the rtems kernel source list

2016-04-21 Thread printk
Hello,  

I am learning RTEMS, where is the kernel source, I find some

place seems holding the kernel source:  



  rtems/c/src/support/ -  only version.c there. 

  rtems/c/src/ada/ -  is this kernel source?



  rtems/cpukit/score/src   why named score? 

  rtems/cpukit/rtems/src   kernel source?   



  rtems/score/cpu/$(ARCH) --- cpu depended code?



is there any more code belone to the RTEMS kernel.  

thanks very much. please kindly help.   



- - 

duhuanpeng


___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel


arm shared code contains not shared code?

2016-04-21 Thread printk
Hello,
I am learning the rtems source, and when i look into this folder:

rtems@book64:~/rtems-lite/rtems/c/src/lib/libcpu/arm/shared$ tree
.
├── arm920
│   └── mmu.c
└── include
├── am335x.h
├── arm-cp15.h
├── cache_.h
├── mmu.h
├── omap3.h
└── omap_timer.h

2 directories, 7 files
rtems@book64:~/rtems-lite/rtems/c/src/lib/libcpu/arm/shared$ 

this folder i guess it means stuff shared by arm arch? or something
else?
why these files here to, it seems can not 'share' with other
arm 'SoC'
  am335x.h
  omap3.h
  omap_timer.h
How about move these file to some place machine depended folders?
I guess the arm just means the arm cpu in the SoC, not a arm SoC.

thanks,
duhuanpeng

___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

build failed: cpukit/posix/src/conddefaultattributes.c

2016-04-23 Thread printk
Hi, I followed this to build the bsp, arm920
- - -
be built to target the BOARD_SUPPORT_PACKAGE board.
mkdir build-rtems
cd build-rtems
../rtems-4.10.0/configure --target= \
--disable-posix --disable-networking --disable-cxx \
--enable-rtemsbsp=\
--prefix=
make all
make install
- - -
$ cd rtems
$ mkdir arm920
$ cd arm920
$ ../configure --target=arm-rtems4.12
$ make RTEMS_BSP="arm920"

it fails here:
- - -
arm-rtems4.12-gcc --pipe -DHAVE_CONFIG_H   -I.. 
-I../../cpukit/../../../arm920/lib/include   -mcpu=arm920 -O2 -g 
-ffunction-sections -fdata-sections -Wall -Wmissing-prototypes 
-Wimplicit-function-declaration -Wstrict-prototypes -Wnested-externs -MT 
src/libposix_a-conddefaultattributes.o -MD -MP -MF 
src/.deps/libposix_a-conddefaultattributes.Tpo -c -o 
src/libposix_a-conddefaultattributes.o `test -f 'src/conddefaultattributes.c' 
|| echo 
'../../../../../../c/src/../../cpukit/posix/'`src/conddefaultattributes.c
../../../../../../c/src/../../cpukit/posix/src/conddefaultattributes.c:31:3: 
error: unknown field 'clock' specified in initializer
   .clock  = CLOCK_REALTIME,
   ^
- - -
I find this breaks many other bsps too, I comment this line out, then I can 
compile all other arm bsps.(about 20hours, 20G :D)
did I do something wrong?

btw, there are many sources include this file
 #include 
but i can not find this file by:
 find -name tree.h
where is this file?

- - -
duhuanpeng


___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel


Re: build failed: cpukit/posix/src/conddefaultattributes.c

2016-04-23 Thread printk
On Sat, Apr 23, 2016 at 08:04:35AM -0500, Joel Sherrill wrote:
> The tools have been updating pretty regularly. This means your tools are
> old. Update the RSB and build me tools.
sorry, i forgot to attach my gcc version, it was built some days ago:

