1
1

30128 Коммитов

Автор SHA1 Сообщение Дата
Brian Barrett
a6d97e2d6d
Merge pull request #7815 from raafatfeki/topic/ime
Topic/ime: Bring over IME component from master to v4.1
2020-06-26 12:40:35 -07:00
Christoph Niethammer
ad1d427d60 Accept UCX 1.8 in configure of btl/uct
The configure script for the btl uct component reports an error for
the new UCX 1.8.0 versions as it was fixed up to UCX 1.7.

This fixes #7612

Signed-off-by: Christoph Niethammer <niethammer@hlrs.de>
(cherry picked from commit 9b10f46126b0a5aa796d9fec063c2d454a9a1bc9)
2020-06-26 21:20:49 +02:00
Brian Barrett
25abbb219a
Merge pull request #7808 from bwbarrett/backports/v4.1.x-collectives-updates
Backport Collective changes from master to v4.1.x
2020-06-26 12:01:51 -07:00
raafatfeki
6e145188d9 fs/ime & fbtl/ime: Support of IME file system
Signed-off-by: raafatfeki <fekiraafat@gmail.com>
2020-06-26 12:26:51 -04:00
Jeff Squyres
7987a7f56e common_ofi: fix preprocessor macro typo
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
(cherry picked from commit f64c30e93c719b0c910e81f716b68d5889de8ce5)
2020-06-26 07:56:53 -07:00
Brian Barrett
339ee6378a dist: Add Collectives backports to NEWS
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:07:26 +00:00
William Zhang
db6ed187b2 coll/tuned: Add NULL check to prevent segfault
Signed-off-by: William Zhang <wilzhang@amazon.com>

cr https://code.amazon.com/reviews/CR-23837553

(cherry picked from commit 771f9c011d2a4daf78a4b26f88c971b3868fe132)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
William Zhang
03758b1ef7 coll/tuned: Fix typos
Signed-off-by: William Zhang <wilzhang@amazon.com>
(cherry picked from commit 50640402ab5765a0dfde71628adfbbaa686555bd)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Mikhail Brinskii
7eb94164a0 COLL/TUNED: Add linear scatter using isend for mlnx platform
Signed-off-by: Mikhail Brinskii <mikhailb@mellanox.com>
(cherry picked from commit f2cbd4806e9a38b5e58c0fc69b41624af79fb99b)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Gilles Gouaillardet
221fad6862 coll/cuda: remove unnecessary references to ORTE
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
(cherry picked from commit 531171ca50955f8b7762f932388f633852454e6f)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Tomislav Janjusic
f51bd8ca0c Coll/hcoll: adding scatterv interface
Signed-off-by: Valentin Petrov valentinp@mellanox.com
(cherry picked from commit 6ea920e225c7ed905949c6afec554b6bf2705f94)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Alex Anenkov
2891a23329 coll/libnbc: add recursive doubling algorithm for MPI_Iallreduce
Signed-off-by: Alex Anenkov <anenkov.ru@gmail.com>
(cherry picked from commit 77d466edf369c9851476b7ec7392f3dfd4cdc0b1)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Mikhail Kurnosov
ba11f31fc8 coll/libnbc: remove debug output
1. Remove debug output in iallgather (I have forgotten to remove it).
2. Remove an incorrect comment in description of ibcast

Signed-off-by: Mikhail Kurnosov <mkurnosov@gmail.com>
(cherry picked from commit 64abd0f405be91b927cd8f37d30cdf41aa6685c2)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Mikhail Kurnosov
bf1c8bb394 coll/libnbc/ireduce: silence Coverity warning CID 1440360
Signed-off-by: Mikhail Kurnosov <mkurnosov@gmail.com>
(cherry picked from commit 8b511c788965e6467b5fd834f1adbdcca5012f55)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Mikhail Kurnosov
5ee1fb62b9 coll/libnbc: add Rabenseifner's algorithm for MPI_Iallreduce
An implementation of R. Rabenseifner's algorithm for MPI_Iallreduce.

This algorithm is a combination of a reduce-scatter implemented with recursive vector halving
and recursive distance doubling, followed either by an allgather.

