ.get("UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS", None) is None:
os.environ.setdefault('UR_L0_ENABLE_RELAXED_ALLOCATION_LIMITS', '1') # Work around the 4G alloc limit on Alchemist
# FP64 emulation causes random UR Errors
#if os.environ.get("OverrideDefaultFP64Settings", None) is None:
# os.environ.setdefault('OverrideDefaultFP64Settings', '1')

8
modules/intel/ipex/hijacks.py
View File

@ -17,14 +17,6 @@ device_supports_fp64 = torch.xpu.has_fp64_dtype() if hasattr(torch.xpu, "has_fp6
if os.environ.get('IPEX_FORCE_ATTENTION_SLICE', '0') == '0':
if torch_version[0] > 2 or (torch_version[0] == 2 and torch_version[1] >= 7):
use_dynamic_attention = False # torch 2.7 has flash atten support
elif (torch.xpu.get_device_properties(devices.device).total_memory / 1024 / 1024 / 1024) > 4.1:
try:
x = torch.ones((33000,33000), dtype=torch.float32, device=devices.device)
del x
torch.xpu.empty_cache()
use_dynamic_attention = False
except Exception:
use_dynamic_attention = True
else:
use_dynamic_attention = True
else: