Hi Kim,
I try to address some of the comments. I am not mentioning things that I
agree with you.
> + */
> +static int get_sg_count(struct scatterlist *sg_list, int nbytes)
> +{
> + struct scatterlist *sg = sg_list;
> + int sg_nents = 0;
> +
> + while (nbytes) {
> + sg_ne
[EMAIL PROTECTED]
Arnd Bergmann and Mijo Safradin called my attention to problems with
libnuma's read_mask() on a big-endian 64-bit machine.
In testing that function I found it to be buggy on little-endian and 32-bit
systems as well.
I propose a function that looks like the below. The patch is
Benjamin Herrenschmidt wrote:
> On Fri, 2008-12-05 at 06:08 -0500, Josh Boyer wrote:
> > Shouldn't there also be a next-level-cache property added to the cpu
> > node that references this?
> >
> It would be nice indeed, it would allow the kernel to expose the cache
> info in sysfs
Currently the k
Ok .
=
Case arch = powerpc
I saw that cdm is modified in two fields.
arch/powerpc/platforms/mpc52xx_common.c file, in function:
mpc52xx_setup_cpu(void)
.
/* Use internal 48 Mhz */
out_8(&cdm->ext_48mhz_en, 0x00);
out_8(&cdm->fd_enable,
- Add gpio-controller node to manage QE GPIO Bank D;
- Add mmc-spi node;
- Modify board file so that it won't use legacy SPI support with the new
device trees.
Signed-off-by: Anton Vorontsov <[EMAIL PROTECTED]>
---
arch/powerpc/boot/dts/mpc832x_rdb.dts | 24
arch/
The bindings describes a case where MMC/SD/SDIO slot directly connected
to a SPI bus. Such setups are widely used on embedded PowerPC boards.
The patch also adds the mmc-spi-slot entry to the OpenFirmware modalias
table.
Signed-off-by: Anton Vorontsov <[EMAIL PROTECTED]>
---
.../powerpc/dts-bind
The advantages of this:
- Don't encourage legacy support;
- Less external symbols, less code to compile-in for !MPC832x_RDB
platforms.
Signed-off-by: Anton Vorontsov <[EMAIL PROTECTED]>
---
arch/powerpc/platforms/83xx/mpc832x_rdb.c | 107 +
arch/powerpc/sysdev/fsl_s
This patch implements full support for OF SPI bindings. Now the driver
can manage its own chip selects without any help from the board files
and/or fsl_soc constructors.
The "legacy" code is well isolated and could be removed as time goes by.
Signed-off-by: Anton Vorontsov <[EMAIL PROTECTED]>
---
The main purpose of this patch is to pass 'struct spi_device' to the
chip select handling routines. This is needed so that we could implement
full-fledged OpenFirmware support for this driver.
While at it, also:
- Replace two {de,activate}_cs routines by single cs_contol().
- Don't duplicate platf
The patch fixes following sparse warnings:
CHECK spi_mpc83xx.c
spi_mpc83xx.c:145:1: warning: symbol 'mpc83xx_spi_rx_buf_u8' was not declared.
Should it be static?
spi_mpc83xx.c:146:1: warning: symbol 'mpc83xx_spi_rx_buf_u16' was not declared.
Should it be static?
spi_mpc83xx.c:147:1: warning
This is needed to not bother with ugly #ifdefs in the drivers.
Signed-off-by: Anton Vorontsov <[EMAIL PROTECTED]>
---
arch/powerpc/sysdev/fsl_soc.h |5 +
1 files changed, 5 insertions(+), 0 deletions(-)
diff --git a/arch/powerpc/sysdev/fsl_soc.h b/arch/powerpc/sysdev/fsl_soc.h
index 60f7
Hi all,
The patch series are used to support SPI via the OF SPI subsystem
(driver/of/of_spi.c). Now the driver is able to manage its own
chip selects, and doesn't need any auxiliary support from the
board files or fsl_soc constructors.
The series also contains PowerPC portions of the MMC SPI supp
Hi Paul,
Please pull the 'merge' branch of the 4xx tree to pick up the fix below.
Once it's in Linus' tree, I'll submit it to -stable as we need the same
fix in .27.
josh
The following changes since commit feaf3848a813a106f163013af6fcf6c4bfec92d9:
Linus Torvalds (1):
Merge branch 'for-
thanks, applied
___
Linuxppc-dev mailing list
Linuxppc-dev@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-dev
The attachment is from Freescale and shows how the divider works.
--
Jon Smirl
[EMAIL PROTECTED]
fractinal_div.pdf
Description: Adobe PDF document
___
Linuxppc-dev mailing list
Linuxppc-dev@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-dev
This function is used to count how many GPIOs are specified for
a device node.
Signed-off-by: Anton Vorontsov <[EMAIL PROTECTED]>
---
drivers/of/gpio.c | 34 ++
include/linux/of_gpio.h |6 ++
2 files changed, 40 insertions(+), 0 deletions(-)
diff -
Given this list (contains three gpio specifiers, one of which is a hole):
gpios = <&phandle1 1 2 3
0 /* a hole */
&phandle2 4 5 6>;
of_parse_phandles_with_args() would report -ENOENT for the `hole'
specifier item, the same error value is used to report the end of the
list, for e
By using 'list++' in the beginning we can simplify the code a
little bit.
Suggested-by: Benjamin Herrenschmidt <[EMAIL PROTECTED]>
Signed-off-by: Anton Vorontsov <[EMAIL PROTECTED]>
---
drivers/of/base.c | 11 ---
1 files changed, 4 insertions(+), 7 deletions(-)
diff --git a/drivers/of
Hi all,
Few patches are needed to implement of_gpio_count() helper function.
