[LTP] fix lib6 test building

[LTP] fix lib6 test building

[Paul Larson]

[-- Attachment #1: Type: text/plain, Size: 188 bytes --]

The lib6 tests seem to fail the first pass through building for me
because ranlib does not seem to get set correctly at any point.  This
seems to fix it up for me.

Thanks,
Paul Larson




[-- Attachment #2: lib6-makefix.patch --]
[-- Type: text/x-patch, Size: 573 bytes --]

Fix build problem with lib6 tests


Signed-off-by: Paul Larson <paul.larson@canonical.com>
---

diff --git a/testcases/network/lib6/Makefile b/testcases/network/lib6/Makefile
index 7c1784d..f622468 100644
--- a/testcases/network/lib6/Makefile
+++ b/testcases/network/lib6/Makefile
@@ -23,6 +23,10 @@
 CFLAGS+=	-I../../../include -g -D_GNU_SOURCE
 LOADLIBES+=	-L../../../lib runcc.a -lltp -lpthread
 
+ifeq ($(RANLIB),)
+RANLIB = ranlib
+endif
+
 SRCS=$(wildcard *.c)
 TARGETS= asapi_01 asapi_02 asapi_03 asapi_04 asapi_05 asapi_06 asapi_07 \
 	getaddrinfo_01 in6_01 in6_02

[-- Attachment #3: Type: text/plain, Size: 355 bytes --]

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july

[-- Attachment #4: Type: text/plain, Size: 155 bytes --]

_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] fix lib6 test building

[Garrett Cooper]

On Thu, Aug 6, 2009 at 7:52 AM, Paul Larson<paul.larson@canonical.com> wrote:
> The lib6 tests seem to fail the first pass through building for me
> because ranlib does not seem to get set correctly at any point.  This
> seems to fix it up for me.

Fixed on HEAD moments ago:

gcooper@orangebox /scratch/ltp-dev2/ltp $ cvs diff
testcases/network/lib6/Makefile
Index: testcases/network/lib6/Makefile
===================================================================
RCS file: /cvsroot/ltp/ltp/testcases/network/lib6/Makefile,v
retrieving revision 1.6
diff -r1.6 Makefile
25a26,27
> RANLIB                ?= ranlib
>

The problem is that while configure specifies RANLIB, make doesn't. So
if you don't run from a top-level make, there's no guarantee [today]
that the right variables will be set. This will be fixed soon.

If cross-compiling, please specify $(CROSS_COMPILE)ranlib, and the
compile will go through in the child make.

Thanks,
-Garrett

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] fix lib6 test building

[Paul Larson]

Garrett Cooper wrote:
> On Thu, Aug 6, 2009 at 7:52 AM, Paul Larson<paul.larson@canonical.com> wrote:
>> The lib6 tests seem to fail the first pass through building for me
>> because ranlib does not seem to get set correctly at any point.  This
>> seems to fix it up for me.
> 
> Fixed on HEAD moments ago:
> 
> gcooper@orangebox /scratch/ltp-dev2/ltp $ cvs diff
> testcases/network/lib6/Makefile
> Index: testcases/network/lib6/Makefile
> ===================================================================
> RCS file: /cvsroot/ltp/ltp/testcases/network/lib6/Makefile,v
> retrieving revision 1.6
> diff -r1.6 Makefile
> 25a26,27
>> RANLIB                ?= ranlib
The patch you checked in does not correct the build failure.

Thanks,
Paul Larson

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] fix lib6 test building

[Henry Yei]

[-- Attachment #1: Type: text/plain, Size: 1768 bytes --]

Paul, 

The patch above applied to version 1.5 of the make file works fo me across multiple architectures. It turned out that RANLIB was not needed, as long as the appropriate AR was used. It had been hardcoded in 1.4. Let me know if that works. 
 

