Support multiple RC implementations, and add tests for that. #36

yosefe · 2014-11-27T17:21:59Z

No description provided.

mellanox-github · 2014-11-27T17:26:10Z

Test FAILed.
Refer to this link for build results (access rights to CI server needed):
http://bgate.mellanox.com:8000/jenkins-secure/job/gh-ucx-pr/63/

Build Log
last 50 lines

[...truncated 737 lines...]
make  all-am
make[3]: Entering directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test/src/tools/info'
  CC       ucx_info.o
  CCLD     ucx_info
make[3]: Leaving directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test/src/tools/info'
make[2]: Leaving directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test/src/tools/info'
Making all in test/perf
make[2]: Entering directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test/test/perf'
  CC       libperf.lo
  CCLD     libucxperf.la
  CCLD     ucx_perftest
make[2]: Leaving directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test/test/perf'
make[2]: Entering directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test'
make[2]: Nothing to be done for `all-am'.
make[2]: Leaving directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test'
make[1]: Leaving directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test'
+ make -j31 distcheck
make  dist-gzip am__post_remove_distdir='@:'
make[1]: Entering directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test'
if test -d "ucx-0.1.69"; then find "ucx-0.1.69" -type d ! -perm -200 -exec chmod u+w {} ';' && rm -rf "ucx-0.1.69" || { sleep 5 && rm -rf "ucx-0.1.69"; }; else :; fi
test -d "ucx-0.1.69" || mkdir "ucx-0.1.69"
 (cd src/ucs && make  top_distdir=../../ucx-0.1.69 distdir=../../ucx-0.1.69/src/ucs \
     am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir)
make[2]: Entering directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test/src/ucs'
make[2]: Leaving directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test/src/ucs'
 (cd src/uct && make  top_distdir=../../ucx-0.1.69 distdir=../../ucx-0.1.69/src/uct \
     am__remove_distdir=: am__skip_length_check=: am__skip_mode_fix=: distdir)
make[2]: Entering directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test/src/uct'
make[2]: *** No rule to make target `ib/rc/rc_mlx5.h', needed by `distdir'.  Stop.
make[2]: Leaving directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test/src/uct'
make[1]: *** [distdir] Error 1
make[1]: Leaving directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test'
make: *** [dist] Error 2
Build step 'Execute shell' marked build as failure
TAP Reports Processing: START
Looking for TAP results report in workspace using pattern: **/*.tap
Did not find any matching files.
[Valgrind] Files to copy:
[Valgrind] Analysing valgrind results
[Valgrind] Ending the valgrind analysis.
Anchor chain: could not read file with links: /var/lib/jenkins/jobs/gh-ucx-pr/workspace/jenkins_sidelinks.txt (No such file or directory)
[copy-to-slave] The build is taking place on the master node, no copy back to the master will take place.
Email was triggered for: Failure - Any
Sending email for trigger: Failure - Any
Request made to compress build log
Sending email to: eugene@mellanox.com miked@mellanox.com yosefe@mellanox.com
[BFA] Scanning build for known causes...

[BFA] Done. 0s

Test FAILed.

mellanox-github · 2014-11-27T17:36:11Z

Test FAILed.
Refer to this link for build results (access rights to CI server needed):
http://bgate.mellanox.com:8000/jenkins-secure/job/gh-ucx-pr/64/

