On Mon, 2017-08-14 at 10:38 -0500, Steve Magnani wrote: > I've been porting patches from mainline into a copy of QLogic's target driver tree > and ran across an anomaly in this changeset of yours from 2013 > (8c0eb596baa51f2b43949c698c644727ef17805c). > > > The commit log says: > > ... Make it easy for Coverity (and for humans) to recognize that there is no > fcport leak in the error path by changing the > > bsg_job->request->msgcode == FC_BSG_HST_ELS_NOLOGIN > > test into > > bsg_job->request->msgcode != FC_BSG_RPT_ELS. > > > But the change actually made was this: > > @@ -399,7 +399,7 @@ done_unmap_sg: > goto done_free_fcport; > > done_free_fcport: > - if (bsg_job->request->msgcode == FC_BSG_HST_ELS_NOLOGIN) > + if (bsg_job->request->msgcode == FC_BSG_RPT_ELS) > kfree(fcport); > done: > return rval; > > > Shouldn't the "== FC_BSG_RPT_ELS" be "!= FC_BSG_RPT_ELS"? Hello Steve, Without having had a close look at the patch, I think you are right. On http://www.spinics.net/lists/linux-scsi/msg66513.html one can see that the change I had posted originally is as follows: done_free_fcport: - if (bsg_job->request->msgcode == FC_BSG_HST_ELS_NOLOGIN) + if (bsg_job->request->msgcode != FC_BSG_RPT_ELS) kfree(fcport); done: I have not been able to find the reposted version of this patch so I can't figure out anymore what went wrong ... Bart.
