Uploaded image for project: 'Lustre'
  1. Lustre
  2. LU-10300

Can the Lustre 2.10.x clients support 64K kernel page?

Details

    • Question/Request
    • Resolution: Cannot Reproduce
    • Major
    • None
    • Lustre 2.10.0
    • None
    • Redhat
    • 9223372036854775807

    Description

      We test Lustre in the kernel 4.11.0 which use 64K memory page, and we found that there are no config items in the configure.

      Attachments

        Issue Links

          is related to

          Bug - A problem which impairs or prevents the functions of the product. LU-12255 sanity-dom test 42e fails with ‘test_42e failed with 1’

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Open

          Bug - A problem which impairs or prevents the functions of the product. LU-4398 mdt_object_open_lock() may not flush conflicting handles

          • Critical - Crashes, loss of data, severe memory leak.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-10073 lnet-selftest test_smoke: lst Error found

          • Critical - Crashes, loss of data, severe memory leak.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-11729 ARM: sanity test_810: BAD WRITE CHECKSUM with adler

          • Critical - Crashes, loss of data, severe memory leak.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-11595 sanity-dom sanityn test 11: LBUG: (file.c:470:ll_dom_finish_open()) ASSERTION( lnb.lnb_file_offset % ((1UL) << 16) == 0 ) failed

          • Major - Major loss of function.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-11596 sanity test_42d/test_42e: FAIL: failed: client:143294464 server: 143110144

          • Major - Major loss of function.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-12014 check correct size in ll_dom_finish_open()

          • Major - Major loss of function.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-15722 IO write gets stuck on some sanityn test cases for 64K PAGE_SIZE

          • Major - Major loss of function.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-11667 sanity test 317: FAIL: Expected Block 8 got 48 for f317.sanity

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-11671 sanity test 45: FAIL: write wasn't cached

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-11787 sanityn test 71a fails with ‘data is not flushed from client’

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-12362 kernel warning 'do not call blocking ops when !TASK_RUNNING ' in ptlrpcd

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-12419 ppc64le: "LNetError: RDMA has too many fragments for peer_ni" when reading two files

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Closed

          Improvement - An improvement or enhancement to an existing feature or task. LU-15223 Improve partial page read/write

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • In Progress
          is related to

          Bug - A problem which impairs or prevents the functions of the product. LU-14970 sanity-flr test_50a: FAIL: Mirrored file consumes 1728 blocks

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Open

          Bug - A problem which impairs or prevents the functions of the product. LU-14346 sanity-pcc test 7a hangs in multiop/down_read

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • In Progress

          Bug - A problem which impairs or prevents the functions of the product. LU-10157 LNET_MAX_IOV hard coded to 256

          • Blocker - Blocks development and/or testing work, production could not run.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-15364 Kernel oops when stripe on Arm64 Server end multiple MDTs

          • Critical - Crashes, loss of data, severe memory leak.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-7650 ko2iblnd map_on_demand can't negotitate when page sizes are different between nodes.

          • Major - Major loss of function.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-11597 sanityn test 16a failed with direct I/O

          • Major - Major loss of function.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-11785 conf-sanity test_98 fails with 'Buffer overflow check failed'

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Resolved

          Bug - A problem which impairs or prevents the functions of the product. LU-11788 sanity test 104a fails with ‘lfs df failed’

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Resolved

          New Feature - A new feature of the product, which has yet to be developed. LU-6387 Add Power8 support to Lustre

          • Major - Major loss of function.
          • Resolved

          New Feature - A new feature of the product, which has yet to be developed. LU-11200 Centos 8 arm64 server support

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Resolved

          New Feature - A new feature of the product, which has yet to be developed. LU-15293 Add lbuild support for latest Arm64 CentOS8

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Resolved

          Activity

            [LU-10300] Can the Lustre 2.10.x clients support 64K kernel page?
            adilger Andreas Dilger added a comment -

            The aarch64 clients are working with any recent Lustre release.

            adilger Andreas Dilger added a comment - The aarch64 clients are working with any recent Lustre release.
            gerrit Gerrit Updater added a comment -

            "James Simmons <jsimmons@infradead.org>" uploaded a new patch: https://review.whamcloud.com/45962
            Subject: LU-10300 test: does arm testing work
            Project: fs/lustre-release
            Branch: master
            Current Patch Set: 1
            Commit: 2036c3298804defa44e0457e65893e91d1da3985

            gerrit Gerrit Updater added a comment - "James Simmons <jsimmons@infradead.org>" uploaded a new patch: https://review.whamcloud.com/45962 Subject: LU-10300 test: does arm testing work Project: fs/lustre-release Branch: master Current Patch Set: 1 Commit: 2036c3298804defa44e0457e65893e91d1da3985
            adilger Andreas Dilger added a comment -

