Re: [PATCH v2 2/3] tools/testing/cxl: Skip the auto region when no volatile window is active

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] cxl_test: mock HMEM and region APIs still hardcode inactive mock_cfmws[0]
--

commit ab47ae118b5deca13e8e911d3065f4b4d5bbbf01
Author: Alison Schofield <alison.schofield@intel.com>

tools/testing/cxl: Skip the auto region when no volatile window is active

This commit updates the mock auto region in the CXL testing infrastructure to
dynamically locate an active volatile window instead of hardcoding CFMWS0.
It sets up the auto region only when a volatile window is available, avoiding
issues where endpoint decoders are programmed at HPA 0 during XOR interleave
tests.

> diff --git a/tools/testing/cxl/test/cxl.c b/tools/testing/cxl/test/cxl.c
> index aa140307e0c3c..46a87633e9621 100644
> --- a/tools/testing/cxl/test/cxl.c
> +++ b/tools/testing/cxl/test/cxl.c
> @@ -423,6 +423,18 @@ struct acpi_cedt_cxims *mock_cxims[1] = {
>  	[0] = &mock_cedt.cxims0.cxims,
>  };
>  
> +static struct acpi_cedt_cfmws *mock_auto_region_cfmws(void)
> +{
> +	for (int i = cfmws_start; i <= cfmws_end; i++) {
> +		struct acpi_cedt_cfmws *window = mock_cfmws[i];
> +
> +		if (window->restrictions & ACPI_CEDT_CFMWS_RESTRICT_VOLATILE)
> +			return window;
> +	}
> +
> +	return NULL;
> +}
> +

[ ... ]

> @@ -1135,9 +1154,9 @@ static bool mock_decoder_handle_saved(struct cxl_decoder *cxld, struct cxl_test_
>  }
>  
>  static void mock_init_hdm_type2_cxled(struct cxl_endpoint_decoder *cxled,
> -				      struct cxl_port *port)
> +				      struct cxl_port *port,
> +				      struct acpi_cedt_cfmws *window)
>  {
> -	struct acpi_cedt_cfmws *window = mock_cfmws[0];
>  	struct cxl_decoder *cxld = &cxled->cxld;

[Severity: Medium]
With this patch introducing `mock_auto_region_cfmws()` and removing the
hardcoded `mock_cfmws[0]` here, should the mock HMEM and region APIs also be
updated to use the dynamic window?

Looking at `mock_walk_hmem_resources()`, `mock_region_intersects()`, and
`mock_region_intersects_soft_reserve()`, they all still unconditionally
reference `mock_cfmws[0]`:

tools/testing/cxl/test/cxl.c:mock_walk_hmem_resources() {
	struct acpi_cedt_cfmws *cfmws = mock_cfmws[0];
	struct resource window =
		DEFINE_RES_MEM(cfmws->base_hpa, cfmws->window_size / 2);
    ...
}

If `interleave_arithmetic=1` (XOR tests) is active, `mock_cfmws[0]` is inactive
and its HPA base is left as 0. Will these functions falsely report soft-reserved
memory and CXL intersections at HPA 0 instead of the real volatile window?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/cover.1791582353.git.alison.schofield@intel.com?part=2