On PATA, we'd cap our transers at UDMA2 if the device reported that an
80 pin cable wasn't detected, but SATA devices do not perform this test.
So alter the check to only apply on PATA devices, so that SATA devices
can run at full UDMA speeds.
Change-Id: Id7aa25f2a702c0af73d707395439d69da1e04719
characters less than the first diacritic in the symbol table (0x300)
return false after checking the MRU table
we gain some performance by eliding the function call all together if less than first diacritic
Change-Id: I02c14e350eb168eca808523affad443cd43888b4
remove nvram and use the existing settings framework for it
add a crc check to the user_settings data to see if we need to save
the user setting file or if we can just save the status file (resume.cfg)
move volume to the system_status struct so we don't write the whole settings file
over volume changes
allow user to still export volume with save sound settings
allow the user to also export pitch and speed
name the file .resume.cfg
Rename all the SYSTEM_STATUS save file variables to TLAs to save space and
discourage tinkering
Cleanup DEBUG_AVAIL_SETTINGS output
when saving user_settings it calls status_save as well this cause the resume
file to be written twice. instead remove the callback for status_save
when setting_save is called
remove header text when saving .resume.cfg
convert status_save() to status_save(bool force)
add SYSTEM_STATUS_UPDATE_TICKS
for ATA device set this to 5 minutes
since we arlready wait for the disk to be up before saving
we don't want to miss our window
for all other every 15 minutes
that way if the battery is too low by the time shutdown comes around you
don't lose much progress
Change-Id: I27214ffd6e5d5494ee5ca83b14f04a41ba426ad7
Use AXP2101's egauge battery percent level if available (hw4 units).
If not available (_battery_level() will return -1 on hw1-hw3 units),
fall back to voltage battery level.
Also fix logic in axp2101_battery_status()
Change-Id: Ic300418532dae6f7772fff8bf5e2b32516f3b973
if you run yesno screens back to back or another screen with
wait_for_release you may never see the release
instead clear anything in the queue but release events
Change-Id: I1b1e42cbb44f2fdfed441ab1f217b6ea4fe07492
Improve ADC behavior, add correct conversion
for die temperature, add readout of EGauge to debug screen.
Add delay to give ADC time to get a good reading. Initial reading
may still be 1-3% optimistic, which is probably ok.
EGauge appears to be pretty good, but no great need to convert
to using it either.
Change-Id: Iab9d50e2427f656ad0875f9a623b03545fc69881
This enables smooth resizing of the window using a
fixed aspect ratio, instead of snapping into the
correct aspect ratio only when the resize operation
has finished, by using an SDL event filter that gets
events delivered during the resize operation
(whereas SDL_PollEvent blocks until done on macOS).
Change-Id: Ie6614e4b6f49a24469c5ee6a69721c9fbd440dae
No need to create a new texture for every
rendered frame, unless the scaling method
has been adjusted.
We also don't need to upload the (unchanged)
player interface to GPU memory repeatedly.
+ Remove unused lcd_display_redraw &
having_new_lcd variables
Change-Id: I5bff6aa2d54347a3f2c3afba8d8d7eb9e39f77f7
Tested on Linux, MacOS, and Windows.
On MacOS and Windows, we constrain the window's aspect
ratio by adjusting the size when responding to resize
events.
On Linux, I've not found a way to do so, that doesn't
result in fairly stuttery behavior and weird jumpy
behavior of the resize handle, possibly depending
on your window manager. So, black bars are displayed
around the content.
Maybe someone, at some point, finds a way.
(SDL3 seems to have SDL_SetWindowAspectRatio)
When the window is in fullscreen, black bars are
display necessarily, of course, on all systems,
unless the player GUI has exactly the same aspect
ratio as the screen...
Change-Id: I535e6617497611ea57a4c19e08e552f990859cfe
in testing this slightly speeds up any of the draw routines using fg or bg
and slightly slows down the ones that don't
being that these are already faster it serves to bring them closer in
ops per second
.. and slightly speeds up the most general case DRMODE_SOLID
Change-Id: I4ea4898354e8bc9dfa7e96f188108ca77ea0aca9
we calculate the size of the scrolling line every scroll update
the line data doesn't ever change without a call to update
instead calculate the string size when the line is updated
re-use this value while scrolling the line
in theory the current font could change but in practice
I can't find a case when the string size needs to be updated
that the scroll engine isn't already updating the line
if this is later an issue we can recalculate the stringsize each time
the line is finished scrolling which still saves quite a few calculations
Other:
I don't think we have any targets with pixel counts exceeding 65535 pixels
where unsigned short stringsize will be an issue
Change-Id: I83d6374377ed648d9a320d4fd69f9d6a17095b0c
unless lcd_update() is called the sim doesn't update scrolling
you CANNOT call it from the scroll thread its simply ignored
I suspect this has something to do with where the call to render
originates as thi is the only thing I can think of besides
a call to disable the render
see demos/rb_info > paths -- observe the lack of scrolling
see any menu in a plugin that exceeds screen width
Change-Id: Ic14dee4a34de29479d739e6a280d6cf1cc283719
Turns out they did not swap the stereo switch lines on us,
so this setting isn't really useful. Repurpose it to allow
the user to force the device into headphone or line output mode.
