Re: crash in entry.S restore_all, 2.6.12-rc2, x86, PAGEALLOC

From: Richard B. Johnson
Date: Thu Apr 07 2005 - 12:19:44 EST


On Thu, 7 Apr 2005, Dave Jones wrote:

On Thu, Apr 07, 2005 at 07:47:41AM -0700, Linus Torvalds wrote:

> So the sysenter sequence might as well look like
>
> pushl $(__USER_DS)
> pushl %ebp
> sti
> pushfl
> ..
>
> which actually does three protected pushes thanks to the one-instruction
> "interrupt shadow" after an sti.

Is this guaranteed on every x86 variant (or rather, every one
that has SEP). ?

Dave

The i486 book says that if the next instruction lets the IF
flag remain enabled, then any interrupts pending occur after
the next instruction. If the next instruction is CLI, no
interrupts will occur. There is a note:

"In case of an NM1, trap, or fault following ST1, the interrupt
will be taken before executing the next sequential instruction
in the code." The use of "NM1" and "ST1" is (sic). I don't
know what NM1 is, it can't be NMI because, by definition NMI
is not maskable so CLI and STI could not affect this pin.

Cheers,
Dick Johnson
Penguin : Linux version 2.6.11 on an i686 machine (5537.79 BogoMips).
Notice : All mail here is now cached for review by Dictator Bush.
98.36% of all statistics are fiction.
-
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/