[LU-6215] Sync Lustre external tree with lustre linux kernel client

• Status: Resolved
• Project: Lustre
• Affects Version/s: Lustre 2.8.0
• Type: Improvement
• Priority: Major
• Reporter: James A Simmons
• Assignee: Oleg Drokin
• Resolution: Duplicate
• Labels: patch, upstream
• Epic/Theme: upstream

Description

This is for book keeping all the work that is being done to bring the Intel branch into sync with what is upstream in the linux kernel. Also at the same ensure important changes upstream get merged back into the Intel branch.

Issue Links

• is related to: LU-6285 Assert fails in staging client module...
• is related to: LU-4416 support for 3.12 linux kernel
• is related to: LU-6548 build errors building against recent ...
• is related to: LU-5030 "lctl {get,set}_param" should also ch...
• is related to: LU-6746 move IT_* definitions to lustre_idl.h
• is related to: LU-7624 fld_proc_hash_seq_write accesses user...
• is related to: LU-9472 FastReg (MLX5) support breaks when ma...
• is related to: LU-4451 Kernel Oops with NFS reexport using m...
• is related to: LU-6490 builds on 3.12 fail in gss
• is related to: LU-6547 build errors building against recent ...
• is related to: LU-7623 Clean up lustre user/kernel pointer t...
• is related to: LU-8025 ll_direct_io code introduced in LU-62...
• is related to: LU-8435 LBUG (osc_cache.c:1290:osc_completion...
• is related to: LU-8540 ib_map_mr_sg() with five arguments
• is related to: LU-4530 Mainline kernel client (3.12-3.14): l...
• is related to: LU-5478 Get rid of typedefs
• is related to: LU-5577 comparison between signed and unsigne...
• is related to: LU-5829 too many EXPORT_SYMBOL in code
• is related to: LU-6303 comparison between signed and unsigne...
• is related to: LU-6428 Lustre client Compilation failed buil...
• is related to: LU-7366 Lustre fails to compile on Ubuntu 15.04
• is related to: LU-8773 fiemap extent uses u64 type, not loff...
• is related to: LU-8835 More 64 bit divides without do_div.
• is related to: LU-9003 Linux kernel 4.8 adds user_namespace ...
• is related to: LU-5396 annotations for sparse static analyzer
• is related to: LU-5710 correcting typos and grammar
• is related to: LU-7311 Linux Kernel 3.18 LTS support
• is related to: LU-9026 Adapt to the removal of ib_get_dma_mr()
• is related to: LU-5628 Dealing with kernels that have lustre...
• is related to: LU-6209 remove old LNDs - ralnd, mxlnd
• is related to: LU-8560 Support for linux 4.6 kernels
• is related to: LU-5609 support for 3.17 kernel
• is related to: LU-6030 cleanup for ldiskfs patches
• is related to: LU-6260 more support for 3.16 linux kernel
• is related to: LU-8056 Support for linux 4.5 kernels
• is related to: LU-8843 comparison between signed and unsigne...
• is related to: LU-8844 Delete llite lloop with fire!!!
• is related to: LU-9456 Change socklnd calls from sock_create...
• is related to: LU-6302 annotations for sparse static analyzer
• is related to: LU-6141 Merge ldiskfs patches into upstream e...
• is related to: LU-8919 Move to new kernel documentation API

Comments

[Doug Oucharek (Inactive)]

So is the work under this ticket for both upstream and downstream?

[Oleg Drokin]

It needs to be in-sync mostly.
E.g. the typedefs needed to be done in both so that the code is in sync.
Same in almost every other item.

[Gerrit Updater]

Andreas Dilger ([e-mail]) uploaded a new patch: http://review.whamcloud.com/13956
Subject: LU-6215 osc: use list_for_each_entry_safe() when delete items
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: be592a496de6b86a75868520b89f4776936e123f

[Andreas Dilger]

Note that I've updated https://wiki.hpdd.intel.com/display/PUB/Commit+Comments to describe requirements for patches ported to/from the upstream kernel, so that we keep them consistent and able to trace patches between Lustre and the kernel.

