From: catalin.marinas@arm.com (Catalin Marinas)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v2 0/7] of: setup dma parameters using dma-ranges and dma-coherent
Date: Tue, 22 Apr 2014 16:25:29 +0100 [thread overview]
Message-ID: <20140422152529.GE10224@arm.com> (raw)
In-Reply-To: <201404221702.19842.arnd@arndb.de>
On Tue, Apr 22, 2014 at 04:02:19PM +0100, Arnd Bergmann wrote:
> On Saturday 19 April 2014, Thomas Petazzoni wrote:
> > I am not sure whether there is an intersection or not, but I wanted to
> > mention that the mvebu platform (in mach-mvebu) supports hardware I/O
> > coherency, which makes it a coherent DMA platform. However, we are not
> > able to use arm_coherent_dma_ops for this platform, because when a
> > transfer is being made DMA_FROM_DEVICE, at the end of the transfer, we
> > need to perform an I/O barrier to wait for the snooping unit to
> > complete its coherency work. So we're coherent, but not with
> > arm_coherent_dma_ops: we have our own dma operation implementation (see
> > arch/arm/mach-mvebu/coherency.c).
>
> I had completely missed the fact that this support was merged already.
>
> It's an interesting question if this should actually be called
> 'coherent' or not. It's certainly more coherent than without that
> support, but then again, you still can't rely on incoming data to
> be visible after a readl() from the device has returned or an MSI
> interrupt has been delivered, which is what we normally expect.
>
> In particular, it means you can't really use arm_coherent_dma_alloc(),
> which is a shame, since that is a significante performance overhead.
It should still work if __io*mb() macros do the extra work specific to
mvebu (similar to the L2x0 outer_sync(), though this was for
non-cacheable DMA buffers).
> I would hope we can find a way to avoid the platform notifiers for
> mvebu as well and come up with a generic way to express this
> 'semi-coherent' mode. I believe x-gene has a similar issue, and
> I wouldn't be surprised if there are others like this.
The solution is for the snooping unit to detect the DSB instruction
(which is propagated outside the CPU) and wait for the completion of the
coherency work (but we need more information from the hardware guys).
--
Catalin
WARNING: multiple messages have this Message-ID (diff)
From: Catalin Marinas <catalin.marinas-5wv7dgnIgG8@public.gmane.org>
To: Arnd Bergmann <arnd-r2nGTMty4D4@public.gmane.org>
Cc: "Thomas Petazzoni"
<thomas.petazzoni-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8@public.gmane.org>,
"Santosh Shilimkar"
<santosh.shilimkar-l0cyMroinI0@public.gmane.org>,
"linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
"devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
<devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
"Grygorii Strashko"
<grygorii.strashko-l0cyMroinI0@public.gmane.org>,
"Russell King" <linux-lFZ/pmaqli7XmaaqVzeoHQ@public.gmane.org>,
"Greg Kroah-Hartman"
<gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org>,
"Linus Walleij"
<linus.walleij-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>,
"grant.likely-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org"
<grant.likely-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>,
"Rob Herring" <robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
"Olof Johansson" <olof-nZhT3qVonbNeoWH0uzbU5w@public.gmane.org>,
"linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org"
<linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org>,
"Gregory Clément"
<gregory.clement-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8@public.gmane.org>,
"Ezequiel Garcia"
<ezequiel.garcia-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8@public.gmane.org>,
"Lior Amsalem" <alior-eYqpPyKDWXRBDgjK7y7TUQ@public.gmane.org>,
"Tawfik Bayouk" <tawfik-eYqpPyKDWXRBDgjK7y7TUQ@public.gmane.org>
Subject: Re: [PATCH v2 0/7] of: setup dma parameters using dma-ranges and dma-coherent
Date: Tue, 22 Apr 2014 16:25:29 +0100 [thread overview]
Message-ID: <20140422152529.GE10224@arm.com> (raw)
In-Reply-To: <201404221702.19842.arnd-r2nGTMty4D4@public.gmane.org>
On Tue, Apr 22, 2014 at 04:02:19PM +0100, Arnd Bergmann wrote:
> On Saturday 19 April 2014, Thomas Petazzoni wrote:
> > I am not sure whether there is an intersection or not, but I wanted to
> > mention that the mvebu platform (in mach-mvebu) supports hardware I/O
> > coherency, which makes it a coherent DMA platform. However, we are not
> > able to use arm_coherent_dma_ops for this platform, because when a
> > transfer is being made DMA_FROM_DEVICE, at the end of the transfer, we
> > need to perform an I/O barrier to wait for the snooping unit to
> > complete its coherency work. So we're coherent, but not with
> > arm_coherent_dma_ops: we have our own dma operation implementation (see
> > arch/arm/mach-mvebu/coherency.c).
