Re: [PATCH 03/26] drm/xe/pf: Add save/restore control state stubs and connect to debugfs

[Michal Wajdeczko <michal.wajdeczko@intel.com>]

On 10/11/2025 9:38 PM, Michał Winiarski wrote:
> The states will be used by upcoming changes to produce (in case of save)
> or consume (in case of resume) the VF migration data.
> 
> Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
> ---
>  drivers/gpu/drm/xe/xe_gt_sriov_pf_control.c   | 270 ++++++++++++++++++
>  drivers/gpu/drm/xe/xe_gt_sriov_pf_control.h   |   6 +
>  .../gpu/drm/xe/xe_gt_sriov_pf_control_types.h |  17 ++
>  drivers/gpu/drm/xe/xe_sriov_pf_control.c      |  96 +++++++
>  drivers/gpu/drm/xe/xe_sriov_pf_control.h      |   4 +
>  drivers/gpu/drm/xe/xe_sriov_pf_debugfs.c      |  38 +++
>  6 files changed, 431 insertions(+)
> 
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_control.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_control.c
> index 2e6bd3d1fe1da..44df984278548 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_control.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_control.c
> @@ -184,6 +184,13 @@ static const char *control_bit_to_string(enum xe_gt_sriov_control_bits bit)
>  	CASE2STR(PAUSE_SAVE_GUC);
>  	CASE2STR(PAUSE_FAILED);
>  	CASE2STR(PAUSED);
> +	CASE2STR(MIGRATION_DATA_WIP);
> +	CASE2STR(SAVE_WIP);
> +	CASE2STR(SAVE_FAILED);
> +	CASE2STR(SAVED);
> +	CASE2STR(RESTORE_WIP);
> +	CASE2STR(RESTORE_FAILED);
> +	CASE2STR(RESTORED);
>  	CASE2STR(RESUME_WIP);
>  	CASE2STR(RESUME_SEND_RESUME);
>  	CASE2STR(RESUME_FAILED);
> @@ -207,6 +214,8 @@ static unsigned long pf_get_default_timeout(enum xe_gt_sriov_control_bits bit)
>  		return HZ / 2;
>  	case XE_GT_SRIOV_STATE_FLR_WIP:
>  	case XE_GT_SRIOV_STATE_FLR_RESET_CONFIG:
> +	case XE_GT_SRIOV_STATE_SAVE_WIP:
> +	case XE_GT_SRIOV_STATE_RESTORE_WIP:
>  		return 5 * HZ;
>  	default:
>  		return HZ;
> @@ -359,6 +368,10 @@ static void pf_queue_vf(struct xe_gt *gt, unsigned int vfid)
>  
>  static void pf_exit_vf_flr_wip(struct xe_gt *gt, unsigned int vfid);
>  static void pf_exit_vf_stop_wip(struct xe_gt *gt, unsigned int vfid);
> +static void pf_exit_vf_save_wip(struct xe_gt *gt, unsigned int vfid);
> +static void pf_exit_vf_saved(struct xe_gt *gt, unsigned int vfid);
> +static void pf_exit_vf_restore_wip(struct xe_gt *gt, unsigned int vfid);
> +static void pf_exit_vf_restored(struct xe_gt *gt, unsigned int vfid);
>  static void pf_exit_vf_pause_wip(struct xe_gt *gt, unsigned int vfid);
>  static void pf_exit_vf_resume_wip(struct xe_gt *gt, unsigned int vfid);
>  
> @@ -380,6 +393,8 @@ static void pf_exit_vf_wip(struct xe_gt *gt, unsigned int vfid)
>  
>  		pf_exit_vf_flr_wip(gt, vfid);
>  		pf_exit_vf_stop_wip(gt, vfid);
> +		pf_exit_vf_save_wip(gt, vfid);
> +		pf_exit_vf_restore_wip(gt, vfid);
>  		pf_exit_vf_pause_wip(gt, vfid);
>  		pf_exit_vf_resume_wip(gt, vfid);
>  
> @@ -399,6 +414,8 @@ static void pf_enter_vf_ready(struct xe_gt *gt, unsigned int vfid)
>  	pf_exit_vf_state(gt, vfid, XE_GT_SRIOV_STATE_PAUSED);
>  	pf_exit_vf_state(gt, vfid, XE_GT_SRIOV_STATE_STOPPED);
>  	pf_exit_vf_state(gt, vfid, XE_GT_SRIOV_STATE_RESUMED);
> +	pf_exit_vf_saved(gt, vfid);
> +	pf_exit_vf_restored(gt, vfid);
>  	pf_exit_vf_mismatch(gt, vfid);
>  	pf_exit_vf_wip(gt, vfid);
>  }
> @@ -675,6 +692,8 @@ static void pf_enter_vf_resumed(struct xe_gt *gt, unsigned int vfid)
>  {
>  	pf_enter_vf_state(gt, vfid, XE_GT_SRIOV_STATE_RESUMED);
>  	pf_exit_vf_state(gt, vfid, XE_GT_SRIOV_STATE_PAUSED);
> +	pf_exit_vf_saved(gt, vfid);
> +	pf_exit_vf_restored(gt, vfid);
>  	pf_exit_vf_mismatch(gt, vfid);
>  	pf_exit_vf_wip(gt, vfid);
>  }
> @@ -776,6 +795,249 @@ int xe_gt_sriov_pf_control_resume_vf(struct xe_gt *gt, unsigned int vfid)
>  	return -ECANCELED;
>  }
>  
> +/**
> + * xe_gt_sriov_pf_control_check_vf_data_wip - check if new SR-IOV VF migration data is expected

