On 11/06/18 01:59, Fu, Siyuan wrote:
> Hi, Leif and Laszlo
>
>
>> -----Original Message-----
>> From: Laszlo Ersek [mailto:[email protected]]
>> Sent: Tuesday, November 6, 2018 6:12 AM
>> To: Leif Lindholm <[email protected]>; Fu, Siyuan
>> <[email protected]>
>> Cc: [email protected]
>> Subject: Re: [edk2] [PATCH v1 0/7] Delete TCP, PXE, iSCSI driver in
>> MdeModulePkg.
>>
>> Hi Leif,
>>
>> On 10/31/18 11:59, Leif Lindholm wrote:
>>> Hi Fu Siyuan,
>>>
>>> Hmm, strange.
>>>
>>> Ming Huang (cc) has also had issues with this in his latest set (but
>>> not before). Can you guys work together to try to figure out what is
>>> going wrong?
>>>
>>> Can you start by testing with an older version of git?
>>
>> this is indeed a git regression. I ran into it myself recently, when I
>> upgraded git from v2.14.1 to v2.19.1. I dug a little into the git source
>> code, but couldn't spend much time on it. Ultimately I rolled back the
>> upgrade.
>>
>> The diffstats are still well-formed (not truncated) in the individual
>> patch emails, under v2.19.1, as long as the "recommended" options are
>> used on the cmdline. However, the cumulative diffstat is truncated
>> regardless. I couldn't convince git v2.19.1 to keep the previous
>> (desired) behavior.
>
> So should I rollback my Git to v2.14.1, or the new truncated cumulative
> diffstat is also acceptable?
Leif identified the culprit git patch as commit 43662b23abbd
("format-patch: keep cover-letter diffstat wrapped in 72 columns",
2018-01-25), and that commit was part of release v2.17.0.
Therefore, if you could roll back to v2.16.5 (the latest stable release
of v2.16), that should mitigate this issue.
(I haven't checked whether running upstream v2.16.* binaries is
advisable from a security perspective.)
Thanks
Laszlo
_______________________________________________
edk2-devel mailing list
[email protected]
https://lists.01.org/mailman/listinfo/edk2-devel
