Re: [edk2-devel] [PATCH v3 0/2] CryptoPkg/OpensslLib: Remove "no-autoalginit" flag from OpenSSL build

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

Jian,

On 02/14/20 01:40, Zurcher, Christopher J wrote:
> In order to implement the EVP interface, the EVP_get_digestbyname function
> requires the desired digest to be already initialized. Removing the
> "no-autoalginit" build option will allow algorithms to be retrieved by name.
> I plan to submit a future patch with an EVP implementation; the OpenSSL
> documentation has stated that the EVP interface "should almost always be
> used" over direct digest interfaces at least as far back as version 0.9.6c.
> 
> This patch set also adds the "sort" keyword to the header file parsing loop
> so the .inf files aren't randomized after every run of process_files.pl.
> 
> Cc: Jian J Wang <jian.j.wang@intel.com>
> Cc: Xiaoyu Lu <xiaoyux.lu@intel.com>
> 
> Christopher J Zurcher (2):
>   CryptoPkg/OpensslLib: Add "sort" keyword to header file parsing loop
>   CryptoPkg/OpensslLib: Remove "no-autoalginit" flag from OpenSSL build
> 
>  CryptoPkg/Library/OpensslLib/OpensslLib.inf       | 66 ++++++++++----------
>  CryptoPkg/Library/OpensslLib/OpensslLibCrypto.inf | 62 +++++++++---------
>  CryptoPkg/Library/Include/openssl/opensslconf.h   |  3 -
>  CryptoPkg/Library/OpensslLib/process_files.pl     |  5 +-
>  4 files changed, 66 insertions(+), 70 deletions(-)
> 

please revert commit 6b7855209ad8e3e077a24faa40f69acdc231ff4f (merged
via your PR <https://github.com/tianocore/edk2/pull/433/>).

I have never given my R-b for the second patch in this series, only for
the first one.

Feel free to reapply the 2nd patch after the revert, but the re-applied
patch should *not* carry my R-b.

It's not that I "disagree" with patch#2 (if that had been the case, I
would have spoken up). I have no input on patch#2, instead.

Please pay more attention to review feedback.

Thanks
Laszlo