Re: mkfs.xfs ignores data stripe on 4k devices?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



That should have been "ignores data device stripe geometry for an internal log ..."

On 12/6/21 1:16 PM, Eric Sandeen wrote:
This seems odd, and unusual to me, but it's been there so long I'm wondering
if it's intentional.

We have various incarnations of this in mkfs since 2003:

         } else if (lsectorsize > XFS_MIN_SECTORSIZE && !lsu && !lsunit) {
                 lsu = blocksize;
                 logversion = 2;
         }

which sets the log stripe unit before we query the device geometry, and so
with the log stripe unit already set, we ignore subsequent device geometry
that may be discovered:

# modprobe scsi_debug dev_size_mb=1024 opt_xferlen_exp=10 physblk_exp=3

# mkfs.xfs -f /dev/sdh
meta-data=/dev/sdh               isize=512    agcount=8, agsize=32768 blks
          =                       sectsz=4096  attr=2, projid32bit=1
          =                       crc=1        finobt=1, sparse=1, rmapbt=0
          =                       reflink=1    bigtime=0 inobtcount=0
data     =                       bsize=4096   blocks=262144, imaxpct=25
          =                       sunit=128    swidth=128 blks
                                  ^^^^^^^^^
naming   =version 2              bsize=4096   ascii-ci=0, ftype=1
log      =internal log           bsize=4096   blocks=2560, version=2
          =                       sectsz=4096  sunit=1 blks, lazy-count=1
                                               ^^^^^^^^^^^^
realtime =none                   extsz=4096   blocks=0, rtextents=0

surely this is unintentional and suboptimal? But please sanity-check me,
I don't know how this could have stood since 2003 w/o being noticed...

Thanks,
-Eric





[Index of Archives]     [XFS Filesystem Development (older mail)]     [Linux Filesystem Development]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux RAID]     [Linux SCSI]


  Powered by Linux