From: Thomas Zimmermann <tzimmermann@suse.de>
To: Gerd Hoffmann <kraxel@redhat.com>, dri-devel@lists.freedesktop.org
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>,
Christian Koenig <christian.koenig@amd.com>,
Huang Rui <ray.huang@amd.com>, David Airlie <airlied@linux.ie>,
Daniel Vetter <daniel@ffwll.ch>,
open list <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 4/8] drm/ttm: factor out ttm_bo_mmap_vma_setup
Date: Fri, 13 Sep 2019 14:56:09 +0200 [thread overview]
Message-ID: <7c8d1570-3b07-719a-01d3-59bd32c51366@suse.de> (raw)
In-Reply-To: <20190913122908.784-5-kraxel@redhat.com>
[-- Attachment #1.1: Type: text/plain, Size: 4019 bytes --]
Hi
Am 13.09.19 um 14:29 schrieb Gerd Hoffmann:
> Factor out ttm vma setup to a new function. Reduces
> code duplication a bit and allows to implement
> &drm_gem_object_funcs.mmap in gem ttm helpers.
>
> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
> ---
> include/drm/ttm/ttm_bo_api.h | 8 ++++++
> drivers/gpu/drm/ttm/ttm_bo_vm.c | 47 ++++++++++++++++++---------------
> 2 files changed, 33 insertions(+), 22 deletions(-)
>
> diff --git a/include/drm/ttm/ttm_bo_api.h b/include/drm/ttm/ttm_bo_api.h
> index 43c4929a2171..88c652f49602 100644
> --- a/include/drm/ttm/ttm_bo_api.h
> +++ b/include/drm/ttm/ttm_bo_api.h
> @@ -734,6 +734,14 @@ int ttm_fbdev_mmap(struct vm_area_struct *vma, struct ttm_buffer_object *bo);
> int ttm_bo_mmap(struct file *filp, struct vm_area_struct *vma,
> struct ttm_bo_device *bdev);
>
> +/**
> + * ttm_bo_mmap_vma_setup - initialize vma for ttm bo mmap
> + *
> + * @bo: The buffer object.
> + * @vma: vma as input from the mmap method.
> + */
> +void ttm_bo_mmap_vma_setup(struct ttm_buffer_object *bo, struct vm_area_struct *vma);
> +
> void *ttm_kmap_atomic_prot(struct page *page, pgprot_t prot);
>
> void ttm_kunmap_atomic_prot(void *addr, pgprot_t prot);
> diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c
> index 4aa007edffb0..7c0e85c10e0e 100644
> --- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
> +++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
> @@ -426,6 +426,29 @@ static struct ttm_buffer_object *ttm_bo_vm_lookup(struct ttm_bo_device *bdev,
> return bo;
> }
>
> +void ttm_bo_mmap_vma_setup(struct ttm_buffer_object *bo, struct vm_area_struct *vma)
> +{
> + vma->vm_ops = &ttm_bo_vm_ops;
> +
> + /*
> + * Note: We're transferring the bo reference to
> + * vma->vm_private_data here.
> + */
> +
> + vma->vm_private_data = bo;
> +
> + /*
> + * We'd like to use VM_PFNMAP on shared mappings, where
> + * (vma->vm_flags & VM_SHARED) != 0, for performance reasons,
> + * but for some reason VM_PFNMAP + x86 PAT + write-combine is very
> + * bad for performance. Until that has been sorted out, use
> + * VM_MIXEDMAP on all mappings. See freedesktop.org bug #75719
> + */
> + vma->vm_flags |= VM_MIXEDMAP;
> + vma->vm_flags |= VM_IO | VM_DONTEXPAND | VM_DONTDUMP;
> +}
> +EXPORT_SYMBOL(ttm_bo_mmap_vma_setup);
> +
> int ttm_bo_mmap(struct file *filp, struct vm_area_struct *vma,
> struct ttm_bo_device *bdev)
> {
> @@ -449,24 +472,7 @@ int ttm_bo_mmap(struct file *filp, struct vm_area_struct *vma,
> if (unlikely(ret != 0))
> goto out_unref;
>
> - vma->vm_ops = &ttm_bo_vm_ops;
> -
> - /*
> - * Note: We're transferring the bo reference to
> - * vma->vm_private_data here.
