summaryrefslogtreecommitdiff
path: root/src/drivers
Commit message (Collapse)AuthorAge
* altos: Make sure the packet format is set reasonably for radio testKeith Packard2013-04-12
| | | | | | Dunno if this matters, but it might as well be set reasonably Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Don't include bufio debug commands by defaultKeith Packard2013-04-12
| | | | | | We shouldn't need these Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Stick a mutex around FAT operationsKeith Packard2013-04-12
| | | | | | This allows the command line and logging operations to occur safely in parallel Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Delay while waking up SD card a bitKeith Packard2013-04-12
| | | | | | | | This seems to make bringing the card from idle to ready mode more reliable. If you spam the card with requests, it will eventually whinge and shut down communications. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Create telebt-v1.0 product. Remove old telebt productsKeith Packard2013-04-08
| | | | Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Try RDF mode for TX calibrationKeith Packard2013-04-06
| | | | | | | | Trying to get the radio to stop modulating the carrier when calibrating the radio, we'll try RDF mode which says no preamble or sync data. This might shift the frequency though? Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Shorten SD initialization timeouts.Keith Packard2013-04-02
| | | | | | This makes failure when no card is present much quicker. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Horrible kludge -- disable radio while talking with SD cardKeith Packard2013-04-01
| | | | | | | The SD card really doesn't like the RFI generated by our enormous radio, so just lock the radio out while working with the card. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Retry SD card I/O. Use time for timeouts instead of countsKeith Packard2013-04-01
| | | | | | | | | | Sometimes I/O operations may fail; give the card a chance and retry the operation in case it works the next time. Replace the loop counts with loops that check the clock so that they'll have consistent timeouts even if the CPU or SPI speed changes. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Support open on multiple simultaneous FAT filesKeith Packard2013-04-01
| | | | | | | Need to be able to see the contents of a log file, even if the logger is running. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Make sure FAT cluster allocation works for size zero filesKeith Packard2013-04-01
| | | | | | | There were some rounding errors mis-computing the number of clusters needed, and the logic to figure out how to re-connect a chain was broken. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: let FAT tracing work in ao_fat_test as neededKeith Packard2013-04-01
| | | | | | | This allows the FAT DBG hooks to be enabled even if some other module turned DBG off. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Flush the on-board mega log after every sample interval.Keith Packard2013-04-01
| | | | | | | | SPI flash parts don't need flushing, but the SD card does. Make sure the SD card contents are sane after every logging interval has passed by flushing all dirty blocks to the device. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Create the log file if it doesn't already existKeith Packard2013-03-31
| | | | | | open will return failure unless the file already exists. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Make ao_fat_readdir return real error values instead of 1/0Keith Packard2013-03-31
| | | | | | | This way, we can distinguish between 'something bad happened' and 'you're at the end of the directory'. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Increase SD card timeout at startup timeKeith Packard2013-03-31
| | | | | | | Sometimes the SD card takes 'a while' to go into idle mode at first power up. Just hang around waiting for a long time. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Fix command-line FAT filename parsingKeith Packard2013-03-31
| | | | | | Pad extension with spaces Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Unmount file system after each testing pass in ao_fat_testKeith Packard2013-03-31
| | | | | | Otherwise, we use stale data and 'bad things' happen. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Don't add fat commands when building ao_fat_testKeith Packard2013-03-31
| | | | Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Do not release interrupts from any pollchar functionKeith Packard2013-03-31
| | | | | | | | | | | getchar relies on interrupts being blocked across the pollchar calls and into the sleep call or it may go to sleep with data pending. This prefixes all pollchar functions with _ to indicate that they are to be called with interrupts blocked and eliminates all interrupt manipulation calls from within the pollchar functions. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Add sdcard read/write tracingKeith Packard2013-03-31
| | | | | | This just dumps info in trace mode about read and write commands Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Move fat mount information to separate command.Keith Packard2013-03-31
| | | | | | | This makes the mount report precise error information and then prints that with the 'M' command. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Disable CC115L debug commandsKeith Packard2013-03-31
| | | | | | now that it appears to work, leave these disabled by default Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Add SD card writing functionKeith Packard2013-03-31
| | | | | | | Now that the FAT code seems to be operational, go back and add SD writing. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Add logging and telem to telegpsKeith Packard2013-03-30
| | | | | | | | | | | | This turns on telemetry, APRS, RDF and data logging for telegps. Data is logged as soon as GPS has a date to create the right filename, using files of the form YYYYMMDD.LOG which just barely fits in a FAT filename. Telemetry/RDF/APRS are all separately controllable. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Export ao_fat_sync and ao_fat_full functionsKeith Packard2013-03-30
| | | | | | | ao_fat_sync() flushes the bufio data to disk along with any fsinfo changes. ao_fat_full() returns whether the file system is full. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Lock cc115l radio mutex when using global radio valuesKeith Packard2013-03-30
