On 04/07/16 14:28, Ard Biesheuvel wrote: > On 7 April 2016 at 14:26, Laszlo Ersek <[email protected]> wrote: >> On 04/07/16 14:04, Ard Biesheuvel wrote: >>> On 7 April 2016 at 14:02, Laszlo Ersek <[email protected]> wrote: >>> Does that mean you are going to ignore my v2? >> >> I certainly don't *want* to ignore it, but I think we have identified >> several aspects that may affect many of the remaining patches: >> >> (a) UINT32 vs. UINTN output parameters in the protocol interface >> >> (b) LibFdt usage / SwapBytes() >> >> (c) (important:) for the commit messages, tracking down the clients of >> the new plugin library instances that set PCDs (to avoid >> multi-setting), recursively; plus tracking down all the consumers >> of those PCDs (to ensure they will all see the updated PCD values). >> >> (d) line lengths >> >> (e) argument indentation in function calls >> >> Issues (a), (b), (d) and (e) will continue tripping me up in the review, >> and you can fix them up without me actually identifying each individual >> location. >> >> Issue (c) requires me to spend multiple tens of minutes per affected >> patch. I would prefer if you spent those minutes, wrote up your findings >> in the commit messages, so I'd only need verify your findings, not find >> them myself from scratch. >> >> In other words, the above issues should be fixed not just for the sake >> of the final code and messages that get into the repository, but also in >> order to help me review the rest of the series. >> >> Normally I do prefer to review all patches in a series, before asking >> for the next version, but I believe the above problems might affect the >> rest of the patches. My main worry is that if I keep getting tripped up >> by them, I will get de-sensitivised for version 3, and maybe miss >> something else that would be important. >> >> Consider, if I review the rest of v2, and point out twenty more >> instances of (d) and (e), then when I review v3, I will have to verify >> whether you fixed up every single one of those instances. This wastes a >> lot of gray matter. :) This is the kind of issue that you can locate >> yourself in the rest of the series (now that I've pointed out a few >> instances), so they wouldn't even exist to begin with, when I looked at >> the rest of the patches, in v3, for the first time. >> >> Of course, I could be wrong too. If you assure me that issues (a) >> through (e) will probably not impede my review in the remaining v2 >> patches :), then I'm more than happy to continue reviewing v2! >> >> To repeat: saving time for me (at the cost of your time) is just the >> small goal. The big goal is to preserve my sensitivity to your patches. >> > > Dude, I was kidding. I haven't even sent out my v2 yet, the patches > you have been looking at were v1.
Sigh. Did I mention de-sensitivization?... ;) I guess I'm lolling at myself. :) Thanks! Laszlo _______________________________________________ edk2-devel mailing list [email protected] https://lists.01.org/mailman/listinfo/edk2-devel