            James, I'd suggest to move your previous comment into a separate ticket related to 64KB PAGE_SIZE on the server, and leave this one for tracking 64KB PAGE_SIZE on the client. I suspect most people care about client-side support more than server-side, and this will help understand which issues are important to fix for the two different cases. In any case, it doesn't make sense to have multiple independent issues being worked on in the same Jira ticket.

            adilger Andreas Dilger added a comment - James, I'd suggest to move your previous comment into a separate ticket related to 64KB PAGE_SIZE on the server, and leave this one for tracking 64KB PAGE_SIZE on the client. I suspect most people care about client-side support more than server-side, and this will help understand which issues are important to fix for the two different cases. In any case, it doesn't make sense to have multiple independent issues being worked on in the same Jira ticket.
            simmonsja James A Simmons added a comment -

            This time I tested ZFS lustre servers using the ethernet interface and it worked. Currently their is a bug in the RHEL7 alt kernel that I'm using in netlink that shows up when using k2oiblnd for some reason. Their appears to be a fix but requires rebuild the RHEL7 alt kernel. Now to test ldiskfs.

            simmonsja James A Simmons added a comment - This time I tested ZFS lustre servers using the ethernet interface and it worked. Currently their is a bug in the RHEL7 alt kernel that I'm using in netlink that shows up when using k2oiblnd for some reason. Their appears to be a fix but requires rebuild the RHEL7 alt kernel. Now to test ldiskfs.
            simmonsja James A Simmons added a comment - - edited

            So I have managed to get Lustre ZFS servers running on Power8 nodes. I got it to mount and then it just locked up with any attempt to use the file system  Well its a start

            simmonsja James A Simmons added a comment - - edited So I have managed to get Lustre ZFS servers running on Power8 nodes. I got it to mount and then it just locked up with any attempt to use the file system  Well its a start
            simmonsja James A Simmons added a comment -

            Once the RHEL ARM/Power8 server support work is complete we can test 64K pages on the server side.

            simmonsja James A Simmons added a comment - Once the RHEL ARM/Power8 server support work is complete we can test 64K pages on the server side.
            adilger Andreas Dilger added a comment -

            It probably makes sense to print a build warning in the server code when PAGE_SIZE isn't 4096, just so people are aware that this isn't being tested. If someone starts testing this in the future, the warning can be removed.

            adilger Andreas Dilger added a comment - It probably makes sense to print a build warning in the server code when PAGE_SIZE isn't 4096, just so people are aware that this isn't being tested. If someone starts testing this in the future, the warning can be removed.
            Red ZhangWei (Inactive) added a comment -

            Yes, I tested the 64KB PAGE_SIZE server with 4KB PAGE_SIZE Client and 64KB PAGE_SIZE Client, there are some errors both in these two clients.Thanks for help !

            Red ZhangWei (Inactive) added a comment - Yes, I tested the 64KB PAGE_SIZE server with 4KB PAGE_SIZE Client and 64KB PAGE_SIZE Client, there are some errors both in these two clients.Thanks for help !
            adilger Andreas Dilger added a comment -

            We definitely used to run Lustre with 64KB PAGE_SIZE on IA64 clients, and in theory this would still work but we haven't tested it in a long time.  We never had much success with 64KB PAGE_SIZE on the server, since this caused problems with 4KB PAGE_SIZE clients doing writes to the 64KB PAGE_SIZE server.

            adilger Andreas Dilger added a comment - We definitely used to run Lustre with 64KB PAGE_SIZE on IA64 clients, and in theory this would still work but we haven't tested it in a long time.  We never had much success with 64KB PAGE_SIZE on the server, since this caused problems with 4KB PAGE_SIZE clients doing writes to the 64KB PAGE_SIZE server.
            Red ZhangWei (Inactive) added a comment -

            Thanks for your replay, with the map_ondemand enabled, can the Lustre only support 4k kernel memory page?

            Red ZhangWei (Inactive) added a comment - Thanks for your replay, with the map_ondemand enabled, can the Lustre only support 4k kernel memory page?

            People

              simmonsja James A Simmons
              Red ZhangWei (Inactive)
              Votes:
              0 Vote for this issue
              Watchers:
              11 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: