> Christoph Raisch writes: > > > The patch is needed. We've seen it on the real system. We did fix it on the > > real system. > > I disagree that the ioremap change is needed. > > > ...and it conforms to theory... although theory is a bit confusing here. > > > > let me try to summarize: > > ioremap checks for 64k boundary (actually page boundary) > > Actually, ioremap itself already does the calculations that your patch > adds - that is, it generates the offset within the page and the > physical address of the start of the page, does the mapping using the > latter, then adds on the offset to the virtual address of the page and > returns that.
Paul, you are right. The calculation is done in your code already. We can't reproduce the problem anymore on latest kernel. Was this calculation there in ioremap right from the start with 64k on POWER or added later on? So Roland, feel free to ignore that line where we do the calculation. > > Paul. _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
