On Mon, Mar 9, 2015 at 3:00 PM, Andy Lutomirski <[email protected]> wrote: >> I'm preparing a patch which does this: >> >> btl $X86_EFLAGS_VM_BIT,PT_EFLAGS(%esp) >> jc restore_nocheck # VM set, not it >> testb $3,PT_CS(%esp) >> jz restore_nocheck # CPL0, not it >> # Note: we access PT_OLDSS only when we know it exists. >> # If PT_CS is from CPL0, this can be not true. >> testb $SEGMENT_TI_MASK,PT_OLDSS(%esp) >> jnz ldt_ss # returning to user-space with LDT SS >> >> All three checks can run in parallel on an OOO CPU. >> Most of the time, none of branches will be taken. > > Why do we need the VM check? Isn't CS == 0 in vm86 mode?
IF VM is set, then we are in vm86 mode, not in 16-bit protected mode where ESPFIX thingy may be necessary. And also, %ss register in vm86 does nit have LDT bit, thus that check would not work properly anyway. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
