Commit Graph

1048 Commits

Author SHA1 Message Date
hathach c14c0cfc9f support class drivers implemented by application 2020-04-29 10:49:58 +07:00
hathach fcdb22b2f9 fix typo 2020-04-28 10:53:43 +07:00
hathach 958b5510cb added comment for hw clearing TXFE 2020-04-27 13:17:47 +07:00
hathach e785b09118 TXFE is read only bit 2020-04-27 12:06:14 +07:00
hathach d0487088ac revert a change to net driver 2020-04-26 23:04:17 +07:00
hathach 00fcf829a1 sync synopsis fix for esp32s2 2020-04-26 22:41:04 +07:00
hathach 8d18d6077b turn off TX FIFO Empty for EPIN if all bytes are written
fix dcd synopsys issue with usbnet #289
2020-04-26 22:14:59 +07:00
hathach 83353dd93f add TODO for usbnet clean up 2020-04-26 22:03:05 +07:00
hathach a1c599f4b6 clean up log message 2020-04-26 22:02:49 +07:00
hathach 9c8517a8d2 fix missing debug log macros 2020-04-26 14:58:23 +07:00
hathach 017c95037f add usbd edpt open
- RTT mode is blocking to prevent log lost
- Improve logging message
2020-04-26 14:51:44 +07:00
Jerzy Kasenberg b949ae596f synopsys: Reduce interrupt time for IN ZLP
For IN endpoints output FIFO is filled in interrupt, therefor before
endpoint is enabled, DIEPTSIZ is set with correct size of packet.
Then endpoint is enabled and FIFO empty interrupt is enabled.

This works fine except for the ZLP. Enabling FIFO empty interrupt
results in interrupt handler being called all the time because
there is nothing to put in the FIFO.
Eventually it ends when IN token is received and empty
packed is transmitted out.

This change does not enable FIFO empty interrupt for ZLP reducing
CPU load.
2020-04-24 17:38:30 +02:00
hathach c59fa77427 Revert "Merge pull request #359 from versioduo/midi-flow-control"
This reverts commit 1d33aa9b6f, reversing
changes made to 718db7e536.
2020-04-23 23:25:20 +07:00
Ha Thach f8bb4c86ef
Merge pull request #376 from hathach/add-more-example
Add dynamic configuration example
2020-04-23 14:50:27 +07:00
Ha Thach c39dc03331
Merge pull request #374 from duempel/stm32f1_synopsys_support
Added support for STM32F1 Connectivity Line MCU
2020-04-23 13:38:48 +07:00
hathach 10e035241f
house keeping 2020-04-22 23:04:21 +07:00
Jan Dümpelmann 551724498a Added support for STM32F1 Connectivity Line MCU
STM32F105 and STM32F107 are using the Synopsys IP
2020-04-22 17:39:53 +02:00
hathach 7acdcc2ebc Merge branch 'master' into add-more-example 2020-04-22 19:50:23 +07:00
hathach 1fc7f54a8a
added swo as logger
tested with feather nrf52840 + jlink
2020-04-22 19:18:03 +07:00
hathach 3aa3c35986 Merge branch 'master' into add-rtt 2020-04-22 13:45:20 +07:00
Ha Thach 379b8ac556
Merge pull request #365 from pigrew/desc_tu_verify
tu_verify for getting descriptors
2020-04-22 12:24:05 +07:00
hathach 3b83813f01 clean up 2020-04-22 00:29:51 +07:00
Ha Thach 1d33aa9b6f
Merge pull request #359 from versioduo/midi-flow-control
MIDI - Add flow control to incoming packet stream
2020-04-22 00:11:27 +07:00
Nathan Conrad 0ec69de77c sof is optional, revert other changes but remove unneeded check. 2020-04-21 10:06:17 -04:00
Nathan Conrad d1656c0b8d tu_verify for getting descriptors 2020-04-20 16:09:15 -04:00
hathach bbcf9241bd add back MIDI multiple jack 2020-04-20 23:46:17 +07:00
hathach d57312602d add extra comma to HID_REPORT_ID
this make the template with Report ID look less weird to the user
2020-04-20 16:09:17 +07:00
hathach e4570c35f7 add uart for ea4357 2020-04-20 00:27:35 +07:00
Kay Sievers a0598ef369 MIDI - Add flow control to incoming packet stream
Larger SysEx transfers get corrupted by incoming packets.

