From: Rodrigo Vivi <rodrigo.vivi@intel.com>
To: "Michal Wajdeczko" <michal.wajdeczko@intel.com>,
"Thomas Hellström" <thomas.hellstrom@linux.intel.com>,
"Lucas De Marchi" <lucas.demarchi@intel.com>
Cc: <intel-xe@lists.freedesktop.org>,
Lucas De Marchi <lucas.demarchi@intel.com>
Subject: Re: [PATCH 13/14] drm/xe/pf: Allow to stop and reset VF using sysfs
Date: Fri, 24 Oct 2025 15:51:39 -0400 [thread overview]
Message-ID: <aPvYy5EH7KOYlqen@intel.com> (raw)
In-Reply-To: <20251020182414.576-14-michal.wajdeczko@intel.com>
On Mon, Oct 20, 2025 at 08:24:13PM +0200, Michal Wajdeczko wrote:
> It is expected that VFs activity will be monitored and in some
> cases admin might want to silence specific VF without killing
> the VM where it was attached.
>
> Add write-only attributes to control GuC scheduling at VF level.
>
> /sys/bus/pci/drivers/xe/BDF/
> ├── sriov_admin/
> ├── vf1/
> │ ├── reset [WO] bool
This reset is the only thing in this series that I'm not 100% comfortable.
in the device symlink you already have the reset that does the right thing.
If I understood correctly this one here has the same behavior, but with
slightly different path... bypassing the pci subsystem.
I understand that your goal is to have it near the stop since you have to
stop and then reset. But It is still a bit strange.
If you get our only PF FLR supported case, we do need to unbind the driver
before echo 1 > /sys/../device/reset and still the unbind is not a file
near the the reset.
So we can perhaps live without this double reset here.
Although not a big blocker from my side....
Lucas, Thomas, thoughts?
> │ └── stop [WO] bool
> ├── vf2/
> │ ├── reset [WO] bool
> │ └── stop [WO] bool
>
> Writing "1" or "y" (or whatever is recognized by the strtobool()
> function) to these files will trigger the change of the VF state
> to STOP (GuC will stop servicing the VF) or back to READY (GuC
> will start servicing this VF again but after completing the full
> sequence of the software initiated VF FLR).
>
> Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
> Cc: Lucas De Marchi <lucas.demarchi@intel.com>
> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
> ---
> drivers/gpu/drm/xe/xe_sriov_pf_sysfs.c | 52 ++++++++++++++++++++++++++
> 1 file changed, 52 insertions(+)
>
> diff --git a/drivers/gpu/drm/xe/xe_sriov_pf_sysfs.c b/drivers/gpu/drm/xe/xe_sriov_pf_sysfs.c
> index c68acf6cd34f..f872013313c1 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_pf_sysfs.c
> +++ b/drivers/gpu/drm/xe/xe_sriov_pf_sysfs.c
> @@ -13,6 +13,7 @@
> #include "xe_pm.h"
> #include "xe_sriov.h"
> #include "xe_sriov_pf.h"
> +#include "xe_sriov_pf_control.h"
> #include "xe_sriov_pf_helpers.h"
> #include "xe_sriov_pf_provision.h"
> #include "xe_sriov_pf_sysfs.h"
> @@ -52,6 +53,8 @@ static int emit_choice(char *buf, int choice, const char * const *array, size_t
> * ├── vf1/
> * │ ├── ...
> * │ ├── device -> ../../../BDF.1
> + * │ ├── reset
> + * │ ├── stop
> * │ └── profile
> * │ ├── exec_quantum_ms
> * │ ├── preempt_timeout_us
> @@ -291,8 +294,57 @@ static const struct attribute_group profile_vf_attr_group = {
> .is_visible = profile_vf_attr_is_visible,
> };
>
> +#define DEFINE_SIMPLE_CONTROL_SRIOV_VF_ATTR(NAME) \
> + \
> +static ssize_t xe_sriov_vf_attr_##NAME##_store(struct xe_device *xe, unsigned int vfid, \
> + const char *buf, size_t count) \
> +{ \
> + bool yes; \
> + int err; \
> + \
> + if (!vfid) \
> + return -EPERM; \
> + \
> + err = kstrtobool(buf, &yes); \
> + if (err) \
> + return err; \
> + if (!yes) \
> + return count; \
> + \
> + err = xe_sriov_pf_control_##NAME##_vf(xe, vfid); \
> + return err ?: count; \
> +} \
> + \
> +static XE_SRIOV_VF_ATTR_WO(NAME)
> +
> +DEFINE_SIMPLE_CONTROL_SRIOV_VF_ATTR(stop);
