Skip to content
Open
Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
200 commits
Select commit Hold shift + click to select a range
ba4580a
add printf
May 18, 2023
4b9c639
Revert "Perftest: replace rand() with getrandom() during MR buffer in…
HassanKhadour Jun 12, 2023
8ff29c1
modify --source_ip to --bind_sounce_ip to fix init connection establi…
May 18, 2023
ff1a84a
Fix issue with PD deallocation.
pim-pesochek Jun 15, 2023
9711d16
Merge pull request #210 from w180112/master
HassanKhadour Jun 19, 2023
4ad453f
Merge pull request #213 from HassanKhadour/master
HassanKhadour Jun 22, 2023
0fc987c
Merge pull request #215 from pim-pesochek/master
HassanKhadour Jun 22, 2023
f226ca2
Perftest: Align WQE length to MTU in case of shared queue
HassanKhadour Jun 19, 2023
7d00c4b
Perftest: Version increase to 6.16
HassanKhadour Jun 22, 2023
5a218cd
Add ipv6 address support for initial communication.
HassanKhadour Jun 28, 2023
abf6dd9
Perftest: Fix and optimize initial communication functions.
HassanKhadour Jul 6, 2023
56a53ec
Merge pull request #218 from HassanKhadour/master
HassanKhadour Jul 11, 2023
dbf395e
Fix race in non-rdmacm ctx_close_connection()
rolandd Jul 17, 2023
d051db0
Add support for DMA-buffers in Neuron devices
YonatanNachum May 10, 2023
9c4f8ed
Add missing HW accelerator flags to perftest's man
YonatanNachum Jul 2, 2023
48c0974
Perftest: Fix limit_bw in ib_send_bw bidir traffic duration mode
HassanKhadour Aug 1, 2023
47a7de5
Merge pull request #222 from YonatanNachum/neuron
HassanKhadour Aug 3, 2023
1e211ce
Merge pull request #220 from rolandd/fix-shutdown
HassanKhadour Aug 7, 2023
c1f8d3b
Perftest: Version increase to 6.17
HassanKhadour Aug 8, 2023
e3f0700
Perftest: changing spec file version to 23.10.0.
HassanKhadour Aug 13, 2023
69bc866
perftest: support set flow_label in GRH
changchengx Aug 27, 2023
5856a7f
Merge pull request #224 from changchengx/flow_label
HassanKhadour Aug 27, 2023
59fb505
Error out if CuDeviceGetByPCIBusId fails
jithinjosepkl Oct 27, 2023
5b7407d
Perftest: Add missing newline characters for error messages
ddmatsu Nov 1, 2023
0b5bfb6
Perftest: Print an error message to stderr
ddmatsu Nov 1, 2023
8c0f1cc
Minor update to error log
jithinjosepkl Nov 10, 2023
c1a401f
Merge pull request #227 from jithinjosepkl/patch-1
HassanKhadour Nov 10, 2023
8eef5cb
Update perftest_resources.c
ecjtusbs Nov 20, 2023
4d05679
Merge pull request #228 from ddmatsu/newlines
HassanKhadour Nov 27, 2023
5607b0b
Merge pull request #229 from ecjtusbs/master
HassanKhadour Nov 27, 2023
c09edf9
Change index to the right data type
HassanKhadour Nov 27, 2023
dffd1dd
perftest: Add minimum rx_depth size in case of UD qp type
HassanKhadour Dec 18, 2023
fd329b1
Add option for write-with-immediate verb for write_lat test
rauteric Nov 15, 2023
d977590
Add write-with-immediate option for write_bw test
rauteric Dec 10, 2023
d18f441
Fix wrong-result bug in write_bw-with-imm duration mode
rauteric Dec 20, 2023
dcfcaa0
Perftest: Version increase to 6.18
HassanKhadour Dec 25, 2023
45eb075
Perftest: changing spec file version to 24.01.0.
