[Blackhole bringup] Wider addresses in NOC cmd buffer programming sequence

[abhullar-tt]

TODO:

• Investigate and summarize changes between noc_nonblocking_api in WH and BH
• Highlight APIs in dataflow_api.h that are candidates to move into noc_nonblocking_api (many (?) low-level APIs added here to workaround the fact that noc_nonblocking_api only exposes one API that programs all registers even if this is not needed on each invocation)
• Zero out "MID" register in FW since we will use the LO/HI registers

[abhullar-tt]

FYI @pgkeller, It doesn't look like any drastic changes introduced for BH noc_nonblocking_api. A lot of the changes to BH version needed because BBE and Metal versions of WH noc_nonblocking_api.h have diverged with BH version being copied directly from BBE.

I am assuming that the other tensix files have also diverged. I can go through those as well and make sure our BH versions are incremental steps from WH versions rather than copied of BBE variants. Should we consider going through a commonization exercise (between Metal and BBE) and add these files to UMD?

API additions/removals in BH noc_nonblocking_api to match diverged Metal WH noc_nonblocking_api

• Note: __attribute__((always_inline)) needs to be added
• NOC_CMD_BUF_WRITE_REG_L1, NOC_CMD_BUF_READ_REG_L1, NOC_STATUS_READ_REG_L1 need to be removed from BH (Metal WH version uses the non L1 specific APIs)
• noc_cmd_buf_ready needs to be added to BH
• ncrisc_noc_fast_read_scatter not in Metal WH version (remove from BH?)
• remove ncrisc_noc_fast_read_l1
• remove ncrisc_noc_reads_flushed_l1
• remove ncrisc_noc_all_reads_flushed and ncrisc_noc_all_reads_flushed_l1 and add BH variant of ncrisc_noc_reads_flushed
• remove ncrisc_noc_fast_read_ok and ncrisc_noc_fast_read_ok_l1
• remove ncrisc_rd_data_word_recv
• remove ncrisc_noc_clear_outstanding_reqs
• remove ncrisc_noc_fast_write_l1
• remove ncrisc_noc_fast_write_ok, ncrisc_noc_fast_write_bufs_ok, ncrisc_noc_fast_write_ok_l1
• remove ncrisc_noc_blitz_write
• remove ncrisc_noc_nonposted_writes_sent_l1
• remove ncrisc_noc_nonposted_all_writes_sent and ncrisc_noc_nonposted_all_writes_sent_l1
• remove ncrisc_noc_nonposted_writes_flushed_l1
• remove ncrisc_noc_nonposted_all_writes_flushed and ncrisc_noc_nonposted_all_writes_flushed_l1
• add noc_local_state_init
• add ncrisc_noc_counters_init
• remove ncrisc_noc_all_flushed
• remove RISC_B0_HW versions of ncrisc_noc_fast_read_any_len
• remove ncrisc_noc_fast_read_any_len_scatter
• remove ncrisc_noc_fast_write_any_len_l1
• add ncrisc_noc_fast_write_any_len_loopback_src
• remove noc_atomic_read_and_increment_l1

API changes

ncrisc_noc_fast_read

• src_addr split between NOC_TARG_ADDR_LO/MID/HI

  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_TARG_ADDR_LO, (uint32_t)src_addr);
  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_TARG_ADDR_MID, (uint32_t)(src_addr >> 32) & 0xF);
  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_TARG_ADDR_HI, (uint32_t)(src_addr >> 36) & 0xFFFFFF);

ncrisc_noc_fast_write

• parameter update: remove transaction_id and add multicast_path_reserve +posted
• dest_addr split between NOC_RET_ADDR_LO/MID/HI

  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_RET_ADDR_LO, (uint32_t)dest_addr);
  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_RET_ADDR_MID, (uint32_t)(dest_addr >> 32) & 0xF);
  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_RET_ADDR_HI, (uint32_t)(dest_addr >> 36) & 0xFFFFFF);

Add ncrisc_noc_fast_write_loopback_src where dest_addr split between NOC_RET_ADDR_LO/MID/HI

ncrisc_noc_blitz_write_setup

• remove transaction_id parameter
• replace poll to wait on noc_cmd_buf_ready
• split dest_addr between NOC_RET_ADDR_MID/HI

  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_RET_ADDR_MID, (uint32_t)(dest_addr >> 32) & 0xF);
  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_RET_ADDR_HI,  (uint32_t)(dest_addr >> 36) & 0xFFFFFF);

Remove transaction_id from:

• ncrisc_noc_reads_flushed
• ncrisc_noc_nonposted_writes_sent
• ncrisc_noc_nonposted_writes_flushed

noc_init (rename from ncrisc_noc_init and port Metal WH version)

• xy_local_addr should be written to NOC_TARG_ADDR_HI
• atomic_ret_addr split between NOC_RET_ADDR_LO/MID/HI????
• xy_local_addr written to NOC_RET_ADDR_HI???

Port WH version of ncrisc_noc_full_sync

noc_fast_posted_write_dw_inline

• Add posted parameter
• dest_addr written to NOC_TARG_ADDR_LO/MID/HI

  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_TARG_ADDR_LO, (uint32_t)(dest_addr & ~(NOC_WORD_BYTES-1)));
  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_TARG_ADDR_MID, (dest_addr >> 32) & 0xF);
  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_TARG_ADDR_HI, (dest_addr >> 36) & 0xFFFFFF);

noc_fast_atomic_increment

• parameter changes: remove read_addr and transaction_id, add posted
• addr written to NOC_TARG_ADDR_LO/MID/HI

  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_TARG_ADDR_LO, (uint32_t)(addr & 0xFFFFFFFF));
  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_TARG_ADDR_MID, (uint32_t)(addr >> 32) & 0xF));
  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_TARG_ADDR_HI, (uint32_t)(addr >> 36) & 0xFFFFFF);

[abhullar-tt]

@davorchap do you anticipate needing to address over 4 GB? If not then we don't need the higher bits of src_addr/dest_addr and can ignore writing to *_MID/*_HI registers.

@pgkeller on WH we do write to *_LO and *_MID on WH, eg:

  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_TARG_ADDR_LO, (uint32_t)src_addr);
  NOC_CMD_BUF_WRITE_REG(noc, cmd_buf, NOC_TARG_ADDR_MID, src_addr >> 32);

[abhullar-tt]

Related to noc_nonblocking_api: seeing a hang in ncrisc_noc_reads_flushed

[abhullar-tt]

Related to noc_nonblocking_api: seeing a hang in ncrisc_noc_reads_flushed

this was due to bug in setting a HI register (value needed to be right shifted by an additional 4 bits)