Re: Linuxppc-embedded Digest, Vol 50, Issue 15

[Duy-Ky Nguyen]

Hello Quan,

You're right, I2C @ 0x50 is backup I2C bootloader in case the regular
NOR/NAND corrupued.
What I'm trying to do is to take advantage of high-level access supported
from Linux where I could use read() write() after opening /dev/i2c-0 as a
file using function open("/dev/i2c-0", RDWR)

I'd install i2c-tools you recommend to see what I could get from there. In
the worst case I will write a light-weight I2C driver and access it at
low-level as I used to do.

Thanks, Quan.

Best Wishes,

Duy-Ky

----- Original Message ----- 
From: <[EMAIL PROTECTED]>
To: <linuxppc-embedded@ozlabs.org>
Sent: Thursday, October 09, 2008 5:03 PM
Subject: Linuxppc-embedded Digest, Vol 50, Issue 15


Send Linuxppc-embedded mailing list submissions to
linuxppc-embedded@ozlabs.org

To subscribe or unsubscribe via the World Wide Web, visit
https://ozlabs.org/mailman/listinfo/linuxppc-embedded
or, via email, send a message with subject or body 'help' to
[EMAIL PROTECTED]

You can reach the person managing the list at
[EMAIL PROTECTED]

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Linuxppc-embedded digest..."


Today's Topics:

  1. Re: MTD map drivers (FSL UPM NAND) (Scott Wood)
  2. API access I2C (Duy-Ky Nguyen)
  3. RE: API access I2C (Nguyen Duc Quan)


----------------------------------------------------------------------

Message: 1
Date: Thu, 09 Oct 2008 16:17:51 -0500
From: Scott Wood <[EMAIL PROTECTED]>
Subject: Re: MTD map drivers (FSL UPM NAND)
To: Alemao <[EMAIL PROTECTED]>
Cc: linuxppc-embedded@ozlabs.org
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Alemao wrote:
Hi all,

Im trying to use a physmap driver for NOR and NAND flash devices.

You can't use physmap with NAND.

  ranges = <0 0 ff000000 1000000     // nor flash,  16 MB
                    1 0 60000000 4000000>;   // nand flash, 64 MB

Is your NAND window really that large?  I'm not too familiar with UPM,
but typically it'd only be a few KiB.

So, how can I load the UPM driver before physmap?

Just stop pointing physmap at the NAND.

-Scott


------------------------------

Message: 2
Date: Thu, 9 Oct 2008 16:11:22 -0700
From: "Duy-Ky Nguyen" <[EMAIL PROTECTED]>
Subject: API access I2C
To: <linuxppc-embedded@ozlabs.org>
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; format=flowed; charset="iso-8859-1";
reply-type=original

Hello,

I have an eva board MPC8313E-RDB with 3 I2C device at address 48, 50, 68.
I'm able to read data from those 3 devices

I follow linux/Documentation/i2c/dev-interface to create a C code to 
access
those 3 devices in Linux env, but all are failed !!!

I really appreciate your help where to find  info or sample code running 
in
Linyux to access I2C devices

Thanks so much,

Duy-Ky

----- Original Message ----- 
From: <[EMAIL PROTECTED]>
To: <linuxppc-embedded@ozlabs.org>
Sent: Thursday, October 09, 2008 2:05 PM
Subject: Linuxppc-embedded Digest, Vol 50, Issue 14


Send Linuxppc-embedded mailing list submissions to
linuxppc-embedded@ozlabs.org

To subscribe or unsubscribe via the World Wide Web, visit
https://ozlabs.org/mailman/listinfo/linuxppc-embedded
or, via email, send a message with subject or body 'help' to
[EMAIL PROTECTED]

You can reach the person managing the list at
[EMAIL PROTECTED]

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Linuxppc-embedded digest..."


Today's Topics:

  1. Re: performance: memcpy vs. __copy_tofrom_user (Leon Woestenberg)
  2. Re: performance: memcpy vs. __copy_tofrom_user (Matt Sealey)
  3. Toolchain for ppc-440 without FPU (Ademir Zanetti Junior)
  4. Re: FS_ENET ERROR(s) 0x12 at second NFS RPC port lookup
     (100005/1) (Remi Lefevre)
  5. MTD map drivers (FSL UPM NAND) (Alemao)
  6. RE: Toolchain for ppc-440 without FPU (Ayman El-Khashab)


----------------------------------------------------------------------

Message: 1
Date: Thu, 9 Oct 2008 14:04:53 +0200
From: "Leon Woestenberg" <[EMAIL PROTECTED]>
Subject: Re: performance: memcpy vs. __copy_tofrom_user
To: "Dominik Bozek" <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED], Paul Mackerras <[EMAIL PROTECTED]>,
linuxppc-embedded@ozlabs.org
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1

Hello all,

On Thu, Oct 9, 2008 at 1:41 PM, Dominik Bozek <[EMAIL PROTECTED]>
wrote:
Paul Mackerras wrote:
Dominik Bozek writes:
Actually I made couple of other tests on that mpc8313. Most of them 
are
to ugly to publish them, but... My problem is that I have to boost the
gigabit interface on the mpc8313. I made simple substitution and

Very interesting.  Can you work out where memcpy is being called on
the network data?  I wouldn't have expected that.


Also see this recent thread David Jander on August 25th, "Efficient
memcpy()/memmove() for G2/G3 cores..."
on [EMAIL PROTECTED]

http://ozlabs.org/pipermail/linuxppc-dev/2008-September/062449.html

BTW, I am interested in this work as well, I'm currently working with
a 8313 and 8315 design, both are e300 cores.

My current goal is PCIe though, but I need fast GbE performance as well.

Also, did you test Freescale's 2.6.24.3 patches that tune the gianfar
interfaces for higher performance?

Regards,
--
Leon


------------------------------

Message: 2
Date: Thu, 09 Oct 2008 10:37:55 -0500
From: Matt Sealey <[EMAIL PROTECTED]>
Subject: Re: performance: memcpy vs. __copy_tofrom_user
To: Paul Mackerras <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED], Dominik Bozek <[EMAIL PROTECTED]>,
linuxppc-embedded@ozlabs.org
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=UTF-8; format=flowed

Paul Mackerras wrote:
Dominik Bozek writes:

Actually I made couple of other tests on that mpc8313. Most of them are
to ugly to publish them, but... My problem is that I have to boost the
gigabit interface on the mpc8313. I made simple substitution and
__copy_tofrom_user was used instead of memcpy. I know, it's wrong, but 
I
speedup that way the network interface for about 10%.

Very interesting.  Can you work out where memcpy is being called on
the network data?  I wouldn't have expected that.

It probably is somewhere.. through some weird and wonderful code path 
that
needs some serious digging to find. At least in 2.4 memcpy was used and
optimizing it (see Freescale's libmotovec benchmarks) did produce a
sizable
performance improvement. That, and offloading TCP checksumming to AltiVec
helped a lot.

No help at all on an 8313 but, relevant anyway.

Since then zero copy networking and other fancy things like the DMA
engine API (for intel ioat at least but also there is fsl dma support)
there's less to actually optimize now so you're less likely to see the
same benefits. All these got into mainline because it's essential to
have this kind of architecture to get reasonable speeds out of >gigabit
network links.

There is actually no strong reason not to use __copy_tofrom_user as
memcpy, in fact, as long as we are sure that source and destination
are both cacheable.

I do think there is probably a good benefit in doing things like zeroing
pages in AltiVec and copying entire pages with AltiVec (for instance
when copy-on-write happens in an application) - NetBSD and QNX implement
at least this because it's faster than using the cache management and
works fine on uncacheable pages too (also since you're always aligned to
a page, zeroing 4kb aligned to a 4kb boundary - or whatever your page
size happens to be, the number of errors that can occur are absolutely
tiny and performance can go through the roof).

Ahem, but nobody here wants AltiVec in the kernel do they?

--
Matt Sealey <[EMAIL PROTECTED]>
Genesi, Manager, Developer Relations


------------------------------

Message: 3
Date: Thu, 9 Oct 2008 16:08:18 -0300
From: "Ademir Zanetti Junior" <[EMAIL PROTECTED]>
Subject: Toolchain for ppc-440 without FPU
To: linuxppc-embedded@ozlabs.org
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset="utf-8"

Hello all,

I am trying to create a toolchain by hand and I am facing some problems
when
I try to configure the glibc 2.7 for a powerpc. When I try to configure
the
glibc to install the headers I get the following error:

configure --prefix=/usr --host=powerpc-440-linux-gnu
--with-headers=${SYSROOT}/usr/include --disable-shared --without-cvs
--with-binutils=/home/azanetti/ml507/powerpc/bin/

checking build system type... i686-pc-linux-gnu
checking host system type... i686-pc-linux-gnu
configure: running configure fragment for add-on nptl
checking sysdep dirs... sysdeps/i386/elf
nptl/sysdeps/unix/sysv/linux/i386/i686 nptl/sysdeps/unix/sysv/linux/i386
sysdeps/unix/sysv/linux/i386 nptl/sysdeps/unix/sysv/linux
nptl/sysdeps/pthread sysdeps/pthread sysdeps/unix/sysv/linux sysdeps/gnu
sysdeps/unix/common sysdeps/unix/mman sysdeps/unix/inet
sysdeps/unix/sysv/i386 nptl/sysdeps/unix/sysv sysdeps/unix/sysv
sysdeps/unix/i386 nptl/sysdeps/unix sysdeps/unix sysdeps/posix
sysdeps/i386/i686/fpu nptl/sysdeps/i386/i686 sysdeps/i386/i686
sysdeps/i386/i486 nptl/sysdeps/i386/i486 sysdeps/i386/fpu
nptl/sysdeps/i386
sysdeps/i386 sysdeps/wordsize-32 sysdeps/ieee754/ldbl-96
sysdeps/ieee754/dbl-64 sysdeps/ieee754/flt-32 sysdeps/ieee754
sysdeps/generic/elf sysdeps/generic
checking for a BSD-compatible install... /usr/bin/install -c
checking whether ln -s works... yes
checking for gcc... gcc
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ISO C89... none needed
checking how to run the C preprocessor... gcc -E
checking for g++... g++

.
.
.
checking for long double... yes
checking size of long double... 12
running configure fragment for sysdeps/powerpc/powerpc32/elf
checking for powerpc32 TLS support... no
running configure fragment for sysdeps/unix/sysv/linux/powerpc
checking whether gcc -B/home/azanetti/ml507/powerpc/bin/ -g -O2
-mlong-double-128 uses IBM extended format... no
checking whether gcc -B/home/azanetti/ml507/powerpc/bin/ -g -O2 supports
-mabi=ibmlongdouble... no
configure: error: this configuration requires -mlong-double-128 IBM
extended
format support


I already have the binutils 2.18 compiled on my PATH. I know that it is
easier to use a script like crosstool to compile it but I do want to
understand what is happening behind the scenes.

Regards.

Ademir
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://ozlabs.org/pipermail/linuxppc-embedded/attachments/20081009/a3dd71d6/attachment-0001.htm>

------------------------------

Message: 4
Date: Thu, 9 Oct 2008 21:46:42 +0200
From: "Remi Lefevre" <[EMAIL PROTECTED]>
Subject: Re: FS_ENET ERROR(s) 0x12 at second NFS RPC port lookup
(100005/1)
To: "Scott Wood" <[EMAIL PROTECTED]>
Cc: linuxppc-embedded@ozlabs.org
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1

Ok, I'm sorry, I have incorrectly set the 60x pipeline depth to 0
instead of 1 in BCR register.
It works fine now. The side effect is interesting to know...

regards,
r?mi


------------------------------

Message: 5
Date: Thu, 9 Oct 2008 18:04:25 -0300
From: Alemao <[EMAIL PROTECTED]>
Subject: MTD map drivers (FSL UPM NAND)
To: linuxppc-embedded@ozlabs.org
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1

Hi all,

Im trying to use a physmap driver for NOR and NAND flash devices.

With NOR, everything ok. It uses CFI driver.

The problem is with NAND, cause it is connected at Freescale
LocalBus UPM, and at a first look, the UPM driver is loading after the
physmap (lines 14-15 in kernel output)

Here some outputs from the board:

Bootloader (U-Boot 1.2.0):

CPU:   e300c1, MPC8360E, Rev: 21 at 499.999 MHz, CSB:  333 MHz
Board: MPC8360KTX
I2C:   ready
DDR RAM: 128 MB
FLASH: 16 MB
NAND:  64 MiB
In:    serial
Out:   serial
Err:   serial


Kernel (linux-2.6.17):

1               physmap nor flash device: 1000000 at ff000000
2               NOR flash: Found 1 x16 devices at 0x0 in 16-bit bank
3                Amd/Fujitsu Extended Query Table at 0x0040
4                number of CFI chips: 1
5               Using physmap partition definition
6               Creating 6 MTD partitions on "NOR flash":
7               0x00000000-0x00040000 : "u-boot"
8               0x00040000-0x00060000 : "env."
9               0x00060000-0x00080000 : "sys"
10              0x00080000-0x000c0000 : "logs"
11              0x000c0000-0x00860000 : "admin"
12              0x00860000-0x01000000 : "users"
13
14              physmap nand flash device: 4000000 at 60000000
15              UPM: User-Programmable Machine NAND driver


Device Tree Source:

