Roland McGrath <[EMAIL PROTECTED]> writes:
> > The nice thing here is that with this interface, we can fix it to use
> > the *right* (but much harder) implementation at any later date.
>
> Indeed. That's why I don't want vm_map to fail even when it can
> immediately tell that all access will fai
Thanks for testing that. I've checked in the fix.
___
Bug-hurd mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-hurd
Roland McGrath <[EMAIL PROTECTED]> writes:
> It should be in a different spot. Please let me know if this works instead
> of your patch:
That works.
> --- main.c.~1.32.~ Tue Nov 19 22:03:35 2002
> +++ main.cWed Nov 20 15:24:21 2002
> @@ -230,6 +230,8 @@ parse_opt (int opt, char *arg, s
Roland McGrath <[EMAIL PROTECTED]> writes:
> If you can implement the interface that takes many objects and ranges from
> each, please do that. The reason to want this is so that we can use it to
> compose proxy memory objects that span adjacent memory objects of logically
> contiguous underlying
> The nice thing here is that with this interface, we can fix it to use
> the *right* (but much harder) implementation at any later date.
Indeed. That's why I don't want vm_map to fail even when it can
immediately tell that all access will fail, for example.
If you can implement the interface that takes many objects and ranges from
each, please do that. The reason to want this is so that we can use it to
compose proxy memory objects that span adjacent memory objects of logically
contiguous underlying objects like files and disks. That lets us use the
> here is my patch for proxy memory objects
Yowza! Does it actually work?
> There is a new file in vm/memory_object_proxy.c and a new interface in
> mach4.defs. I have added a couple of skip's in the !MACH_PCSAMPLE case,
> because I think reusing msgid's is bad mojo, but I am not sure if
> MACH
The implementation of memory_object_create_proxy should allow a proxy
object to be passed as the value of OBJECT, and dereference it, and
then make a proxy object with the minimum set of the two.
Note that if you add the range-restriction to this, the result will be
that proxy objects need to spe
Marcus Brinkmann <[EMAIL PROTECTED]> writes:
> There is a new file in vm/memory_object_proxy.c and a new interface in
> mach4.defs. I have added a couple of skip's in the !MACH_PCSAMPLE case,
> because I think reusing msgid's is bad mojo, but I am not sure if
> MACH_PCSAMPLE is something to worry
Roland McGrath <[EMAIL PROTECTED]> writes:
> > Yeah, I was thinking of that. In fact, the reason I became aware of that is
> > that I wanted to use another pseudo device with no ops for the proxy memory
> > objects (but for those I need a close function, so I made my own device ops
> > structure)
> Roland's earlier change hasn't made it to the server for some reason,
I don't know what this means.
> 2002-11-20 David Walter <[EMAIL PROTECTED]>
>
> * hurdio.c (hurdio_init): added hurdio_assert_dtr () call to force
> the open of a device during a change of arg/type.
It should
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Following up on the hangup when using fsysopts /dev/console, the
following change fixes the problem.
This immediately opens the new argument the term translator is pointed
to (eg. /dev/vcs/10/console or your preference)
Roland's earlier change hasn'
On Wed, Nov 20, 2002 at 11:22:57PM +0100, M. Gerards wrote:
> Hi,
>
> Here is a small plugin I wrote for the console client using the generic-speaker
> sourcecode and some hints I found in driver.h. It is a quick hack, but it works :).
>
> To use it just supply the parameter "-d visual_bell" to t
Hi,
Here is a small plugin I wrote for the console client using the generic-speaker
sourcecode and some hints I found in driver.h. It is a quick hack, but it works :).
To use it just supply the parameter "-d visual_bell" to the console client. It
should work with all output drivers.
Hi,
here is my patch for proxy memory objects, with a new kernel type, and with
the behaviour I assumed to be the correct one in my last mail (changing that
if I was wrong is trivial, the work is in all the infrastructure of setting
up a new kernel object and adding the interface).
There is a new
Du bureau de Didier Bonneville-Roussy
St-Barthélemy
Cher ami et entrepreneur,
Ceci est le premier numéro de la Lettre du marketing géométrique. Le coût de
l'abonnement pour un an est de 197$.
Par contre, j'ai décidé de vous envoyer cette première lettre gratuitement et de vous
offrir un abonn
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Marcus Brinkmann <[EMAIL PROTECTED]> writes:
> (btw, [EMAIL PROTECTED] is not one of my addresses)
Apologies, don't know what that was from %-|
> On Wed, Nov 20, 2002 at 09:23:37AM -0500, David Walter wrote:
>> I've been noticing that if the initial
Hi Neal,
I think you need a change similar to this for libthreads in libpthread:
2002-08-26 Roland McGrath <[EMAIL PROTECTED]>
* cthreads.c (cthread_body) [HAVE_USELOCALE]: Call uselocale to
initialize the thread to the global locale.
The failure is that functions like isprint
FYI, I've just run make check on glibc for the Hurd (2.3.1 with various
Debian patches)
I'm pleased that all of the locales stuff appears to pass (I last ran
make check when we still used stdio)
This post is mostly informational and in case some brave hacker feels
like figuring them out over wint
Warning: l0c is not a valid urgency; it will be treated as low by testing.
Accepted:
hurd-dev_20021118-1_hurd-i386.deb
to pool/main/h/hurd/hurd-dev_20021118-1_hurd-i386.deb
hurd_20021118-1.diff.gz
to pool/main/h/hurd/hurd_20021118-1.diff.gz
hurd_20021118-1.dsc
to pool/main/h/hurd/hurd_200211
(btw, [EMAIL PROTECTED] is not one of my addresses)
On Wed, Nov 20, 2002 at 09:23:37AM -0500, David Walter wrote:
> I've been noticing that if the initial term process running on console
> isn't killed console output doesn't change to the new device.
>
> If process 7 is killed (/hurd/term /dev/co
It turns out that there is another issue (at least one).
I've been wondering about this in a different context, and hopefully
the following makes sense.
In hurd/init/init.c the hardware device is opened.
|| device_open (device_master, D_WRITE, "console", &consdev))
But when multiuser mode
22 matches
Mail list logo
