All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Kuehling, Felix" <Felix.Kuehling-5C7GfCeVMHo@public.gmane.org>
To: "Koenig,
	Christian" <Christian.Koenig-5C7GfCeVMHo@public.gmane.org>,
	"amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org"
	<amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org>
Cc: "Zeng, Oak" <Oak.Zeng-5C7GfCeVMHo@public.gmane.org>
Subject: Re: [PATCH 3/4] drm/amdgpu: Support snooped PTE flag
Date: Mon, 26 Aug 2019 22:32:25 +0000	[thread overview]
Message-ID: <bd1f6713-ae1a-37fd-3996-3bdc9c7223d8@amd.com> (raw)
In-Reply-To: <49ba47da-a225-5a21-9014-ccc316c55b60-5C7GfCeVMHo@public.gmane.org>

On 2019-08-26 2:12 p.m., Koenig, Christian wrote:
>
> We should probably just revert back the to doing most of the mapping in
> amdgpu_vm_bo_split_mapping().
>
> Here we already have a whole bunch of ASIC specific handling anyway:
>
>>          if (!(mapping->flags & AMDGPU_PTE_READABLE))
>>                  flags &= ~AMDGPU_PTE_READABLE;
>>          if (!(mapping->flags & AMDGPU_PTE_WRITEABLE))
>>                  flags &= ~AMDGPU_PTE_WRITEABLE;
>>
>>          flags &= ~AMDGPU_PTE_EXECUTABLE;
>>          flags |= mapping->flags & AMDGPU_PTE_EXECUTABLE;
>>
>>          if (adev->asic_type == CHIP_NAVI10) {
>>                  flags &= ~AMDGPU_PTE_MTYPE_NV10_MASK;
>>                  flags |= (mapping->flags & AMDGPU_PTE_MTYPE_NV10_MASK);
>>          } else {
>>                  flags &= ~AMDGPU_PTE_MTYPE_VG10_MASK;
>>                  flags |= (mapping->flags & AMDGPU_PTE_MTYPE_VG10_MASK);
>>          }
>>
>>          if ((mapping->flags & AMDGPU_PTE_PRT) &&
>>              (adev->asic_type >= CHIP_VEGA10)) {
>>                  flags |= AMDGPU_PTE_PRT;
>>                  if (adev->asic_type >= CHIP_NAVI10) {
>>                          flags |= AMDGPU_PTE_SNOOPED;
>>                          flags |= AMDGPU_PTE_LOG;
>>                          flags |= AMDGPU_PTE_SYSTEM;
>>                  }
>>                  flags &= ~AMDGPU_PTE_VALID;
>>          }
> And now that you mentioned it, the code you proposed wouldn't have
> worked anyway because the AMDGPU_PTE_SNOOPED would have been filtered
> out here :)

OK, and that does look like the right place to set the PTE_SNOOPED bit 
for XGMI mappings on Arcturus. I'll send out an updated patch series 
that no longer needs the AMDGPU_VM_PAGE_INVALIDATE_PROBE bit in the UAPI.

Thanks for the pointer,
   Felix


>
> Regards,
> Christian.
>
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

  parent reply	other threads:[~2019-08-26 22:32 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-23 21:33 [PATCH 0/4] KFD: Mapping-specific MTYPEs on Arcturus Kuehling, Felix
     [not found] ` <20190823213249.10749-1-Felix.Kuehling-5C7GfCeVMHo@public.gmane.org>
2019-08-23 21:33   ` [PATCH 2/4] drm/amdgpu: Support new arcturus mtype Kuehling, Felix
2019-08-23 21:33   ` [PATCH 1/4] drm/amdgpu: Extends amdgpu vm definitions Kuehling, Felix
     [not found]     ` <20190823213249.10749-2-Felix.Kuehling-5C7GfCeVMHo@public.gmane.org>
2019-08-24 11:12       ` Christian König
2019-08-23 21:33   ` [PATCH 3/4] drm/amdgpu: Support snooped PTE flag Kuehling, Felix
     [not found]     ` <20190823213249.10749-4-Felix.Kuehling-5C7GfCeVMHo@public.gmane.org>
2019-08-24 11:13       ` Christian König
     [not found]         ` <f09e6893-347b-4ade-76e5-ad37d8e4e782-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2019-08-24 16:24           ` Kuehling, Felix
     [not found]             ` <a7f9ad48-2e46-5415-e2a8-1738a101d716-5C7GfCeVMHo@public.gmane.org>
2019-08-24 18:59               ` Christian König
     [not found]                 ` <96b6ac1d-de87-3fdb-a531-af4b0a42f1d5-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2019-08-26 15:45                   ` Kuehling, Felix
     [not found]                     ` <4b48fc1a-6ee2-bb60-0518-ea9c6346b8d6-5C7GfCeVMHo@public.gmane.org>
2019-08-26 17:16                       ` Koenig, Christian
     [not found]                         ` <89bf5baa-4b63-e40a-7995-fa35bad988b7-5C7GfCeVMHo@public.gmane.org>
2019-08-26 18:03                           ` Kuehling, Felix
     [not found]                             ` <216f63db-78c5-1098-bea5-2f379b0bf051-5C7GfCeVMHo@public.gmane.org>
2019-08-26 18:12                               ` Koenig, Christian
     [not found]                                 ` <49ba47da-a225-5a21-9014-ccc316c55b60-5C7GfCeVMHo@public.gmane.org>
2019-08-26 22:32                                   ` Kuehling, Felix [this message]
2019-08-23 21:33   ` [PATCH 4/4] drm/amdgpu: Determing PTE flags separately for each mapping Kuehling, Felix

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=bd1f6713-ae1a-37fd-3996-3bdc9c7223d8@amd.com \
    --to=felix.kuehling-5c7gfcevmho@public.gmane.org \
    --cc=Christian.Koenig-5C7GfCeVMHo@public.gmane.org \
    --cc=Oak.Zeng-5C7GfCeVMHo@public.gmane.org \
    --cc=amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org \
    /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.