[EMAIL PROTECTED] {
 compatible = "fsl,board-localbus";
 #address-cells = <2>;
 #size-cells = <1>;
 reg = <e0005000 d8>;    // BRx, ORx, etc

 ranges = <0 0 ff000000 1000000     // nor flash,  16 MB
                   1 0 60000000 4000000>;   // nand flash, 64 MB

 [EMAIL PROTECTED],0 {
     compatible = "atmel,29LV256", "cfi-flash";
     reg = <0 0 1000000>;
     bank-width = <2>;
     device-width = <1>;
 };

 [EMAIL PROTECTED],0 {
     compatible = "stmicro,NAND512W3A", "fsl,upm-nand";
     reg = <1 0 4000000>;   //reg = <1 0 1>;
     width = <1>;
     upm = "A";
     upm-addr-offset = <16>;
     upm-cmd-offset = <8>;
     gpios = <4 18>;
     gpio-parent = <&qe_pio>;
     wait-pattern;
     wait-write;
 };
};

The CFI driver is at linux/drivers/mtd/chips
The UPM driver is at linux/drivers/mtd/nand

So, how can I load the UPM driver before physmap?

Thanks in advance,

--
Alemao


------------------------------

Message: 6
Date: Thu, 9 Oct 2008 16:05:47 -0500
From: "Ayman El-Khashab" <[EMAIL PROTECTED]>
Subject: RE: Toolchain for ppc-440 without FPU
To: "Ademir Zanetti Junior" <[EMAIL PROTECTED]>,
<linuxppc-embedded@ozlabs.org>
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset="us-ascii"

I was compiling for the 440FP under cygwin, I believe that I had to use
--enable-long-long when I configured gcc in order to build glibc.

________________________________

From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On
Behalf Of Ademir Zanetti Junior
Sent: Thursday, October 09, 2008 2:08 PM
To: linuxppc-embedded@ozlabs.org
Subject: Toolchain for ppc-440 without FPU


Hello all,

I am trying to create a toolchain by hand and I am facing some problems
when I try to configure the glibc 2.7 for a powerpc. When I try to
configure the glibc to install the headers I get the following error:

configure --prefix=/usr --host=powerpc-440-linux-gnu
--with-headers=${SYSROOT}/usr/include --disable-shared --without-cvs
--with-binutils=/home/azanetti/ml507/powerpc/bin/

checking build system type... i686-pc-linux-gnu
checking host system type... i686-pc-linux-gnu
configure: running configure fragment for add-on nptl
checking sysdep dirs... sysdeps/i386/elf
nptl/sysdeps/unix/sysv/linux/i386/i686 nptl/sysdeps/unix/sysv/linux/i386
sysdeps/unix/sysv/linux/i386 nptl/sysdeps/unix/sysv/linux
nptl/sysdeps/pthread sysdeps/pthread sysdeps/unix/sysv/linux sysdeps/gnu
sysdeps/unix/common sysdeps/unix/mman sysdeps/unix/inet
sysdeps/unix/sysv/i386 nptl/sysdeps/unix/sysv sysdeps/unix/sysv
sysdeps/unix/i386 nptl/sysdeps/unix sysdeps/unix sysdeps/posix
sysdeps/i386/i686/fpu nptl/sysdeps/i386/i686 sysdeps/i386/i686
sysdeps/i386/i486 nptl/sysdeps/i386/i486 sysdeps/i386/fpu
nptl/sysdeps/i386 sysdeps/i386 sysdeps/wordsize-32
sysdeps/ieee754/ldbl-96 sysdeps/ieee754/dbl-64 sysdeps/ieee754/flt-32
sysdeps/ieee754 sysdeps/generic/elf sysdeps/generic
checking for a BSD-compatible install... /usr/bin/install -c
checking whether ln -s works... yes
checking for gcc... gcc
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ISO C89... none needed
checking how to run the C preprocessor... gcc -E
checking for g++... g++

.
.
.
checking for long double... yes
checking size of long double... 12
running configure fragment for sysdeps/powerpc/powerpc32/elf
checking for powerpc32 TLS support... no
running configure fragment for sysdeps/unix/sysv/linux/powerpc
checking whether gcc -B/home/azanetti/ml507/powerpc/bin/ -g -O2
-mlong-double-128 uses IBM extended format... no
checking whether gcc -B/home/azanetti/ml507/powerpc/bin/ -g -O2 supports
-mabi=ibmlongdouble... no
configure: error: this configuration requires -mlong-double-128 IBM
extended format support


I already have the binutils 2.18 compiled on my PATH. I know that it is
easier to use a script like crosstool to compile it but I do want to
understand what is happening behind the scenes.

