Question on examples compiling
Hello,
I'm just writing a support for a new board based on mips (au1100) and I cannot understand why I have to apply this patch:
examples/Makefile +2 -2
@@ -60,7 +60,7 @@ endif include $(TOPDIR)/config.mk SREC = hello_world.srec -BIN = hello_world.bin hello_world +BIN = hello_world hello_world.bin ifeq ($(CPU),mpc8xx) SREC = test_burst.srec
@@ -122,7 +122,7 @@ clibdir := $(shell dirname `$(CC) $(CFLA CPPFLAGS += -I.. -all: .depend $(OBJS) $(LIB) $(SREC) $(BIN) +all: .depend $(OBJS) $(LIB) $(BIN) $(SREC) ######################################################################### $(LIB): .depend $(LIBOBJS)
in order to have files into example directory to be compiled? =:-o
Rodolfo
Signature · 4 lines
GNU/Linux Solutions e-mail: giometti at enneenne.com Linux Device Driver giometti at gnudd.com Embedded Systems giometti at linux.it UNIX programming phone: +39 349 2432127
Quoted 3 lines · In message <[email protected]> you wrote:
I'm just writing a support for a new board based on mips (au1100) and I cannot understand why I have to apply this patch:
...
Quoted 1 line
in order to have files into example directory to be compiled? =:-o
Because your toolchain (especially the "make" command) is seriously broken.
Best regards,
Wolfgang Denk
Signature · 3 lines
Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de A rolling stone gathers momentum.
Quoted 3 lines · Wolfgang Denk
Because your toolchain (especially the "make" command) is seriously broken.
I'm using this version (default Debian unstable):
giometti at zaigor:~$ make --version
GNU Make 3.81rc2
Copyright (C) 2006 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.
This program built for i486-pc-linux-gnu
Is this version broken?
Rodolfo
Signature · 11 lines
GNU/Linux Solutions e-mail: giometti at enneenne.com Linux Device Driver giometti at gnudd.com Embedded Systems giometti at linux.it UNIX programming phone: +39 349 2432127 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://lists.denx.de/pipermail/u-boot/attachments/20060411/09f17c28/attachment.pgp
Quoted 2 lines · In message <[email protected]> you wrote:
I'm using this version (default Debian unstable):
-----------------------------------------^^^^^^^^^
Do I need to say more?
Quoted 2 lines
giometti at zaigor:~$ make --version GNU Make 3.81rc2
...
Quoted 1 line
Is this version broken?
Just try a stable version, and you will not need to ask.
Yes, it is broken. See previous reports on this list. Anything up to and including 3.80 seems to be fine. 3.81 and later seems to be broken.
Best regards,
Wolfgang Denk
Signature · 5 lines
Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de The more complex the mind, the greater the need for the simplicity of play. -- Kirk, "Shore Leave", stardate 3025.8
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Wolfgang Denk writes:
Rodolfo> I'm using this version (default Debian unstable):
Wolfgang> -----------------------------------------^^^^^^^^^
Wolfgang> Do I need to say more?
I won't exaggerate. Debian unstable' is stable enough even for production use. For really unstable version, you have to use experimental.'
Rodolfo> giometti at zaigor:~$ make --version GNU Make 3.81rc2
Rodolfo> Is this version broken?
Wolfgang> Just try a stable version, and you will not need to ask.
Wolfgang> Yes, it is broken. See previous reports on this Wolfgang> list. Anything up to and including 3.80 seems to be Wolfgang> fine. 3.81 and later seems to be broken.
There are many discussions about what's really broken - make or Makefiles. The 3.81 caused entire Linux kernel to compile each time any single file was changed. Make developers claim it's a kernel Makefiles problem. Fortunately, for the specific problem which is discussed here the future outcome of this argument does not matter because there is a simple fix. I changed the order of dependencies in the examples/Makefile and the problem disappeared. I have no time to investigate if it's make or Makefile problem (though I personally think it's the latter). I attached the diff just in case you'd like to use it.
Signature · 68 lines
======================================================================== Yuli Barcohen | Phone +972-9-765-1788 | Software Project Leader yuli at arabellasw.com | Fax +972-9-765-7494 | Arabella Software, Israel ======================================================================== -------------- next part -------------- Index: examples/Makefile =================================================================== RCS file: /home/CVS/u-boot/u-boot/examples/Makefile,v retrieving revision 1.1.1.12.2.3 diff -p -u -r1.1.1.12.2.3 Makefile --- examples/Makefile 29 May 2005 19:40:40 -0000 1.1.1.12.2.3 +++ examples/Makefile 11 Apr 2006 11:37:37 -0000 @@ -56,39 +56,39 @@ endif include $(TOPDIR)/config.mk SREC = hello_world.srec -BIN = hello_world.bin hello_world +BIN = hello_world hello_world.bin ifeq ($(CPU),mpc8xx) SREC = test_burst.srec -BIN = test_burst.bin test_burst +BIN = test_burst test_burst.bin endif ifeq ($(ARCH),i386) SREC += 82559_eeprom.srec -BIN += 82559_eeprom.bin 82559_eeprom +BIN += 82559_eeprom 82559_eeprom.bin endif ifeq ($(ARCH),ppc) SREC += sched.srec -BIN += sched.bin sched +BIN += sched sched.bin endif # The following example is pretty 8xx specific... ifeq ($(CPU),mpc8xx) SREC += timer.srec -BIN += timer.bin timer +BIN += timer timer.bin endif # The following example is 8260 specific... ifeq ($(CPU),mpc8260) SREC += mem_to_mem_idma2intr.srec -BIN += mem_to_mem_idma2intr.bin mem_to_mem_idma2intr +BIN += mem_to_mem_idma2intr mem_to_mem_idma2intr.bin endif # Utility for resetting i82559 EEPROM ifeq ($(BOARD),oxc) SREC += eepro100_eeprom.srec -BIN += eepro100_eeprom.bin eepro100_eeprom +BIN += eepro100_eeprom eepro100_eeprom.bin endif ifeq ($(BIG_ENDIAN),y) @@ -113,7 +113,7 @@ clibdir := $(shell dirname `$(CC) $(CFLA CPPFLAGS += -I.. -all: .depend $(OBJS) $(LIB) $(SREC) $(BIN) +all: .depend $(OBJS) $(LIB) $(BIN) $(SREC) ######################################################################### $(LIB): .depend $(LIBOBJS)
Dear Yuli,
Quoted 3 lines · in message <[email protected]> you wrote:
I won't exaggerate. Debian `unstable' is stable enough even for production use. For really unstable version, you have to use
Did you ever try to explain this to any manager in a company with more than 200 employees? But agreed, that's off topic here.
Quoted 2 lines
There are many discussions about what's really broken - make or Makefiles. The 3.81 caused entire Linux kernel to compile each time any
It is *obviously* make which is broken.
Quoted 2 lines
problem. Fortunately, for the specific problem which is discussed here the future outcome of this argument does not matter because there is a
Yes, it does matter. Broken tools are a serious problem to all of us, especiallyu when it's such basic tools as make ot gcc or ld.
Quoted 3 lines
simple fix. I changed the order of dependencies in the examples/Makefile and the problem disappeared. I have no time to investigate if it's make or
Makefileproblem (though I personally think it's the latter). I
No. The result of a make run MUST NEVER depend on the order of the dependencies. This is plain unacceptable to me. I'm not even going to discuss this.
Best regards,
Wolfgang Denk
Signature · 5 lines
Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de Here is an Appalachian version of management's answer to those who are concerned with the fate of the project: "Don't worry about the mule. Just load the wagon." - Mike Dennison's hillbilly uncle