Hi,
Le 07/03/2015 22:48, Bill Allombert a écrit :
On Sat, Mar 07, 2015 at 12:59:36PM +0100, Julien Puydt wrote:
How does cb_pari_err_recover() work ?
Their cb_pari_err_handle detects if it's a short stack problem
(errnum == e_STACK), in which case it doubles the stack and doesn't
throw an exception. If it's another problem, it blocks exceptions,
throws one and unblock.
Then cb_pari_err_recover gets called: if no exception has been
thrown, it decides the stack was doubled and retries. If an
exception has been thrown, it throws it further.
Why cannot this be done directly by cb_pari_err_handle ?
Rethrow the error seems needlessly complex.
Yes, I'll try to simplify that. I have now found that in each and every
segmentation fault case, I only get an 'ok3' displayed before the error,
with the following code :
print('ok3')
s = closure_func_err()
if s is not NULL:
print('ok4')
pari_error_string = s.decode('ascii') + ": " +
pari_error_string
print('ok5')
raise PariError(errnum, pari_error_string,
pari_instance.new_gen_noclear(E))
So my conclusion is that the segfault is triggered by calling
closure_func_err! It looks like this function is supposed to make the
error messages more precise : how come it sometimes just crashes?
We're slowly getting there...
Snark on #debian-science
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
