All of lore.kernel.org
 help / color / mirror / Atom feed
From: Lang Yu <Lang.Yu@amd.com>
To: Felix Kuehling <felix.kuehling@amd.com>,
	Jonathan Kim <Jonathan.Kim@amd.com>,
	David Yat Sin <David.YatSin@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>,
	Huang Rui <ray.huang@amd.com>,
	amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/amdkfd: add family_id property for kfd_node
Date: Thu, 18 Aug 2022 09:53:34 +0800	[thread overview]
Message-ID: <Yv2bntl7o8B59C6k@lang-desktop> (raw)
In-Reply-To: <f1bbde75-c273-9a90-f966-bb670e3b24d6@amd.com>

On 08/17/ , Felix Kuehling wrote:
> Am 2022-08-17 um 11:04 schrieb Felix Kuehling:
> > Am 2022-08-16 um 23:09 schrieb Lang Yu:
> > > Then we can remove the burden of maintaining codes to
> > > parse family_id from gfx version in rocr,
> > > i.e., remove DevIDToAddrLibFamily().
> > 
> > I'm OK with the change. But you won't be able to remove
> > DevIDToAddrLibFamily as long as ROCr needs to support older kernels that
> > don't provide this info.
> As an alternative you should be able to use the AMDGPU_INFO_DEV_INFO ioctl
> from the render node. The Thunk already has the render nodes of all the
> GPUs. This will work on older kernels as well and will allow you to remove
> DevIDToAddrLibFamily.

Yes, topology_get_marketing_name() has called amdgpu_device_initialize()
and got family id info in Thunk. I will put the family id info into
HsaNodeProperties. Thanks for your suggestion.

Regards,
Lang

> Regards,
>   Felix
> 
> 
> > 
> > Regards,
> >   Felix
> > 
> > 
> > > 
> > > Signed-off-by: Lang Yu <Lang.Yu@amd.com>
> > > ---
> > >   drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 3 +++
> > >   drivers/gpu/drm/amd/amdkfd/kfd_topology.h | 1 +
> > >   2 files changed, 4 insertions(+)
> > > 
> > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
> > > b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
> > > index 3f0a4a415907..7e0331e853d5 100644
> > > --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
> > > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
> > > @@ -507,6 +507,8 @@ static ssize_t node_show(struct kobject *kobj,
> > > struct attribute *attr,
> > >                     dev->node_props.vendor_id);
> > >       sysfs_show_32bit_prop(buffer, offs, "device_id",
> > >                     dev->node_props.device_id);
> > > +    sysfs_show_32bit_prop(buffer, offs, "family_id",
> > > +                  dev->node_props.family_id);
> > >       sysfs_show_32bit_prop(buffer, offs, "location_id",
> > >                     dev->node_props.location_id);
> > >       sysfs_show_32bit_prop(buffer, offs, "domain",
> > > @@ -1690,6 +1692,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
> > >       dev->node_props.gfx_target_version =
> > > gpu->device_info.gfx_target_version;
> > >       dev->node_props.vendor_id = gpu->pdev->vendor;
> > >       dev->node_props.device_id = gpu->pdev->device;
> > > +    dev->node_props.family_id = gpu->adev->family;
> > >       dev->node_props.capability |=
> > >           ((dev->gpu->adev->rev_id << HSA_CAP_ASIC_REVISION_SHIFT) &
> > >               HSA_CAP_ASIC_REVISION_MASK);
> > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
> > > b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
> > > index 9f6c949186c1..4ff8dd2c9549 100644
> > > --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
> > > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
> > > @@ -56,6 +56,7 @@ struct kfd_node_properties {
> > >       uint32_t gfx_target_version;
> > >       uint32_t vendor_id;
> > >       uint32_t device_id;
> > > +    uint32_t family_id;
> > >       uint32_t location_id;
> > >       uint32_t domain;
> > >       uint32_t max_engine_clk_fcompute;

      reply	other threads:[~2022-08-18  1:54 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-17  3:09 [PATCH] drm/amdkfd: add family_id property for kfd_node Lang Yu
2022-08-17 15:04 ` Felix Kuehling
2022-08-17 16:25   ` Felix Kuehling
2022-08-18  1:53     ` Lang Yu [this message]

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=Yv2bntl7o8B59C6k@lang-desktop \
    --to=lang.yu@amd.com \
    --cc=David.YatSin@amd.com \
    --cc=Jonathan.Kim@amd.com \
    --cc=alexander.deucher@amd.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=felix.kuehling@amd.com \
    --cc=ray.huang@amd.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.