- - - - - - - - - -
rtems@book64:~$ arm-rtems4.12-gcc --version
arm-rtems4.12-gcc (GCC) 6.0.0 20160124 (RTEMS 4.12, RSB 
662c2e73028a8a8ac293d295a04629312f4f6e5a-modified, Newlib 2.3.0.20160104)
Copyright (C) 2016 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

rtems@book64:~$ arm-rtems4.12-gcc -v
Using built-in specs.
COLLECT_GCC=arm-rtems4.12-gcc
COLLECT_LTO_WRAPPER=/home/du/dev/rtems/4.12/libexec/gcc/arm-rtems4.12/6.0.0/lto-wrapper
Target: arm-rtems4.12
Configured with: ../gcc-6-20160124/configure --prefix=/home/du/dev/rtems/4.12 
--bindir=/home/du/dev/rtems/4.12/bin --exec_prefix=/home/du/dev/rtems/4.12 
--includedir=/home/du/dev/rtems/4.12/include 
--libdir=/home/du/dev/rtems/4.12/lib 
--libexecdir=/home/du/dev/rtems/4.12/libexec 
--mandir=/home/du/dev/rtems/4.12/share/man 
--infodir=/home/du/dev/rtems/4.12/share/info 
--datadir=/home/du/dev/rtems/4.12/share --build=x86_64-linux-gnu 
--host=x86_64-linux-gnu --target=arm-rtems4.12 --disable-libstdcxx-pch 
--with-gnu-as --with-gnu-ld --verbose --with-newlib --with-system-zlib 
--disable-nls --without-included-gettext --disable-win32-registry 
--enable-version-specific-runtime-libs --disable-lto 
--enable-newlib-io-c99-formats --enable-newlib-iconv 
--enable-newlib-iconv-encodings=big5,cp775,cp850,cp852,cp855,cp866,euc_jp,euc_kr,euc_tw,iso_8859_1,iso_8859_10,iso_8859_11,iso_8859_13,iso_8859_14,iso_8859_15,iso_8859_2,iso_8859_3,iso_8859_4,iso_8859_5,iso_8859_6,iso_8859_7,iso_8859_8,iso_8859_9,iso_
 
ir_111,koi8_r,koi8_ru,koi8_u,koi8_uni,ucs_2,ucs_2_internal,ucs_2be,ucs_2le,ucs_4,ucs_4_internal,ucs_4be,ucs_4le,us_ascii,utf_16,utf_16be,utf_16le,utf_8,win_1250,win_1251,win_1252,win_1253,win_1254,win_1255,win_1256,win_1257,win_1258
 --enable-threads --disable-plugin --enable-libgomp --enable-languages=c,c++
Thread model: rtems
gcc version 6.0.0 20160124 (RTEMS 4.12, RSB 
662c2e73028a8a8ac293d295a04629312f4f6e5a-modified, Newlib 2.3.0.20160104) (GCC) 
rtems@book64:~$ 
- - - - - - - - - -
it is this toolchain too old? 

> 
> Between gcc and newlib activity, we have been bumping them every few weeks.
> One at the first of March added new POSIX methods and condition variable
> attributes.
I will build a new one and try again. :D, it will take anther day to built it 
again.
> 
> --joel
thanks very much!


> On Apr 23, 2016 6:50 AM, "printk"  wrote:
> 
> Hi, I followed this to build the bsp, arm920
> - - -
> be built to target the BOARD_SUPPORT_PACKAGE board.
> mkdir build-rtems
> cd build-rtems
> ../rtems-4.10.0/configure --target= \
> --disable-posix --disable-networking --disable-cxx \
> --enable-rtemsbsp=\
> --prefix=
> make all
> make install
> - - -
> $ cd rtems
> $ mkdir arm920
> $ cd arm920
> $ ../configure --target=arm-rtems4.12
> $ make RTEMS_BSP="arm920"
> 
> it fails here:
> - - -
> arm-rtems4.12-gcc --pipe -DHAVE_CONFIG_H   -I..
> -I../../cpukit/../../../arm920/lib/include   -mcpu=arm920 -O2 -g
> -ffunction-sections -fdata-sections -Wall -Wmissing-prototypes
> -Wimplicit-function-declaration -Wstrict-prototypes -Wnested-externs -MT
> src/libposix_a-conddefaultattributes.o -MD -MP -MF
> src/.deps/libposix_a-conddefaultattributes.Tpo -c -o
> src/libposix_a-conddefaultattributes.o `test -f
> 'src/conddefaultattributes.c' || echo
> '../../../../../../c/src/../../cpukit/posix/'`src/conddefaultattributes.c
> ../../../../../../c/src/../../cpukit/posix/src/conddefaultattributes.c:31:3:
> error: unknown field 'clock' specified in initializer
>.clock  = CLOCK_REALTIME,
>^
> - - -
> I find this breaks many other bsps too, I comment this line out, then I can
> compile all other arm bsps.(about 20hours, 20G :D)
> did I do something wrong?
> 
> btw, there are many sources include this file
>  #include 
> but i can not find this file by:
>  find -name tree.h
> where is this file?
> 
> - - -
> duhuanpeng
> 
> 
> ___
> devel mailing list
> devel@rtems.org
> http://lists.rtems.org/mailman/listinfo/devel

...
duhuanpeng.

