This is the mail archive of the crossgcc@sourceware.org mailing list for the crossgcc project.

See the CrossGCC FAQ for lots more information.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Bug: compile i686-nptl-linux-gnu with crosstool-ng-1.5.2


Hello He!
All,

On Friday 18 December 2009 04:30:16 He Gang wrote:
> I am very glad to see you make crosstool continued, I am encountering
> a bug with the lastest build (crosstool-ng-1.5.2) to compile
> i686-nptl-linux-gnu on Ubuntu 9.10 or Fedora 12,
> The attached is the configuration file .config and buildlog, in fact,
> I just copied the default configuration file from samples.

The process to use an existing configuration is not to copy files, but to
tell crosstool-NG to use an existing sample:

  ct-ng i686-nptl-linux-gnu

> Could you take a look, see if there is a fix?

(Excerpt from your build.log:)
[ERROR]    i686-nptl-linux-gnu-gcc: Internal error: Segmentation fault
(program as)
[ALL  ]    Please submit a full bug report.
[ALL  ]    See <http://gcc.gnu.org/bugs.html> for instructions.


As you can see by yourself, it's not crosstool-NG that fails, but 'as' that
is segfaulting.

It is difficult to remotely solve a segfault. There can be various causes
for a segfault. To try to debug this, ask yourself the following questions:

1) Does it happen again if you re-run the build?
   - no: see (A), below
   - yes: go to question 2

2) Does it happen systematically at the same point?
  - no: see (A) below
  - yes: go to question 3

3) Does it happen if you select other components for the toolchain?
   - no, see (B), below
   - yes, go to question 4

4) Does running other CPU/memory intensive applications (eg. video encoding
   or other big compilation...) also segfault?
   - no: see (A), below
   - yes: see (B), below


(A) There is a non-reproducible glitch somewhere, most probably due to
    any or many of:
    - CPU/memory over-heating
    - CPU/memory over-clocking
    - one or more bad memory module(s)
    This is known as the infamous "SIG11 problem", with explanations and
    solutions described there: http://www.bitwizard.nl/sig11/

(B) Indeed, there is a problem when building this very source code.
    In this case, try re-running the same compilation command manually
    with debug turned on ( eg.: -v --verbose -Wa,-a=as-debug.log ) and
    see what it says.

Sorry I can't help you more. Next time, please post to the crossgcc mailing
list, there are many nice and knowledgeable people hanging on there.

Regards,
Yann E. MORIN.

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
`------------------------------^-------^------------------^--------------------'



--
For unsubscribe information see http://sourceware.org/lists.html#faq


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]