From: Matthew Auld <matthew.auld@intel.com>
To: Xin Wang <x.wang@intel.com>, igt-dev@lists.freedesktop.org
Subject: Re: [PATCH 3/4] tests/xe_pat: add bo-wb-comp-1way-bind PAT test
Date: Wed, 7 Jan 2026 10:17:45 +0000 [thread overview]
Message-ID: <cacc459b-f946-4f8a-9c7a-4a152dc07284@intel.com> (raw)
In-Reply-To: <20260107052406.21678-4-x.wang@intel.com>
On 07/01/2026 05:24, Xin Wang wrote:
> Add a xe_pat subtest that looks up a WB+compression+1way PAT entry
> from the debugfs PAT table and validates that binding a WB-cached
> BO with that PAT index succeeds on Xe3+.
>
> Skip the test when the platform doesn't expose a suitable PAT entry.
>
> Cc: Matthew Auld <matthew.auld@intel.com>
> Signed-off-by: Xin Wang <x.wang@intel.com>
> ---
> tests/intel/xe_pat.c | 76 ++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 76 insertions(+)
>
> diff --git a/tests/intel/xe_pat.c b/tests/intel/xe_pat.c
> index 8af6ad174..2c98acd2c 100644
> --- a/tests/intel/xe_pat.c
> +++ b/tests/intel/xe_pat.c
> @@ -112,6 +112,43 @@ static int xe_fetch_pat_sw_config(int fd, struct intel_pat_cache *pat_sw_config)
> return parsed;
> }
>
> +static bool find_wb_comp_1way_pat_index(int fd, uint8_t *pat_index_out)
> +{
> + struct intel_pat_cache pat_sw_config = {};
> + int32_t parsed;
> + int i;
> +
> + parsed = xe_fetch_pat_sw_config(fd, &pat_sw_config);
> +
> + for (i = 0; i < parsed; i++) {
> + uint32_t pat = pat_sw_config.entries[i].pat;
> +
> + if (pat_sw_config.entries[i].rsvd)
> + continue;
> +
> + if (!(pat & XE2_COMP_EN))
> + continue;
> +
> + if (REG_FIELD_GET(XE2_COH_MODE, pat) != COH_MODE_1WAY)
> + continue;
> +
> + if (REG_FIELD_GET(XE2_L3_POLICY, pat) != L3_CACHE_POLICY_WB)
> + continue;
> +
> + /*
> + * Matches Xe3 compressed+coherent entry (see kernel xe3_lpg_pat_table[16]).
> + * Note: L4 policy is UC in that entry.
> + */
> + if (REG_FIELD_GET(XE2_L4_POLICY, pat) != L4_CACHE_POLICY_UC)
> + continue;
> +
> + *pat_index_out = i;
> + return true;
> + }
> +
> + return false;
> +}
> +
> /**
> * SUBTEST: pat-sanity
> * Test category: functionality test
> @@ -958,6 +995,42 @@ static void bo_comp_disable_bind(int fd)
> xe_vm_destroy(fd, vm);
> }
>
> +/**
> + * SUBTEST: bo-wb-comp-1way-bind
> + * Test category: functionality test
> + * Description: Validate binding a WB-cached BO using a WB+compression+1way PAT
> + * index (as advertised by the debugfs PAT table) succeeds. Skip if the
> + * platform doesn't expose such a PAT entry.
> + */
> +static void bo_wb_comp_1way_bind(int fd)
> +{
> + size_t size = xe_get_default_alignment(fd);
> + uint16_t dev_id = intel_get_drm_devid(fd);
> + uint8_t wb_comp_pat_index;
> + bool supported;
> + uint32_t vm, bo;
> + int ret;
> +
> + igt_require(intel_get_device_info(dev_id)->graphics_ver >= 30);
> +
> + supported = find_wb_comp_1way_pat_index(fd, &wb_comp_pat_index);
> + igt_require_f(supported, "No WB+compression+1way PAT index found, skipping.\n");
> +
> + vm = xe_vm_create(fd, 0, 0);
> + bo = xe_bo_create_caching(fd, 0, size, system_memory(fd), 0,
> + DRM_XE_GEM_CPU_CACHING_WB);
> +
> + ret = __xe_vm_bind(fd, vm, 0, bo, 0, 0x100000,
> + size, 0, 0, NULL, 0,
> + 0, wb_comp_pat_index, 0);
I think we also need something that actually uses and validates
compression with cpu:wb from hw pov (which is a significant change),
similar to what we do with other interesting PAT modes and adding them
to pat_index_modes[], and then hitting them with blit, render-copy etc.
Otherwise only testing the software bind interface is big test coverage
gap IMO.
I think you were reworking the pat_index_modes stuff somewhere else, so
it might change a bit but in the meantime I think we really need to have
at least something trigger the hw flow with this new mode in this
series. Or does that already exist and I missed it? What are your
thoughts here?
> +
> + igt_assert_eq(ret, 0);
> + xe_vm_unbind_sync(fd, vm, 0, 0x100000, size);
> +
> + gem_close(fd, bo);
> + xe_vm_destroy(fd, vm);
> +}
> +
> /**
> * SUBTEST: userptr-comp
> * Test category: functionality test
> @@ -1461,6 +1534,9 @@ int igt_main_args("V", NULL, help_str, opt_handler, NULL)
> igt_subtest("bo-comp-disable-bind")
> bo_comp_disable_bind(fd);
>
> + igt_subtest("bo-wb-comp-1way-bind")
> + bo_wb_comp_1way_bind(fd);
> +
> igt_subtest_with_dynamic("pat-index-xelp") {
> igt_require(intel_graphics_ver(dev_id) <= IP_VER(12, 55));
> subtest_pat_index_modes_with_regions(fd, xelp_pat_index_modes,
next prev parent reply other threads:[~2026-01-07 10:17 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-07 5:24 [PATCH 0/4] xe_pat: add compression subtests Xin Wang
2026-01-07 5:24 ` [PATCH 1/4] tests/xe_pat: add userptr-comp PAT test Xin Wang
2026-01-07 9:57 ` Matthew Auld
2026-01-07 5:24 ` [PATCH 2/4] tests/xe_pat: add prime-external-import-comp " Xin Wang
2026-01-07 10:01 ` Matthew Auld
2026-01-07 5:24 ` [PATCH 3/4] tests/xe_pat: add bo-wb-comp-1way-bind " Xin Wang
2026-01-07 10:17 ` Matthew Auld [this message]
2026-01-08 6:57 ` Wang, X
2026-01-07 5:24 ` [PATCH 4/4] intel-ci: add xe_pat compression subtests Xin Wang
2026-01-07 6:48 ` ✓ Xe.CI.BAT: success for xe_pat: add " Patchwork
2026-01-07 7:08 ` ✓ i915.CI.BAT: " Patchwork
2026-01-07 8:33 ` ✗ Xe.CI.Full: failure " Patchwork
2026-01-07 9:59 ` ✗ i915.CI.Full: " Patchwork
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=cacc459b-f946-4f8a-9c7a-4a152dc07284@intel.com \
--to=matthew.auld@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=x.wang@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox