<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-dev/net/rds/ib.c, branch linus/master</title>
<subtitle>Linux kernel development work - see feature branches</subtitle>
<id>https://git.zx2c4.com/linux-dev/atom/net/rds/ib.c?h=linus%2Fmaster</id>
<link rel='self' href='https://git.zx2c4.com/linux-dev/atom/net/rds/ib.c?h=linus%2Fmaster'/>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/'/>
<updated>2022-04-06T18:02:13Z</updated>
<entry>
<title>RDMA: Split kernel-only global device caps from uverbs device caps</title>
<updated>2022-04-06T18:02:13Z</updated>
<author>
<name>Jason Gunthorpe</name>
<email>jgg@nvidia.com</email>
</author>
<published>2022-04-04T15:26:42Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=e945c653c8e972d1b81a88e474d79f801b60213a'/>
<id>urn:sha1:e945c653c8e972d1b81a88e474d79f801b60213a</id>
<content type='text'>
Split out flags from ib_device::device_cap_flags that are only used
internally to the kernel into kernel_cap_flags that is not part of the
uapi. This limits the device_cap_flags to being the same bitmap that will
be copied to userspace.

This cleanly splits out the uverbs flags from the kernel flags to avoid
confusion in the flags bitmap.

Add some short comments describing which each of the kernel flags is
connected to. Remove unused kernel flags.

Link: https://lore.kernel.org/r/0-v2-22c19e565eef+139a-kern_caps_jgg@nvidia.com
Reviewed-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Max Gurtovoy &lt;mgurtovoy@nvidia.com&gt;
Signed-off-by: Jason Gunthorpe &lt;jgg@nvidia.com&gt;
</content>
</entry>
<entry>
<title>rds: stop using dmapool</title>
<updated>2020-11-17T19:22:06Z</updated>
<author>
<name>Christoph Hellwig</name>
<email>hch@lst.de</email>
</author>
<published>2020-11-06T18:19:36Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=42f2611cc1738b201701e717246e11e86bef4e1e'/>
<id>urn:sha1:42f2611cc1738b201701e717246e11e86bef4e1e</id>
<content type='text'>
RDMA ULPs should only perform DMA through the ib_dma_* API instead of
using the hidden dma_device directly.  In addition using the dma coherent
API family that dmapool is a part of can be very ineffcient on plaforms
that are not DMA coherent.  Switch to use slab allocations and the
ib_dma_* APIs instead.

Link: https://lore.kernel.org/r/20201106181941.1878556-6-hch@lst.de
Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;
Acked-by: Santosh Shilimkar &lt;santosh.shilimkar@oracle.com&gt;
Signed-off-by: Jason Gunthorpe &lt;jgg@nvidia.com&gt;
</content>
</entry>
<entry>
<title>RDMA: Remove 'max_fmr'</title>
<updated>2020-06-02T23:32:54Z</updated>
<author>
<name>Jason Gunthorpe</name>
<email>jgg@mellanox.com</email>
</author>
<published>2020-05-28T19:45:54Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=649392bf75a423287a9c4936b341677f12e8cf0b'/>
<id>urn:sha1:649392bf75a423287a9c4936b341677f12e8cf0b</id>
<content type='text'>
Now that FMR support is gone, this attribute can be deleted from all
places.

Link: https://lore.kernel.org/r/12-v3-f58e6669d5d3+2cf-fmr_removal_jgg@mellanox.com
Reviewed-by: Max Gurtovoy &lt;maxg@mellanox.com&gt;
Reviewed-by: Bernard Metzler &lt;bmt@zurich.ibm.com&gt;
Signed-off-by: Jason Gunthorpe &lt;jgg@mellanox.com&gt;
</content>
</entry>
<entry>
<title>RDMA/rds: Remove FMR support for memory registration</title>
<updated>2020-06-02T23:32:53Z</updated>
<author>
<name>Max Gurtovoy</name>
<email>maxg@mellanox.com</email>
</author>
<published>2020-05-28T19:45:45Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=07549ee21ce5247143ffb069bf838025d86b908c'/>
<id>urn:sha1:07549ee21ce5247143ffb069bf838025d86b908c</id>
<content type='text'>
Use FRWR method for memory registration by default and remove the ancient
and unsafe FMR method.

Link: https://lore.kernel.org/r/3-v3-f58e6669d5d3+2cf-fmr_removal_jgg@mellanox.com
Signed-off-by: Max Gurtovoy &lt;maxg@mellanox.com&gt;
Signed-off-by: Jason Gunthorpe &lt;jgg@mellanox.com&gt;
</content>
</entry>
<entry>
<title>RDMA: Allow ib_client's to fail when add() is called</title>
<updated>2020-05-06T14:57:33Z</updated>
<author>
<name>Jason Gunthorpe</name>
<email>jgg@mellanox.com</email>
</author>
<published>2020-04-21T17:24:40Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=11a0ae4c4bff9b2a471b54dbe910fc0f60e58e62'/>
<id>urn:sha1:11a0ae4c4bff9b2a471b54dbe910fc0f60e58e62</id>
<content type='text'>
When a client is added it isn't allowed to fail, but all the client's have
various failure paths within their add routines.

This creates the very fringe condition where the client was added, failed
during add and didn't set the client_data. The core code will then still
call other client_data centric ops like remove(), rename(), get_nl_info(),
and get_net_dev_by_params() with NULL client_data - which is confusing and
unexpected.

If the add() callback fails, then do not call any more client ops for the
device, even remove.

