Hi Finn,
Am 09.04.2019 um 17:25 schrieb Finn Thain:
On Tue, 9 Apr 2019, Michael Schmitz wrote:
Seems real enough to me.
I agree. I was wondering if you'd seen anything like it before but I see
that the question is redundant. Had anything like it come up before, I
reckon you would have mentioned it when bus_error030() was discussed a
year ago.
Quite so.
Now the question I'm asking myself is: why does pagefault_disable() not
suppress the bus error here:
long __probe_kernel_read(void *dst, const void *src, size_t size)
{
long ret;
mm_segment_t old_fs = get_fs();
set_fs(KERNEL_DS);
pagefault_disable();
current->kernel_uaccess_faults_ok++;
ret = __copy_from_user_inatomic(dst,
(__force const void __user *)src, size);
current->kernel_uaccess_faults_ok--;
pagefault_enable();
set_fs(old_fs);
return ret ? -EFAULT : 0;
}
If I read the m68k mm/fault.c code right, we should see user space take
a SIGBUS, not SIGSEGV.
And our implementation of generic_copy_from_user() has exception
handling, but that is only used to fix up the copied byte count returned?
Cheers,
Michael
