X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=encoder%2Fencoder.c;h=a0b01fb943d4edf7ad588fe6869c9a48cc4ccac6;hb=d23d18655249944c1ca894b451e2c82c7a584c62;hp=b0a47a865f283421a81f1ce8301d1af436e7d591;hpb=f08b1c6b8e186ff5a931e9a80e8923e42efff0e4;p=x264

diff --git a/encoder/encoder.c b/encoder/encoder.c
index b0a47a86..a0b01fb9 100644
--- a/encoder/encoder.c
+++ b/encoder/encoder.c
@@ -1,7 +1,7 @@
 /*****************************************************************************
  * encoder.c: top-level encoder functions
  *****************************************************************************
- * Copyright (C) 2003-2015 x264 project
+ * Copyright (C) 2003-2016 x264 project
  *
  * Authors: Laurent Aimar <fenrir@via.ecp.fr>
  *          Loren Merritt <lorenm@u.washington.edu>
@@ -91,22 +91,24 @@ static void x264_frame_dump( x264_t *h )
 
     /* Write the frame in display order */
     int frame_size = FRAME_SIZE( h->param.i_height * h->param.i_width * sizeof(pixel) );
-    fseek( f, (uint64_t)h->fdec->i_frame * frame_size, SEEK_SET );
-    for( int p = 0; p < (CHROMA444 ? 3 : 1); p++ )
-        for( int y = 0; y < h->param.i_height; y++ )
-            fwrite( &h->fdec->plane[p][y*h->fdec->i_stride[p]], sizeof(pixel), h->param.i_width, f );
-    if( !CHROMA444 )
-    {
-        int cw = h->param.i_width>>1;
-        int ch = h->param.i_height>>CHROMA_V_SHIFT;
-        pixel *planeu = x264_malloc( (cw*ch*2+32)*sizeof(pixel) );
-        if( planeu )
+    if( !fseek( f, (int64_t)h->fdec->i_frame * frame_size, SEEK_SET ) )
+    {
+        for( int p = 0; p < (CHROMA444 ? 3 : 1); p++ )
+            for( int y = 0; y < h->param.i_height; y++ )
+                fwrite( &h->fdec->plane[p][y*h->fdec->i_stride[p]], sizeof(pixel), h->param.i_width, f );
+        if( !CHROMA444 )
         {
-            pixel *planev = planeu + cw*ch + 16;
-            h->mc.plane_copy_deinterleave( planeu, cw, planev, cw, h->fdec->plane[1], h->fdec->i_stride[1], cw, ch );
-            fwrite( planeu, 1, cw*ch*sizeof(pixel), f );
-            fwrite( planev, 1, cw*ch*sizeof(pixel), f );
-            x264_free( planeu );
+            int cw = h->param.i_width>>1;
+            int ch = h->param.i_height>>CHROMA_V_SHIFT;
+            pixel *planeu = x264_malloc( (cw*ch*2+32)*sizeof(pixel) );
+            if( planeu )
+            {
+                pixel *planev = planeu + cw*ch + 16;
+                h->mc.plane_copy_deinterleave( planeu, cw, planev, cw, h->fdec->plane[1], h->fdec->i_stride[1], cw, ch );
+                fwrite( planeu, 1, cw*ch*sizeof(pixel), f );
+                fwrite( planev, 1, cw*ch*sizeof(pixel), f );
+                x264_free( planeu );
+            }
         }
     }
     fclose( f );
@@ -474,12 +476,12 @@ static int x264_validate_parameters( x264_t *h, int b_open )
 