___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel


Re: the rtems kernel source list

2016-04-23 Thread printk
Now I seems understand a bit more.
the rtems/c/src folder is seems cantain the rtems itself writen in C
language.
   ada/   (what is it?, ada interface?)
   lib/   hw depended, seems very different naming from other project
   libchip/   some device driver
   librtems++/(a hack for C++ interface, thanks chris)
   support/   version.c

cpukit/   why do you called cpukit?
the rtems/cpukit/ folder is other middle-ware or applications, but 
but also rtems itself...
   dev/i2c driver, seems need to clean up.
libi2c/i2c driver, do you want to live in dev/ ?
libpci/driver?
libdrvmgr/ driver framwork?
---> does rtems have it's own driver framework?

   mghttpd/app?
   pppd/   app
   telnetd/app
   ftpd/   app

   libnetworking/  net 

   dtc/host app, Device Tree Compiler

   libfs/  fs, mid-ware
   libblock/   mid-ware, block

   librpc/ ?, need to dig

   libcrypt/   crypto, mid-ware
libmd/ crypto, mid-ware
   zlib/   compress, mid-ware?

   libcsupport/c lib? what kind of c lib calls from newlib, what from here?

   libdl/  need to dig

   rtems/  api(rtems classic api)
sapi/  ??
libstdthreads/ api layer?
posix/ api layer
score/ api layer
libgnat/   api layer?
libqos/thread qos? only a header there..


   libmisc/mini cpukit in cpukit?

:D, just a leaning homework, please kindly corret me.

 

On Fri, Apr 22, 2016 at 01:02:03PM +1000, Chris Johns wrote:
> On 22/04/2016 12:19, printk wrote:
> >Hello,
> >I am learning RTEMS, where is the kernel source, I find some
> >place seems holding the kernel source:
> >
> >   rtems/c/src/support/ -  only version.c there.
> >   rtems/c/src/ada/ -  is this kernel source?
> 
> No.
> 
> >
> >   rtems/cpukit/score/src   why named score?
> 
> Supercore or score, it sits behind the POSIX and Classic APIs and is
> a superset of the various APIs that export the kernel to users, eg
> POSIX.
> 
> >   rtems/cpukit/rtems/src   kernel source?
> 
> RTEMS Classic API.
> 
> >   rtems/score/cpu/$(ARCH) --- cpu depended code?
> 
> Yes.
> 
> >is there any more code belone to the RTEMS kernel.
> >thanks very much. please kindly help.
> 
> The BSP code, and a chip driver library (libchip).
> 
> Chris
> ___
> devel mailing list
> devel@rtems.org
> http://lists.rtems.org/mailman/listinfo/devel

...
duhuanpeng.

___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel


Re: build failed: cpukit/posix/src/conddefaultattributes.c

2016-04-24 Thread printk
Ah, I works now. thank you all!
I guess rtems maybe not easy to compile with other
toolchain. eg. armcc, arm-none-eabi-gcc(ARM: 
https://launchpad.net/gcc-arm-embedded).

...
duhuanpeng.

___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel


Re: build failed: cpukit/posix/src/conddefaultattributes.c

2016-04-24 Thread printk
On Sun, Apr 24, 2016 at 05:55:09AM -0500, Joel Sherrill wrote:
> We do not support building RTEMS with anything but the RTEMS toolchain.
> There are very specific configurations of the language run times and C
> Library for RTEMS. Those account for our thread model and SMP safety.
could you please tell me more about the thread model and SMP saftety?
where is the code to read or ?

