On 22/02/2012 00:36, Meelis Roos wrote:
Meelis, can you please apply the following patch before& after the
offending commit, boot with "memblock=debug" added as kernel param and
post the boot log? The patch will generate some offset warnings after
the commit but should work fine.
Before the commit (v3.2-rc3-75-g0ee332c): memblock1.gz (attached)
After the commit (v3.2-rc3-76-g7bd0b0f): memblock2.gz (attached)
Its a long time since I regularly had to worry about SPARC boxes (not)
booting so may be the difference between virtual & physical addresses
but I notice that some of the addresses in the register dump have
non-zero values in the upper 32 bits but the memblock values have zero
in the upper half.
memblock reserved: ADD [0x0000007fcc0a40-0x0000007fcc0a4e] node 1
memblock reserved: add [0x0000007fcc0a40-000000007fcc0a4e] node 1 @767
But a similar address in the registers has fffff800 in there.
o4: fffff8007fcc0a4d
I know that there are a number of explanations why things would be
different (32 bit acesses etc) but it could explain things plus we would
be talking 64 bit addresses in the kernel.
Just a thought.
Richard
In addition, a third type of sparc machines breaks in a third way - V210
and V240 just hang after telling
console [tty0] enabled, bootconsole disabled
and before calibrating the delay loop. Bisect has led to the same commit.
--
To unsubscribe from this list: send the line "unsubscribe sparclinux" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