HassanKhadour Dec 25, 2023
3665d98
Merge pull request #230 from rauteric/lat_write_imm_simplified
HassanKhadour Dec 25, 2023
a43ab3c
feat: add more valid RATE_VALUES
FujiZ Dec 26, 2023
5ff3afa
Perftest: fix completion count with special combination of tx_depth/c…
sshaulnv Dec 28, 2023
0da9a02
Fix Neuron dmabuf uninitialized offset
mrgolin Dec 31, 2023
f9e82f7
Make output JSON standard
blochl Jan 2, 2024
9afcc72
Merge pull request #239 from blochl/lb-dev2
HassanKhadour Jan 3, 2024
22ac823
Merge pull request #237 from mrgolin/fix-neuron-offset
HassanKhadour Jan 3, 2024
aadec43
Merge pull request #235 from FujiZ/add-rate
HassanKhadour Jan 3, 2024
d34bf0f
Display a warning if BW peak measurement was disabled implicitly
blochl Jan 8, 2024
7fa7374
Merge pull request #238 from blochl/lb-dev
HassanKhadour Jan 8, 2024
5191f3e
perftest: Add Broadcom gen p7 adapter device ids
selvintxavier Jan 11, 2024
ecab99a
Merge pull request #242 from selvintxavier/genp7_devid
HassanKhadour Jan 16, 2024
8f441d3
Perftest: define use_write_with_imm_flag outside HAVE_AES_XTS ifdef
HassanKhadour Jan 16, 2024
def27db
Perftest: Version increase to 6.19
HassanKhadour Jan 16, 2024
d414c55
Revert "perftest: Add minimum rx_depth size in case of UD qp type"
HassanKhadour Jan 17, 2024
5039890
Add support for loongarch64.
cnmushiba Jan 16, 2024
afe39a9
Perftest: Version increase to 6.20
HassanKhadour Jan 17, 2024
4a83eea
add support for ROCm6.0 API changes
edgargabriel Jan 18, 2024
ada1a29
add support for ROCm6.0 API changes (II)
edgargabriel Jan 23, 2024
921730a
Both the server and client call rdma_disconnect() synchronously, When…
Anumula-Murali-Mohan-Reddy Jan 24, 2024
248d1b7
Fix man page handling for out-of-tree builds
chuckcranor Feb 14, 2024
7044f45
Merge pull request #244 from cnmushiba/la64-dev
HassanKhadour Feb 15, 2024
f6fdf37
Merge pull request #245 from edgargabriel/topic/rocm-6.0-support
HassanKhadour Feb 15, 2024
05317fd
Merge pull request #246 from Anumula-Murali-Mohan-Reddy/master
HassanKhadour Feb 15, 2024
6f30da3
Merge pull request #248 from chuckcranor/master
HassanKhadour Feb 15, 2024
f5b054f
Perftest: Version increase to 6.21
HassanKhadour Feb 15, 2024
c3ecbd8
Perftest: Add condition to force stop polling in write_with_imm
HassanKhadour Apr 10, 2024
7e34f40
Perftest: changing spec file version to 24.04.0.
HassanKhadour Feb 15, 2024
366b209
Fix type in print
cliffburdick Mar 26, 2024
fe1d7aa
Ignore generated config files from git
agirault Mar 14, 2024
9ed5637
Fix out-of-tree man page build when top_srcdir is an absolute path
chuckcranor Feb 20, 2024
592e09e
Fix Integer overflow issue
sshaulnv Feb 26, 2024
ed44d77
Support RDMA with Tegra integrated GPUs
agirault Mar 14, 2024
1d202bb
Display proper unit for MiB/sec results
agirault Mar 19, 2024
2cdf25b
Perftest: changing spec file version to 24.07.0.
HassanKhadour Apr 8, 2024
8bbfe3c
Perftest: Version increase to 6.22
HassanKhadour Apr 8, 2024
1a11412
Perftest: Fix indentation between results after the fix of the BW unit
HassanKhadour Apr 8, 2024
b61a9e7
Fix format of man pages
Apr 9, 2024
6ab70de
Fix syntax of script run_perftest_loopback
Apr 9, 2024
7c282ef
Implement get_cycles for hppa architecture
Apr 9, 2024
8867689
Perftest: Fix rx_depth check for XRC
May 24, 2024
cdc9e27
Perftest: Support selecting congestion control algorithms
Jun 28, 2024
2e3aa2f
perftest: Add minimum rx_depth size in case of SRQ and UD/UC qp type
sshaulnv Jun 30, 2024
febba14
Add support for Cambricon devices
Jun 13, 2024
463300b
Add 0xefa3 pciid to the database
mrgolin Jul 1, 2024
2e88853
Perftest: add error message for DC runs with small queue depth
sshaulnv Aug 11, 2024
