[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <iiyzek47iwn4av7bohszmhbgwkjpk3ytqkuwq7sk2qpoqdzrvg@msh5rt5bp3iy>
Date: Tue, 5 Mar 2024 15:08:04 -0300
From: "Ricardo B. Marliere" <ricardo@...liere.net>
To: Felix Kuehling <felix.kuehling@....com>
Cc: Alex Deucher <alexander.deucher@....com>,
Christian König <christian.koenig@....com>, David Airlie <airlied@...il.com>,
Daniel Vetter <daniel@...ll.ch>, amd-gfx@...ts.freedesktop.org, dri-devel@...ts.freedesktop.org,
linux-kernel@...r.kernel.org, Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Subject: Re: [PATCH] drm/amdkfd: make kfd_class constant
On 5 Mar 12:38, Felix Kuehling wrote:
> On 2024-03-05 7:15, Ricardo B. Marliere wrote:
> > Since commit 43a7206b0963 ("driver core: class: make class_register() take
> > a const *"), the driver core allows for struct class to be in read-only
> > memory, so move the kfd_class structure to be declared at build time
> > placing it into read-only memory, instead of having to be dynamically
> > allocated at boot time.
> >
> > Cc: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
> > Suggested-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
> > Signed-off-by: Ricardo B. Marliere <ricardo@...liere.net>
>
> The patch looks good to me. Do you want me to apply this to Alex's
> amd-staging-drm-next?
>
> Reviewed-by: Felix Kuehling <felix.kuehling@....com>
Hi Felix!
Thanks for reviewing. The patch was based in agd5f/linux/drm-next. But I
think amd-staging-drm-next would be fine, too.
Best regards,
- Ricardo.
>
>
> > ---
> > drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 21 +++++++++++----------
> > 1 file changed, 11 insertions(+), 10 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> > index f030cafc5a0a..dfa8c69532d4 100644
> > --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> > @@ -63,8 +63,10 @@ static const struct file_operations kfd_fops = {
> > };
> >
> > static int kfd_char_dev_major = -1;
> > -static struct class *kfd_class;
> > struct device *kfd_device;
> > +static const struct class kfd_class = {
> > + .name = kfd_dev_name,
> > +};
> >
> > static inline struct kfd_process_device *kfd_lock_pdd_by_id(struct kfd_process *p, __u32 gpu_id)
> > {
> > @@ -94,14 +96,13 @@ int kfd_chardev_init(void)
> > if (err < 0)
> > goto err_register_chrdev;
> >
> > - kfd_class = class_create(kfd_dev_name);
> > - err = PTR_ERR(kfd_class);
> > - if (IS_ERR(kfd_class))
> > + err = class_register(&kfd_class);
> > + if (err)
> > goto err_class_create;
> >
> > - kfd_device = device_create(kfd_class, NULL,
> > - MKDEV(kfd_char_dev_major, 0),
> > - NULL, kfd_dev_name);
> > + kfd_device = device_create(&kfd_class, NULL,
> > + MKDEV(kfd_char_dev_major, 0),
> > + NULL, kfd_dev_name);
> > err = PTR_ERR(kfd_device);
> > if (IS_ERR(kfd_device))
> > goto err_device_create;
> > @@ -109,7 +110,7 @@ int kfd_chardev_init(void)
> > return 0;
> >
> > err_device_create:
> > - class_destroy(kfd_class);
> > + class_unregister(&kfd_class);
> > err_class_create:
> > unregister_chrdev(kfd_char_dev_major, kfd_dev_name);
> > err_register_chrdev:
> > @@ -118,8 +119,8 @@ int kfd_chardev_init(void)
> >
> > void kfd_chardev_exit(void)
> > {
> > - device_destroy(kfd_class, MKDEV(kfd_char_dev_major, 0));
> > - class_destroy(kfd_class);
> > + device_destroy(&kfd_class, MKDEV(kfd_char_dev_major, 0));
> > + class_unregister(&kfd_class);
> > unregister_chrdev(kfd_char_dev_major, kfd_dev_name);
> > kfd_device = NULL;
> > }
> >
> > ---
> > base-commit: 8bc75586ea01f1c645063d3472c115ecab03e76c
> > change-id: 20240305-class_cleanup-drm-amd-bdc7255b7540
> >
> > Best regards,
Powered by blists - more mailing lists