This changes the FIFOs not to overwrite their data. MIDI should not be
a transport that drops packets. A potentially blocking device is easier
to detect and handle than a device that silently corrupts the packet
stream at random overflows, especially when SysEx messages are involved.
2020-04-19 11:44:15 +02:00
Ha Thach f9262007ac
Merge pull request #258 from versioduo/midi-packets
MIDI packet interface
2020-04-19 15:43:07 +07:00
hathach 1f442c0a9a also add wakeup event 2020-04-18 23:48:36 +07:00
hathach c3fc5f1595 session end interrupt doesn't trigger on esp32 saola board
it is possibly due to the board design without vbus sense. Revisit
later.
2020-04-18 23:42:51 +07:00
hathach fa71402e17 implement disconnection detection for stm32 synopsys
- disconnection is OTG INT session end bit
- add USE_SOF to disable 1ms interrupt on mcu which isn't used now by
the stack
- add suspend detection
2020-04-18 23:40:27 +07:00
hathach ce6a81e74d
fix typo 2020-04-17 22:10:31 +07:00
hathach 464b1e8e89
correct return for write flush 2020-04-17 15:57:24 +07:00
hathach 4571ce0d29
add a bit of document for cdc device API. also improve cdc write flush when complete. 2020-04-17 15:54:20 +07:00
Ha Thach 0b1b9a2b9e
Merge pull request #260 from majbthrd/cdczlp
CDC device: fix behavior for transfers that are a whole multiple of endpoint buffer
2020-04-17 15:00:12 +07:00
hathach 6f9c256ad0
complete remove dcd_set_config(), fix unit test 2020-04-17 13:52:34 +07:00
hathach 05476d9ad7 Merge branch 'master' into remove-dcd-set-config 2020-04-17 13:38:22 +07:00
Ha Thach 85f0894fff
Merge pull request #347 from majbthrd/nuc505_config
nuc505: change dcd_set_config() behavior
2020-04-17 13:36:16 +07:00
Ha Thach 1f69807621
Merge pull request #354 from hathach/cxd56-disconnect-connect
Cxd56 disconnect connect
2020-04-17 13:12:47 +07:00
hathach dddf30f532
added disconnect for spresense 2020-04-17 12:53:42 +07:00
hathach 50be9d7c3a
mass rename tud/dcd_irq_handler to tud/dcd_init_handler 2020-04-17 12:27:53 +07:00
hathach 89f99426fc
add code for disconnect/connect (not tested) 2020-04-17 11:48:49 +07:00
Ha Thach 4bbbba48cb
Merge pull request #351 from hathach/valentyusb-dcd-disconnect-connect
Valentyusb dcd disconnect connect
2020-04-17 09:52:29 +07:00
Ha Thach 19e2eb66c0
Merge pull request #350 from hathach/synopsys-msp430-dcd-disconnect-connect
Synopsys msp430 dcd disconnect connect
2020-04-17 09:51:15 +07:00
Ha Thach 9aa553ccc2
Merge pull request #352 from hathach/samg-dcd-disconnect-connect
samg implement dcd connect/disconnect
2020-04-16 23:14:42 +07:00
Ha Thach 33610751d7
Merge pull request #336 from pigrew/edpt_close
> If you notice my chain of events above, the bulk transfer was started BEFORE the SET_INTERFACE call. The USB device hardware swaps the order of them being delivered. On STM32, it gives priority to the lower-numbered EP index.

It shouldn't be a matter, control is 2+ stage, before sending the setup. Host should stop all communication to the endpoint that It wants to close.
2020-04-16 23:10:36 +07:00
hathach 36d4efa077
samg implement dcd connect/disconnect 2020-04-16 22:59:19 +07:00
hathach 6377699d77
implement dcd disconnect connect 2020-04-16 22:30:33 +07:00