aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* notes/interrupts.txt: Fix typos and other errors.Marti Bolivar2012-08-231-29/+27
| | | | Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Add notes/interrupts.txt.Marti Bolivar2012-08-231-0/+348
| | | | | | | | This is a developer's doc explaining how interrupts are handled in libmaple. It's time we exposed this information to enable people who know what they're doing to move around more easily. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Revert "Re-work the constraints on delay_us(). Ask GCC to allocate a register"Marti Bolivar2012-08-221-4/+4
| | | | | | | | This reverts commit 715a63dccb3a39d7d0502fb39b900c8302a60a87. It fails to build using the 2010 CodeSourcery toolchain, which is still the latest officially supported version. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_serial.cpp: Neaten up reset signal RX hook.Marti Bolivar2012-08-221-34/+33
| | | | | | | | | | Move magic bytes from stack to Flash. De-magic-number sizeof magic byte buffer. Use uintptr_t instead of unsigned int when casting address of wait_reset(). Based on a patch by Michael Hope. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Re-work the constraints on delay_us(). Ask GCC to allocate a registerMichael Hope2012-08-221-4/+4
| | | | | | | instead of forcing r0. Use the same operand as input and output to remove the move. Signed-off-by: Michael Hope <michael.hope@linaro.org>
* Merge branch 'reset-script'Marti Bolivar2012-08-221-7/+27
|\ | | | | | | | | | | Weed out non-Maple ttyACMs on Linux. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
| * reset.py: Not every Unix is Linux.Marti Bolivar2012-08-221-8/+20
| | | | | | | | | | | | | | | | Move the sysfs tests for Maple out of unix_get_maple_path() and into a new linux_get_maple_path(). This prevents unnecessary probing for a nonexistent /sys on e.g. OS X. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
| * Improve Maple device detection in reset.py on LinuxDavid Kiliani2012-08-221-4/+12
|/ | | | | | | | Adds a function to lookup the USB vendor & product id from the udev info in the sysfs tree. This removes invalid choices and reduces user queries for the correct ttyACM device. Signed-off-by: David Kiliani <mail@davidkiliani.de>
* reset.py fixups for OS X 10.8.Marti Bolivar2012-08-201-2/+3
| | | | | | | | As discussed on this forum thread: http://forums.leaflabs.com/topic.php?id=2451 Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Make PRIMASK set/clear functions __always_inline.Marti Bolivar2012-08-142-4/+4
| | | | | | | This is just to ensure that -Os doesn't prevent these from getting inlined. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* wirish/ext_interrupts.h: DoxygenMarti Bolivar2012-08-141-6/+7
| | | | Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* stm32loader.py: replace obsolete exception catch syntax.Marti Bolivar2012-08-101-1/+1
| | | | | | For Python 3 support. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Makefile: Don't assume Cortex M3.Marti Bolivar2012-08-052-5/+14
| | | | | | | | | Instead, derive -march and -mcpu from the target MCU in target-config.mk. Also (on STM32F4) add the necessary floating point flags. The CodeSourcery toolchains don't support these, so we'll need to prepare alternative toolchains when F4 support is official. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* fixup! Add support for ARM's GCC ARM embedded toolchain.Marti Bolivar2012-08-052-0/+7
| | | | | | Oops. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Add support for ARM's GCC ARM embedded toolchain.Marti Bolivar2012-08-055-22/+44
| | | | | | | Based on patches provided by Hanspeter Portner: http://forums.leaflabs.com/topic.php?id=1717#post-11812 Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Makefile: move flag settings after build-rules and -templates.Marti Bolivar2012-08-051-7/+7
| | | | | | | | | We'll need to mess with the linker flags to add support for ARM toolchains other than CodeSourcery's. Since the toolchain is determined in build-rules.mk, decisions that depend on it have to happen after it's included. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm.c: Fix up descriptor definitions.Marti Bolivar2012-08-041-39/+25
| | | | Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm.h: Add convenience for declaring device descriptor.Marti Bolivar2012-08-041-0/+21
| | | | Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* <libmaple/usb.h>: Fix usb_descriptor_string.Marti Bolivar2012-08-041-3/+3
| | | | | | uint16 is the wrong type for the bString array. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Move contents of usb_descriptors.h to <libmaple/usb.h>.Marti Bolivar2012-08-043-139/+94
| | | | | | | | These form and name of structs etc. is basically determined by the USB spec and the libmaple style guide, so there's little reason for them to change, and less reason for them to stay hidden. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Prep usb_descriptors.h for removal.Marti Bolivar2012-08-043-63/+84
| | | | | | | | | | | | Just change names and things to fit the libmaple style. Its contents will move to the top-level <libmaple/usb.h> header next. This API is still unstable, but we need to expose pieces of it in order to get the Maple-specific pieces of usb_cdcacm.c into Wirish. Adjust uses of the API appropriately. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm: Remove unused/redundant variables.Marti Bolivar2012-08-041-11/+3
| | | | Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm: Expose line coding state.Marti Bolivar2012-08-042-22/+55
| | | | | | | This was already supported, so just move the line coding structure to the header and add some functions for accessing the current values. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm: Set endpoint to NAK before checking byte count.Marti Bolivar2012-08-041-4/+4
| | | | | | Fixes an (admittedly very unlikely) race condition. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Add FIXME for usb_get_ep_tx_count().Marti Bolivar2012-08-041-1/+3
| | | | | | | | | | | We definitely need to mask out the high bits, as RM0008 doesn't specify that they read as zero. Calling this function seems to crash the board, though. I'm not sure why, especially since usb_set_ep_tx_count(), used by usb_cdcacm.c, seems to work. Confusing. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm: various fixups.Marti Bolivar2012-08-042-64/+60