Regards.

Ademir

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://ozlabs.org/pipermail/linuxppc-embedded/attachments/20081009/c91ac7a9/attachment.htm>

------------------------------

_______________________________________________
Linuxppc-embedded mailing list
Linuxppc-embedded@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-embedded

End of Linuxppc-embedded Digest, Vol 50, Issue 14
*************************************************




------------------------------

Message: 3
Date: Thu, 9 Oct 2008 16:56:26 -0700 (PDT)
From: Nguyen Duc Quan <[EMAIL PROTECTED]>
Subject: RE: API access I2C
To: Duy-Ky Nguyen <[EMAIL PROTECTED]>,
linuxppc-embedded@ozlabs.org
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="utf-8"

I think I2C-tools is what you need as sample code to access I2C in Linux. 
Here is the link
http://dl.lm-sensors.org/i2c-tools/releases/i2c-tools-3.0.1.tar.bz2
There are a lot of tools to access, set/get/dump register, read/write the 
I2C eeprom...
But please use them with care, especially when you try to read/write the 
device at address 0x50 ( it's is bootstrap eeprom).
You might not boot up your board after playing with this device :-)
Quan Nguyen



----- Th? g?c ----
T?: Duy-Ky Nguyen <[EMAIL PROTECTED]>
??n: linuxppc-embedded@ozlabs.org
G?i ng?y: Th? S?u, 10 th?ng 10, 2008 6:11:22
Ch? ??: API access I2C

Hello,

I have an eva board MPC8313E-RDB with 3 I2C device at address 48, 50, 68.
I'm able to read data from those 3 devices

I follow linux/Documentation/i2c/dev-interface to create a C code to 
access
those 3 devices in Linux env, but all are failed !!!

I really appreciate your help where to find  info or sample code running 
in
Linyux to access I2C devices

Thanks so much,

Duy-Ky

----- Original Message ----- 
From: <[EMAIL PROTECTED]>
To: <linuxppc-embedded@ozlabs.org>
Sent: Thursday, October 09, 2008 2:05 PM
Subject: Linuxppc-embedded Digest, Vol 50, Issue 14


Send Linuxppc-embedded mailing list submissions to
linuxppc-embedded@ozlabs.org

To subscribe or unsubscribe via the World Wide Web, visit
https://ozlabs.org/mailman/listinfo/linuxppc-embedded
or, via email, send a message with subject or body 'help' to
[EMAIL PROTECTED]

You can reach the person managing the list at
[EMAIL PROTECTED]

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Linuxppc-embedded digest..."


Today's Topics:

  1. Re: performance: memcpy vs. __copy_tofrom_user (Leon Woestenberg)
  2. Re: performance: memcpy vs. __copy_tofrom_user (Matt Sealey)
  3. Toolchain for ppc-440 without FPU (Ademir Zanetti Junior)
  4. Re: FS_ENET ERROR(s) 0x12 at second NFS RPC port lookup
     (100005/1) (Remi Lefevre)
  5. MTD map drivers (FSL UPM NAND) (Alemao)
  6. RE: Toolchain for ppc-440 without FPU (Ayman El-Khashab)


----------------------------------------------------------------------

Message: 1
Date: Thu, 9 Oct 2008 14:04:53 +0200
From: "Leon Woestenberg" <[EMAIL PROTECTED]>
Subject: Re: performance: memcpy vs. __copy_tofrom_user
To: "Dominik Bozek" <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED], Paul Mackerras <[EMAIL PROTECTED]>,
linuxppc-embedded@ozlabs.org
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1

Hello all,

On Thu, Oct 9, 2008 at 1:41 PM, Dominik Bozek <[EMAIL PROTECTED]>
wrote:
Paul Mackerras wrote:
Dominik Bozek writes:
Actually I made couple of other tests on that mpc8313. Most of them 
are
to ugly to publish them, but... My problem is that I have to boost the
gigabit interface on the mpc8313. I made simple substitution and

Very interesting.  Can you work out where memcpy is being called on
the network data?  I wouldn't have expected that.


Also see this recent thread David Jander on August 25th, "Efficient
memcpy()/memmove() for G2/G3 cores..."
on [EMAIL PROTECTED]

http://ozlabs.org/pipermail/linuxppc-dev/2008-September/062449.html

BTW, I am interested in this work as well, I'm currently working with
a 8313 and 8315 design, both are e300 cores.

My current goal is PCIe though, but I need fast GbE performance as well.

Also, did you test Freescale's 2.6.24.3 patches that tune the gianfar
interfaces for higher performance?

