Hi, Christopher, Bjorn, Apologies for leaving these issues in. I don’t currently have resources to easily test compiling, yet alone running on most architectures. I did not receive the buildbot warnings but have now had a look at them on the web archive since my attention was brought to them. Attempting to remove constness from the pointer to map_irq functions was tricky as I was nervous to blindly change a bunch of functions for hardware I cannot easily compile for and cannot test on as it is hard for me to be 100% sure I have not broken behaviour without trying the code even if visually the changes look fine. I did not understand all the arch specific code well enough to verify for sure what behaviour was ok to change so ended up making the minimum changes to functionality of arch code possible. I expect the ideal thing to do is to fix all of those pointers to be non const even where the buildbot did not give warnings as removing constness as far as I know should never break anything that isn't already broken but leaving it in incorrectly could cause hard to find issues. If it later becomes clear that it should be there it can always be re-added in another patch-set. However a second opinion on this would be greatly appreciated as I dont know if some architecture code has non obvious restrictions here. Also please note I will not have much time to spend developing and working on this over the next couple of weeks due to Christmas related events but will try and keep up to date on emails. But thanks everyone so far for the feedback. Many thanks, Matthew Minter -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
