diff --git a/libavcodec/h261dec.c b/libavcodec/h261dec.c
index 5c25aa9cb3..eb544e6046 100644
--- a/libavcodec/h261dec.c
+++ b/libavcodec/h261dec.c
@@ -678,6 +678,6 @@ AVCodec ff_h261_decoder = {
     .close          = h261_decode_end,
     .decode         = h261_decode_frame,
     .capabilities   = AV_CODEC_CAP_DR1,
-    .caps_internal  = FF_CODEC_CAP_INIT_THREADSAFE | FF_CODEC_CAP_INIT_CLEANUP,
+    .caps_internal  = FF_CODEC_CAP_INIT_THREADSAFE,
     .max_lowres     = 3,
 };
diff --git a/libavcodec/h263dec.c b/libavcodec/h263dec.c
index dafa54d8d4..e8b4d83e6e 100644
--- a/libavcodec/h263dec.c
+++ b/libavcodec/h263dec.c
@@ -771,7 +771,7 @@ AVCodec ff_h263_decoder = {
     .decode         = ff_h263_decode_frame,
     .capabilities   = AV_CODEC_CAP_DRAW_HORIZ_BAND | AV_CODEC_CAP_DR1 |
                       AV_CODEC_CAP_TRUNCATED | AV_CODEC_CAP_DELAY,
-    .caps_internal  = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM | FF_CODEC_CAP_INIT_CLEANUP,
+    .caps_internal  = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM,
     .flush          = ff_mpeg_flush,
     .max_lowres     = 3,
     .pix_fmts       = ff_h263_hwaccel_pixfmt_list_420,
@@ -789,7 +789,7 @@ AVCodec ff_h263p_decoder = {
     .decode         = ff_h263_decode_frame,
     .capabilities   = AV_CODEC_CAP_DRAW_HORIZ_BAND | AV_CODEC_CAP_DR1 |
                       AV_CODEC_CAP_TRUNCATED | AV_CODEC_CAP_DELAY,
-    .caps_internal  = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM | FF_CODEC_CAP_INIT_CLEANUP,
+    .caps_internal  = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM,
     .flush          = ff_mpeg_flush,
     .max_lowres     = 3,
     .pix_fmts       = ff_h263_hwaccel_pixfmt_list_420,
diff --git a/libavcodec/mpeg12dec.c b/libavcodec/mpeg12dec.c
index 92dd6a0b24..94221da2c1 100644
--- a/libavcodec/mpeg12dec.c
+++ b/libavcodec/mpeg12dec.c
@@ -2880,7 +2880,8 @@ static av_cold int mpeg_decode_end(AVCodecContext *avctx)
 {
     Mpeg1Context *s = avctx->priv_data;
 
-    ff_mpv_common_end(&s->mpeg_enc_ctx);
+    if (s->mpeg_enc_ctx_allocated)
+        ff_mpv_common_end(&s->mpeg_enc_ctx);
     av_buffer_unref(&s->a53_buf_ref);
     return 0;
 }
@@ -2897,7 +2898,7 @@ AVCodec ff_mpeg1video_decoder = {
     .capabilities          = AV_CODEC_CAP_DRAW_HORIZ_BAND | AV_CODEC_CAP_DR1 |
                              AV_CODEC_CAP_TRUNCATED | AV_CODEC_CAP_DELAY |
                              AV_CODEC_CAP_SLICE_THREADS,
-    .caps_internal         = FF_CODEC_CAP_INIT_THREADSAFE | FF_CODEC_CAP_INIT_CLEANUP |
+    .caps_internal         = FF_CODEC_CAP_INIT_THREADSAFE |
                              FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM,
     .flush                 = flush,
     .max_lowres            = 3,
@@ -2931,7 +2932,7 @@ AVCodec ff_mpeg2video_decoder = {
     .capabilities   = AV_CODEC_CAP_DRAW_HORIZ_BAND | AV_CODEC_CAP_DR1 |
                       AV_CODEC_CAP_TRUNCATED | AV_CODEC_CAP_DELAY |
                       AV_CODEC_CAP_SLICE_THREADS,
-    .caps_internal  = FF_CODEC_CAP_INIT_THREADSAFE | FF_CODEC_CAP_INIT_CLEANUP |
+    .caps_internal  = FF_CODEC_CAP_INIT_THREADSAFE |
                       FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM,
     .flush          = flush,
     .max_lowres     = 3,
@@ -2976,7 +2977,7 @@ AVCodec ff_mpegvideo_decoder = {
     .close          = mpeg_decode_end,
     .decode         = mpeg_decode_frame,
     .capabilities   = AV_CODEC_CAP_DRAW_HORIZ_BAND | AV_CODEC_CAP_DR1 | AV_CODEC_CAP_TRUNCATED | AV_CODEC_CAP_DELAY | AV_CODEC_CAP_SLICE_THREADS,
-    .caps_internal  = FF_CODEC_CAP_INIT_THREADSAFE | FF_CODEC_CAP_INIT_CLEANUP |
+    .caps_internal  = FF_CODEC_CAP_INIT_THREADSAFE |
                       FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM,
     .flush          = flush,
     .max_lowres     = 3,
diff --git a/libavcodec/mpeg4videodec.c b/libavcodec/mpeg4videodec.c
index 2ca804fdc6..2c440a5026 100644
--- a/libavcodec/mpeg4videodec.c
+++ b/libavcodec/mpeg4videodec.c
@@ -3579,8 +3579,7 @@ AVCodec ff_mpeg4_decoder = {
                              AV_CODEC_CAP_TRUNCATED | AV_CODEC_CAP_DELAY |
                              AV_CODEC_CAP_FRAME_THREADS,
     .caps_internal         = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM |
-                             FF_CODEC_CAP_ALLOCATE_PROGRESS |
-                             FF_CODEC_CAP_INIT_CLEANUP,
+                             FF_CODEC_CAP_ALLOCATE_PROGRESS,
     .flush                 = ff_mpeg_flush,
     .max_lowres            = 3,
     .pix_fmts              = ff_h263_hwaccel_pixfmt_list_420,
diff --git a/libavcodec/mpegvideo.c b/libavcodec/mpegvideo.c
index 728e591958..e03f174024 100644
--- a/libavcodec/mpegvideo.c
+++ b/libavcodec/mpegvideo.c
@@ -932,17 +932,17 @@ av_cold int ff_mpv_common_init(MpegEncContext *s)
     for (i = 0; i < MAX_PICTURE_COUNT; i++) {
         s->picture[i].f = av_frame_alloc();
         if (!s->picture[i].f)
-            return AVERROR(ENOMEM);
+            goto fail_nomem;
     }
 
