All of lore.kernel.org
 help / color / mirror / Atom feed
From: Timur Tabi <timur-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
To: Sinan Kaya <okaya-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>,
	Christoph Hellwig <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	cov-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org,
	Yishai Hadas <yishaih-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>,
	netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH V2] net: ethernet: mellanox: correct page conversion
Date: Mon, 18 Apr 2016 10:15:56 -0500	[thread overview]
Message-ID: <5714FA2C.4030209@codeaurora.org> (raw)
In-Reply-To: <5714E5D6.7050600-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>

Sinan Kaya wrote:
>
> VMAP allows you to make several pages look contiguous to the CPU.
> It can only be used against logical addresses returned from kmalloc
> or alloc_page.
>
> You cannot take several virtually mapped addresses returned by dma_alloc_coherent
> and try to make them virtually contiguous again.
>
> The code happens to work on other architectures by pure luck. AFAIK, dma_alloc_coherent
> returns logical addresses on Intel systems until it runs out of DMA memory. After
> that intel arch will also start returning virtually mapped addresses and this code
> will also fail. ARM64 on the other hand always returns a virtually mapped address.
>
> The goal of this code is to allocate a bunch of page sized memory and make it look
> contiguous. It is just using the wrong API. The correct API is either kmalloc or
> alloc_page map it with dma_map_page not dma_alloc_coherent.
>
> The proper usage of dma_map_page requires code to call dma_sync API in correct
> places to be compatible with noncoherent systems. This code is already assuming
> coherency. It would be a nice to have dma_sync APIs in right places. There is no
> harm in calling dma_sync API for coherent systems as they are no-ops in DMA mapping
> layer whereas it is a cache flush for non-coherent systems.

The text would be a great addition to the patch description.

-- 
Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora
Forum, a Linux Foundation collaborative project.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

WARNING: multiple messages have this Message-ID (diff)
From: Timur Tabi <timur@codeaurora.org>
To: Sinan Kaya <okaya@codeaurora.org>, Christoph Hellwig <hch@infradead.org>
Cc: linux-rdma@vger.kernel.org, cov@codeaurora.org,
	Yishai Hadas <yishaih@mellanox.com>,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH V2] net: ethernet: mellanox: correct page conversion
Date: Mon, 18 Apr 2016 10:15:56 -0500	[thread overview]
Message-ID: <5714FA2C.4030209@codeaurora.org> (raw)
In-Reply-To: <5714E5D6.7050600@codeaurora.org>

Sinan Kaya wrote:
>
> VMAP allows you to make several pages look contiguous to the CPU.
> It can only be used against logical addresses returned from kmalloc
> or alloc_page.
>
> You cannot take several virtually mapped addresses returned by dma_alloc_coherent
> and try to make them virtually contiguous again.
>
> The code happens to work on other architectures by pure luck. AFAIK, dma_alloc_coherent
> returns logical addresses on Intel systems until it runs out of DMA memory. After
> that intel arch will also start returning virtually mapped addresses and this code
> will also fail. ARM64 on the other hand always returns a virtually mapped address.
>
> The goal of this code is to allocate a bunch of page sized memory and make it look
> contiguous. It is just using the wrong API. The correct API is either kmalloc or
> alloc_page map it with dma_map_page not dma_alloc_coherent.
>
> The proper usage of dma_map_page requires code to call dma_sync API in correct
> places to be compatible with noncoherent systems. This code is already assuming
> coherency. It would be a nice to have dma_sync APIs in right places. There is no
> harm in calling dma_sync API for coherent systems as they are no-ops in DMA mapping
> layer whereas it is a cache flush for non-coherent systems.

The text would be a great addition to the patch description.

