#11771: sage crashes on some degenerate flint xgcd's
--------------------------------+-------------------------------------------
Reporter: lftabera | Owner: AlexGhitza
Type: defect | Status: new
Priority: critical | Milestone: sage-4.7.2
Component: basic arithmetic | Keywords: flint, crash, xgcd fmpq_poly
Work_issues: | Upstream: N/A
Reviewer: | Author:
Merged: | Dependencies:
--------------------------------+-------------------------------------------
Comment(by leif):
Apparently a proper fix ''to the insufficient size of `temp`'' is:
{{{
#!diff
diff --git a/sage/libs/flint/fmpq_poly.c b/sage/libs/flint/fmpq_poly.c
--- a/sage/libs/flint/fmpq_poly.c
+++ b/sage/libs/flint/fmpq_poly.c
@@ -2360,15 +2360,14 @@
/* rop->den rop->num ==
*/
/* (s->num a->den / s->den) a + (t->num b->den / t->den)
b. */
- limbs = FLINT_MAX(s->num->limbs, t->num->limbs);
- limbs = FLINT_MAX(limbs, fmpz_size(s->den));
- limbs = FLINT_MAX(limbs, fmpz_size(t->den) + fmpz_size(rop->den) +
fmpz_size(lead));
- temp = fmpz_init(limbs);
-
s->den = fmpz_realloc(s->den, fmpz_size(s->den) + fmpz_size(rop->den)
+ fmpz_size(lead));
if (!fmpz_is_one(a->den))
fmpz_poly_scalar_mul_fmpz(s->num, s->num, a->den);
+
+ limbs = fmpz_size(s->den) + fmpz_size(rop->den);
+ temp = fmpz_init(limbs);
+
fmpz_mul(temp, s->den, rop->den);
fmpz_mul(s->den, temp, lead);
@@ -2376,6 +2375,10 @@
+ fmpz_size(lead));
if (!fmpz_is_one(b->den))
fmpz_poly_scalar_mul_fmpz(t->num, t->num, b->den);
+
+ limbs = fmpz_size(t->den) + fmpz_size(rop->den);
+ temp = fmpz_realloc(temp, limbs);
+
fmpz_mul(temp, t->den, rop->den);
fmpz_mul(t->den, temp, lead);
}}}
But someone more knowledgeable than me should IMHO (re-)review ''the whole
function'' w.r.t. the sizes used in memory (re)allocations.
Reluctantly I dispense with cc-ing all the reviewers of #4000... ;-)
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/11771#comment:18>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
