Re: [lfs-dev] automake -1.15 test failures fixes

[Pierre Labastie]

On 08/05/2017 07:46, Paul Rogers wrote:
>>> Personally, I don't find adding that handful of lines to my build
>>> scripts at all cumbersome.  But, yes, if one is typing it all on the
>>> console line...  Well, that's why I haven't done that since LFS-4.1!
>>
>> I take it you don't use jhalfs. How do you build your scripts?
>>
>> Alan
> 
> I don't.  I provided a sample template along with the hint I contributed
> for pio, to show how I use my package manager.  You can find it there.
> 
The links in the hint are broken. I've found the files at:
http://www.linuxfromscratch.org/hints/downloads/files/pio-files/

While the links in the text are:
http://www.linuxfromscratch.org/hints/pio-files/

Actually, it seems the right place for attachments is:
http://www.linuxfromscratch.org/hints/downloads/files/ATTACHMENTS
(that is, for this case:
http://www.linuxfromscratch.org/hints/downloads/files/ATTACHMENTS/pio-files)

Pierre

-- 
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

Re: [lfs-dev] automake -1.15 test failures fixes

[Paul Rogers]

>> Personally, I don't find adding that handful of lines to my build
>> scripts at all cumbersome.  But, yes, if one is typing it all on the
>> console line...  Well, that's why I haven't done that since LFS-4.1!
>
>I take it you don't use jhalfs. How do you build your scripts?
>
>Alan

I don't.  I provided a sample template along with the hint I contributed
for pio, to show how I use my package manager.  You can find it there.

-- 
Paul Rogers
paulgrog...@fastmail.fm
Rogers' Second Law: "Everything you do communicates."
(I do not personally endorse any additions after this line. TANSTAAFL
:-)

-- 
http://www.fastmail.com - Does exactly what it says on the tin

-- 
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

Re: [lfs-dev] automake -1.15 test failures fixes

[Alan Feuerbacher]

On 5/7/2017 1:28 AM, Paul Rogers wrote:


For automake, the process of removing the bad tests is cumbersome, but
we do say:  Four tests are known to fail.


Personally, I don't find adding that handful of lines to my build
scripts at all cumbersome.  But, yes, if one is typing it all on the
console line...  Well, that's why I haven't done that since LFS-4.1!


I take it you don't use jhalfs. How do you build your scripts?

Alan
--
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

Re: [lfs-dev] automake -1.15 test failures fixes

[Paul Rogers]

> It would be useful to know which pages you consider a problem.  I've
> recently fixed glibc, vim, and am in the process of fixing the one in
> flex via an updated patch.

I also accepted a failure in my tar-1.28 fallback, sparse3 IIRC.  AIUI
it tries to allocate a huge file, but I'm building in a 15GB partition,
for the moment.  Ch6-8 are done and dusted, boots and runs.

> For automake, the process of removing the bad tests is cumbersome, but
> we do say:  Four tests are known to fail.

Personally, I don't find adding that handful of lines to my build
scripts at all cumbersome.  But, yes, if one is typing it all on the
console line...  Well, that's why I haven't done that since LFS-4.1!

> Would it be sufficient to just name the failing tests?

Indeed, that would be most helpful.  The check12 fail was what Ken told
me that gave me confidence to go on, not throwing it all away!  I can't
emphasize enough how important it is to have condifence in the
reliability of the system we're building.  Not knowing which failures
have been accepted by the book developers, leaves doubt.

I hope you do.
-- 
Paul Rogers
paulgrog...@fastmail.fm
Rogers' Second Law: "Everything you do communicates."
(I do not personally endorse any additions after this line. TANSTAAFL
:-)

-- 
http://www.fastmail.com - A no graphics, no pop-ups email service

-- 
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

Re: [lfs-dev] automake -1.15 test failures fixes

[Bruce Dubbs]

Paul Rogers wrote:

These are all good points.  There's one that hasn't been mentioned that
seems to be at the root in automake, changed behavior of a dependency in
a test, which could cause a, strictly speaking, unnecessary up/downgrade
downstream.  Another is that the (B)LFS build environment imposes its
own restrictions while a package may presume it is being built in a
full-
function environment, e.g. tar-1.29.  The issue is complex.

But the real problem remains, the LFS builder is left with a test error
which (s)he may well not be able to properly judge in severity and may
have doubts about ignoring/accepting.  This creates doubts about the
reliability of the system they are building, and whether building a
flawed system has become "a waste of time".  All things considered,
(s)he must decide when diagnosing a test problem becomes a waste of
time, especially if it is a common issue.  Google may or may not be a
help.  It's one thing to expect basic competency with the Linux/GNU
toolset operations, but we can't all be expert diagnosticians.

There are some places in the book where test errors are identified
specifically acceptable, some places where they are removed from the
test stream.  So, clearly the (B)LFS developers have met certain faults.
That's why I contacted Ken off-list, to ask if these were unique to my
build, or have been met before.  More information in the book about test
results would be very helpful.


It would be useful to know which pages you consider a problem.  I've 
recently fixed glibc, vim, and am in the process of fixing the one in flex 
via an updated patch.


For automake, the process of removing the bad tests is cumbersome, but we 
do say:  Four tests are known to fail.


Would it be sufficient to just name the failing tests?

