news for uml_hcd driver ?
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 12 lines · you
Hello
I'm a french student, (engineer school). I work on a very interessant project. We want to develop a USB kernel driver for a joystick. In order to test our driver, we'd want to run it in user space (with UML andUML_HCD), doing this way we are sure not to crash the host kernel... We success to compile a 2.4.18 kernel with your patch, but theuml_hcdmodule don't seem to work (we don't see anything in/proc/bus/usb, when usbdevfs in mounted).
Is there any developement on this driver today?
Is there a way to port this driver for the 2.6 kernel ?
Does it exist a way to run a device driver in user space (emulation ?)
Quoted 5 lines
Thank you in advance
your sincerellyHubert FONGARNAND
CPE Student
It should still work what was your host kernel (the one you were running UML on top of)
I think that 2.4.19-39um as a guest kernel also worked but about 2.4.20 Greg Kroh back-ported the 2.5/2.6 USB structures to 2.4 and broke the umusb-2.4.19-39 patch. Later 2.4.19-XXum versions should also work but are untested.
From the web page (which seems to be down?)
Operation of the user mode linux host controller requires that the host system has /proc/bus/usb mounted and writable by the user running the user mode linux kernel.
Usually when I had problems it was that user mcmechan could not write to the /proc/bus/usb files or that the host kernel driver had found a driver for the device on the host kernel which caused the host USB driver to run not the UML USB driver
As to porting the uml-hcd to 2.6 it should be possible but there has not been that much interest in running USB from UML though it seems to be picking up again.
As I recall it will enumerate the devices on the host but has a problem creating the /proc/bus/usb files in the UML
The proper solution was to setup a new style HCD in the 2.6 tree but UML is rather different than USB expected e.g. no DMA and at the time creating the absent DMA infrastructure seemed like much work to hit a moving target.
As to emulation:
UML is my preferred way to work on USB or I would have done it differently. VMWARE might have a way to run USB also but I don't remember seeing any last time I looked at the web pages, and it would not help nearly as much for debugging.
-------------------------------------------------------
The SF.Net email is sponsored by: Beat the post-holiday blues Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek. It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt
_______________________________________________
User-mode-linux-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 2 lines · Hubert FONGARNAND
Hello!!!
The Umusb driver works now on the 2.4.27 kernel... I can use now, my
force
Quoted 7 lines
feedback joystick and my usbmouse on the 2.4.27-1um kernel... But I'm in very beta testing phase and I experience many "kernel panic : Kernel mode fault"
I try to run : ./linux debug
A window with gdb appear, but the kernel looks like stopped... in there another way to debug the kernel?
Your's sincerely
It did not start up when you entered "c" to continue after setting your breakpoints?
It would appear that you are running in TT mode "tracing thread" not SKAS mode.
Have you looked at http://user-mode-linux.sf.net/debugging.html the debugging tutorial? or checked on the mailing lists?
Quoted 6 lines
Quoted 6 lines · Hubert FONGARNAND
Quoted 4 lines
Quoted 3 lines · James McMechan
Quoted 2 lines · you
Hello
I'm a french student, (engineer school). I work on a very
interessant
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 1 line
Quoted 1 line
project. We want to develop a USB kernel driver for a joystick. In
order
Quoted 3 lines
Quoted 3 lines
to test our driver, we'd want to run it in user space (with UML and
UML_HCD), doing this way we are sure not to crash the host kernel... We success to compile a 2.4.18 kernel with your patch, but the
uml_hcd
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 1 line
Quoted 1 line
module don't seem to work (we don't see anything in
/proc/bus/usb,when
Quoted 4 lines
Quoted 4 lines
usbdevfs in mounted).
Is there any developement on this driver today? Is there a way to port this driver for the 2.6 kernel ? Does it exist a way to run a device driver in user space
(emulation ?)
Quoted 11 lines
Quoted 11 lines
Quoted 11 lines
Quoted 8 lines
Quoted 6 lines
Thank you in advance
your sincerellyHubert FONGARNAND
CPE StudentIt should still work what was your host kernel (the one you were
running
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
UML on top of)
I think that 2.4.19-39um as a guest kernel also worked but about
2.4.20
Quoted 1 line
Greg Kroh back-ported the 2.5/2.6 USB structures to 2.4 and broke
the
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
umusb-2.4.19-39 patch. Later 2.4.19-XXum versions should also work
but
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 4 lines
are untested.
From the web page (which seems to be down?)
Operation of the user mode linux host controller requires that thehost
Quoted 1 line
system has
/proc/bus/usbmounted and writable by the user running
the
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
user mode linux kernel.
Usually when I had problems it was that user mcmechan could not
write
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
to
Quoted 1 line
the
/proc/bus/usbfiles or that the host kernel driver had found adriver
Quoted 1 line
for the device on the host kernel which caused the host USB driver
to
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
run
Quoted 3 lines
not the UML USB driver
As to porting the uml-hcd to 2.6 it should be possible but there has
not
Quoted 1 line
been that much interest in running USB from UML though it seems to
be
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 3 lines
picking up again.
As I recall it will enumerate the devices on the host but has a
problem
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 2 lines
creating the
/proc/bus/usbfiles in the UML
The proper solution was to setup a new style HCD in the 2.6 tree butUML
Quoted 2 lines
is rather different than USB expected e.g. no DMA and at the time creating the absent DMA infrastructure seemed like much work to hit
a
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 5 lines
moving target.
As to emulation:
UML is my preferred way to work on USB or I would have done it differently. VMWARE might have a way to run USB also but I don'tremember
Quoted 1 line
seeing any last time I looked at the web pages, and it would not
help
Quoted 22 lines
Quoted 22 lines
Quoted 3 lines
Quoted 1 line
nearly as much for debugging.
I use your umusb driver with some success on the 2.4.19 kernel!
I am glad to see a report of your success
Quoted 3 lines
I've done some modifications for interrupt transfer. now I can use my usb mouse very well. usb-storage driver works well too!
Good, I had not thought to try with a USB mouse, I think I tried USB storage, a smart card reader, and a USB Ethernet adapter
Quoted 2 lines
I've modified the 2.4.19 um kernel in order it could run on a 2.6 host kernel... and it works too.
Ah, very good... you are I hope contemplating sharing your fixes, I have CC'ed the [email protected]
Quoted 2 lines
I've problem with unloading usb modules, sometimes i have some "Kernel Mode Fault at ip 0x...." how could I debug that?
Quoted 8 lines
Quoted 8 lines
That is where UML shines, you can use the debugger on a UML kernel and put a breakpoint at the panic/Oops and then do back-traces/examine symbols
The UML sites even include examples of getting the symbols from modules All of which makes tracing down "Kernel Mode Fault..." much easier I usually have those sort of errors happen when I forget to unregister part of the module and then one of the kernel subsystems is still
trying
Quoted 3 lines
Quoted 3 lines
to use the module after the rmmod.
Quoted 1 line
We are looking to use the driver on a 2.4.27 kernel, but we
experienced
Quoted 9 lines
Quoted 9 lines
Quoted 5 lines
some problem with dma. We are making a fake dma support (we use your
um_pcidriver for the 2.5 kernel)... I hope it will work! I will send you our patches if it works!Your's sincerely
It is a pleasure to see that you are finding it useful
James McMechan
-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting Tool for open source databases. Create drag-&-drop reports. Save time by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc. Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________
User-mode-linux-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel