Currently, I more like lockless readdir+, which can much improve "ls -l" performance. I am wonder whether it can be an option for ORNL's project1 or not.
nasf (Inactive)
added a comment - Currently, I more like lockless readdir+, which can much improve "ls -l" performance. I am wonder whether it can be an option for ORNL's project1 or not.
This attachment is the original design with my review comments added.
I'm very happy with the level of detail. I'd like to explore the use
of readdir+ over the wire rather than readdir + / aggregate stat in
more detail.
Eric Barton (Inactive)
added a comment - This attachment is the original design with my review comments added.
I'm very happy with the level of detail. I'd like to explore the use
of readdir+ over the wire rather than readdir + / aggregate stat in
more detail.
Please can you attach a Microsoft Word version of the document so that I can turn on change tracking and edit it.
Eric Barton (Inactive)
added a comment - Please can you attach a Microsoft Word version of the document so that I can turn on change tracking and edit it.
Replaced by
LU-14393.