hl_state_dum...nc_to_engine(9)

🌸 NAME 🌸

okay so bestie, this is the hl_state_dum...nc_to_engine(9) page from the Kernel API Manual, and the function we're obsessing over is hl_state_dump_get_sync_to_engine — it transforms sync_id into a hl_sync_to_engine_map_entry if that's available for the current id. periodt 💅✨

---

✨ SYNOPSIS ✨

the whole signature is literally:

struct hl_sync_to_engine_map_entry * hl_state_dump_get_sync_to_engine
(struct hl_sync_to_engine_map *map , u32 sync_id );

---

🫶 ARGUMENTS 🫶

• map — the sync object to engine map. it's giving lookup energy 🗺️
• sync_id — the sync object id, aka the id we're trying to translate 🔎

---

💖 DESCRIPTION 💖

Returns the translation entry if found, or NULL if it's not.

bestie, hear me: a returned NULL shall not be considered as a failure 😭 the map does not cover all possible, it is a best effort sync ids. so don't spiral, it's just built like that 🫶

---

🔥 SEE ALSO 🔥

Kernel file ./drivers/accel/habanalabs/common/state_dump.c

• hl_format_as_binary(9)
• resize_to_fit(9)
• hl_snprintf_resize(9)
• hl_sync_engine_to_string(9)
• hl_print_resize_sync_engine(9)
• hl_state_dump_get_sync_name(9)
• hl_state_dump_get_monitor_name(9)
• hl_state_dump_free_sync_to_engine_map(9)
• hl_state_dump_read_sync_objects(9)
• hl_state_dump_free_sync_objects(9)
• hl_state_dump_print_syncs_single_block(9)
• hl_state_dump_print_syncs(9)
• hl_state_dump_alloc_read_sm_block_monitors(9)
• hl_state_dump_free_monitors(9)
• hl_state_dump_print_monitors_single_block(9)
• hl_state_dump_print_monitors(9)
• hl_state_dump_print_engine_fences(9)
• hl_state_dump_print_fences(9)
• hl_state_dump(9)

---

luv u, bestie — this one lives in ./drivers/accel/habanalabs/common, it's dated October 2026, and it's serving hl_state_dum...nc_to_engine(9) manual-page realness. byeee 💅🎀🦋