> On Apr 24, 2016 10:14 AM, "printk"  wrote:
> 
> > Ah, I works now. thank you all!
> > I guess rtems maybe not easy to compile with other
> > toolchain. eg. armcc, arm-none-eabi-gcc(ARM:
> > https://launchpad.net/gcc-arm-embedded).
> >
> > ...
> > duhuanpeng.
> >
> >

> ___
> devel mailing list
> devel@rtems.org
> http://lists.rtems.org/mailman/listinfo/devel


...
duhuanpeng.
___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel


start trying to build rtems with kbuild

2016-04-25 Thread printk
hi, a11:
I am trying to build rtems with kbuild on linux.
just a hobby, make rtems easier to build on linux.
not support freebsd, windows...
(I guess some company don't use windows or linux,
securty is critical for them. for small companies,
e.g. shenzhen small companies, convenien is more
important.)
but this may make it hard to ship a bsp in lib.
anyway, just for fun. :D
for more infomation about kbuild building and config,
there maybe some video on youtube: compile linux kernel.

repo: (push later)
http://github.com/zhang3/rtems-kbuild

here is some build log:
- - - -
rtems@book:~/rtems-kbuild$ make
  CC  c/src/lib/libcpu/arm/s3c24xx/clock/clockdrv.o
  CC  c/src/lib/libcpu/arm/s3c24xx/clock/support.o
  LD  c/src/lib/libcpu/arm/s3c24xx/clock/built-in.o
  CC  c/src/lib/libcpu/arm/s3c24xx/irq/irq.o
  LD  c/src/lib/libcpu/arm/s3c24xx/irq/built-in.o
  CC  c/src/lib/libcpu/arm/s3c24xx/timer/timer.o
  LD  c/src/lib/libcpu/arm/s3c24xx/timer/built-in.o
  LD  c/src/lib/libcpu/arm/s3c24xx/built-in.o
  LINKvmlinux
  LD  vmlinux.o
arm-rtems-ld: cannot find arch/arm/kernel/head.o: No such file or directory
c/src/lib/libcpu/arm/s3c24xx/: file not recognized: Is a directory
Makefile:970: recipe for target 'vmlinux' failed
make: *** [vmlinux] Error 1

