Hi!
This is the 19980830 release of the UUSBD, the Linux USB
Driver stack.
News:
- I have rewritten the UHCI scheduler to use a linear list
for control and bulk transfers, also coded interrupt
transfers. All work and are tested except bulk, which is
untested.
- UHCI bandwidth reclamation slows down my system, so I've
added a #define at uhci.h to disable it completely. If you
feel like trying, enable the #define. If you see your
processes CPU% share raising, that's the problem [please
mail me your /proc/pci or lspci].
- The problem of zombie khubdd is gone with a work-around, but
the _real_problem_ is still there. Now you can have the
modules load as long as you want. I don't know how to make
those zombies really die without waiting for the pid
[waitpid()] to finish ... any kernel guru out there?
- There's still an small problem with simultaneous
drivers. I'm trying to hunt it down, but it is harmless. It
will also leave some memory allocations unfreed.
- I've added a retry count on ports, so when an attachment
fails, the reset logic can use it to increase the reset
times. It counts up to ten, then sets back to one and so
on.
- I've started coding the HID driver and there is basic
keyboard support using the Boot Protocol. The matrix I have
seems to be broken, so I cannot verify it really works. Use
the program `uusbd/hidd/bkbd-test' to read from a char
device (major 60, minor 0-8) if you have a keyboard and the
logs report it supports the boot protocol. It does not much,
but enough to see something.
- My Longshine hub doesn't work yet ...
I'll keep on coding, but meanwhile, I'd like everyone who has
USB stuff to insmod all these modules (usb.o, uhci/usb-uhci.o,
audd/usb-audd, hidd/usb-hidd.o and hubd/usb-hubd.o) and send me the
logs and /proc/pci (be warned it may crash! :).
Have a nice time!
--Linux-USB! http://peloncho.fis.ucm.es/~inaky/USB.html - - Inaky Perez Gonzalez -- PGP pubkey fingerprint - inaky@peloncho.fis.ucm.es -- 8E 34 3A 62 64 99 E2 44 - http://peloncho.fis.ucm.es/~inaky -- AD 7B 30 D9 DD FF 3E 4C - --------------------------------- -- ----------------------- - The loneliness of the long distance runner .....
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html