| | | | | | | | | | Mark internal state static. Properly reset the RX/TX state on USB reset. Choose better names for countTx and newBytes. Move the exposed configuration back into the .c; this information is not beneficial to hooks. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm: Improve write speed.Marti Bolivar2012-08-031-4/+3
| | | | | | | | | Actually fill the TX buffer in the packet memory area, so we can send more bytes per transaction. Using the test-session benchmark, we're now clocking in the high 700KB/sec range via PySerial. With screen, rate is now about 550 KB/sec. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm: CosmeticsMarti Bolivar2012-08-031-1/+1
| | | | Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* test session: Add SerialUSB benchmark to 'U' command.Marti Bolivar2012-08-031-4/+20
| | | | | | | | I'm currently measuring over 500 KB/sec with screen as my serial monitor. If I don't display the output at all, I get over 600 KB/sec. Nice! Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm: minor fixups.Marti Bolivar2012-08-031-17/+41
| | | | | | | | | | Rename USB_Line_Coding to usb_line_coding, and rename its fields to match the spec to aid searches. Alter the line_coding global to match, fixing an incorrect comment and avoiding magic numbers. Other minor fixups and comment improvements. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm: Rename "notification" -> "management".Marti Bolivar2012-08-012-11/+11
| | | | | | This keeps things in line with the specification, helping readability. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Move CDC ACM config. descriptor into usb_cdcacm.c.Marti Bolivar2012-07-312-15/+15
| | | | Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Move bootloader details out of libmaple.h.Marti Bolivar2012-07-312-8/+12
| | | | | | Finally! Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* HardwareSerial: make read() block until data arrive.Marti Bolivar2012-07-311-0/+3
| | | | | | | This makes the HardwareSerial behavior match that of SerialUSB, and should help avoid confusion. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Move bootloader reset signal detection to Wirish.Marti Bolivar2012-07-313-127/+140
| | | | | | | | Use usb_cdcacm hooks to move the DTR edge and "1EAF" magic packet detection to usb_serial.cpp. We'll later be able to extend this system to support Leonardo-style reset signalling. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm: Add usb_cdcacm_peek().Marti Bolivar2012-07-312-27/+39
| | | | | | | | | | | We need this so the bootloader reset signal detector can look ahead in the byte stream without stealing bytes from the user when "1EAF" isn't received. While we're doing this, take the time to fix a bunch of RX-related code (dead code removal, adding volatile, etc.) Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdacm: Use config instead of magic numbers.Marti Bolivar2012-07-311-2/+2
| | | | | | Sigh. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm: Add hook system.Marti Bolivar2012-07-312-0/+44
| | | | | | | | Provide hooks so users can reach into the CDC ACM callbacks with their own code. We'll use this to move the bootloader reset signals to Wirish. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm: Expose some internal configuration and values.Marti Bolivar2012-07-312-84/+93
| | | | | | | | | Mark the API as unstable so people don't rely on these too hard. (This header is not part of the official libmaple docs). We'll need some of this configuration elsewhere later, and it doesn't make much sense to expose it piecemeal. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm.c: Add another verification TODO.Marti Bolivar2012-07-311-0/+2
| | | | | | | Not sure what the comment above the new TODO is supposed to mean, but it looks like something that should be checked. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_cdcacm.c: Resolve TODO and improve comments.Marti Bolivar2012-07-311-5/+7
| | | | | | | | | The disc pin code was copy-pasted when the CDC ACM pieces were extracted from the rest of the USB code, and I didn't really understand what was going on when I moved it over. Having checked the spec, the comments can be safely improved. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Hack: Provide a hook so Mini doesn't disable JTAG/SWD.Marti Bolivar2012-07-311-1/+3
| | | | | | | The nonexistence of a configuration system is starting to become kind of a pain. Oh well; let's keep piling hacks on until things break. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* usb_serial.cpp: Minor tweak.Marti Bolivar2012-07-251-2/+1
| | | | Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* Add pin alias enums to other boards.Marti Bolivar2012-07-245-0/+41
| | | | | | | | Repeat Anton's VLDiscovery trick on the rest of the boards. Leave out st_stm3220g_eval since that's just a teaser to get people to play with F2. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* stm32loader.py: Cosmetics.Marti Bolivar2012-07-241-23/+22
| | | | | | | Move the actual QUIET setting to above mdebug(); that was confusing. Rename _wait_for_ask to _wait_for_ack. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* stm32loader.py: Updates for F2 and F4.Marti Bolivar2012-07-241-7/+51
| | | | | | | | | | | | | Support the extended erase memory command, which replaced the erase memory command in the bootloader protocol as of version 3.0. This takes a long time, so tweak _wait_for_ask to take an extra timeout for special cases like this. Use this under if __name__ == '__main__' to support F2 and F4. Additionally, produce a human-readable report of the target chip from its chip ID. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* stm32loader.py: Improve error message on failed init.Marti Bolivar2012-07-241-1/+1
| | | | Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* stm32loader.py: Use new-style classes.Marti Bolivar2012-07-241-1/+1
| | | | Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* stm32loader.py: Don't use Psyco.Marti Bolivar2012-07-241-8/+0
| | | | | | This script is I/O bound. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
* stm32loader.py: Use print function.Marti Bolivar2012-07-241-10/+12
| | | | | | Towards py3k compatibility. Signed-off-by: Marti Bolivar <mbolivar@leaflabs.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-27 21:37:41 +0000