On 12/15/21 19:09, Liam Howlett wrote: >> - the future lookup of address vma->vm_end will return this vma, although >> the address doesn't belong to it. Wouldn't be surprised if this caused >> infrequent mysterious bugs? >> >> Both will resolve with vmacache removal later, but better still fix this. > This is certainly worth fixing. I am surprised that I did not hit a > random bug as you said above, I would expect it to occur as well. > Although, I didn't run each patch individually for long unless I was > tracking down an issue on a rebase. Ah, it doesn't cause a bug because vmacache_find() checks the boundaries of vmas in the vmacache.