Regards,
--
Leon


------------------------------

Message: 2
Date: Thu, 09 Oct 2008 10:37:55 -0500
From: Matt Sealey <[EMAIL PROTECTED]>
Subject: Re: performance: memcpy vs. __copy_tofrom_user
To: Paul Mackerras <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED], Dominik Bozek <[EMAIL PROTECTED]>,
linuxppc-embedded@ozlabs.org
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=UTF-8; format=flowed

Paul Mackerras wrote:
Dominik Bozek writes:

Actually I made couple of other tests on that mpc8313. Most of them are
to ugly to publish them, but... My problem is that I have to boost the
gigabit interface on the mpc8313. I made simple substitution and
__copy_tofrom_user was used instead of memcpy. I know, it's wrong, but 
I
speedup that way the network interface for about 10%.

Very interesting.  Can you work out where memcpy is being called on
the network data?  I wouldn't have expected that.

It probably is somewhere.. through some weird and wonderful code path 
that
needs some serious digging to find. At least in 2.4 memcpy was used and
optimizing it (see Freescale's libmotovec benchmarks) did produce a
sizable
performance improvement. That, and offloading TCP checksumming to AltiVec
helped a lot.

No help at all on an 8313 but, relevant anyway.

Since then zero copy networking and other fancy things like the DMA
engine API (for intel ioat at least but also there is fsl dma support)
there's less to actually optimize now so you're less likely to see the
same benefits. All these got into mainline because it's essential to
have this kind of architecture to get reasonable speeds out of >gigabit
network links.

There is actually no strong reason not to use __copy_tofrom_user as
memcpy, in fact, as long as we are sure that source and destination
are both cacheable.

I do think there is probably a good benefit in doing things like zeroing
pages in AltiVec and copying entire pages with AltiVec (for instance
when copy-on-write happens in an application) - NetBSD and QNX implement
at least this because it's faster than using the cache management and
works fine on uncacheable pages too (also since you're always aligned to
a page, zeroing 4kb aligned to a 4kb boundary - or whatever your page
size happens to be, the number of errors that can occur are absolutely
tiny and performance can go through the roof).

Ahem, but nobody here wants AltiVec in the kernel do they?

--
Matt Sealey <[EMAIL PROTECTED]>
Genesi, Manager, Developer Relations


------------------------------

Message: 3
Date: Thu, 9 Oct 2008 16:08:18 -0300
From: "Ademir Zanetti Junior" <[EMAIL PROTECTED]>
Subject: Toolchain for ppc-440 without FPU
To: linuxppc-embedded@ozlabs.org
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset="utf-8"

Hello all,

I am trying to create a toolchain by hand and I am facing some problems
when
I try to configure the glibc 2.7 for a powerpc. When I try to configure
the
glibc to install the headers I get the following error:

configure --prefix=/usr --host=powerpc-440-linux-gnu
--with-headers=${SYSROOT}/usr/include --disable-shared --without-cvs
--with-binutils=/home/azanetti/ml507/powerpc/bin/

checking build system type... i686-pc-linux-gnu
checking host system type... i686-pc-linux-gnu
configure: running configure fragment for add-on nptl
checking sysdep dirs... sysdeps/i386/elf
nptl/sysdeps/unix/sysv/linux/i386/i686 nptl/sysdeps/unix/sysv/linux/i386
sysdeps/unix/sysv/linux/i386 nptl/sysdeps/unix/sysv/linux
nptl/sysdeps/pthread sysdeps/pthread sysdeps/unix/sysv/linux sysdeps/gnu
sysdeps/unix/common sysdeps/unix/mman sysdeps/unix/inet
sysdeps/unix/sysv/i386 nptl/sysdeps/unix/sysv sysdeps/unix/sysv
sysdeps/unix/i386 nptl/sysdeps/unix sysdeps/unix sysdeps/posix
sysdeps/i386/i686/fpu nptl/sysdeps/i386/i686 sysdeps/i386/i686
sysdeps/i386/i486 nptl/sysdeps/i386/i486 sysdeps/i386/fpu
nptl/sysdeps/i386
sysdeps/i386 sysdeps/wordsize-32 sysdeps/ieee754/ldbl-96
sysdeps/ieee754/dbl-64 sysdeps/ieee754/flt-32 sysdeps/ieee754
sysdeps/generic/elf sysdeps/generic
checking for a BSD-compatible install... /usr/bin/install -c
checking whether ln -s works... yes
checking for gcc... gcc
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ISO C89... none needed
checking how to run the C preprocessor... gcc -E
checking for g++... g++