- - - -

rtems@book:~/rtems-kbuild$ make V=1
make -f ./scripts/Makefile.build obj=c/src/lib/libcpu/arm/s3c24xx
make -f ./scripts/Makefile.build obj=c/src/lib/libcpu/arm/s3c24xx/clock
  arm-rtems-gcc -Wp,-MD,c/src/lib/libcpu/arm/s3c24xx/clock/.clockdrv.o.d  
-Ic/src/lib/libcpu/arm/s3c24xx/include -Ic/src/lib/libcpu/arm/include 
-Ic/src/lib/libcpu/arm/shared/include -Icpukit/score/include 
-Icpukit/dev/include -Icpukit/include -Icpukit/score/cpu/arm -Ibsp-include 
-DHAVE_CONFIG_H -DCPU_S3C2410 -D__rtems__ -mlittle-endian 
-Ic/src/lib/libcpu/arm/s3c24xx/include -fno-dwarf2-cfi-asm -fno-ipa-sra -marm 
-D__LINUX_ARM_ARCH__=5 -march=armv5te -mtune=arm9tdmi -msoft-float -Uarm 
-fno-delete-null-pointer-checks -O2 --param=allow-store-data-races=0 
-fno-stack-protector -Wno-unused-but-set-variable -fomit-frame-pointer 
-fno-var-tracking-assignments -Wdeclaration-after-statement -Wno-pointer-sign 
-fno-strict-overflow -fconserve-stack -Werror=implicit-int 
-Werror=strict-prototypes -Werror=date-time -DCPU_S3C2410 -specs 
./c/src/lib/libbsp/arm/smdk2410/bsp_specs-D"KBUILD_STR(s)=#s" 
-D"KBUILD_BASENAME=KBUILD_STR(clockdrv)"  
-D"KBUILD_MODNAME=KBUILD_STR(clockdrv)" -c -o c/src/lib/lib
 cpu/arm/s3c24xx/clock/clockdrv.o c/src/lib/libcpu/arm/s3c24xx/clock/clockdrv.c
  arm-rtems-gcc -Wp,-MD,c/src/lib/libcpu/arm/s3c24xx/clock/.support.o.d  
-Ic/src/lib/libcpu/arm/s3c24xx/include -Ic/src/lib/libcpu/arm/include 
-Ic/src/lib/libcpu/arm/shared/include -Icpukit/score/include 
-Icpukit/dev/include -Icpukit/include -Icpukit/score/cpu/arm -Ibsp-include 
-DHAVE_CONFIG_H -DCPU_S3C2410 -D__rtems__ -mlittle-endian 
-Ic/src/lib/libcpu/arm/s3c24xx/include -fno-dwarf2-cfi-asm -fno-ipa-sra -marm 
-D__LINUX_ARM_ARCH__=5 -march=armv5te -mtune=arm9tdmi -msoft-float -Uarm 
-fno-delete-null-pointer-checks -O2 --param=allow-store-data-races=0 
-fno-stack-protector -Wno-unused-but-set-variable -fomit-frame-pointer 
-fno-var-tracking-assignments -Wdeclaration-after-statement -Wno-pointer-sign 
-fno-strict-overflow -fconserve-stack -Werror=implicit-int 
-Werror=strict-prototypes -Werror=date-time -DCPU_S3C2410 -specs 
./c/src/lib/libbsp/arm/smdk2410/bsp_specs-D"KBUILD_STR(s)=#s" 
-D"KBUILD_BASENAME=KBUILD_STR(support)"  -D"KBUILD_MODNAME=KBUILD_STR(support)" 
-c -o c/src/lib/libcpu
 /arm/s3c24xx/clock/support.o c/src/lib/libcpu/arm/s3c24xx/clock/support.c
   arm-rtems-ld -EL-r -o c/src/lib/libcpu/arm/s3c24xx/clock/built-in.o 
c/src/lib/libcpu/arm/s3c24xx/clock/clockdrv.o 
c/src/lib/libcpu/arm/s3c24xx/clock/support.o
(cat /dev/null; ) > c/src/lib/libcpu/arm/s3c24xx/clock/modules.order
make -f ./scripts/Makefile.build obj=c/src/lib/libcpu/arm/s3c24xx/irq
  arm-rtems-gcc -Wp,-MD,c/src/lib/libcpu/arm/s3c24xx/irq/.irq.o.d  
-Ic/src/lib/libcpu/arm/s3c24xx/include -Ic/src/lib/libcpu/arm/include 
-Ic/src/lib/libcpu/arm/shared/include -Icpukit/score/include 
-Icpukit/dev/include -Icpukit/include -Icpukit/score/cpu/arm -Ibsp-include 
-DHAVE_CONFIG_H -DCPU_S3C2410 -D__rtems__ -mlittle-endian 
-Ic/src/lib/libcpu/arm/s3c24xx/include -fno-dwarf2-cfi-asm -fno-ipa-sra -marm 
-D__LINUX_ARM_ARCH__=5 -march=armv5te -mtune=arm9tdmi -msoft-float -Uarm 
-fno-delete-null-pointer-checks -O2 --param=allow-store-data-races=0 
-fno-stack-protector -Wno-unused-but-set-variable -fomit-frame-pointer 
-fno-var-tracking-assignments -Wdeclaration-after-statement -Wno-pointer-sign 
-fno-strict-overflow -fconserve-stack -Werror=implicit-int 
-Werror=strict-prototypes -Werror=date-time -DCPU_S3C2410 -specs 
./c/src/lib/libbsp/arm/smdk2410/bsp_specs-D"KBUILD_STR(s)=#s" 
-D"KBUILD_BASENAME=KBUILD_STR(irq)"  -D"KBUILD_MODNAME=KBUILD_STR(irq)" -c -o 
c/src/lib/libcpu/arm/s3c24xx/i
 rq/irq.o c/src/l

when the getcwd.c is compiled?

2016-04-27 Thread printk
Hello, I am translating Makefile.am(s) to kbuild

> # 
> # DIRECTORY_SCAN_C_FILES =
> # ## Newlib SHOULD have provided this one
> # DIRECTORY_SCAN_C_FILES += src/readdir_r.c
> # ## Comment out when using these from newlib's unix directory
> # DIRECTORY_SCAN_C_FILES += src/getcwd.c
> # 
> # ID_C_FILES = src/getegid.c src/geteuid.c src/getgid.c src/getgroups.c \
> # src/getlogin.c src/getpgrp.c src/getpid.c src/getppid.c src/getuid.c \
> # src/seteuid.c src/setgid.c src/setuid.c src/setegid.c src/setpgid.c \
> # src/setsid.c
> 
> CONFIG_DIRECTORY_SCAN_C_FILES := y
> obj-$(CONFIG_DIRECTORY_SCAN_C_FILES) += readdir_r.o
> #ERROR obj-$(CONFIG_DIRECTORY_SCAN_C_FILES) += getcwd.o

--> # ## Comment out when using these from newlib's unix directory
is this means when compiling with newlib, the getcwd.c is not needed?
now the arm-rtemsX.XX-gcc is with newlib. is this file not compiled?

