On Thu, 2015-01-29 at 20:21 +1100, Alexey Kardashevskiy wrote:
This checks that the TCE table page size is not bigger that the size of
a page we just pinned and going to put its physical address to the table.
Otherwise the hardware gets unwanted access to physical memory between
the end of
Hi Aneesh,
yes. We do use jump label. I also verified that looking at .s
#APP
# 23 ./arch/powerpc/include/asm/jump_label.h 1
1:
nop
.pushsection __jump_table, aw
.llong 1b, .L201, __tracepoint_hash_fault+8 #,
.popsection
# 0 2
So we
On Mon, Feb 02, 2015 at 09:47:57AM -0600, Bjorn Helgaas wrote:
I think it makes more sense to merge this along with the other 14
patches that remove the #ifdefs (at least, I assume that's what they
do; I haven't seen them).
Acked-by: Bjorn Helgaas bhelg...@google.com
But please edit the
On Thu, 2015-01-29 at 20:21 +1100, Alexey Kardashevskiy wrote:
The existing implementation accounts the whole DMA window in
the locked_vm counter which is going to be even worse with multiple
containers and huge DMA windows.
This introduces 2 ioctls to register/unregister DMA memory which
On Thu, 2015-01-29 at 20:21 +1100, Alexey Kardashevskiy wrote:
There moves locked pages accounting to helpers.
Later they will be reused for Dynamic DMA windows (DDW).
While we are here, update the comment explaining why RLIMIT_MEMLOCK
might be required to be bigger than the guest RAM. This
Add a stub for pci_device_to_OF_node() so drivers don't need to
use #ifdef CONFIG_OF around calls to it.
Signed-off-by: Kevin Hao haoke...@gmail.com
Acked-by: Bjorn Helgaas bhelg...@google.com
---
Hi Bjorn,
Here is the revised version with the update of the subject and commit log
in case that
On Thu, 2015-01-29 at 20:21 +1100, Alexey Kardashevskiy wrote:
Modern IBM POWERPC systems support multiple (currently two) TCE tables
per IOMMU group (a.k.a. PE). This adds a powerpc_iommu container
for TCE tables. Right now just one table is supported.
Signed-off-by: Alexey Kardashevskiy
On Mon, 2015-02-02 at 09:47 -0600, Bjorn Helgaas wrote:
On Sun, Feb 1, 2015 at 8:28 PM, Michael Ellerman m...@ellerman.id.au wrote:
On Sat, 2015-01-31 at 21:47 +0800, Kevin Hao wrote:
So we can avoid the ugly #ifdef in some drivers.
Signed-off-by: Kevin Hao haoke...@gmail.com
---
On Fri, 2015-01-30 at 10:00 +0100, Geert Uytterhoeven wrote:
Hi Michael,
On Fri, Jan 30, 2015 at 5:09 AM, Michael Ellerman m...@ellerman.id.au wrote:
On Wed, 2015-14-01 at 13:51:57 UTC, Geert Uytterhoeven wrote:
of_find_node_by_name() calls of_node_put() on its from parameter,
which must
On Fri, Jan 30, 2015 at 05:08:03PM -0600, Bjorn Helgaas wrote:
On Thu, Jan 15, 2015 at 10:28:04AM +0800, Wei Yang wrote:
On PowrNV platform, resource position in M64 implies the PE# the resource
belongs to. In some particular case, adjustment of a resource is necessary
to locate it to a correct
Hi,
ebizzy with -S 30 -t 1 -P gave
13627 records/s - Without patch
13546 records/s - With patch with tracepoint disabled
OK. So that's about -0.6%. Are we happy with that? I'm not sure.
Can you do a few more runs and see if that's a stable result.
Surprisingly large. Is
On Sun, Feb 01, 2015 at 01:51:50PM +0800, Kevin Hao wrote:
diff --git a/drivers/video/fbdev/imsttfb.c b/drivers/video/fbdev/imsttfb.c
index aae10ce74f14..91a80bb8f988 100644
--- a/drivers/video/fbdev/imsttfb.c
+++ b/drivers/video/fbdev/imsttfb.c
@@ -1470,7 +1470,6 @@ static int
Sometimes the KVM code on powerpc needs to emulate load or store
instructions from the guest, which can include both normal and byte
reversed forms.
We currently (AFAICT) handle this correctly, but some variable names are
very misleading. In particular we use is_bigendian in several places to
The powerpc specific st_le*() and ld_le*() functions in
arch/powerpc/asm/swab.h no longer have any users. They are also
misleadingly named, since they always byteswap, even on a little-endian
host.
This patch removes them.
Signed-off-by: David Gibson da...@gibson.dropbear.id.au
---
On 02/03/2015 11:11 AM, Alex Williamson wrote:
On Thu, 2015-01-29 at 20:21 +1100, Alexey Kardashevskiy wrote:
The existing implementation accounts the whole DMA window in
the locked_vm counter which is going to be even worse with multiple
containers and huge DMA windows.
This introduces 2
The bt8xx PCI DVB driver includes a powerpc specific hack, using one of
the powerpc specific byteswapping functions in an IO helper macro.
There's no reason to use the powerpc specific function instead of a
generic byteswap, so this patch removes it. I'm not sure if the powerpc
specific memory
When the MXC MMUC driver is used on a Freescale MPC512x machine, it
contains some additional byteswapping code (I'm assuming this is a
workaround for a hardware defect). This uses the ppc specific st_le32()
function, but there's no reason not to use the generic swab32() function
instead. gcc is
On Thu, 2015-01-29 at 19:03 +0100, Greg Kurz wrote:
On Wed, 17 Dec 2014 10:40:46 +0100
Greg Kurz gk...@linux.vnet.ibm.com wrote:
Hi,
This series addresses remarks from Ben and Michael (see individual patches).