-----Original Message-----
From: Paul Larson [mailto:paul.larson@canonical.com] 
Sent: Wednesday, August 12, 2009 8:13 AM
To: Garrett Cooper
Cc: ltp-list@lists.sourceforge.net
Subject: Re: [LTP] fix lib6 test building

Garrett Cooper wrote:
> On Thu, Aug 6, 2009 at 7:52 AM, Paul Larson<paul.larson@canonical.com> wrote:
>> The lib6 tests seem to fail the first pass through building for me 
>> because ranlib does not seem to get set correctly at any point.  This 
>> seems to fix it up for me.
> 
> Fixed on HEAD moments ago:
> 
> gcooper@orangebox /scratch/ltp-dev2/ltp $ cvs diff 
> testcases/network/lib6/Makefile
> Index: testcases/network/lib6/Makefile 
> ===================================================================
> RCS file: /cvsroot/ltp/ltp/testcases/network/lib6/Makefile,v
> retrieving revision 1.6
> diff -r1.6 Makefile
> 25a26,27
>> RANLIB                ?= ranlib
The patch you checked in does not correct the build failure.

Thanks,
Paul Larson

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now.  http://p.sf.net/sfu/bobj-july _______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

[-- Attachment #2: lib6-Makefile.patch --]
[-- Type: application/octet-stream, Size: 282 bytes --]

--- ltp-full-20090531/testcases/network/lib6/Makefile	2009-01-22 01:26:10.000000000 -0800
+++ ltp/testcases/network/lib6/Makefile	2009-07-20 12:49:00.403512700 -0700
@@ -38,4 +38,4 @@
 	rm -f $(TARGETS) *.a *.o
 
 runcc.a: runcc.o
-	ar cr runcc.a runcc.o
+	$(AR) cr runcc.a runcc.o

[-- Attachment #3: Type: text/plain, Size: 355 bytes --]

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july

[-- Attachment #4: Type: text/plain, Size: 155 bytes --]

_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] fix lib6 test building

[Garrett Cooper]

On Aug 12, 2009, at 8:13 AM, Paul Larson wrote:

> Garrett Cooper wrote:
>> On Thu, Aug 6, 2009 at 7:52 AM, Paul  
>> Larson<paul.larson@canonical.com> wrote:
>>> The lib6 tests seem to fail the first pass through building for me
>>> because ranlib does not seem to get set correctly at any point.   
>>> This
>>> seems to fix it up for me.
>>
>> Fixed on HEAD moments ago:
>>
>> gcooper@orangebox /scratch/ltp-dev2/ltp $ cvs diff
>> testcases/network/lib6/Makefile
>> Index: testcases/network/lib6/Makefile
>> ===================================================================
>> RCS file: /cvsroot/ltp/ltp/testcases/network/lib6/Makefile,v
>> retrieving revision 1.6
>> diff -r1.6 Makefile
>> 25a26,27
>>> RANLIB                ?= ranlib
> The patch you checked in does not correct the build failure.

     Error message please?
Thanks,
-Garrett

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] fix lib6 test building

[Paul Larson]

Garrett Cooper wrote:
> On Aug 12, 2009, at 8:13 AM, Paul Larson wrote:
> 
>> Garrett Cooper wrote:
>>> On Thu, Aug 6, 2009 at 7:52 AM, Paul
>>> Larson<paul.larson@canonical.com> wrote:
>>>> The lib6 tests seem to fail the first pass through building for me
>>>> because ranlib does not seem to get set correctly at any point.  This
>>>> seems to fix it up for me.
>>>
>>> Fixed on HEAD moments ago:
>>>
>>> gcooper@orangebox /scratch/ltp-dev2/ltp $ cvs diff
>>> testcases/network/lib6/Makefile
>>> Index: testcases/network/lib6/Makefile
>>> ===================================================================
>>> RCS file: /cvsroot/ltp/ltp/testcases/network/lib6/Makefile,v
>>> retrieving revision 1.6
>>> diff -r1.6 Makefile
>>> 25a26,27
>>>> RANLIB                ?= ranlib
>> The patch you checked in does not correct the build failure.
> 
>     Error message please?
Same error as before:
make[3]: Entering directory `/home/plars/ltp2/testcases/network/lib6'
cc -Wall  -I../../../include -g -D_GNU_SOURCE   -c -o runcc.o runcc.c
ar cr runcc.a runcc.o
runcc.a
make[3]: runcc.a: Command not found
make[3]: *** [runcc.a] Error 127
make[3]: Leaving directory `/home/plars/ltp2/testcases/network/lib6'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/home/plars/ltp2/testcases/network'
make[1]: *** [all] Error 2
make[1]: Leaving directory `/home/plars/ltp2/testcases'
make: *** [all] Error 2