-- 
Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora
Forum, a Linux Foundation collaborative project.

  parent reply	other threads:[~2016-04-18 15:15 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-16 22:23 [PATCH V2] net: ethernet: mellanox: correct page conversion Sinan Kaya
     [not found] ` <1460845412-13120-1-git-send-email-okaya-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2016-04-18  4:00   ` David Miller
2016-04-18  4:00     ` David Miller
2016-04-18  5:06     ` okaya
     [not found]       ` <87c17d3f979cf0167cd37077f39d0534-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2016-04-18 15:59         ` David Miller
2016-04-18 15:59           ` David Miller
     [not found]           ` <20160418.115902.1053705461620271779.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
2016-04-18 16:06             ` Sinan Kaya
2016-04-18 16:06               ` Sinan Kaya
2016-04-18 12:12   ` Christoph Hellwig
2016-04-18 12:12     ` Christoph Hellwig
2016-04-18 13:06     ` okaya
2016-04-18 13:10       ` Christoph Hellwig
2016-04-18 13:49         ` Sinan Kaya
2016-04-18 13:59           ` Christoph Hellwig
     [not found]           ` <5714E5D6.7050600-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2016-04-18 15:15             ` Timur Tabi [this message]
2016-04-18 15:15               ` Timur Tabi
     [not found]               ` <5714FA2C.4030209-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2016-04-18 15:22                 ` Sinan Kaya
2016-04-18 15:22                   ` Sinan Kaya
2016-04-19 18:22   ` Christoph Hellwig
2016-04-19 18:22     ` Christoph Hellwig
2016-04-19 18:37     ` Sinan Kaya
2016-04-20 11:08       ` Eran Ben Elisha
     [not found]     ` <20160419182212.GA8441-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2016-04-20 13:35       ` Sinan Kaya
2016-04-20 13:35         ` Sinan Kaya
     [not found]         ` <571785A5.5040306-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2016-04-20 13:38           ` Sinan Kaya
2016-04-20 13:38             ` Sinan Kaya
     [not found]             ` <57178663.4050503-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2016-04-20 13:41               ` Timur Tabi
2016-04-20 13:41                 ` Timur Tabi
2016-04-20 18:40           ` Eran Ben Elisha
2016-04-20 18:40             ` Eran Ben Elisha
     [not found]             ` <CAKHjkjmYNLABO10V1DZQmZ_zczjbfDZU0TDPHoMmv_1FMi9_gA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-04-20 18:42               ` Sinan Kaya
2016-04-20 18:42                 ` Sinan Kaya
2016-04-21 13:37               ` Or Gerlitz
2016-04-21 13:39                 ` Christoph Hellwig
     [not found]                   ` <20160421133903.GA19633-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2016-04-21 13:45                     ` Or Gerlitz
2016-04-25 13:47                   ` Eran Ben Elisha
2016-04-18  6:54 ` Eli Cohen
2016-04-18 13:53   ` Sinan Kaya
2016-04-18 14:05     ` Eli Cohen
2016-04-18 14:32   ` Christoph Hellwig
2016-04-18 14:39     ` Eli Cohen
     [not found]       ` <DB5PR05MB1848424EC9DC8D4150EADB03C56B0-8IvNv+8VlcDdorXcTtKhldqRiQSDpxhJvxpqHgZTriW3zl9H0oFU5g@public.gmane.org>
2016-04-18 15:17         ` Christoph Hellwig
2016-04-18 15:17           ` Christoph Hellwig
     [not found]           ` <20160418151702.GA26565-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2016-04-18 15:21             ` Sinan Kaya
2016-04-18 15:21               ` Sinan Kaya
     [not found]               ` <5714FB68.3020604-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2016-04-18 15:40                 ` Christoph Hellwig
2016-04-18 15:40                   ` Christoph Hellwig
2016-04-18 17:47                   ` Sinan Kaya
2016-04-19  7:50                   ` Sinan Kaya

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=5714FA2C.4030209@codeaurora.org \
    --to=timur-sgv2jx0feol9jmxxk+q4oq@public.gmane.org \
    --cc=cov-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org \
    --cc=hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org \
    --cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=okaya-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org \
    --cc=yishaih-VPRAkNaXOzVWk0Htik3J/w@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.