Sounds about right to me. Do it before base_cpu_load, which reloads the
tss anyway. It falls appropriately into the various *_init calls right there.
___
Bug-hurd mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-hurd
On Fri, Oct 12, 2001 at 04:21:28PM -0400, Roland McGrath wrote:
> It's easy enough to either have a different ktss and switch to it, or just
> to link-time override the base_tss definition (the oskit puts it in a
> separate file to make this easy).
I think link-time is better, and I am working th
> On Sat, Oct 13, 2001 at 02:03:41AM -0400, Roland McGrath wrote:
> > Don't worry so much. Just add an alternative to the union inside struct
> > device in ds_oskit.h.
>
> Ok.
>
> > Having 8kb there for a whole bitmap might be a bit
> > much, so use a pointer to a bitmap allocated with kmalloc
On Sat, Oct 13, 2001 at 02:03:41AM -0400, Roland McGrath wrote:
> Don't worry so much. Just add an alternative to the union inside struct
> device in ds_oskit.h.
Ok.
> Having 8kb there for a whole bitmap might be a bit
> much, so use a pointer to a bitmap allocated with kmalloc or zalloc.
What
Don't worry so much. Just add an alternative to the union inside struct
device in ds_oskit.h. Having 8kb there for a whole bitmap might be a bit
much, so use a pointer to a bitmap allocated with kmalloc or zalloc.
___
Bug-hurd mailing list
[EMAIL PRO
On Wed, Oct 10, 2001 at 06:46:22PM -0400, Roland McGrath wrote:
> gnumach is never going to have SMP support anyway, so it's all moot (and
> the macros you are using expand to empty). In oskit-mach, I think we
> should get rid of the lookup table and just hang the data directly off the
> device_t
> I think I will work on a solution for oskit-mach and forget gnumach for now.
> I have a solution for gnumach ready, but there are bugs and it is more
> difficult to debug than oskit-mach. (I have a second computer do to remote
> debugging now).
:-) You won't get any argument from me!
> In osk
On Wed, Oct 10, 2001 at 06:46:22PM -0400, Roland McGrath wrote:
> gnumach is never going to have SMP support anyway, so it's all moot (and
> the macros you are using expand to empty). In oskit-mach, I think we
> should get rid of the lookup table and just hang the data directly off the
> device_t
Sure, but that is not particular to io ports. You do the same for any
privileged instruction that traps. You could call page faults and FPU
emulation the same thing.
___
Bug-hurd mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-
On Wed, Oct 10, 2001 at 06:46:22PM -0400, Roland McGrath wrote:
> > Can I/O permission violations be catched in userspace, by emulating the
> > io_emulate exception? Then I guess we can just let the default
> > implementation return an error anyway.
>
> I'm not really sure what you are asking he
> I guess we will drop the automatic io access entirely?
Yes, I think it was always a stupid feature.
> Can I/O permission violations be catched in userspace, by emulating the
> io_emulate exception? Then I guess we can just let the default
> implementation return an error anyway.
I'm not re
On Mon, Oct 08, 2001 at 07:34:25PM -0400, Roland McGrath wrote:
> (This is also where the insane automatic io access if your
> task has some send right feature is implemented.) However, it doesn't look
> like emulate_io will actually dtrt if you don't have the iopl device port,
> but almost.
I g
> I don't know yet enough. It's okay to not fix the broken code and put the
> new implementation there immediately, but I would like to have at least an
> idea why the code doesn't work "as is".
I won't discourage you from working on it. But I've always found debugging
this stuff to be pretty a
put the
new implementation there immediately, but I would like to have at least an
idea why the code doesn't work "as is".
[snipped explanation of what linux does]
> So let's do that. It's easy and you probably don't have as much weirdo
> Intel magic to debug
he bitmap in the single tss.
So let's do that. It's easy and you probably don't have as much weirdo
Intel magic to debug than with tss switching.
When we make it task based, we just move the code into the existing
old-task!=new-task test used for switching the address space.
W
Hi,
I want to start a new thread for this. Below is my current patch to fix the
I/O permission code. All of it works. However, it is broken nevertheless:
Although the user tss is correctly (AFAICS) setup, it doesn't come
effective: The program is killed with Illegal Instruction. I am at my w
16 matches
Mail list logo