My make-foo is a bit rusty, but I'm guessing that the higher level
export is exporting an empty RANLIB, so the RANLIB ?= ranlib is not
doing anything since RANLIB already exists.

Thanks,
Paul Larson

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] fix lib6 test building

[Garrett Cooper]

On Wed, Aug 12, 2009 at 2:09 PM, Paul Larson<paul.larson@canonical.com> wrote:
> Garrett Cooper wrote:
>> On Aug 12, 2009, at 8:13 AM, Paul Larson wrote:
>>
>>> Garrett Cooper wrote:
>>>> On Thu, Aug 6, 2009 at 7:52 AM, Paul
>>>> Larson<paul.larson@canonical.com> wrote:
>>>>> The lib6 tests seem to fail the first pass through building for me
>>>>> because ranlib does not seem to get set correctly at any point.  This
>>>>> seems to fix it up for me.
>>>>
>>>> Fixed on HEAD moments ago:
>>>>
>>>> gcooper@orangebox /scratch/ltp-dev2/ltp $ cvs diff
>>>> testcases/network/lib6/Makefile
>>>> Index: testcases/network/lib6/Makefile
>>>> ===================================================================
>>>> RCS file: /cvsroot/ltp/ltp/testcases/network/lib6/Makefile,v
>>>> retrieving revision 1.6
>>>> diff -r1.6 Makefile
>>>> 25a26,27
>>>>> RANLIB                ?= ranlib
>>> The patch you checked in does not correct the build failure.
>>
>>     Error message please?
> Same error as before:
> make[3]: Entering directory `/home/plars/ltp2/testcases/network/lib6'
> cc -Wall  -I../../../include -g -D_GNU_SOURCE   -c -o runcc.o runcc.c
> ar cr runcc.a runcc.o
> runcc.a
> make[3]: runcc.a: Command not found
> make[3]: *** [runcc.a] Error 127
> make[3]: Leaving directory `/home/plars/ltp2/testcases/network/lib6'
> make[2]: *** [all] Error 2
> make[2]: Leaving directory `/home/plars/ltp2/testcases/network'
> make[1]: *** [all] Error 2
> make[1]: Leaving directory `/home/plars/ltp2/testcases'
> make: *** [all] Error 2
>
> My make-foo is a bit rusty, but I'm guessing that the higher level
> export is exporting an empty RANLIB, so the RANLIB ?= ranlib is not
> doing anything since RANLIB already exists.

You're right. I've just fixed it on HEAD with the following comment:

Signed-off-by: Garrett Cooper <yanegomi@gmail.com>

Index: Makefile
===================================================================
RCS file: /cvsroot/ltp/ltp/Makefile,v
retrieving revision 1.40
diff -u -r1.40 Makefile
--- Makefile	10 Jul 2009 23:01:27 -0000	1.40
+++ Makefile	12 Aug 2009 21:27:24 -0000
@@ -7,6 +7,11 @@
 CC=$(CROSS_COMPILER)gcc
 AR=$(CROSS_COMPILER)ar
 RANLIB=$(CROSS_COMPILER)ranlib
+else
+# RANLIB isn't a standard Make variable, which means that when it's exported
+# to leaf callers as an empty value, builds fail.
+# testcases/network/lib6/Makefile is a prime example.
+RANLIB?=ranlib
 endif

 HAS_NUMA=$(shell sh tools/scripts/numa_test.sh)

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list