1This driver is supposed to support: 2 3 Mojave cards (single port PCI Gigabit) both copper and fiber 4 Oasis cards (single and dual port PCI-x Gigabit) copper and fiber 5 Kalahari cards (dual and quad port PCI-e Gigabit) copper and fiber 6 7The driver was actually tested on Oasis and Kalahari cards. 8 9TODO: 10 - move firmware loading to request_firmware() 11 - remove direct memory access of structures 12 - any remaining sparse and checkpatch.pl warnings 13 14 - use net_device_ops 15 - use dev->stats rather than adapter->stats 16 - don't cast netdev_priv it is already void 17 - use compare_ether_addr 18 - GET RID OF MACROS 19 - work on all architectures 20 - without CONFIG_X86_64 confusion 21 - do 64 bit correctly 22 - don't depend on order of union 23 - get rid of ASSERT(), use BUG() instead but only where necessary 24 looks like most aren't really useful 25 - no new SIOCDEVPRIVATE ioctl allowed 26 - don't use module_param for configuring interrupt mitigation 27 use ethtool instead 28 - reorder code to elminate use of forward declarations 29 - don't keep private linked list of drivers. 30 - remove all the gratiutous debug infrastructure 31 - use PCI_DEVICE() 32 - do ethtool correctly using ethtool_ops 33 - NAPI? 34 - wasted overhead of extra stats 35 - state variables for things that are 36 easily available and shouldn't be kept in card structure, cardnum, ... 37 slotnumber, events, ... 38 - get rid of slic_spinlock wrapper 39 - volatile == bad design => bad code 40 - locking too fine grained, not designed just throw more locks 41 at problem 42 43 44Please send patches to: 45 Greg Kroah-Hartman <gregkh@linuxfoundation.org> 46and Cc: Lior Dotan <liodot@gmail.com> and Christopher Harrer 47<charrer@alacritech.com> as well as they are also able to test out any 48changes. 49