On Wed, Sep 25, 2019 at 06:33:35PM -0700, Bart Van Assche wrote: > On 2019-09-25 05:39, Daniel Wagner wrote: > > So I after starring on the code I am not sure if the WARN_ON_ONCE is > > correct. It assumes that after processing one status continuation, > > there is no more work. Though it looks like there is another element > > to process. Is it possible that two sense continuation are possible? > > According to the firmware documentation it is possible that multiple > status continuations are emitted by the firmware. Do you want to submit > a patch or do you prefer that I do this myself? I just send out a patch, which removes the WARN_ON_ONCE. This solves the obvious warning but drops the consistency check. Maybe qla2x00_status_cont_entry() could return a status and the outer loop could check at the end if sp->done() has been called. But I suppose this should be done for all parsing functions not just for one.