b6f957f
Perftest: Add support for TD lock-free mode
Aug 13, 2024
bbb237d
Perftest: Version increase to 6.23
sshaulnv Aug 20, 2024
ccae524
perftest: Set Ack timeout for rdma_cm connection id
Aug 1, 2024
8f87972
Perftest: changing spec file version to 24.10.0
sshaulnv Aug 27, 2024
d7989b2
Perftest: Fix TD lock-free mode not working for QP
Aug 19, 2024
279d92e
Write cuda device id to json file
shaulerez Aug 20, 2024
ad2a685
Add support for SRD unsolicited write w/ imm. receive
mrgolin Aug 28, 2024
fba7ce7
Perftest: Fix failure in creating cq when create cq ex is not support…
Aug 30, 2024
2fb0e05
Merge pull request #286 from hginjgerx/td
sshaulnv Sep 2, 2024
d2def67
Merge pull request #268 from mrgolin/unsolicited-write-recv
sshaulnv Sep 2, 2024
f136038
create_comm_struct: Copy in user_param->qp_timeout
Sep 5, 2024
b8aa202
Set qp_timeout for create_rdma_cm_connection path
Sep 5, 2024
5c29996
Merge pull request #288 from raphael-s-norwitz/fix-qp-timeout-for-cm
sshaulnv Sep 9, 2024
91fadb5
Perftest: fix qp_timeout with rdma_cm and UD
sshaulnv Sep 16, 2024
01be1f4
Support hl (#1)
sdashevsky-ai Oct 28, 2024
ecaccf0
Perftest: add DDP support
sshaulnv Jul 23, 2024
1ed34b1
perftest: Turn on comp_mask create flag and add DDP indication
sshaulnv Aug 29, 2024
60e87cc
Perftest: Add flag to disable DDP
sshaulnv Sep 1, 2024
b88cad2
Perftest: add a check for the max ooo_recv_wrs_caps
sshaulnv Sep 24, 2024
afc2277
Perftest: Version increase to 6.24
sshaulnv Nov 10, 2024
a765cb2
Perftest: changing spec file version to 25.01.0
sshaulnv Nov 10, 2024
f51c4f0
Perftest: change exit flow in run_infinitely mode
sshaulnv Sep 18, 2024
9616ab9
Perftest: disable infinitely trap over client in SEND
sshaulnv Sep 23, 2024
a21bdbe
Merge pull request #293 from sshaulnv/master
sshaulnv Nov 10, 2024
289c058
Perftest: Add connectionless server support for multicast traffic
sshaulnv Sep 23, 2024
45f9f32
Perftest: Add support for CONNECTX9
sshaulnv Nov 12, 2024
d6e1c86
Use zero size receive buffers for write w/imm
Dec 11, 2024
3f012ee
Merge pull request #301 from mrgolin/write-recv-buf
sshaulnv Jan 5, 2025
aec645e
perftest: support set flow_label list val in GRH with RR method
changchengx Aug 22, 2024
d61f8fb
Merge pull request #279 from changchengx/flow_label
sshaulnv Jan 6, 2025
1926a1f
fix core dump issue when disable HAVE_IBV_WR_API
tianx666 Dec 27, 2024
5ae53ba
dmabuf: Add data-direct option to dmabuf
ShacharKagan Jul 9, 2024
937074b
debian: Rely on dh_shlibdeps for determining the dependencies
bdrung Jan 15, 2025
384add6
debian: add missing build dependency on libibumad-dev and libpci-dev
bdrung Jan 15, 2025
54d8147
debian: Bump debhelper to version 10
bdrung Jan 15, 2025
bba6e13
perftest: update payload_file_path & flow_label format man page
changchengx Jan 20, 2025
f5c3035
Enable dmabuf to ROCm
paklui Jan 15, 2025
9c7235c
fix some build and linker issues
paklui Jan 30, 2025
2d6ce96
Fixed issue with loss of one payload part due to skip of one address
Feb 6, 2025
43a5e4f
fix compilation warnings
paklui Feb 12, 2025
d125f1e
dmabuf: Update dmabuf flag for data-direct traffic
ShacharKagan Oct 29, 2024
6730e97
Perftest: enable pcie mapping type only if supported by cuda
sshaulnv Feb 19, 2025
6bd09ff
Merge pull request #304 from tianx666/master
sshaulnv Feb 24, 2025
bfd9b40
Merge branch 'linux-rdma:master' into master
sdashevsky-ai Feb 25, 2025
e97f4cf
Fixed CR styling notes
sdashevsky-ai Feb 25, 2025
9f0b86c
Perftest: Fix for loop initial declarations
sshaulnv Feb 26, 2025
b58e46e
Add DMA-BUF option for ROCm in man page
paklui Feb 26, 2025
70d3aa2
Merge pull request #309 from paklui/master-dmabuf-rocm
