| Commit message (Collapse) | Author | Files | Lines |
|---|
|
commands should really not behave differently if they're issued
inside a command list or not; so stop having special handler
functions to deal with them. "update" was the only command
that used this functionality and I changed that in the last
commit to serialize access.
|
|
Now the "update" command can be issued multiple times regardless
of whether the client is in list mode or not.
We serialize the update tasks to prevent updates from trampling
over each other and will spawn another update task
once the current one is finished updating and reaped.
Right now we cap the queue size to 32 which is probably enough (I
bet most people usually run update with no argument anyways);
but we can make it grow/shrink dynamically if needed. There'll
still be a hard-coded limit to prevent DoS attacks, though.
|
|
It'll be easier to keep track of what code runs in what
task/thread this way.
|
|
Instead of relying on the shortname, just pass the song pointer
to prevent redundant lookups during deletes.
|
|
This is still not SMP-safe yet, as it needs at least a barrier
before calling tag_free(old_tag).
Locks may be simpler to implement and the potential performance
gain of avoiding locks may not be worth it on infrequently
modified data structures...
|
|
Hopefully this makes the code feel less claustrophobic...
|
|
This actually opened us up to making lock dependencies more
difficult than they needed to be now that we have threaded
updates. We would always use the memory anyways, just in
the stack instead of bss.
|
|
Seeing the "mpd_" prefix _everywhere_ is mind-numbing as the
mind needs to retrain itself to skip over the first 4 tokens of
a type to get to its meaning. So avoid having extra characters
on my terminal to make it easier to follow code at 2:30 am in
the morning.
Please report any new issues you may come across on Free
toolchains. I realize how difficult it can be to build/maintain
cross-compiling toolchains and I have no intention of forcing
people to upgrade their toolchains to build mpd.
Tested with gcc 2.95.4 and and gcc 4.3.1 on x86-32.
|
|
tfing wrote:
> I have quite some files with an empty album tag as they do not come
> from a particular album.
>
> If I want to look for those files and browse them, this happens:
> :: nc localhost 6600
> OK MPD 0.12.0
> find album ""
> ACK [2@0] {find} too few arguments for "find"
>
> I'd like to be able to browse those files in a client like gmpc.
> So these 2 items would have to be developed:
> - list album should report that some files have an empty tag
> - it should be possible to search for an empty tag with the find command
Patch-by: Marc Pavot
ref: http://musicpd.org/mantis/view.php?id=464
|
|
We already sanitize and duplicated our paths before calling
updateInit() to get pre-pthread_create() error-checking along
with heap allocation reduction because we don't have to redupe
because our parent stack went away.
|
|
Remove yet another use of our old malloc-happy linked list
implementation and replace it with a simple array of strings.
This also implements more eager error handling of invalid
paths (still centralized in updateInit) so we can filter out
bad paths before we spawn a thread.
This also does its part to fix the "update" command inside list mode
which lost its static variable in
ada24f9a921ff95d874195acf253b5a9dd12213d (although it was broken and
requires the fix in 769939b62f7557f8e7c483223d68a8b39af43e37, too).
|
|
This only breaks "update" under list command mode and
no other commands. This can be done more optimally
without the extra heap allocation via xstrdup(); but is
uncommon enough to not matter.
|
|
chunk_length can be converted to a local variable, because it is
always reset to 0 after it was used.
|
|
Since flacSendChunk() is a trivial function and is only used in one
location, move the code there. The advantage is that calling
decoder_data() directly returns the decoder_command value, so we can
eliminate one decoder_get_command() call.
[ew: using the terser API in the main branch, it always
eliminated the dc_intr()/dc_seek() call, but I agree that
killing a camel, err.. camelCase function is a good thing]
|
|
Support for bit rates except 16 bits (and 8 bits on little endian) has
always been broken. Since we added optimized functions for 8, 16,
24/32 bits, we can remove the generic flac_convert() function.