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/13956/
Subject: LU-6215 osc: use list_for_each_entry_safe() when delete items
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: eaba382df06cd6f570b16c5f759f612b057e5324

[Gerrit Updater]

Andreas Dilger ([e-mail]) uploaded a new patch: http://review.whamcloud.com/14013
Subject: LU-6215 llite: __GFP_NOWARN doesn't need to be conditional
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: e7875908d0735eca128dfd2635ef1d5fd56c7c90

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/14598
Subject: LU-6215 kernel: use f_path for inode dentry
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: c32e140e1b8e2623f48994bb5def9a139e4910cf

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/14662
Subject: LU-6215 libcfs: shrinker code needs atomic.h
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 2b63acbaee251fd50df20ccadb374c5df235d381

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/14664
Subject: LU-6215 llite: cleanup inode_dio_XXX functions
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 52c1524628eaa53bde552ddf67727299ce113c7f

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/14669
Subject: LU-6215 client: handle dirty page cleanup api changes
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 12fb3b605daf699d5327125a902f2cd1f574c53f

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/14681
Subject: LU-6215 llite: filldir_t api changed for 3.18 kernels
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 9e11765b0b39c124471ea9b0b6a864320fb854b5

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/14683
Subject: LU-6215 llite: handle removal of ki_nbytes in struct kiocb
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 7794c47b118d292cb75445d148d26da89fdf46f8

[James A Simmons]

As you can see I pushed several patches. I think it will cover Ubuntu 15 support but I haven't got a setup yet to try it out. Still more patches are needed to bring us in line with greg's lustre-staging tree but we are almost there.

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/14662/
Subject: LU-6215 libcfs: shrinker code needs atomic.h
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: d1b9bfc290dfafb3f7aa1b4ac92a6c01cd1f92a7

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/14664/
Subject: LU-6215 llite: cleanup inode_dio_XXX functions
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: dd2a3f8fa0e95c707121855c6d5acf01b2e1c855

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/14598/
Subject: LU-6215 kernel: use f_path for inode dentry
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 4a8f90cfdf8868d25f14b02214e51e7423d70160

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/14669/
Subject: LU-6215 client: handle dirty page cleanup api changes
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: bcc52fdccc92c45f657d637eaf8403decc6d9877

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/15165
Subject: LU-6215 llite: handle new_sync_[read|write] removal
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 7258a537c1d880559dcb6faecb9d4078a4694aea

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/15182
Subject: LU-6215 llite: handle backing_dev_info removal
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 96b1d7838b2acbb23089f4768603f97c095fb196

[James A Simmons]

Last patch has been pushed that is needed to allow master testing against the linux staging tree.

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/14683/
Subject: LU-6215 llite: handle removal of ki_nbytes in struct kiocb
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 9d3a33b176a1b353603f3a204ae1e818c05226c5

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/14681/
Subject: LU-6215 llite: filldir_t api changed for 3.18 kernels
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 7e1758e5621df902c5942d100a75cebde8df0327

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/15165/
Subject: LU-6215 llite: handle new_sync_[read|write] removal
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 0c080f1e8791b6b1aebba9d2165235122697db68

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/15182/
Subject: LU-6215 llite: handle backing_dev_info removal
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 7e0e87e6f2dbd04e6fb94123f453ad8f2806dc8b

[James A Simmons]

With the latest patch landing masters client support is up to an 4.1 linux kernel now.

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/16049
Subject: LU-6215 libcfs: handle topology_thread_cpumask rename
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 6475eae1e1a2c3fa1c12ff7e7a0af4cb07e434da

[James A Simmons]

Linux kernel version 4.2 will soon be release so its time for updates.

