Method alloc

tinygrad/runtime/support/system.py:258–272

(self, size:int, host=False, uncached=False, cpu_access=False, contiguous=False, force_devmem=False, **kwargs)

    self.dev, self.vram_bar, self.count = dev, vram_bar, len(hcq_filter_visible_devices(System.list_devices(vendor, devices, base_class), dn))

  def alloc(self, size:int, host=False, uncached=False, cpu_access=False, contiguous=False, force_devmem=False, **kwargs) -> HCQBuffer:
    should_use_sysmem = host or ((cpu_access if self.is_bar_small() else (uncached and cpu_access)) and not force_devmem)

    # Align size to huge pages for large allocations, otherwise the unaligned tail falls back to 4KB pages, increasing TLB pressure.
    size = round_up(size, mmap.PAGESIZE if should_use_sysmem else ((2 << 20) if size >= (8 << 20) else (4 << 10)))

    if should_use_sysmem:
      vaddr = self.dev_impl.mm.alloc_vaddr(size:=round_up(size, mmap.PAGESIZE), align=mmap.PAGESIZE)
      memview, paddrs = self.pci_dev.alloc_sysmem(size, vaddr=vaddr, contiguous=contiguous)
      mapping = self.dev_impl.mm.map_range(vaddr, size, [(paddr, 0x1000) for paddr in paddrs], aspace=AddrSpace.SYS, snooped=True, uncached=True)
      return HCQBuffer(vaddr, size, meta=PCIAllocationMeta(mapping, has_cpu_mapping=True, hMemory=paddrs[0]), view=memview, owner=self.dev)

    mapping = self.dev_impl.mm.valloc(size:=round_up(size, 0x1000), uncached=uncached, contiguous=cpu_access)
    barview = self.pci_dev.map_bar(bar=self.vram_bar, off=mapping.paddrs[0][0], size=mapping.size) if cpu_access else None
    return HCQBuffer(mapping.va_addr, size, view=barview, meta=PCIAllocationMeta(mapping, cpu_access, hMemory=mapping.paddrs[0][0]), owner=self.dev)

  def free(self, b:HCQBuffer):
    if b.owner != self.dev: self.dev.iface.dev_impl.mm.unmap_range(b.va_addr, b.size)