[openib-general] Re: [PATCH] set eq->nent earlier in mthca_create_eq

[Michael S. Tsirkin]

Quoting r. Roland Dreier <rolandd at cisco.com>:
> Subject: Re: [PATCH] set eq->nent earlier in mthca_create_eq
> 
> Thanks, good catch.
> 
> How about if we get rid of any confusion by just using eq->nent all
> the time?  And since we're touching the code anyway, we might as well
> us roundup_pow_of_two()...
> 
> How's this seem to you?
> 
> --- infiniband/hw/mthca/mthca_eq.c	(revision 3432)
> +++ infiniband/hw/mthca/mthca_eq.c	(working copy)
> @@ -476,12 +476,8 @@ static int __devinit mthca_create_eq(str
>  	int i;
>  	u8 status;
>  
> -	/* Make sure EQ size is aligned to a power of 2 size. */
> -	for (i = 1; i < nent; i <<= 1)
> -		; /* nothing */
> -	nent = i;
> -
> -	eq->dev = dev;
> +	eq->dev  = dev;
> +	eq->nent = roundup_pow_of_two(nent);

Looks good.
One nit: the min. eq size is 2, and roundup_pow_of_two cant handle 0
parameter. Of course create_eq is an internal function, so we can either
rely on all callers to pass at least 2 as a parameter, or write
 
	roundup_pow_of_two(max(nent, 2))

I suggest the later as a form of documentation.
Thanks,

-- 
MST