...
duhuanpeng.
u74147@gmail.com


___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

Re: when the getcwd.c is compiled?

2016-04-27 Thread printk
On Wed, Apr 27, 2016 at 11:30:59AM -0500, Joel Sherrill wrote:
> On Apr 27, 2016 12:31 PM, "printk"  wrote:
> >
> > Hello, I am translating Makefile.am(s) to kbuild
> >
> > > #
> > > # DIRECTORY_SCAN_C_FILES =
> > > # ## Newlib SHOULD have provided this one
> > > # DIRECTORY_SCAN_C_FILES += src/readdir_r.c
> > > # ## Comment out when using these from newlib's unix directory
> > > # DIRECTORY_SCAN_C_FILES += src/getcwd.c
> > > #
> > > # ID_C_FILES = src/getegid.c src/geteuid.c src/getgid.c src/getgroups.c
> \
> > > # src/getlogin.c src/getpgrp.c src/getpid.c src/getppid.c
> src/getuid.c \
> > > # src/seteuid.c src/setgid.c src/setuid.c src/setegid.c
> src/setpgid.c \
> > > # src/setsid.c
> > >
> > > CONFIG_DIRECTORY_SCAN_C_FILES := y
> > > obj-$(CONFIG_DIRECTORY_SCAN_C_FILES) += readdir_r.o
> > > #ERROR obj-$(CONFIG_DIRECTORY_SCAN_C_FILES) += getcwd.o
> >
> > --> # ## Comment out when using these from newlib's unix directory
> > is this means when compiling with newlib, the getcwd.c is not needed?
> > now the arm-rtemsX.XX-gcc is with newlib. is this file not compiled?
> 
> That's how I would interpret it. Are the files sti present?
> 
> Are the methods really coming from newlib?

I pull the lastest(20160428) newlib, find getcwd.c there:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
> du@book:~/source/newlib-cygwin$ find -name getcwd.c
> ./newlib/libc/unix/getcwd.c
> ./libgloss/spu/getcwd.c
> 
> du@book:~/source/newlib-cygwin$ head -3 ./newlib/libc/unix/getcwd.c 
> #ifndef _NO_GETCWD
> /*
> * Copyright (c) 1989, 1991 The Regents of the University of California.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
in RTEMS, the getcwd.c is compiled by checking the build log.

so shall we remove this file or add a macro to control? like:

> ifdef NEWLIB
> # DIRECTORY_SCAN_C_FILES += src/getcwd.c
> endif
> not comment it out by hand.
> 
> by the way,(1) what is this libcsupport's goal?

is files exist in standar C lib(newlib) can be removed?
or mark as a option?

> 
> > ...
> > duhuanpeng.
> > u74147@gmail.com
> >
> >
> > ___
> > devel mailing list
> > devel@rtems.org
> > http://lists.rtems.org/mailman/listinfo/devel

> ___
> devel mailing list
> devel@rtems.org
> http://lists.rtems.org/mailman/listinfo/devel


...
duhuanpeng.
u74147@gmail.com
___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

dtc: compile error

2016-04-27 Thread printk
Hi, I use this cflags to compile rtems/cpukit/dtc/*,
I get error message below:

- - - - - - - - - - - - - - - - - - - - 
make -f ./scripts/Makefile.build obj=cpukit/dtc/libfdt
  arm-rtems-gcc -Wp,-MD,cpukit/dtc/libfdt/.fdt_addresses.o.d  
-Ic/src/lib/libbsp/shared/include -Ic/src/lib/libcpu/arm/s3c24xx/include 
-Ic/src/lib/libcpu/arm/in
clude -Ic/src/lib/libcpu/arm/shared/include -Icpukit -Icpukit/score/include 
-Icpukit/include -Icpukit/score/cpu/arm -DHAVE_CONFIG_H -DCPU_S3C2410 
-D__rtems__ -I
include -mlittle-endian -Ic/src/lib/libcpu/arm/s3c24xx/include 
-fno-dwarf2-cfi-asm -mno-unaligned-access -fno-ipa-sra -marm 
-D__LINUX_ARM_ARCH__=5 -march=armv5t
e -mtune=arm9tdmi -msoft-float -Uarm -fno-delete-null-pointer-checks -O2 
--param=allow-store-data-races=0 -fno-stack-protector 
-Wno-unused-but-set-variable -fom
it-frame-pointer -fno-var-tracking-assignments -Wno-pointer-sign 
-fno-strict-overflow -fconserve-stack -Werror=implicit-int 
-Werror=strict-prototypes -Werror=da
te-time -Icpukit/dtc/libfdt-D"KBUILD_STR(s)=#s" 
-D"KBUILD_BASENAME=KBUILD_STR(fdt_addresses)"  
-D"KBUILD_MODNAME=KBUILD_STR(fdt_addresses)" -c -o cpukit/dtc
/libfdt/fdt_addresses.o cpukit/dtc/libfdt/fdt_addresses.c
In file included from cpukit/dtc/libfdt/fdt_addresses.c:54:0:
cpukit/dtc/libfdt/libfdt.h: In function 'fdt_set_magic':
cpukit/dtc/libfdt/libfdt.h:192:7: error: dereferencing pointer to incomplete 
type 'struct fdt_header'
   fdth->name = cpu_to_fdt32(val); \
   ^
cpukit/dtc/libfdt/libfdt.h:194:1: note: in expansion of macro '__fdt_set_hdr'
 __fdt_set_hdr(magic);
 ^
cpukit/dtc/libfdt/libfdt.h: In function 'fdt_set_totalsize':
cpukit/dtc/libfdt/libfdt.h:192:7: error: dereferencing pointer to incomplete 
type 'struct fdt_header'
   fdth->name = cpu_to_fdt32(val); \
   ^
cpukit/dtc/libfdt/libfdt.h:195:1: note: in expansion of macro '__fdt_set_hdr'
 __fdt_set_hdr(totalsize);
 ^
cpukit/dtc/libfdt/libfdt.h: In function 'fdt_set_off_dt_struct':
cpukit/dtc/libfdt/libfdt.h:192:7: error: dereferencing pointer to incomplete 
type 'struct fdt_header'
- - - - - - - - - - - - - - - - - - - - 

the original cc flags is:

- - - - - - - - - - - - - - - - - - - - 
arm-rtems4.12-gcc --pipe -DHAVE_CONFIG_H   -I../.. 
-I../../../cpukit/../../../smdk2410/lib/include 
-I../../../../../../../rtems/c/src/../../cpukit/dtc/libfdt   -mcpu=arm920t 
-DCPU_S3C2410 -O2 -g -ffunction-sections -fdata-sections -Wall 
-Wmissing-prototypes -Wimplicit-function-declaration -Wstrict-prototypes 
-Wnested-externs -MT libfdt_a-fdt_addresses.o -MD -MP -MF 
.deps/libfdt_a-fdt_addresses.Tpo -c -o libfdt_a-fdt_addresses.o `test -f 
'fdt_addresses.c 
- - - - - - - - - - - - - - - - - - - - 

