[OpenWrt-Devel] [PATCH v2 00/18] atheros: I/O cleanups

[Felix Fietkau]

On 2014-07-15 15:53, Jonas Gorski wrote:
> On Tue, Jul 15, 2014 at 1:57 AM, Sergey Ryazanov <ryazanov.s.a at gmail.com> wrote:
>> Main goals of this series:
>>  * Simplify interface between arch code and SoC drivers
>>  * Simplify internal realization of arch code
>>  * Make code consistent with mainstream kernel rules and practice
>>
>> This series extensively tested with FON2202 (ar2315 based) and D-Link
>> DWL-2100AP (ar2313 based), but tests are not done with AR2317/8 and
>> AR5312, since I have no access to such boards now.
>>
>> Changes since v1:
>>  * Add missed patch 'atheros[ar2315-wdt]: update initialization'
>>
>> Sergey Ryazanov (18):
>>   atheros[ar2315-wdt]: update initialization
>>   atheros[ar2315-wdt]: rename config symbol
>>   atheros: use correct address space and pointer type for register
>>     access
>>   atheros[ar2315-wdt]: update interrupt handling
>>   atheros[ar2315-wdt]: update I/O handling
>>   atheros[ar231x-eth]: update MAC and PHY reset method
>>   atheros[ar231x-eth]: move driver to atheros subdirectory
>>   atheros: pass UART IRQ number via function argument
>>   atheros: move AR2315 misc IRQ dispatching to separate function
>>   atheros: rename some interrupt control handlers
>>   atheros: simplify AR2315 misc IRQ (un)masking
>>   atheros: use irq_set_chained_handler()
>>   atheros: simplify gpiolib realization
>>   atheros[ar231x-eth]: pass phys address of I/O memory via platform res
>>   atheros[ar231x-eth]: pass PHY I/O memory via device resources
>>   atheros[uart]: use 32-bit aligned I/O
>>   atheros[uart]: pass only physical I/O mem address to 8250 driver
>>   atheros: update macroses names
> 
> as a nitpick: for the subject, instead of "atheros[foo]:" it would be
> better to use "atheros: foo:" for the subject; this is the canonical
> version of providing subsystem tags. Git tends to strip everything in
> brackets, so these can easily get lost.
In this case, it worked. git apparently only strips stuff if the bracket
part is surrounded by whitespace.
But yes, for future submissions it's probably better to avoid that.

- Felix
_______________________________________________
openwrt-devel mailing list
openwrt-devel at lists.openwrt.org
https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel