| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
In one spot (usbatm_cancel_send) we were calling dev_kfree_skb with irqs
disabled. This mistake is just too easy to make, so systematically use
dev_kfree_skb_any rather than dev_kfree_skb.
Signed-off-by: Duncan Sands <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
We weren't always returning -ENOMEM.
Signed-off-by: Duncan Sands <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
This patch causes vcc_release_async to be applied to any open
vcc's when the modem is disconnected. This signals a socket
shutdown, letting the socket user know that the game is up.
I wrote this patch because of reports that pppd would keep
connections open forever when the modem is disconnected.
This patch does not fix that problem, but it's a step in the
right direction. It doesn't help because the pppoatm module
doesn't yet monitor state changes on the ATM socket, so simply
never realises that the ATM connection has gone down (meaning
it doesn't tell the ppp layer). But at least there is a socket
state change now. Unfortunately this patch may create problems
for those rare users like me who use routed IP or some other
non-ppp connection method that goes via the ATM ARP daemon: the
daemon is buggy, and with this patch will crash when the modem
is disconnected. Users with a buggy atmarpd can simply restart
it after disconnecting the modem.
Signed-off-by: Duncan Sands <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
Convert kmalloc + memset to kzalloc.
Signed-off-by: Duncan Sands <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
Have minidrivers and the core signal special requirements
using a flags field in struct usbatm_data. For the moment
this is only used to replace the need_heavy_init bind
parameter, but there'll be new flags in later patches.
Signed-off-by: Duncan Sands <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
Formatting, changes to variable names, comments, log level changes,
printk rate limiting.
Signed-off-by: Duncan Sands <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
Remove the "inline" keyword from a bunch of big functions in the kernel with
the goal of shrinking it by 30kb to 40kb
Signed-off-by: Arjan van de Ven <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Acked-by: Jeff Garzik <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
This patch contains the following cleanups:
- make needlessly global functions static
- every file should #include the headers containing the prototypes for
it's global functions
Signed-off-by: Adrian Bunk <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
atm_dev_deregister() removes device from atm_dev list immediately to
prevent operations on a phantom device. Decision to free device based
only on ->refcnt now. Remove shutdown_atm_dev() use atm_dev_deregister()
instead. atm_dev_deregister() also asynchronously releases all vccs
related to device.
Signed-off-by: Stanislaw Gruszka <[email protected]>
Signed-off-by: Chas Williams <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This patch converts kcalloc(1, ...) calls to use the new kzalloc() function.
Signed-off-by: Pekka Enberg <[email protected]>
Cc: Greg KH <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
you seem to have applied the original, not the new improved one with
whiter teeth that uses kcalloc instead of kmalloc + memset. Here's a
patch that goes on top of the one you applied.
Signed-off-by: Duncan Sands <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
Zero the entire instance, not just the struct usbatm_data head.
Make sure the just allocated urb is freed if we fail to allocate
a buffer. Based on a patch by Stanislaw W. Gruszka.
Signed-off-by: Duncan Sands <[email protected]>
Acked-by: Pete Zaitcev <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
Reduce the number of "unknown vpi/vci" debug messages to (usually) at most
one per-urb, rather than one per-cell. This is only an issue when (a) many
packets come in but no connection is open; and (b) CONFIG_USB_DEBUG is set.
Signed-off-by: Duncan Sands <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
Rework the core usbatm code: minidrivers (i.e. drivers for particular
modems) now register themselves with the usbatm core, supplying methods
for binding/unbinding etc. The design was inspired by usb-serial and
usbnet. At the same time, more common code from the speedtch and
cxacru (patch 3/5) drivers was generalized and moved into the core. The
transmission and reception parts have been unified and simplified. Since
this is a major change and I don't like underscores in file names,
usb_atm.[ch] has been renamed usbatm.[ch].
Many thanks to Roman Kagan, who did a lot of the coding.
Signed-off-by: Duncan Sands <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|