Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
To: "Piotr Piórkowski" <piotr.piorkowski@intel.com>
Cc: <intel-xe@lists.freedesktop.org>
Subject: Re: [PATCH 1/4] drm/xe/pf: Promote VFs provisioning helpers
Date: Thu, 16 Oct 2025 13:33:24 +0200	[thread overview]
Message-ID: <d264b175-76ea-4336-92a0-e377dd7e5b0e@intel.com> (raw)
In-Reply-To: <20251015130033.fsiaela2k64kicm6@intel.com>



On 10/15/2025 3:00 PM, Piotr Piórkowski wrote:
> Michal Wajdeczko <michal.wajdeczko@intel.com> wrote on śro [2025-paź-15 11:12:06 +0200]:
>> As we plan to add more VFs provisioning methods, start moving
>> related code into single place.
>>
>> Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
>> ---
>>  drivers/gpu/drm/xe/Makefile                |  1 +
>>  drivers/gpu/drm/xe/xe_pci_sriov.c          | 45 ++----------
>>  drivers/gpu/drm/xe/xe_sriov_pf_provision.c | 83 ++++++++++++++++++++++
>>  drivers/gpu/drm/xe/xe_sriov_pf_provision.h | 14 ++++
>>  4 files changed, 102 insertions(+), 41 deletions(-)
>>  create mode 100644 drivers/gpu/drm/xe/xe_sriov_pf_provision.c
>>  create mode 100644 drivers/gpu/drm/xe/xe_sriov_pf_provision.h
>>
>> diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
>> index 84321fad3265..edae0c2a6e2f 100644
>> --- a/drivers/gpu/drm/xe/Makefile
>> +++ b/drivers/gpu/drm/xe/Makefile
>> @@ -176,6 +176,7 @@ xe-$(CONFIG_PCI_IOV) += \
>>  	xe_sriov_pf.o \
>>  	xe_sriov_pf_control.o \
>>  	xe_sriov_pf_debugfs.o \
>> +	xe_sriov_pf_provision.o \
>>  	xe_sriov_pf_service.o \
>>  	xe_tile_sriov_pf_debugfs.o
>>  
>> diff --git a/drivers/gpu/drm/xe/xe_pci_sriov.c b/drivers/gpu/drm/xe/xe_pci_sriov.c
>> index 9c1c9e669b04..735f51effc7a 100644
>> --- a/drivers/gpu/drm/xe/xe_pci_sriov.c
>> +++ b/drivers/gpu/drm/xe/xe_pci_sriov.c
>> @@ -19,46 +19,9 @@
>>  #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_printk.h"
>>  
>> -static int pf_needs_provisioning(struct xe_gt *gt, unsigned int num_vfs)
>> -{
>> -	unsigned int n;
>> -
>> -	for (n = 1; n <= num_vfs; n++)
>> -		if (!xe_gt_sriov_pf_config_is_empty(gt, n))
>> -			return false;
>> -
>> -	return true;
>> -}
>> -
>> -static int pf_provision_vfs(struct xe_device *xe, unsigned int num_vfs)
>> -{
>> -	struct xe_gt *gt;
>> -	unsigned int id;
>> -	int result = 0, err;
>> -
>> -	for_each_gt(gt, xe, id) {
>> -		if (!pf_needs_provisioning(gt, num_vfs))
>> -			continue;
>> -		err = xe_gt_sriov_pf_config_set_fair(gt, VFID(1), num_vfs);
>> -		result = result ?: err;
>> -	}
>> -
>> -	return result;
>> -}
>> -
>> -static void pf_unprovision_vfs(struct xe_device *xe, unsigned int num_vfs)
>> -{
>> -	struct xe_gt *gt;
>> -	unsigned int id;
>> -	unsigned int n;
>> -
>> -	for_each_gt(gt, xe, id)
>> -		for (n = 1; n <= num_vfs; n++)
>> -			xe_gt_sriov_pf_config_release(gt, n, true);
>> -}
>> -
>>  static void pf_reset_vfs(struct xe_device *xe, unsigned int num_vfs)
>>  {
>>  	unsigned int n;
>> @@ -168,7 +131,7 @@ static int pf_enable_vfs(struct xe_device *xe, int num_vfs)
>>  	 */
>>  	xe_pm_runtime_get_noresume(xe);
>>  
>> -	err = pf_provision_vfs(xe, num_vfs);
>> +	err = xe_sriov_pf_provision_vfs(xe, num_vfs);
>>  	if (err < 0)
>>  		goto failed;
>>  
>> @@ -192,7 +155,7 @@ static int pf_enable_vfs(struct xe_device *xe, int num_vfs)
>>  	return num_vfs;
>>  
>>  failed:
>> -	pf_unprovision_vfs(xe, num_vfs);
>> +	xe_sriov_pf_unprovision_vfs(xe, num_vfs);
>>  	xe_pm_runtime_put(xe);
>>  out:
>>  	xe_sriov_notice(xe, "Failed to enable %u VF%s (%pe)\n",
>> @@ -218,7 +181,7 @@ static int pf_disable_vfs(struct xe_device *xe)
>>  
>>  	pf_reset_vfs(xe, num_vfs);
>>  
>> -	pf_unprovision_vfs(xe, num_vfs);
>> +	xe_sriov_pf_unprovision_vfs(xe, num_vfs);
>>  
>>  	/* not needed anymore - see pf_enable_vfs() */
>>  	xe_pm_runtime_put(xe);
>> diff --git a/drivers/gpu/drm/xe/xe_sriov_pf_provision.c b/drivers/gpu/drm/xe/xe_sriov_pf_provision.c
>> new file mode 100644
>> index 000000000000..c24803312231
>> --- /dev/null
>> +++ b/drivers/gpu/drm/xe/xe_sriov_pf_provision.c
>> @@ -0,0 +1,83 @@
>> +// SPDX-License-Identifier: MIT
>> +/*
>> + * Copyright © 2025 Intel Corporation
>> + */
>> +
>> +#include "xe_assert.h"
>> +#include "xe_device.h"
>> +#include "xe_gt_sriov_pf_config.h"
>> +#include "xe_sriov.h"
>> +#include "xe_sriov_pf_helpers.h"
>> +#include "xe_sriov_pf_provision.h"
>> +
>> +static int pf_needs_provisioning(struct xe_gt *gt, unsigned int num_vfs)
> 
> NIT: Why don't you use bool here?

