Hello Peter, Any suggestions regarding this topic ?
On Sun, Sep 10, 2017 at 8:33 PM, Ramy Sameh <ramysame...@gmail.com> wrote: > Thanks Peter. > > I have made a program to read a string from the UART, then write it again > to the UART. > I made a function to manipulate values in the pl011 registers (bit > flipping the flags inside the registers). > The target is to simulate hardware fault injection. > > For each run of the program, I made a bit-flip in *only one flag* in *one > register*, and I observed the output of the program. > > *My question is*: where to invoke the call of this fault injection > function to cause *the maximum effect* on the program's output? > > *p.s.* I invoked it once inside pl011_read function, and the program gave > a certain output, and I invoked it again in pl011_update, and it gave > another output. > When I invoked the function in pl011_write, the faults injected had no > effect on the output of the program! > > Do you have an explanation for this behaviour? > > Thanks in advance. > > > On Wed, Sep 6, 2017 at 2:18 PM, Peter Maydell <peter.mayd...@linaro.org> > wrote: > >> On 6 September 2017 at 13:12, Ramy Sameh <ramysame...@gmail.com> wrote: >> > Do the emulated baud rate registers have any effect? (I think they would >> > have no effect, because there is no real clock that can be used to >> produce >> > the baud rate). >> >> No, they don't have any effect. (In this UART model we don't implement >> the feature that you can connect a guest serial port to a host serial >> port and have the guest baud rate etc settings propagate through. >> We do that in other UART models and maybe one day we'll add it here, >> but for now, baud rate settings are ignored.) >> >> thanks >> -- PMM >> > > > > -- > Best Regards, > Ramy Sameh > Embedded Software Engineer > +2-010-172-777-14 > -- Best Regards, Ramy Sameh Embedded Software Engineer +2-010-172-777-14