hcq: better errors for ifaces (#11751)

* hcq: better errors for ifaces

* fix linter

* typo

* space
This commit is contained in:
nimlgen
2025-08-20 17:50:51 +03:00
committed by GitHub
parent be7b0b6970
commit 6589c9e643
2 changed files with 10 additions and 5 deletions

View File

@@ -438,12 +438,13 @@ class HCQCompiled(Compiled, Generic[SignalType]):
return buf, realloced
def _select_iface(self, *ifaces:Type):
errs:str = ""
errs, err_short = "", ""
if val:=getenv(f'{type(self).__name__[:-6].upper()}_IFACE', ""): ifaces = tuple(x for x in ifaces if x.__name__.startswith(val.upper()))
for iface_t in ifaces:
try: return iface_t(self, self.device_id)
except Exception: errs += f"\n{iface_t.__name__}: {traceback.format_exc()}"
raise RuntimeError(f"Cannot find a usable interface for {type(self).__name__[:-6]}:{self.device_id}:\n{errs}")
except Exception as e: errs, err_short = errs + f"\n{iface_t.__name__}: {traceback.format_exc()}", err_short + f"\n{iface_t.__name__}: {e}"
raise RuntimeError(f"{errs}\nNo interface for {type(self).__name__[:-6]}:{self.device_id} is available:{err_short}\n" \
f"\nForce an interface with {type(self).__name__[:-6].upper()}_IFACE={('|'.join(x.__name__[:-5] for x in ifaces))}.")
def _is_cpu(self) -> bool: return hasattr(self, 'device') and self.device.split(":")[0] in ("CPU", "LLVM")

View File

@@ -1,4 +1,4 @@
import os, mmap, array, functools, ctypes, select, contextlib, dataclasses, sys
import os, mmap, array, functools, ctypes, select, contextlib, dataclasses, sys, errno
from typing import cast, ClassVar
from tinygrad.helpers import round_up, to_mv, getenv, OSX, temp
from tinygrad.runtime.autogen import libc, vfio
@@ -84,7 +84,11 @@ class PCIDevice:
for i in resize_bars or []:
if FileIOInterface.exists(rpath:=f"/sys/bus/pci/devices/{self.pcibus}/resource{i}_resize"):
try: FileIOInterface(rpath, os.O_RDWR).write(str(int(FileIOInterface(rpath, os.O_RDONLY).read(), 16).bit_length() - 1))
except OSError as e: raise RuntimeError(f"Cannot resize BAR {i}: {e}. Ensure the resizable BAR option is enabled on your system.") from e
except OSError as e:
if e.errno == errno.EPERM:
raise RuntimeError(f"Cannot resize BAR {i}: {e}. Permission error: run `extra/amdpci/setup_python_cap.sh`"
" to allow python accessing device or run with sudo") from e
raise RuntimeError(f"Cannot resize BAR {i}: {e}. Ensure the resizable BAR option is enabled on your system.") from e
if getenv("VFIO", 0) and (vfio_fd:=System.vfio()) is not None:
FileIOInterface(f"/sys/bus/pci/devices/{self.pcibus}/driver_override", os.O_WRONLY).write("vfio-pci")