From: Alexey Kardashevskiy <aik@ozlabs.ru>
To: Christoph Hellwig <hch@lst.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Joerg Roedel <joro@8bytes.org>,
linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org,
iommu@lists.linux-foundation.org,
"Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>,
Robin Murphy <robin.murphy@arm.com>,
Lu Baolu <baolu.lu@linux.intel.com>
Subject: Re: [PATCH 1/2] dma-mapping: add a dma_ops_bypass flag to struct device
Date: Tue, 24 Mar 2020 15:55:33 +1100 [thread overview]
Message-ID: <f02f9a98-532b-d457-bbb5-21d874b43e8c@ozlabs.ru> (raw)
In-Reply-To: <d4bf6058-aa77-d0bc-8196-f4c27fb21b74@ozlabs.ru>
On 24/03/2020 14:37, Alexey Kardashevskiy wrote:
>
>
> On 24/03/2020 04:20, Christoph Hellwig wrote:
>> On Mon, Mar 23, 2020 at 07:58:01PM +1100, Alexey Kardashevskiy wrote:
>>>>> 0x100.0000.0000 .. 0x101.0000.0000
>>>>>
>>>>> 2x4G, each is 1TB aligned. And we can map directly only the first 4GB
>>>>> (because of the maximum IOMMU table size) but not the other. And 1:1 on
>>>>> that "pseries" is done with offset=0x0800.0000.0000.0000.
>>>>>
>>>>> So we want to check every bus address against dev->bus_dma_limit, not
>>>>> dev->coherent_dma_mask. In the example above I'd set bus_dma_limit to
>>>>> 0x0800.0001.0000.0000 and 1:1 mapping for the second 4GB would not be
>>>>> tried. Does this sound reasonable? Thanks,
>>>>
>>>> bus_dma_limit is just another limiting factor applied on top of
>>>> coherent_dma_mask or dma_mask respectively.
>>>
>>> This is not enough for the task: in my example, I'd set bus limit to
>>> 0x0800.0001.0000.0000 but this would disable bypass for all RAM
>>> addresses - the first and the second 4GB blocks.
>>
>> So what about something like the version here:
>>
>> http://git.infradead.org/users/hch/misc.git/shortlog/refs/heads/dma-bypass.3
>
>
> dma_alloc_direct() and dma_map_direct() do the same thing now which is
> good, did I miss anything else?
>
> This lets us disable bypass automatically if this weird memory appears
> in the system but does not let us have 1:1 after that even for normal
> RAM. Thanks,
Ah no, does not help much, simple setting dma_ops_bypass will though.
But eventually, in this function:
static inline bool dma_map_direct(struct device *dev,
const struct dma_map_ops *ops)
{
if (likely(!ops))
return true;
if (!dev->dma_ops_bypass)
return false;
return min_not_zero(*dev->dma_mask, dev->bus_dma_limit) >=
dma_direct_get_required_mask(dev);
}
we rather want it to take a dma handle and a size, and add
if (dev->bus_dma_limit)
return dev->bus_dma_limit > dma_handle + size;
where dma_handle=phys_to_dma(dev, phys) (I am not doing it here as unmap
needs the same test and it does not receive phys as a parameter).
--
Alexey
next prev parent reply other threads:[~2020-03-24 4:57 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-20 14:16 generic DMA bypass flag v2 Christoph Hellwig
2020-03-20 14:16 ` [PATCH 1/2] dma-mapping: add a dma_ops_bypass flag to struct device Christoph Hellwig
2020-03-20 15:02 ` Greg Kroah-Hartman
2020-03-23 1:28 ` Alexey Kardashevskiy
2020-03-23 8:37 ` Christoph Hellwig
2020-03-23 8:50 ` Christoph Hellwig
2020-03-23 15:37 ` Aneesh Kumar K.V
2020-03-23 17:22 ` Christoph Hellwig
2020-03-24 3:05 ` Alexey Kardashevskiy
2020-03-24 6:30 ` Aneesh Kumar K.V
2020-03-24 7:55 ` Christoph Hellwig
2020-03-24 7:54 ` Christoph Hellwig
2020-03-25 4:51 ` Alexey Kardashevskiy
2020-03-25 8:37 ` Christoph Hellwig
2020-03-26 1:26 ` Alexey Kardashevskiy
2020-04-03 8:38 ` Alexey Kardashevskiy
2020-04-06 11:50 ` Christoph Hellwig
2020-04-06 13:25 ` Alexey Kardashevskiy
2020-04-06 17:17 ` Christoph Hellwig
2020-04-07 10:12 ` Alexey Kardashevskiy
2020-04-14 6:21 ` Alexey Kardashevskiy
2020-04-14 6:30 ` Christoph Hellwig
2020-03-23 8:58 ` Alexey Kardashevskiy
2020-03-23 17:20 ` Christoph Hellwig
2020-03-24 3:37 ` Alexey Kardashevskiy
2020-03-24 4:55 ` Alexey Kardashevskiy [this message]
2020-03-24 7:52 ` Christoph Hellwig
2020-03-23 12:14 ` Robin Murphy
2020-03-23 12:55 ` Christoph Hellwig
2020-03-20 14:16 ` [PATCH 2/2] powerpc: use the generic dma_ops_bypass mode Christoph Hellwig
-- strict thread matches above, loose matches on Subject: below --
2020-03-24 9:39 [PATCH 1/2] dma-mapping: add a dma_ops_bypass flag to, struct device Christian Zigotzky
2019-11-13 13:37 generic DMA bypass flag Christoph Hellwig
2019-11-13 13:37 ` [PATCH 1/2] dma-mapping: add a dma_ops_bypass flag to struct device Christoph Hellwig
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=f02f9a98-532b-d457-bbb5-21d874b43e8c@ozlabs.ru \
--to=aik@ozlabs.ru \
--cc=aneesh.kumar@linux.ibm.com \
--cc=baolu.lu@linux.intel.com \
--cc=gregkh@linuxfoundation.org \
--cc=hch@lst.de \
--cc=iommu@lists.linux-foundation.org \
--cc=joro@8bytes.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=robin.murphy@arm.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).