| | | | | | | This moves the locking up above the global state variable uses so that multiple radio users (as if we had any) won't collide. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Configure cc115l sync byte count for each radio modeKeith Packard2013-03-30
| | | | | | two sync bytes for packet mode, disable sync for rdf/aprs mode. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Get CC115L radio working.Keith Packard2013-03-29
| | | | | | | | | | | | | | | | This involved figuring out which GPIO signal would reliably indicate that the transmitter was finished; I ended up using the PA_PD bit for this. This also converts all of the radio users to the long packet support as the CC115L has only a 64-byte fifo, not large enough to hold either an RDF tone or a regular AltOS telemetry packet. This also renames the public API for sending APRS packets from ao_radio_send_lots to ao_radio_send_aprs, which is at least more accurate. The workings of that API haven't changed, just the name. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Add temporary RF power settingsKeith Packard2013-03-29
| | | | | | | | | | These expose the raw cc115l and rfpa0133 register settings so that we can calibrate them against measured power outputs. I've tested them to verify that they change how much power the board consumes, so they're clearly doing something... Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Add FAT32 support. And lots more testing.Keith Packard2013-03-29
| | | | | | | | Generalizes the FAT code to deal with either 16-bit or 32-bit versions. The testing code now runs over a variety of disk images to check for compatibility on all of them. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: seek forward on FAT cluster chain instead of restartingKeith Packard2013-03-28
| | | | | | | | This improves sequential file performance by taking advantage of any previous cached cluster/offset pair and starting from there when the cluster changes rather than starting from scratch at the begining again. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Add a simple cache for the FAT position->cluster computationKeith Packard2013-03-28
| | | | | | | This improves read/write performance with large files by not re-walking the cluster chain for every operation Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Clean up fat driver API. Improve fat testKeith Packard2013-03-28
| | | | | | | | Make FAT api provide reasonable error return values, change the tests to write and then read a pile of files, checking that the contents are correct (using md5sum). Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Add sanity checking to busy counts in bufio driverKeith Packard2013-03-28
| | | | | | Make sure the busy counts don't underflow or overflow. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Add SDCARD and FAT16 filesystem supportKeith Packard2013-03-27
| | | | | | | This adds a fairly primitive FAT16 file system implementation along with support for SD cards. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Improve CC115L driver. Generates carrier now.Keith Packard2013-03-26
| | | | | | Still no data, but at least the carrier comes up on frequency now. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Add RFPA0133 amplifier driverKeith Packard2013-03-26
| | | | | | No configuration of power level yet, just the bare driver. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Add cc115l driver (untested)Keith Packard2013-03-24
| | | | | | Includes support for sending telemetry, RDF and APRS tones Signed-off-by: Keith Packard <keithp@keithp.com>
* altos/driver: Make HMC5883 driver build againKeith Packard2013-02-24
| | | | | | Adapt to changes in OS interfaces Signed-off-by: Keith Packard <keithp@keithp.com>
* Merge branch 'telescience-v0.2'Keith Packard2013-02-24
|\
| * altos: Change CC1120 SPI speed to 4MHz.Keith Packard2013-01-13
| | | | | | | | | | | | | | | | Most of the chip can run at 8MHz, but extended register access is limited to 6.1MHz. Instead of pushing things, just run the SPI bus at 4MHz. Signed-off-by: Keith Packard <keithp@keithp.com>
* | altos: Document which MPU6000 revs have broken accel valuesKeith Packard2013-02-04
|/ | | | | | | | | From Tridge -- MPU6000 rev C4 and C5 are broken, having accelerometer values in the wrong range. This commit just adds comments which note this; experimentation will be required to actually sort out what's going on. Signed-off-by: Keith Packard <keithp@keithp.com>
* Merge remote-tracking branch 'origin/micropeak-logging'Keith Packard2013-01-10
|\
| * Allow CC1120 to sit on other SPI bussesKeith Packard2012-12-31
| | | | | | | | | | | | | | | | Reading the incoming data bypasses the SPI API and touches the SPI data register directly; which port that is needs to be specified in the pins file Signed-off-by: Keith Packard <keithp@keithp.com>
* | altos: Time out reading packet data from cc1120 after 100msKeith Packard2013-01-10
|/ | | | | | | | | | Sometimes the radio will give a spurious wakeup indicating that a preamble seems to have arrived, but no packet data will appear. In this case, abandon the packet reception and go back to waiting for a preamble again. This releases the SPI bus for other users and also avoids missing packets. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Fix cc1120 debug code to build on megadongleKeith Packard2012-12-28
| | | | | | RDF function had changed, and APRS isn't available on megadongle. Signed-off-by: Keith Packard <keithp@gag.com>
* altos: Wire up another CC1120 GPIO to get MARC status changesKeith Packard2012-12-16
| | | | | | | | | | | | | When the radio drops out of RX or TX mode due to an error, it changes the MARC status, and sends pulse down a configured GPIO. Use this to tell when something 'bad' happened during TX or RX so that we can recover from losing the SPI bus in the middle of transmission or reception. Without this, the radio would change state and we'd never know, leaving the radio code waiting for an interrupt that would never arrive. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Document what HAS_BOOT_RADIO does in the m25 driverKeith Packard2012-12-16
| | | | | | | HAS_BOOT_RADIO causes the m25 driver to abort any ongoing receive in case that is holding the SPI bus. Signed-off-by: Keith Packard <keithp@keithp.com>
* altos: Round APRS data correctlyKeith Packard2012-12-16
| | | | | | | Apply rounding once at the start of the computation, then truncate after that. Signed-off-by: Keith Packard <keithp@keithp.com>
generated by cgit v1.2.3 (git 2.39.1) at 2025-12-27 14:57:36 +0000