Note that settings other than "Automatic" will cause jack detection
to not work.
Default Automatic.
Change-Id: If94499fbb8f0d92111811fe51f6107ce213b66ad
Check for the ES9018K2M dac in the bootloader for
hw1/hw2 devices. Assume that all devices newer than
hw2 have ES9018K2M DAC unconditionally.
All devices will now report the correct hw revision
in the debug menu under Device Data.
Add devicedata.version field, current version 0.
Rename device_data.lcd_version to device_data.hw_rev.
hw2 devices with older bootloaders which ID as hw1 are special-
cased to keep using hwvol on them. They should still upgrade though.
Change-Id: If0fd5ce3bc6e85e511047721ec18e42fb89312e7
Event handling must happen on the main
thread for MacOS.
Not sure if button_queue_wait is the
best place for doing the SDL event
polling, but seems to work ok.
Change-Id: If928282df84bdd74e24a48afd7dbc4c4bfcc49e2
* Make the partial sector logic a little clearer (no functional change)
* Corrections for debugging messages
* Also use MAX_VIRT_SECTOR_SIZE in BOUNCE_BUFFER calculations
Change-Id: I89363824b092b2e3bddd5e0f75bf81200c9bc513
* if MAX_VARIABLE_LOG_SECTOR is defined, use that
* if MAX_PHYS_SECTOR_SIZE is defined, use that
* use SECTOR_SIZE
Change-Id: Ia314f590fdb300e99df2c60587657c9563b9ce80
* If the device returns a valid upper limit (word 47) use that.
* If the current limit (word 59) is valid, use that.
* Fall back to 1 as a default, not 16!
Note this is only used for PIO transfers!
Change-Id: I269b751466242bbcce91ee991d9ade449cc84b6b
Support hw4 units with AXP2101 PMU
Bootloader successfully compiles and loads onto device.
The LCD appears to be identical to hw3 units.
Scroll wheel and buttons work
Audio output works, including volume.
HP/LO detect works
Rockbox build is generic
GPIO gating logic seems to be working as intended now.
- Added new GPIO definitions - some significant overlaps with pins
from previous hardware revisions...
- Added some GPIO definitions for older players we didn't know about
- Add register definitions for AXP2101 from datasheet
(these are very different from AXP192!)
- Add AXP2101 regulator definitions, need to support multiple step
sizes per regulator.
- Verify AXP2101 voltage set multi-range logic
- Verify AXP2101 voltage get multi-range logic
- Make AXP2101 its own driver
- AXP2101 driver should be "minimally viable", though I think
there is some extra functionality that could be implemented.
- Disabling the coulomb counter stuff - we could maybe make
the E-Gauge work for the same purpose, but it only appears to
be used on the debug screen at the moment so it doesn't seem
like it's worth the effort.
- Found new button GPIOs
- Found error in my GPIO setting logic, blue light works now!
- Set LDO/DCDC output voltages to OF's settings, as far as
I can tell.
- Determined we probably want TCS1421_CFG1:0 to be 0x00,
for UFP behavior
- Tested this rb build with both old and new bootloaders on hw1.5,
hw2, hw4 in as many configurations as I can think of, works across
the board.
- Bootloader can install itself on hw4, so nand chip isn't novel
- Uninstallation file can be made by patcher script, works on hw4
- Installation file can be made by patcher script, works on hw4
- Added HW4 to rbutil, manual
Change-Id: I5b75782273e81c2c6f2b9c79501c8b7cbf88391f
Normally, if a device uses larger physical sector size than the logical
size and supports so-called "512e" mode, we let the device deal with
partial sector reads/writes.
However, if MAX_VIRT_SECTOR_SIZE is defined, we support
partitioning/filesystems that use a larger "virtual" sector than the
logical sector. typically this matches the physical sector size of the
drive, which means that despite a small logical sector size, all I/O
is done in terms of the physical sector size.
Therefore, when MAX_VIRT_SECTOR_SIZE and MAX_PHYS_SECTOR_SIZE are
enabled (currently only ipod5g and ipod6g), prefer software-based
partial sector I/O.
Change-Id: I0815ad0a2f987b89bb2debfbf3d0ed64cdf85525
If identify device word 209 is valid, check to see if the drive report
on the alignment of LBA0 with respect to physical sector 0.