So far there are two potential users for the new helper:
- PPC4xx SPI driver
http://www.mail-archive.com/linuxppc-dev@ozlabs.org/msg26847.html
^^ The driver implements of_num_gpios() by itself. Though its
implemen
On Freitag, 5. Dezember 2008, [EMAIL PROTECTED] wrote:
>
>
> Now plugging-in USB:
>
>
>
> [EMAIL PROTECTED]: ~ #[ 17.228078] hub 1-0:1.0: state 7 ports 2 chg evt
> 0004 [EMAIL PR
Kumar Gala wrote:
> this doesn't seem critical right now so I'd say we wait for .29
I agree. None of the defconfigs or default behavior expose this bug. I don't
expect anyone to actually turn off the QE in a real-world situation.
--
Timur Tabi
Linux kernel developer at Freescale
_
On Dec 5, 2008, at 10:55 AM, Anton Vorontsov wrote:
This patch implements traditional way of !QE case handling.
Alternative version is coming (w/o ifdefs in the board files).
p.s. I don't know if it is 2.6.28 material...
arch/powerpc/platforms/83xx/mpc832x_rdb.c |3 ++-
arch/powerpc/platfo
Some 83xx boards were not ready for the optional QUICC Engine support.
This patch fixes following build errors:
arch/powerpc/platforms/built-in.o: In function `flush_disable_caches':
(.text+0xb308): undefined reference to `par_io_data_set'
arch/powerpc/platforms/built-in.o: In function `flush_dis
Some 83xx boards were not ready for the optional QUICC Engine support.
This patch fixes following build errors:
arch/powerpc/platforms/built-in.o: In function `flush_disable_caches':
(.text+0xb308): undefined reference to `par_io_data_set'
arch/powerpc/platforms/built-in.o: In function `flush_dis
This patch fixes following sparse warnings:
CHECK 83xx/usb.c
83xx/usb.c:205:5: warning: symbol 'mpc837x_usb_cfg' was not declared. Should it
be static?
CHECK 83xx/mpc831x_rdb.c
83xx/mpc831x_rdb.c:45:13: warning: symbol 'mpc831x_rdb_init_IRQ' was not
declared. Should it be static?
CHECK
Using 2.6.26.3 nothing changes.
Reading MPC5200 user guide I calculate the value of port_config to
configure USB on ethernet port. So I'm using this value in arch=ppc
everyting works.
With arch=powerpc I see that the kernel read the device tree blob, so I think
that I need to modify it.
How
On Fri, Dec 5, 2008 at 8:16 AM, <[EMAIL PROTECTED]> wrote:
> Hi Grant
>
> we have already tested what you are suggesting, but nothing happens.
>
> I'm supposing the problem comes from interrupts.
Unlikely. The interrupt routing is hard wired into the SoC and
probably not messed up (unless the US
Hi Grant
we have already tested what you are suggesting, but nothing happens.
I'm supposing the problem comes from interrupts.
In fact USB is still alive, every 5 sec. we can see some timeout messages.
(from previous post)
[EMAIL PROTECTED]: ~ #[ 24.027842] usb 1-2: khubd timed out on ep0in
On Thursday 04 December 2008, Arnd Bergmann wrote:
> The read-mask function assumes that it is running in 32-bit mode,
> by addressing the bitmask as a series of int values, instead of
> longs. This is broken as can easily be reproduced by running numademo
> on a bit-endian 64-bit system.
>
> Chan
On Fri, Dec 5, 2008 at 4:22 AM, Angelo <[EMAIL PROTECTED]> wrote:
> Hi all.
>
> I'd like to know if it's necessary to modify my DTS.
>
> On my board I have:
> * u-boot 1.2.0
> * kernel 2.6.21
> * arch=powerpc
> * dts based on lite5200b.dts
> * USB port (USB2) configured on the Ethernet port
>
> Is
Angelo wrote:
> Hi all.
>
> I saw that someones trying to work on a custom board based on mpc5200.
> I have:
> * u-boot 1.2.0
> * kernel 2.6.21
> * arch=powerpc
> * custom dts based on lite5200b.dts
> * USB port (USB2) configured on the Ethernet port
>
> Is it necessary to combine Ethernet and us
Hi all.
I saw that someones trying to work on a custom board based on mpc5200.
I have:
* u-boot 1.2.0
* kernel 2.6.21
* arch=powerpc
* custom dts based on lite5200b.dts
* USB port (USB2) configured on the Ethernet port
Is it necessary to combine Ethernet and usb nodes?
However this is the log whe
With this patch the L2 cache is enabled on Canyonlands to increase the
overall performance. There is a known cache coherency issue with the L2
cache, but this is related to the high bandwidth (HB) PLB segment where
the memory address is 0x8.. (low bandwidth PLB segment is mapped
to 0x0.
On Friday 05 December 2008, Benjamin Herrenschmidt wrote:
> On Fri, 2008-12-05 at 06:08 -0500, Josh Boyer wrote:
> > Shouldn't there also be a next-level-cache property added to the cpu
> > node that references this?
>
> It would be nice indeed, it would allow the kernel to expose the cache
> info
On Fri, 2008-12-05 at 06:08 -0500, Josh Boyer wrote:
> Shouldn't there also be a next-level-cache property added to the cpu
> node that references this?
>
It would be nice indeed, it would allow the kernel to expose the cache
info in sysfs
Cheers,
Ben.
__
On Fri, 5 Dec 2008 07:08:52 +0100
Stefan Roese <[EMAIL PROTECTED]> wrote:
> With this patch the L2 cache is enabled on Canyonlands to increase the
> overall performance. There is a known cache coherency issue with the L2
> cache, but this is related to the high bandwidth (HB) PLB segment where
>
36 matches
Mail list logo
