Alexander, I misstated the question about using dd - it seems to be used to erase a label for testing - but how does clearing blocks of slice 0 accomplish this? Thank you, William
William Schumann wrote: > Alexander, > The code assumes that if prtvtoc fails for any reason, that format(1m) > should write the vtoc. I am not sure how solid an assumption this is, > but it seems reasonable. > > What does the 'dd' command accomplish, zeroing the 1st 16 blocks of > slice 0 as shown in the example below on a volume that has no vtoc? > Please explain how this works. > > The boolean_t disk_is_unlabeled is unnecessary. The code to invoke > format -d to write the label should be within the block in which > disk_is_unlabeled is set. > Since a disk is being labeled, please clearly comment the code and add > a log message stating that the format command was used to label the disk. > > Thank you, > William > > Alexander Eremin wrote: >> I need review for Bug 3136 - TI fails to create VTOC on unlabeled disk >> on Sparc. >> >> webrev : http://cr.opensolaris.org/~alhazred/3136/ >> >> Tests: >> Usual test_ti: >> # dd if=/dev/zero of=/dev/dsk/c0t3d0s0 count=16 >> 16+0 records in >> 16+0 records out >> #./test_ti -c -x 3 -t v -d c0t3d0 >> Test TI started in real mode... >> Target type specified: VTOC >> Config file not specified, default VTOC will be created >> VTOC target prepared successfully >> ERR: creating of VTOC target failed >> >> test_ti_static with fixed libti: >> >> # ./test_ti_static -c -x 3 -t v -d c0t3d0 >> Test TI started in real mode... >> Target type specified: VTOC >> Config file not specified, default VTOC will be created >> VTOC target prepared successfully >> VTOC target created successfully >> >> >> Thanks, >> >> > _______________________________________________ > caiman-discuss mailing list > caiman-discuss at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
