Re: [PATCH v2] x86/fault: Send a SIGBUS to user process always for hwpoison page access.
From: Andy Lutomirski <luto@kernel.org>
Date: 2021-02-01 16:59:32
Also in:
lkml
On Mon, Feb 1, 2021 at 12:17 AM Aili Yao [off-list ref] wrote:
When one page is already hwpoisoned by AO action, process may not be killed, the process mapping this page may make a syscall include this page and result to trigger a VM_FAULT_HWPOISON fault, if it's in kernel mode it may be fixed by fixup_exception. Current code will just return error code to user process. This is not sufficient, we should send a SIGBUS to the process and log the info to console, as we can't trust the process will handle the error correctly.
Does this happen when one process gets SIGBUSed due to memory failure and another process independently hits the poisoned memory? I'm not entirely convinced that this is a problem. In any case, this patch needs rebasing on top of my big fault series -- as it stands, it's way too difficult to keep track of which paths even call your new code.. And the various signal paths need to be consolidated -- we already have three of them, and the last thing we need is a fourth.