Files: re-encode a video the browser can't play, on the machine's own GPU, and play that

The viewer tries the file as it is first. When the browser gives up — Safari and an MKV,
HEVC without a hardware decoder — or on a tap, a picker: codec (H.264, HEVC, AV1, or the
original codec repackaged) and size (original, 1080p, 720p, 480p), each codec labelled
with the engine this machine would use. POST /api/files/play runs ffmpeg into an HLS
event playlist of fragmented-MP4 segments under <data dir>/transcode/<id>/; Safari plays
HLS itself and everything else goes through hls.js, loaded only then. The recipes are
tried best first: a card on the bus before the integrated GPU, decoding on the same card
before decoding on the CPU and uploading, the software encoders last; a recipe that
produces no first segment is dropped for the next. The seekable range grows as ffmpeg
gets ahead; status says how far. Streams are reused for the same request, closed when
the player goes, swept after ten idle minutes, and the folder is emptied at start. Root
mode runs ffmpeg as root through a helper that ends when the daemon lets go of it.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
Mitchell-Johnson
2026-10-11 17:42:25 -07:00
co-authored by Claude Fable 5.1
parent aa4ae57c22
commit f3c547adfa
18 changed files with 1351 additions and 32 deletions
+49 -12
View File
@@ -39,6 +39,8 @@ type GPU struct {
Device string `json:"device"` // /dev/dri/renderD128; "0" for cuda (its device index)
Name string `json:"name,omitempty"` // "Intel Arc A380"
Codecs []string `json:"codecs,omitempty"` // ffmpeg codec names it decodes; nil when nothing could say
Encodes []string `json:"encodes,omitempty"` // ffmpeg codec names it encodes, from vainfo
Integrated bool `json:"integrated,omitempty"` // on the CPU die; a card on the bus is preferred for encoding
Reason string `json:"reason,omitempty"` // Unusable only: why
Permission bool `json:"permission,omitempty"` // Unusable only: the node exists but this user may not open it
}
@@ -49,7 +51,8 @@ type Caps struct {
FFprobe string `json:"ffprobe,omitempty"`
Nice string `json:"nice,omitempty"`
Version string `json:"version,omitempty"`
GPUs []GPU `json:"gpus"` // usable, in preference order
GPUs []GPU `json:"gpus"` // usable, in node order
Encoders []string `json:"encoders,omitempty"` // the encoders ffmpeg itself was built with that matter here
Unusable []GPU `json:"unusable,omitempty"` // render nodes that exist but cannot be used, with the reason
Error string `json:"error,omitempty"` // why nothing can be made at all
ProbedAt time.Time `json:"probed_at"`
@@ -139,13 +142,14 @@ func Probe(ctx context.Context) Caps {
return c
}
c.Version = ffmpegVersion(ctx, c.FFmpeg)
c.Encoders = encoders(ctx, c.FFmpeg)
accels := hwaccels(ctx, c.FFmpeg)
if accels["vaapi"] {
nodes, _ := filepath.Glob("/dev/dri/renderD*")
sort.Strings(nodes)
vainfo, _ := exec.LookPath("vainfo")
for _, dev := range nodes {
g := GPU{API: "vaapi", Device: dev, Name: metrics.RenderNodeName(dev)}
g := GPU{API: "vaapi", Device: dev, Name: metrics.RenderNodeName(dev), Integrated: metrics.RenderNodeIntegrated(dev)}
if err := openable(dev); err != nil {
g.Reason, g.Permission = err.Error(), errors.Is(err, os.ErrPermission)
c.Unusable = append(c.Unusable, g)
@@ -157,7 +161,7 @@ func Probe(ctx context.Context) Caps {
continue
}
if vainfo != "" {
g.Codecs = vainfoCodecs(ctx, vainfo, dev)
g.Codecs, g.Encodes = vainfoCodecs(ctx, vainfo, dev)
}
c.GPUs = append(c.GPUs, g)
}
@@ -246,28 +250,61 @@ func hwaccels(ctx context.Context, ffmpeg string) map[string]bool {
return m
}
var vaProfileRe = regexp.MustCompile(`VAProfile([A-Za-z0-9]+?)(?:Simple|Main|High|Profile|Baseline|Constrained|Scc|Advanced|Version|Stereo|Multiview)\w*\s*:\s*VAEntrypointVLD`)
var vaProfileRe = regexp.MustCompile(`VAProfile([A-Za-z0-9]+?)(?:Simple|Main|High|Profile|Baseline|Constrained|Scc|Advanced|Version|Stereo|Multiview)\w*\s*:\s*VAEntrypoint(VLD|EncSlice|EncSliceLP)\b`)
// wantedEncoders are the encoders this package may ask ffmpeg for.
var wantedEncoders = set("libx264 libx265 libsvtav1 libaom-av1 libopenh264 h264_vaapi hevc_vaapi av1_vaapi aac")
// encoders lists the wanted encoders this ffmpeg was built with.
func encoders(ctx context.Context, ffmpeg string) []string {
ctx, cancel := context.WithTimeout(ctx, 5*time.Second)
defer cancel()
out, _ := exec.CommandContext(ctx, ffmpeg, "-hide_banner", "-encoders").Output()
var found []string
for _, l := range strings.Split(string(out), "\n") {
f := strings.Fields(l)
if len(f) >= 2 && wantedEncoders[f[1]] {
found = append(found, f[1])
}
}
sort.Strings(found)
return found
}
// vaCodec maps a VAProfile family to the ffmpeg codec it decodes.
var vaCodec = map[string]string{"H264": "h264", "HEVC": "hevc", "VP9": "vp9", "AV1": "av1", "MPEG2": "mpeg2video", "VC1": "vc1", "VP8": "vp8", "JPEG": "mjpeg", "MPEG4": "mpeg4", "H263": "h263"}
func vainfoCodecs(ctx context.Context, vainfo, dev string) []string {
func vainfoCodecs(ctx context.Context, vainfo, dev string) (decodes, encodes []string) {
ctx, cancel := context.WithTimeout(ctx, 10*time.Second)
defer cancel()
out, _ := exec.CommandContext(ctx, vainfo, "--display", "drm", "--device", dev).Output()
return parseVainfo(string(out))
}
// parseVainfo keeps the decode (VLD) profiles and names their codecs the way ffmpeg does.
func parseVainfo(text string) []string {
seen := map[string]bool{}
// parseVainfo names the codecs the decode (VLD) and encode (EncSlice) profiles cover,
// the way ffmpeg names them.
func parseVainfo(text string) (decodes, encodes []string) {
dec, enc := map[string]bool{}, map[string]bool{}
for _, m := range vaProfileRe.FindAllStringSubmatch(text, -1) {
if c, ok := vaCodec[m[1]]; ok {
seen[c] = true
c, ok := vaCodec[m[1]]
if !ok {
continue
}
if m[2] == "VLD" {
dec[c] = true
} else {
enc[c] = true
}
}
out := make([]string, 0, len(seen))
for c := range seen {
return keys(dec), keys(enc)
}
func keys(m map[string]bool) []string {
if len(m) == 0 {
return nil
}
out := make([]string, 0, len(m))
for c := range m {
out = append(out, c)
}
sort.Strings(out)