On Mon, Nov 27, 2023 at 4:21 AM Paul E Luse <paul.e.luse@xxxxxxxxxxxxxxx> wrote: > > On Sun, 26 Nov 2023 22:44:45 -0800 > Christoph Hellwig <hch@xxxxxxxxxxxxx> wrote: > > > Hi Shushu, > > > > the work certainly looks interesting! > > > > However: > > > > > Optimized by using fine-grained locks, customized data structures, > > > and scattered address space. Achieves significant improvements in > > > both throughput and latency. > > > > this is a lot of work for a single Linux patch, we usually do that > > work pice by pice instead of complete rewrite, and for such > > signigicant changes the commit logs also tend to be a bit extensive. > > > > I'm also not quite sure what scattered address spaces are - I bet > > reading the paper (I plan to get to that) would explain it, but it > > also helps to explain the idea in the commit message. > > > > That's my high level nitpicking for now, I'll try to read the paper > > and the patch in detail and come back later. > > > > > > For sure the paper provides a lot more context. Is there more > performance data avialble, like a general sweep of various IO sizes, > patterns and queue depths? ALso, what kind of data integrity testing > has been done? Echo comments from Christoph and Paul. Also, please run ./scripts/checkpatch.pl on the patches before sending. It will catch a lot of issues. Please also clean up the code before sending the patches. Something like the following should not be included in the patch. + // new_bmclocks = kcalloc(num_bmclocks, sizeof(*new_bmclocks), GFP_KERNEL); Thanks, Song