.
.
.
checking for long double... yes
checking size of long double... 12
running configure fragment for sysdeps/powerpc/powerpc32/elf
checking for powerpc32 TLS support... no
running configure fragment for sysdeps/unix/sysv/linux/powerpc
checking whether gcc -B/home/azanetti/ml507/powerpc/bin/ -g -O2
-mlong-double-128 uses IBM extended format... no
checking whether gcc -B/home/azanetti/ml507/powerpc/bin/ -g -O2 supports
-mabi=ibmlongdouble... no
configure: error: this configuration requires -mlong-double-128 IBM
extended
format support


I already have the binutils 2.18 compiled on my PATH. I know that it is
easier to use a script like crosstool to compile it but I do want to
understand what is happening behind the scenes.

Regards.

Ademir
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://ozlabs.org/pipermail/linuxppc-embedded/attachments/20081009/a3dd71d6/attachment-0001.htm>

------------------------------

Message: 4
Date: Thu, 9 Oct 2008 21:46:42 +0200
From: "Remi Lefevre" <[EMAIL PROTECTED]>
Subject: Re: FS_ENET ERROR(s) 0x12 at second NFS RPC port lookup
(100005/1)
To: "Scott Wood" <[EMAIL PROTECTED]>
Cc: linuxppc-embedded@ozlabs.org
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1

Ok, I'm sorry, I have incorrectly set the 60x pipeline depth to 0
instead of 1 in BCR register.
It works fine now. The side effect is interesting to know...

regards,
r?mi


------------------------------

Message: 5
Date: Thu, 9 Oct 2008 18:04:25 -0300
From: Alemao <[EMAIL PROTECTED]>
Subject: MTD map drivers (FSL UPM NAND)
To: linuxppc-embedded@ozlabs.org
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset=ISO-8859-1

Hi all,

Im trying to use a physmap driver for NOR and NAND flash devices.

With NOR, everything ok. It uses CFI driver.

The problem is with NAND, cause it is connected at Freescale
LocalBus UPM, and at a first look, the UPM driver is loading after the
physmap (lines 14-15 in kernel output)

Here some outputs from the board:

Bootloader (U-Boot 1.2.0):

CPU:   e300c1, MPC8360E, Rev: 21 at 499.999 MHz, CSB:  333 MHz
Board: MPC8360KTX
I2C:   ready
DDR RAM: 128 MB
FLASH: 16 MB
NAND:  64 MiB
In:    serial
Out:   serial
Err:   serial


Kernel (linux-2.6.17):

1               physmap nor flash device: 1000000 at ff000000
2               NOR flash: Found 1 x16 devices at 0x0 in 16-bit bank
3                Amd/Fujitsu Extended Query Table at 0x0040
4                number of CFI chips: 1
5               Using physmap partition definition
6               Creating 6 MTD partitions on "NOR flash":
7               0x00000000-0x00040000 : "u-boot"
8               0x00040000-0x00060000 : "env."
9               0x00060000-0x00080000 : "sys"
10              0x00080000-0x000c0000 : "logs"
11              0x000c0000-0x00860000 : "admin"
12              0x00860000-0x01000000 : "users"
13
14              physmap nand flash device: 4000000 at 60000000
15              UPM: User-Programmable Machine NAND driver


Device Tree Source:

[EMAIL PROTECTED] {
 compatible = "fsl,board-localbus";
 #address-cells = <2>;
 #size-cells = <1>;
 reg = <e0005000 d8>;    // BRx, ORx, etc

 ranges = <0 0 ff000000 1000000     // nor flash,  16 MB
                   1 0 60000000 4000000>;   // nand flash, 64 MB

 [EMAIL PROTECTED],0 {
     compatible = "atmel,29LV256", "cfi-flash";
     reg = <0 0 1000000>;
     bank-width = <2>;
     device-width = <1>;
 };

 [EMAIL PROTECTED],0 {
     compatible = "stmicro,NAND512W3A", "fsl,upm-nand";
     reg = <1 0 4000000>;   //reg = <1 0 1>;
     width = <1>;
     upm = "A";
     upm-addr-offset = <16>;
     upm-cmd-offset = <8>;
     gpios = <4 18>;
     gpio-parent = <&qe_pio>;
     wait-pattern;
     wait-write;
 };
};

The CFI driver is at linux/drivers/mtd/chips
The UPM driver is at linux/drivers/mtd/nand

So, how can I load the UPM driver before physmap?

