Re: [RFC PATCH] x86/cpu: Fix MSR value truncation issue

From: Andy Lutomirski
Date: Fri Oct 30 2015 - 15:34:28 EST


On Fri, Oct 30, 2015 at 12:32 PM, Borislav Petkov <bp@xxxxxxxxx> wrote:
> On Fri, Oct 30, 2015 at 12:26:42PM -0700, Andy Lutomirski wrote:
>> Want to add that to the patch or make it another patch?
>
> Yeah, I'll make another one as it is going to document why we're
> explicitly ANDing with 0xffffffffull.
>
>> Fair enough. I suppose that this thing is a handful of separate
>> fields as opposed to being just a number.
>
> You mean MSR_STAR?
>
> Just the two upper 16-bit values. The lower 32-bit are reserved on Intel
> while on AMD they're "32-bit SYSCALL Target EIP", meaning that you can
> use SYSCALL on 32-bit too.

Yeah, exactly. When I wrote the "make wrmsrl a function" patch, I was
dealing with MSRs that were genuinely 64-bit numbers, except that in
some cases they provably fit in 32 bits, and gcc was unhappy.

--Andy
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/