On Wed, Nov 13, 2013 at 5:05 PM, Paul Fertser <fercer...@gmail.com> wrote:
> Hi,
>
> On Mon, Nov 11, 2013 at 02:35:03PM +0100, Andreas Fritiofson wrote:
> > On Mon, Nov 11, 2013 at 2:22 PM, Paul Fertser <[1]fercer...@gmail.com>
> wrote:
> >
> > It has an SDR 33065440 bits long (4133180 bytes) and so the "buffer is
> > not enough, report to author" message is output.
> >
> > What was the reason you commented out realloc() code initially, are
> > there any gotchas that should be taken into account?
> >
> > I haven't looked at the code, but why not split the scan into multiple
> smaller
> > ones as an alternative to allocating larger and larger buffers?
>
> A typical SDR command looks like this:
>
> SDR 32 TDI (00000000) SMASK (ffffffff) TDO (f1448093) MASK (0fffffff) ;
>
> So to split the scan one would need some trickery with at least 3 file
> pointers.
>
Yes you're right, that would be awkward.
>
> Jiri has also provided me with an SVF file that has 104465920 (~13M)
> SDR command (targetting XC3S4000).
>
Even that amount of data shouldn't be a match for even the most resource
limited systems reasonably capable of running OpenOCD. I think just
reallocing to fit the data would be best and if the code is there but
disabled it shouldn't be too hard to see if it works. Push a fix and we can
review it.
/Andreas
On Wed, Nov 13, 2013 at 5:05 PM, Paul Fertser <fercer...@gmail.com> wrote:
> Hi,
>
> On Mon, Nov 11, 2013 at 02:35:03PM +0100, Andreas Fritiofson wrote:
> > On Mon, Nov 11, 2013 at 2:22 PM, Paul Fertser <[1]fercer...@gmail.com>
> wrote:
> >
> > It has an SDR 33065440 bits long (4133180 bytes) and so the "buffer is
> > not enough, report to author" message is output.
> >
> > What was the reason you commented out realloc() code initially, are
> > there any gotchas that should be taken into account?
> >
> > I haven't looked at the code, but why not split the scan into multiple
> smaller
> > ones as an alternative to allocating larger and larger buffers?
>
> A typical SDR command looks like this:
>
> SDR 32 TDI (00000000) SMASK (ffffffff) TDO (f1448093) MASK (0fffffff) ;
>
> It has bit count, then TDI data, then SMASK (source mask, no idea what
> and where it can have any meaning did the SVF standard creators really
> intended for TDI to go Hi-Z here? Anyway, OpenOcd just ignores it),
> the output to be observed on TDO and the mask that should be applied
> both to the expected and received TDO data before comparison.
>
> So to split the scan one would need some trickery with at least 3 file
> pointers. Sounds doable but I'm not feeling like trying to do that
> myself as it's non-trivial, error-prone and also I have no hardware to
> test.
>
> Jiri has also provided me with an SVF file that has 104465920 (~13M)
> SDR command (targetting XC3S4000).
>
> I think this SVF player limitation needs to be fixed, the sooner the
> better. Folks seem to be really interested in flashing their FPGAs
> with cheap and common adapters.
>
> --
> Be free, use free (http://www.gnu.org/philosophy/free-sw.html) software!
> mailto:fercer...@gmail.com
>
------------------------------------------------------------------------------
DreamFactory - Open Source REST & JSON Services for HTML5 & Native Apps
OAuth, Users, Roles, SQL, NoSQL, BLOB Storage and External API Access
Free app hosting. Or install the open source package on any LAMP server.
Sign up and see examples for AngularJS, jQuery, Sencha Touch and Native!
http://pubads.g.doubleclick.net/gampad/clk?id=63469471&iu=/4140/ostg.clktrk
_______________________________________________
OpenOCD-devel mailing list
OpenOCD-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openocd-devel