nit: add () to function name:

	xe_gt_sriov_pf_control_check_vf_data_wip() - Check ...

> + * @gt: the &struct xe_gt
> + * @vfid: the VF identifier
> + *
> + * This function is for PF only.
> + *
> + * Return: true when new migration data is expected to be produced, false otherwise
> + */
> +bool xe_gt_sriov_pf_control_check_vf_data_wip(struct xe_gt *gt, unsigned int vfid)
> +{
> +	return pf_check_vf_state(gt, vfid, XE_GT_SRIOV_STATE_MIGRATION_DATA_WIP);
> +}
> +
> +static void pf_exit_vf_save_wip(struct xe_gt *gt, unsigned int vfid)
> +{
> +	pf_exit_vf_state(gt, vfid, XE_GT_SRIOV_STATE_SAVE_WIP);
> +}
> +
> +static void pf_enter_vf_saved(struct xe_gt *gt, unsigned int vfid)
> +{
> +	if (!pf_enter_vf_state(gt, vfid, XE_GT_SRIOV_STATE_SAVED))
> +		pf_enter_vf_state_machine_bug(gt, vfid);
> +
> +	xe_gt_sriov_info(gt, "VF%u saved!\n", vfid);
> +
> +	pf_exit_vf_mismatch(gt, vfid);
> +	pf_exit_vf_wip(gt, vfid);
> +}
> +
> +static void pf_exit_vf_saved(struct xe_gt *gt, unsigned int vfid)
> +{
> +	pf_exit_vf_state(gt, vfid, XE_GT_SRIOV_STATE_SAVED);
> +}
> +
> +static bool pf_handle_vf_save_wip(struct xe_gt *gt, unsigned int vfid)
> +{
> +	if (!pf_check_vf_state(gt, vfid, XE_GT_SRIOV_STATE_SAVE_WIP))
> +		return false;
> +
> +	pf_exit_vf_save_wip(gt, vfid);
> +	pf_enter_vf_saved(gt, vfid);
> +
> +	return true;
> +}
> +
> +static bool pf_enter_vf_save_wip(struct xe_gt *gt, unsigned int vfid)
> +{
> +	if (pf_enter_vf_state(gt, vfid, XE_GT_SRIOV_STATE_SAVE_WIP)) {
> +		pf_enter_vf_state(gt, vfid, XE_GT_SRIOV_STATE_MIGRATION_DATA_WIP);
> +		pf_exit_vf_restored(gt, vfid);
> +		pf_enter_vf_wip(gt, vfid);
> +		pf_queue_vf(gt, vfid);
> +		return true;
> +	}
> +
> +	return false;
> +}
> +

