remote bitbang failing on halt command and sampling wrong IR capture

I am running remote bitbang server with RTL simulation(CVA6 ariane core + riscv-dbg RTL +simJTAG). When I try to connect openocd I am getting following error:


Error: 222 127573 core.c:1242 jtag_validate_ircapture(): riscv.cpu: IR capture error; saw 0x0a not 0x01
Debug: 223 127573 core.c:327 jtag_call_event_callbacks(): jtag event: TAP reset
Warn : 224 127573 core.c:1462 jtag_init_inner(): Bypassing JTAG setup events due to errors
Debug: 225 127573 command.c:143 script_debug(): command - dap dap init
Debug: 226 127573 log.c:409 keep_alive(): keep_alive() was not invoked in the 1000ms timelimit (127573). This may cause trouble with GDB connections.
Debug: 229 127573 arm_dap.c:105 dap_init_all(): Initializing all DAPs ...
Debug: 230 127573 openocd.c:158 handle_init_command(): Examining targets...
Debug: 231 127573 target.c:1616 target_call_event_callbacks(): target event 17 (examine-start) for core riscv.cpu
Debug: 232 127573 riscv.c:1015 riscv_examine(): riscv_examine()
Debug: 233 136596 riscv.c:445 dtmcontrol_scan(): DTMCONTROL: 0x0 -> 0x20e2
Debug: 234 136596 riscv.c:1025 riscv_examine(): dtmcontrol=0x20e2
Debug: 235 136596 riscv.c:1027 riscv_examine():   version=0x2
Error: 236 136596 riscv.c:465 get_target_type(): Unsupported DTM version: 2
Debug: 237 136596 openocd.c:160 handle_init_command(): target examination failed
Debug: 238 136596 command.c:143 script_debug(): command - flash flash init
Debug: 239 136597 log.c:409 keep_alive(): keep_alive() was not invoked in the 1000ms timelimit (9024). This may cause trouble with GDB connections.
Debug: 242 136597 tcl.c:1222 handle_flash_init_command(): Initializing flash devices...
Debug: 243 136597 command.c:143 script_debug(): command - nand nand init
Debug: 245 136597 tcl.c:498 handle_nand_init_command(): Initializing NAND devices...
Debug: 246 136597 command.c:143 script_debug(): command - pld pld init
Debug: 248 136597 pld.c:206 handle_pld_init_command(): Initializing PLDs...
Debug: 249 136597 gdb_server.c:3503 gdb_target_start(): starting gdb server for riscv.cpu on 3333
Info : 250 136597 server.c:311 add_service(): Listening on port 3333 for gdb connections
Debug: 251 136597 command.c:143 script_debug(): command - halt halt
Debug: 253 136597 target.c:3070 handle_halt_command(): -
Error: 254 136597 target.c:567 target_halt(): Target not examined yet
Debug: 255 136597 command.c:630 run_command(): Command 'halt' failed with error code -4

The configuration file I use is as below:


interface remote_bitbang
remote_bitbang_host localhost
remote_bitbang_port 8088

set _CHIPNAME riscv
jtag newtap $_CHIPNAME cpu -irlen 5

set _TARGETNAME $_CHIPNAME.cpu
target create $_TARGETNAME riscv -chain-position $_TARGETNAME -coreid 0

gdb_report_data_abort enable

riscv set_reset_timeout_sec 120
riscv set_command_timeout_sec 120

init
halt

The DTMCONTROL as per RTL is:


 dtmcs_d  = '{
                      zero1        : '0,
                      dmihardreset : 1'b0,
                      dmireset     : 1'b0,
                      zero0        : '0,
                      idle         : 3'd1, // 1: Enter Run-Test/Idle and leave it immediately
                      dmistat      : dmi_error_i, // 0: No error, 2: Op failed, 3: too fast
                      abits        : 6'd7, // The size of address in dmi
                      version      : 4'd1  // Version described in spec version 0.13 (and later?)
                    };

version is mentioned 1 here but jtag is reporting it as 2. Has anyone tried this before or faced such issue before ?

pulp-platform / riscv-dbg

remote bitbang failing on halt command and sampling wrong IR capture #153