On Fri, Jan 09, 2015 at 10:55:29AM +0000, Leif Lindholm wrote: > On Fri, Jan 09, 2015 at 10:19:50AM +0000, Mark Rutland wrote: > > On Thu, Jan 08, 2015 at 07:09:22PM +0000, Ard Biesheuvel wrote: > > > On 8 January 2015 at 19:04, Mark Rutland <mark.rutland@xxxxxxx> wrote: > > > > Hi Ard, > > > > > > > > On Thu, Jan 08, 2015 at 05:51:47PM +0000, Ard Biesheuvel wrote: > > > >> This fixes two minor issues in the implementation of get_memory_map(): > > > >> - Currently, it assumes that sizeof(efi_memory_desc_t) == desc_size, > > > >> which is usually true, but not mandated by the spec. (This was added > > > >> intentionally to allow future additions to the definition of > > > >> efi_memory_desc_t). The way the loop is implemented currently, the > > > >> added slack space may be insufficient if desc_size is larger, which in > > > >> some corner cases could result in the loop never terminating. > > > >> - It allocates 32 efi_memory_desc_t entries first (again, using the size > > > >> of the struct instead of desc_size), and frees and reallocates if it > > > >> turns out to be insufficient. Few implementations of UEFI have such small > > > >> memory maps, which results in a unnecessary allocate/free pair on each > > > >> invocation. > > > >> > > > >> Fix this by calling the get_memory_map() boot service first with a '0' > > > >> input value for map size to retrieve the map size and desc size from the > > > >> firmware and only then perform the allocation, using desc_size rather > > > >> than sizeof(efi_memory_desc_t). > > > > > > > > Is the desc_size guaranteed to be set up correctly if the size is too > > > > small? > > > > > > > > As far as I can see, for that case the spec only mandates that > > > > MemoryMapSize is updated nd EFI_BUFFER_TOO_SMALL is returned. > > > > > > > > It's not clear to me whether DescriptorSize or DescriptorVersion are > > > > initialised in cases other than success. > > > > > > > > > > The way I read it, descriptor size and descriptor version are always > > > returned, e.g., as opposed to MapKey, which is only returned on > > > success, and the spec mentions that specifically. > > > > I agree that that would be the sensible reading of the spec. I just fear > > that there's room for an implementor to read it slightly differently, > > and cause pain for us. > > I disagree. > The intent is clear, and you can not follow the current text and > provide a version that does not do the right thing. Only the MapKey > return is conditional on success. > > > > We could ask for clarification just to be sure. > > > > I think we should. > > > > Given it could take a while for any conclusion to be reached and > > published, I'm happy to go with the patch below in the meantime, so long > > as the issue gets raised. > > We could ask for clarification of the spec, but I see no need to ask > for guidance on the current text. I'm fine with that. Mark. -- To unsubscribe from this list: send the line "unsubscribe linux-efi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