please add diagram for the inner state machines for both new SAVE/RESTORE states

and also update master diagram how SAVE/RESTORE interacts with existing states

with diagrams that include expected flows it will be easier to review the impl

> +/**
> + * xe_gt_sriov_pf_control_save_vf - Save SR-IOV VF migration data.
> + * @gt: the &struct xe_gt
> + * @vfid: the VF identifier
> + *
> + * This function is for PF only.
> + *
> + * Return: 0 on success or a negative error code on failure.
> + */
> +int xe_gt_sriov_pf_control_save_vf(struct xe_gt *gt, unsigned int vfid)
> +{
> +	if (pf_check_vf_state(gt, vfid, XE_GT_SRIOV_STATE_STOPPED)) {
> +		xe_gt_sriov_dbg(gt, "VF%u is stopped!\n", vfid);
> +		return -EPERM;
> +	}
> +
> +	if (!pf_check_vf_state(gt, vfid, XE_GT_SRIOV_STATE_PAUSED)) {
> +		xe_gt_sriov_dbg(gt, "VF%u is not paused!\n", vfid);
> +		return -EPERM;
> +	}
> +
> +	if (pf_check_vf_state(gt, vfid, XE_GT_SRIOV_STATE_RESTORE_WIP)) {
> +		xe_gt_sriov_dbg(gt, "VF%u restore is in progress!\n", vfid);
> +		return -EPERM;
> +	}
> +
> +	if (!pf_enter_vf_save_wip(gt, vfid)) {
> +		xe_gt_sriov_dbg(gt, "VF%u save already in progress!\n", vfid);
> +		return -EALREADY;
> +	}
> +
> +	return 0;
> +}
> +
> +static int pf_wait_vf_save_done(struct xe_gt *gt, unsigned int vfid)
> +{
> +	unsigned long timeout = pf_get_default_timeout(XE_GT_SRIOV_STATE_SAVE_WIP);
> +	int err;
> +
> +	err = pf_wait_vf_wip_done(gt, vfid, timeout);
> +	if (err) {
> +		xe_gt_sriov_notice(gt, "VF%u SAVE didn't finish in %u ms (%pe)\n",
> +				   vfid, jiffies_to_msecs(timeout), ERR_PTR(err));
> +		return err;
> +	}
> +
> +	if (!pf_expect_vf_not_state(gt, vfid, XE_GT_SRIOV_STATE_SAVE_FAILED))
> +		return -EIO;
> +
> +	return 0;
> +}
> +
> +/**
> + * xe_gt_sriov_pf_control_wait_save_done() - Wait for a VF Save to complete
> + * @gt: the &struct xe_gt
> + * @vfid: the VF identifier
> + *
> + * This function is for PF only.
> + *
> + * Return: 0 on success or a negative error code on failure.
> + */
> +int xe_gt_sriov_pf_control_wait_save_done(struct xe_gt *gt, unsigned int vfid)
> +{
> +	if (!pf_check_vf_state(gt, vfid, XE_GT_SRIOV_STATE_SAVE_WIP))
> +		return 0;
> +
> +	return pf_wait_vf_save_done(gt, vfid);
> +}
> +
> +static void pf_exit_vf_restore_wip(struct xe_gt *gt, unsigned int vfid)
> +{
> +	pf_expect_vf_not_state(gt, vfid, XE_GT_SRIOV_STATE_MIGRATION_DATA_WIP);
> +	pf_exit_vf_state(gt, vfid, XE_GT_SRIOV_STATE_RESTORE_WIP);
> +}
> +
> +static void pf_enter_vf_restored(struct xe_gt *gt, unsigned int vfid)
> +{
> +	if (!pf_enter_vf_state(gt, vfid, XE_GT_SRIOV_STATE_RESTORED))
> +		pf_enter_vf_state_machine_bug(gt, vfid);
> +
> +	xe_gt_sriov_info(gt, "VF%u restored!\n", vfid);

