diff options
author | wm4 <wm4@nowhere> | 2015-02-20 22:24:37 +0100 |
---|---|---|
committer | Diogo Franco (Kovensky) <diogomfranco@gmail.com> | 2015-02-21 21:00:50 +0900 |
commit | 192318a216d63d3163d551a3e3e493156ae60f3d (patch) | |
tree | 51767731cf1c8b09f7759139b1d2726cdca95f56 | |
parent | 822e6e0dd14ed0ab00c8480bdaacd7d3a52da9f3 (diff) | |
download | mpv-192318a216d63d3163d551a3e3e493156ae60f3d.tar.bz2 mpv-192318a216d63d3163d551a3e3e493156ae60f3d.tar.xz |
video: un-discourage "vaapi-copy" hwdec mode
Maybe I don't know what I'm doing. I'm fairly certain though that Intel
does not know what they're doing.
(cherry picked from commit d71bbcbc98c1fa8a934ee656b13adda089a11681)
-rw-r--r-- | DOCS/man/options.rst | 3 | ||||
-rw-r--r-- | video/decode/vaapi.c | 5 |
2 files changed, 2 insertions, 6 deletions
diff --git a/DOCS/man/options.rst b/DOCS/man/options.rst index 6409cd9b0e..4aa93e098a 100644 --- a/DOCS/man/options.rst +++ b/DOCS/man/options.rst @@ -551,7 +551,8 @@ Video The ``vaapi-copy`` mode allows you to use vaapi with any VO. Because this copies the decoded video back to system RAM, it's likely less efficient - than the ``vaapi`` mode. + than the ``vaapi`` mode. But there are reports that this is actually faster + as well, and avoids many issues with ``vaapi``. .. note:: diff --git a/video/decode/vaapi.c b/video/decode/vaapi.c index 82de0a667e..f78a01d86e 100644 --- a/video/decode/vaapi.c +++ b/video/decode/vaapi.c @@ -69,7 +69,6 @@ struct priv { int rt_format; struct mp_image_pool *sw_pool; - bool printed_readback_warning; }; #define PE(av_codec_id, ff_profile, vdp_profile) \ @@ -450,10 +449,6 @@ static struct mp_image *copy_image(struct lavc_ctx *ctx, struct mp_image *img) struct mp_image *simg = va_surface_download(img, p->sw_pool); if (simg) { - if (!p->printed_readback_warning) { - MP_WARN(p, "Using GPU readback. This is usually inefficient.\n"); - p->printed_readback_warning = true; - } talloc_free(img); return simg; } |