Limitations:
-- count >= 2^{\floor{\log_2 p}}
-- commutative operations only
-- intra-communicators only

Signed-off-by: Mikhail Kurnosov <mkurnosov@gmail.com>
(cherry picked from commit 73e048b62a92325fc3fca80c2ade5f5e9bf3192a)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
George Bosilca
fd29cce114 Remove few warnings in libnbc identified by clang-1000.11.45.2
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
(cherry picked from commit 66182a294d5e8cf03a00fba579b05f59e764133c)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Mikhail Kurnosov
91a4b4c799 coll/libnbc: add recursive doubling algorithm for MPI_Iallgather
Implements recursive doubling algorithm for MPI_Iallgather.
The algorithm can be used only for power-of-two number of processes.

Signed-off-by: Mikhail Kurnosov <mkurnosov@gmail.com>
(cherry picked from commit a7386c1e09fb274991ca5b50d9d418a0d6b77b6c)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Mikhail Kurnosov
6971dab943 coll/libnbc: add knomial tree algorithm for MPI_Ibcast
Signed-off-by: Mikhail Kurnosov <mkurnosov@gmail.com>
(cherry picked from commit b0429d25dfb1ed3f9adfa45169478cada0ba2675)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Mikhail Kurnosov
a318f117f6 coll/libnbc: add Rabenseifner's algorithm for MPI_Ireduce
An implementation of R. Rabenseifner's algorithm for MPI_Ireduce.
This algorithm is a combination of a reduce-scatter implemented with recursive vector halving
and recursive distance doubling, followed either by a gather.

Limitations:
-- count >= 2^{\floor{\log_2 p}}
-- commutative operations only
-- intra-communicators only

Signed-off-by: Mikhail Kurnosov <mkurnosov@gmail.com>
(cherry picked from commit 7bd63e79c865080c801a45ed852602bdc4eb4d8f)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Brian Barrett
6f6d8180a3 coll libnbc: Remove dead code
Remove dead code that was causing warnings about unused static
functions.

Signed-off-by: Brian Barrett <bbarrett@amazon.com>
(cherry picked from commit 2e24e6ec082d29f76dcbc75c6f214d2d0d647701)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Mikhail Kurnosov
de5e435dee coll/libnbc: add recursive doubling algorithm for MPI_Iexscan
Implements recursive doubling algorithm for MPI_Iexscan.
The algorithm preserves order of operations so it can be used both
by commutative and non-commutative operations.

The MCA parameter 'coll_libnbc_iexscan_algorithm' was added for dynamic
algorithm selection.

Signed-off-by: Mikhail Kurnosov <mkurnosov@gmail.com>
(cherry picked from commit dfe203e167f5d8abc3b55226c6f17a468c9567dd)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Mikhail Kurnosov
65990af3ad coll/libnbc: add recursive doubling algorithm for MPI_Iscan
Implements recursive doubling algorithm for MPI_Iscan. The algorithm preserves order of operations so it can be used both by commutative and non-commutative operations.

The MCA parameter coll_libnbc_iscan_algorithm was added for dynamic algorithm selection.

Signed-off-by: Mikhail Kurnosov <mkurnosov@gmail.com>
(cherry picked from commit 3d43ff0f3209d5bf4713c6696acb0acb8f1756e4)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Jeff Squyres
547fb3d933 libnbc: remove some stale/dead code
Gcc 8 identified hb_tree_csearch() as an infinite recursion, and it
turns out that we never call this function, anyway.  So just remove
it.

Fixes #5670.

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
(cherry picked from commit 06c1bf73da875f4a6449f38a993530d6fae7817d)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Aurelien Bouteiller
2692840d40 Always return a valid error code from collective operations
Signed-off-by: Aurelien Bouteiller <bouteill@icl.utk.edu>
(cherry picked from commit 466217fadda0391698b383f2792de7bcbdff7e97)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Gilles Gouaillardet
d9d84d5dd6 coll/libnbc: fix NBC_Unpack()
always initialize 'size'.

Only the a2a_sched_diss() alltoall algorithm is impacted,
and this algo is currently unused, so there is no need
to backport nor update the NEWS file for now.

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
(cherry picked from commit ff48e9286430b37aac3146efe2b355f255db94d5)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Mikhail Kurnosov
ba221e1a08 coll/base/allgatherv: fix MPI_IN_PLACE processing
The call of MPI_Allgatherv with sendbuf and sendtype parameters equal to MPI_IN_PLACE and NULL correspondingly, produces the segmentation fault.

The problem is that sendtype is used even when sendbuf value is MPI_IN_PLACE. But according to the standard, sendtype and sendcount parameters should be ignored in this case.