> +DEFINE_SIMPLE_CONTROL_SRIOV_VF_ATTR(reset);
> +
> +static struct attribute *control_vf_attrs[] = {
> + &xe_sriov_vf_attr_stop.attr,
> + &xe_sriov_vf_attr_reset.attr,
> + NULL
> +};
> +
> +static umode_t control_vf_attr_is_visible(struct kobject *kobj,
> + struct attribute *attr, int index)
> +{
> + struct xe_sriov_kobj *vkobj = to_xe_sriov_kobj(kobj);
> +
> + if (vkobj->vfid == PFID)
> + return 0;
> +
> + return attr->mode;
> +}
> +
> +static const struct attribute_group control_vf_attr_group = {
> + .attrs = control_vf_attrs,
> + .is_visible = control_vf_attr_is_visible,
> +};
> +
> static const struct attribute_group *xe_sriov_vf_attr_groups[] = {
> &profile_vf_attr_group,
> + &control_vf_attr_group,
> NULL
> };
>
> --
> 2.47.1
>
next prev parent reply other threads:[~2025-10-24 19:51 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-20 18:24 [PATCH 00/14] PF: Add sriov_admin sysfs tree Michal Wajdeczko
2025-10-20 18:24 ` [PATCH 01/14] drm/xe/pf: Prepare sysfs for SR-IOV admin attributes Michal Wajdeczko
2025-10-24 19:43 ` Rodrigo Vivi
2025-10-27 17:11 ` Lucas De Marchi
2025-10-27 17:59 ` Michal Wajdeczko
2025-10-27 18:30 ` Lucas De Marchi
2025-10-20 18:24 ` [PATCH 02/14] drm/xe/pf: Take RPM during calls to SR-IOV attr.store() Michal Wajdeczko
2025-10-27 17:14 ` Lucas De Marchi
2025-10-20 18:24 ` [PATCH 03/14] drm/xe/pf: Allow change PF and VFs EQ/PT using sysfs Michal Wajdeczko
2025-10-24 19:45 ` Rodrigo Vivi
2025-10-27 17:27 ` Lucas De Marchi
2025-10-27 18:09 ` Michal Wajdeczko
2025-10-27 18:32 ` Lucas De Marchi
2025-10-20 18:24 ` [PATCH 04/14] drm/xe/pf: Relax report helper to accept PF in bulk configs Michal Wajdeczko
2025-10-27 18:50 ` Lucas De Marchi
2025-10-20 18:24 ` [PATCH 05/14] drm/xe/pf: Add functions to bulk configure EQ/PT on GT Michal Wajdeczko
2025-10-27 19:03 ` Lucas De Marchi
2025-10-27 20:12 ` Michal Wajdeczko
2025-10-20 18:24 ` [PATCH 06/14] drm/xe/pf: Add functions to bulk provision EQ/PT Michal Wajdeczko
2025-10-27 19:18 ` Lucas De Marchi
2025-10-20 18:24 ` [PATCH 07/14] drm/xe/pf: Allow bulk change all VFs EQ/PT using sysfs Michal Wajdeczko
2025-10-24 19:46 ` Rodrigo Vivi
2025-10-27 19:28 ` Lucas De Marchi
2025-10-27 20:15 ` Michal Wajdeczko
2025-10-20 18:24 ` [PATCH 08/14] drm/xe/pf: Add functions to provision scheduling priority Michal Wajdeczko
2025-10-28 11:17 ` Piotr Piórkowski
2025-10-20 18:24 ` [PATCH 09/14] drm/xe/pf: Allow bulk change all VFs priority using sysfs Michal Wajdeczko
2025-10-24 19:47 ` Rodrigo Vivi
2025-10-20 18:24 ` [PATCH 10/14] drm/xe/pf: Allow change PF scheduling " Michal Wajdeczko
2025-10-24 19:47 ` Rodrigo Vivi
2025-10-20 18:24 ` [PATCH 11/14] drm/xe/pf: Promote xe_pci_sriov_get_vf_pdev Michal Wajdeczko
2025-10-28 9:57 ` Piotr Piórkowski
2025-10-28 12:22 ` Michal Wajdeczko
2025-10-28 16:03 ` Piotr Piórkowski
2025-10-20 18:24 ` [PATCH 12/14] drm/xe/pf: Add sysfs device symlinks to enabled VFs Michal Wajdeczko
2025-10-24 19:47 ` Rodrigo Vivi
2025-10-20 18:24 ` [PATCH 13/14] drm/xe/pf: Allow to stop and reset VF using sysfs Michal Wajdeczko
2025-10-24 19:51 ` Rodrigo Vivi [this message]
2025-10-27 20:58 ` Lucas De Marchi
2025-10-20 18:24 ` [PATCH 14/14] drm/xe/pf: Add documentation for sriov_admin attributes Michal Wajdeczko
2025-10-27 16:44 ` Rodrigo Vivi
2025-10-21 4:35 ` ✗ CI.checkpatch: warning for PF: Add sriov_admin sysfs tree Patchwork
2025-10-21 4:36 ` ✓ CI.KUnit: success " Patchwork
2025-10-21 10:19 ` ✗ Xe.CI.Full: failure " 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=aPvYy5EH7KOYlqen@intel.com \
--to=rodrigo.vivi@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=lucas.demarchi@intel.com \
--cc=michal.wajdeczko@intel.com \
--cc=thomas.hellstrom@linux.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.