Re: [edk2-devel] [PATCH v8 36/46] OvmfPkg/ResetVector: Add support for a 32-bit SEV check

["Laszlo Ersek" <lersek@redhat.com>]

On 05/26/20 18:31, Tom Lendacky wrote:
> On 5/25/20 11:50 AM, Laszlo Ersek wrote:

>> Now I'm reviewing the differences (incrementally from v6 to v8), and I'm
>> shocked how many changes you incorporated into preexistent patches,
>> while keeping my R-b's.
> 
> My apologies for this. I was experimenting with cleaning things up and
> making the code more readable and I guess I forgot to either remove it
> or note it as a change to be re-reviewed - thinking I had done one or
> the other.

I agree that it's not easy to track previously given R-b's across
rebases (reordering patches, moving hunks between them, etc), especially
if there's no huge need to touch up the commit messages themselves.

What helps IME is to review the full set (just about to be posted)
before sending, with "git-range-diff", against the last posted version.
If code changes are shown for any patch (not just context differences),
take notes about those patches, then perform a final rebase, dropping
all the previously given tags from the affected patches.

(Clearly such a "git-range-diff" is incomparably faster on the author's
end than on the reviewer's end, as the author has it all in their
"working set".)

>> (1) Unless there was an actual bug in the v6 version of this patch,
>> please let's go back to that. IOW, if the v6->v8 changes are only
>> cleanups or optimizations, let's please postpone them.
> 
> Yes, I'll revert it back to the original version. Again, sorry for all
> of this churn.

Thank you!
Laszlo