[PATCH v8 4/6] drm/xe/guc: Extract GuC error capture lists
Teres Alexis, Alan Previn
alan.previn.teres.alexis at intel.com
Sat May 11 01:43:18 UTC 2024
On Mon, 2024-05-06 at 18:47 -0700, Zhanjun Dong wrote:
> Upon the G2H Notify-Err-Capture event, parse through the
> GuC Log Buffer (error-capture-subregion) and generate one or
> more capture-nodes. A single node represents a single "engine-
> instance-capture-dump" and contains at least 3 register lists:
> global, engine-class and engine-instance. An internal link
> list is maintained to store one or more nodes.
>
> Because the link-list node generation happen before the call
> to devcoredump, duplicate global and engine-class register
> lists for each engine-instance register dump if we find
> dependent-engine resets in a engine-capture-group.
>
alan:snip
> diff --git a/drivers/gpu/drm/xe/xe_guc_capture.c b/drivers/gpu/drm/xe/xe_guc_capture.c
> index d2df027081b5..71d7c4a58925 100644
> --- a/drivers/gpu/drm/xe/xe_guc_capture.c
> +++ b/drivers/gpu/drm/xe/xe_guc_capture.c
> @@ -520,6 +520,560 @@ static void check_guc_capture_size(struct xe_guc *guc)
> buffer_size, spare_size, capture_size);
> }
>
alan:snip
> +static struct __guc_capture_parsed_output *
> +guc_capture_get_prealloc_node(struct xe_guc *guc)
> +{
> + struct __guc_capture_parsed_output *found = NULL;
> +
> + if (!list_empty(&guc->capture->cachelist)) {
> + struct __guc_capture_parsed_output *n, *ntmp;
> +
> + /* get first avail node from the cache list */
> + list_for_each_entry_safe(n, ntmp, &guc->capture->cachelist, link) {
> + found = n;
> + list_del(&n->link);
> + break;
> + }
> + } else {
> + struct __guc_capture_parsed_output *n, *ntmp;
> +
> + /* traverse down and steal back the oldest node already allocated */
> + list_for_each_entry_safe(n, ntmp, &guc->capture->outlist, link) {
> + found = n;
> + }
> + if (found)
> + list_del(&found->link);
> + }
> + if (found)
> + guc_capture_init_node(guc, found);
> +
> + return found;
> +}
alan: I mentioned this in rev6, you cannot start pre-allocated nodelist anywhere
in this patch when you are only allocating it in patch 6. Look back at my rev 6
comments on this. Also, take a look at the original i915 patch on how to implement
guc_capture_alloc/delete_one_node without preallocated nodelist:
https://patchwork.freedesktop.org/patch/479022/?series=101604&rev=1
(note: watch especially for the use of new->reginfo[i].regs which needed
additional allocation step. Alternatively we could squash patch 4 and patch 6
together and change patch 4's comment but not sure it might be too large a
patch (can discuss offline).
> +static int
> +guc_capture_extract_reglists(struct xe_guc *guc, struct __guc_capture_bufstate *buf)
> +{
> + struct xe_gt *gt = guc_to_gt(guc);
> + struct guc_state_capture_group_header_t ghdr = {0};
> + struct guc_state_capture_header_t hdr = {0};
> + struct __guc_capture_parsed_output *node = NULL;
> + struct guc_mmio_reg *regs = NULL;
> + int i, numlists, numregs, ret = 0;
> + enum guc_capture_type datatype;
> + struct guc_mmio_reg tmp;
> + bool is_partial = false;
alan:snip
> + if (!node) {
> + node = guc_capture_get_prealloc_node(guc);
alan: see above comment on the use of prealloc_node (as per rev 6's comments)
alan:snip
> +static void __guc_capture_process_output(struct xe_guc *guc)
> +{
> + unsigned int buffer_size, read_offset, write_offset, full_count;
> + struct xe_uc *uc = container_of(guc, typeof(*uc), guc);
> + struct guc_log_buffer_state log_buf_state_local;
> + struct guc_log_buffer_state *log_buf_state;
> + struct __guc_capture_bufstate buf;
> + bool new_overflow;
> + int ret;
> + u32 log_buf_state_offset;
> + u32 src_data_offset;
> +
> + log_buf_state = (struct guc_log_buffer_state *)((ulong)guc->log.bo->vmap.vaddr +
> + (sizeof(struct guc_log_buffer_state) * GUC_CAPTURE_LOG_BUFFER));
alan: once again, i dont think we can use vmap.vaddr directly this this anymore right?
i dont think we use "log_buf_state" until the end of this function to set the new read_ptr
and flush flag. We ought to use xe_map_wr below?
> +
> + log_buf_state_offset = sizeof(struct guc_log_buffer_state) * GUC_CAPTURE_LOG_BUFFER;
> + src_data_offset = xe_guc_get_log_buffer_offset(&guc->log, GUC_CAPTURE_LOG_BUFFER);
> +
> + /*
> + * Make a copy of the state structure, inside GuC log buffer
> + * (which is uncached mapped), on the stack to avoid reading
> + * from it multiple times.
> + */
> + xe_map_memcpy_from(guc_to_xe(guc), &log_buf_state_local, &guc->log.bo->vmap,
> + log_buf_state_offset, sizeof(struct guc_log_buffer_state));
alan:snip
More information about the Intel-xe
mailing list