From: Sebastian Ott <sebott@linux.ibm.com>
To: Halil Pasic <pasic@linux.ibm.com>
Cc: kvm@vger.kernel.org, linux-s390@vger.kernel.org,
Cornelia Huck <cohuck@redhat.com>,
Martin Schwidefsky <schwidefsky@de.ibm.com>,
virtualization@lists.linux-foundation.org,
Christian Borntraeger <borntraeger@de.ibm.com>,
Viktor Mihajlovski <mihajlov@linux.ibm.com>,
Vasily Gorbik <gor@linux.ibm.com>,
Janosch Frank <frankja@linux.ibm.com>,
Claudio Imbrenda <imbrenda@linux.ibm.com>,
Farhan Ali <alifm@linux.ibm.com>,
Eric Farman <farman@linux.ibm.com>
Subject: Re: [RFC PATCH 04/12] s390/cio: introduce cio DMA pool
Date: Thu, 11 Apr 2019 20:25:01 +0200 (CEST) [thread overview]
Message-ID: <alpine.LFD.2.21.1904111955520.2283@schleppi> (raw)
In-Reply-To: <20190404231622.52531-5-pasic@linux.ibm.com>
On Fri, 5 Apr 2019, Halil Pasic wrote:
> To support protected virtualization cio will need to make sure the
> memory used for communication with the hypervisor is DMA memory.
>
> Let us introduce a DMA pool to cio that will help us in allocating
> deallocating those chunks of memory.
>
> We use a gen_pool backed with DMA pages to avoid each allocation
> effectively wasting a page, as we typically allocate much less
> than PAGE_SIZE.
I don't think we should use this global DMA pool. I guess it's OK for
stuff like airq (where we don't have a struct device at hand) but for
CCW we should use the device we have. Yes, this way we waste some memory
but all dma memory a device uses should fit in a page - so the wastage
is not too much.
Sebastian
WARNING: multiple messages have this Message-ID (diff)
From: Sebastian Ott <sebott@linux.ibm.com>
To: Halil Pasic <pasic@linux.ibm.com>
Cc: Farhan Ali <alifm@linux.ibm.com>,
linux-s390@vger.kernel.org, Eric Farman <farman@linux.ibm.com>,
Claudio Imbrenda <imbrenda@linux.ibm.com>,
Vasily Gorbik <gor@linux.ibm.com>,
kvm@vger.kernel.org, Cornelia Huck <cohuck@redhat.com>,
virtualization@lists.linux-foundation.org,
Martin Schwidefsky <schwidefsky@de.ibm.com>,
Viktor Mihajlovski <mihajlov@linux.ibm.com>,
Janosch Frank <frankja@linux.ibm.com>
Subject: Re: [RFC PATCH 04/12] s390/cio: introduce cio DMA pool
Date: Thu, 11 Apr 2019 20:25:01 +0200 (CEST) [thread overview]
Message-ID: <alpine.LFD.2.21.1904111955520.2283@schleppi> (raw)
In-Reply-To: <20190404231622.52531-5-pasic@linux.ibm.com>
On Fri, 5 Apr 2019, Halil Pasic wrote:
> To support protected virtualization cio will need to make sure the
> memory used for communication with the hypervisor is DMA memory.
>
> Let us introduce a DMA pool to cio that will help us in allocating
> deallocating those chunks of memory.
>
> We use a gen_pool backed with DMA pages to avoid each allocation
> effectively wasting a page, as we typically allocate much less
> than PAGE_SIZE.
I don't think we should use this global DMA pool. I guess it's OK for
stuff like airq (where we don't have a struct device at hand) but for
CCW we should use the device we have. Yes, this way we waste some memory
but all dma memory a device uses should fit in a page - so the wastage
is not too much.
