Solaris newfs pukes when trying to calculate the numbers to create a very large partition. We discovered this when trying to make a 550G RAID into a single filesystem. 

bash-2.03$ sudo newfs -v -f 8192 -i 16384 /dev/rdsk/c1t2d0s3
newfs: /dev/rdsk/c1t2d0s3 last mounted as /export
newfs: construct a new file system /dev/rdsk/c1t2d0s3: (y/n)? y
mkfs -F ufs /dev/rdsk/c1t2d0s3 1131661440 127 255 8192 8192 251 1 90 16384 t 0 -1 8 127
Cylinder groups must have a multiple of 16 cylinders with the given parameters
Rounded cgsize up to 256
Warning: insufficient space in super block for
rotational layout tables with nsect 127, ntrack 255, and nrpos 8.
Omitting tables - file system performance may be impaired.
/dev/rdsk/c1t2d0s3:     1131661440 sectors in 34944 cylinders of 255
tracks, 127 sectors
        552569.1MB in 2184 cyl groups (16 c/g, 253.01MB/g, 16064 i/g)
The work around for this problem is bypass newfs and invoke mkfs directly. First, run prtvtoc to get the number of blocks in the partition, then, run mkfs as follows: 
bash-2.03$ sudo prtvtoc /dev/rdsk/c1t2d0s3
* /dev/rdsk/c1t2d0s3 partition map
*
* Dimensions:
*     512 bytes/sector
*     127 sectors/track
*     255 tracks/cylinder
*   32385 sectors/cylinder
*   35877 cylinders
*   35875 accessible cylinders
*
* Flags:
*   1: unmountable
*  10: read-only
*
*                          First     Sector    Last
* Partition  Tag  Flags    Sector     Count    Sector  Mount Directory
       0      2    00    8420100  20985480  29405579   
       1      3    01          0   8420100   8420099
       2      5    01          0 1161811875 1161811874
       3      7    00   29405580 1132406295 1161811874
bash-2.03$ 
bash-2.03$ sudo mkfs -F ufs -o \
nsect=64,ntrack=127,bsize=8192,fragsize=1024,cgsize=13,free=1,rps=90,nbpi=8256,opt=t,apc=0,gap=0,nrpos=8,maxcontig=64 \
/dev/rdsk/c1t2d0s3 1132406295

Notes: