mbox series

[RFC,0/4] cachestat: a new syscall for page cache state of files

Message ID 20221115182901.2755368-1-nphamcs@gmail.com
Headers
Series cachestat: a new syscall for page cache state of files |

Message

Nhat Pham Nov. 15, 2022, 6:28 p.m. UTC 
  There is currently no good way to query the page cache state of large
file sets and directory trees. There is mincore(), but it scales poorly:
the kernel writes out a lot of bitmap data that userspace has to
aggregate, when the user really doesn not care about per-page information
in that case. The user also needs to mmap and unmap each file as it goes
along, which can be quite slow as well.

This series of patches introduces a new system call, cachestat, that
summarizes the page cache statistics (number of cached pages, dirty
pages, pages marked for writeback, evicted pages etc.) of a file, in a
specified range of bytes. It also include a selftest suite that tests some
typical usage

This interface is inspired by past discussion and concerns with fincore,
which has a similar design (and as a result, issues) as mincore.
Relevant links:

https://lkml.indiana.edu/hypermail/linux/kernel/1302.1/04207.html
https://lkml.indiana.edu/hypermail/linux/kernel/1302.1/04209.html

For comparison with mincore, I ran both syscalls on a 2TB sparse file:

Using mincore:
real    0m37.510s
user    0m2.934s
sys     0m34.558s

Using cachestat:
real    0m0.009s
user    0m0.000s
sys     0m0.009s

Mincore takes about 4000 times longer to obtain less aggregated
information!

Some open questions:

* What other fields might be useful?
* Huge pages: another useful stat to include is the number of huge pages
  cached. However, as the size of a huge page can vary, having just a
  single field is not very meaningful.
* An alternative would be to have one field for each possible size - but
  this is not future-proof, as bigger sizes might be introduced later.

Johannes Weiner (1):
  workingset: fix confusion around eviction vs refault container

Nhat Pham (3):
  workingset: refactor LRU refault to expose refault recency check
  cachestat: implement cachestat syscall
  selftests: Add selftests for cachestat

 MAINTAINERS                                   |   8 +
 arch/x86/entry/syscalls/syscall_32.tbl        |   1 +
 arch/x86/entry/syscalls/syscall_64.tbl        |   1 +
 include/linux/swap.h                          |   1 +
 include/linux/syscalls.h                      |   2 +
 include/uapi/asm-generic/unistd.h             |   5 +-
 include/uapi/linux/mman.h                     |   8 +
 kernel/sys_ni.c                               |   1 +
 mm/Makefile                                   |   2 +-
 mm/cachestat.c                                | 109 +++++++++++
 mm/workingset.c                               | 142 +++++++++-----
 tools/testing/selftests/Makefile              |   1 +
 tools/testing/selftests/cachestat/.gitignore  |   2 +
 tools/testing/selftests/cachestat/Makefile    |   9 +
 .../selftests/cachestat/test_cachestat.c      | 184 ++++++++++++++++++
 15 files changed, 430 insertions(+), 46 deletions(-)
 create mode 100644 mm/cachestat.c
 create mode 100644 tools/testing/selftests/cachestat/.gitignore
 create mode 100644 tools/testing/selftests/cachestat/Makefile
 create mode 100644 tools/testing/selftests/cachestat/test_cachestat.c

--
2.30.2

Comments

Nhat Pham Nov. 16, 2022, 11:18 p.m. UTC | #1 
There are some build errors that has to do with specific arch/config -
but these are not related to the the main logic of the implementation.
Of course, I'll address these in v2 of the patch set, but in the
meantime, please feel free to take a look at this RFC and give me your
thoughts and suggestion.