The 'fe-lock' branch of
git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/netdev-2.6.git fe-lock
contains the following changes that I would like to get tested:
[netdrvr] forcedeth: make NAPI unconditional
[netdrvr] forcedeth: interrupt handling cleanup
[netdrvr] forcedeth: process TX completions using NAPI
[netdrvr] forcedeth: internal simplification and cleanups
[netdrvr] forcedeth: timer overhaul
These are intended for feedback and testing, NOT for merging.
The goals of these changes are:
* move the driver towards a more sane, simple, easy to verify locking
setup -- irq handler would often acquire/release the lock twice
for each interrupt -- and hopefully
* to eliminate a rarely used, apparently fragile locking scheme that
includes heavy use of disable_irq(). this tool is most often employed
during NIC reset/reconfiguration, so satisfying this goal implies
changing the way NIC reset and config are accomplished.
Miscellaneous notes:
* using the new napi_struct stuff in net-2.6.24, the TX completion
process has been moved to a -separate- NAPI polling channel. thus
there are now two napi_structs, one for RX and one for TX, independent
of each other.
* I feel TX NAPI is a useful tool, because it provides an independent TX
process control point and system load feedback point.
Thus I felt this was slightly superior to tasklets.
* But who knows if this is a good idea? :) I am interested in
feedback and criticism on this issue.
-
