[Pixman] Cleaning patchwork

[Oded Gabbay]

Hi Ben,

I would like to clean pixman's patchwork and you have about 20
outstanding patches.

Could you please take a look and tell me which ones are still relevant
and which ones are not ?

You can take a look at http://patchwork.freedesktop.org/project/pixman/patches/
I also copied here there names:

[4/4] pixman-fast-path: Make bilinear cover fetcher use COVER_CLIP_TIGHT flag
[3/4] armv7/mips/sse2: Make bilinear cover fast paths use COVER_CLIP_TIGHT flag
[2/4] Introduce FAST_PATH_SAMPLES_COVER_CLIP_TIGHT_BILINEAR flag
Resolve implementation-defined behaviour for division rounded to -infinity
[24/37,v2] armv6: Add optimised scanline fetcher for a1r5g5b5
[23/37,v2] armv6: Add optimised scanline fetchers and writeback for
r5g6b5 and a8
Require destination images to be bitmaps
[2/5,v2] armv7: Add in_8888_8 fast path
[11/37,v2] armv6: Add in_8888_8 fast path
[10/37,v2] pixman-fast-path: Add in_8888_8 fast path
[05/37,v2] composite flags: Early detection of fully opaque 1x1
repeating source images
[5/5] armv7: Add src_1555_8888 fast path
[4/5] armv7: Add in_reverse_8888_8888 fast path
[3/5] armv7: Add in_n_8888 fast path
[1/5] armv7: Use prefetch for small-width images too
armv7: Re-use existing fast paths in more cases
armv7: Re-use existing fast paths in more cases
[25/37,v2] armv6: Add src_1555_8888 fast path
[3/3] armv7: Use VLD-to-all-lanes
[2/3] armv7: Faster fill operations
[1/3] armv7: Coalesce scalar accesses where possible

I also suggest that for the relevant ones (if there are any), you
would rebase them on the latest master and re-send them as a single
series in order to restart the review process.

Thanks,

Oded