From: "Bernatowicz, Marcin" <marcin.bernatowicz@linux.intel.com>
To: "Laguna, Lukasz" <lukasz.laguna@intel.com>,
igt-dev@lists.freedesktop.org
Cc: adam.miszczak@linux.intel.com, jakub1.kolakowski@intel.com,
kamil.konieczny@linux.intel.com
Subject: Re: [PATCH i-g-t 1/4] tests/intel/xe_sriov_vfio: Add module load/unload subtests
Date: Thu, 26 Mar 2026 12:41:55 +0100 [thread overview]
Message-ID: <8d4b9edb-0aee-4cce-b611-b93ba64860de@linux.intel.com> (raw)
In-Reply-To: <756e09f7-c2b9-4529-92b7-ad00ecac3d53@intel.com>
On 3/24/2026 2:54 PM, Laguna, Lukasz wrote:
>
> On 3/18/2026 17:14, Marcin Bernatowicz wrote:
>> Introduce subtests that verify xe_vfio_pci can be loaded and unloaded,
>> while reporting already-loaded or already-unloaded initial state as
>> success. The unload path still skips when xe-vfio-pci has bound devices,
>> to avoid attempting removal while the driver is in use.
>>
>> Signed-off-by: Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com>
>> Cc: Adam Miszczak <adam.miszczak@linux.intel.com>
>> Cc: Jakub Kolakowski <jakub1.kolakowski@intel.com>
>> Cc: Kamil Konieczny <kamil.konieczny@linux.intel.com>
>> Cc: Lukasz Laguna <lukasz.laguna@intel.com>
>> ---
>> tests/intel/xe_sriov_vfio.c | 88 +++++++++++++++++++++++++++++++++++++
>> tests/meson.build | 1 +
>> 2 files changed, 89 insertions(+)
>> create mode 100644 tests/intel/xe_sriov_vfio.c
>>
>> diff --git a/tests/intel/xe_sriov_vfio.c b/tests/intel/xe_sriov_vfio.c
>> new file mode 100644
>> index 000000000..2fe0137e1
>> --- /dev/null
>> +++ b/tests/intel/xe_sriov_vfio.c
>> @@ -0,0 +1,88 @@
>> +// SPDX-License-Identifier: MIT
>> +/*
>> + * Copyright(c) 2026 Intel Corporation. All rights reserved.
>> + */
>> +
>> +#include <ctype.h>
>> +#include <dirent.h>
>> +
>> +#include "igt_core.h"
>> +#include "igt_kmod.h"
>> +
>> +/**
>> + * TEST: xe_sriov_vfio
>> + * Category: Core
>> + * Mega feature: SR-IOV
>> + * Sub-category: SR-IOV VFIO
>> + * Functionality: VFIO module
>> + * Description: Verify basic xe-vfio-pci module operations.
>> + *
>> + * SUBTEST: load-xe-vfio-pci
>> + * Description: Attempt to load xe_vfio_pci module.
>> + *
>> + * SUBTEST: unload-xe-vfio-pci
>> + * Description: Attempt to unload xe_vfio_pci module (skips if busy).
>> + */
>> +
>> +IGT_TEST_DESCRIPTION("Xe SR-IOV VFIO tests (xe-vfio-pci)");
>> +
>> +static const char *XE_VFIO_PCI_MOD = "xe_vfio_pci";
>> +static const char *XE_VFIO_PCI_DRV = "xe-vfio-pci";
>> +
>> +static bool pci_driver_has_any_bound_devices(const char *driver)
>> +{
>> + char path[PATH_MAX];
>> + DIR *dir;
>> + struct dirent *de;
>> +
>> + snprintf(path, sizeof(path), "/sys/bus/pci/drivers/%s", driver);
>> + dir = opendir(path);
>> + if (!dir)
>> + return false;
>> +
>> + while ((de = readdir(dir))) {
>> + if (de->d_type != DT_LNK)
>> + continue;
>> + if (!isdigit(de->d_name[0]))
>> + continue;
>> +
>> + closedir(dir);
>> + return true;
>> + }
>> +
>> + closedir(dir);
>> + return false;
>> +}
>> +
>> +int igt_main()
>> +{
>> + igt_describe("Attempt to load xe_vfio_pci module.");
>> + igt_subtest("load-xe-vfio-pci") {
>> + int ret;
>> +
>> + if (igt_kmod_is_loaded(XE_VFIO_PCI_MOD))
>> + igt_info("%s already loaded\n", XE_VFIO_PCI_MOD);
>
> I think it would be better to skip in such case.
