Janne Grunau <[email protected]> writes: > On 2012-02-01 11:55:59 +0000, Måns Rullgård wrote: >> "Ronald S. Bultje" <[email protected]> writes: >> > 2012/2/1 Måns Rullgård <[email protected]>: >> >> >> >> This patch should not be accepted without a better understanding of >> >> what problem, if any, it is intended to solve. >> > >> > A quick look at fate suggests that this tries to solve the issue of >> > av_realloc() not giving aligned memory, and thus using this (which >> > only needs a memcpy() if a new allocation was done) would fix some of >> > the failures we see after bswap32 was SSE'ified. >> >> Setting memory to zero only to immediately memcpy on top of it strikes >> me as stupid. > > it doesn't memcpy(), the content preserving property seems not to be > used. At least on first glance at the patches making use of > av_fast_padded_malloc(). I'll post a patch with the av_mallocz() > removed.
For the purposes of this discussion, bswap_buf() shall be considered equivalent to memcpy(). -- Måns Rullgård [email protected] _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