Instead of removing it, convert it to a wrapper function for
flac_convert_*().
|
|
Same optimization for 8 and 32 bit files, like the previous patch for
16 bit. Along the way, this patch adds 24 bit FLAC support!
|
|
flac_convert_16() runs a lot faster than the generic (and quite buggy)
function flac_convert(). flac_convert_16() is only used for
non-stereo files, since there is already flac_convert_stereo16().
|
|
By mistake, I casted the sample value to uint16_t, which is wrong.
This patch simplifies the code by using a int16_t pointer instead of
casting to int16_t* every time.
|
|
There is still a lot of duplicated code in flac_plugin.c and
oggflac_plugin.c. Move code from flac_plugin.c to _flac_common.c, and
use the new function flac_common_write() also in oggflac_plugin.c,
porting lots of optimizations over to it.
|
|
The previous patch on this topic was incomplete: it still added
data->chunk_length when calling flac_convert(). Remove this, too.
|
|
Make the code more readable by moving the range checks to pcm_range().
gcc does quite a good job at optimizing it: the resulting binary is
exactly the same, although it contains a parametrized shift instead of
hard-coded boundaries.
|
|
Merge some code into an inline function, so we can optimize it later
only once.
|
|
Move error reporting to command.c so directory.c does not deal
with client error handling any more.
|
|
Improving the signal to noise ratio...
|
|
If a write failed, it's a good sign subsequent writes will fail,
too, so propgate errors all the way up the stack.
|
|
A long time ago in an mpd far away...
|
|
MPD has supported more audio formats than just MP3
for over five years...
|
|
We don't change the song pointer there, either.
|
|
We don't modify the Song when we delete it
|
|
It was a huge confusing mess of parameter passing around
and around. Add a few extra assertions to ensure we're
handling parent/child relationships properly.
|
|
This is like basename(3) but with predictable semantics independent
of C library or build options used. This is also much more strict
and does not account for trailing slashes (mpd should never deal with
trailing slashes on internal functions).
|
|
We no longer fork for directory updates, so we
no longer have children to reap.
|
|
Small memory reduction compared to songvec since most users have
much fewer dirs than songs, but still nice to have.
|
|
We no longer for for updates
|
|
"free" implies the songvec structure itself is freed,
which is not the case.
|
|
|
|
If we updated the mpd metadata database; then there's a chance
some of those songs in the playlist will have updated metadata.
So be on the safe side and increment the playlist version number
if _any_ song changed (this is how all released versions of mpd
did it, too).
This bug was introduced recently when making "update" threaded.
Thanks to stonecrest for the bug report.
|
|
We forgot to update the playlist.queued marker if
playlist.current changed.
Additionally, if the queue cleared in any other mode,
attempt to requeue (as it's a harmless no-op otherwise).
Thanks to stonecrest for the bug report.
|
|
If repeat is off, we reset (and reshuffle in random mode)
the playlist.
|
|
Fix this regression introduced in the core rewrite so that we
now skip to the next song when we encounter an error with the
song we tried to decode.
|
|
SongList has been superseded by struct songvec.
|
|
With patch 8d2830b3, I broke "addid": it did not return the id of the
new song, because of a typo in the return condition (== instead of
!=).
|
|
This reverts commit efefaee1f9535012be2fbfea8f0f870904daad5d.
Conflicts:
src/directory.c
|
|
Potentially broken free() implementations don't like it
|
|
|
|
Use freeList() instead of free() to free all elements in
the list.
|
|
It's possible the playlist will be accessing a song that is to
be freed in the update thread. Rather than going through the
complexity (and potential to make mistakes) of locking the
playlist (as well as losing CPU cycles/pipelining due to
barriers with mutexes), we'll just line up all songs to
be freed in the main thread.
It's relatively uncommon to call freeSong() heavily (as it is to
update); so the extra, temporary memory usage won't be very
noticeable.
Additionally, if a song is renamed and it contains unique tag
item; this has the additional side effect of preventing
unnecessary fragmentation where an item is freed and shortly
reallocated.
|
|
The umask calls were remants of when we used fopen().
|
|
open(2) should only interrupt on "slow" devices, afaik...
|
|
|