since commit ac43294e8ec2 for GT-level state changes we use dbg() messages

> +
> +	pf_exit_vf_mismatch(gt, vfid);
> +	pf_exit_vf_wip(gt, vfid);
> +}
> +
> +static void pf_exit_vf_restored(struct xe_gt *gt, unsigned int vfid)
> +{
> +	pf_exit_vf_state(gt, vfid, XE_GT_SRIOV_STATE_RESTORED);
> +}
> +
> +static bool pf_handle_vf_restore_wip(struct xe_gt *gt, unsigned int vfid)
> +{
> +	if (!pf_check_vf_state(gt, vfid, XE_GT_SRIOV_STATE_RESTORE_WIP))
> +		return false;
> +
> +	pf_exit_vf_restore_wip(gt, vfid);
> +	pf_enter_vf_restored(gt, vfid);
> +
> +	return true;
> +}
> +
> +static bool pf_enter_vf_restore_wip(struct xe_gt *gt, unsigned int vfid)
> +{
> +	if (pf_enter_vf_state(gt, vfid, XE_GT_SRIOV_STATE_RESTORE_WIP)) {
> +		pf_enter_vf_state(gt, vfid, XE_GT_SRIOV_STATE_MIGRATION_DATA_WIP);
> +		pf_exit_vf_saved(gt, vfid);
> +		pf_enter_vf_wip(gt, vfid);
> +		pf_enter_vf_restored(gt, vfid);
> +		return true;
> +	}
> +
> +	return false;
> +}
> +
> +/**
> + * xe_gt_sriov_pf_control_restore_vf - Restore SR-IOV VF migration data.
> + * @gt: the &struct xe_gt
> + * @vfid: the VF identifier
> + *
> + * This function is for PF only.
> + *
> + * Return: 0 on success or a negative error code on failure.
> + */
> +int xe_gt_sriov_pf_control_restore_vf(struct xe_gt *gt, unsigned int vfid)
> +{
> +	if (pf_check_vf_state(gt, vfid, XE_GT_SRIOV_STATE_STOPPED)) {
> +		xe_gt_sriov_dbg(gt, "VF%u is stopped!\n", vfid);
> +		return -EPERM;
> +	}
> +
> +	if (!pf_check_vf_state(gt, vfid, XE_GT_SRIOV_STATE_PAUSED)) {
> +		xe_gt_sriov_dbg(gt, "VF%u is not paused!\n", vfid);
> +		return -EPERM;
> +	}
> +
> +	if (pf_check_vf_state(gt, vfid, XE_GT_SRIOV_STATE_SAVE_WIP)) {
> +		xe_gt_sriov_dbg(gt, "VF%u save is in progress!\n", vfid);
> +		return -EPERM;
> +	}
> +
> +	if (!pf_enter_vf_restore_wip(gt, vfid)) {
> +		xe_gt_sriov_dbg(gt, "VF%u restore already in progress!\n", vfid);
> +		return -EALREADY;
> +	}
> +
> +	return 0;
> +}
> +
> +static int pf_wait_vf_restore_done(struct xe_gt *gt, unsigned int vfid)
> +{
> +	unsigned long timeout = pf_get_default_timeout(XE_GT_SRIOV_STATE_RESTORE_WIP);
> +	int err;
> +
> +	err = pf_wait_vf_wip_done(gt, vfid, timeout);
> +	if (err) {
> +		xe_gt_sriov_notice(gt, "VF%u RESTORE didn't finish in %u ms (%pe)\n",
> +				   vfid, jiffies_to_msecs(timeout), ERR_PTR(err));
> +		return err;
> +	}
> +
> +	if (!pf_expect_vf_not_state(gt, vfid, XE_GT_SRIOV_STATE_RESTORE_FAILED))
> +		return -EIO;
> +
> +	return 0;
> +}
> +
> +/**
> + * xe_gt_sriov_pf_control_wait_restore_done() - Wait for a VF Restore to complete
> + * @gt: the &struct xe_gt
> + * @vfid: the VF identifier
> + *
> + * This function is for PF only.
> + *
> + * Return: 0 on success or a negative error code on failure.
> + */
> +int xe_gt_sriov_pf_control_wait_restore_done(struct xe_gt *gt, unsigned int vfid)
> +{
