On 05/26/2014 07:40 PM, arei.gong...@huawei.com wrote: > From: Gonglei <arei.gong...@huawei.com> > > Spotted by Coverity: > > (1) Event unreachable: This code cannot be reached: "abort();". > > 112 abort(); > > Signed-off-by: Gonglei <arei.gong...@huawei.com> > --- > hw/i2c/pm_smbus.c | 1 - > 1 file changed, 1 deletion(-) > > diff --git a/hw/i2c/pm_smbus.c b/hw/i2c/pm_smbus.c > index fedb5fb..da09eb0 100644 > --- a/hw/i2c/pm_smbus.c > +++ b/hw/i2c/pm_smbus.c > @@ -109,7 +109,6 @@ static void smb_transaction(PMSMBus *s) > default: > goto error; > } > - abort();
Given the questions that have been raised on this being dead code, would
it be worth just getting rid of the 'default' label, as in:
switch () {
existing normal cases
}
goto error;
so that you are trading the abort() with an unconditional goto that gets
reached via the fact that the switch no longer has a default?
--
Eric Blake eblake redhat com +1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