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/16069
Subject: LU-6215 client: cancel_dirty_page revival
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 8328e5f366382356b7c2b6a2c59736e13ddd1e5e

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/16118
Subject: LU-6215 o2iblnd: handle new struct ib_cq_init_attr
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 0c54d62a9c5e5db1ce7466d3f0e5c8989498cdc2

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/16119
Subject: LU-6215 llite: file->f_dentry no longer exist
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 4da3cd8157bcf92429db7b792b9a9ebcc02a9d69

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16049/
Subject: LU-6215 libcfs: handle topology_thread_cpumask rename
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: aa637a7be1dd34840fc512365e312857c4b0e756

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16069/
Subject: LU-6215 client: cancel_dirty_page revival
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 014e2c56ba8ffb30cc50a724a402f72e38418e05

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16119/
Subject: LU-6215 llite: file->f_dentry no longer exist
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 5da417890577f7c173906bc19246ea302a9c83c4

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16118/
Subject: LU-6215 o2iblnd: handle new struct ib_cq_init_attr
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: d9aa04c7b18d194c5bc1e851dd5d1b89e9cfb45a

[James A Simmons]

With the release of 4.2 a few more patches are now need.

1) misc_deregister no longer returns a int
2) bio_endio only take one argument now.
3) the follow_sym and put_sym api has changed.

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/16275
Subject: LU-6215 libcfs: misc_unregister is now void function
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 7052ef9c0a4c13a2089ffd3692171a889349a4a5

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/16278
Subject: LU-6215 llite: bio_endio only takes one argument for 4.2
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: df6b717f48220e8e85d980dcb316d4fc4a2dc083

[James A Simmons]

One last patch to go. In the upstream client Al Viro ripped out the THREAD_SIZE handling in ll_follow_link. Should we do the same thing>

[Alexander Boyko]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16118/
Subject: LU-6215 o2iblnd: handle new struct ib_cq_init_attr

This patch introduce a regression. It checks ib_cq_init_attr at configure path, but use it later for a ib_create_cq. It is wrong way, because ib_cq_init_attr and ib_create_cq modification was added at kernel by different commits. So kernel could have ib_cq_init_attr patch, but ib_create_cq() prototype does not include ib_cq_init_attr as parameter.

[James A Simmons]

Yes it was two separate commits which were committed together on the same day. True it is possible a future distribution could include only one of those patches. I have seen strange things with some distribution.

[James A Simmons]

Alex I looked into different OFED stacks and saw the tendency is to keep the kernel headers in sync with the latest kernels but the source code itself is a few versions behind. I did run into the issue due to this, sigh I guess you can never under estimate bad back porting. So for any infinband version checking always test the function changes, never the header api changes.

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/16342
Subject: LU-6215 ofed: strengthen ib_create_cq in 4.2+ kernels
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 378de434a39de02d396479bc3472f03ee491d55d

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/16376
Subject: LU-6215 llite: handle kernel symlink api changes in 4.2+ kernels
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 440992f546c3f48f69621a4d1d0c5b8c65a1024d

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16275/
Subject: LU-6215 libcfs: misc_unregister is now void function
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: ab2a180545986bfebbdbef320cab75be1603f856

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16342/
Subject: LU-6215 ofed: strengthen ib_create_cq test in 4.2+ kernels
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: d7223c0ccda11f120e50cd0195d84583d6ee4a29

[James A Simmons]

For 4.3-rc1 the return value for seq_print has been removed Very big patch coming.

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16376/
Subject: LU-6215 llite: handle kernel symlink api changes in 4.2+
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: ab8b977f77e44c3f4d57be2244e9df898fd15316

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/16649
Subject: LU-6215 ldlm: handle percpu_counter_init change in 3.18+ kernels
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: d62311d55d400daf74354e67164c657d82e2d443

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16278/
Subject: LU-6215 llite: bio_endio only takes one argument for 4.2
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: d410aeef76f79e0fc7b09227f850a4ebd462ad3f

[James A Simmons]

Latest landing enabled kernel 4.2.1 client support. One patch outstanding that enables server side, using ZFS, support for 4.2.1 kernels.

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16649/
Subject: LU-6215 ldlm: handle percpu_counter_init change in 3.18+ kernels
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: a2844eb7d4ea15b8164f219aec1900037f0c506b

[James A Simmons]

With the latest merge Lustre clients and Lustre servers using the latest ZFS can now work with linux kernel versions up to 4.2.1

[Andreas Dilger]

James, it would be useful to add this information to the lustre/ChangeLog file - separately for the client and server, and making it clear that the server support is only for ZFS and not ldiskfs (due to the lack of ldiskfs patch series).

