On Wed, Nov 27, 2024 at 05:17:15PM -0800, Joe Damato wrote:
>
> Guenter: would you mind sending me your cocci script? Mostly for
> selfish reasons; I'd like to see how you did it so I can learn more
> :) Feel free to do so off list if you prefer.
>
You mean because it is so clumsy ? :-). No worries, I attached
it below. It is way too complicated, but it was good enough
for a quick hack.
> I tried to write my first coccinelle script (which you can find
> below) that is probably wrong, but it attempts to detect:
> - interrupt routines that hold locks
> - in drivers that call napi_enable between a lock/unlock
>
> I couldn't figure out how to get regexps to work in my script, so I
> made a couple variants of the script for each of the spin_lock_*
> variants and ran them all.
I pretty much did the same, only in one script.
>
> Only one offender was detected: pcnet32.
>
Your script doesn't take into account that the spinlock may have been
released before the call to napi_enable(). Other than that it should
work just fine. I didn't try to track the interrupt handler because
tracking that through multiple functions can be difficult.
Thanks,
Guenter
---
virtual report
@f1@
identifier flags;
position p;
expression e;
@@
<...
spin_lock_irqsave@p(e, flags);
... when != spin_unlock_irqrestore(e, flags);
napi_enable(...);
... when any
spin_unlock_irqrestore(e, flags);
...>
@f2@
position p;
expression e;
@@
<...
spin_lock_irq@p(e);
... when != spin_unlock_irq(e);
napi_enable(...);
... when any
spin_unlock_irq(e);
...>
@f3@
position p;
expression e;
@@
<...
spin_lock@p(e);
... when != spin_unlock(e);
napi_enable(...);
... when any
spin_unlock(e);
...>
@script:python@
p << f1.p;
@@
print("napi_enable called under spin_lock_irqsave from %s:%s" % (p[0].file, p[0].line))
@script:python@
p << f2.p;
@@
print("napi_enable called under spin_lock_irq from %s:%s" % (p[0].file, p[0].line))
@script:python@
p << f3.p;
@@
print("napi_enable called under spin_lock from %s:%s" % (p[0].file, p[0].line))
