2008/5/1 Jeff Dike <[EMAIL PROTECTED]>:
> On Tue, Apr 29, 2008 at 06:35:06PM +0200, Flavio wrote:
> > Any news?
>
> The current state of things as I understand them is that we have the
> eject ioctl being called in exactly the same way from eject and from
> UML and it behaves differently. This, of
On Tue, Apr 29, 2008 at 06:35:06PM +0200, Flavio wrote:
> Any news?
The current state of things as I understand them is that we have the
eject ioctl being called in exactly the same way from eject and from
UML and it behaves differently. This, of course, shouldn't happen, so
there must be somethi
2008/4/4 Flavio <[EMAIL PROTECTED]>:
>
> On 04/04/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> > On Wed, Apr 02, 2008 at 01:34:00PM +0200, Flavio wrote:
> > > I guess you mean the difference between eject on the host and eject on
> > > the UML guest. If so, this is the eject strace output on the
On 04/04/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> On Wed, Apr 02, 2008 at 01:34:00PM +0200, Flavio wrote:
> > I guess you mean the difference between eject on the host and eject on
> > the UML guest. If so, this is the eject strace output on the host:
> > ioctl(3, CDROMEJECT, 0x1)
On Wed, Apr 02, 2008 at 01:34:00PM +0200, Flavio wrote:
> I guess you mean the difference between eject on the host and eject on
> the UML guest. If so, this is the eject strace output on the host:
> ioctl(3, CDROMEJECT, 0x1) = 0
>
> This is the eject strace output in UML:
> ioctl(16
On 31/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> Now, if you straced eject at this point, you would see
>
> ioctl(, CDROMEJECT, 0x1) = 0
> ?
Yes, it should be. But it isn't.
>
> If so, what's the difference between eject and UML?
I guess you mean the difference between eje
On Mon, Mar 24, 2008 at 04:58:11PM +0100, Flavio wrote:
> This is the strace result after modifying the corresponding line as you want.
> ioctl(16, CDROMEJECT, 0x1) = -1 EIO (Input/output error)
Now, if you straced eject at this point, you would see
ioctl(, CDROMEJECT, 0x1)
On 24/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> Yup - I'm mystified as to why arg == 1 isn't happening with the UML.
> Can you change the "arg" to just a literal "1" and see what the strace
> looks like?
Hello Jeff, I'm mystified too.
This is the strace result after modifying the correspon
On Wed, Mar 19, 2008 at 09:39:13PM +0100, Flavio wrote:
> On 19/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> > On Tue, Mar 18, 2008 at 08:17:52PM +0100, Flavio wrote:
> > > Ok, I applied your last patch. I've done the test, and this is the
> > result:
> > > ioctl(16, CDROMEJECT, 0)
On 19/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> On Tue, Mar 18, 2008 at 08:17:52PM +0100, Flavio wrote:
> > Ok, I applied your last patch. I've done the test, and this is the result:
> > ioctl(16, CDROMEJECT, 0)= -1 EIO (Input/output error)
>
>
> I want that 0 to be a 1. Ar
On Tue, Mar 18, 2008 at 08:17:52PM +0100, Flavio wrote:
> Ok, I applied your last patch. I've done the test, and this is the result:
> ioctl(16, CDROMEJECT, 0)= -1 EIO (Input/output error)
I want that 0 to be a 1. Are you sure you applied the patch? The
relevant line in ubd_kern.
On 18/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> On Tue, Mar 18, 2008 at 06:53:06PM +0100, Flavio wrote:
> > No I didn't. Excuse me, but I've never done a patch back out before.
> > Could you tell me how to do it, please?
> > Note that, your last patch has been successfully applied.
>
>
> B
On Tue, Mar 18, 2008 at 06:53:06PM +0100, Flavio wrote:
> No I didn't. Excuse me, but I've never done a patch back out before.
> Could you tell me how to do it, please?
> Note that, your last patch has been successfully applied.
Backing out a patch is the same as applying it, except you add -R to
On 18/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> Did you back out the previous patch first? This is a replacement for
> it, not an update.
No I didn't. Excuse me, but I've never done a patch back out before.
Could you tell me how to do it, please?
Note that, your last patch has been success
On Tue, Mar 18, 2008 at 04:41:37PM +0100, Flavio wrote:
> I applied your last patch to my kernel tree doing nothing else before but...
>
> I had problems during compilation:
>
> arch/um/drivers/ubd_kern.c: In function 'path_requires_switch':
> arch/um/drivers/ubd_kern.c:1203: error: case label no
On 18/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> I just noticed something. I forgot to pass down the ioctl argument.
> See if the patch below makes any differece.
I applied your last patch to my kernel tree doing nothing else before but...
I had problems during compilation:
# make linux AR
On Tue, Mar 18, 2008 at 12:30:53PM +0100, Flavio wrote:
> >
> > As a sanity-check, you can do this again, look for the line above, and
> > make sure the file descriptor (16 in this case) refers to /dev/cdrom.
> > Look in /proc//fd for the same pid that you strace.
> I don't know why I don't have
On 18/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> On Mon, Mar 17, 2008 at 09:50:43PM +0100, Flavio wrote:
> > Here's the strace log file attached to this mail.
