[U-Boot-Users] [PATCH] fix for "make doesn't know how to make target in examples/"

newer
Re: [U-Boot-Users] ld error : EABI...

older
[U-Boot-Users] striping u-boot

Wen-chien Jesse Sung

10 Jul 2006 10 Jul '06

9:45 a.m.

Hi,

Make complains that it doesn't know how to make hello_world.srec.

make[1]: *** No rule to make target `hello_world.srec', needed by `all'. Stop.

Introduce $(TARGET) let make to build % before %.srec and %.bin.

CHANGELOG

* Introduce $(TARGET) in Makefile of examples/.

-- Best Regards, Wen-chien Jesse Sung

Attachments:

0002-fix-for-make-doesn-t-know-how-to-make-target-in-examples.txt.gz (application/x-gzip — 1010 bytes)

Show replies by date

Wolfgang Denk

10 Jul 10 Jul

10:14 a.m.

In message 44B20597.4010103@cola.voip.idv.tw you wrote:

...

Make complains that it doesn't know how to make hello_world.srec.

Is it correct to assume that you are using a (broken) BETA version of the GNU make tool?

I reject your patch. AFAIK, any sane version of make handles the current Makefile correctly.

Please get working tools.

Best regards,

Wolfgang Denk

-- Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de Man did not weave the web of life; he is merely a strand in it. Whatever he does to the web, he does to himself. - Seattle [1854]

Wen-chien Jesse Sung

11:27 a.m.

New subject: [U-Boot-Users] [PATCH] fix for "make doesn't know how to make target in examples/"

Wolfgang Denk wrote:

...

...
Make complains that it doesn't know how to make hello_world.srec.

Is it correct to assume that you are using a (broken) BETA version of the GNU make tool?

I reject your patch. AFAIK, any sane version of make handles the current Makefile correctly.

Please get working tools.

Hi,

I'm using make 3.81-2 that is in Debian etch and sid currently.

$ make --version GNU Make 3.81

$ dpkg -l make ii make 3.81-2 The GNU version of the "make" utility.

A test with the attached Makefile: $ touch 1.src $ make make: *** No rule to make target `1.out', needed by `all'. Stop.

I would file a bug against make on Debian BTS if that is believed to be one.

-- Best Regards, Wen-chien Jesse Sung %.out: % cp $< $@ echo $@ %: %.src cp $< $@ echo $@ all: 1.out echo $@

Andreas Schweigstill

12:28 p.m.

New subject: [U-Boot-Users] [PATCH] fix for "make doesn't know how to make target in examples/"

Dear all!

Wen-chien Jesse Sung schrieb:

...

%.out: % cp $< $@ echo $@

%: %.src cp $< $@ echo $@

all: 1.out echo $@

I have tested the Makefile on a Suse 9.3 Professional system with GNU make 3.80 (RPM package version make-3.80-187) and it fails there, too.

But if I add the line

1.out: 1

everything is working.

With best regards Andreas Schweigstill

-- Dipl.-Phys. Andreas Schweigstill Schweigstill IT | Embedded Systems Schauenburgerstraße 116, D-24118 Kiel, Germany Phone: (+49) 431 5606-435, Fax: (+49) 431 5606-436 Mobile: (+49) 171 6921973, Web: http://www.schweigstill.de/

Wolfgang Denk

2:51 p.m.

Hello,

in message 44B21D7F.5070603@cola.voip.idv.tw you wrote:

...

I'm using make 3.81-2 that is in Debian etch and sid currently.

This is supposed to be ok.

...

A test with the attached Makefile: $ touch 1.src $ make make: *** No rule to make target `1.out', needed by `all'. Stop.

This Makefile is not equivlent to what we use in the U-Boot code.

...

all: 1.out

The "examples/Makefile" is different; it corresponds to:

all: 1.out 1

...and this works fine for me with make 3.81

Best regards,

Wolfgang Denk

-- Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de The first 90% of a project takes 90% of the time, the last 10% takes the other 90% of the time.

Wen-chien Jesse Sung

3:45 p.m.

New subject: [U-Boot-Users] [PATCH] fix for "make doesn't know how to make target in examples/"

Wolfgang Denk wrote:

...

...
A test with the attached Makefile: $ touch 1.src $ make make: *** No rule to make target `1.out', needed by `all'. Stop.

This Makefile is not equivlent to what we use in the U-Boot code.

...
all: 1.out

The "examples/Makefile" is different; it corresponds to:

all: 1.out 1

...and this works fine for me with make 3.81

Hi,

I modify the testing Makefile (attached).

$ make make: *** No rule to make target `1.out', needed by `all'. Stop.

And then I replace "all: 1.out 1" with "all: 1 1.out", it works.

The sequence of the prerequisites affect the result for me, and that's what the patch does: place $(TARGET) before $(SREC) and $(BIN). Besides, it removes the duplicated entries, SREC = foo.srec BIN = foo.bin foo now becomes SREC = foo.srec only.

I'm just curious about the difference of make's behavior. :)

-- Best Regards, Wen-chien Jesse Sung %.out: % cp $< $@ echo $@ %: %.src cp $< $@ echo $@ all: 1.out 1 echo $@

Wolfgang Denk

6:41 p.m.

In message 44B259F8.8090902@cola.voip.idv.tw you wrote:

...

$ make make: *** No rule to make target `1.out', needed by `all'. Stop.

And then I replace "all: 1.out 1" with "all: 1 1.out", it works.

Aha! This is exactly what I wanted to know. This is the specific failure mode tha has been reported couple times before. See for example http://sourceforge.net/mailarchive/message.php?msg_id=15318222

...

The sequence of the prerequisites affect the result for me, and that's

Yes, this is the observation all users of GNU make version 3.81 made, and which I consider a bug in make. Nothing in the 3.80 documentation gives any hint that the order of the file names in the prerequisite list (to be precise, we're talking about `normal' prerequisites here, not about `order-only" prerequisites) might be important for the result of the depemdency checking. Actually I am convinced that there must be no such dependency on the order, as this can lead to circular dependencies that would be impossible to solve.

So as a summary: your version of make is broken.

Best regards,

Wolfgang Denk

-- Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de In a survey taken several years ago, all incoming freshmen at MIT were asked if they expected to graduate in the top half of their class. Ninety-seven percent responded that they did.

6876

Age (days ago)

6876

Last active (days ago)

List overview

Download

6 comments

3 participants

tags (0)

participants (3)

Andreas Schweigstill
Wen-chien Jesse Sung
Wolfgang Denk