> - */
> -
> - vma->vm_private_data = bo;
> -
> - /*
> - * We'd like to use VM_PFNMAP on shared mappings, where
> - * (vma->vm_flags & VM_SHARED) != 0, for performance reasons,
> - * but for some reason VM_PFNMAP + x86 PAT + write-combine is very
> - * bad for performance. Until that has been sorted out, use
> - * VM_MIXEDMAP on all mappings. See freedesktop.org bug #75719
> - */
> - vma->vm_flags |= VM_MIXEDMAP;
> - vma->vm_flags |= VM_IO | VM_DONTEXPAND | VM_DONTDUMP;
> + ttm_bo_mmap_vma_setup(bo, vma);
> return 0;
> out_unref:
> ttm_bo_put(bo);
> @@ -481,10 +487,7 @@ int ttm_fbdev_mmap(struct vm_area_struct *vma, struct ttm_buffer_object *bo)
>
> ttm_bo_get(bo);
>
> - vma->vm_ops = &ttm_bo_vm_ops;
> - vma->vm_private_data = bo;
> - vma->vm_flags |= VM_MIXEDMAP;
> - vma->vm_flags |= VM_IO | VM_DONTEXPAND;
> + ttm_bo_mmap_vma_setup(bo, vma);
Just double-checking: ttm_bo_mmap_vma_setup() will set VM_DONTDUMP in
vm_flags. Is that OK?
Best regards
Thomas
> return 0;
> }
> EXPORT_SYMBOL(ttm_fbdev_mmap);
>
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Linux GmbH, Maxfeldstrasse 5, 90409 Nuernberg, Germany
GF: Felix Imendörffer, Mary Higgins, Sri Rasiah
HRB 21284 (AG Nürnberg)
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
next prev parent reply other threads:[~2019-09-13 12:56 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20190913122908.784-1-kraxel@redhat.com>
2019-09-13 12:29 ` [PATCH 1/8] drm: add mmap() to drm_gem_object_funcs Gerd Hoffmann
2019-09-17 13:27 ` Daniel Vetter
2019-09-13 12:29 ` [PATCH 2/8] drm/shmem: switch shmem helper to &drm_gem_object_funcs.mmap Gerd Hoffmann
2019-09-13 14:03 ` Steven Price
2019-09-17 8:14 ` Gerd Hoffmann
2019-09-13 12:29 ` [PATCH 3/8] drm/shmem: drop DEFINE_DRM_GEM_SHMEM_FOPS Gerd Hoffmann
2019-09-16 22:07 ` Rob Herring
2019-09-17 8:18 ` Gerd Hoffmann
2019-09-13 12:29 ` [PATCH 4/8] drm/ttm: factor out ttm_bo_mmap_vma_setup Gerd Hoffmann
2019-09-13 12:56 ` Thomas Zimmermann [this message]
2019-09-17 8:26 ` Gerd Hoffmann
2019-09-13 13:05 ` Thomas Zimmermann
2019-09-17 8:34 ` Gerd Hoffmann
2019-09-17 9:38 ` Thomas Zimmermann
2019-09-17 9:54 ` Gerd Hoffmann
2019-09-13 12:29 ` [PATCH 5/8] drm/ttm: add drm_gem_ttm_mmap() Gerd Hoffmann
2019-09-13 13:02 ` Thomas Zimmermann
2019-09-13 12:29 ` [PATCH 6/8] drm/vram: switch vram helper to &drm_gem_object_funcs.mmap() Gerd Hoffmann
2019-09-13 13:10 ` Thomas Zimmermann
2019-09-13 12:29 ` [PATCH 7/8] drm/vram: drop verify_access Gerd Hoffmann
2019-09-13 13:11 ` Thomas Zimmermann
2019-09-13 12:29 ` [PATCH 8/8] drm/vram: drop DRM_VRAM_MM_FILE_OPERATIONS Gerd Hoffmann
2019-09-13 13:18 ` Thomas Zimmermann
2019-09-17 8:39 ` Gerd Hoffmann
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=7c8d1570-3b07-719a-01d3-59bd32c51366@suse.de \
--to=tzimmermann@suse.de \
--cc=airlied@linux.ie \
--cc=christian.koenig@amd.com \
--cc=daniel.vetter@ffwll.ch \
--cc=daniel@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=kraxel@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).