> +	if (!pf_check_vf_state(gt, vfid, XE_GT_SRIOV_STATE_RESTORE_WIP))
> +		return 0;
> +
> +	return pf_wait_vf_restore_done(gt, vfid);
> +}
> +
>  /**
>   * DOC: The VF STOP state machine
>   *
> @@ -817,6 +1079,8 @@ static void pf_enter_vf_stopped(struct xe_gt *gt, unsigned int vfid)
>  
>  	pf_exit_vf_state(gt, vfid, XE_GT_SRIOV_STATE_RESUMED);
>  	pf_exit_vf_state(gt, vfid, XE_GT_SRIOV_STATE_PAUSED);
> +	pf_exit_vf_saved(gt, vfid);
> +	pf_exit_vf_restored(gt, vfid);
>  	pf_exit_vf_mismatch(gt, vfid);
>  	pf_exit_vf_wip(gt, vfid);
>  }
> @@ -1461,6 +1725,12 @@ static bool pf_process_vf_state_machine(struct xe_gt *gt, unsigned int vfid)
>  	if (pf_exit_vf_pause_save_guc(gt, vfid))
>  		return true;
>  
> +	if (pf_handle_vf_save_wip(gt, vfid))
> +		return true;
> +
> +	if (pf_handle_vf_restore_wip(gt, vfid))
> +		return true;
> +
>  	if (pf_exit_vf_resume_send_resume(gt, vfid))
>  		return true;
>  
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_control.h b/drivers/gpu/drm/xe/xe_gt_sriov_pf_control.h
> index 8a72ef3778d47..2e121e8132dcf 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_control.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_control.h
> @@ -14,8 +14,14 @@ struct xe_gt;
>  int xe_gt_sriov_pf_control_init(struct xe_gt *gt);
>  void xe_gt_sriov_pf_control_restart(struct xe_gt *gt);
>  
> +bool xe_gt_sriov_pf_control_check_vf_data_wip(struct xe_gt *gt, unsigned int vfid);
> +
>  int xe_gt_sriov_pf_control_pause_vf(struct xe_gt *gt, unsigned int vfid);
>  int xe_gt_sriov_pf_control_resume_vf(struct xe_gt *gt, unsigned int vfid);
> +int xe_gt_sriov_pf_control_save_vf(struct xe_gt *gt, unsigned int vfid);
> +int xe_gt_sriov_pf_control_wait_save_done(struct xe_gt *gt, unsigned int vfid);
> +int xe_gt_sriov_pf_control_restore_vf(struct xe_gt *gt, unsigned int vfid);
> +int xe_gt_sriov_pf_control_wait_restore_done(struct xe_gt *gt, unsigned int vfid);
>  int xe_gt_sriov_pf_control_stop_vf(struct xe_gt *gt, unsigned int vfid);
>  int xe_gt_sriov_pf_control_trigger_flr(struct xe_gt *gt, unsigned int vfid);
>  int xe_gt_sriov_pf_control_sync_flr(struct xe_gt *gt, unsigned int vfid, bool sync);
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_control_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_pf_control_types.h
> index c80b7e77f1ad2..02b517533ee8a 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_control_types.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_control_types.h
> @@ -31,6 +31,13 @@
>   * @XE_GT_SRIOV_STATE_PAUSE_SAVE_GUC: indicates that the PF needs to save the VF GuC state.
>   * @XE_GT_SRIOV_STATE_PAUSE_FAILED: indicates that a VF pause operation has failed.
>   * @XE_GT_SRIOV_STATE_PAUSED: indicates that the VF is paused.
> + * @XE_GT_SRIOV_STATE_MIGRATION_DATA_WIP: indicates that the new data is expected in migration ring.

