[PATCH v3] ath10k: save firmware debug log messages.

[Kalle Valo]

greearb at candelatech.com writes:

> From: Ben Greear <greearb at candelatech.com>
>
> They may be dumped through the firmware dump debugfs
> file.
>
> Signed-off-by: Ben Greear <greearb at candelatech.com>
> ---
>
> v3:  Re-worked byte-order related stuff again,
>    thanks to Michael's comments on IRC.
>    Fix problem reported by smatch, possible NPE.
>
>
> v2:  Remove some #ifdefs
>      rebase:  Fix logging, deal with byte-order differences
>        when reading firmware memory.
>
> Tested on x86-64.  smatch is clean, but it seems it cannot detect
> le v/s cpu endian issues on x86-64, so I could have easily missed
> something.

Yeah, sparse warns about endian stuff. But let me do a proper review
before sending a new version.

drivers/net/wireless/ath/ath10k/debug.c:830:34: warning: incorrect type in assignment (different base types)
drivers/net/wireless/ath/ath10k/debug.c:830:34:    expected unsigned int [unsigned] [usertype] next_idx
drivers/net/wireless/ath/ath10k/debug.c:830:34:    got restricted __le32 [usertype] <noident>
drivers/net/wireless/ath/ath10k/debug.c:831:35: warning: incorrect type in assignment (different base types)
drivers/net/wireless/ath/ath10k/debug.c:831:35:    expected unsigned int [unsigned] [usertype] first_idx
drivers/net/wireless/ath/ath10k/debug.c:831:35:    got restricted __le32 [usertype] <noident>

-- 
Kalle Valo