Re: [ccache] ccache test failures

[Greg Forte via ccache]

 I figured out what the problem was with both of those tests - the
directory I was building in had a symlink high up in the path - a 'cd
$(readlink -f .)' before the build fixed it, and make/make test/make
install now succeeds on our SLES12 systems.

Now I have two different test cases failing only on even-older SLES11
systems:

one is in the cpp1 suite, and manifests as:

   Manifest has bad magic number: 1013343290
   Corrupt manifest file
   Error reading manifest file
   Did not find object file hash in manifest

in the ccache.log.  The "bad magic number" value is always the same.  This
only occurs when building via a rather convoluted script that we use to
crank out builds of (mostly) 3rd party packages - the failure occurs
consistently in that script, but I have so far not managed to reproduce it
outside.  So I'm pretty certain it's got something to do with the
environment rendered by the script ... no idea what, exactly, though. Any
thoughts?  The same script is driving the successful builds on SLES12, for
whatever that's worth.

The other one appears to be due to a too-old or buggy (or both) version of
objdump on the oldest platform we still support - when the test suite
(debug_prefix_map) runs 'objdump -g' on the file, objdump just prints the
"file format" line and nothing about the debug symbols .. but objdump from
a slightly less ancient os image prints the expected output.  So I think I
can safely ignore that one ... though that also means the same suite could
be falsely passing when it checks for `pwd`.  I'd submit a patch, but not
sure what the best alternative would be ... fall back to readelf?

-Greg Forte


On Wed, Mar 21, 2018 at 7:41 AM, Forte, Greg via ccache <
ccache@lists.samba.org> wrote:

> Thanks, Joel!  I haven't had much time to look at this since my previous
> post, but I did discover that, for the first failing test (CCACHE_NOHASHDIR
> in base suite), forcing CCACHE_BASEDIR to a common parent dir for just that
> test cleared the error.  Still not sure if this is actually a problem with
> the test or due to some other peculiarity in my environment.
> CCACHE_BASEDIR is already turned on for the other failing test, though; I
> will try looking at that one with CCACHE_DEBUG_HASH when I have some time.
>
> -Original Message-
> From: joel.rosd...@gmail.com [mailto:joel.rosd...@gmail.com] On Behalf Of
> Joel Rosdahl
> Sent: Monday, March 19, 2018 3:57 PM
> To: Forte, Greg <greg.fo...@msx.bala.susq.com>
> Cc: ccache@lists.samba.org
> Subject: Re: [ccache] ccache test failures
>
> On 16 March 2018 at 12:52, Forte, Greg via ccache <ccache@lists.samba.org>
> wrote:
> > Tried this with and without CCACHE_DIR set and exported - with it, I
> > see the same output you showed, including the unset, and the test
> > still fails: [...]
>
> OK.
>
> What's common between the two tests that fail for you is that you don't
> get a cache hit for a second compilation in another directory. One way to
> understand why is to log all data that is hashed by ccache in the two cases
> and see where it starts to differ. There is a crude way of doing that
> already: Build ccache with -DCCACHE_DEBUG_HASH and set CCACHE_DEBUG_HASH at
> runtime. Each ccache invocation will then create a ccache-debug-hash.bin
> that contains the hashed data, so you could compare the files in the two
> different directories. Or send them to me (private mail) and I'll have a
> look.
>
> By the way, here's a sketch of a yet unimplemented feature what would make
> this kind of debugging easier: https://github.com/ccache/ccache/issues/226
>
> -- Joel
>
> 
>
> IMPORTANT: The information contained in this email and/or its attachments
> is confidential. If you are not the intended recipient, please notify the
> sender immediately by reply and immediately delete this message and all its
> attachments. Any review, use, reproduction, disclosure or dissemination of
> this message or any attachment by an unintended recipient is strictly
> prohibited. Neither this message nor any attachment is intended as or
> should be construed as an offer, solicitation or recommendation to buy or
> sell any security or other financial instrument. Neither the sender, his or
> her employer nor any of their respective affiliates makes any warranties as
> to the completeness or accuracy of any of the information contained herein
> or that this message or any of its attachments is free of viruses.
> ___
> ccache mailing list
> ccache@lists.samba.org
> https://lists.samba.org/mailman/listinfo/ccache
>
___
ccache mailing list
ccache@lists.samba.org
https://lists.samba.org/mailman/listinfo/ccache

Re: [ccache] ccache test failures

[Forte, Greg via ccache]

Thanks, Joel!  I haven't had much time to look at this since my previous post, 
but I did discover that, for the first failing test (CCACHE_NOHASHDIR in base 
suite), forcing CCACHE_BASEDIR to a common parent dir for just that test 
cleared the error.  Still not sure if this is actually a problem with the test 
or due to some other peculiarity in my environment.  CCACHE_BASEDIR is already 
turned on for the other failing test, though; I will try looking at that one 
with CCACHE_DEBUG_HASH when I have some time.