hmm, that looks like a wrong optimization, likely we should have separate:

> + * @XE_GT_SRIOV_STATE_SAVE_WIP: indicates that VF save operation is in progress.

       XE_GT_SRIOV_STATE_SAVE_WAIT_DATA - indicates that the new SAVE data is expected

> + * @XE_GT_SRIOV_STATE_SAVE_FAILED: indicates that VF save operation has failed.
> + * @XE_GT_SRIOV_STATE_SAVED: indicates that VF is saved.
> + * @XE_GT_SRIOV_STATE_RESTORE_WIP: indicates that VF restore operation is in progress.

and
       XE_GT_SRIOV_STATE_RESTORE_WAIT_DATA - indicates that the new RESTORE data is expected 

> + * @XE_GT_SRIOV_STATE_SAVE_FAILED: indicates that VF restore operation has failed.
> + * @XE_GT_SRIOV_STATE_SAVED: indicates that VF is restored.

2x copy/paste typo

>   * @XE_GT_SRIOV_STATE_RESUME_WIP: indicates the a VF resume operation is in progress.
>   * @XE_GT_SRIOV_STATE_RESUME_SEND_RESUME: indicates that the PF is about to send RESUME command.
>   * @XE_GT_SRIOV_STATE_RESUME_FAILED: indicates that a VF resume operation has failed.
> @@ -63,6 +70,16 @@ enum xe_gt_sriov_control_bits {
>  	XE_GT_SRIOV_STATE_PAUSE_FAILED,
>  	XE_GT_SRIOV_STATE_PAUSED,
>  
> +	XE_GT_SRIOV_STATE_MIGRATION_DATA_WIP,
> +
> +	XE_GT_SRIOV_STATE_SAVE_WIP,
> +	XE_GT_SRIOV_STATE_SAVE_FAILED,
> +	XE_GT_SRIOV_STATE_SAVED,
> +
> +	XE_GT_SRIOV_STATE_RESTORE_WIP,
> +	XE_GT_SRIOV_STATE_RESTORE_FAILED,
> +	XE_GT_SRIOV_STATE_RESTORED,
> +
>  	XE_GT_SRIOV_STATE_RESUME_WIP,
>  	XE_GT_SRIOV_STATE_RESUME_SEND_RESUME,
>  	XE_GT_SRIOV_STATE_RESUME_FAILED,
> diff --git a/drivers/gpu/drm/xe/xe_sriov_pf_control.c b/drivers/gpu/drm/xe/xe_sriov_pf_control.c
> index 416d00a03fbb7..e64c7b56172c6 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_pf_control.c
> +++ b/drivers/gpu/drm/xe/xe_sriov_pf_control.c
> @@ -149,3 +149,99 @@ int xe_sriov_pf_control_sync_flr(struct xe_device *xe, unsigned int vfid)
>  
>  	return 0;
>  }
> +
> +/**
> + * xe_sriov_pf_control_save_vf - Save VF migration data on all GTs.
> + * @xe: the &xe_device
> + * @vfid: the VF identifier
> + *
> + * This function is for PF only.
> + *
> + * Return: 0 on success or a negative error code on failure.
> + */
> +int xe_sriov_pf_control_save_vf(struct xe_device *xe, unsigned int vfid)
> +{
> +	struct xe_gt *gt;
> +	unsigned int id;
> +	int ret;
> +
> +	for_each_gt(gt, xe, id) {
> +		ret = xe_gt_sriov_pf_control_save_vf(gt, vfid);
> +		if (ret)
> +			return ret;
> +	}
> +
> +	return 0;
> +}
> +
> +/**
> + * xe_sriov_pf_control_wait_save_vf - Wait until VF migration data was saved on all GTs
> + * @xe: the &xe_device
> + * @vfid: the VF identifier
> + *
> + * This function is for PF only.
> + *
> + * Return: 0 on success or a negative error code on failure.
> + */
> +int xe_sriov_pf_control_wait_save_vf(struct xe_device *xe, unsigned int vfid)
> +{
> +	struct xe_gt *gt;
> +	u8 id;

