michaelh/zephyr
1
0
Fork 0
Code Releases Activity

soc/intel_adsp: DMA Stability fix for cavstool

Browse source 
There is a hardware bug with stream reset, it won't stop the stream.
I thought it was limited to just the early versions, but it turns out
that 2.5 has it too in some variants.  Which is frustrating, because
the TGL chromebook I have doesn't like the START clear.  But all
systems work with an extra delay between them, so do that.

Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
This commit is contained in:
Andy Ross 2022-01-24 19:50:55 -08:00 • committed by Anas Nashif
commit 140837ce3a
1 changed files with 8 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

13
soc/xtensa/intel_adsp/tools/cavstool.py
View file

@ -256,11 +256,14 @@ def load_firmware(fw_file):
log.warning(f"Load failed? FW_STATUS = 0x{dsp.SRAM_FW_STATUS:x}")
# Turn DMA off and reset the stream. Clearing START first is a
# noop per the spec, but required for early versions (apparently
# the reset doesn't stop the stream and the next load fails), and
# makes the load on 2.5 unstable. Go figure.
if not cavs25:
sd.CTL &= ~2 # clear START
# noop per the spec, but absolutely required for stability.
# Apparently the reset doesn't stop the stream, and the next load
# starts before it's ready and kills the load (and often the DSP).
# The sleep too is required, on at least one board (a fast
# chromebook) putting the two writes next each other also hangs
# the DSP!
sd.CTL &= ~2 # clear START
time.sleep(0.1)
sd.CTL |= 1
log.info(f"cAVS firmware load complete")