In article <10782.1287539...@splode.eterna.com.au>,
matthew green <m...@eterna.com.au> wrote:
>
>hi folks.
>
[:stuff deleted:]
>
>i have a slightly different version of this in testing for netbsd-5
>as well (netbsd-5 has no getdisksize(9)) and will send pullups for
>that when i'm more confident. you can find that patch at:
>
> http://www.netbsd.org/~mrg/rf-2tb-netbsd-5.diff
>
>for anyone extremely keen.
Thank, great stuff. I have to go and buy myself some disks now!
christos
>+ if (c_label->version == RF_COMPONENT_LABEL_VERSION)
>+ c_label->partitionSizeHi =
>+ raidPtr->Disks[fcol].partitionSize >> 32;
All these tests should really be:
+ if (c_label->version > RF_COMPONENT_LABEL_VERSION_1)
+ c_label->partitionSizeHi =
+ raidPtr->Disks[fcol].partitionSize >> 32;
because if you bump the version to version 3, then you will
not execute the code for version 2 which already supported it.
>! blocknum = bp->b_blkno << DEV_BSHIFT >>
>raidPtr->logBytesPerSector;
This should be a macro:
#define BLOCK_TO_SECTOR(block, bytes_per_sector) \
(((block) << DEV_BSHIFT) >> (bytes_per_sector))
>! bp->b_blkno = startSect << logBytesPerSector >> DEV_BSHIFT;
And this:
#define SECTOR_TO_BLOCK(sector, bytes_per_sector) \
(((sector) << (bytes_per_sector)) >> DEV_BSHIFT)
christos