-Original Message-
From: joel.rosd...@gmail.com [mailto:joel.rosd...@gmail.com] On Behalf Of Joel 
Rosdahl
Sent: Monday, March 19, 2018 3:57 PM
To: Forte, Greg <greg.fo...@msx.bala.susq.com>
Cc: ccache@lists.samba.org
Subject: Re: [ccache] ccache test failures

On 16 March 2018 at 12:52, Forte, Greg via ccache <ccache@lists.samba.org> 
wrote:
> Tried this with and without CCACHE_DIR set and exported - with it, I
> see the same output you showed, including the unset, and the test
> still fails: [...]

OK.

What's common between the two tests that fail for you is that you don't get a 
cache hit for a second compilation in another directory. One way to understand 
why is to log all data that is hashed by ccache in the two cases and see where 
it starts to differ. There is a crude way of doing that already: Build ccache 
with -DCCACHE_DEBUG_HASH and set CCACHE_DEBUG_HASH at runtime. Each ccache 
invocation will then create a ccache-debug-hash.bin that contains the hashed 
data, so you could compare the files in the two different directories. Or send 
them to me (private mail) and I'll have a look.

By the way, here's a sketch of a yet unimplemented feature what would make this 
kind of debugging easier: https://github.com/ccache/ccache/issues/226

-- Joel



IMPORTANT: The information contained in this email and/or its attachments is 
confidential. If you are not the intended recipient, please notify the sender 
immediately by reply and immediately delete this message and all its 
attachments. Any review, use, reproduction, disclosure or dissemination of this 
message or any attachment by an unintended recipient is strictly prohibited. 
Neither this message nor any attachment is intended as or should be construed 
as an offer, solicitation or recommendation to buy or sell any security or 
other financial instrument. Neither the sender, his or her employer nor any of 
their respective affiliates makes any warranties as to the completeness or 
accuracy of any of the information contained herein or that this message or any 
of its attachments is free of viruses.
___
ccache mailing list
ccache@lists.samba.org
https://lists.samba.org/mailman/listinfo/ccache

Re: [ccache] ccache test failures

[Joel Rosdahl via ccache]

On 16 March 2018 at 12:52, Forte, Greg via ccache
 wrote:
> Tried this with and without CCACHE_DIR set and exported - with it, I see the
> same output you showed, including the unset, and the test still fails: [...]

OK.

What's common between the two tests that fail for you is that you don't get a
cache hit for a second compilation in another directory. One way to understand
why is to log all data that is hashed by ccache in the two cases and see where
it starts to differ. There is a crude way of doing that already: Build ccache
with -DCCACHE_DEBUG_HASH and set CCACHE_DEBUG_HASH at runtime. Each ccache
invocation will then create a ccache-debug-hash.bin that contains the hashed
data, so you could compare the files in the two different directories. Or send
them to me (private mail) and I'll have a look.

By the way, here's a sketch of a yet unimplemented feature what would make this
kind of debugging easier: https://github.com/ccache/ccache/issues/226

-- Joel

___
ccache mailing list
ccache@lists.samba.org
https://lists.samba.org/mailman/listinfo/ccache

Re: [ccache] ccache test failures

[Forte, Greg via ccache]

Thanks for the response!

> On 15 March 2018 at 20:33, Forte, Greg via ccache  
> wrote:
> > I've run into a few failures with 'make test' after building ccache
> > and I'm hoping someone here has insight into what might be causing
> > them, or can provide some leads on how to investigate further. [...]
>
> [...] What does
>
> env | sed -n 's/^\(CCACHE_[A-Z0-9_]*\)=.*$/\1/p'
>
> return for you?

Nothing.

>  If it looks OK, you can try running the test suite like this:
>
> bash -x test/run debug_prefix_map
> [...]
> + unset CCACHE_DIR
>
> Note the “unset”.

Tried this with and without CCACHE_DIR set and exported - with it, I see the 
same output you showed, including the unset, and the test still fails:

Test suite: debug_prefix_map
Test case:  Mapping of debug info CWD
Failure reason: Expected "cache hit (direct)" to be 1, actual 0

Same results without CCACHE_DIR preset, except of course for no "unset 
CCACHE_DIR" line.

> > IMPORTANT: The information contained in this email and/or its
> > attachments is confidential. [...]
>
>  Are you sure that you are allowed to post to public lists from your email 
> address? :-P
>
> -- Joel

They don't leave me much choice - it was either that or scratch out that msg on 
my phone.  If the boilerplate is offensive (which I can see how it would be), 
I'll try to use gmail, but they block it here.

-greg



IMPORTANT: The information contained in this email and/or its attachments is 
confidential. If you are not the intended recipient, please notify the sender 
immediately by reply and immediately delete this message and all its 
attachments. Any review, use, reproduction, disclosure or dissemination of this 
message or any attachment by an unintended recipient is strictly prohibited. 
Neither this message nor any attachment is intended as or should be construed 
as an offer, solicitation or recommendation to buy or sell any security or 
other financial instrument. Neither the sender, his or her employer nor any of 
their respective affiliates makes any warranties as to the completeness or 
accuracy of any of the information contained herein or that this message or any 
of its attachments is free of viruses.
___
ccache mailing list
ccache@lists.samba.org
https://lists.samba.org/mailman/listinfo/ccache