The most notable changes are:
- the parsing code being pull out into a
Anton Blanchard an...@samba.org writes:
Hi Aneesh,
yes. We do use jump label. I also verified that looking at .s
#APP
# 23 ./arch/powerpc/include/asm/jump_label.h 1
1:
nop
.pushsection __jump_table, aw
.llong 1b, .L201, __tracepoint_hash_fault+8
On Sun, 1 Feb 2015, Geert Uytterhoeven wrote:
On Sun, Feb 1, 2015 at 4:39 AM, Finn Thain wrote:
On Sun, 4 Jan 2015, Geert Uytterhoeven wrote:
On Sun, Jan 4, 2015 at 8:21 AM, Finn Thain wrote:
On Thu, 1 Jan 2015, Rickard Strandqvist wrote:
Removes some functions that are not used
We currently read the information about idle states from the DT
so as to find out the cpu idle states supported by the platform.
Use those APIs to read from the DT that can avoid endianness
conversions of the property values.
Signed-off-by: Preeti U Murthy pre...@linux.vnet.ibm.com
---
On Thu, 2015-01-29 at 20:22 +1100, Alexey Kardashevskiy wrote:
This adds create/remove window ioctls to create and remove DMA windows.
This changes VFIO_IOMMU_SPAPR_TCE_GET_INFO handler to return additional
information such as a number of supported windows and maximum number
levels of TCE
ppc has special instruction forms to efficiently load and store values
in non-native endianness. These can be accessed via the arch-specific
{ld,st}_le{16,32}() inlines in arch/powerpc/include/asm/swab.h.
However, gcc is perfectly capable of generating the byte-reversing
load/store instructions
arch/powerpc/include/asm/swab.h includes some powerpc specific
byteswapping functions, which are implemented in terms of powerpc's
built in byte reversed load/store instructions. There are two problems with
this:
1) They're not necessary - gcc is perfectly capable of generating the
On Sun, 1 Feb 2015, Russell King - ARM Linux wrote:
On Sun, Feb 01, 2015 at 02:39:42PM +1100, Finn Thain wrote:
I find the ARM support in drivers/char/nvram to be surprising, not to
say questionable. The /proc/driver/nvram implementation, given
defined(__arm__), decodes the NVRAM
The device tree now exposes the residency values for different idle states. Read
these values instead of calculating residency from the latency values. The
values
exposed in the DT are validated for optimal power efficiency. However to
maintain
compatibility with the older firmware code which
On Sun, Feb 1, 2015 at 8:28 PM, Michael Ellerman m...@ellerman.id.au wrote:
On Sat, 2015-01-31 at 21:47 +0800, Kevin Hao wrote:
So we can avoid the ugly #ifdef in some drivers.
Signed-off-by: Kevin Hao haoke...@gmail.com
---
include/linux/pci.h | 2 ++
1 file changed, 2 insertions(+)
Hi
On Tue, 03 Feb 2015 13:47:35 +1100
Michael Ellerman m...@ellerman.id.au wrote:
On Thu, 2015-01-29 at 19:03 +0100, Greg Kurz wrote:
On Wed, 17 Dec 2014 10:40:46 +0100
Greg Kurz gk...@linux.vnet.ibm.com wrote:
Hi,
This series addresses remarks from Ben and Michael (see individual
The actual IOV BAR range is determined by the start address and the actual
size for vf_num VFs BAR. After shifting the IOV BAR, there would be a
chance the actual end address exceed the limit and overlap with other
devices.
This patch adds a check to make sure after shifting, the range will not
Anton Blanchard an...@samba.org writes:
Hi,
ebizzy with -S 30 -t 1 -P gave
13627 records/s - Without patch
13546 records/s - With patch with tracepoint disabled
OK. So that's about -0.6%. Are we happy with that? I'm not sure.
Can you do a few more runs and see if that's a stable
Michael Ellerman m...@ellerman.id.au writes:
On Wed, 2015-01-21 at 14:15 +0530, Aneesh Kumar K.V wrote:
Michael Ellerman m...@ellerman.id.au writes:
On Tue, 2015-01-20 at 17:05 +0530, Aneesh Kumar K.V wrote:
This enables us to understand how many hash fault we are taking
when running
Freescale ethernet controllers have the capability to re-assemble fragmented
data into a single ethernet frame. This patch uses this capability and
implements NETIP_F_SG feature into the fs_enet ethernet driver.
On a MPC885, I get 53% performance improvement on a ftp transfer of a 15Mb file:
*
Hi Kevin,
On Sun, 1 Feb 2015 13:51:50 +0800 Kevin Hao haoke...@gmail.com wrote:
That was my first thought, but the codes protected by the PPC_OF seem not
ppc specific and should be safe for other archs which also support OF. So I
drop the PPC_OF completely. Did I miss something?
Ah, ok.
Hi Kevin,
On Tue, 3 Feb 2015 10:20:02 +0800 Kevin Hao haoke...@gmail.com wrote:
I took a second look at this. It seems that there is a declaration of
struct device_node in linux/device.h and there is also no access to the
member of device_node in this driver. So we are safe to not include
We currently read the information about idle states from the DT
so as to populate the cpuidle table. Use those APIs to read from
the DT that can avoid endianness conversions of the property values
in the cpuidle driver.
Signed-off-by: Preeti U Murthy pre...@linux.vnet.ibm.com
---
This patch is
On 02/02/2015 12:09 PM, Michael Ellerman wrote:
On Mon, 2015-02-02 at 10:40 +0530, Preeti U Murthy wrote:
The device tree now exposes the residency values for different idle states.
Read
these values instead of calculating residency from the latency values. The
values
exposed in the DT are
36 matches
Mail list logo
