Re: [EXTERNAL] RE: [edk2-devel] [RedfishPkg PATCH v5 1/4] RedfishPkg: Initial commit of RedfishPkg.

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

On 10/06/20 18:29, Leif Lindholm wrote:
> Hi Abner,
> 
> On Tue, Oct 6, 2020 at 5:20 PM Chang, Abner (HPS SW/FW Technologist) <
> abner.chang@hpe.com> wrote:
>>>> diff --git a/RedfishPkg/RedfishPkg.dsc b/RedfishPkg/RedfishPkg.dsc new
>>>> file mode 100644 index 0000000000..d5c65f68c3
>>>> --- /dev/null
>>>> +++ b/RedfishPkg/RedfishPkg.dsc
>>>> @@ -0,0 +1,40 @@
>>>> +## @file
>>>> +# Redfish Package
>>>> +#
>>>> +# Copyright (c) 2019, Intel Corporation. All rights reserved.<BR> #
>>>> +(C) Copyright 2020 Hewlett-Packard Enterprise Development LP.
>>>> +#
>>>> +#    SPDX-License-Identifier: BSD-2-Clause-Patent
>>>> +#
>>>> +##
>>>> +
>>>> +[Defines]
>>>> +  PLATFORM_NAME                  = RedfishPkg
>>>> +  PLATFORM_GUID                  =
> c4352870-5232-11e7-9522-005056c00008
>>>> +  PLATFORM_VERSION               = 1.0
>>>> +  DSC_SPECIFICATION              = 0x0001001c
>>>> +  OUTPUT_DIRECTORY               = Build/RedfishPkg
>>>> +  SUPPORTED_ARCHITECTURES        = IA32|X64|ARM|AARCH64|RISCV64
>>>> +  BUILD_TARGETS                  = DEBUG|RELEASE
>>>
>>> v3 included NOOPT here. That was the version Laszlo gave his R-b for.
>>>
>>> v4 cover letter does mention NOOPT being removed, but I can see no
>>> conversation in which this is discussed. What have I missed?
>>
>> No Leif, you did miss nothing. There was some Q/A conversations between
> Bret and me about the build target of CI.
> 
> This should really be referenced in the cover letter.
> 
>> We would like to keep NO-TARGET for the non-build CI test but remove
> NOOPT for now because we do not have
>> the host-based unit test at the moment. We would like to add NOOPT back
> once we have the corresponding unit test on RedfishPkg.
> 
> Right, but dropping it from the .dsc BUILD_TARGETS is not required for that.
> And it's not clear to me that Laszlo noticed this change.

I missed it.

The v4 blurb said, "Note for v4: NOOPT is removed from RedfishPkg.dsc
[...]", and then I wrote in my v4 patch#1 response, "According to the v4
blurb, this patch has not undergone any changes since v3".

So I clearly missed the actual meaning of the blurb. Sorry about that. I
guess I'm more used to patch-granularity changelogs.

Also I believe I didn't expect this series to reach v5, so I didn't do
my usual thing where I apply every version of a patch set separately on
a local topic branch, and then do incremental reviews with git-range-diff.

So it's my fault. A better (patch-granularity) changelog might have
mitigated my mistake, perhaps.

(BTW I've not been CC'd on any version of the blurb.)

Thanks
Laszlo