 
here is my gcc version:
- - - - - - - - - - - - - - - - - - - - 
rtems@book:~/rtems-kbuild$ arm-rtems-gcc -v
Using built-in specs.
COLLECT_GCC=arm-rtems-gcc
COLLECT_LTO_WRAPPER=/home/opt/gcc-arm-rtems/libexec/gcc/arm-rtems/6.0.1/lto-wrapper
Target: arm-rtems
Configured with: ../gcc-6.0.1-RC-20160415/configure 
--prefix=/home/opt/gcc-arm-rtems --bindir=/home/opt/gcc-arm-rtems/bin 
--exec_prefix=/home/opt/gcc-arm-rtems 
--includedir=/home/opt/gcc-arm-rtems/include 
--libdir=/home/opt/gcc-arm-rtems/lib 
--libexecdir=/home/opt/gcc-arm-rtems/libexec 
--mandir=/home/opt/gcc-arm-rtems/share/man 
--infodir=/home/opt/gcc-arm-rtems/share/info 
--datadir=/home/opt/gcc-arm-rtems/share --build=x86_64-linux-gnu 
--host=x86_64-linux-gnu --target=arm-rtems --disable-libstdcxx-pch 
--with-gnu-as --with-gnu-ld --verbose --with-newlib --with-system-zlib 
--disable-nls --without-included-gettext --disable-win32-registry 
--enable-version-specific-runtime-libs --disable-lto 
--enable-newlib-io-c99-formats --enable-newlib-iconv 
--enable-newlib-iconv-encodings=big5,cp775,cp850,cp852,cp855,cp866,euc_jp,euc_kr,euc_tw,iso_8859_1,iso_8859_10,iso_8859_11,iso_8859_13,iso_8859_14,iso_8859_15,iso_8859_2,iso_8859_3,iso_8859_4,iso_8859_5,iso_8859_6,iso_8859_7,iso_8859_8,iso_8859_9,iso_ir_111,koi8_r,koi8_ru,koi8_u,koi8_uni,ucs_2,ucs_2_internal,ucs_2be,ucs_2le,ucs_4,ucs_4_internal,ucs_4be,ucs_4le,us_ascii,utf_16,utf_16be,utf_16le,utf_8,win_1250,win_1251,win_1252,win_1253,win_1254,win_1255,win_1256,win_1257,win_1258
 --enable-threads --disable-plugin --enable-libgomp --enable-languages=c,c++
Thread model: rtems
gcc version 6.0.1 20160415 (RTEMS 4.12, RSB 
8bcc58b249fbf1f6e7bbd4151d9cf9b04c94b87e-modified, Newlib 2.4.0, little-endian) 
(GCC)
- - - - - - - - - - - - - - - - - - - - 

what is wrong?
thanks!!

...
duhuanpeng.
u74147@gmail.com


___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

codingstyle: usage of #include and #include "head.h"

2016-04-28 Thread printk
What is the difference in RTEMS?

there are code use #include "xxx.h" reference to xxx.h not located
in the current folder.
how about just use #include "xxx.h" to include the file in current folder?
and use #include  to include files in -isystem -I...

this usage find in 
  cpukit/lib/networking/

...
duhuanpeng.
u74147@gmail.com


___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

waf: wil the new build system copy header files to include as autoconf?

2016-04-30 Thread printk
Hi, will the new build system copy the headers to include too?

...
duhuanpeng.
u74147@gmail.com


___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

Re: waf: wil the new build system copy header files to include as autoconf?

2016-05-03 Thread printk
On Sun, May 01, 2016 at 11:44:23PM +, Amar Takhar wrote:
> On 2016-04-30 23:50 +0800, printk wrote:
> > Hi, will the new build system copy the headers to include too?
> 
> No, the new build system which you can see here:
> 
>   https://git.rtems.org/amar/waf.git/
> 
> Does not do this.
> 
> You can see how to build it and get more details by following this thread:
> 
>   https://lists.rtems.org/pipermail/devel/2015-February/009837.html
> 
> 
> Amar.
> ___
> devel mailing list
> devel@rtems.org
> http://lists.rtems.org/mailman/listinfo/devel

hello, Amar,

very happy to heard from you and thank you. I am trinng compile & config RTEMS 
with kbuild:
so i need to fix the includes. (-I) option.


 .config - RTEMS/arm 4.12 Kernel Configuration  
 -- 
  +-- RTEMS/arm 4.12 Kernel Configuration --+   
  |  Arrow keys navigate the menu.   selects submenus ---> (or empty |   
  |  submenus ).  Highlighted letters are hotkeys.  Pressing |   
  |  includes,  excludes,  modularizes features.  Press  to |   
  |  exit,  for Help,  for Search.  Legend: [*] built-in  [ ] |   
  | +-+ |   
  | |Linux example  --->  | |   
  | |*** RTEMS ***| |   
  | |libcpu  ---> | |   
  | |libbsp  ---> | |   
  | |c  --->  | |   
  | |cpukit  ---> | |   
  | | | |   
  | | | |   
  | | | |   
  | | | |   
  | | | |   
  | +-+ |   
  +-+   
  |< Exit >< Help >< Save >< Load > |   
  +-+   





...
duhuanpeng.
u74147@gmail.com


___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel