On Fri, 2011-06-17 at 11:58 -0500, Scott Wood wrote:
> When did this change from "considered an internal implementation
> issue, and not really an interface" to "all new interfaces"?
Interesting blurb... that's not everybody's opinion and I would argue
that supporting AMP kernels isn't something w
On Fri, 2011-06-17 at 11:32 -0500, Scott Wood wrote:
> On Fri, 17 Jun 2011 12:00:50 +1000
> Benjamin Herrenschmidt wrote:
>
> > Does this completely replace your previous series of 7 patches ? (IE.
> > Should I ditch them in patchwork ?) Or does it apply on top of them ?
>
> It replaces them.
>
On Fri, 2011-06-17 at 10:03 -0600, Dave Carroll wrote:
> > On Fri, 2011-06-17 at 15:39 +1000, Benjamin Herrenschmidt wrote:
> >
> > Hrm.. none of your 2 pending patches applies cleanly (factoring
> > free_initmem & that one). Haven't had a chance to check why yet, can you
> > check on your side ?
>
On Fri, 17 Jun 2011 16:54:27 -0400
"Matthew L. Creech" wrote:
> Hi, I posted this on the Linux-MTD list but haven't gotten any hits.
> Since it looks like it could be MPC83xx-specific, I'm reposting here.
> Rick Johnson noted a problem in fsl_elbc_nand.c back in May which
> might be related:
>
>
Hi, I posted this on the Linux-MTD list but haven't gotten any hits.
Since it looks like it could be MPC83xx-specific, I'm reposting here.
Rick Johnson noted a problem in fsl_elbc_nand.c back in May which
might be related:
http://lists.infradead.org/pipermail/linux-mtd/2011-May/035372.html
We've
+ addir1,r1,STACKFRAMESIZE
+
+ .align 5
Do we know that the blank will be filled with something harmless ?
Yes. See ppc_handle_align() in gas/config/tc-ppc.c : it fills with nops
(ori 0,0,0), and a branch if there are more than four nops, and for
POWER6
and POWER7 it puts a
On Fri, Jun 17, 2011 at 03:12:56PM +0400, Dmitry Eremin-Solenikov wrote:
> What about drivers/cpufreq/powerpc, or it's an unnecessary?
We haven't done it so far for x86 & arm, so for now at least, just keeping
them in drivers/cpufreq/ should be sufficient.
> Should I resumbit it, or there will
On Sun, Jun 12, 2011 at 7:23 PM, Shaohui Xie wrote:
> when nand_get_flash_type() is called, it will read 8 bytes of ID instead of 5,
> but the driver only read 5 bytes, so kernel will dump error messages like:
>
> fsl-lbc ffe124000.localbus: read_byte beyond end of buffer
> fsl-lbc ffe124000.local
From: Holger Brunck
Date: Fri, 17 Jun 2011 10:30:39 +0200
> From: Clive Stubbings
>
> The RIPTR and TIPTR (receive/transmit internal temporary data pointer),
> used by microcode as a temporary buffer for data, must be 32-byte aligned
> according to the RM for MPC8247.
>
> Tested on mgcoge.
>
On Fri, 17 Jun 2011 15:33:04 +1000
Benjamin Herrenschmidt wrote:
> On Tue, 2011-05-31 at 14:19 -0500, Meador Inge wrote:
> > +void mpic_msgr_enable(struct mpic_msgr *msgr)
> > +{
> > + out_be32(msgr->mer, in_be32(msgr->mer) | (1 << msgr->num));
> > +}
> > +EXPORT_SYMBOL(mpic_msgr_enable);
>
>
On Fri, 17 Jun 2011 12:00:50 +1000
Benjamin Herrenschmidt wrote:
> Does this completely replace your previous series of 7 patches ? (IE.
> Should I ditch them in patchwork ?) Or does it apply on top of them ?
It replaces them.
> > #define SET_IVOR(vector_number, vector_offset) \
> > diff -
On Fri, 17 Jun 2011 15:12:00 +1000
Benjamin Herrenschmidt wrote:
> On Tue, 2011-05-24 at 06:51 +1000, Benjamin Herrenschmidt wrote:
> > On Mon, 2011-05-23 at 15:26 -0500, Scott Wood wrote:
> > > On Sat, 21 May 2011 08:32:58 +1000
> > > Benjamin Herrenschmidt wrote:
> > >
> > > > On Fri, 2011-05
> For finding structures, the following pattern tends to work well:
>
> git grep 'struct device_node {'
In general 'typedef.*\' will find types.
eg: grep -r -n --include '*.h' 'typedef.*\'
base_of_source_tree
Finding functions (in .c files) is easy if names start in
column 1 '^function_name\>'
On Fri, Jun 17, 2011 at 7:53 AM, Guillaume Dargaud
wrote:
>> I'm sorry, I should have mentioned that struct of_device was replaced by
>> struct platform_device. So in order to get your code compiled, you have
>> to change the probe function to
>>
>> static int xad_driver_probe(struct platform_devi
> I'm sorry, I should have mentioned that struct of_device was replaced by
> struct platform_device. So in order to get your code compiled, you have
> to change the probe function to
>
> static int xad_driver_probe(struct platform_device* dev, const struct
> of_device_id *match) {
Thanks, that wo
On 06/17/2011 12:33 AM, Benjamin Herrenschmidt wrote:
> On Tue, 2011-05-31 at 14:19 -0500, Meador Inge wrote:
>> Some MPIC implementations contain one or more blocks of message registers
>> that are used to send messages between cores via IPIs. A simple API has
>> been added to access (get/put, r
Enable functions used to access SCOM if PPC_MAPLE is defined: they are
used by cpufreq driver to control hardware.
