Hi Muchun,

> Thank you very much for the test, so in your scenario, this is also
> the behavior of your real workload (mainly read/write-based, rather
> than mmap), right?

Yes. We mount read-only rootfs images over virtio-pmem and FS-DAX. Most
file access goes through normal read/write. mmap is only used for
loading binaries and shared libraries at startup, which is a small part
of the image.

> To be fair, it does fix the issue, but it's not very clean. I might
> do some code cleanup first to solve this more gracefully. I want to
> move the initialization to where the template page is allocated,
> similar to what section-based vmemmap optimization does.

Makes sense. Initializing it at allocation time is cleaner.

> Thanks for the analysis and suggestions. However, the reason I
> haven't sent out v2 yet is that I've been trying to refactor the
> vmemmap optimization logic to make FS_DAX support cleaner. I think
> I'll probably do some code refactoring first.

Understood. It would also help if the refactor could support subsection
(2 MiB) alignment, so any 2 MiB-aligned device can use the optimization
without padding to a 128 MiB multiple.

> I'cc you next time if you'd like to help me test.

Sounds good. Please Cc me when you post v2 and I'll be glad to test.

Best regards,
Yilei

Reply via email to