[RFC PATCH v1 0/2] selftests/x86: sysret_rip update for FRED system

From: Ammar Faizi
Date: Mon Jan 23 2023 - 19:26:42 EST


From: Ammar Faizi <ammarfaizi2@xxxxxxxxxxx>

On Mon, 23 Jan 2023 15:58:12 -0800, "H. Peter Anvin" wrote:
> On 1/23/23 15:43, Ammar Faizi wrote:
> >
> > Align them to spot differences:
> >
> > 0x200893 = 0b1000000000100010010011
> > 0x200a93 = 0b1000000000101010010011
> > ^
> >
> > Or just xor them to find the differences:
> >
> > (gdb) p/x 0x200893 ^ 0x200a93
> > $3 = 0x200
> >
> > ** Checks my Intel SDM cheat sheets. **
> >
> > Then, I was like "Oh, that's (1 << 9) a.k.a. IF. Of course we can't
> > change rflags[IF] from userspace!!!".
> >
> > In short, we can't use 0x200893 as the rflags_sentinel value because it
> > clears the interrupt flag.
> >
>
> Right, my mistake.

I changed it to 0x200a93. The test passed on my machine. But I don't
have a FRED system to test the special case.

Didn't manage to apply the feedback from Andrew about the way to handle
redzone properly, though.

Something like this...

----------

This is just an RFC patchset.

Xin Li reported sysret_rip test fails at:

assert(ctx->uc_mcontext.gregs[REG_EFL] ==
ctx->uc_mcontext.gregs[REG_R11]);

in a FRED system. Handle the FRED system scenario too. There are two
patches in this series. Comments welcome...

Note: Only tested for 'syscall' sets %rcx=%rip and %r11=%rflags case.
I don't have a FRED system to test it.

How to test this:

$ make -C tools/testing/selftests/x86
$ tools/testing/selftests/x86/sysret_rip_64

Link: https://lore.kernel.org/lkml/5d4ad3e3-034f-c7da-d141-9c001c2343af@xxxxxxxxx
Signed-off-by: Ammar Faizi <ammarfaizi2@xxxxxxxxxxx>
---

Ammar Faizi (2):
selftests/x86: sysret_rip: Handle syscall in a FRED system
selftests/x86: sysret_rip: Add more syscall tests with respect to `%rcx` and `%r11`

tools/testing/selftests/x86/sysret_rip.c | 105 ++++++++++++++++++++++-
1 file changed, 104 insertions(+), 1 deletion(-)


base-commit: e12ad468c22065a2826b2fc4c11d2113a7975301
--
Ammar Faizi