Mikhail Brinskiy
Mikhail Brinskiy
> @brminich how was this issue found and can be reproduced? Is this ep reconfiguration issue? does it mean AM lane will use 2nd path now? I got an assert...
> @brminich the following still fails with this PR, is it expected? > No, will check it
> @brminich the following still fails with this PR, is it expected? > > ``` > $ taskset -c 18,19 make -C build-devel/test/gtest test GTEST_FILTER=rcx/test_ucp_am_nbx_seg_size.single/7 GTEST_REPEAT=1000 > ... I can...
> @brminich the following still fails with this PR, is it expected? > @yosefe fixed in #8472
@yosefe, force-pushed just a single fix for am lane reusage by BW lanes
@yosefe, updated in place, because PR is small enough
> maybe "inter" or "ipc" ? we already have `rkey` (stands for **r**emote), which is similar to inter. ipc is better in this sense
@dqwu, is there any error like `Transport retry count exceeded on ... `? Also are these errors seen on the specific nodes, or the failing nodes can vary from run...
@rakhmets, can you please review?