sshaulnv Feb 27, 2025
2465f54
Perftest: Fix Cppcheck warnings in rocm_memory.c
sshaulnv Feb 27, 2025
02bb98c
Fix size_t formatting
sdashevsky-ai Mar 2, 2025
7131375
Merge pull request #307 from bdrung/debian-dependencies
sshaulnv Mar 2, 2025
59e2c61
Merge pull request #292 from sdashevsky-ai/master
sshaulnv Mar 3, 2025
27acd3a
Properly handle the error case when running data_direct with binaries…
drossetti Mar 4, 2025
3a37aef
add runtime check for data direct support in the GPU driver
drossetti Mar 4, 2025
f339bdf
fix typo
drossetti Mar 5, 2025
38acb9d
Perftest: Fix configure variable for data_direct
sshaulnv Mar 6, 2025
ec748de
fix autoconf issue with detection of CUDA_MEM_RANGE_FLAG_DMA_BUF_MAPP…
drossetti Mar 7, 2025
629ae8e
fix another typo
drossetti Mar 7, 2025
32e807a
improve error diagnostic
drossetti Mar 7, 2025
84a5f73
Merge pull request #317 from drossetti/fixes
sshaulnv Mar 25, 2025
fea0734
Perftest: Clarify flag symmetry in app help
sshaulnv Feb 24, 2025
f9e4f50
Perftest: Add IPv6 support to bind_source_ip
sshaulnv Mar 16, 2025
05df6d6
Perftest: Optimize cqe polling batch
sshaulnv Feb 23, 2025
f5adde7
Perftest: Fix TD lock-free mode not working for SRQ/XRC QP
Apr 15, 2025
81b94cf
Merge pull request #310 from changchengx/data_pattern
sshaulnv Apr 21, 2025
1598b00
fix typo and align code format
changchengx Nov 19, 2024
989d224
Perftest: remove is_contig_supported
changchengx Nov 19, 2024
9a0e76a
rename var to align with semantic
changchengx Nov 19, 2024
d59fa4d
remove unnecessary parameter
changchengx Nov 20, 2024
8428b92
Merge pull request #299 from changchengx/part_clean
sshaulnv Apr 21, 2025
800ae2d
Perftest: random buffer initialization optimization
sshaulnv Apr 21, 2025
008457c
Perftest: Do not align SRQ recv length to MTU for hns
Apr 27, 2025
5b86637
Perftest: Adding support for new CUDA memory types
Aug 14, 2024
059911f
Perftest: Add gpu_touch flag to test GPU buffer accesses interference
Jul 29, 2024
a4bd71e
Perftest: Add OpenCL memory type support
Jul 29, 2024
89aafcf
Perftest: Adding CUDA support for gpu_touch and lunching CUDA kernels
Sep 10, 2024
a46bfac
Perftest: Adding CCFLAGS and LDFLAGS to load libcudart
RoeyAzran1992 Mar 24, 2025
b8900f8
fixing code review comments
RoeyAzran1992 Apr 20, 2025
f5010dc
Bugfix - using the same stop flag for all running GPU kerenls (redund…
RoeyAzran1992 Apr 27, 2025
5588860
Perftest: optimize buffer init with 32-bit writes
sshaulnv Apr 27, 2025
c17219b
initializing stop touch GPU mem indicator only when touching flag is set
RoeyAzran1992 Apr 28, 2025
22e7c7f
code review - fixing printing type
RoeyAzran1992 Apr 29, 2025
c04922f
Merge pull request #319 from RoeyAzran1992/master
sshaulnv May 6, 2025
045b162
Cuda: Use pcie mapping regardless of data direct
dkkranz May 8, 2025
41d8dba
Perftest: Version increase to 6.25
sshaulnv Feb 24, 2025
446dea7
Perftest: changing spec file version to 25.04.0
sshaulnv Feb 24, 2025
9ab7ec0
Perftest: explicitly include standard C++ library
sshaulnv May 11, 2025
6a3ddbe
Perftest: Make cudart linkage optional due to nvcc and GCC incompatib…
sshaulnv May 12, 2025
ff45fd4
Perftest: add cudart support info to README
sshaulnv May 13, 2025
d3330e5
Merge pull request #312 from TamaraBabayan/master
sshaulnv May 20, 2025
766c13b
Perftest: Fix perform warm up process stuck
Apr 15, 2025
640b064
Merge pull request #323 from hginjgerx/td
sshaulnv May 20, 2025
0b26bc3
Merge pull request #324 from hginjgerx/srq
sshaulnv May 20, 2025
abc99f2
Merge pull request #325 from dkkranz/use_pcie_mapping
sshaulnv May 26, 2025
0bdfc10
Add support for DMA-buffers in Cambricon devices
Nov 7, 2024
0490645