     int i_csp = h->param.i_csp & X264_CSP_MASK;
 #if X264_CHROMA_FORMAT
-    if( CHROMA_FORMAT != CHROMA_420 && i_csp >= X264_CSP_I420 && i_csp <= X264_CSP_NV12 )
+    if( CHROMA_FORMAT != CHROMA_420 && i_csp >= X264_CSP_I420 && i_csp < X264_CSP_I422 )
     {
         x264_log( h, X264_LOG_ERROR, "not compiled with 4:2:0 support\n" );
         return -1;
     }
-    else if( CHROMA_FORMAT != CHROMA_422 && i_csp >= X264_CSP_I422 && i_csp <= X264_CSP_V210 )
+    else if( CHROMA_FORMAT != CHROMA_422 && i_csp >= X264_CSP_I422 && i_csp < X264_CSP_I444 )
     {
         x264_log( h, X264_LOG_ERROR, "not compiled with 4:2:2 support\n" );
         return -1;
@@ -492,7 +494,7 @@ static int x264_validate_parameters( x264_t *h, int b_open )
 #endif
     if( i_csp <= X264_CSP_NONE || i_csp >= X264_CSP_MAX )
     {
-        x264_log( h, X264_LOG_ERROR, "invalid CSP (only I420/YV12/NV12/I422/YV16/NV16/I444/YV24/BGR/BGRA/RGB supported)\n" );
+        x264_log( h, X264_LOG_ERROR, "invalid CSP (only I420/YV12/NV12/NV21/I422/YV16/NV16/I444/YV24/BGR/BGRA/RGB supported)\n" );
         return -1;
     }
 
@@ -537,7 +539,13 @@ static int x264_validate_parameters( x264_t *h, int b_open )
     }
 
     if( h->param.i_threads == X264_THREADS_AUTO )
+    {
         h->param.i_threads = x264_cpu_num_processors() * (h->param.b_sliced_threads?2:3)/2;
+        /* Avoid too many threads as they don't improve performance and
+         * complicate VBV. Capped at an arbitrary 2 rows per thread. */
+        int max_threads = X264_MAX( 1, (h->param.i_height+15)/16 / 2 );
+        h->param.i_threads = X264_MIN( h->param.i_threads, max_threads );
+    }
     int max_sliced_threads = X264_MAX( 1, (h->param.i_height+15)/16 / 4 );
     if( h->param.i_threads > 1 )
     {
@@ -591,11 +599,20 @@ static int x264_validate_parameters( x264_t *h, int b_open )
         h->param.i_dpb_size = 1;
     }
 
-    if( h->param.i_frame_packing < -1 || h->param.i_frame_packing > 6 )
+    if( h->param.i_frame_packing < -1 || h->param.i_frame_packing > 7 )
     {
         x264_log( h, X264_LOG_WARNING, "ignoring unknown frame packing value\n" );
         h->param.i_frame_packing = -1;
     }
+    if( h->param.i_frame_packing == 7 &&
+        ((h->param.i_width - h->param.crop_rect.i_left - h->param.crop_rect.i_right)  % 3 ||
+         (h->param.i_height - h->param.crop_rect.i_top - h->param.crop_rect.i_bottom) % 3) )
+    {
+        x264_log( h, X264_LOG_ERROR, "cropped resolution %dx%d not compatible with tile format frame packing\n",
+                  h->param.i_width - h->param.crop_rect.i_left - h->param.crop_rect.i_right,
+                  h->param.i_height - h->param.crop_rect.i_top - h->param.crop_rect.i_bottom );
+        return -1;
+    }
 
     /* Detect default ffmpeg settings and terminate with an error. */
     if( b_open )
@@ -993,9 +1010,9 @@ static int x264_validate_parameters( x264_t *h, int b_open )
         h->param.i_fps_num = 25;
         h->param.i_fps_den = 1;
     }
-    float fps = (float) h->param.i_fps_num / h->param.i_fps_den;
+    float fps = (float)h->param.i_fps_num / h->param.i_fps_den;
     if( h->param.i_keyint_min == X264_KEYINT_MIN_AUTO )
-        h->param.i_keyint_min = X264_MIN( h->param.i_keyint_max / 10, fps );
+        h->param.i_keyint_min = X264_MIN( h->param.i_keyint_max / 10, (int)fps );
     h->param.i_keyint_min = x264_clip3( h->param.i_keyint_min, 1, h->param.i_keyint_max/2+1 );
     h->param.rc.i_lookahead = x264_clip3( h->param.rc.i_lookahead, 0, X264_LOOKAHEAD_MAX );
     {
@@ -1392,7 +1409,7 @@ x264_t *x264_encoder_open( x264_param_t *param )
 {
     x264_t *h;
     char buf[1000], *p;
-    int qp, i_slicetype_length;
+    int i_slicetype_length;
 