"unsigned int" like above?

> +	int ret;
> +
> +	for_each_gt(gt, xe, id) {
> +		ret = xe_gt_sriov_pf_control_wait_save_done(gt, vfid);
> +		if (ret)
> +			break;
> +	}
> +
> +	return ret;
> +}
> +
> +/**
> + * xe_sriov_pf_control_restore_vf - Restore VF migration data on all GTs.
> + * @xe: the &xe_device
> + * @vfid: the VF identifier
> + *
> + * This function is for PF only.
> + *
> + * Return: 0 on success or a negative error code on failure.
> + */
> +int xe_sriov_pf_control_restore_vf(struct xe_device *xe, unsigned int vfid)
> +{
> +	struct xe_gt *gt;
> +	u8 id;
> +	int ret;
> +
> +	for_each_gt(gt, xe, id) {
> +		ret = xe_gt_sriov_pf_control_restore_vf(gt, vfid);
> +		if (ret)
> +			return ret;
> +	}
> +
> +	return ret;
> +}
> +
> +/**
> + * xe_sriov_pf_control_wait_save_vf - Wait until VF migration data was restored on all GTs
> + * @xe: the &xe_device
> + * @vfid: the VF identifier
> + *
> + * This function is for PF only.
> + *
> + * Return: 0 on success or a negative error code on failure.
> + */
> +int xe_sriov_pf_control_wait_restore_vf(struct xe_device *xe, unsigned int vfid)
> +{
> +	struct xe_gt *gt;
> +	u8 id;
> +	int ret;
> +
> +	for_each_gt(gt, xe, id) {
> +		ret = xe_gt_sriov_pf_control_wait_restore_done(gt, vfid);
> +		if (ret)
> +			break;
> +	}
> +
> +	return ret;
> +}
> diff --git a/drivers/gpu/drm/xe/xe_sriov_pf_control.h b/drivers/gpu/drm/xe/xe_sriov_pf_control.h
> index 2d52d0ac1b28f..512fd21d87c1e 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_pf_control.h
> +++ b/drivers/gpu/drm/xe/xe_sriov_pf_control.h
> @@ -13,5 +13,9 @@ int xe_sriov_pf_control_resume_vf(struct xe_device *xe, unsigned int vfid);
>  int xe_sriov_pf_control_stop_vf(struct xe_device *xe, unsigned int vfid);
>  int xe_sriov_pf_control_reset_vf(struct xe_device *xe, unsigned int vfid);
>  int xe_sriov_pf_control_sync_flr(struct xe_device *xe, unsigned int vfid);
> +int xe_sriov_pf_control_save_vf(struct xe_device *xe, unsigned int vfid);
> +int xe_sriov_pf_control_wait_save_vf(struct xe_device *xe, unsigned int vfid);
> +int xe_sriov_pf_control_restore_vf(struct xe_device *xe, unsigned int vfid);
> +int xe_sriov_pf_control_wait_restore_vf(struct xe_device *xe, unsigned int vfid);
>  
>  #endif
> diff --git a/drivers/gpu/drm/xe/xe_sriov_pf_debugfs.c b/drivers/gpu/drm/xe/xe_sriov_pf_debugfs.c
> index 97636ed86fb8b..74eeabef91c57 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_pf_debugfs.c
> +++ b/drivers/gpu/drm/xe/xe_sriov_pf_debugfs.c
> @@ -75,11 +75,31 @@ static void pf_populate_pf(struct xe_device *xe, struct dentry *pfdent)
>   *      │   │   ├── reset
>   *      │   │   ├── resume
>   *      │   │   ├── stop
> + *      │   │   ├── save
> + *      │   │   ├── restore
>   *      │   │   :
>   *      │   ├── vf2
>   *      │   │   ├── ...
>   */
>  
> +static int from_file_read_to_vf_call(struct seq_file *s,
> +				     int (*call)(struct xe_device *, unsigned int))
> +{
> +	struct dentry *dent = file_dentry(s->file)->d_parent;
> +	struct xe_device *xe = extract_xe(dent);
> +	unsigned int vfid = extract_vfid(dent);
> +	int ret;
> +
> +	xe_pm_runtime_get(xe);
> +	ret = call(xe, vfid);
> +	xe_pm_runtime_put(xe);
> +
> +	if (ret < 0)
> +		return ret;
> +
> +	return s->count;