> > Note that it's a compressed text file (as you can see), but once it
> > has been uncompressed, it occupies about 75 MBytes.
>
>
(Forgot to mention - I can see you're running with skas4. Nice work,
and I trust it's working well for you...)
Jeff
--
Work email - jdike at linux dot intel dot com
-
This SF.net email is sponsored by: Mic
On Mon, Mar 17, 2008 at 09:50:43PM +0100, Flavio wrote:
> Here's the strace log file attached to this mail.
> Note that it's a compressed text file (as you can see), but once it
> has been uncompressed, it occupies about 75 MBytes.
This is the relevant bit:
ioctl(16, CDROMEJECT, 0)
On 17/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> On Mon, Mar 17, 2008 at 07:46:36PM +0100, Flavio wrote:
> > Thanks Jeff, here's the output:
> >
> > ~# strace -e ioctl eject /dev/ubd2
> > ioctl(3, CDROMEJECT, 0) = -1 EIO (Input/output error)
>
>
> Nope, that's strace inside
On Mon, Mar 17, 2008 at 07:46:36PM +0100, Flavio wrote:
> Thanks Jeff, here's the output:
>
> ~# strace -e ioctl eject /dev/ubd2
> ioctl(3, CDROMEJECT, 0) = -1 EIO (Input/output error)
Nope, that's strace inside UML. What I want is strace on UML on the
host. I want to see what's
On 17/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> On Fri, Mar 14, 2008 at 11:59:38PM +0100, Flavio wrote:
> > Let me give you the strace eject /dev/ubd2 again, it may helps you to
> > understand.
> > (I don't know if it's the same strace I posted before)
>
>
> It changed from
>
>
> > > > i
On Fri, Mar 14, 2008 at 11:59:38PM +0100, Flavio wrote:
> Let me give you the strace eject /dev/ubd2 again, it may helps you to
> understand.
> (I don't know if it's the same strace I posted before)
It changed from
> > > ioctl(3, CDROMEJECT, 0) = -1 EINVAL (Invalid argument)
to
On 14/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> On Thu, Mar 13, 2008 at 07:22:06PM +0100, Flavio wrote:
> > ioctl(3, CDROMEJECT, 0) = -1 EINVAL (Invalid argument)
>
>
> Try the patch below - compile tested only.
Thanks Jeff,
I've just applied that patch successfully.
Let me
On Thu, Mar 13, 2008 at 07:22:06PM +0100, Flavio wrote:
> ioctl(3, CDROMEJECT, 0) = -1 EINVAL (Invalid argument)
Try the patch below - compile tested only.
Jeff
--
Work email - jdike at linux dot intel dot com
Index: linux-2.6.22/arch/um/drivers/ubd_kern.c
On 13/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> > 2.1) If I change CD-ROM and I try to mount it, I have no problem. It's all
> OK.
> > 2.2) If I have a CD-ROM inside the reader (and also mounted) and I try
> > to do 'eject /dev/ubd2' this is the result:
> >
> >DebianUM:~# eject /dev
On Thu, Mar 13, 2008 at 12:14:12AM +0100, Flavio wrote:
> 1) If I run UML using the following command line
> linux ubd0=rootfs.debian.etch ubd1=swapfs.debian ubd2=/dev/cdrom
> eth0=tuntap,,,192.168.1.100 mem=512M
> and we have no disk inside the cd-rom reader, thus I get the following error:
>
>
On 12/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> On Wed, Mar 12, 2008 at 08:00:10PM +0100, Flavio wrote:
> > So, here's what I was trying to do: binding the /dev/cdrom with the
> > /dev/ubd2 virtual device, passing the ubd2=/dev/cdrom in the command
> > line kernel option (no cd-rom inserte
On Wed, Mar 12, 2008 at 08:00:10PM +0100, Flavio wrote:
> So, here's what I was trying to do: binding the /dev/cdrom with the
> /dev/ubd2 virtual device, passing the ubd2=/dev/cdrom in the command
> line kernel option (no cd-rom inserted up to here).
> Once UML has finished to boot I do 'eject /dev
On 12/03/2008, Jeff Dike <[EMAIL PROTECTED]> wrote:
> On Wed, Mar 12, 2008 at 05:00:31PM +0100, Flavio wrote:
> > I would like to ask you a simple question about cdrom and usb pen
> > drives utilisation.
> >
> > I would like to know if I must insert a cd-rom in the cd-rom reader
> > before uml
On Wed, Mar 12, 2008 at 05:00:31PM +0100, Flavio wrote:
> I would like to ask you a simple question about cdrom and usb pen
> drives utilisation.
>
> I would like to know if I must insert a cd-rom in the cd-rom reader
> before uml starts, in case I have to use the cd-rom under UML.
>
> If I put t
Hello all,
I would like to ask you a simple question about cdrom and usb pen
drives utilisation.
I would like to know if I must insert a cd-rom in the cd-rom reader
before uml starts, in case I have to use the cd-rom under UML.
If I put the cd in before uml starts, there's no problem to mount an
33 matches
Mail list logo