     if (!(s->next_picture.f    = av_frame_alloc()) ||
         !(s->last_picture.f    = av_frame_alloc()) ||
         !(s->current_picture.f = av_frame_alloc()) ||
         !(s->new_picture.f     = av_frame_alloc()))
-        return AVERROR(ENOMEM);
+        goto fail_nomem;
 
     if ((ret = init_context_frame(s)))
-        return AVERROR(ENOMEM);
+        goto fail;
 
     s->parse_context.state = -1;
 
@@ -956,10 +956,10 @@ av_cold int ff_mpv_common_init(MpegEncContext *s)
             if (i) {
                 s->thread_context[i] = av_memdup(s, sizeof(MpegEncContext));
                 if (!s->thread_context[i])
-                    return AVERROR(ENOMEM);
+                    goto fail_nomem;
             }
             if ((ret = init_duplicate_context(s->thread_context[i])) < 0)
-                return ret;
+                goto fail;
             s->thread_context[i]->start_mb_y =
                 (s->mb_height * (i) + nb_slices / 2) / nb_slices;
             s->thread_context[i]->end_mb_y   =
@@ -967,7 +967,7 @@ av_cold int ff_mpv_common_init(MpegEncContext *s)
         }
     } else {
         if ((ret = init_duplicate_context(s)) < 0)
-            return ret;
+            goto fail;
         s->start_mb_y = 0;
         s->end_mb_y   = s->mb_height;
     }
@@ -975,6 +975,11 @@ av_cold int ff_mpv_common_init(MpegEncContext *s)
 //     }
 
     return 0;
+ fail_nomem:
+    ret = AVERROR(ENOMEM);
+ fail:
+    ff_mpv_common_end(s);
+    return ret;
 }
 
 /**
@@ -1067,17 +1072,17 @@ int ff_mpv_common_frame_size_change(MpegEncContext *s)
 
     if ((s->width || s->height) &&
         (err = av_image_check_size(s->width, s->height, 0, s->avctx)) < 0)
-        return err;
+        goto fail;
 
     /* set chroma shifts */
     err = av_pix_fmt_get_chroma_sub_sample(s->avctx->pix_fmt,
                                            &s->chroma_x_shift,
                                            &s->chroma_y_shift);
     if (err < 0)
-        return err;
+        goto fail;
 
     if ((err = init_context_frame(s)))
-        return err;
+        goto fail;
 
