Skip to contentSkip to conversation

RDMA/InfiniBand merge plans for 2.6.30

DISCUSSION Roland Dreier 3 participants 4 messages Mar 25, 2009

lore
Share & tools
lore
#0 lore raw

Since 2.6.29 is out, I should update everyone on 2.6.30 merge plans. All the pending things that I'm aware of are listed below. As usual, if something isn't already in my tree and it isn't listed below, I probably missed it or dropped it unintentionally. Please remind me again.

Boilerplate:

As usual, when submitting a patch:

- Give a good changelog that explains what issue your patch addresses, how you address the issue, how serious the issue is, and any other information that would be useful to someone evaluating your patch or reading it years from now.

- Please make sure that you include a "Signed-off-by:" line, and put any extra junk that should not go into the final kernel log *after* the "---" line so that git tools strip it off automatically. Make the subject line be appropriate for inclusion in the kernel log as well once the leading "[PATCH ...]" stuff is stripped off. I waste a lot of time fixing patches by hand that could otherwise be spent doing something productive like watching youtube.

- Run your patch through checkpatch.pl so I don't have to nag you to fix trivial issues (or spend time fixing them myself).

- Read your patch over so I don't see a memory leak or deadlock as soon as I look at it.

- Build your patch with sparse checking ("C=2 CF=-D__CHECK_ENDIAN__") and make sure it doesn't introduce new warnings. (A big bonus in goodwill for sending patches that fix old warnings)

- Test your patch on a kernel with things like slab debugging and lockdep turned on.

And while you're waiting for me to get to your patch, I sure wouldn't mind if you read and commented on someone else's patch. None of this means you shouldn't remind me about pending patches, since I often lose track of things and drop them accidentally.

Core:

- Fix for sysfs lifetime problem that could allow userspace to call into a device driver after it's unloaded.

- Several fixes for the MAD handling layer.

ULPs:

- I have one IPoIB fix merge, and I have the nagging feeling that there are still IPoIB crash bugs that I lost track of.

HW specific:

- I have a few cxgb3 fixes merged already, and I'll merge the EEH patch after the net-next tree lands, just to avoid merge hassles.

- Many nes fixes.

- Support for port type (IB or Ethernet) autonegotiation for mlx4 devices.

Here are a few topics that I believe will not be ready in time for the 2.6.30 window and will need to wait for 2.6.31 at least:

- Jack's XRC patch set. I just need time to work with Jack and clean up the code.

Here all the patches I already have in my for-next branch:

Chien Tung (3):
RDMA/nes: Update copyright to new legal entity and year RDMA/nes: Report correct vendor_id and vendor_part_id RDMA/nes: Fix tmp_addr compilation warning

Don Wood (3):
RDMA/nes: Account for freed PBL after HW operation RDMA/nes: Inform hardware that asynchronous event has been handled RDMA/nes: Improve use of PBLs

Faisal Latif (2):
RDMA/nes: Remove LLTX
RDMA/nes: Handle MPA Reject message properly

Harvey Harrison (1):
IB: Remove __constant_{endian} uses

Jack Morgenstein (3):
IPoIB: In unicast_arp_send(), only free newly-created paths
mlx4_core: Add device IDs for MT25458 10GigE devices IB/sa_query: Fix AH leak due to update_sm_ah() race

Moni Shoua (2):
IB/mlx4: Fix dispatch of IB_EVENT_LID_CHANGE event IB/mthca: Fix dispatch of IB_EVENT_LID_CHANGE event

Or Gerlitz (1):
IB/iser: Remove hard setting of path MTU

Ralph Campbell (3):
IB/mad: Fix null pointer dereference in local_completions() IB/mad: initialize mad_agent_priv before putting on lists IB/mad: Fix ib_post_send_mad() returning 0 with no generate send comp

Ramachandra K (1):
IB/mad: Fix RMPP header RRespTime manipulation

Roland Dreier (5):
IB/ipath: Fix memory leak in init_shadow_tids() error path
IB/ipath: Really run work in ipath_release_user_pages_on_close()
IB: Remove sysfs files before unregistering device
IB: Remove useless ibdev_is_alive() tests from sysfs code
Merge branches 'cxgb3', 'endian', 'ipath', 'ipoib', 'iser', 'mad', 'misc', 'mlx4', 'mthca', 'nes' and 'sysfs' into for-next

Steve Wise (3):
RDMA/cxgb3: sgl/pbl offset calculation needs 64 bits RDMA/cxgb3: Connection termination fixes RDMA/cxgb3: Remove modulo math from build_rdma_recv()

Yevgeny Petrilin (3):
mlx4_core: Don't perform SET_PORT command for Ethernet ports
mlx4_core: Add link type autosensing
IB/mlx4: Unregister IB device prior to CLOSE PORT command

#1 lore raw
Quoted 4 lines · Roland Dreier

Since 2.6.29 is out, I should update everyone on 2.6.30 merge plans. All the pending things that I'm aware of are listed below. As usual, if something isn't already in my tree and it isn't listed below, I probably missed it or dropped it unintentionally. Please remind me again.

There are some patches in the ofed release by Eli and me to support weak ordering on DMA ("ib_core: Use weak ordering for data registered memory"). When they got merged into OFED, I assumed they would find their way into mainline along with most of the other patches in there.

Did they get forgotten about, or did I misunderstand the process?

	Arnd <><
#2 lore raw
Quoted 6 lines

There are some patches in the ofed release by Eli and me to support weak ordering on DMA ("ib_core: Use weak ordering for data registered memory"). When they got merged into OFED, I assumed they would find their way into mainline along with most of the other patches in there.

Did they get forgotten about, or did I misunderstand the process?

Oh yeah, forgot about that weak ordering. I still want to have a think about that (I don't like having a module parameter for it but maybe there's not better way)

However in general getting into OFED is pretty unrelated to getting upstream...

- R.

#3 lore raw
In reply to Roland Dreier
Quoted 6 lines · Roland Dreier

Here are a few topics that I believe will not be ready in time for the 2.6.30 window and will need to wait for 2.6.31 at least:

- Jack's XRC patch set. I just need time to work with Jack and clean up the code.

Any chance we can make it for 2.6.30?

Tziporet