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/16767
Subject: LU-6215 lnet: make o2iblnd buildable for 4.2.1 kernels
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 400be89ec5170b88da8b50bbd2ad789a882c17d4

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/16769
Subject: LU-6215 kernel: report 4.2.1 kernel support
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 7e739adcab379ee3abe4a562593de5f888248d55

[James A Simmons]

Andreas I did push a patch for the ChangeLog. Bob was surprise that no lbuild changes was included due to no setup for this. Is it okay to push a patch with no lbuild support? Also now that we support 4.2.1 perhaps it is time to look at adding new kernel testing to the matrix.

[Andreas Dilger]

I don't think the ability of Lustre to build against a certain kernel depends on lbuild support for same. I never use lbuild locally, just "make" or "make rpms" and leave the kernel alone.

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16767/
Subject: LU-6215 lnet: make o2iblnd buildable for 4.2.1 kernels
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: cc0733fe2d3c97435c4619f9123972641a8a6681

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/16933
Subject: LU-6215 lprocfs: handle seq_printf api change
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: e899950c68a564cce6981a3eb80cc0736cded658

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16769/
Subject: LU-6215 kernel: report 4.2.1 kernel support
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 95557d5c1080f68be2c24dad9f3acf83d6fcbcf9

[Gerrit Updater]

Li Dongyang ([e-mail]) uploaded a new patch: http://review.whamcloud.com/18728
Subject: LU-6215 gss: cache_head is now on a hlist in 4.3+ kernels
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 86809c26dc558061b372896ea0288288cbe0aafa

[Gerrit Updater]

Li Dongyang ([e-mail]) uploaded a new patch: http://review.whamcloud.com/18729
Subject: LU-6215 llite: handle locks API change in 4.4+ kernels
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 02724adaa8551024b9f1b0c638d3838ff16b03b5

[Gerrit Updater]

Li Dongyang ([e-mail]) uploaded a new patch: http://review.whamcloud.com/18730
Subject: LU-6215 llite: make_request_fn returns blk_qc_t in kernel 4.4
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 0eebe20c825d18c11dfc5908528919e257dc6bda

[Gerrit Updater]

Li Dongyang ([e-mail]) uploaded a new patch: http://review.whamcloud.com/18731
Subject: LU-6215 gss: key->payload.data is an array in 4.4+ kernels
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 8bb597ca9635aca7901a917ee67b414293d94283

[Gerrit Updater]

Li Dongyang ([e-mail]) uploaded a new patch: http://review.whamcloud.com/18759
Subject: LU-6215 o2iblnd: rdma_create_id() takes extra parameter
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: a2872b6ab3c2374ffda2546ac65af5abd41bf195

[Gerrit Updater]

Li Dongyang ([e-mail]) uploaded a new patch: http://review.whamcloud.com/18760
Subject: LU-6215 o2iblnd: handle ib_send_wr changes in 4.4+ kernels
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 809cfab31ef5dddcd1eaddb3362c3da7780fe19d

[James A Simmons]

Thanks Li for finishing off the 4.4 kernel support

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/18728/
Subject: LU-6215 gss: cache_head is now on a hlist in 4.3+ kernels
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 4ad19ae1a30de4c5f1d7a3a4e2fa2cd7dc72877f

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/18729/
Subject: LU-6215 llite: handle locks API change in 4.4+ kernels
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: dbaa768769a4075ca765dc76a138ef7a39c54eb0

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/18730/
Subject: LU-6215 llite: make_request_fn returns blk_qc_t in kernel 4.4
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 9806eb0efdba6d815808f53ada4423850f4493f7

[Gerrit Updater]

Dmitry Eremin ([e-mail]) uploaded a new patch: http://review.whamcloud.com/19168
Subject: LU-6215 lnet: split struct ib_send_wr
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 9cc37ce60ec2dbf3dcce583f4d2b0b0b968d84b7

[Gerrit Updater]

Li Dongyang ([e-mail]) uploaded a new patch: http://review.whamcloud.com/19185
Subject: LU-6215 o2iblnd: ib_alloc_fast_reg_mr() removed in 4.3
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 2e1ead93804ccb5c91259f994138059fb255f463

