Did some tests with a 12TB disk and I think I found the issue, or at least what's causing it.
Note that large disks formatted with v6.9.x or below don't have this issue after upgrading to v6.10, problem is with large disks formatted in the array with v6.10, looks like it's caused by the MD driver showing the disk as:
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 4096 bytes / 131072 bytes
vs v6.9
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
actual disk values
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
With v6.9.x and earlier the physical sector size also shows as 512 and there's no problem, but the I/O size is also 512 unlike v6.10, where I/O sizes seem wrong, and likely the reason for not getting errors with disks formatted with v6.9 on v6.10.
Output of xfs_growfs, note the agcount and agsize differences, also sunit and swidth, not sure if those are relevant (also sector size when formatting without the MD driver but that's expected).
Disk formatted using the MD driver in v6.9, output with v6.9 or v6.10:
meta-data=/dev/md1 isize=512 agcount=11, agsize=268435455 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1
data = bsize=4096 blocks=2929721331, imaxpct=5
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=521728, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
Disk formatted using the MD driver in v6.10:
meta-data=/dev/md1 isize=512 agcount=32, agsize=91553791 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1 bigtime=0 inobtcount=0
data = bsize=4096 blocks=2929721312, imaxpct=5
= sunit=1 swidth=32 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=521728, version=2
= sectsz=512 sunit=1 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
xfs_growfs: XFS_IOC_FSGROWFSDATA xfsctl failed: No space left on device
Disk formatted without the MD driver in v6.10:
meta-data=/dev/sdb1 isize=512 agcount=11, agsize=268435455 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=2929721331, imaxpct=5
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=521728, version=2
= sectsz=4096 sunit=1 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
Only disks formatted using the MD driver in v6.10 display the "xfs_growfs: XFS_IOC_FSGROWFSDATA xfsctl failed: No space left on device" error (and experimental warning) after running xfs_growfs, so this looks to me like an Unraid issue, XFS maintainers unlikely to investigate this if the problem only manifests itself after formatting the disk with the out of tree MD driver.