Perftest supports MLU latency tests with read/send verbs only
Sep 20, 2024
9569a3f
Perftest: Add GitHub actions support
sshaulnv May 27, 2025
64330bb
Merge pull request #329 from sshaulnv/master
sshaulnv May 27, 2025
f961e40
Perftest: Add null-mr support over server side
sshaulnv May 27, 2025
ae68798
Added TCU support
iyangsj May 29, 2025
2af4110
add Yunsilicon dev types
tianx666 May 29, 2025
4d645d4
Merge pull request #320 from hc235280/support_ib_send_lat
sshaulnv Jun 15, 2025
ce4f20c
Merge pull request #330 from iyangsj/master
sshaulnv Jun 16, 2025
e503581
Merge branch 'master' into master
tianx666 Jun 17, 2025
9600494
Merge pull request #331 from tianx666/master
sshaulnv Jun 17, 2025
cd69ae0
Perftest: disable Scatter2CQE with gpudirect send_lat
sshaulnv Jun 16, 2025
920087f
Perftest: fix OOO_RECV_WRS enablement flow
sshaulnv Jun 19, 2025
5be2b4e
Perftest: Add print in case blueflame is not supported
maorgottlieb May 20, 2025
4cef9b5
Perftest: Version increase to 6.26
sshaulnv May 19, 2025
14ae7a0
Perftest: changing spec file version to 25.07.0
sshaulnv May 19, 2025
c543c84
Perftest: Dynamic CUDA linking
sshaulnv May 12, 2025
89d0d54
feat(perftest): add --report-min-bw=X to measure the min bandwidth ov…
antgun42 Jun 12, 2023
c5c58ba
feat(perftest): report_min_bw_cycles must be uint64 to prevent wrappi…
antgun42 Jun 12, 2023
97510f1
fix(perftest): more simple and robust logic for measuring batch duration
antgun42 Jun 16, 2023
d21ca8d
feat(perftest): add report-min-bw to man page
antgun42 Jun 16, 2023
0adb524
feat(perftest): check dependencies for report-min-bw
antgun42 Jun 16, 2023
c7e9508
Merge branch 'master' into ly/merge-min-bw-from-downstream
liayan Aug 21, 2025
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions src/perftest_parameters.c
Original file line number Diff line number Diff line change
Expand Up @@ -604,6 +604,7 @@ static void usage(const char *argv0, VerbType verb, TestType tst, int connection
printf(" --run_infinitely ");
printf(" Run test forever, print results every <duration> seconds (SYMMETRIC)\n");


printf(" --report-min-bw=<sample iterations>\n");
printf(" Sample minimum bandwidth over X iterations\n");
}
Expand Down Expand Up @@ -4127,6 +4128,7 @@ void print_report_bw (struct perftest_parameters *user_param, struct bw_report_d
} else {
my_bw_rep->bw_min = 0;
}

if (!user_param->duplex || ((user_param->verb == SEND || user_param->verb == WRITE_IMM) && user_param->test_type == DURATION)
|| user_param->test_method == RUN_INFINITELY || user_param->connection_type == RawEth)
print_full_bw_report(user_param, my_bw_rep, NULL);
Expand Down Expand Up @@ -4250,6 +4252,7 @@ static void write_test_info_to_file(int out_json_fds, struct perftest_parameters
if (user_param->report_min_bw) {
dprintf(out_json_fds, "report_min_bw: %d\n",user_param->report_min_bw);
}

dprintf(out_json_fds, "\n},\n");
}

Expand Down
3 changes: 3 additions & 0 deletions src/perftest_parameters.h
Original file line number Diff line number Diff line change
Expand Up @@ -193,6 +193,9 @@
#define USEC "usec"
/* The format of the results */

#define RESULT_FMT " #bytes #iterations BW peak[MB/sec] BW average[MB/sec] MsgRate[Mpps] BW min[MB/sec]"
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This seems redundant



#define RESULT_FMT " #bytes #iterations BW peak[MB/sec] BW average[MB/sec] MsgRate[Mpps] BW min[MB/sec]"

#define RESULT_FMT_PER_PORT " #bytes #iterations BW peak[MiB/sec] BW average[MiB/sec] MsgRate[Mpps] BW Port1[MiB/sec] MsgRate Port1[Mpps] BW Port2[MiB/sec] MsgRate Port2[Mpps]"
Expand Down
You are viewing a condensed version of this merge commit. You can view the full changes here.