Setting the cache later can result in the system trying to create cache files
in the current working directory, since an empty cache path will be treated as
current directory. If this happens set the system temporary path.
Change-Id: I1623330013d387a966015edd0e1d278c922d9475
HttpGet used to use QHttp which has been deprecated since a while and has been
removed from Qt5. Rewrite the class based on QNetworkAccessManager which is the
recommended way these days.
Change-Id: I4902309c433a85ec18e157ef3a9f5e60fd0f4b1f
The Themes Window reuses the download object for getting preview images. Since
the object is reused it isn't necessary to set the cache path on each network
request.
Change-Id: Ibc531e09ad19ede3cab7326d9230bd5188710d98
Added three bitmaps with the new 70x14 graphics and added link to them in /apps/plugins/bitmaps/native/SOURCES
Change-Id: I1f4b9e3504011b80a7cd40e0fcd71ba8fd100424
Reviewed-on: http://gerrit.rockbox.org/389
Reviewed-by: Jonathan Gordon <rockbox@jdgordon.info>
Tested-by: Jonathan Gordon <rockbox@jdgordon.info>
The build of bootloader with HAVE_ATA_DMA fails because of missing
commit_discard_dcache(). This seems to be bigger problem as
bootloader builds don't call cache_init() also which seems wrong.
but I know too little about the PP platform to judge.
For now use ATA DMA only in regular builds.
Change-Id: I82873cb1771f5a95ebfbef91ce26744e3abd743c
PP502x ATA DMA was disabled in 55fab77 because it caused various
instability on all targets using it. This instability is fixed in
583c948 via FS#12391. Here, ATA DMA is enabled for all PP502x targets
except nano1g and hdd6330. For nano1g, there may be other causes of
instability and more testing is needed. ATA DMA does not work on hdd6330.
Change-Id: I786b9edb19e74e6eb957ab205ea026f0969200ac
Reviewed-on: http://gerrit.rockbox.org/340
Tested-by: David Hooper <dave@beermex.com>
Tested-by: Boris Gjenero <boris.gjenero@gmail.com>
Reviewed-by: Szymon Dziok <b0hoon@o2.pl>
Tested-by: Szymon Dziok <b0hoon@o2.pl>
Reviewed-by: Marcin Bukat <marcin.bukat@gmail.com>
In commit_discard_idcache(), cache lines were marked as invalid. When
some cache lines are marked as invalid, memory corruption can occur.
This caused instability when using PP502x ATA DMA because of the many
more calls to that function. Here, commit_discard_idcache() is changed
to avoid the problem. Also, the cache is filled after being enabled to
to ensure there are never any cache lines that aren't marked as valid.
Change-Id: Ia26300acef6b0573c1f40299c496ee5cbda3dac8
Reviewed-on: http://gerrit.rockbox.org/339
Reviewed-by: Szymon Dziok <b0hoon@o2.pl>
Tested-by: Szymon Dziok <b0hoon@o2.pl>
Reviewed-by: Marcin Bukat <marcin.bukat@gmail.com>
RFC850 timeformat isn't y2k compliant. Furthermore, parsing turned out to be
broken and it doesn't seem supporting the format is necessary anyway.
Change-Id: I062e636a9e016ff1ac49d2a55f00adb1af182576
If a file is not available in the cache immediately sent a GET request. Using a
HEAD request to retrieve the file timestamp on the server is not necessary and
only creates an unnecessary network request.
Change-Id: I358507dcc0c6b837ff47e5fd710b5262d03cb7b0
Speeds up decoding of 128k opus files by 1.2MHz on AMSv2. Rounding
error is 1 bit due to KissFFT using a 15 bit shift instead of a 16 bit shift.
Also, change an LDMIA in the armv4 code to LDM as the pointer should not
increment.
Change-Id: I626a207c6a056a1984e33cfe89415c35d0caed93
Reviewed-on: http://gerrit.rockbox.org/377
Reviewed-by: Michael Giacomelli <giac2000@hotmail.com>
Tested-by: Michael Giacomelli <giac2000@hotmail.com>
This will allow the user to set the file colour and icon for all
unknown files in the file browser.
Change-Id: I4e601f9d2ea31ac972d8066252bf80b6c0d22268
Sometimes (and perhaps only on some devices) Android's YesNo dialog
loses focus and is put to the background. Since the native code waits on the
result (which is then impossible to happen) it would become impossible
to further control Rockbox. This is an attempt to fix as I cannot reproduce
the problem on my device(s).
Change-Id: Iff849ff4be5e8f41922fb7d36491d860486c6441
There were some bugs, especially when the user scrolled above the
list viewport. One bug made Rockbox completely unusable once triggered.
Change-Id: I9bb4722ff4381db189058e9a19ea30b2c69e87d9
readlink() might return the full size of the target buffer
and we write a '\0' to the returned length offset.
cppecheck reported:
[rockbox/apps/tagcache.c:4335]: (warning, inconclusive) readlink() might return the full size of 'target'. Lower the supplied size by one.
(the check was actually written by me)
Change-Id: Ibb42f732aa42c38bb6cb92cdccd3e6a0d3aa9b9f
The SD driver doesn't initialised drives at the beginning but
upon request to handle removable drives. Since means that the
init should call init_drive() and not init_sd_card() otherwise
the check for WINDOW flag is bypasses. This breaks the zenxfi3
bootloader and has been overlooked for some time.
Change-Id: I7325f7164d16d7e7e54eeb4645e98517a08e0836
AA was broken when an RTL language is used, causing it to not
be displayed.
Also fix th out of bounds handling when width/height > viewport
Change-Id: I4899cd32ec58107c987e3cc0e8df582963bdcf62
Offline Mode was intended to allow performing an installation without network
access. However, to get the required files cached the same installation has to
be performed with network access, which is a rather strange prerequisite. A
better way would be a way to direct Rockbox Utility to some local "repository"
that holds the required files.
Furthermore, Offline Mode hasn't been tested since long and is likely to be
broken since the caching mechanism has been extended.
For now remove this functionality. As far as I know it's been rarely used (if
at all) anyway.
Change-Id: Ib2af4892708e0440bd0a7940c131f04182ddb39a
This function returns an internal enum value of QHttp (which is also
deprecated). It was only used for showing an error to the user / showing it in
the system trace. Since it is an enum value it doesn't have much value. Log /
show the error string instead.
Change-Id: I54b9b6026969f8108f779b02a04477f0ad9201ab
Move to a table based approach (scales better) and distinguish
between upward changes (increase frequency) and downward changes
(decrease frequency). This provides a better ordering of
operations and in particular it allows to avoid changing the
regulator while running at low speed since it takes a long time !
This should result in a much smoother scaling.
Change-Id: Iad7e5b61277e215f31c07877fbbad07ddde1171f
For some reason it is the responsability of the driver to send
this event so do it. This might fix some non-updating screens.
Change-Id: Ib5fdc94bf266c3497a8ac4e89d0418c0e876ff9f
The lcd kind is always set to st7783 in case we can't read the ID
so don't bother handling impossible cases
Change-Id: I352fd43b26068b460e69190d37c4cd4627e1db9a
The flip and invert settings can potentially be reset to their
value accross a disable/enable cycle, so save the value of the
impacted registers and apply it after each enable. Also avoid
poking registers when the lcd is not on.
Change-Id: Ica98f166c060aade7eb205f5628b58aae692024f