Signed-off-by: Dmitry Eremin-Solenikov
---
arch/powerpc/kernel/misc_64.S |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/kernel/misc_64.S b/
Add simple cpufreq driver for Maple-based boards (ppc970fx evaluation
kit and others). Driver is based on a cpufreq driver for 64-bit powermac
boxes with all pmac-dependant features removed and simple cleanup
applied.
Signed-off-by: Dmitry Eremin-Solenikov
---
drivers/cpufreq/Kconfig |
On 06/17/2011 02:57 PM, Philipp Ittershagen wrote:
> The struct device_node was moved to struct device in order to make the
> CONFIG_OF more generic. See
>
> commit d706c1b050274b3bf97d7cb0542c0d070c9ccb8b
> Author: Grant Likely
> Date: Tue Apr 13 16:12:28 2010 -0700
>
> driver-core: Add d
On 06/17/2011 01:22 PM, Guillaume Dargaud wrote:
>> Not answering your primary question here, but in order to print the most
>> recent tag (along with a uniqe suffix, see man page) that is reachable
>> from your current branch, you can use "git describe". It will print out
>> the kernel version on
If second CPU is not enabled, CPC925 EDAC driver will spill out warnings
about errors on second Processor Interface. Support masking that out,
by detecting at runtime which CPUs are present in device tree.
Signed-off-by: Dmitry Eremin-Solenikov
Cc: Harry Ciao
Cc: Doug Thompson
Signed-off-by: Dm
Currently Maple setup code creates cpc925_edac device only on
Motorola ATCA-6101 blade. Make setup code check bridge revision
and enable EDAC on all U3H bridges.
Verified on Momentum MapleD (ppc970fx kit) board.
Signed-off-by: Dmitry Eremin-Solenikov
---
arch/powerpc/platforms/maple/setup.c |
Hello, colleagues,
Another respin of this patchset.
Changes since previous version:
* Fix comments in cpc925_edac.c
* Rename cpc925_cpu_getmask() to cpc925_cpu_mask_disabled()
* Don't return an error from init code if it's a U3 and not U3H
arch/powerpc/platforms/maple/setup.c | 41 +--
> Not answering your primary question here, but in order to print the most
> recent tag (along with a uniqe suffix, see man page) that is reachable
> from your current branch, you can use "git describe". It will print out
> the kernel version on your latest Xilinx tree.
$ git describe
v2.6.37-719-
On Friday 17 June 2011 13:12:56 Dmitry Eremin-Solenikov wrote:
> > I think new cpufreq drivers should live in drivers/cpufreq, not in arch/.
> > We've
> > started moving other drivers away from arch/x86 and arch/arm already.
>
> What about drivers/cpufreq/powerpc, or it's an unnecessary?
drivers/
Hello,
On 6/17/11, Arnd Bergmann wrote:
> On Friday 17 June 2011 10:49:16 Dmitry Eremin-Solenikov wrote:
>> Add simple cpufreq driver for Maple-based boards (ppc970fx evaluation
>> kit and others). Driver is based on a cpufreq driver for 64-bit powermac
>> boxes with all pmac-dependant features r
On 06/17/2011 11:19 AM, Guillaume Dargaud wrote:
> [I'm on the latest Xilinx tree, FWIW, so that's 2.6.25 still ?]
Not answering your primary question here, but in order to print the most
recent tag (along with a uniqe suffix, see man page) that is reachable
from your current branch, you can use "
A couple linux versions ago, I had the following working code in my driver:
static int xad_driver_probe(struct of_device* dev, const struct of_device_id
*match) {
struct device_node *dn = dev->dev.of_node;
But now it won't compile and I can't figure out how to get the device_node
struct
> Hello,
> Motioned to the memory aligned, now there is such requirement:
> When the driver send an packet to hardware, the skb's address passed
by
> stack do a dma map into hardware, the skb's dma address must
> be 64-byte aligned.
Does the hardware support buffer chaining?
In which case you o
Hello,
Motioned to the memory aligned, now there is such requirement:
When the driver send an packet to hardware, the skb's address passed by
stack do a dma map into hardware, the skb's dma address must be 64-byte
aligned.
My method:
Allocate a new skb which is 64-byte aligned, then copy the passe
On Friday 17 June 2011 10:49:16 Dmitry Eremin-Solenikov wrote:
> Add simple cpufreq driver for Maple-based boards (ppc970fx evaluation
> kit and others). Driver is based on a cpufreq driver for 64-bit powermac
> boxes with all pmac-dependant features removed and simple cleanup
> applied.
>
> Signe
> On Fri, Jun 17, 2011 at 02:54:00PM +1000, Anton Blanchard wrote:
> > Implement a POWER7 optimised memcpy using VMX. For large aligned
> > copies this new loop is over 10% faster and for large unaligned
> > copies it is over 200% faster.
...
> BTW: do you have any statistics on the size distrib
Add simple cpufreq driver for Maple-based boards (ppc970fx evaluation
kit and others). Driver is based on a cpufreq driver for 64-bit powermac
boxes with all pmac-dependant features removed and simple cleanup
applied.
Signed-off-by: Dmitry Eremin-Solenikov
---
arch/powerpc/kernel/misc_64.S
On Fri, Jun 17, 2011 at 02:54:00PM +1000, Anton Blanchard wrote:
> Implement a POWER7 optimised memcpy using VMX. For large aligned
> copies this new loop is over 10% faster and for large unaligned
> copies it is over 200% faster.
>
> On POWER7 unaligned stores rarely slow down - they only flush w
34 matches
Mail list logo
