Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Update nroff-generated man pages #9930

Merged
merged 1 commit into from
Mar 21, 2024
Merged

Update nroff-generated man pages #9930

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
32 changes: 31 additions & 1 deletion man/man7/fi_cxi.7
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
.\"t
.\" Automatically generated by Pandoc 2.9.2.1
.\"
.TH "fi_cxi" "7" "2024\-02\-01" "Libfabric Programmer\[cq]s Manual" "#VERSION#"
.TH "fi_cxi" "7" "2024\-03\-21" "Libfabric Programmer\[cq]s Manual" "#VERSION#"
.hy
.SH NAME
.PP
Expand Down Expand Up @@ -269,6 +269,19 @@ application processes.
.IP "7." 3
Application processes select from the list of available service IDs and
VNIs to form an authorization key to use for Endpoint allocation.
.SS Endpoint Protocols
.PP
The provider supports multiple endpoint protocols.
The default protocol is FI_PROTO_CXI and fully supports the messaging
requirements of parallel applicaitons.
.PP
The FI_PROTO_CXI_RNR endpoint protocol is an optional protocol that
targets client/server environments where send-after-send ordering is not
required and messaging is generally to pre-posted buffers; FI_MULTI_RECV
is recommended.
It utilizes a receiver-not-ready implementation where
\f[I]FI_CXI_RNR_MAX_TIMEOUT_US\f[R] can be tuned to control the maximum
retry duration.
.SS Address Vectors
.PP
The CXI provider supports both \f[I]FI_AV_TABLE\f[R] and
Expand Down Expand Up @@ -514,6 +527,16 @@ Using Pinned mode avoids any overhead due to network page faults but
requires all buffers to be backed by physical memory.
Copy-on-write semantics are broken when using pinned memory.
See the Fork section for more information.
.PP
The CXI provider supports DMABUF for device memory registration.
If the ROCR and CUDA libraries support it, the CXI provider will default
to use DMA-buf.
There may be situations with CUDA that may double the BAR consumption.
Until this is fixed in the CUDA stack, the environment variable
\f[I]FI_CXI_DISABLE_DMABUF_CUDA\f[R] can be used to fall back to the
nvidia peer-memory interface.
Also, \f[I]FI_CXI_DISABLE_DMABUF_ROCR\f[R] can be used to fall back to
the amdgpu peer-memory interface.
.SS Translation Cache
.PP
Mapping a buffer for use by the NIC is an expensive operation.
Expand Down Expand Up @@ -1300,6 +1323,13 @@ queue becomes empty.
Default VNI value used only for service IDs where the VNI is not
restricted.
.TP
\f[I]FI_CXI_RNR_MAX_TIMEOUT_US\f[R]
When using the endpoint FI_PROTO_CXI_RNR protocol, this setting is used
to control the maximum time from the original posting of the message
that the message should be retried.
A value of 0 will return an error completion on the first RNR ack
status.
.TP
\f[I]FI_CXI_EQ_ACK_BATCH_SIZE\f[R]
Number of EQ events to process before writing an acknowledgement to HW.
Batching ACKs amortizes the cost of event acknowledgement over multiple
Expand Down
Loading