Signed-off-by: Mikhail Kurnosov <mkurnosov@gmail.com>
(cherry picked from commit b45e190e6629b664872f7f872cfffd916180bb9a)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-25 23:06:51 +00:00
Jeff Squyres
465414953d
Merge pull request #7828 from edgargabriel/pr/v4.1.x-avg-fview-size
common/ompio: use avg. file view size in the aggregator selection logic
2020-06-25 11:45:48 -04:00
Brian Barrett
173142bf32
Merge pull request #7824 from hoopoepg/topic/ucx-test-external-events-v4.1
COMMON/UCX: improved missing events test - v4.1
2020-06-25 08:10:54 -07:00
Joseph Schuchart
d4219f8144 OSC UCX: make sure no-op fetch in rget/rput is properly aligned
Signed-off-by: Joseph Schuchart <schuchart@hlrs.de>
(cherry picked from commit c1f7776341254b825a1b5f12fb93088c73c2c2ef)
2020-06-25 17:05:48 +02:00
Jeff Squyres
a3258afad9
Merge pull request #7814 from raafatfeki/topic/gpfs_4.1.x
fs/gpfs: Bring over GPFS component from master to v4.1
2020-06-25 10:47:39 -04:00
Jeff Squyres
470b1c518d
Merge pull request #7812 from hppritcha/topic/cobalt_for_orte
RAS:ALPS add support for ANL Cobalt
2020-06-25 10:40:57 -04:00
Jeff Squyres
2ba26c746d
Merge pull request #7831 from hppritcha/backports/v4.1.x-psm2-updates
Backports/v4.1.x psm2 updates
2020-06-25 10:40:38 -04:00
Jeff Squyres
4e95b7ae60
Merge pull request #7836 from jsquyres/pr/v4.1.x/fix-test-asm-run-tests-script
v4.1.x: tests/asm/run_tests: fix basename usage
2020-06-25 10:40:18 -04:00
Brian Barrett
9ffe9535f3
Merge pull request #7837 from markalle/interception_early_toc_read_v41x
noinline to avoid compiler reading TOC before PATCHER_BEGIN
2020-06-24 10:55:35 -07:00
Brian Barrett
9550cb2a97
Merge pull request #7863 from hppritcha/topic/patch_for_ofi_v41x
add a common ofi whitelist/blacklist
2020-06-24 10:55:03 -07:00
Howard Pritchard
8fcf9cee3b add a common ofi whitelist/blacklist
also add common verbose variable.

Note the verbosity thing is a little tricky owing to the way the MCA frameworks and components are registered and
and initialized.  The BTL's are registered/initialized prior to the MTL components even getting registered.

Here's the change in ofi mtl mca parameters.  Before commit:

            MCA mtl ofi: parameter "mtl_ofi_provider_include" (current value: "psm2", data source: environment, level: 1 user/basic, type: string)
                          Comma-delimited list of OFI providers that are considered for use (e.g., "psm,psm2"; an empty value means that all providers will be considered). Mutually exclusive with mtl_ofi_provider_exclude.
             MCA mtl ofi: parameter "mtl_ofi_provider_exclude" (current value: "shm,sockets,tcp,udp,rstream", data source: default, level: 1 user/basic, type: string)
                          Comma-delimited list of OFI providers that are not considered for use (default: "sockets,mxm"; empty value means that all providers will be considered). Mutually exclusive with mtl_ofi_provider_include.

After commit:

             MCA btl ofi: parameter "btl_ofi_provider_include" (current value: "", data source: default, level: 1 user/basic, type: string, synonym of: opal_common_ofi_provider_include)
                          Comma-delimited list of OFI providers that are considered for use (e.g., "psm,psm2"; an empty value means that all providers will be considered). Mutually exclusive with mtl_ofi_provider_exclude.
             MCA btl ofi: parameter "btl_ofi_provider_exclude" (current value: "shm,sockets,tcp,udp,rstream", data source: default, level: 1 user/basic, type: string, synonym of: opal_common_ofi_provider_exclude)
                          Comma-delimited list of OFI providers that are not considered for use (default: "sockets,mxm"; empty value means that all providers will be considered). Mutually exclusive with mtl_ofi_provider_include.
             MCA mtl ofi: parameter "mtl_ofi_provider_exclude" (current value: "shm,sockets,tcp,udp,rstream", data source: default, level: 1 user/basic, type: string, synonym of: opal_common_ofi_provider_exclude)
                          Comma-delimited list of OFI providers that are not considered for use (default: "sockets,mxm"; empty value means that all providers will be considered). Mutually exclusive with mtl_ofi_provider_include.
             MCA mtl ofi: parameter "mtl_ofi_verbose" (current value: "0", data source: default, level: 3 user/all, type: int, synonym of: opal_common_ofi_verbose)

