From: Li Yang-r58472 <hidden> Date: 2004-06-09 04:21:27
Hi
I am confused about the maintaining mechanism of kernel.org tree and linuxppc tree. How a patch goes into linuxppc tree? And how patch from linuxppc tree goes into kernel.org tree? Is there any patch that goes directly into kernel.org tree? Who are the decision makers of merging a patch for both trees?
Regards,
Leo
** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/
From: Paul Mackerras <hidden> Date: 2004-06-09 04:49:46
Li Yang-r58472 writes:
I am confused about the maintaining mechanism of kernel.org tree and
linuxppc tree. How a patch goes into linuxppc tree? And how patch
from linuxppc tree goes into kernel.org tree? Is there any patch
that goes directly into kernel.org tree? Who are the decision
makers of merging a patch for both trees?
I am the overall ppc32 maintainer, meaning that I ultimately get to
say what goes into arch/ppc and include/asm-ppc. Andrew Morton is the
maintainer of the 2.6 kernel.org tree, meaning that he gets to say
what goes in or doesn't, anywhere in the tree.
The only way that any patch gets into the 2.6 kernel.org tree is by
someone sending a patch to Andrew Morton. In the ppc parts of the
tree, it would be me or one of the maintainers for the ppc subsystems:
Tom Rini for bootloader and 6xx/7xx/7xxx embedded stuff, Ben
Herrenschmidt for Power Macintosh stuff, Matt Porter for 4xx embedded
stuff, or Dan Malek or Tom Rini for 8xx embedded stuff. There is
nothing to prevent other people from sending patches directly to
Andrew Morton, of course, but unless it's a completely trivial and
obvious patch, Andrew will ask me or one of the people listed above
for my/their opinion before accepting it.
As for the linuxppc trees, there are about 10 people that can make
changes in those trees. Changes checked in to the linuxppc trees do
*not* automatically flow upstream to the kernel.org trees - I, or one
of the subsystem maintainers, has to prepare a patch with a suitably
detailed description and send it to Andrew Morton. This means that a
change checked into linuxppc-2.5 is unlikely to go upstream unless
either (a) one of the ppc subsystem maintainers notices it and makes
the effort to send it upstream or (b) the author of the change
prepares a patch with description and asks one of the ppc subsystem
maintainers to send it upstream.
Patches can go directly into the kernel.org tree without first going
into linuxppc-2.5. Such patches will automatically get imported to
the linuxppc-2.5 tree when someone does a pull from the kernel.org
tree (this is a manual process, but it's easy). Since ppc patches are
going into the kernel.org tree quickly and easily these days, this is
actually the least-effort way to go in most cases. One day perhaps we
will be able to get away from having any linuxppc tree and just work
with the kernel.org tree all the time.
Paul.
** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/
Hi,
I met a strange problem on FLASH programming with
BDI2000 on a 8xx board.After programming a 11.7MB
ramdisk to 4*8bit Intel FLASH[4*28F128J3A=64MB],the
main menu of application in ramdisk couldn't
appear.But if I did this job with U-BOOT,everything
went well.Also the ramdisk programmed by BDI2000 was
checked "right" as following with crc32 by
U-BOOT.Meanwhile,I have been using BDI2000 good for
programming U-BOOT and Linux kernel.So weird... I hope
this is not a bug of BDI2000 fireware... Could anyone
encounter such strangeness before?
Image at FE000000:
Image Name: ramdisk for 5-25 test
Image Type: PowerPC Linux RAMDisk Image (gzip
compressed)
Data Size: 12235275 Bytes = 11.7 MB
Load Address: 00000000
Entry Point: 00000000
Verifying Checksum ... OK
[root@sam bdi]# ./bdisetup -v -p/dev/ttyS1 -b57 -s
BDI Type : BDI2000 Rev.C (SN: 94402134)
Loader : V1.05
Firmware : V1.12 bdiGDB for MPC8xx/MPC5xx
Logic : V1.02 MPC8xx/MPC5xx
MAC : 00-0c-01-94-40-21
IP Addr : 172.16.115.8
Subnet : 255.255.255.0
Gateway : 255.255.255.255
Host IP : 172.16.115.6
Config : dmtless+.cfg
----------------------------------------------------
; bdiGDB configuration file for custom 8xx board
[INIT]
; init core register
WSPR 149 0x2002000F ;DER : set
debug enable register
;;WSPR 149 0x2006000F ;DER : enable
SYSIE for BDI flash progr.
WSPR 638 0xFA200000 ;IMMR :
internal memory at 0xFFF00000
; init SIU register
WM32 0xFA200000 0x00000C00 ;SIUMCR
WM32 0xFA200004 0xFFFFFF89 ;SYPCR
[TARGET]
CPUCLOCK 40000000 ;; the CPU clock rate after
processing the init list
BDIMODE AGENT ;; the BDI working mode
(LOADONLY | AGENT)
BREAKMODE HARD ;; SOFT or HARD, HARD uses PPC
hardware breakpoints
[HOST]
IP 172.16.115.6
FILE u-boot
FORMAT ELF
LOAD MANUAL ;load code MANUAL or AUTO
after reset
[FLASH]
WORKSPACE 0xFA202000
CHIPTYPE STRATAX8 ;; Flash type (AM29F |
AM29BX8 | AM29BX16 | I28BX8 | I28BX16)
CHIPSIZE 0x1000000 ;; The size of one flash
chip in bytes
BUSWIDTH 32 ;; The width of the flash
memory bus in bits (8 | 16 | 32)
FILE u-boot.bin ;; The file to program
FORMAT BIN 0x00000000
ERASE 0xfff00000 BLOCK
ERASE 0xfff80000 BLOCK
[REGS]
DMM1 0xFA200000
FILE reg823e.def
** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/
Hi,
I added 8xx RTC driver both on U-Boot and Linux
kernel.And I could set current time right under
U-Boot.But I have no idea on making it through Linux
application?I could read it with following code but
how to set it?
#include <time.h>
time_t t;
struct tm * tm;
time(&t);
tm=localtime(&t);
printf("\nThe corrent time is : %02d.%02d.%02d
%02d:%02d:%02d\n",tm->tm_year+1900,tm->tm_mon+1,tm->tm_mday,tm->tm_hour,tm->tm_min,tm->tm_sec);
Thanks in advance!
Sam
** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/
If you are using ELDK-3.0 and it's development root filesystem, hwclock
does the trick.
David
I added 8xx RTC driver both on U-Boot and Linux
kernel.And I could set current time right under
U-Boot.But I have no idea on making it through Linux
application?I could read it with following code but
how to set it?