This makes it far more useful, as before it insisted on scanning from
the root directory and putting the generated db files in /.rockbox
Change-Id: I0d0e3c6c6c7ee1fc5473185482fbfb9a2a2be29f
This was introduced by adding multivolume support to the hibyplayer units,
and might affect other things, if so a more general fix might be needed.
Change-Id: Ia9c1b6b1eb50e67221ce440d30d6770daaeac74e
Three separate problems:
1) Make 4.4 defaults to a named pipe for jobserver control, this
was incompatible with how glibc did recursive makes
2) Make 4.4 passed long arguments ( --foo ) into MAKEFLAGS and this
broke stuff that wasn't expecting it
3) Circular dependency in a header generation due do make 4.4 being
stricter/more consistent in how it orders rules vs which makefiles
they came from. This one was the real !@#!@ to deal with.
This patch set fixes the hosted arm (glibc 2.19) and mips (glibc 2.25)
toolchains.
Change-Id: Ie6d6a0ab7e1b36f24c43e524fee4afc0bd3a14d6
eg, our current toolchian generates this:
GNU ld (GNU Binutils) 2.26.1
But the system toolchain on the buildserver generates this:
GNU ld version 2.37-37.fc36
Change-Id: I0166f12d660654c1456289954fb256db98213618
NO_CROSSREFS_TO() was first introduced in binutils 2.27.
It is used to have the linker report errors when symbols
from a list of sections refer to a specified section,
which is useful for verifying that normal code does not
refer to INIT_ATTR code.
Note this doesn't actually start using NO_CROSSREFS_TO()
for builds -- that will have to wait until the toolchain
upgrade when we'll have a new enough binutils -- it just
detects support for the feature so it can be conditionally
enabled in linker scripts.
Change-Id: If1553eaa671fcbbdfe6af357a6331931e7c4f997
Previously, it was hardcoded to the english convention of units-last, so
"100%" would be voiced as "one hundred percent". This adds a new
language flag that makes the units be voiced first, ie "100%" will be
voiced as "percent one hundred".
So far only the Chinese-traditional and Chinese-simplified languages
utilize this feature (taken from an old ticket, FS#10340) but I'm sure
others would want this feature too.
Change-Id: Idf825ec9299dc0ed09921cf67aec61b1ab262fc6
Notes:
* Currently limited to 32-bit sector addresses due to internal Rockbox
APIs. So this means a practical limit of 2TiB per drive.
* Only 'General Data' GPT partition type is recognised, as that's
what SD cards seem to use for exFAT/FAT32.
Note that _booting_ off GPT-partitioned drive will require rebuilding
the various rockbox bootloaders, and even then there may be platform
limitations that preclude this.
Change-Id: Ibfaae1960adcb1e81976d4b60dd596c6d16318e4
I couldn't seem to reproduce the issue here:
https://forums.rockbox.org/index.php/topic,54165.msg252081.html#msg252081
but I figure its probably a threading issue
so we now have a mutex on the cache
and to top it all off each cached entry has a crc and length that are
checked before writing the entry to the file otherwise it is prepended
with # FAILED - so hopefully scrobbler 'parsers?' don't barf on the log
Other changes:
there is now a MRU table for tracks this should help prevent duplicates
it is configurable..
the cache buffer now no longer uses fixed chunks allowing more tracks
to be written between flushes
Change-Id: Iaab7e3f6a76abfc61130f3233379a51c9a6d12e5
First was an off-by-one bug in the loop over buffers. The condition was
`i < this->hidden->n_buffers', but should have been <=, not <. This wasn't
a critical issue, because the last iteration would check the currently
playing buffer - an unlikely situation.
The second logic bug would make the code always think it had found an empty
buffer, even when it had not, since `idx' was being unconditionally set
every loop, rendering it unusable for a sentinel value. Fix that.
Change-Id: I137330965d699acf827e59b6aa5c5f9ddf2ca617
Added fw modifications required to unpack real world player dumps.
Documented more fwu header fields, magic numbers and finite field arithmetics (extended Euclidean for inverse, long division for reducing modulo field_poly).
v3 encryption used is standard RC4 with the key additionally ciphered by the Elliptic Curve Integrated Encryption Scheme.
Either sect233k1 (NIST K-233) or sect163r2 (NIST B-163) curves can be used, with the former overwhelmingly prevailing, being hardwired in SDK's maker.exe. Using a private/public key scheme is superfluous because both are stored in the firmware, with the added level of complexity likely serving the purpose of obfuscation. The private key is generated at random with each invokation.
None of KDF or MAC from ECIES are used, RC4 key is directly xored with the shared secret. The random number r used to calculate rG isn't stored, but that's unimportant since only krG == rkG is actually used in the encryption.
Change-Id: Ieacf8cc744bc90c7c5582dd724b2c10a41bfc191
Targets with remote LCDs need to have the lcd scroll driver instantiated,
even in the bootloader.
This is because lcd_remote_scroll_stop() is called, in addition to
lcd_scroll_stop(). This could probably be excised.
Change-Id: I475269f9361eb469c62ad0f74abc1eb3baa4c134
a new slot gets added but it is set to 0 and an entry gets discarded
instead set it to the last entry read
Change-Id: I3d6f9b73030c8a5083d299c99314f9c40052689a
This turns on -fwrapv and -fwrapv-pointers.
As per GCC docs:
"This option instructs the compiler to assume that signed arithmetic
overflow of addition, subtraction and multiplication wraps around
using twos-complement representation. This flag enables some
optimizations and disables others."
All of our targets fall under this category.
For some time now we've been using -Wall which implies
-Wstrict-overflow=2, which only warns when the compiler actually
implements a possibly questionable optimization, so I'd expect this to
not affect our code size at all.
Change-Id: I8e1c10183e390c6d0897e81f84a5df4590b02d84
a plugin to view lastfm scrobbler logs
uses print cell to give a spreadsheet view of scrobbler logs
buffers the whole file if possible otherwise it reads entries from disk
rudimentary text searching for columns
include / exclude; all/any and case sensitive
Change-Id: Id9616e5796658952fba4ea747f596cb77d6f34c0
When displaying Track Info for multiple tracks,
the value for combined file sizes or length was
capped at 2 GiB / ~596h.
Limit has been raised by a factor of 1000.
Change-Id: I942c64e81864cba3f719c83a24912883fafeb70e
- get_albumart_for_index_from_db:
slide_index not allowed to be -1
- Fix visual glitch changing scroll direction:
An errant frame was displayed when switching
from one scroll direction immediately to the other
- Rename/combine functions and replace
magic numbers for zoom animation
- iPods: Fix inability to skip or reverse zoom animation
using Back button
Change-Id: I888c3c437bb123325813bf79be56b320cfbfda0c
When data is not in packed-16-bit mode, flushing the fifo
may result in swapping left and right channels if there
happens to be an odd number of entries in the FIFO.
This is especially likely when switching sample frequencies
for some reason.
When stopping PCM DMA, disable DMA and Underrun Interrupts
and then wait for FIFO to be empty before stopping AIC's playback.
Change-Id: I45b6b022c9e3889627842663cd9b7d2e0affb7c6
opx shortcuts allow an easy way to add parameters to plugins
you should be able to set them as shortcuts now
the plugin hd a use after free bug on the dat file file descriptor
Change-Id: I5fe3b0628e7da003c03b9b54f5788b0d0dbfea74
save tsr plugin path for later
resume tsr plugin when user stops the interrupting plugin
expand return of tsr_exit function to allow
continue, suspend, terminate
tsr plugins check parameter at start to determine if
the plugin is being resumed
Change-Id: I6fc70de664c7771e7dbc9a1af7a831e7b50b1d15
Control cache entries cost 24 bytes per command, but cacheable
commands are always less than that when written out to the file.
We can actually cache *more* data by writing commands directly
to the fd (native Rockbox has a 512-byte cache per fd) and it's
much simpler.
Change-Id: Ibb1b9ffa56ef17431b281419a04082e14d0cbd85
The intent behind the check is to only do this if the filename
comes from an on-disk file. With the RAM buffer is gone this is
now "always" the case.
Change-Id: I4c26e32c482dde176b69a6071e9562e9955d7171
This checks for an out of memory condition when the filenames
were buffered in RAM. The playlist no longer buffers anything
in RAM so the code is unreachable.
Change-Id: I4c2d0b2ac4ba3b7e7c94fe554a0cf7cb1a798e09
Use the playlist control file for directory playback instead of
storing filenames in RAM. The implementation of that feature is
very iffy and probably responsible for random crashes that may
occur when skipping through directories rapidly.
Change-Id: I3863940cd4542410d8046a3ca47508b5d97309a1
GCC 4.9 always emits assembly with divided syntax. Setting unified
syntax in inline assembly causes the assembler to complain about
GCC's generated code, because the directive extends past the scope
of the inline asm. Fix this by setting divided mode at the end of
the inline assembly block.
The assembler directives are hidden behind macros because later
versions of GCC won't need this workaround: they can be told to
use the unified syntax with -masm-syntax-unified.
Change-Id: Ic09e729e5bbb6fd44d08dac348daf6f55c75d7d8
Any modifications to the playlist (insert, delete, shuffle, etc)
will cause the modified flag to be set. The flag is cleared when
the playlist is saved. Code that generates playlists can manually
clear the modified flag if appropriate; there is now a proper API
for this so the tagcache and pictureflow don't need to resort to
hacks.
Change-Id: I8d3c723265a41db07a13de3f1d2abb0444528d57
Replace the minimum version bound with a check on the size of
the API struct. The version only needs to be incremented for
ABI breaking changes. Additions to the API won't need to touch
the version number, resulting in fewer merge conflicts.
Change-Id: I916a04a7bf5890dcf5d615ce30087643165f8e1f
Make 4.3 and newer doesn't interpret comments in a shell call,
whereas Make 4.2 and older do. Escaping the comment directly
works on old makes, but on new makes the backslash is passed
as well -- which we want to avoid.
The safe way to pass a literal "#" character to the shell on
both versions is by embedding it in a variable and expanding
that. It's ugly, but it works...
Change-Id: I1a217c42d747fd5aa83f9990c234e06966ac1a00
GNU grep 3.8 warns about "\#" when compiling. The "#" has to be
escaped for Makefiles and shell, but it isn't special to grep so
grep thinks it's an invalid escape sequence.
Change-Id: Ieb2607d42b7daa1939ca0101b915ec0c7afd4298
Enabling locked actions for all softlock targets accidentally
broke keylock on touchscreens because the generic touchscreen
keymap was missed. Trying to lookup CONTEXT_WPS|CONTEXT_LOCKED
returned the mapping for CONTEXT_STD because the locked version
wasn't explicitly handled.
But on almost all cases, a context's keymap does not change when
the screen is locked. It makes more sense to mask out the locked
flag and only check for it where needed.
Change-Id: I65cda2de82950d272d4394fd772286699e7c3779
