linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Marek Szyprowski <m.szyprowski@samsung.com>
To: Sumit Semwal <sumit.semwal@linaro.org>,
	linux-kernel@vger.kernel.org, linux-media@vger.kernel.org,
	dri-devel@lists.freedesktop.org, linaro-mm-sig@lists.linaro.org,
	linux-arm-kernel@lists.infradead.org, linux-mm@kvack.org
Cc: linaro-kernel@lists.linaro.org, robdclark@gmail.com,
	daniel@ffwll.ch, stanislawski.tomasz@googlemail.com,
	robin.murphy@arm.com
Subject: Re: [RFCv3 2/2] dma-buf: add helpers for sharing attacher constraints with dma-parms
Date: Wed, 11 Feb 2015 09:28:37 +0100	[thread overview]
Message-ID: <54DB12B5.4080000@samsung.com> (raw)
In-Reply-To: <1422347154-15258-2-git-send-email-sumit.semwal@linaro.org>

Hello,

On 2015-01-27 09:25, Sumit Semwal wrote:
> Add some helpers to share the constraints of devices while attaching
> to the dmabuf buffer.
>
> At each attach, the constraints are calculated based on the following:
> - max_segment_size, max_segment_count, segment_boundary_mask from
>     device_dma_parameters.
>
> In case the attaching device's constraints don't match up, attach() fails.
>
> At detach, the constraints are recalculated based on the remaining
> attached devices.
>
> Two helpers are added:
> - dma_buf_get_constraints - which gives the current constraints as calculated
>        during each attach on the buffer till the time,
> - dma_buf_recalc_constraints - which recalculates the constraints for all
>        currently attached devices for the 'paranoid' ones amongst us.
>
> The idea of this patch is largely taken from Rob Clark's RFC at
> https://lkml.org/lkml/2012/7/19/285, and the comments received on it.
>
> Cc: Rob Clark <robdclark@gmail.com>
> Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>

The code looks okay, although it will probably will work well only with 
typical
cases like 'contiguous memory needed' or 'no constraints at all' (iommu).