[Gerrit Updater]

Li Dongyang ([e-mail]) uploaded a new patch: http://review.whamcloud.com/19186
Subject: LU-6215 o2iblnd: port to new fast reg API introduced in 4.4
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 444fe1fcb62060ddcb1e5f2d493e5ab9c37e3849

[Andreas Dilger]

What do people think about having a separate LU ticket for each upstream kernel (e.g. "Build client with 4.3 kernel", "Build client with Linux 4.4 kernel", etc)? That would allow closing the ticket when the code is known to build with that particular kernel, and would go into the release tracker, instead of keeping this ticket open for multiple releases.

Also, as a reminder, please remember to update lustre/ChangeLog for the "vanilla Linux" client build version in the last patch of the series for that kernel so that it is clear to users which kernels the client can build with.

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/16933/
Subject: LU-6215 lprocfs: handle seq_printf api change
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: aff399789fe456d8dc58f49cd490c94916b2a46f

[James A Simmons]

I took your advise Andreas and opened a new ticket for 4.5 kernel support. Since many patches are in flight and ready to land for 4.4 I left this attached to this ticket.

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/18731/
Subject: LU-6215 gss: key->payload.data is an array in 4.4+ kernels
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 859c947848fbd7a0b681da6cc90eb701583121fd

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/19185/
Subject: LU-6215 o2iblnd: ib_alloc_fast_reg_mr() removed in 4.3
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 868bacf42bd3edfa47c76a06a7a32acaa18d02cc

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/18759/
Subject: LU-6215 o2iblnd: rdma_create_id() takes extra parameter
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 3e95945ff970426a5904c859feed46176e678305

[Gerrit Updater]

Andreas Dilger ([e-mail]) uploaded a new patch: http://review.whamcloud.com/19890
Subject: LU-6215 llite: remove obsolete conditional code
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 1c010827e96bd31a14a700d0c130acc61ff758ca

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/19890/
Subject: LU-6215 llite: remove obsolete conditional code
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 9f4042fa3256d8b5146a711872759187f25965fe

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/19186/
Subject: LU-6215 o2iblnd: port to new fast reg API introduced in 4.4
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 2366ba2b6f8ac2a6eeff796e89f1df9ce7ac546e

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/19168/
Subject: LU-6215 lnet: split struct ib_send_wr
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: ac25785328d31f63bd76a03f9cbb76f7f31f2ab0

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/20543
Subject: LU-6215 build: announce linxu kernel 4.4.6
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 4f5d60ddb17ddf1aa0aa5220260a75d7b709597c

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/20543/
Subject: LU-6215 build: announce linux kernel 4.4.6 support
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 0a0f9f7dafefcb4b9e3c3fc6e9fd8ecd368e7c00

[Jian Yu]

Hi James,
So far, all of the existing patches for this ticket have landed. Are there any new patches?

[Andreas Dilger]

If all of the current patches are landed, it would be best to close this bug for 2.9.0 and open a new one for 2.10.0 (linked to this one). That makes it a lot easier to track.

[James A Simmons]

I first need to open the new ticket and add all the open links on this ticket to that new one.

[Peter Jones]

I cannot mark the ticket as resolved because it has technical tasks still open so I have removed the 2.9 fixversion. IMHO the best thing to do from a tracking point of view would be to have a ticket for each kernel version so it is really clear to understand which patches are needed for a given kernel version.

[James A Simmons]

Future kernel patches are being directed to separate tickets. Currently kernel 4.5 support is being done under LU-8056. This ticket is really meant to be a task tracking ticket for the upstream <-> master work.

[Gerrit Updater]

James Simmons ([e-mail]) uploaded a new patch: http://review.whamcloud.com/21970
Subject: LU-6215 build: announce linux kernel 4.5.7 support
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 92266f98adea3aa0a0f8166cd858e65be85aad6d

[James A Simmons]

My bad. Wrong ticket number was used by mistake. Now its under LU-8056.

[James A Simmons]

Its no longer easy to see what work is left so I started a new ticket to track the remaining work, LU-9679. Please migrate to that ticket.