     memset(s->thread_context, 0, sizeof(s->thread_context));
     s->thread_context[0]   = s;
@@ -1089,11 +1094,12 @@ int ff_mpv_common_frame_size_change(MpegEncContext *s)
                 if (i) {
                     s->thread_context[i] = av_memdup(s, sizeof(MpegEncContext));
                     if (!s->thread_context[i]) {
-                        return AVERROR(ENOMEM);
+                        err = AVERROR(ENOMEM);
+                        goto fail;
                     }
                 }
                 if ((err = init_duplicate_context(s->thread_context[i])) < 0)
-                    return err;
+                    goto fail;
                 s->thread_context[i]->start_mb_y =
                     (s->mb_height * (i) + nb_slices / 2) / nb_slices;
                 s->thread_context[i]->end_mb_y   =
@@ -1102,7 +1108,7 @@ int ff_mpv_common_frame_size_change(MpegEncContext *s)
         } else {
             err = init_duplicate_context(s);
             if (err < 0)
-                return err;
+                goto fail;
             s->start_mb_y = 0;
             s->end_mb_y   = s->mb_height;
         }
@@ -1110,6 +1116,9 @@ int ff_mpv_common_frame_size_change(MpegEncContext *s)
     }
 
     return 0;
+ fail:
+    ff_mpv_common_end(s);
+    return err;
 }
 
 /* init common structure for both encoder and decoder */
diff --git a/libavcodec/msmpeg4dec.c b/libavcodec/msmpeg4dec.c
index 405fda4b83..63f9faa1bb 100644
--- a/libavcodec/msmpeg4dec.c
+++ b/libavcodec/msmpeg4dec.c
@@ -870,7 +870,7 @@ AVCodec ff_msmpeg4v1_decoder = {
     .close          = ff_h263_decode_end,
     .decode         = ff_h263_decode_frame,
     .capabilities   = AV_CODEC_CAP_DRAW_HORIZ_BAND | AV_CODEC_CAP_DR1,
-    .caps_internal  = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM | FF_CODEC_CAP_INIT_CLEANUP,
+    .caps_internal  = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM,
     .max_lowres     = 3,
     .pix_fmts       = (const enum AVPixelFormat[]) {
         AV_PIX_FMT_YUV420P,
@@ -888,7 +888,7 @@ AVCodec ff_msmpeg4v2_decoder = {
     .close          = ff_h263_decode_end,
     .decode         = ff_h263_decode_frame,
     .capabilities   = AV_CODEC_CAP_DRAW_HORIZ_BAND | AV_CODEC_CAP_DR1,
-    .caps_internal  = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM | FF_CODEC_CAP_INIT_CLEANUP,
+    .caps_internal  = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM,
     .max_lowres     = 3,
     .pix_fmts       = (const enum AVPixelFormat[]) {
         AV_PIX_FMT_YUV420P,
@@ -906,7 +906,7 @@ AVCodec ff_msmpeg4v3_decoder = {
     .close          = ff_h263_decode_end,
     .decode         = ff_h263_decode_frame,
     .capabilities   = AV_CODEC_CAP_DRAW_HORIZ_BAND | AV_CODEC_CAP_DR1,
-    .caps_internal  = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM | FF_CODEC_CAP_INIT_CLEANUP,
+    .caps_internal  = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM,
     .max_lowres     = 3,
     .pix_fmts       = (const enum AVPixelFormat[]) {
         AV_PIX_FMT_YUV420P,
@@ -924,7 +924,7 @@ AVCodec ff_wmv1_decoder = {
     .close          = ff_h263_decode_end,
     .decode         = ff_h263_decode_frame,
     .capabilities   = AV_CODEC_CAP_DRAW_HORIZ_BAND | AV_CODEC_CAP_DR1,
-    .caps_internal  = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM | FF_CODEC_CAP_INIT_CLEANUP,
+    .caps_internal  = FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM,
     .max_lowres     = 3,
     .pix_fmts       = (const enum AVPixelFormat[]) {
         AV_PIX_FMT_YUV420P,
diff --git a/libavcodec/rv10.c b/libavcodec/rv10.c
index d5c7480611..89b838ad15 100644
--- a/libavcodec/rv10.c
+++ b/libavcodec/rv10.c
@@ -687,7 +687,6 @@ AVCodec ff_rv10_decoder = {
     .close          = rv10_decode_end,
     .decode         = rv10_decode_frame,
     .capabilities   = AV_CODEC_CAP_DR1,
-    .caps_internal  = FF_CODEC_CAP_INIT_CLEANUP,
     .max_lowres     = 3,
     .pix_fmts       = (const enum AVPixelFormat[]) {
         AV_PIX_FMT_YUV420P,
@@ -705,7 +704,6 @@ AVCodec ff_rv20_decoder = {
     .close          = rv10_decode_end,
     .decode         = rv10_decode_frame,
     .capabilities   = AV_CODEC_CAP_DR1 | AV_CODEC_CAP_DELAY,
-    .caps_internal  = FF_CODEC_CAP_INIT_CLEANUP,
     .flush          = ff_mpeg_flush,
     .max_lowres     = 3,
     .pix_fmts       = (const enum AVPixelFormat[]) {