Build Log
last 50 lines

[...truncated 1096 lines...]
../../../src/uct/ib/rc/rc_mlx5.c:69: error: expected ';' before 'qp_info'
cc1: warnings being treated as errors
../../../src/uct/ib/rc/rc_mlx5.c:74: error: implicit declaration of function 'uct_ib_mlx5_get_qp_info'
../../../src/uct/ib/rc/rc_mlx5.c:74: error: 'qp_info' undeclared (first use in this function)
../../../src/uct/ib/rc/rc_mlx5.c:81: error: 'MLX5_SEND_WQE_BB' undeclared (first use in this function)
../../../src/uct/ib/rc/rc_mlx5.c:95: error: 'MLX5_SND_DBR' undeclared (first use in this function)
../../../src/uct/ib/rc/rc_mlx5.c: In function 'uct_rc_mlx5_ep_put_short':
../../../src/uct/ib/rc/rc_mlx5.c:129: error: 'MLX5_SEND_WQE_BB' undeclared (first use in this function)
../../../src/uct/ib/rc/rc_mlx5.c:132: error: 'MLX5_OPCODE_RDMA_WRITE' undeclared (first use in this function)
../../../src/uct/ib/rc/rc_mlx5.c:134: error: 'MLX5_WQE_CTRL_CQ_UPDATE' undeclared (first use in this function)
../../../src/uct/ib/rc/rc_mlx5.c:135: error: implicit declaration of function 'htonll'
../../../src/uct/ib/rc/rc_mlx5.c:139: error: duplicate case value
../../../src/uct/ib/rc/rc_mlx5.c:139: error: previously used here
../../../src/uct/ib/rc/rc_mlx5.c:140: error: 'MLX5_INLINE_SEG' undeclared (first use in this function)
../../../src/uct/ib/rc/rc_mlx5.c: In function 'uct_rc_mlx5_iface_progress':
../../../src/uct/ib/rc/rc_mlx5.c:185: error: invalid application of 'sizeof' to incomplete type 'struct mlx5_cqe64' 
../../../src/uct/ib/rc/rc_mlx5.c:186: error: implicit declaration of function 'uct_ib_mlx5_cqe_hw_owned'
../../../src/uct/ib/rc/rc_mlx5.c:195: error: dereferencing pointer to incomplete type
../../../src/uct/ib/rc/rc_mlx5.c: In function 'uct_rc_mlx5_iface_query':
../../../src/uct/ib/rc/rc_mlx5.c:207: error: 'MLX5_SEND_WQE_BB' undeclared (first use in this function)
../../../src/uct/ib/rc/rc_mlx5.c: In function 'uct_rc_mlx5_iface_t_init':
../../../src/uct/ib/rc/rc_mlx5.c:229: error: 'uct_ib_mlx5_cq_info_t' undeclared (first use in this function)
../../../src/uct/ib/rc/rc_mlx5.c:229: error: expected ';' before 'cq_info'
../../../src/uct/ib/rc/rc_mlx5.c:241: error: implicit declaration of function 'uct_ib_mlx5_get_cq_info'
../../../src/uct/ib/rc/rc_mlx5.c:241: error: 'cq_info' undeclared (first use in this function)
make[3]: *** [ib/rc/libuct_la-rc_verbs.lo] Error 1
make[3]: *** [ib/rc/libuct_la-rc_mlx5.lo] Error 1
make[3]: Leaving directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test/ucx-0.1.70/_build/src/uct'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test/ucx-0.1.70/_build'
make[1]: *** [all] Error 2
make[1]: Leaving directory `/scrap/jenkins/jobs/gh-ucx-pr/workspace/build-test/ucx-0.1.70/_build'
make: *** [distcheck] Error 1
Build step 'Execute shell' marked build as failure
TAP Reports Processing: START
Looking for TAP results report in workspace using pattern: **/*.tap
Did not find any matching files.
[Valgrind] Files to copy:
[Valgrind] Analysing valgrind results
[Valgrind] Ending the valgrind analysis.
Anchor chain: could not read file with links: /var/lib/jenkins/jobs/gh-ucx-pr/workspace/jenkins_sidelinks.txt (No such file or directory)
[copy-to-slave] The build is taking place on the master node, no copy back to the master will take place.
Email was triggered for: Failure - Any
Sending email for trigger: Failure - Any
Request made to compress build log
Sending email to: eugene@mellanox.com miked@mellanox.com yosefe@mellanox.com
[BFA] Scanning build for known causes...

[BFA] Done. 0s

Test FAILed.

* Remove file from automake * Remove unneeded files

mellanox-github · 2014-11-27T17:41:45Z

Test PASSed.
Refer to this link for build results (access rights to CI server needed):
http://bgate.mellanox.com:8000/jenkins-secure/job/gh-ucx-pr/65/
Test PASSed.

mellanox-github · 2014-11-27T17:42:19Z

Test PASSed.
Refer to this link for build results (access rights to CI server needed):
http://bgate.mellanox.com:8000/jenkins-secure/job/gh-ucx-pr/66/
Test PASSed.

shamisp · 2014-11-27T17:54:07Z

src/uct/ib/rc/rc_mlx5.c

    /* Advance queue pointer */
    if (ucs_unlikely((ep->tx.seg += MLX5_SEND_WQE_BB) >= ep->tx.qend)) {
        ep->tx.seg = ep->tx.qstart;
    }

