[PATCH] rhel4.8 lvm rounding up pv problem

[Joel Granados Moreno <jgranado redhat com>]

Hi list:

About 480793.  It seems that rhel4.8 is doing something strange when it
rounds up the lv size in the lvgrow function.  This patch seems to fix
the traceback seen in the bug.  As I see it, the cause of the problem
was that the grow function was clamping the volume group size with
lvm.clampPVSize, which is the right thing to do.  But it did not
lvm.clampLVSizeRequest at the end of the calculation.  The result, a
"missing one pe" problem.

Tested on my machine.  Pretty sure this will fix the issue.

Review appreciated.

regards.