cli: ensure analyzer finishes before USB shutdown.
by
whitequark
5 years 4 months
target.analyzer: fix initial IO/OE update.
Broken during nMigen migration.
by
whitequark
5 years 4 months
access.direct.multiplexer: fix FIFO throttling.
Broken during nMigen migration.
by
whitequark
5 years 4 months
hardware/boards/glasgow: explain the function of R40.
by
whitequark
5 years 4 months
firmware: let bitstream overflow into FX2_MEM.
The HX8K bitstream is very slightly larger than ICE_MEM. It is not
practical to use a larger memory due to I2C addressing issues. Due
to my foresight of issues like this, I made FX2_MEM twice as large
as the largest FX2 firmware, so now we can stuff the ~0x1000 extra
bytes of the bitstream into FX2_MEM, and still have 0x3000 bytes
of Free Real Estate left.
This commit also reports failures during loading of flashed bitstream
by lighting up the error LED instead of ignoring them.
Fixes #164.
by
whitequark
5 years 4 months
cli: fix `glasgow flash`.
Broken in 3ff1f20a.
by
whitequark
5 years 4 months
applet.audio.yamaha_opx: raise file size limit to 1 MB.
There are some quite large Touhou VGZ files.
by
whitequark
5 years 4 months
access.direct.demultiplexer: fix a race condition cancelling writes.
Before this commit, if _out_task was cancelled inside bulk_write(),
it would not decrement _out_inflight, and this would later result
in a crash in write pushback code:
File ".../Glasgow/software/glasgow/support/task_queue.py", line 62, in wait_one
await asyncio.wait(self._live, return_when=asyncio.FIRST_COMPLETED)
by
whitequark
5 years 4 months
applet.audio.yamaha_opx: allow providing a link to VGM/VGZ as well.
by
whitequark
5 years 4 months
applet.audio.yamaha_opx: rename DAC handlers. NFC.
by
whitequark
5 years 4 months