>
> I had completely missed the fact that this support was merged already.
>
> It's an interesting question if this should actually be called
> 'coherent' or not. It's certainly more coherent than without that
> support, but then again, you still can't rely on incoming data to
> be visible after a readl() from the device has returned or an MSI
> interrupt has been delivered, which is what we normally expect.
>
> In particular, it means you can't really use arm_coherent_dma_alloc(),
> which is a shame, since that is a significante performance overhead.
It should still work if __io*mb() macros do the extra work specific to
mvebu (similar to the L2x0 outer_sync(), though this was for
non-cacheable DMA buffers).
> I would hope we can find a way to avoid the platform notifiers for
> mvebu as well and come up with a generic way to express this
> 'semi-coherent' mode. I believe x-gene has a similar issue, and
> I wouldn't be surprised if there are others like this.
The solution is for the snooping unit to detect the DSB instruction
(which is propagated outside the CPU) and wait for the completion of the
coherency work (but we need more information from the hardware guys).
--
Catalin
--
To unsubscribe from this list: send the line "unsubscribe devicetree" 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: Catalin Marinas <catalin.marinas@arm.com>
To: Arnd Bergmann <arnd@arndb.de>
Cc: "Thomas Petazzoni" <thomas.petazzoni@free-electrons.com>,
"Santosh Shilimkar" <santosh.shilimkar@ti.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"devicetree@vger.kernel.org" <devicetree@vger.kernel.org>,
"Grygorii Strashko" <grygorii.strashko@ti.com>,
"Russell King" <linux@arm.linux.org.uk>,
"Greg Kroah-Hartman" <gregkh@linuxfoundation.org>,
"Linus Walleij" <linus.walleij@linaro.org>,
"grant.likely@linaro.org" <grant.likely@linaro.org>,
"Rob Herring" <robh+dt@kernel.org>,
"Olof Johansson" <olof@lixom.net>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>,
"Gregory Clément" <gregory.clement@free-electrons.com>,
"Ezequiel Garcia" <ezequiel.garcia@free-electrons.com>,
"Lior Amsalem" <alior@marvell.com>,
"Tawfik Bayouk" <tawfik@marvell.com>
Subject: Re: [PATCH v2 0/7] of: setup dma parameters using dma-ranges and dma-coherent
Date: Tue, 22 Apr 2014 16:25:29 +0100 [thread overview]
Message-ID: <20140422152529.GE10224@arm.com> (raw)
In-Reply-To: <201404221702.19842.arnd@arndb.de>
On Tue, Apr 22, 2014 at 04:02:19PM +0100, Arnd Bergmann wrote:
> On Saturday 19 April 2014, Thomas Petazzoni wrote:
> > I am not sure whether there is an intersection or not, but I wanted to
> > mention that the mvebu platform (in mach-mvebu) supports hardware I/O
> > coherency, which makes it a coherent DMA platform. However, we are not
> > able to use arm_coherent_dma_ops for this platform, because when a
> > transfer is being made DMA_FROM_DEVICE, at the end of the transfer, we
> > need to perform an I/O barrier to wait for the snooping unit to
> > complete its coherency work. So we're coherent, but not with
> > arm_coherent_dma_ops: we have our own dma operation implementation (see
> > arch/arm/mach-mvebu/coherency.c).
>
> I had completely missed the fact that this support was merged already.
>
> It's an interesting question if this should actually be called
> 'coherent' or not. It's certainly more coherent than without that
> support, but then again, you still can't rely on incoming data to
> be visible after a readl() from the device has returned or an MSI
> interrupt has been delivered, which is what we normally expect.
>
> In particular, it means you can't really use arm_coherent_dma_alloc(),
> which is a shame, since that is a significante performance overhead.
It should still work if __io*mb() macros do the extra work specific to
mvebu (similar to the L2x0 outer_sync(), though this was for
non-cacheable DMA buffers).
> I would hope we can find a way to avoid the platform notifiers for
> mvebu as well and come up with a generic way to express this
> 'semi-coherent' mode. I believe x-gene has a similar issue, and
> I wouldn't be surprised if there are others like this.
The solution is for the snooping unit to detect the DSB instruction
(which is propagated outside the CPU) and wait for the completion of the
coherency work (but we need more information from the hardware guys).