Thanks in advance,

--
Alemao


------------------------------

Message: 6
Date: Thu, 9 Oct 2008 16:05:47 -0500
From: "Ayman El-Khashab" <[EMAIL PROTECTED]>
Subject: RE: Toolchain for ppc-440 without FPU
To: "Ademir Zanetti Junior" <[EMAIL PROTECTED]>,
<linuxppc-embedded@ozlabs.org>
Message-ID:
<[EMAIL PROTECTED]>
Content-Type: text/plain; charset="us-ascii"

I was compiling for the 440FP under cygwin, I believe that I had to use
--enable-long-long when I configured gcc in order to build glibc.

________________________________

From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On
Behalf Of Ademir Zanetti Junior
Sent: Thursday, October 09, 2008 2:08 PM
To: linuxppc-embedded@ozlabs.org
Subject: Toolchain for ppc-440 without FPU


Hello all,

I am trying to create a toolchain by hand and I am facing some problems
when I try to configure the glibc 2.7 for a powerpc. When I try to
configure the glibc to install the headers I get the following error:

configure --prefix=/usr --host=powerpc-440-linux-gnu
--with-headers=${SYSROOT}/usr/include --disable-shared --without-cvs
--with-binutils=/home/azanetti/ml507/powerpc/bin/

checking build system type... i686-pc-linux-gnu
checking host system type... i686-pc-linux-gnu
configure: running configure fragment for add-on nptl
checking sysdep dirs... sysdeps/i386/elf
nptl/sysdeps/unix/sysv/linux/i386/i686 nptl/sysdeps/unix/sysv/linux/i386
sysdeps/unix/sysv/linux/i386 nptl/sysdeps/unix/sysv/linux
nptl/sysdeps/pthread sysdeps/pthread sysdeps/unix/sysv/linux sysdeps/gnu
sysdeps/unix/common sysdeps/unix/mman sysdeps/unix/inet
sysdeps/unix/sysv/i386 nptl/sysdeps/unix/sysv sysdeps/unix/sysv
sysdeps/unix/i386 nptl/sysdeps/unix sysdeps/unix sysdeps/posix
sysdeps/i386/i686/fpu nptl/sysdeps/i386/i686 sysdeps/i386/i686
sysdeps/i386/i486 nptl/sysdeps/i386/i486 sysdeps/i386/fpu
nptl/sysdeps/i386 sysdeps/i386 sysdeps/wordsize-32
sysdeps/ieee754/ldbl-96 sysdeps/ieee754/dbl-64 sysdeps/ieee754/flt-32
sysdeps/ieee754 sysdeps/generic/elf sysdeps/generic
checking for a BSD-compatible install... /usr/bin/install -c
checking whether ln -s works... yes
checking for gcc... gcc
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ISO C89... none needed
checking how to run the C preprocessor... gcc -E
checking for g++... g++

.
.
.
checking for long double... yes
checking size of long double... 12
running configure fragment for sysdeps/powerpc/powerpc32/elf
checking for powerpc32 TLS support... no
running configure fragment for sysdeps/unix/sysv/linux/powerpc
checking whether gcc -B/home/azanetti/ml507/powerpc/bin/ -g -O2
-mlong-double-128 uses IBM extended format... no
checking whether gcc -B/home/azanetti/ml507/powerpc/bin/ -g -O2 supports
-mabi=ibmlongdouble... no
configure: error: this configuration requires -mlong-double-128 IBM
extended format support


I already have the binutils 2.18 compiled on my PATH. I know that it is
easier to use a script like crosstool to compile it but I do want to
understand what is happening behind the scenes.

Regards.

Ademir

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://ozlabs.org/pipermail/linuxppc-embedded/attachments/20081009/c91ac7a9/attachment.htm>

------------------------------

_______________________________________________
Linuxppc-embedded mailing list
Linuxppc-embedded@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-embedded

End of Linuxppc-embedded Digest, Vol 50, Issue 14
*************************************************


_______________________________________________
Linuxppc-embedded mailing list
Linuxppc-embedded@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-embedded



     B?n c? ? ki?n v? M? T?m?
H?y c?ng chia s? v?i m?i ng??i t?i Yahoo! Vi?t Nam Music
http://vn.music.yahoo.com/artist/100000056?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 
<http://ozlabs.org/pipermail/linuxppc-embedded/attachments/20081009/0c4f9884/attachment.htm>

------------------------------

_______________________________________________
Linuxppc-embedded mailing list
Linuxppc-embedded@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-embedded

End of Linuxppc-embedded Digest, Vol 50, Issue 15
*************************************************


_______________________________________________
Linuxppc-embedded mailing list
Linuxppc-embedded@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-embedded