Is test xfs/096 correct?

[Brian Foster]

On Fri, Dec 04, 2015 at 01:20:26PM +0100, Jan Tulak wrote:
> Hi.
> 
> I'm looking on test xfs/096 and I'm not sure if I got it right:
> 
> "test out mkfs_xfs output on IRIX/Linux and some of its error handling,
> ensure pv#920679 is addressed" - this, and things like "$max_lr_size +
> 4096" all looks like mkfs should be catching invalid input. Yet the .out
> file instead looks like it should create the FS correctly (it contains the
> created fs stats instead of mkfs's usage and some error).
> 

max_lr_size refers to the maximum log record size (256k). When a larger
log stripe unit is passed, mkfs warns about it and adjusts to the
default:

$ mkfs.xfs -f -l su=266240 ./tmp
log stripe unit (266240 bytes) is too large (maximum is 256KiB)
log stripe unit adjusted to 32KiB
meta-data=./tmp                  isize=256    agcount=4, agsize=655360
blks
         =                       sectsz=512   attr=2, projid32bit=1
         =                       crc=0        finobt=0
data     =                       bsize=4096   blocks=2621440, imaxpct=25
         =                       sunit=0      swidth=0 blks
naming   =version 2              bsize=4096   ascii-ci=0 ftype=0
log      =internal log           bsize=4096   blocks=2560, version=2
         =                       sectsz=512   sunit=8 blks, lazy-count=1
realtime =none                   extsz=4096   blocks=0, rtextents=0

The mkfs_filter() function in xfs/096 filters out the warning messages
(iirc, the warning is a relatively recent addition), so the test output
file expects the typical (filtered) mkfs output.

Brian

> So either I'm reading the test wrong, or the patch approves invalid
> behaviour. I hit this test because when I added a stricter input
> validation, this patch started to fail as mkfs is now refusing to create
> the fs with these arguments.
> 
> BTW: I tried to look for the mentioned pv number, but didn't found it -
> where should I look further?
> ​Cheers
> Jan​
> 
> -- 
> Jan Tulak
> jtulak at redhat.com / jan at tulak.me

> _______________________________________________
> xfs mailing list
> xfs at oss.sgi.com
> http://oss.sgi.com/mailman/listinfo/xfs