Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH hmm v2 5/5] mm/hmm: remove the customizable pfn format
 from hmm_range_fault
To:     Jason Gunthorpe <jgg@ziepe.ca>, <linux-mm@kvack.org>
CC:     Alex Deucher <alexander.deucher@amd.com>,
        <amd-gfx@lists.freedesktop.org>, Ben Skeggs <bskeggs@redhat.com>,
        =?UTF-8?Q?Christian_K=c3=b6nig?= <christian.koenig@amd.com>,
        "David (ChunMing) Zhou" <David1.Zhou@amd.com>,
        <dri-devel@lists.freedesktop.org>,
        Felix Kuehling <Felix.Kuehling@amd.com>,
        Christoph Hellwig <hch@lst.de>,
        <intel-gfx@lists.freedesktop.org>,
        =?UTF-8?B?SsOpcsO0bWUgR2xpc3Nl?= <jglisse@redhat.com>,
        John Hubbard <jhubbard@nvidia.com>,
        <linux-kernel@vger.kernel.org>,
        "Niranjana Vishwanathapura" <niranjana.vishwanathapura@intel.com>,
        <nouveau@lists.freedesktop.org>,
        "Yang, Philip" <Philip.Yang@amd.com>
References: <5-v2-b4e84f444c7d+24f57-hmm_no_flags_jgg@mellanox.com>
From:   Ralph Campbell <rcampbell@nvidia.com>
Message-ID: <3c06a94c-c17f-dc31-537e-f3f6e1ace9a2@nvidia.com>
Date:   Fri, 1 May 2020 17:53:26 -0700
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.2.2
MIME-Version: 1.0
In-Reply-To: <5-v2-b4e84f444c7d+24f57-hmm_no_flags_jgg@mellanox.com>
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On 5/1/20 11:20 AM, Jason Gunthorpe wrote:
> From: Jason Gunthorpe <jgg@mellanox.com>
> 
> Presumably the intent here was that hmm_range_fault() could put the data
> into some HW specific format and thus avoid some work. However, nothing
> actually does that, and it isn't clear how anything actually could do that
> as hmm_range_fault() provides CPU addresses which must be DMA mapped.
> 
> Perhaps there is some special HW that does not need DMA mapping, but we
> don't have any examples of this, and the theoretical performance win of
> avoiding an extra scan over the pfns array doesn't seem worth the
> complexity. Plus pfns needs to be scanned anyhow to sort out any
> DEVICE_PRIVATE pages.
> 
> This version replaces the uint64_t with an usigned long containing a pfn
> and fixed flags. On input flags is filled with the HMM_PFN_REQ_* values,
> on successful output it is filled with HMM_PFN_* values, describing the
> state of the pages.
> 
> amdgpu is simple to convert, it doesn't use snapshot and doesn't use
> per-page flags.
> 
> nouveau uses only 16 hmm_pte entries at most (ie fits in a few cache
> lines), and it sweeps over its pfns array a couple of times anyhow. It
> also has a nasty call chain before it reaches the dma map and hardware
> suggesting performance isn't important:
> 
>     nouveau_svm_fault():
>       args.i.m.method = NVIF_VMM_V0_PFNMAP
>       nouveau_range_fault()
>        nvif_object_ioctl()
>         client->driver->ioctl()
> 	  struct nvif_driver nvif_driver_nvkm:
> 	    .ioctl = nvkm_client_ioctl
> 	   nvkm_ioctl()
> 	    nvkm_ioctl_path()
> 	      nvkm_ioctl_v0[type].func(..)
> 	      nvkm_ioctl_mthd()
> 	       nvkm_object_mthd()
> 		  struct nvkm_object_func nvkm_uvmm:
> 		    .mthd = nvkm_uvmm_mthd
> 		   nvkm_uvmm_mthd()
> 		    nvkm_uvmm_mthd_pfnmap()
> 		     nvkm_vmm_pfn_map()
> 		      nvkm_vmm_ptes_get_map()
> 		       func == gp100_vmm_pgt_pfn
> 			struct nvkm_vmm_desc_func gp100_vmm_desc_spt:
> 			  .pfn = gp100_vmm_pgt_pfn
> 			 nvkm_vmm_iter()
> 			  REF_PTES == func == gp100_vmm_pgt_pfn()
> 			    dma_map_page()
> 
> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
> Tested-by: Ralph Campbell <rcampbell@nvidia.com>
> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
>   Documentation/vm/hmm.rst                |  26 ++--
>   drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c |  35 ++----
>   drivers/gpu/drm/nouveau/nouveau_dmem.c  |  27 +---
>   drivers/gpu/drm/nouveau/nouveau_dmem.h  |   3 +-
>   drivers/gpu/drm/nouveau/nouveau_svm.c   |  87 ++++++++-----
>   include/linux/hmm.h                     |  99 ++++++---------
>   mm/hmm.c                                | 160 +++++++++++-------------
>   7 files changed, 192 insertions(+), 245 deletions(-)
> 

...snip...

>   
> +static void nouveau_hmm_convert_pfn(struct nouveau_drm *drm,
> +				    struct hmm_range *range, u64 *ioctl_addr)
> +{
> +	unsigned long i, npages;
> +
> +	/*
> +	 * The ioctl_addr prepared here is passed through nvif_object_ioctl()
> +	 * to an eventual DMA map in something like gp100_vmm_pgt_pfn()
> +	 *
> +	 * This is all just encoding the internal hmm reprensetation into a

s/reprensetation/representation/

Looks good and still tests OK with nouveau.