If it's not aligned, bail immediately. Supporting this properly
won't be hard, but it's not someting we want to do unless necessary.
Change-Id: I3d5bb8fad9e32fff43dfb6454393728d7c01b93b
This isn't strictly needed for FAT32, but the core file cache code
needs to be able to reference >32bit sector addresses.
Change-Id: I57838f1228c1d45f6a8c4755c5d1f9063c13b3dd
When enabled this allows 512n and 4Kn drives to be used with a single build.
(So far all ATA SSDs use 512 byte logical sectors)
Change-Id: I902d2318ca8abb581699c0bca68d6e3ec227d064
Only used if MAX_LOG_SECTOR_SIZE is defined
This allows a single build to seamlessly work with (eg) 512B or 4K sectors.
Change-Id: I85d2a6612afca6a1d7a3bd49c588b5745ab2b220
Basically this requires un-hardcoding SECTOR_SIZE everywhere, in
favor of using a variable containing what was reported in IDENTIFY INFO.
Note that the rest of the storage subsystem still needs to be fixed up!
Change-Id: I7f2dbd54ff2bc16b15010721e011949cbf308e12
This lets us *natively* handle varying physical sector sizes
without playing games and lying about the logical sector size.
(The original drives use 4K _physical_ sectors with 512B logical
sectors, but you have to access everything in 4K blocks...)
Achieve this by splitting the MAX_PHYS_SECTOR_SIZE code out
of the main ATA driver and re-using it.
Change-Id: I0bc615ab4562f1e3e83171a8633c74fb60c7da1f
....It's specified in 16-bit words, not bytes. So multiply it by 2.
(This hasn't been a problem in practice as everything uses 512B logical
sectors so far..)
Change-Id: I0b1abd0f6184330f0b7f5c000c5ad547038f7c95
* FLUSH_EXT is used if featureflag is set and we are using LBA48
(unconditionally used for CE-ATA on ipod6g)
* FLUSH is used if featureflag is set (ATA6+) or if device claims to be ATA5+
* Rename ata_disk_can_power_off() to ata_disk_can_sleep() as that is
what it actually tests for. Only use it to gate issuing the
STANDBY IMMEDIATE command.
* Restore behavior of ata_disk_is_active() to return 1 if drive is
"spinning" or powered up.
* Allow poweroff if drive claims PM support OR we are able to issue
FLUSH/FLUSH_EXT commands.
* Added ata_flush() to explicitly trigger a flush operation, and hook it
up to storage_flush() in the device shutdown path. (Flushes were
only previously used in the storage device power management path)
* After issuing all settings, re-issue IDENTIFY_DEVICE to make sure
it reflects everything we've enabled.
* Update manual section on Flash/SSD mods.
Change-Id: I6770a54ef3a87f4c47120bcb96c944a6652f1bf4
This information is available in the RDA5802N or RDA5807 datasheet. I suspect
that register SYSCONFIG6 is not actually used by the tuner. Based on an
obscure statement in the datasheet, I guess SYSCONFIG5 is a frequency in Khz
called freq_direct that offsets the lower band frequency (56/67 MHz) but
1) it is not enabled (FREQ_MODE=0) and 2) it is not clear if this offset is
really just to avoid interference (ie the RF and IF are simply shifted by this
amount) or an actual offset.
Change-Id: Ia469f5370aee7c8c3390324d0cef1193c64df755
The current code relies on the initial value of some RDS register(which is
documented) but this assumption will be broken when we start enabling RDS.
This commit changes the code to really read the chip ID, although it is more
involved.
Change-Id: I0ed630322a94523612d2f0297dbcbea5f869eb6d
ipod6g was configured with SECTOR_SIZE of 4096, but this ATA driver
unconditionally translated these to 512B operations on the actual
storage device.
Rockbox's storage layer already has robust support for "logical sectors
larger than physical storage sectors" through use of
MAX_LOG_SECTOR_SIZE. So switch to that mechanism, allowing the ipod6g
ATA driver to be simplified.
If we want to support drives with physical sector sizes > 512B, then
we need to port the MAX_PHYS_SECTOR_SIZE logic from the primary ATA driver.
Additional changes:
* Simplify MWDMA/UDMA selection logic
* Report CE-ATA mode in debug menu
* Use LBA48 commands only if drive is over 128GiB.
* Drop default sleep/poweroff time from 20s to 7s (matching main ATA driver)
Finally, the bulk of the changes are the first phase of a badly needed
style cleanup that made reading this driver a lot harder than it should
be. I intend to split this into a separate patch.
Change-Id: I2feca9fd319c8d6cfb3c2610208970428d2fa947
