On Tue, Apr 22, 2014 at 06:25:13PM +0300, Dan Carpenter wrote:
> On Tue, Apr 22, 2014 at 04:28:35PM +0300, Dan Carpenter wrote:
> > Thanks for the report. This was introduced in commit a5d654e8e937
> > ('missing_break: use fewer states'). I didn't notice this because I
> > have some code in my testing tree which cancels out this bug. Very
> > weird. Sorry, about that.
>
> Hm... Nope. That's not it. The missing_break change can't possibly
> be it. Anyway, I will continue looking at this.
I can't tell whether this commit is the actual source of the issue but
"git bisect" pointed me to the same commit:
git bisect start
# bad: [f28e8d607f8dd8d5bc56e11706711935235b6a31] db: function with too many pointers to it cause slow down
git bisect bad f28e8d607f8dd8d5bc56e11706711935235b6a31
# good: [0a454f0b4b513666351f4e50c72e057bde0ab9ca] start_states: update get_start_states() to use stree
git bisect good 0a454f0b4b513666351f4e50c72e057bde0ab9ca
# good: [9d6962bad2b6f2a142edf03690bd31d3caf01f25] implied: smatch was saying somethings were implied when they weren't
git bisect good 9d6962bad2b6f2a142edf03690bd31d3caf01f25
# bad: [9960e84cfe20a7b252d6bb821d955cd2707bca6b] states: add another read_only check
git bisect bad 9960e84cfe20a7b252d6bb821d955cd2707bca6b
# good: [6405846fab6bbbb43355b5a95e27b71755817505] implied: add the stree_id to debug output
git bisect good 6405846fab6bbbb43355b5a95e27b71755817505
# good: [3db2b11ed17de4e7d7f2c511336984ec43727808] unconstant_macros: allow (foo & _PAGE_NX) even when _PAGE_NX is zero
git bisect good 3db2b11ed17de4e7d7f2c511336984ec43727808
# good: [a10ac1c796a4dc02ed205699bf76dae0eaccf07b] avl: add a counter for when you allocate or free a stree
git bisect good a10ac1c796a4dc02ed205699bf76dae0eaccf07b
# bad: [a5d654e8e937da4ca25e8621cc55a59187299cad] missing_break: use fewer states
git bisect bad a5d654e8e937da4ca25e8621cc55a59187299cad
# first bad commit: [a5d654e8e937da4ca25e8621cc55a59187299cad] missing_break: use fewer states
Cheers,
Silvan
--
To unsubscribe from this list: send the line "unsubscribe smatch" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