related to #7755

Signed-off-by: Howard Pritchard <howardp@lanl.gov>
(cherry picked from commit 9f1081a07ac3c7b7277a27277ed970ed713207c9)
(cherry picked from commit 45b643d0cfa46f1abb9a5f43cf0ff304cf6a5fea)
2020-06-23 14:02:33 -06:00
Jeff Squyres
6d550a32ca
Merge pull request #7856 from hjelmn/v4.1.x_add_a_missing_osc_rdma_commit_for_dynamic_memory_windows_that_was_forgotten_before_the_last_release_doh
osc/rdma: fix bug in attach for non-debug builds
2020-06-22 17:16:42 -04:00
Jeff Squyres
4a466e4f08
Merge pull request #7853 from devreal/osc-rdma-noncontig-requests-v4.1.x
osc rdma: check for outstanding fragments before completing a request (v4.1.x)
2020-06-22 11:53:48 -04:00
Jeff Squyres
c8e5227e1a
Merge pull request #7173 from jjhursey/v4-jsm-schizo
Add detection for JSM direct launch
2020-06-22 11:51:52 -04:00
Nathan Hjelm
059c9614a6 osc/rdma: fix bug in attach for non-debug builds
This commit fixes an issue with non-debug builds where adding an
attachment to the attachment list doesn't actually happen. This
causes all MPI_Win_detach calls to fail. The call was within an
assert which is optimized out in optimized builds.

Signed-off-by: Nathan Hjelm <hjelmn@google.com>
(cherry picked from commit 8ee80d885561c4349ab97725416613614d63c77f)
2020-06-22 08:34:43 -06:00
Joseph Schuchart
1af51d07d6 osc rdma: check for outstanding fragments before completing a request
Signed-off-by: Joseph Schuchart <schuchart@hlrs.de>
(cherry picked from commit 85ed26f2f859283522188a68dea493cdc99d76aa)
2020-06-22 08:32:14 +02:00
Mark Allen
1f5adc65bc noinline to avoid compiler reading TOC before PATCHER_BEGIN
This bug was first seen in a different product that's using the same
interception code as OMPI.  But I think it's potentially in OMPI too.

In my vanilla build of OMPI master on RH8 if I "gdb libopen-pal.so" and
"disassemble intercept_brk", I'm seeing a suspicious extra instruction
in front of PATCHER_BEGIN:
   0x00000000000d6778 <+40>:    std     r2,24(r1) // something gcc put in front
   0x00000000000d677c <+44>:    std     r2,96(r1) // PATCHER_BEGIN's toc_save
   0x00000000000d6780 <+48>:    nop               // NOPs from PATCHER_BEGIN
   0x00000000000d6784 <+52>:    nop               // that get replaced
   0x00000000000d6788 <+56>:    nop               // by instructions that
   0x00000000000d678c <+60>:    nop               // change r2
   0x00000000000d6790 <+64>:    nop               //

Later there are loads from that location like
   0x000000000019e0e4 <+132>:   ld      r2,24(r1)
that make me nervous since that's the pre-updated value.

I believe this is the same thing Nathan is describing way back in a9bc692d
and his solution was to put a second call around each interception, where
the outer call is just
    intercept_brk():
        PATCHER_BEGIN
        _intercept_brk()
        PATCHER_END
and the inner call _intercept_brk() is where the bulk of the code goes.

What I'm seeing is that _intercept_brk() is being inlined and probably
negating Nathan's fix.  So I want to add __opal_attribute_noinline__ to
restore the fix.

With this commit in place, the disassembly of intercept_brk becomes tiny
because it's no longer inlining _intercept_brk() and the susipicious
early save of r2 is gone.  I made the same fix to all the intercept_*
functions, although intercept_brk was the only one that had a suspicious
save of r2.

As far as empirical failures though, we only have those from the non-OMPI
product that's using the same patcher code.  I'm not actually getting OMPI
to fail from the above suspicious data being saved in r1+24.