  -- Bruce


--
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

Re: [lfs-dev] automake-1.15 test failures fixes

[Bruce Dubbs]

akhiezer wrote:

From: Paul Rogers 
Date: Thu, 04 May 2017 00:14:40 -0700

For some uses I'd like my build scripts to not fail on tests that will
be ignored anyhow.  In automake-1.15 three FAILs seem to be common and
due to flawed tests, notably check12{,-w}.  I have added the following
commands to my scripts to avert these FAILed tests:

# check12 is known to fail, debian removes them so we will too
rm -f t/check12{,-w}.sh &&
sed  -e '/t\/check12.sh/d' -e '/t\/check12-w.log/d' -e
's/t\/check12-w.sh//' \
   -i Makefile.in &&
sed '/t\/check12-w/d' -i t/testsuite-part.am &&
sed '/t\/check12/d' -i t/list-of-tests.mk &&
# remove another known failure
sed -e '/t\/distcheck-no-prefix-or-srcdir-override.sh/d' -i Makefile.in
&&




It's better to let tests run, log what happens, then analyse the results,
and do any actions as wanted/reqd: rather than suppress at source -
which can be whac-a-mole, and at times opening security risks.


When I test a new package for the book, I always remove special code like 
this to see it they tests have been removed or fixed.  I then make  a 
decision whether to just document the problem or to add code to remove the 
problem.


I jsut went though this with vim and decided that a one line fix to remove 
the test was acceptable.  For automake, IMO, it is too intrusive.


So the security issues are probably not an issue.  It is a bit painful 
when upstream doesn't care about the tests enough to fix them or remove them.


  -- Bruce

--
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

Re: [lfs-dev] automake-1.15 test failures fixes

[akhiezer]

> From: Paul Rogers 
> Date: Thu, 04 May 2017 00:14:40 -0700
>
> For some uses I'd like my build scripts to not fail on tests that will
> be ignored anyhow.  In automake-1.15 three FAILs seem to be common and
> due to flawed tests, notably check12{,-w}.  I have added the following
> commands to my scripts to avert these FAILed tests:
>
> # check12 is known to fail, debian removes them so we will too
> rm -f t/check12{,-w}.sh &&
> sed  -e '/t\/check12.sh/d' -e '/t\/check12-w.log/d' -e
> 's/t\/check12-w.sh//' \
>   -i Makefile.in &&
> sed '/t\/check12-w/d' -i t/testsuite-part.am &&
> sed '/t\/check12/d' -i t/list-of-tests.mk &&
> # remove another known failure
> sed -e '/t\/distcheck-no-prefix-or-srcdir-override.sh/d' -i Makefile.in
> &&
>


It's better to let tests run, log what happens, then analyse the results,
and do any actions as wanted/reqd: rather than suppress at source -
which can be whac-a-mole, and at times opening security risks.



akh





--
-- 
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

Re: [lfs-dev] automake-1.15 test failures fixes

[Bruce Dubbs]

Ken Moffat wrote:

On Thu, May 04, 2017 at 12:14:40AM -0700, Paul Rogers wrote:

For some uses I'd like my build scripts to not fail on tests that will
be ignored anyhow.  In automake-1.15 three FAILs seem to be common and
due to flawed tests, notably check12{,-w}.  I have added the following
commands to my scripts to avert these FAILed tests:

# check12 is known to fail, debian removes them so we will too
rm -f t/check12{,-w}.sh &&
sed  -e '/t\/check12.sh/d' -e '/t\/check12-w.log/d' -e
's/t\/check12-w.sh//' \
   -i Makefile.in &&
sed '/t\/check12-w/d' -i t/testsuite-part.am &&
sed '/t\/check12/d' -i t/list-of-tests.mk &&
# remove another known failure
sed -e '/t\/distcheck-no-prefix-or-srcdir-override.sh/d' -i Makefile.in
&&


Paul spoke to me off-list about this, so although I don't think it
is necessary to not run tests which will fail, I suggested he post
this here.

However, looking back at my logs - my first build of 8.0-rc1 was on
my i7 and there I *did* get 4 failures like the book says.  In *all*
my subsequent builds (8.0 and svn) on all my machines I only got the
three failures Paul has suppressed above.  But as I said to him -
different people get different failures.

The fourth was t/subobj.sh, a quick look suggests it usually
passes.


I think the code above is a little too much for the book.  We do document 
the failures.  If it was one or two lines, then I'd be OK with it.


Don't misunderstand.  I like the concept and will put it in my script for 
automake.  I just think it puts a little too much emphasis on the tests in 
the book.


  -- Bruce

--
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

Re: [lfs-dev] automake-1.15 test failures fixes

[Ken Moffat]

On Thu, May 04, 2017 at 12:14:40AM -0700, Paul Rogers wrote:
> For some uses I'd like my build scripts to not fail on tests that will
> be ignored anyhow.  In automake-1.15 three FAILs seem to be common and
> due to flawed tests, notably check12{,-w}.  I have added the following
> commands to my scripts to avert these FAILed tests:
> 
> # check12 is known to fail, debian removes them so we will too
> rm -f t/check12{,-w}.sh &&
> sed  -e '/t\/check12.sh/d' -e '/t\/check12-w.log/d' -e
> 's/t\/check12-w.sh//' \
>   -i Makefile.in &&
> sed '/t\/check12-w/d' -i t/testsuite-part.am &&
> sed '/t\/check12/d' -i t/list-of-tests.mk &&
> # remove another known failure
> sed -e '/t\/distcheck-no-prefix-or-srcdir-override.sh/d' -i Makefile.in
> &&
> 
Paul spoke to me off-list about this, so although I don't think it
is necessary to not run tests which will fail, I suggested he post
this here.

However, looking back at my logs - my first build of 8.0-rc1 was on
my i7 and there I *did* get 4 failures like the book says.  In *all*
my subsequent builds (8.0 and svn) on all my machines I only got the
three failures Paul has suppressed above.  But as I said to him -
different people get different failures.

The fourth was t/subobj.sh, a quick look suggests it usually
passes.

ĸen
-- 
I live in a city. I know sparrows from starlings.  After that
everything is a duck as far as I'm concerned.  -- Monstrous Regiment
-- 
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page