--
Catalin
next prev parent reply other threads:[~2014-04-22 15:25 UTC|newest]
Thread overview: 92+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-04-19 14:32 [PATCH v2 0/7] of: setup dma parameters using dma-ranges and dma-coherent Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 14:32 ` [PATCH v2 1/7] device: introduce per device dma_pfn_offset Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 14:32 ` [PATCH v2 2/7] of: introduce of_dma_get_range() helper Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-21 15:29 ` Rob Herring
2014-04-21 15:29 ` Rob Herring
2014-04-21 15:29 ` Rob Herring
2014-04-22 14:56 ` Grygorii Strashko
2014-04-22 14:56 ` Grygorii Strashko
2014-04-22 14:56 ` Grygorii Strashko
2014-04-22 4:09 ` Joel Fernandes
2014-04-22 4:09 ` Joel Fernandes
2014-04-22 4:09 ` Joel Fernandes
2014-04-19 14:32 ` [PATCH v2 3/7] of: introduce of_dma_is_coherent() helper Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 14:32 ` [PATCH v2 4/7] of: configure the platform device dma parameters Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-21 14:58 ` Rob Herring
2014-04-21 14:58 ` Rob Herring
2014-04-21 14:58 ` Rob Herring
2014-04-22 15:09 ` Grygorii Strashko
2014-04-22 15:09 ` Grygorii Strashko
2014-04-22 15:09 ` Grygorii Strashko
2014-04-22 14:44 ` Rob Herring
2014-04-22 14:44 ` Rob Herring
2014-04-22 14:44 ` Rob Herring
2014-04-22 15:44 ` Grygorii Strashko
2014-04-22 15:44 ` Grygorii Strashko
2014-04-19 14:32 ` [PATCH v2 5/7] ARM: dma: Use dma_pfn_offset for dma address translation Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 19:43 ` Russell King - ARM Linux
2014-04-19 19:43 ` Russell King - ARM Linux
2014-04-19 19:43 ` Russell King - ARM Linux
2014-04-21 13:38 ` Santosh Shilimkar
2014-04-21 13:38 ` Santosh Shilimkar
2014-04-21 13:38 ` Santosh Shilimkar
2014-04-19 14:32 ` [PATCH v2 6/7] ARM: dma: implement set_arch_dma_coherent_ops() Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 14:32 ` [PATCH v2 7/7] ARM: mm: use phys_addr_t in __dma_page_[cpu_to_dev/dev_to_cpu] Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-19 14:32 ` Santosh Shilimkar
2014-04-21 13:39 ` Santosh Shilimkar
2014-04-21 13:39 ` Santosh Shilimkar
2014-04-21 13:39 ` Santosh Shilimkar
2014-04-19 16:25 ` [PATCH v2 0/7] of: setup dma parameters using dma-ranges and dma-coherent Thomas Petazzoni
2014-04-19 16:25 ` Thomas Petazzoni
2014-04-19 16:25 ` Thomas Petazzoni
2014-04-21 13:35 ` Santosh Shilimkar
2014-04-21 13:35 ` Santosh Shilimkar
2014-04-21 13:35 ` Santosh Shilimkar
2014-04-21 18:19 ` Thomas Petazzoni
2014-04-21 18:19 ` Thomas Petazzoni
2014-04-21 19:17 ` Santosh Shilimkar
2014-04-21 19:17 ` Santosh Shilimkar
2014-04-21 19:17 ` Santosh Shilimkar
2014-04-22 14:54 ` Catalin Marinas
2014-04-22 14:54 ` Catalin Marinas
2014-04-22 15:02 ` Arnd Bergmann
2014-04-22 15:02 ` Arnd Bergmann
2014-04-22 15:02 ` Arnd Bergmann
2014-04-22 15:15 ` Santosh Shilimkar
2014-04-22 15:15 ` Santosh Shilimkar
2014-04-22 15:15 ` Santosh Shilimkar
2014-04-22 15:25 ` Catalin Marinas [this message]
2014-04-22 15:25 ` Catalin Marinas
2014-04-22 15:25 ` Catalin Marinas
2014-04-22 15:30 ` Rob Herring
2014-04-22 15:30 ` Rob Herring
2014-04-22 15:30 ` Rob Herring
2014-04-22 16:01 ` Catalin Marinas
2014-04-22 16:01 ` Catalin Marinas
2014-04-22 16:01 ` Catalin Marinas
2014-04-21 14:37 ` Rob Herring
2014-04-21 14:37 ` Rob Herring
2014-04-21 14:37 ` Rob Herring
2014-04-21 15:13 ` Santosh Shilimkar
2014-04-21 15:13 ` Santosh Shilimkar
2014-04-21 15:35 ` Rob Herring
2014-04-21 15:35 ` Rob Herring
2014-04-21 15:35 ` Rob Herring
2014-04-21 15:36 ` Santosh Shilimkar
2014-04-21 15:36 ` Santosh Shilimkar
2014-04-21 15:36 ` Santosh Shilimkar
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=20140422152529.GE10224@arm.com \
--to=catalin.marinas@arm.com \
--cc=linux-arm-kernel@lists.infradead.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.