Граф коммитов

3621 коммит

Автор SHA1 Сообщение Дата
sago35
71b44e79b3 machine/usb/hid/joystick: allow joystick settings override 2023-04-03 00:50:30 +02:00
deadprogram
9e97566b5f machine/usb/hid/joystick: move joystick under HID as it belongs and also remove duplicate code
Signed-off-by: deadprogram <ron@hybridgroup.com>
2023-04-03 00:50:30 +02:00
Ayke van Laethem
7d83e76833 rp2040: use 4MHz as default frequency for SPI
This matches other SPI implementations. I think the original value of
115200 was from a confusion with UART.
2023-04-02 01:04:37 +02:00
Damian Gryski
9e7882b1b7 reflect: uncomment a another test the fails and doc some that don't 2023-04-01 22:46:46 +02:00
Damian Gryski
0c4f9d1f19 reflect; uncomment SetZero (but comment out the parts that fail) 2023-04-01 22:46:46 +02:00
Damian Gryski
a85cb22193 reflect: uncomment TestAppend and fix a bug it found 2023-04-01 22:46:46 +02:00
Damian Gryski
60bb832c89 reflect: handle Convert'ing between identical underlying types
Needed for go-jose/v3
2023-04-01 22:46:46 +02:00
Ayke van Laethem
6eda52a289 rp2040: remove SPI deadline
Removing it improves SPI performance by about 20% for me (updating the
display of a Gopher Badge).
2023-04-01 11:18:51 +02:00
Damian Gryski
8badf79af9 testing: remove unused variable left over from count prototyping 2023-03-31 19:17:02 -07:00
Damian Gryski
66d3c4edb8 main: fix typos in flag usage messages 2023-03-31 19:17:02 -07:00
Damian Gryski
63aaa43072 testing: add test.skip
Fixes #3056
2023-03-31 19:17:02 -07:00
Damian Gryski
ee81c31884 testing: import new version of match.go 2023-03-31 19:17:02 -07:00
Damian Gryski
50d681359d main: set WASMTIME_BACKTRACE_DETAILS when running in wasmtime.
I find myself consistently running tests, seeing them panic, and then
immediately running them again with this environment variable set.  It's
easier to just have tinygo do this for me.
2023-03-31 13:06:23 -07:00
Damian Gryski
d50c54fce0 Makefile: compress/lzw seems to work on wasi now. 2023-03-31 20:41:51 +02:00
Damian Gryski
4a81cac53b main: make sure all testing output goes to the same place 2023-03-31 09:07:13 +02:00
Damian Gryski
84a3273131 main: fix tests with default TestConfig.Count=0 doesn't skip all tests 2023-03-31 09:07:13 +02:00
Damian Gryski
9182664845 testing: make test output unbuffered when verbose
Fixes #3579
2023-03-31 09:07:13 +02:00
Damian Gryski
a2f95d6b87 main: stuff test runner options into their own struct
Fixes #2406
2023-03-31 09:07:13 +02:00
Damian Gryski
698b1f19c6 testing: support -test.count
This makes running benchmarks repeatedly easier.
2023-03-31 09:07:13 +02:00
Damian Gryski
e6ccdd9d1a reflect: another obscure RO bug 2023-03-31 01:08:04 +02:00
Damian Gryski
b39a982067 reflect: uncomment another test and fix RO logic issues it uncovered 2023-03-31 01:08:04 +02:00
Ayke van Laethem
e0bf376068 rp2040: unify all linker scripts using LDFLAGS
The only thing that's different between all these chips is the flash
size, which can easily be passed as a linker flag instead. This removes
a bunch of duplicate code in an uncommon language (linker script).

I've also fixed a few boards with incorrect flash sizes:

  * nano-rp2040 has 16MB instead of 2MB
  * macropad-rp2040 has 8MB instead of 2MB
  * gopher-badge has 8MB instead of 1MB
2023-03-30 23:49:02 +02:00
Damian Gryski
b044d4ff3d reflect: add more RO checks 2023-03-30 21:10:54 +02:00
Damian Gryski
0cd93a3a9e reflect: add valueFlagRO 2023-03-30 21:10:54 +02:00
Damian Gryski
5faff2e13a reflect: add sipmlified strconv.Quote() implementation for struct tags 2023-03-30 21:10:54 +02:00
Damian Gryski
195de23d3b reflect: Fix Kind(-1).String() and enable test 2023-03-30 21:10:54 +02:00
Damian Gryski
d4bdd836bc reflect: implement and test Value.Comparable 2023-03-30 21:10:54 +02:00
Damian Gryski
a11f2436e3 reflect: TestAliasNames passes 2023-03-30 21:10:54 +02:00
Damian Gryski
017ab4c352 reflect: fix TestCanSetField 2023-03-30 21:10:54 +02:00
Damian Gryski
181d2ad2b4 reflect: add CanInt() and friends and uncomments tests that pass 2023-03-30 21:10:54 +02:00
Damian Gryski
53b95cad08 reflect: uncomment Type.String() tests that pass 2023-03-30 21:10:54 +02:00
Damian Gryski
e7bd22edf2 reflect: print struct tags in Type.String() (with a caveat) 2023-03-30 21:10:54 +02:00
Damian Gryski
1a60a1f526 reflect: stub channel select routines/types 2023-03-30 21:10:54 +02:00
Damian Gryski
3fbd3c4d93 compiler,reflect: support channel directions 2023-03-30 21:10:54 +02:00
deadprogram
1213a45197 build: add GH workflow to build LLVM image when needed
Signed-off-by: deadprogram <ron@hybridgroup.com>
2023-03-30 12:40:02 +02:00
Damian Gryski
4e4e4eee04 runtime: use unsafe.Add() in hashmap code 2023-03-30 00:08:07 +02:00
Damian Gryski
bf20c652e2 runtime: take max hash size into account when preallocating with size hint 2023-03-30 00:08:07 +02:00
Damian Gryski
a3afd4e8ac runtime: factor duplicate hashmap snippts to functions 2023-03-30 00:08:07 +02:00
Damian Gryski
cb9c6f0074 runtime: zero map key/value on deletion to so GC doesn't see them 2023-03-30 00:08:07 +02:00
Damian Gryski
5fee3428bc runtime: preallocate maps to size hint actually works 2023-03-30 00:08:07 +02:00
Ayke van Laethem
464ebc4fe1 compiler: implement most math/bits functions
These functions can be implemented more efficiently using LLVM
intrinsics. That makes them the Go equivalent of functions like
__builtin_clz which are also implemented using these LLVM intrinsics.

I believe the Go compiler does something very similar: IIRC it converts
calls to these functions into optimal instructions for the given
architecture.

I tested these by running `tinygo test math/bits` after uncommenting the
tests that would always fail (the *PanicZero and *PanicOverflow tests).
2023-03-29 20:55:09 +02:00
Ayke van Laethem
568c2a4363 rp2040: remove SPI DataBits property
As discussed on Slack, I believe this property does more harm than good:

  * I don't think it's used anywhere. None of the drivers use it.
  * It is not fully implemented. While values <= 8 might work fine,
    values larger than 8 result in extra zero bits (instead of anything
    sensible).
  * Worse, it doesn't return an error when it's out of range. This is
    not an optional property: if the SPI peripheral doesn't support a
    particular number of bits, it should return an error instead of
    silently limiting the number of bits. This will be confusing to
    users.

Therefore, I propose we drop it. Maybe there are good uses for it
(perhaps for displays that use big endian 16-bit values?), but without a
good use case like a driver in tinygo.org/x/drivers, I think it's more
trouble than it's worth.
2023-03-29 09:30:16 +02:00
Kenneth Bell
c611c72526 build: test net on linux & mac only (random CI fails on windows) 2023-03-29 07:57:55 +02:00
deadprogram
dfb8c996a1 machine/lorae5: correct mapping for I2C bus, add pin mapping to enable power
Signed-off-by: deadprogram <ron@hybridgroup.com>
2023-03-28 15:28:14 +02:00
Achille Roussel
85da9a0aac fix resource leak in os.(*File).Close
Signed-off-by: Achille Roussel <achille.roussel@gmail.com>
2023-03-28 13:12:21 +02:00
Ayke van Laethem
17bc0d6663 compiler: only support //go:wasmimport on declared functions
Don't support this pragma on defined functions. It is only meant for
importing, not for exporting.
2023-03-28 09:31:09 +02:00
Ayke van Laethem
31043628d8 reflect: use direct calls to runtime string functions
The runtime.stringFromBytesTyped and runtime.stringToBytesTyped
functions aren't really necessary, because they have the same LLVM IR
signature. Therefore, remove them and link directly to the functions
that the compiler uses internally.
2023-03-27 22:24:20 +02:00
Damian Gryski
2c0f61cad1 reflect: fix bug found by Convert() tests on wasi 2023-03-27 18:53:37 +02:00
Damian Gryski
97ece754f6 reflect: add missing Uintptr type and some numerical tests 2023-03-27 18:53:37 +02:00
Damian Gryski
39f76f43fc reflect: fix indirect issues with makeInt/makeUint/makeFloat 2023-03-27 18:53:37 +02:00