this is pure copy-paste, it was "int" already there

but I can change it why applying

> 
> 
>> +{
>> +	unsigned int n;
>> +
>> +	for (n = 1; n <= num_vfs; n++)
>> +		if (!xe_gt_sriov_pf_config_is_empty(gt, n))
>> +			return false;
>> +
>> +	return true;
>> +}
>> +
>> +static int pf_provision_vfs(struct xe_device *xe, unsigned int num_vfs)
>> +{
>> +	struct xe_gt *gt;
>> +	unsigned int id;
>> +	int result = 0;
>> +	int err;
>> +
>> +	for_each_gt(gt, xe, id) {
>> +		if (!pf_needs_provisioning(gt, num_vfs))
>> +			continue;
>> +		err = xe_gt_sriov_pf_config_set_fair(gt, VFID(1), num_vfs);
>> +		result = result ?: err;
>> +	}
>> +
>> +	return result;
>> +}
>> +
>> +static void pf_unprovision_vfs(struct xe_device *xe, unsigned int num_vfs)
>> +{
>> +	struct xe_gt *gt;
>> +	unsigned int id;
>> +	unsigned int n;
>> +
>> +	for_each_gt(gt, xe, id)
>> +		for (n = 1; n <= num_vfs; n++)
>> +			xe_gt_sriov_pf_config_release(gt, n, true);
>> +}
>> +
>> +/**
>> + * xe_sriov_pf_provision_vfs() - Provision VFs in auto-mode.
>> + * @xe: the PF &xe_device
>> + * @num_vfs: the number of VFs to auto-provision
>> + *
>> + * This function can only be called on PF.
>> + *
>> + * Return: 0 on success or a negative error code on failure.
>> + */
>> +int xe_sriov_pf_provision_vfs(struct xe_device *xe, unsigned int num_vfs)
>> +{
>> +	xe_assert(xe, IS_SRIOV_PF(xe));
>> +
>> +	return pf_provision_vfs(xe, num_vfs);
>> +}
>> +
>> +/**
>> + * xe_sriov_pf_unprovision_vfs() - Unprovision VFs in auto-mode.
>> + * @xe: the PF &xe_device
>> + * @num_vfs: the number of VFs to unprovision
>> + *
>> + * This function can only be called on PF.
>> + *
>> + * Return: 0 on success or a negative error code on failure.
>> + */
>> +int xe_sriov_pf_unprovision_vfs(struct xe_device *xe, unsigned int num_vfs)
>> +{
>> +	xe_assert(xe, IS_SRIOV_PF(xe));
>> +
>> +	pf_unprovision_vfs(xe, num_vfs);
>> +	return 0;
>> +}
> 
> NIT: Perhaps it would be worth adding the assertion num_vfs > 0 here.

