|SUCCESS| [1/1] ml/cnxk: enable data caching for all MRVL layers

[dpdklab at iol.unh.edu]

Test-Label: ai-code-review
Test-Status: SUCCESS
https://dpdk.org/patch/162790

_AI Code Review_

=== Patch Review: series.patch (via Claude) ===
Review date: 2026-03-31

git sha: 20260331085350.1105103-1-syalavarthi at marvell.com
Author: Srikanth Yalavarthi <syalavarthi at marvell.com>
Subject: [PATCH 1/1] ml/cnxk: enable data caching for all MRVL layers

This patch enables data caching for all MRVL layers in TVM models by
removing conditional checks that previously limited caching to specific
model types and subtypes.

> diff --git a/drivers/ml/cnxk/cn10k_ml_ops.c b/drivers/ml/cnxk/cn10k_ml_ops.c
> index b30af7c7a44..628ff963c3c 100644
> --- a/drivers/ml/cnxk/cn10k_ml_ops.c
> +++ b/drivers/ml/cnxk/cn10k_ml_ops.c
> @@ -997,13 +997,8 @@ cn10k_ml_layer_start(void *device, uint16_t model_id, const char *layer_name)
>  	if (ret < 0) {
>  		cn10k_ml_layer_stop(device, model_id, layer_name);
>  	} else {
> -		if (cn10k_mldev->cache_model_data) {
> -			if ((model->type == ML_CNXK_MODEL_TYPE_GLOW &&
> -			     model->subtype == ML_CNXK_MODEL_SUBTYPE_GLOW_MRVL) ||
> -			    (model->type == ML_CNXK_MODEL_TYPE_TVM &&
> -			     model->subtype == ML_CNXK_MODEL_SUBTYPE_TVM_MRVL))
> -				ret = cn10k_ml_cache_model_data(cnxk_mldev, layer);
> -		}
> +		if (cn10k_mldev->cache_model_data)
> +			ret = cn10k_ml_cache_model_data(cnxk_mldev, layer);

Does this change introduce potential issues for models that were
intentionally excluded from caching? The original code explicitly limited
caching to GLOW_MRVL and TVM_MRVL subtypes. If other model types are
incompatible with caching or if cn10k_ml_cache_model_data() was not
designed to handle them, could this cause runtime errors or incorrect
behavior?

Is cn10k_ml_cache_model_data() safe to call for all model types and
subtypes? If not, does this code need validation within the cache function
itself, or should the conditional logic be retained here?

>  	}
>  
>  	return ret;