linux.git/arch/powerpc, branch v3.1-rc2

powerpc: Really fix build without CONFIG_PCI

2011-08-10T15:15:44+00:00

Brown paper bag day, previous commit wouldn't work very well with modules
enabled. Move the exports into the ifdef.

Signed-off-by: Benjamin Herrenschmidt

powerpc: Fix build without CONFIG_PCI

2011-08-05T06:01:20+00:00

Commit fea80311a939a746533a6d7e7c3183729d6a3faf
"iomap: make IOPORT/PCI mapping functions conditional"

Broke powerpc build without CONFIG_PCI as we would still define
pci_iomap(), which overlaps with the new empty inline in the headers.

Make our implementation conditional on CONFIG_PCI

Signed-off-by: Benjamin Herrenschmidt

powerpc/4xx: Fix build of PCI code on 405

2011-08-05T05:59:40+00:00

Commit 112d1fe9f7715db423ffeec5ac1beccff6093dc4
"powerpc/4xx: Add check_link to struct ppc4xx_pciex_hwops" inadvertently
broke 405 builds due to some functions being over protected by an
ifdef CONFIG_44x.

Move them back out.

Signed-off-by: Benjamin Herrenschmidt

powerpc/pseries: Simplify vpa deregistration functions

2011-08-05T04:47:58+00:00

The VPA, SLB shadow and DTL degistration functions do not need an
address, so simplify things and remove it.

Also cleanup pseries_kexec_cpu_down a bit by storing the cpu IDs
in local variables.

Signed-off-by: Anton Blanchard 
Signed-off-by: Benjamin Herrenschmidt

powerpc/pseries: Cleanup VPA registration and deregistration errors

2011-08-05T04:47:57+00:00

Make the VPA, SLB shadow and DTL registration and deregistration
functions print consistent messages on error. I needed the firmware
error code while chasing a kexec bug but we weren't printing it.

Signed-off-by: Anton Blanchard 
Signed-off-by: Benjamin Herrenschmidt

powerpc/pseries: Fix kexec on recent firmware versions

2011-08-05T04:47:57+00:00

Recent versions of firmware will fail to unmap the virtual processor
area if we have a dispatch trace log registered. This causes kexec
to fail.

If a trace log is registered this patch unregisters it before the
SLB shadow and virtual processor areas, fixing the problem.

The address argument is ignored by firmware on unregister so we
may as well remove it.

Signed-off-by: Anton Blanchard 
Cc: 
Signed-off-by: Benjamin Herrenschmidt

powerpc: Make KVM_GUEST default to n

2011-08-05T04:47:57+00:00

KVM_GUEST adds a 1 MB array to the kernel (kvm_tmp) which grew
my kernel enough to cause it to fail to boot.

Dynamically allocating or reducing the size of this array is a
good idea, but in the meantime I think it makes sense to make
KVM_GUEST default to n in order to minimise surprises.

Signed-off-by: Anton Blanchard 
Signed-off-by: Benjamin Herrenschmidt

powerpc/kvm: Fix build errors with older toolchains

2011-08-05T04:47:56+00:00

On a box with gcc 4.3.2, I see errors like:

arch/powerpc/kvm/book3s_hv_rmhandlers.S:1254: Error: Unrecognized opcode: stxvd2x
arch/powerpc/kvm/book3s_hv_rmhandlers.S:1316: Error: Unrecognized opcode: lxvd2x

Signed-off-by: Nishanth Aravamudan 
Signed-off-by: Benjamin Herrenschmidt

powerpc: Lack of ibm,io-events not that important!

2011-08-05T04:47:56+00:00

The ibm,io-events code is a bit verbose with its error messages.
Reverse the reporting so we only print when we successfully enable
I/O event interrupts.

Signed-off-by: Anton Blanchard 
Signed-off-by: Benjamin Herrenschmidt

powerpc: Move kdump default base address to half RMO size on 64bit

2011-08-05T04:47:56+00:00

We are seeing boot failures on some very large boxes even with
commit b5416ca9f824 (powerpc: Move kdump default base address to
64MB on 64bit).

This patch halves the RMO so both kernels get about the same
amount of RMO memory. On large machines this region will be
at least 256MB, so each kernel will get 128MB.

We cap it at 256MB (small SLB size) since some early allocations need
to be in the bolted SLB region. We could relax this on machines with
1TB SLBs in a future patch.

Signed-off-by: Anton Blanchard 
Signed-off-by: Benjamin Herrenschmidt