     CHECKED_MALLOCZERO( h, sizeof(x264_t) );
 
@@ -1559,15 +1576,8 @@ x264_t *x264_encoder_open( x264_param_t *param )
         p += sprintf( p, " none!" );
     x264_log( h, X264_LOG_INFO, "%s\n", buf );
 
-    float *logs = x264_analyse_prepare_costs( h );
-    if( !logs )
+    if( x264_analyse_init_costs( h ) )
         goto fail;
-    for( qp = X264_MIN( h->param.rc.i_qp_min, QP_MAX_SPEC ); qp <= h->param.rc.i_qp_max; qp++ )
-        if( x264_analyse_init_costs( h, logs, qp ) )
-            goto fail;
-    if( x264_analyse_init_costs( h, logs, X264_LOOKAHEAD_QP ) )
-        goto fail;
-    x264_free( logs );
 
     static const uint16_t cost_mv_correct[7] = { 24, 47, 95, 189, 379, 757, 1515 };
     /* Checks for known miscompilation issues. */
@@ -1692,6 +1702,7 @@ x264_t *x264_encoder_open( x264_param_t *param )
         else if( !x264_is_regular_file( f ) )
         {
             x264_log( h, X264_LOG_ERROR, "dump_yuv: incompatible with non-regular file %s\n", h->param.psz_dump_yuv );
+            fclose( f );
             goto fail;
         }
         fclose( f );
@@ -3025,9 +3036,8 @@ static void x264_thread_sync_context( x264_t *dst, x264_t *src )
 
 static void x264_thread_sync_stat( x264_t *dst, x264_t *src )
 {
-    if( dst == src )
-        return;
-    memcpy( &dst->stat.i_frame_count, &src->stat.i_frame_count, sizeof(dst->stat) - sizeof(dst->stat.frame) );
+    if( dst != src )
+        memcpy( &dst->stat, &src->stat, offsetof(x264_t, stat.frame) - offsetof(x264_t, stat) );
 }
 
 static void *x264_slices_write( x264_t *h )
@@ -3229,6 +3239,12 @@ int     x264_encoder_encode( x264_t *h,
     /* ------------------- Setup new frame from picture -------------------- */
     if( pic_in != NULL )
     {
+        if( h->lookahead->b_exit_thread )
+        {
+            x264_log( h, X264_LOG_ERROR, "lookahead thread is already stopped\n" );
+            return -1;
+        }
+
         /* 1: Copy the picture to a frame and move it to a buffer */
         x264_frame_t *fenc = x264_frame_pop_unused( h, 0 );
         if( !fenc )
@@ -3939,13 +3955,13 @@ static int x264_encoder_frame_end( x264_t *h, x264_t *thread_current,
     {
         pic_out->prop.f_ssim = h->stat.frame.f_ssim / h->stat.frame.i_ssim_cnt;
         h->stat.f_ssim_mean_y[h->sh.i_type] += pic_out->prop.f_ssim * dur;
-        snprintf( psz_message + strlen(psz_message), 80 - strlen(psz_message),
-                  " SSIM Y:%.5f", pic_out->prop.f_ssim );
+        int msg_len = strlen(psz_message);
+        snprintf( psz_message + msg_len, 80 - msg_len, " SSIM Y:%.5f", pic_out->prop.f_ssim );
     }
     psz_message[79] = '\0';
 
     x264_log( h, X264_LOG_DEBUG,
-                  "frame=%4d QP=%.2f NAL=%d Slice:%c Poc:%-3d I:%-4d P:%-4d SKIP:%-4d size=%d bytes%s\n",
+              "frame=%4d QP=%.2f NAL=%d Slice:%c Poc:%-3d I:%-4d P:%-4d SKIP:%-4d size=%d bytes%s\n",
               h->i_frame,
               h->fdec->f_qp_avg_aq,
               h->i_nal_ref_idc,