Re: [PATCH 3/4] x86: save user rsp in pt_regs->sp on SYSCALL64 fastpath

[Denys Vlasenko]

On Tue, Mar 10, 2015 at 2:21 PM, Ingo Molnar <mingo@xxxxxxxxxx> wrote:
>> Since this patch does add two extra MOVs,
>> I did benchmark these patches. They add exactly one cycle
>> to system call code path on my Sandy Bridge CPU.
>
> Hm, but that's the wrong direction, we should try to make it faster,
> and to clean it up

As you know, goals of "faster" and "cleaner" are often mutually exclusive.
C'est la vie :(

entry.S seem to lean towards "faster" to the extent where it became
a tangled maze of obscure optimizations.

Such as the mysterious, and not at all obvious existence of 8 byte
"safety padding" at the top of the 32-bit kernel stack. Before Andy
stumbled into it, it was not at all obvious that it is even there.
I had to write a test patch to verify it.
There is a long-standing latent bug in the code where this padding
is not accounted for.

> - but making it slower without really good reasons isn't good.

The thinking here is that cleaning up entry.S is a good reason.

We won't do anything which would slow it down by, say, 5%,
but one cycle may be considered acceptable loss.
--
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/