File system blocks alignment
scottl at samsco.org
Sun Jan 3 06:36:55 UTC 2010
On Jan 2, 2010, at 3:37 PM, Poul-Henning Kamp wrote:
> In message <4B3FC8A2.1090901 at elischer.org>, Julian Elischer writes:
>> Poul-Henning Kamp wrote:
>>> You overlook that MBR/Fdisk requires bootable slices to start at a
>>> "track". That means that the propper slice-alignmen typically
>>> will be 8*63=504 sectors.
>> No it doesn't, (or at least it didn't) but it has become custom to
>> do so.
> Yes it does, for all slices not starting on the first head.
> We've been over this maddness in the past multiple times.
Julian is correct, LBA and packet mode has left the CHS madness as
nothing more than a historical blemish. For many many years, we set
up our FreeBSD systems at Yahoo to create a bootable Slice 1 at sector
512, and it worked just fine. Given that we use a multitude of
vendors and BIOSes, our sample size is fairly significant. IIRC, BIOS
vendors switched to packet mode around 2001-2002, and maybe as late as
2003 for some late adopters, but certainly no later than that. That
was 8-9 years ago, roughly half of FreeBSD's present age. I don't
know for sure, but I'd bet that packet mode has been a WHQL item since
at least WinXP, and thus pretty much guaranteed to be present in the
vast majority of operational hardware out there.
Despite all of this, I do recognize and appreciate your concern for
compatibility. That's why I really like John Baldwin's hybrid MBR+GPT
solution, as it appears to bridge compatibility while breaking away
from the CHS and MBR constraints. We switched to it at Yahoo several
years ago, and it also works very well for us.
More information about the freebsd-arch