Remove all the now redundant checks for NULL client_data in ops callbacks.

Update all the add() callbacks to return error codes
appropriately. EOPNOTSUPP is used for cases where the ULP does not support
the ib_device - eg because it only works with IB.

Link: https://lore.kernel.org/r/20200421172440.387069-1-leon@kernel.org
Signed-off-by: Leon Romanovsky &lt;leonro@mellanox.com&gt;
Acked-by: Ursula Braun &lt;ubraun@linux.ibm.com&gt;
Signed-off-by: Jason Gunthorpe &lt;jgg@mellanox.com&gt;
</content>
</entry>
<entry>
<title>net/rds: Handle ODP mr registration/unregistration</title>
<updated>2020-01-18T09:48:19Z</updated>
<author>
<name>Hans Westgaard Ry</name>
<email>hans.westgaard.ry@oracle.com</email>
</author>
<published>2020-01-15T12:43:39Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=2eafa1746f17872483d1033b0116ec71435ea19d'/>
<id>urn:sha1:2eafa1746f17872483d1033b0116ec71435ea19d</id>
<content type='text'>
On-Demand-Paging MRs are registered using ib_reg_user_mr and
unregistered with ib_dereg_mr.

Signed-off-by: Hans Westgaard Ry &lt;hans.westgaard.ry@oracle.com&gt;
Acked-by: Santosh Shilimkar &lt;santosh.shilimkar@oracle.com&gt;
Signed-off-by: Leon Romanovsky &lt;leonro@mellanox.com&gt;
</content>
</entry>
<entry>
<title>net/rds: Remove unnecessary null check</title>
<updated>2019-10-17T19:23:03Z</updated>
<author>
<name>YueHaibing</name>
<email>yuehaibing@huawei.com</email>
</author>
<published>2019-10-15T11:47:36Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=ce753e66dcc37e19572a87f70585ec6537dede81'/>
<id>urn:sha1:ce753e66dcc37e19572a87f70585ec6537dede81</id>
<content type='text'>
Null check before dma_pool_destroy is redundant, so remove it.
This is detected by coccinelle.

Signed-off-by: YueHaibing &lt;yuehaibing@huawei.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>net/rds: Add missing include file</title>
<updated>2019-10-06T16:34:10Z</updated>
<author>
<name>YueHaibing</name>
<email>yuehaibing@huawei.com</email>
</author>
<published>2019-10-06T07:08:32Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=d131c5bb60123f29ed15dd2f829b6644c2deec87'/>
<id>urn:sha1:d131c5bb60123f29ed15dd2f829b6644c2deec87</id>
<content type='text'>
Fix build error:

net/rds/ib_cm.c: In function rds_dma_hdrs_alloc:
net/rds/ib_cm.c:475:13: error: implicit declaration of function dma_pool_zalloc; did you mean mempool_alloc? [-Werror=implicit-function-declaration]
   hdrs[i] = dma_pool_zalloc(pool, GFP_KERNEL, &amp;hdr_daddrs[i]);
             ^~~~~~~~~~~~~~~
             mempool_alloc

net/rds/ib.c: In function rds_ib_dev_free:
net/rds/ib.c:111:3: error: implicit declaration of function dma_pool_destroy; did you mean mempool_destroy? [-Werror=implicit-function-declaration]
   dma_pool_destroy(rds_ibdev-&gt;rid_hdrs_pool);
   ^~~~~~~~~~~~~~~~
   mempool_destroy

Reported-by: Hulk Robot &lt;hulkci@huawei.com&gt;
Fixes: 9b17f5884be4 ("net/rds: Use DMA memory pool allocation for rds_header")
Signed-off-by: YueHaibing &lt;yuehaibing@huawei.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net</title>
<updated>2019-10-05T20:37:23Z</updated>
<author>
<name>David S. Miller</name>
<email>davem@davemloft.net</email>
</author>
<published>2019-10-05T20:37:23Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=6f4c930e02355664d89c976eccea5d999a90de16'/>
<id>urn:sha1:6f4c930e02355664d89c976eccea5d999a90de16</id>
<content type='text'>
</content>
</entry>
<entry>
<title>net/rds: Use DMA memory pool allocation for rds_header</title>
<updated>2019-10-03T19:11:08Z</updated>
<author>
<name>Ka-Cheong Poon</name>
<email>ka-cheong.poon@oracle.com</email>
</author>
<published>2019-10-03T04:11:08Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=9b17f5884be4484e4d9090a9dccf17e763e0589b'/>
<id>urn:sha1:9b17f5884be4484e4d9090a9dccf17e763e0589b</id>
<content type='text'>
Currently, RDS calls ib_dma_alloc_coherent() to allocate a large piece
of contiguous DMA coherent memory to store struct rds_header for
sending/receiving packets.  The memory allocated is then partitioned
into struct rds_header.  This is not necessary and can be costly at
times when memory is fragmented.  Instead, RDS should use the DMA
memory pool interface to handle this.  The DMA addresses of the pre-
allocated headers are stored in an array.  At send/receive ring
initialization and refill time, this arrary is de-referenced to get
the DMA addresses.  This array is not accessed at send/receive packet
processing.

Suggested-by: Håkon Bugge &lt;haakon.bugge@oracle.com&gt;
Signed-off-by: Ka-Cheong Poon &lt;ka-cheong.poon@oracle.com&gt;
Acked-by: Santosh Shilimkar &lt;santosh.shilimkar@oracle.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
</feed>
