linux-kernel - Re: [PATCH] x86/fault: Send SIGBUS to user process always for hwpoison page access.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20210129225529.GA42172@agluck-desk2.amr.corp.intel.com>
Date:   Fri, 29 Jan 2021 14:55:29 -0800
From:   "Luck, Tony" <tony.luck@...el.com>
To:     Aili Yao <yaoaili@...gsoft.com>
Cc:     x86@...nel.org, naoya.horiguchi@....com,
        linux-kernel@...r.kernel.org, yangfeng1@...gsoft.com
Subject: Re: [PATCH] x86/fault: Send SIGBUS to user process always for
 hwpoison page access.

Thanks for the explanation and test code. I think I see better what
is going on here.

[I took your idea for using madvise(...MADV_HWPOISON) and added a new "-S"
option to my einj_mem_uc test program to use madvise instead of ACPI/EINJ
for injections. Update pushed here:
	git://git.kernel.org/pub/scm/linux/kernel/git/aegl/ras-tools.git ]

There have been some small changes to arch/x86/mm/fault.c since you wrote
the patch.  Can you rebase to v5.11-rc5?

Also maybe this might be a case to use IS_ENABLED() instead of #ifdef to
make the code a little less ugly. At least for the 2nd hunk in your patch
this would work well:

	if (IS_ENABLED(CONFIG_MEMORY_FAILURE) &&
	    (fault & (VM_FAULT_HWPOISON|VM_FAULT_HWPOISON_LARGE)))
		no_context(regs, error_code, address, SIGBUS, BUS_MCEERR_AR);
	else
		no_context(regs, error_code, address, SIGBUS, BUS_ADRERR);

The first hunk might need a bit more thought.

-Tony