Page MenuHomeFreeBSD

pms: handle maximum size IO with any alignment
ClosedPublic

Authored by rlibby on Jan 8 2021, 7:31 PM.
Tags
None
Referenced Files
Unknown Object (File)
Mon, Dec 9, 5:20 PM
Unknown Object (File)
Nov 7 2024, 1:59 AM
Unknown Object (File)
Oct 9 2024, 7:55 PM
Unknown Object (File)
Oct 9 2024, 9:29 AM
Unknown Object (File)
Oct 8 2024, 5:39 PM
Unknown Object (File)
Oct 7 2024, 7:50 PM
Unknown Object (File)
Oct 7 2024, 2:42 PM
Unknown Object (File)
Oct 7 2024, 10:57 AM
Subscribers

Details

Summary

Define the maximum numbers of segments to allow for non-page alignment
at the beginning and end of a maxphys size transfer. Also set
ccb_pathinq.maxio consistent with maxphys.

Sponsored by: Dell EMC Isilon

Test Plan

Here, just make buildkernel. Isilon has tested this downstream, but
has not yet caught up to making maxphys tunable.

Diff Detail

Repository
rS FreeBSD src repository - subversion
Lint
Lint Passed
Unit
No Test Coverage
Build Status
Buildable 36039
Build 32928: arc lint + arc unit

Event Timeline

rlibby requested review of this revision.Jan 8 2021, 7:31 PM
sys/dev/pms/freebsd/driver/ini/src/agtiapi.c
1836

Shouldn't this be min(maxphys, 1024*1024)? Or is there no silicon limit?

sys/dev/pms/freebsd/driver/ini/src/agtiapi.c
1836

Thanks for the feedback. I'm looking into whether we have a spec. If we can't come up with a definitive answer, I'll go with your suggestion.

sys/dev/pms/freebsd/driver/ini/src/agtiapi.c
1836

Sounds good. Or whatever value you are able to test. I'm not super particular nor do I want to artificially limit something we know (or have good reason to believe) works.

imp feedback: keep the limit

This revision is now accepted and ready to land.Jan 13 2021, 10:06 PM