Acked-by: Marek Szyprowski <m.szyprowski@samsung.com>

> ---
> v3:
> - Thanks to Russell's comment, remove dma_mask and coherent_dma_mask from
>    constraints' calculation; has a nice side effect of letting us use
>    device_dma_parameters directly to list constraints.
> - update the debugfs output to show constraint info as well.
>    
> v2: split constraints-sharing and allocation helpers
>
>   drivers/dma-buf/dma-buf.c | 126 +++++++++++++++++++++++++++++++++++++++++++++-
>   include/linux/dma-buf.h   |   7 +++
>   2 files changed, 132 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c
> index 5be225c2ba98..f363f1440803 100644
> --- a/drivers/dma-buf/dma-buf.c
> +++ b/drivers/dma-buf/dma-buf.c
> @@ -264,6 +264,66 @@ static inline int is_dma_buf_file(struct file *file)
>   	return file->f_op == &dma_buf_fops;
>   }
>   
> +static inline void init_constraints(struct device_dma_parameters *cons)
> +{
> +	cons->max_segment_count = (unsigned int)-1;
> +	cons->max_segment_size = (unsigned int)-1;
> +	cons->segment_boundary_mask = (unsigned long)-1;
> +}
> +
> +/*
> + * calc_constraints - calculates if the new attaching device's constraints
> + * match, with the constraints of already attached devices; if yes, returns
> + * the constraints; else return ERR_PTR(-EINVAL)
> + */
> +static int calc_constraints(struct device *dev,
> +			    struct device_dma_parameters *calc_cons)
> +{
> +	struct device_dma_parameters cons = *calc_cons;
> +
> +	cons.max_segment_count = min(cons.max_segment_count,
> +					dma_get_max_seg_count(dev));
> +	cons.max_segment_size = min(cons.max_segment_size,
> +					dma_get_max_seg_size(dev));
> +	cons.segment_boundary_mask &= dma_get_seg_boundary(dev);
> +
> +	if (!cons.max_segment_count ||
> +	    !cons.max_segment_size ||
> +	    !cons.segment_boundary_mask) {
> +		pr_err("Dev: %s's constraints don't match\n", dev_name(dev));
> +		return -EINVAL;
> +	}
> +
> +	*calc_cons = cons;
> +
> +	return 0;
> +}
> +
> +/*
> + * recalc_constraints - recalculates constraints for all attached devices;
> + *  useful for detach() recalculation, and for dma_buf_recalc_constraints()
> + *  helper.
> + *  Returns recalculated constraints in recalc_cons, or error in the unlikely
> + *  case when constraints of attached devices might have changed.
> + */
> +static int recalc_constraints(struct dma_buf *dmabuf,
> +			      struct device_dma_parameters *recalc_cons)
> +{
> +	struct device_dma_parameters calc_cons;
> +	struct dma_buf_attachment *attach;
> +	int ret = 0;
> +
> +	init_constraints(&calc_cons);
> +
> +	list_for_each_entry(attach, &dmabuf->attachments, node) {
> +		ret = calc_constraints(attach->dev, &calc_cons);
> +		if (ret)
> +			return ret;
> +	}
> +	*recalc_cons = calc_cons;
> +	return 0;
> +}
> +
>   /**
>    * dma_buf_export_named - Creates a new dma_buf, and associates an anon file
>    * with this buffer, so it can be exported.
> @@ -313,6 +373,9 @@ struct dma_buf *dma_buf_export_named(void *priv, const struct dma_buf_ops *ops,
>   	dmabuf->ops = ops;
>   	dmabuf->size = size;
>   	dmabuf->exp_name = exp_name;
> +
> +	init_constraints(&dmabuf->constraints);
> +
>   	init_waitqueue_head(&dmabuf->poll);
>   	dmabuf->cb_excl.poll = dmabuf->cb_shared.poll = &dmabuf->poll;
>   	dmabuf->cb_excl.active = dmabuf->cb_shared.active = 0;
> @@ -422,7 +485,7 @@ struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf,
>   					  struct device *dev)
>   {
>   	struct dma_buf_attachment *attach;
> -	int ret;
> +	int ret = 0;
>   
>   	if (WARN_ON(!dmabuf || !dev))
>   		return ERR_PTR(-EINVAL);
> @@ -436,6 +499,9 @@ struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf,
>   
>   	mutex_lock(&dmabuf->lock);
>   
> +	if (calc_constraints(dev, &dmabuf->constraints))
> +		goto err_constraints;
> +
>   	if (dmabuf->ops->attach) {
>   		ret = dmabuf->ops->attach(dmabuf, dev, attach);
>   		if (ret)
> @@ -448,6 +514,7 @@ struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf,
>   
>   err_attach:
>   	kfree(attach);
> +err_constraints:
>   	mutex_unlock(&dmabuf->lock);
