[ofa-general] Re: libmlx4 wc flash

[Michael S. Tsirkin]

> Quoting Roland Dreier <rdreier at cisco.com>:
> Subject: Re: libmlx4 wc flash
> 
> >  >  		memcpy(ctx->bf_page + ctx->bf_offset, ctrl, align(size * 16, 64));
> 
> By the way, if we have an SQ with 32-byte WQEs, and we do blueflame
> from a WQE at the end of the buffer, we might end up reading off the
> end of the buffer.  Not very likely, I guess.

Hmm, is 32-byte SQ wqe shift actually possible? Which parameters give this?

> I wonder if memset(,0,) for the remaining bytes might be faster anyway?

In some early testing it seemed much slower. Try it :)

-- 
MST