Skip to content

mctp-bench: Support recv requests#100

Merged
jk-ozlabs merged 1 commit intoCodeConstruct:mainfrom
mkj:dev/bench-receive-request
Jul 31, 2025
Merged

mctp-bench: Support recv requests#100
jk-ozlabs merged 1 commit intoCodeConstruct:mainfrom
mkj:dev/bench-receive-request

Conversation

@mkj
Copy link
Copy Markdown
Member

@mkj mkj commented Jul 31, 2025

"mctp-bench recv" can now optionally request that a peer start sending with a specified message count and message length.

"mctp-bench recv eid 20 len 987 count 10000"

The send equivalent is not yet implemented. Currently mctp-bench will continue listening forever, since it won't necessarily know that all packets have been received (if drops occur). That could be changed in future.

With default arguments "mctp-bench recv" behaviour is unchanged.

@jk-ozlabs
Copy link
Copy Markdown
Member

PR #100 ! 🎉

Copy link
Copy Markdown
Member

@jk-ozlabs jk-ozlabs left a comment

Choose a reason for hiding this comment

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

Couple of comments inline, but looks good in general.

static const uint8_t COMMAND_VERSION = 1;

enum command {
COMMAND_RESPONSE = 0x00,
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

Will responses to all (future) commands use the same opcode?

(I assume the IID will let us catch mismatched responses, but would a request-specific-response type help to diagnnose?)

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

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

I was intending to use COMMAND_RESPONSE for all, given IID should be reliable.

"mctp-bench recv" can now optionally request that a peer start sending
with a specified message count and message length.

"mctp-bench recv eid 20 len 987 count 10000"

The send equivalent is not yet implemented. Currently mctp-bench will
continue listening forever, since it won't necessarily know that all
packets have been received (if drops occur). That could be changed in
future.

With default arguments "mctp-bench recv" behaviour is unchanged.

Signed-off-by: Matt Johnston <matt@codeconstruct.com.au>
@mkj mkj force-pushed the dev/bench-receive-request branch from 9f15500 to bfbbb34 Compare July 31, 2025 04:56
@jk-ozlabs jk-ozlabs merged commit bfbbb34 into CodeConstruct:main Jul 31, 2025
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants