search for: max_segments

Add max_segment argument to drm_prime_pages_to_sg(). When set pass it through to the __sg_alloc_table_from_pages() call, otherwise use SCATTERLIST_MAX_SEGMENT. Also add max_segment field to drm driver and pass it to drm_prime_pages_to_sg() calls in drivers and helpers. v2: place max_segment in drm driver not gem object. Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> ---

Add max_segment argument to drm_prime_pages_to_sg(). When set pass it through to the __sg_alloc_table_from_pages() call, otherwise use SCATTERLIST_MAX_SEGMENT. Also add max_segment field to gem objects and pass it to drm_prime_pages_to_sg() calls in drivers and helpers. Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- include/drm/drm_gem.h | 8 ++++++++

Add max_segment argument to drm_prime_pages_to_sg(). When set pass it through to the __sg_alloc_table_from_pages() call, otherwise use SCATTERLIST_MAX_SEGMENT. Also add max_segment field to drm driver and pass it to drm_prime_pages_to_sg() calls in drivers and helpers. v2: place max_segment in drm driver not gem object. v3: move max_segment next to the other gem fields. Signed-off-by: Gerd

On Tue, Aug 18, 2020 at 11:20:16AM +0200, Gerd Hoffmann wrote: > Add max_segment argument to drm_prime_pages_to_sg(). When set pass it > through to the __sg_alloc_table_from_pages() call, otherwise use > SCATTERLIST_MAX_SEGMENT. > > Also add max_segment field to drm driver and pass it to > drm_prime_pages_to_sg() calls in drivers and helpers. > > v2: place max_segment in

Am 18.08.20 um 09:48 schrieb Gerd Hoffmann: > Add max_segment argument to drm_prime_pages_to_sg(). When set pass it > through to the __sg_alloc_table_from_pages() call, otherwise use > SCATTERLIST_MAX_SEGMENT. > > Also add max_segment field to gem objects and pass it to > drm_prime_pages_to_sg() calls in drivers and helpers. > > Signed-off-by: Gerd Hoffmann <kraxel at

> > + /** > > + * @max_segment: > > + * > > + * Max size for scatter list segments. When unset the default > > + * (SCATTERLIST_MAX_SEGMENT) is used. > > + */ > > + size_t max_segment; > > Is there no better place for this then "at the bottom"? drm_device is a > huge structure, piling stuff up randomly doesn't make it better

When initializing call virtio_max_dma_size() to figure the scatter list limit. Needed to make virtio-gpu work properly with SEV. v2: place max_segment in drm driver not gem object. Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_kms.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/virtio/virtgpu_kms.c

When initializing call virtio_max_dma_size() to figure the scatter list limit. Needed to make virtio-gpu work properly with SEV. v2: place max_segment in drm driver not gem object. Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_kms.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/virtio/virtgpu_kms.c

Hi, I have (very) poor write performance with Xen PV, used over RBD, and I suppose it''s because of very short values for max_segments and max_segment_size. (I have 250MB/s write speed in Dom0 vs 40MB/s write speed in DomU) From what I read, there is already a patch to fix that, called "multi page ring support for block devices" : http://lists.xen.org/archives/html/xen-devel/2012-03/msg00388.html But it doesn'&...

We must make sure our scatterlist segments are not too big, otherwise we might see swiotlb failures (happens with sev, also reproducable with swiotlb=force). Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_object.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/virtio/virtgpu_object.c

We must make sure our scatterlist segments are not too big, otherwise we might see swiotlb failures (happens with sev, also reproducable with swiotlb=force). Suggested-by: Laszlo Ersek <lersek at redhat.com> Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_object.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git

We must make sure our scatterlist segments are not too big, otherwise we might see swiotlb failures (happens with sev, also reproducable with swiotlb=force). Suggested-by: Laszlo Ersek <lersek at redhat.com> Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_object.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git

We must make sure our scatterlist segments are not too big, otherwise we might see swiotlb failures (happens with sev, also reproducable with swiotlb=force). Suggested-by: Laszlo Ersek <lersek at redhat.com> Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_object.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git

From: Gerd Hoffmann <kraxel at redhat.com> [ Upstream commit 9b2a0a1ef66f96bf34921a3865581eca32ff05ec ] We must make sure our scatterlist segments are not too big, otherwise we might see swiotlb failures (happens with sev, also reproducable with swiotlb=force). Suggested-by: Laszlo Ersek <lersek at redhat.com> Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> Reviewed-by:

When initializing gem objects call virtio_max_dma_size() to figure the scatter list limit. Needed to make virtio-gpu work properly with SEV. Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/virtio/virtgpu_object.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/virtio/virtgpu_object.c b/drivers/gpu/drm/virtio/virtgpu_object.c index

Add drm_device argument to drm_prime_pages_to_sg(), so we can call dma_max_mapping_size() to figure the segment size limit and call into __sg_alloc_table_from_pages() with the correct limit. This fixes virtio-gpu with sev. Possibly it'll fix other bugs too given that drm seems to totaly ignore segment size limits so far ... v2: place max_segment in drm driver not gem object. v3: move

...+ continue; + } + + break; + } + + closedir(ffs); + + pstrcat(path, MAXPATHLEN, d->d_name); + pstrcat(path, MAXPATHLEN, "/queue/"); + + read_queue_limit(path, "max_sectors_kb", &bs->max_sectors); + read_queue_limit(path, "max_segments", &bs->max_segments); + read_queue_limit(path, "max_segment_size", &bs->max_segment_size); +} + static int hdev_open(BlockDriverState *bs, const char *filename, int flags) { BDRVRawState *s = bs->opaque; @@ -868,6 +925,7 @@ static int hdev_open(BlockDriverS...

...+ continue; + } + + break; + } + + closedir(ffs); + + pstrcat(path, MAXPATHLEN, d->d_name); + pstrcat(path, MAXPATHLEN, "/queue/"); + + read_queue_limit(path, "max_sectors_kb", &bs->max_sectors); + read_queue_limit(path, "max_segments", &bs->max_segments); + read_queue_limit(path, "max_segment_size", &bs->max_segment_size); +} + static int hdev_open(BlockDriverState *bs, const char *filename, int flags) { BDRVRawState *s = bs->opaque; @@ -868,6 +925,7 @@ static int hdev_open(BlockDriverS...

...wake_up(&blkbk->pending_free_wq); } - +/* + * Retrieve from the ''pending_reqs'' a free pending_req structure to be used. + */ +static struct pending_req *alloc_req(void) +{ + struct pending_req *req = NULL; + unsigned long flags; + unsigned int max_seg = BLKIF_MAX_SEGMENTS_PER_REQUEST; + + spin_lock_irqsave(&blkbk->pending_free_lock, flags); + if (!list_empty(&blkbk->pending_free)) { + req = list_entry(blkbk->pending_free.next, struct pending_req, + free_list); + list_del(&req->free_list); + } +...

On Mon, Sep 7, 2020 at 1:24 PM Gerd Hoffmann <kraxel at redhat.com> wrote: > > Add drm_device argument to drm_prime_pages_to_sg(), so we can > call dma_max_mapping_size() to figure the segment size limit > and call into __sg_alloc_table_from_pages() with the correct > limit. > > This fixes virtio-gpu with sev. Possibly it'll fix other bugs > too given that drm