On 2016-01-18 15:22, Luca Barbato wrote:
On 18/01/16 08:23, Arttu Ylä-Outinen wrote:
I sent this to FFmpeg as well and they noticed that the multiplication
might overflow. I'll send an updated patch soon.

Ticks per frame is 1 or 2, any other value is highly unlikely if I
understand what's the problem at hand.

Here's the updated version anyway (attached).

- Arttu
>From 81b3f152f9b9b22be6479a677ba32c9e176108ed Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Arttu=20Yl=C3=A4-Outinen?= <[email protected]>
Date: Fri, 15 Jan 2016 13:47:10 +0200
Subject: [PATCH v2] libkvazaar: Set frame rate as a rational number
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Updates libkvazaar to pass the exact frame rate to Kvazaar by setting
the numerator and denominator separately instead of a single floating
point number. The exact frame rate is needed for writing timing info to
the bitstream.

Requires Kvazaar version 0.8.1.

Signed-off-by: Arttu Ylä-Outinen <[email protected]>
---
 configure               | 2 +-
 libavcodec/libkvazaar.c | 9 +++++++--
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/configure b/configure
index 8518e69..1067e00 100755
--- a/configure
+++ b/configure
@@ -4472,7 +4472,7 @@ enabled libgsm            && { for gsm_hdr in "gsm.h" "gsm/gsm.h"; do
                                    check_lib "${gsm_hdr}" gsm_create -lgsm && break;
                                done || die "ERROR: libgsm not found"; }
 enabled libilbc           && require libilbc ilbc.h WebRtcIlbcfix_InitDecode -lilbc
-enabled libkvazaar        && require_pkg_config "kvazaar >= 0.7.1" kvazaar.h kvz_api_get
+enabled libkvazaar        && require_pkg_config "kvazaar >= 0.8.1" kvazaar.h kvz_api_get
 enabled libmfx            && require_pkg_config libmfx "mfx/mfxvideo.h" MFXInit
 enabled libmp3lame        && require "libmp3lame >= 3.98.3" lame/lame.h lame_set_VBR_quality -lmp3lame
 enabled libopencore_amrnb && require libopencore_amrnb opencore-amrnb/interf_dec.h Decoder_Interface_init -lopencore-amrnb
diff --git a/libavcodec/libkvazaar.c b/libavcodec/libkvazaar.c
index 89b024a..748b352 100644
--- a/libavcodec/libkvazaar.c
+++ b/libavcodec/libkvazaar.c
@@ -80,8 +80,13 @@ static av_cold int libkvazaar_init(AVCodecContext *avctx)
     cfg->width  = avctx->width;
     cfg->height = avctx->height;
 
-    cfg->framerate =
-      avctx->time_base.den / (double)(avctx->time_base.num * avctx->ticks_per_frame);
+    if (avctx->ticks_per_frame > INT_MAX / avctx->time_base.num) {
+        av_log(avctx, AV_LOG_ERROR,
+               "Could not set framerate for kvazaar: integer overflow\n");
+        return AVERROR(EINVAL);
+    }
+    cfg->framerate_num   = avctx->time_base.den;
+    cfg->framerate_denom = avctx->time_base.num * avctx->ticks_per_frame;
     cfg->target_bitrate = avctx->bit_rate;
     cfg->vui.sar_width  = avctx->sample_aspect_ratio.num;
     cfg->vui.sar_height = avctx->sample_aspect_ratio.den;
-- 
2.7.0

_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to