Matt had asked for an ability for readdir chunking to be able to absorb more than one chunks worth of entries if it made sense for the FSAL to read a larger number of entries in one go.
It turns out that it was very simple to support this. And for gravy, there's an example of FSAL_VFS which increases the getdents buffer size and supports readahead. Along the way, I also realized there was a bug in FSAL_RGW... It may be worth asking if we should make the FSAL_VFS readdir buffer resizable and allow enabling readahead. You can find it in my readahead branch: https://github.com/ffilz/nfs-ganesha/commits/readahead Or review on gerrithub: https://review.gerrithub.io/#/c/355552/ https://review.gerrithub.io/#/c/355563/ https://review.gerrithub.io/#/c/355564/ https://review.gerrithub.io/#/c/355565/ You can also pull the whole branch from gerrithub with this command (which will create and checkout a branch called readahead in your repo): git fetch https://ff...@review.gerrithub.io/a/ffilz/nfs-ganesha refs/changes/65/355565/1 && git checkout FETCH_HEAD -b readahead This last is taken from the download menu in the final patch, you chose the checkout option, cut and paste that. I always add the -b {branchname} to the end of it to actually checkout a branch. This can be used to pull down anyone's patch set even if they don't have a github repo (or you don't know what it is, or they didn't push to it). I use this all the time for doing weekly merges if someone has a patchset with more than 2 or 3 patches. Thanks Frank --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