Sebastian
next prev parent reply other threads:[~2019-04-11 18:25 UTC|newest]
Thread overview: 87+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-04 23:16 [RFC PATCH 00/12] s390: virtio: support protected virtualization Halil Pasic
2019-04-04 23:16 ` [RFC PATCH 01/12] virtio/s390: use vring_create_virtqueue Halil Pasic
2019-04-08 11:01 ` Cornelia Huck
2019-04-08 11:01 ` Cornelia Huck
2019-04-08 12:37 ` Michael S. Tsirkin
2019-04-08 12:37 ` Michael S. Tsirkin
2019-04-08 13:20 ` Halil Pasic
2019-04-04 23:16 ` [RFC PATCH 02/12] virtio/s390: DMA support for virtio-ccw Halil Pasic
2019-04-09 9:57 ` Cornelia Huck
2019-04-09 9:57 ` Cornelia Huck
2019-04-09 11:29 ` Halil Pasic
2019-04-09 13:01 ` Cornelia Huck
2019-04-09 13:01 ` Cornelia Huck
2019-04-09 13:23 ` Halil Pasic
2019-04-09 15:47 ` Cornelia Huck
2019-04-09 15:47 ` Cornelia Huck
2019-04-04 23:16 ` [RFC PATCH 03/12] s390/mm: force swiotlb for protected virtualization Halil Pasic
2019-04-09 10:16 ` Cornelia Huck
2019-04-09 10:16 ` Cornelia Huck
2019-04-09 10:54 ` Halil Pasic
2019-04-09 17:18 ` Cornelia Huck
2019-04-09 17:18 ` Cornelia Huck
2019-04-09 12:22 ` Christoph Hellwig
2019-04-09 12:22 ` Christoph Hellwig
2019-04-09 12:39 ` Halil Pasic
2019-04-04 23:16 ` [RFC PATCH 04/12] s390/cio: introduce cio DMA pool Halil Pasic
2019-04-09 10:44 ` Cornelia Huck
2019-04-09 10:44 ` Cornelia Huck
2019-04-09 12:11 ` Halil Pasic
2019-04-09 17:14 ` Cornelia Huck
2019-04-09 17:14 ` Cornelia Huck
2019-04-10 15:31 ` Halil Pasic
2019-04-10 16:07 ` Cornelia Huck
2019-04-10 16:07 ` Cornelia Huck
2019-04-10 16:52 ` Halil Pasic
2019-04-11 18:25 ` Sebastian Ott [this message]
2019-04-11 18:25 ` Sebastian Ott
2019-04-12 11:20 ` Halil Pasic
2019-04-12 12:12 ` Sebastian Ott
2019-04-12 12:12 ` Sebastian Ott
2019-04-12 15:30 ` Halil Pasic
2019-04-16 12:50 ` Sebastian Ott
2019-04-16 12:50 ` Sebastian Ott
2019-04-16 13:31 ` Halil Pasic
2019-04-04 23:16 ` [RFC PATCH 05/12] s390/cio: add protected virtualization support to cio Halil Pasic
2019-04-09 17:55 ` Cornelia Huck
2019-04-09 17:55 ` Cornelia Huck
2019-04-10 0:10 ` Halil Pasic
2019-04-10 8:25 ` Cornelia Huck
2019-04-10 8:25 ` Cornelia Huck
2019-04-10 13:02 ` Halil Pasic
2019-04-10 16:16 ` Cornelia Huck
2019-04-10 16:16 ` Cornelia Huck
2019-04-11 14:15 ` Sebastian Ott
2019-04-11 14:15 ` Sebastian Ott
2019-04-12 11:29 ` Halil Pasic
2019-04-04 23:16 ` [RFC PATCH 06/12] s390/airq: use DMA memory for adapter interrupts Halil Pasic
2019-04-04 23:16 ` [RFC PATCH 07/12] virtio/s390: use DMA memory for ccw I/O Halil Pasic
2019-04-10 8:42 ` Cornelia Huck
2019-04-10 8:42 ` Cornelia Huck
2019-04-10 14:42 ` Halil Pasic
2019-04-10 16:21 ` Cornelia Huck
2019-04-10 16:21 ` Cornelia Huck
2019-04-04 23:16 ` [RFC PATCH 08/12] virtio/s390: add indirection to indicators access Halil Pasic
2019-04-04 23:16 ` [RFC PATCH 09/12] virtio/s390: use DMA memory for notifiers Halil Pasic
2019-04-04 23:16 ` [RFC PATCH 10/12] virtio/s390: consolidate DMA allocations Halil Pasic
2019-04-10 8:46 ` Cornelia Huck
2019-04-10 8:46 ` Cornelia Huck
2019-04-10 15:12 ` Halil Pasic
2019-04-10 16:36 ` Cornelia Huck
2019-04-10 16:36 ` Cornelia Huck
2019-04-10 17:48 ` Halil Pasic
2019-04-11 9:24 ` Cornelia Huck
2019-04-11 9:24 ` Cornelia Huck
2019-04-11 10:10 ` Halil Pasic
2019-04-04 23:16 ` [RFC PATCH 11/12] virtio/s390: use the cio DMA pool Halil Pasic
2019-04-04 23:16 ` [RFC PATCH 12/12] virtio/s390: make airq summary indicators DMA Halil Pasic
2019-04-10 9:20 ` [RFC PATCH 00/12] s390: virtio: support protected virtualization Cornelia Huck
2019-04-10 9:20 ` Cornelia Huck
2019-04-10 15:57 ` Halil Pasic
2019-04-10 16:24 ` Cornelia Huck
2019-04-10 16:24 ` Cornelia Huck
2019-04-12 13:47 ` David Hildenbrand
2019-04-12 13:47 ` David Hildenbrand
2019-04-16 11:10 ` Halil Pasic
2019-04-16 11:50 ` David Hildenbrand
2019-04-16 11:50 ` David Hildenbrand
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=alpine.LFD.2.21.1904111955520.2283@schleppi \
--to=sebott@linux.ibm.com \
--cc=alifm@linux.ibm.com \
--cc=borntraeger@de.ibm.com \
--cc=cohuck@redhat.com \
--cc=farman@linux.ibm.com \
--cc=frankja@linux.ibm.com \
--cc=gor@linux.ibm.com \
--cc=imbrenda@linux.ibm.com \
--cc=kvm@vger.kernel.org \
--cc=linux-s390@vger.kernel.org \
--cc=mihajlov@linux.ibm.com \
--cc=pasic@linux.ibm.com \
--cc=schwidefsky@de.ibm.com \
--cc=virtualization@lists.linux-foundation.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.