+    /* Completion counters */
+    ++ep->tx.sw_pi;
+    ++ucs_derived_of(ep->super.super.iface, uct_rc_iface_t)->tx.outstanding;


Do you want to introduce IB_INC() macro. Once we introduce thread safety, you will want to change it to atomic increment. If we have macro it will be easier to do

not right now, because i'm not sure making this atomic will be needed for thread safety. maybe there will just be per-iface lock.

mellanox-github · 2014-11-29T14:22:58Z

Test PASSed.
Refer to this link for build results (access rights to CI server needed):
http://bgate.mellanox.com:8000/jenkins-secure/job/gh-ucx-pr/67/
Test PASSed.

shamisp · 2014-11-29T15:02:54Z

👍

Support multiple RC implementations, and add tests for that.

…int3 IODEMO: data integrity check

fix bug for big messages of TCP

Refactor data handling

yosefe added 3 commits November 26, 2014 19:48

BUILD: Define HAVE_MPI in config.h

68e868e

UCT/IB: Add RC over Verbs transport.

540a01d

TEST: Run transport tests on all available devices and transports.

1d70cc4

yosefe changed the title ~~TEST: Run transport tests on all available devices and transports.~~ Support multiple RC implementations, and add tests for that. Nov 27, 2014

yosefe assigned shamisp Nov 27, 2014

yosefe force-pushed the topic/multi-transport-rc branch from a4fb240 to 42d8b03 Compare November 27, 2014 17:37

UCT/TEST: Build fixes

a5add1a

* Remove file from automake * Remove unneeded files

yosefe force-pushed the topic/multi-transport-rc branch from 42d8b03 to a5add1a Compare November 27, 2014 17:38

shamisp reviewed Nov 27, 2014
View reviewed changes

IB: remove __builtin_memcpy

927a5dd

yosefe added a commit that referenced this pull request Nov 29, 2014

Merge pull request #36 from yosefe/topic/multi-transport-rc

4a2ae30

Support multiple RC implementations, and add tests for that.

yosefe merged commit 4a2ae30 into openucx:master Nov 29, 2014

yosefe deleted the topic/multi-transport-rc branch November 29, 2014 15:03

amaslenn pushed a commit to amaslenn/ucx that referenced this pull request Jun 2, 2020

CI: check steps generator openucx#36

542ee74

evgeny-leksikov pushed a commit to evgeny-leksikov/ucx that referenced this pull request Sep 21, 2020

Merge pull request openucx#36 from evgeny-leksikov/iodemo_data_check_…

cb805e2

…int3 IODEMO: data integrity check

shizhibao pushed a commit to shizhibao/ucx that referenced this pull request Feb 10, 2021

Merge pull request openucx#36 from zheng871026/huawei

c4ea496

fix bug for big messages of TCP

dmitrygx pushed a commit to dmitrygx/ucx that referenced this pull request Dec 1, 2021

Merge pull request openucx#36 from TomAugspurger/data-handling

3a50105

Refactor data handling

nmorey mentioned this pull request Jan 13, 2022

[Q] MALLOC_HOOKS default mode description #5644

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Support multiple RC implementations, and add tests for that. #36

Support multiple RC implementations, and add tests for that. #36

yosefe commented Nov 27, 2014

mellanox-github commented Nov 27, 2014

mellanox-github commented Nov 27, 2014

mellanox-github commented Nov 27, 2014

mellanox-github commented Nov 27, 2014

shamisp Nov 27, 2014

yosefe Nov 29, 2014

mellanox-github commented Nov 29, 2014

shamisp commented Nov 29, 2014

Support multiple RC implementations, and add tests for that. #36

Support multiple RC implementations, and add tests for that. #36

Conversation

yosefe commented Nov 27, 2014

mellanox-github commented Nov 27, 2014

mellanox-github commented Nov 27, 2014

mellanox-github commented Nov 27, 2014

mellanox-github commented Nov 27, 2014

shamisp Nov 27, 2014

Choose a reason for hiding this comment

yosefe Nov 29, 2014

Choose a reason for hiding this comment

mellanox-github commented Nov 29, 2014

shamisp commented Nov 29, 2014