[spi-devel-general] Re: [PATCH/RFC] SPI: add DMAUNSAFE analog to David Brownell's core

[Vitaly W. <vw...@ru...>]

David Brownell wrote:

>On Tuesday 13 December 2005 2:15 pm, Vitaly Wool wrote:
>
>  
>
>>Take for instance spi_w8r8 function used in lotsa places in the drivers 
>>you and Stephen have posted.
>>This function has a) *implicit memcpy* inherited from 
>>spi_write_then_read b) *implicit priority inversion* inherited from the 
>>same place.
>>    
>>
>
>No, (a) is explicit, along with comments not to use that family of
>calls when such things matter more than the convenience of those
>RPC-ish calls.  And (b) was fixed a small patch, now merged.
>  
>
Hmm. Why not use a) when it's convenient? You're placing an artificial 
requirement then.
Yep, saw patch for b). One more kmalloc introduces which can be avoided 
easily with my patch ;)

Vitaly