it's harmless here and in pf_unprovision_vfs() where the loop is from 1

	for (n = 1; n <= num_vfs; n++)

> 
>> diff --git a/drivers/gpu/drm/xe/xe_sriov_pf_provision.h b/drivers/gpu/drm/xe/xe_sriov_pf_provision.h
>> new file mode 100644
>> index 000000000000..f6a902190ad7
>> --- /dev/null
>> +++ b/drivers/gpu/drm/xe/xe_sriov_pf_provision.h
>> @@ -0,0 +1,14 @@
>> +/* SPDX-License-Identifier: MIT */
>> +/*
>> + * Copyright © 2025 Intel Corporation
>> + */
>> +
>> +#ifndef _XE_SRIOV_PF_PROVISION_H_
>> +#define _XE_SRIOV_PF_PROVISION_H_
>> +
>> +struct xe_device;
>> +
>> +int xe_sriov_pf_provision_vfs(struct xe_device *xe, unsigned int num_vfs);
>> +int xe_sriov_pf_unprovision_vfs(struct xe_device *xe, unsigned int num_vfs);
>> +
>> +#endif
> 
> Minor comments, but the code looks okay:
> Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
> 
>> -- 
>> 2.47.1
>>
> 


  reply	other threads:[~2025-10-16 11:33 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-15  9:12 [PATCH 0/4] PF: Update auto-provisioning Michal Wajdeczko
2025-10-15  9:12 ` [PATCH 1/4] drm/xe/pf: Promote VFs provisioning helpers Michal Wajdeczko
2025-10-15 13:00   ` Piotr Piórkowski
2025-10-16 11:33     ` Michal Wajdeczko [this message]
2025-10-15  9:12 ` [PATCH 2/4] drm/xe/pf: Automatically provision VFs only in auto-mode Michal Wajdeczko
2025-10-15 16:30   ` Piotr Piórkowski
2025-10-15  9:12 ` [PATCH 3/4] drm/xe/pf: Disable auto-provisioning if changed using debugfs Michal Wajdeczko
2025-10-16 15:26   ` Piotr Piórkowski
2025-10-15  9:12 ` [PATCH 4/4] drm/xe/pf: Allow to restore auto-provisioning mode Michal Wajdeczko
2025-10-16 15:46   ` Piotr Piórkowski
2025-10-15 12:10 ` ✗ CI.checkpatch: warning for PF: Update auto-provisioning Patchwork
2025-10-15 12:11 ` ✓ CI.KUnit: success " Patchwork
2025-10-15 13:07 ` ✓ Xe.CI.BAT: " Patchwork
2025-10-15 22:12 ` ✗ Xe.CI.Full: failure " Patchwork
2025-10-16 10:55   ` Michal Wajdeczko
2025-10-20  9:49     ` Bernatowicz, Marcin
2025-10-20 11:08       ` Bernatowicz, Marcin

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=d264b175-76ea-4336-92a0-e377dd7e5b0e@intel.com \
    --to=michal.wajdeczko@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=piotr.piorkowski@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