| Joe Perches | 510fa408 | 2014-02-10 16:37:56 -0800 | [diff] [blame] | 1 | TODO: |
| 2 | - move firmware loading to request_firmware() |
| 3 | - remove direct memory access of structures |
| 4 | - any remaining sparse and checkpatch.pl warnings |
| 5 | |
| 6 | - use net_device_ops |
| 7 | - use dev->stats rather than adapter->stats |
| 8 | - don't cast netdev_priv it is already void |
| 9 | - GET RID OF MACROS |
| 10 | - work on all architectures |
| 11 | - without CONFIG_X86_64 confusion |
| 12 | - do 64 bit correctly |
| 13 | - don't depend on order of union |
| 14 | - get rid of ASSERT(), use BUG() instead but only where necessary |
| 15 | looks like most aren't really useful |
| 16 | - no new SIOCDEVPRIVATE ioctl allowed |
| 17 | - don't use module_param for configuring interrupt mitigation |
| 18 | use ethtool instead |
| 19 | - reorder code to elminate use of forward declarations |
| 20 | - don't keep private linked list of drivers. |
| Joe Perches | 510fa408 | 2014-02-10 16:37:56 -0800 | [diff] [blame] | 21 | - use PCI_DEVICE() |
| 22 | - do ethtool correctly using ethtool_ops |
| 23 | - NAPI? |
| 24 | - wasted overhead of extra stats |
| 25 | - state variables for things that are |
| 26 | easily available and shouldn't be kept in card structure, cardnum, ... |
| 27 | slotnumber, events, ... |
| 28 | - get rid of slic_spinlock wrapper |
| 29 | - volatile == bad design => bad code |
| 30 | - locking too fine grained, not designed just throw more locks |
| 31 | at problem |
| 32 | |
| 33 | Please send patches to: |
| 34 | Greg Kroah-Hartman <gregkh@linuxfoundation.org> |
| 35 | and Cc: Lior Dotan <liodot@gmail.com> and Christopher Harrer |
| 36 | <charrer@alacritech.com> as well as they are also able to test out any |
| 37 | changes. |