since we don't expect to put anything into "s", maybe we can explicitly return 0 here?

> +}
> +
>  static ssize_t from_file_write_to_vf_call(struct file *file, const char __user *userbuf,
>  					  size_t count, loff_t *ppos,
>  					  int (*call)(struct xe_device *, unsigned int))
> @@ -118,10 +138,26 @@ static ssize_t OP##_write(struct file *file, const char __user *userbuf,	\
>  }										\
>  DEFINE_SHOW_STORE_ATTRIBUTE(OP)
>  
> +#define DEFINE_VF_RW_CONTROL_ATTRIBUTE(OP)					\

rename it to have RW as suffix (like other ATTR macros):

	DEFINE_VF_CONTROL_ATTRIBUTE_RW

> +static int OP##_show(struct seq_file *s, void *unused)				\
> +{										\
> +	return from_file_read_to_vf_call(s,					\
> +					 xe_sriov_pf_control_wait_##OP);	\
> +}										\
> +static ssize_t OP##_write(struct file *file, const char __user *userbuf,	\
> +			  size_t count, loff_t *ppos)				\
> +{										\
> +	return from_file_write_to_vf_call(file, userbuf, count, ppos,		\
> +					  xe_sriov_pf_control_##OP);		\
> +}										\
> +DEFINE_SHOW_STORE_ATTRIBUTE(OP)
> +
>  DEFINE_VF_CONTROL_ATTRIBUTE(pause_vf);
>  DEFINE_VF_CONTROL_ATTRIBUTE(resume_vf);
>  DEFINE_VF_CONTROL_ATTRIBUTE(stop_vf);
>  DEFINE_VF_CONTROL_ATTRIBUTE(reset_vf);
> +DEFINE_VF_RW_CONTROL_ATTRIBUTE(save_vf);
> +DEFINE_VF_RW_CONTROL_ATTRIBUTE(restore_vf);
>  
>  static void pf_populate_vf(struct xe_device *xe, struct dentry *vfdent)
>  {
> @@ -129,6 +165,8 @@ static void pf_populate_vf(struct xe_device *xe, struct dentry *vfdent)
>  	debugfs_create_file("resume", 0200, vfdent, xe, &resume_vf_fops);
>  	debugfs_create_file("stop", 0200, vfdent, xe, &stop_vf_fops);
>  	debugfs_create_file("reset", 0200, vfdent, xe, &reset_vf_fops);
> +	debugfs_create_file("save", 0600, vfdent, xe, &save_vf_fops);
> +	debugfs_create_file("restore", 0600, vfdent, xe, &restore_vf_fops);
>  }
>  
>  static void pf_populate_with_tiles(struct xe_device *xe, struct dentry *dent, unsigned int vfid)