[PATCH] [powerpc] update xmon slb code
Olof Johansson
olof at lixom.net
Tue Oct 30 07:52:54 EST 2007
On Mon, Oct 29, 2007 at 02:59:27PM -0500, Will Schmidt wrote:
>
> [powerpc] update xmon slb code
>
> adds a bit more detail to the xmon SLB output. When the valid bit is
> set, This displays the ESID and VSID values, as well as decoding the
> segment size. (1T or 256M). This supresses the output for any slb entries
> that contain only zeros.
>
> I debated a bit on whether to check for just (valid) versus checking for
> (valid|esid|vsid). By inspection on power5, I do have SLB entries that
> contain values but without the valid bit set, so opted to display any
> non-zero values.
Yeah, newer versions of the architecture specify that invalid entries
must read as 0, while POWER5 doesn't. Printing them doesn't hurt.
> sample output from power6 (1T segment support):
>
> 00 c000000008000000 40004f7ca3000500 1T ESID=c00000 VSID=40004f7ca3
> 01 d000000008000000 4000eb71b0000400 1T ESID=d00000 VSID=4000eb71b0
> 24 cf00000008000000 400011b260000500 1T ESID=cf0000 VSID=400011b260
> 25 0000040008000000 4000a9e949000c80 1T ESID=4 VSID=4000a9e949
> 26 0000000018000000 00005e93bfd49c80 256M ESID=1 VSID=5e93bfd49
> 27 00000f0008000000 4000e262a4000c80 1T ESID=f VSID=4000e262a4
> 28 0000000008000000 00005dd45172ec80 256M ESID=0 VSID=5dd45172e
>
> sample output from power5 (notice the non-valid but non-zero entries)
>
> 54 0000000048000000 0000cf33bb059c80 256M ESID=4 VSID=cf33bb059
> 55 0000000018000000 0000ccf56fe08c80 256M ESID=1 VSID=ccf56fe08
> 56 0000000010000000 0000dd82ce799c80
> 57 cf00000008000000 0000d59aca40f500 256M ESID=cf0000000 VSID=d59aca40f
> 58 c000000078000000 000045cb97751500 256M ESID=c00000007 VSID=45cb97751
> 59 0000040000000000 000061552db1bc80
>
> Tested on power5 and power6.
Nice, I like it! I wonder if it would make sense to (space) pad the
ESID/VSID fields so they line up, it'd make output just a little tidier.
(If you end up changing that, please also break the long printf lines
in two.)
Beside that it looks good to me! :)
-Olof
More information about the Linuxppc-dev
mailing list