search

pulp-platform / riscv-dbg

RISC-V Debug Support for our PULP RISC-V Cores
Other
218 stars 72 forks source link

Assertion failure expected? #137

Open dpetrisko opened 2 years ago

dpetrisko commented 2 years ago

Hi, I'm trying to use the pulp debug module for my own chip. Things seems to be mainly working, but I'm trying to sort out the last few bugs. I'm using a custom DMI/DTM, but standard riscv-openocd, PULP DM and debug_rom.

Openocd version:

Open On-Chip Debugger 0.11.0+dev-02377-g911d68ef2 (2022-09-12-11:22)
Licensed under GNU GPL v2
For bug reports, read
    http://openocd.org/doc/doxygen/bugs.html

testchip.cfg

adapter driver remote_bitbang
remote_bitbang host localhost
remote_bitbang port 12999
transport select jtag

set _CHIPNAME testchip
set _CPUTAPID 0x04f5484d

jtag newtap $_CHIPNAME cpu -irlen 5 -expected-id $_CPUTAPID
set _TARGETNAME $_CHIPNAME.cpu
target create $_TARGETNAME riscv -chain-position $_TARGETNAME -coreid 0

riscv set_reset_timeout_sec 2000
riscv set_command_timeout_sec 2000

# prefer to use sba for system bus access
riscv set_prefer_sba off

init
halt

I added a print debug to see what DTM requests I'm receiving and found this violated assertion.

...
[          1516600000] '{addr:'h5, op:DTM_READ, data:'h0}
"PATH/opentitan/hw/ip/prim/rtl/prim_fifo_sync.sv", 193: testbench.dm.i_dm_top.i_dm_csrs.i_fifo.DataKnown_A: started at 1516601500ps failed at 1516601500ps
    Offending '(!$isunknown(rdata_o))'
Error: "PATH/opentitan/hw/ip/prim/rtl/prim_fifo_sync.sv", 193: testbench.dm.i_dm_top.i_dm_csrs.i_fifo.DataKnown_A: at time 1516601500 ps
1516602000: (PATH/opentitan/hw/ip/prim/rtl/prim_fifo_sync.sv:193) [testbench.dm.i_dm_top.i_dm_csrs.i_fifo.DataKnown_A] [ASSERT FAILED] DataKnown_A

Looking back through my log, there is no DTM_WRITE to addr 5 before this read which explains the Xs. However, I was wondering if it were perhaps a software/driver problem that these writes were not happening or getting dropped. Have you seen this problem before?

Thanks!

bluewww commented 2 years ago

I haven't seen this kind of error before.

You seem to be using a very recent openocd version and a different FIFO IP it seems. The FIFO IP that is supported by riscv-dbg is the one from common_cells.

dpetrisko commented 2 years ago

Oh, I hadn't noticed that opentitan patches the pulp DM with different primitives, so it's not standard. My mistake!

Which version of Openocd do you suggest? I just pulled down latest and built