This may be problematic in CI and depends on tests execution order and
configuration.
> And igt_kmod_load will print "Module %s already inserted\n", so this
> info seems to be redundant anyway.
igt_kmod_load does not print that info if no options are passed, which
is our case.
>
>
>> +
>> + ret = igt_kmod_load(XE_VFIO_PCI_MOD, NULL);
>> + igt_assert_f(ret == 0, "Failed to load %s (%d)\n",
>> XE_VFIO_PCI_MOD, ret);
>> + igt_assert(igt_kmod_is_loaded(XE_VFIO_PCI_MOD));
>> + }
>> +
>> + igt_describe("Attempt to unload xe_vfio_pci module (skips if
>> busy).");
>> + igt_subtest("unload-xe-vfio-pci") {
>> + int ret;
>> +
>> + if (!igt_kmod_is_loaded(XE_VFIO_PCI_MOD)) {
>> + igt_info("%s already unloaded\n", XE_VFIO_PCI_MOD);
>> + igt_success();
>
> Maybe it would be better to have one subtest "xe-vfio-pci-load-unload" ?
If we do not need load/unload tests leaving the driver loaded/unloaded
then we can do this
and simplify some other parts.
>> + }
>> +
>> + if (pci_driver_has_any_bound_devices(XE_VFIO_PCI_DRV))
>> + igt_skip("%s has bound devices; not attempting
>> unload\n", XE_VFIO_PCI_DRV);
>> +
>> + ret = igt_kmod_unload(XE_VFIO_PCI_MOD);
>> + igt_assert_f(ret == 0, "Failed to unload %s (%d)\n",
>> XE_VFIO_PCI_MOD, ret);
>> + igt_assert_f(!igt_kmod_is_loaded(XE_VFIO_PCI_MOD), "%s still
>> loaded\n",
>> + XE_VFIO_PCI_MOD);
>> + }
>> +}
>> diff --git a/tests/meson.build b/tests/meson.build
>> index cecb4a8ae..26d9345ec 100644
>> --- a/tests/meson.build
>> +++ b/tests/meson.build
>> @@ -337,6 +337,7 @@ intel_xe_progs = [
>> 'xe_sriov_admin',
>> 'xe_sriov_auto_provisioning',
>> 'xe_sriov_flr',
>> + 'xe_sriov_vfio',
>> 'xe_sriov_scheduling',
>> 'xe_survivability',
>> 'xe_sriov_vram',
next prev parent reply other threads:[~2026-03-26 11:42 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-18 16:14 [PATCH i-g-t 0/4] tests/intel/xe_sriov_vfio: Add basic VFIO coverage Marcin Bernatowicz
2026-03-18 16:14 ` [PATCH i-g-t 1/4] tests/intel/xe_sriov_vfio: Add module load/unload subtests Marcin Bernatowicz
2026-03-24 13:54 ` Laguna, Lukasz
2026-03-26 11:41 ` Bernatowicz, Marcin [this message]
2026-03-18 16:14 ` [PATCH i-g-t 2/4] tests/intel/xe_sriov_vfio: Add dynamic bind-unbind-vfs subtest Marcin Bernatowicz
2026-03-24 14:14 ` Laguna, Lukasz
2026-03-26 11:49 ` Bernatowicz, Marcin
2026-03-18 16:14 ` [PATCH i-g-t 3/4] tests/intel/xe_sriov_vfio: Add open-basic subtest Marcin Bernatowicz
2026-03-24 14:17 ` Laguna, Lukasz
2026-03-26 11:44 ` Bernatowicz, Marcin
2026-03-18 16:14 ` [PATCH i-g-t 4/4] tests/intel/xe_sriov_vfio: Add region-info subtest Marcin Bernatowicz
2026-03-18 18:00 ` ✓ Xe.CI.BAT: success for tests/intel/xe_sriov_vfio: Add basic VFIO coverage Patchwork
2026-03-18 18:17 ` ✓ i915.CI.BAT: " Patchwork
2026-03-19 23:18 ` ✗ i915.CI.Full: failure " Patchwork
2026-03-20 6:16 ` ✗ Xe.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=8d4b9edb-0aee-4cce-b611-b93ba64860de@linux.intel.com \
--to=marcin.bernatowicz@linux.intel.com \
--cc=adam.miszczak@linux.intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=jakub1.kolakowski@intel.com \
--cc=kamil.konieczny@linux.intel.com \
--cc=lukasz.laguna@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