On Mon, 2012-06-25 at 14:28 +0800, Dong Aisheng wrote:
> From: Dong Aisheng
>
> prom_update_property() currently fails if the property doesn't
> actually exist yet which isn't what we want. Change to add-or-update
> instead of update-only, then we can remove a lot duplicated lines.
>
> Suggested
Fixes built-in Bluetooth not working on Apple PowerBooks, regression from
commit 75fb0e324daa48ec458fb5c2960eb07b80cfad9d ('Bluetooth: Fix init sequence
for some CSR based controllers').
Cc: sta...@vger.kernel.org [v3.4]
Signed-off-by: Michel Dänzer
---
net/bluetooth/hci_core.c |2 +-
1 file
Add the ability to inject IOMMU faults. We enable this per device
via a fail_iommu sysfs property, similar to fault injection on other
subsystems.
An example:
# lspci
...
0003:01:00.1 Ethernet controller: Emulex Corporation OneConnect 10Gb NIC (be3)
(rev 02)
To inject one error to this device:
The DMA API debug code has hooks to verify all DMA entries have been
freed at time of hot unplug. We need to call dma_debug_add_bus for
this to work.
Signed-off-by: Anton Blanchard
---
Index: linux-build/arch/powerpc/kernel/vio.c
===
Similar to PCI, separate the bus probe from device probe. This allows
us to attach bus notifiers for DMA debug and IOMMU fault injection
before devices have been probed.
Signed-off-by: Anton Blanchard
---
Index: linux-build/arch/powerpc/kernel/vio.c
===
During boot we see a number of these warnings:
vio 3000: Warning: IOMMU dma not supported: mask 0x, table
unavailable
The reason for this is that we set IOMMU properties for all VIO
devices even if they are not DMA capable.
Only set DMA ops, table and mask for devices with
With current implementation, there is one function to retrieve
the corresponding host bridge (struct pci_host_bridge) according
to the given PCI device (struct pci_dev) and that function has
been declared as "static". Further, we don't have the public
function to retrieve host bridge from PCI bus y
On some powerpc platforms, device BARs need to be assigned to separate
"segments" of the address space in order for the error isolation and HW
virtualization mechanisms (EEH) to work properly. Those "segments" have
a minimum size that can be fairly large (16M). In order to be able to
use the generi
We had a request for a fast method of getting CPU and NUMA node IDs
from userspace. Ben suggested we use SPRG3 which is userspace
readable. This is a quick hack to try that out.
I have a glibc patch to implement sched_getcpu using this. Testing
on a POWER7:
baseline: 538 cycles
vdso: 30 cy