>   	return ERR_PTR(ret);
>   }
> @@ -470,6 +537,8 @@ void dma_buf_detach(struct dma_buf *dmabuf, struct dma_buf_attachment *attach)
>   	if (dmabuf->ops->detach)
>   		dmabuf->ops->detach(dmabuf, attach);
>   
> +	recalc_constraints(dmabuf, &dmabuf->constraints);
> +
>   	mutex_unlock(&dmabuf->lock);
>   	kfree(attach);
>   }
> @@ -770,6 +839,56 @@ void dma_buf_vunmap(struct dma_buf *dmabuf, void *vaddr)
>   }
>   EXPORT_SYMBOL_GPL(dma_buf_vunmap);
>   
> +/**
> + * dma_buf_get_constraints - get the *current* constraints of the dmabuf,
> + *  as calculated during each attach(); returns error on invalid inputs
> + *
> + * @dmabuf:		[in]	buffer to get constraints of
> + * @constraints:	[out]	current constraints are returned in this
> + */
> +int dma_buf_get_constraints(struct dma_buf *dmabuf,
> +			    struct device_dma_parameters *constraints)
> +{
> +	if (WARN_ON(!dmabuf || !constraints))
> +		return -EINVAL;
> +
> +	mutex_lock(&dmabuf->lock);
> +	*constraints = dmabuf->constraints;
> +	mutex_unlock(&dmabuf->lock);
> +	return 0;
> +}
> +EXPORT_SYMBOL_GPL(dma_buf_get_constraints);
> +
> +/**
> + * dma_buf_recalc_constraints - *recalculate* the constraints for the buffer
> + *  afresh, from the list of currently attached devices; this could be useful
> + *  cross-check the current constraints, for exporters that might want to be
> + *  'paranoid' about the device constraints.
> + *
> + *  returns error on invalid inputs
> + *
> + * @dmabuf:		[in]	buffer to get constraints of
> + * @constraints:	[out]	recalculated constraints are returned in this
> + */
> +int dma_buf_recalc_constraints(struct dma_buf *dmabuf,
> +			    struct device_dma_parameters *constraints)
> +{
> +	struct device_dma_parameters calc_cons;
> +	int ret = 0;
> +
> +	if (WARN_ON(!dmabuf || !constraints))
> +		return -EINVAL;
> +
> +	mutex_lock(&dmabuf->lock);
> +	ret = recalc_constraints(dmabuf, &calc_cons);
> +	if (!ret)
> +		*constraints = calc_cons;
> +
> +	mutex_unlock(&dmabuf->lock);
> +	return ret;
> +}
> +EXPORT_SYMBOL_GPL(dma_buf_recalc_constraints);
> +
>   #ifdef CONFIG_DEBUG_FS
>   static int dma_buf_describe(struct seq_file *s)
>   {
> @@ -801,6 +920,11 @@ static int dma_buf_describe(struct seq_file *s)
>   				buf_obj->file->f_flags, buf_obj->file->f_mode,
>   				file_count(buf_obj->file),
>   				buf_obj->exp_name);
> +		seq_printf(s, "\tConstraints: Seg Count: %08u, Seg Size: %08u",
> +				buf_obj->constraints.max_segment_count,
> +				buf_obj->constraints.max_segment_size);
> +		seq_printf(s, " seg boundary mask: %08lx\n",
> +				buf_obj->constraints.segment_boundary_mask);
>   
>   		seq_puts(s, "\tAttached Devices:\n");
>   		attach_count = 0;
> diff --git a/include/linux/dma-buf.h b/include/linux/dma-buf.h
> index 694e1fe1c4b4..489ad9b2e5ae 100644
> --- a/include/linux/dma-buf.h
> +++ b/include/linux/dma-buf.h
> @@ -34,6 +34,7 @@
>   #include <linux/wait.h>
>   
>   struct device;
> +struct device_dma_parameters;
>   struct dma_buf;
>   struct dma_buf_attachment;
>   
> @@ -116,6 +117,7 @@ struct dma_buf_ops {
>    * @ops: dma_buf_ops associated with this buffer object.
>    * @exp_name: name of the exporter; useful for debugging.
>    * @list_node: node for dma_buf accounting and debugging.
> + * @constraints: calculated constraints of attached devices.
>    * @priv: exporter specific private data for this buffer object.
>    * @resv: reservation object linked to this dma-buf
>    */
> @@ -130,6 +132,7 @@ struct dma_buf {
>   	void *vmap_ptr;
>   	const char *exp_name;
>   	struct list_head list_node;
> +	struct device_dma_parameters constraints;
>   	void *priv;
>   	struct reservation_object *resv;
>   
> @@ -211,4 +214,8 @@ void *dma_buf_vmap(struct dma_buf *);
>   void dma_buf_vunmap(struct dma_buf *, void *vaddr);
>   int dma_buf_debugfs_create_file(const char *name,
>   				int (*write)(struct seq_file *));
> +
> +int dma_buf_get_constraints(struct dma_buf *, struct device_dma_parameters *);
> +int dma_buf_recalc_constraints(struct dma_buf *,
> +					struct device_dma_parameters *);
>   #endif /* __DMA_BUF_H__ */

