Re: [PATCH v3 16/28] tcg: Replace region.end with region.total_size

[Alex Bennée]

Richard Henderson <richard.henderson@linaro.org> writes:

> A size is easier to work with than an end point,
> particularly during initial buffer allocation.
>
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> ---
>  tcg/region.c | 29 +++++++++++++++++------------
>  1 file changed, 17 insertions(+), 12 deletions(-)
>
> diff --git a/tcg/region.c b/tcg/region.c
> index 9a1e039824..a17f342f38 100644
> --- a/tcg/region.c
> +++ b/tcg/region.c
> @@ -48,7 +48,7 @@ struct tcg_region_state {
>      /* fields set at init time */
>      void *start;
>      void *start_aligned;
> -    void *end;
> +    size_t total_size; /* size of entire buffer */
>      size_t n;
>      size_t size; /* size of one region */
>      size_t stride; /* .size + guard size */

I'd shuffle that to the end so it scans size < stride < total_size. I
see we have special handling bellow:

> @@ -279,7 +279,7 @@ static void tcg_region_bounds(size_t curr_region, void 
> **pstart, void **pend)
>          start = region.start;
>      }
>      if (curr_region == region.n - 1) {
> -        end = region.end;
> +        end = region.start_aligned + region.total_size;

So why isn't this end = start_aligned + (n * stride)? do we not line up
for the last region?

Anyway:

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>

-- 
Alex Bennée