Re: development process failure [was: remove TPM related ppi from Depex for Fsp wrapper PEIM driver]

Yao, Jiewen

Hi Laszlo
Thanks. I agree 1, 2, 3. I take the blame. It is my fault.

For 4, it is out of my scope. I cannot find this by my eyes. Everything works well on my side.
Can we improve patch checker to catch this in CI ?
I don’t think I can find any Unicode in code or commit message easily.
I prefer to let a tool to do that work.

Thank you
Yao Jiewen

-----Original Message-----
From: Laszlo Ersek <>
Sent: Wednesday, September 16, 2020 4:43 PM
To: Chiu, Chasel <>; Yao, Jiewen <>
Cc:; Zhang, Qi1 <>; Desimone,
Nathaniel L <>; Zeng, Star
<>; Wang, Jian J <>
Subject: development process failure [was: remove TPM related ppi from Depex
for Fsp wrapper PEIM driver]

Jiewen, Chasel,

On 09/15/20 08:21, Qi Zhang wrote:
Some open board are TPM disabled. So the boot may hang because
these PPIs can't arrive. And gEdkiiTcgPpiGuid will be notified where
it is used. So we need to remove these PPIs from Depex for Fsp wrapper
PEI and PeiTpmMeasurementLib.

Cc: Chasel Chiu <>
Cc: Nate DeSimone <>
Cc: Star Zeng <>
Cc: Jiewen Yao <>
Cc: Jian J Wang <>

Qi Zhang (2):
IntelFsp2WrapperPkg: remove gPeiTpmInitializationDonePpiGuid from
SecurityPkg/PeiTpmMeasurementLib: remove gEfiTpmDeviceSelectedGuid

IntelFsp2WrapperPkg/FspmWrapperPeim/FspmWrapperPeim.inf | 3 +--
IntelFsp2WrapperPkg/FspsWrapperPeim/FspsWrapperPeim.inf | 3 +--
.../Library/PeiTpmMeasurementLib/PeiTpmMeasurementLib.inf | 3 +--
3 files changed, 3 insertions(+), 6 deletions(-)
Please adopt a *much more* disciplined approach when merging patch series.

(1) When you merge a patch set, please report back on the list. Identify
both the pull request URL, and the commit reange.

In this case, the pull request was

and the commit range is a62fb4229d14..7bcb021a6d54.

(2) The associated Bugzilla:

has been completely neglected, by both submitter and maintainers.

- The original BZ report is *absolute trash*.

- No URL into the mailing list archive has been captured in the BZ,
about the posted series.

- The BZ status is still CONFIRMED.

- No mention of the pull request, or the resultant commit, range in the
BZ ticket.

(3) The github pull request at
<> does contain *any*
indication of the bugzilla ticket, or the cover letter on the list.

Basically we have random artifacts in three different places (Bugzilla,, mailing list), and nobody of the involved parties
(reviewers, maintainers, constributors) on this patch set have made
*any* effort to cross-reference them. We now have to hunt down
everything separately.

(4) Worst of all, the subject line of commit 414d7d11e6ea contains a
Unicode code point called FULLWIDTH COLON (U+FF1A) rather than a normal
colon (U+003A).


- bad (current): IntelFsp2WrapperPkg: remove [...]
- good (should have been): IntelFsp2WrapperPkg: remove [...]

It makes absolutely no sense to use non-ASCII code points in subject
lines, for something as trivial as a colon.

I've been here for 8-9 years now and it's incredibly frustrating that I
*still* have to whine about basic stuff like this on a regular basis.

I don't even know whom I should CC at Intel (management or otherwise) to
see an improvement in attitude here.

I guess this community cannot be saved.


Join to automatically receive all group messages.