Phillip Susi wrote: > On 1/12/2012 10:24 AM, Jim Meyering wrote: >> It was addressed by the change right after the one that added that >> new test: 713322fa88e097b5bcaae36ad5b6a41d1acc6db3 > > My understanding of that patch is that it was supposed to make sure > that the partition geometry got adjusted so that the blkpg request > would not fail. As such, it appears to not be working, and the test > is validating that the patch is not working.
That patch makes it so parted tells the kernel about the slightly larger size so that it will object not just for a simple overlap, but also if we try to create a partition that overlaps those last two sectors.
