On Mon, 21 Mar 2016 15:36:54 -0400 Greg KH <gregkh@xxxxxxxxxxxxxxxxxxx> wrote: > On Mon, Mar 21, 2016 at 03:08:41PM +0530, PrasannaKumar Muralidharan wrote: > > > Are you sure you can just remove these markings? Does the code work the > > > same? What is properly locking these values? Why were they marked this > > > way in the first place? > > > > I could not test the change due to lack of hardware. > > > > Given that the code works without lock and as volatile does not > > guarantee any sort of locking I understand that lock is not necessary. > > Please correct me if I am wrong. > > > > Could not find the purpose of 'volatile' in the code, > > 'request_counter' and 'service_counter' are only accessed by the CPU. > > So I think it can be removed. > > I will need the maintainers of the code to ack this before I can accept > it... > Since it is _not_ assured that the code will run under all circumstances, I don't want this patch to be applied at this point. To be able to decide on the removal of the volatile keyword, we need to run some tests on real hardware. This ain't something we can tell by just looking at the code. We'll add this to our todo list. Regards, Chris _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