Signed-off-by: Mark Allen <markalle@us.ibm.com>
(cherry picked from commit ddd1f578ecfc443d05250e09bf5e5077c6d6f304)
2020-06-17 18:45:54 -04:00
Brian Barrett
a4e8f2b2cb
Merge pull request #7807 from bwbarrett/backports/v4.1.x-ofi-updates
Backport OFI changes from master to v4.1.x
2020-06-17 15:07:32 -07:00
Brian Barrett
204922fff6 dist: Add OFI backports to NEWS
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-17 20:49:19 +00:00
Harumi Kuno
61b11e4101 mtl_btl_ofi_rcache_init() before creating domain
mtl_btl_ofi_rcache_init() initializes patcher which should only take
place things are single threaded.  OFI providers may start spawn threads,
so initialize the rcache before creating OFI objects to prevent races.

Authored-by: John L. Byrne <john.l.byrne@hpe.com>
Signed-off-by: Harumi Kuno <harumi.kuno@hpe.com>
(cherry picked from commit f1b21cb77680106be870ad29e5a7534862fceed7)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-17 20:49:19 +00:00
Nikola Dancejic
c51917675c common/ofi: Fixing compilation issue with ofi versions that do not support fi_info.nic
Added the flag OPAL_OFI_PCI_DATA_AVAILABLE to remove accessing the nic
object in
fi_info when the ofi version does not support that structure.

Signed-off-by: Nikola Dancejic dancejic@amazon.com
(cherry picked from commit ae2a447b0eddaac057beecdba99e10903051a2a7)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-17 20:49:19 +00:00
Jeff Squyres
c8659a98c4 ofi: revamp OPAL_CHECK_OFI configury
Update the OPAL_CHECK_OFI configury macro:

- Make it safe to call the macro multiple times:
  - The checks only execute the first time it is invoked
  - Subsequent invocations, it just emits a friendly "checking..."
    message so that configure output is sensible/logical
- With the goal of ultimately removing opal/mca/common/ofi, rename the
  output variables from OPAL_CHECK_OFI to be
  opal_ofi_{happy|CPPFLAGS|LDFLAGS|LIBS}.
- Update btl/ofi, btl/usnic, and mtl/ofi for these new conventions.
- Also, don't use AC_REQUIRE to invoke OPAL_CHECK_OFI because that
  causes the macro to be invoked at a fairly random time, which makes
  configure stdout confusing / hard to grok.
- Remove a little left-over kruft in OPAL_CHECK_OFI, too (which
  resulted in an indenting change, making the change to
  opal_check_ofi.m4 look larger than it really is).

Thanks Alastair McKinstry for the report and initial fix.
Thanks Rashika Kheria for the reminder.

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
(cherry picked from commit f5e1a672ccd5db127e85e1e8f6bcfeb8a8b04527)

NOTE: This patch was cherry-picked into the v4.0.x branch as 9ad871fc,
but the OFI BTL changes were skipped, because the OFI BTL was not in
the v4.0.x branch.  This version of the cherry pick brings in the
changes to the OFI BTL.

Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-17 20:49:19 +00:00
Nikola Dancejic
d9bfb1c4da common/ofi: Added multi-NIC support to provider selection
Adds the capability to select a NIC based on hardware locality.
Creates a list of NICs that share the same cpuset as the process,
then selects the NIC based on the (local rank) % (number of NICs).
If no NICs are available that share the same cpuset, the selection process
will create a list of all available NICs and make a selection based on
(local rank) % (number of NICs)

Signed-off-by: Nikola Dancejic <dancejic@amazon.com>
(cherry picked from commit 167d75b42ac3ca4770d59c796c011d72e0fffde3)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-17 20:49:19 +00:00
Brian Barrett
62bfd52870 ofi: Call add_procs through PML
Change ompi_mtl_ofi_get_endpoint() to call the active PML's
add_procs() rather than the OFI MTL add_procs() directly when
discovering a new process during operation.

Functionally, this has no impact in correct operation.  However,
the current behavior means that the heterogenous and active PML
checks are not being executed in the dynamic discovery case.

Signed-off-by: Brian Barrett <bbarrett@amazon.com>
(cherry picked from commit 64d70b3076186b044f39f11d0d7d3e5bbcaec469)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-17 20:49:19 +00:00
Harumi Kuno
2e137ae31c set ep to NULL to avoid double close
Per suggestion of @awlauria

Signed-off-by: Harumi Kuno <harumi.kuno@hpe.com>
(cherry picked from commit ab4875ddc2b76ceced120fbfe09d8dcbde6e6ff3)
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
2020-06-17 20:49:19 +00:00