Mark Einon <mark.einon@xxxxxxxxx> :
[...]
> I'd like to propose moving it out of the staging area, or at least
> getting some review comments regarding what work is still needed.
Is there a chance we could have a single file driver - may be with a .h for
the registers layout ? It should be less than 200kbytes, especially if
forward declarations go away.
rx_ring.fbr{0, 1} can probably share a common structure.
Use of kmem_cache seems a bit unusual.
Use dma_alloc_... in place of pci_alloc_...
It's too late stopping the tx queue when there is no room for the current
packet. The condition should be detected for the next packet.
PCI_VDEVICE would not hurt.
My 0.02 € (whatever the CDS spread).
--
Ueimor
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