Best regards
-- 
Marek Szyprowski, PhD
Samsung R&D Institute Poland


  parent reply	other threads:[~2015-02-11  8:28 UTC|newest]

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-27  8:25 [RFCv3 1/2] device: add dma_params->max_segment_count Sumit Semwal
2015-01-27  8:25 ` [RFCv3 2/2] dma-buf: add helpers for sharing attacher constraints with dma-parms Sumit Semwal
2015-01-29 14:16   ` Maarten Lankhorst
2015-01-29 14:39   ` Russell King - ARM Linux
2015-01-29 15:30     ` Sumit Semwal
2015-01-29 15:47       ` Russell King - ARM Linux
2015-01-29 16:55         ` Sumit Semwal
2015-01-29 18:52         ` Rob Clark
2015-01-29 19:26           ` Russell King - ARM Linux
2015-01-29 22:18             ` Rob Clark
2015-01-29 22:31               ` Russell King - ARM Linux
2015-01-29 23:19                 ` Rob Clark
2015-02-02 16:54               ` Daniel Vetter
2015-02-02 20:30                 ` Rob Clark
2015-02-02 21:46                   ` Russell King - ARM Linux
2015-02-02 22:36                     ` Rob Clark
2015-02-03  7:50                       ` Daniel Vetter
2015-02-03  7:46                     ` Daniel Vetter
2015-02-03  7:48                   ` Daniel Vetter
2015-02-03 12:28                     ` Russell King - ARM Linux
2015-02-03 13:00                       ` Daniel Vetter
2015-02-03 13:28                       ` Christian Gmeiner
2015-02-03 14:32                         ` Russell King - ARM Linux
2015-02-03 14:25                       ` Rob Clark
2015-02-03 14:04                     ` Rob Clark
2015-02-03 14:17                       ` Arnd Bergmann
2015-02-03 14:41                         ` Russell King - ARM Linux
2015-02-03 14:52                           ` Arnd Bergmann
2015-02-03 15:22                             ` Russell King - ARM Linux
2015-02-03 15:31                               ` [Linaro-mm-sig] " Arnd Bergmann
2015-02-03 15:54                                 ` Russell King - ARM Linux
2015-02-03 16:12                                   ` Arnd Bergmann
2015-02-03 16:22                                     ` Rob Clark
2015-02-03 16:36                                       ` Arnd Bergmann
2015-02-03 20:04                                         ` Daniel Vetter
2015-02-03 21:42                                           ` Arnd Bergmann
2015-02-03 22:07                                             ` Daniel Vetter
2015-02-04  0:14                                             ` Russell King - ARM Linux
2015-02-03 17:01                                       ` Russell King - ARM Linux
2015-02-03 16:58                                     ` Russell King - ARM Linux
2015-02-03 17:35                                       ` Rob Clark
2015-02-03 20:08                                         ` Daniel Vetter
2015-02-03 21:44                                         ` Arnd Bergmann
2015-02-03 15:25                             ` Rob Clark
2015-02-03 15:19                           ` Rob Clark
2015-02-03 14:37                       ` Russell King - ARM Linux
2015-02-03 14:44                         ` Rob Clark
2015-02-03 14:58                           ` Russell King - ARM Linux
2015-02-02  5:53             ` Sumit Semwal
2015-02-11  8:28   ` Marek Szyprowski [this message]
2015-02-11 11:12     ` Russell King - ARM Linux
2015-02-11 11:23       ` Rob Clark
2015-02-11 12:56         ` Daniel Vetter
2015-02-11 13:30           ` Rob Clark
2015-02-11 12:20       ` Marek Szyprowski
2015-02-11 16:23         ` Russell King - ARM Linux
2015-05-05 14:41           ` Sumit Semwal
2015-06-03  6:39             ` [Linaro-mm-sig] " Hans Verkuil
2015-06-03  8:41               ` Russell King - ARM Linux
2015-06-03  9:37                 ` Hans Verkuil
2015-06-04  5:24                   ` Sumit Semwal
2015-01-28 14:09 ` [RFCv3 1/2] device: add dma_params->max_segment_count Marek Szyprowski
2015-06-03  6:13 ` [Linaro-mm-sig] " Hans Verkuil

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=54DB12B5.4080000@samsung.com \
    --to=m.szyprowski@samsung.com \
    --cc=daniel@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=linaro-kernel@lists.linaro.org \
    --cc=linaro-mm-sig@lists.linaro.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=robdclark@gmail.com \
    --cc=robin.murphy@arm.com \
    --cc=stanislawski.tomasz@googlemail.com \
    --cc=sumit.semwal@linaro.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 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).