Hi,
On 26/07/2019 13:47, Volodymyr Babchuk wrote:
Julien Grall writes:
The definition of PRIregister varies between Arm32 and Arm64 (32-bit vs
64-bit). However, some of the users uses the wrong padding.
type: "users use"
For more consistency, the padding is now moved into the PRIregister and
varies depending on the architecture.
I'm not sure this is the right thing to do. There are lots of code
(especially in vgic) that does not use padding at all. Now it will print
padding, even if original author does not wanted to. And, honestly it is
hard to parse 15-16 zeroes in a row.
Well, I am usually starting to read from the right to left. So, for me, 15-16
zeroes are easy to ignore ;).
I am suggesting to add another macro like PRIregister_pad or something
like that.
No, we should print register the same way everywhere. I am clearly against
providing two different formats here for the same type. Otherwise this will lead
to endless debate on which one you will chose in the code.
Looking at the vGIC, they are mostly print for debug. If you reach them then
there are probably already something wrong happening. So I think we can leave
the extra zeroes there.
Cheers,
--
Julien Grall
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
