New implementation of pseudo console support (experimental)

Ken Brown kbrown@cornell.edu
Wed Jul 22 12:13:45 GMT 2020


On 7/22/2020 7:49 AM, Corinna Vinschen wrote:
> On Jul 22 17:45, Takashi Yano via Cygwin-developers wrote:
>> On Wed, 22 Jul 2020 03:17:51 +0900
>> Takashi Yano via Cygwin-developers <cygwin-developers@cygwin.com> wrote:
>>> Hi Corinna,
>>>
>>> On Mon, 20 Jul 2020 10:06:13 +0200
>>> Corinna Vinschen wrote:
>>>> On Jul 18 14:30, Takashi Yano via Cygwin-developers wrote:
>>>>> Hi Corinna,
>>>>>
>>>>> On Fri, 17 Jul 2020 13:19:12 +0200
>>>>> Corinna Vinschen wrote:
>>>>>> Hi Takashi,
>>>>>>
>>>>>> On Jul  1 20:47, Takashi Yano via Cygwin-developers wrote:
>>>>>>> [...]
>>>>>>> Revise the patch to fit the current git head.
>>>>>>
>>>>>> are you satisfied with the code?  If you want to merge it,
>>>>>> I'd bump Cygwin to 3.2.
>>>>>
>>>>> Since this new implementation has both advantages and disadvantages,
>>>>> there might be some options.
>>>>>
>>>>> 1) Default to new implementation and leave the current one as an
>>>>>    alternative. Switch them using the environment CYGWIN.
>>>>> 2) Default to current implementation and add the new one as an
>>>>>    alternative. Switch them using the environment CYGWIN.
>>>>> 3) Adopt only new implementation and throw the current one away.
>>>>>
>>>>> What do you think?
>>>>
>>>> Do you really want to maintain twice as much code doing the same stuff
>>>> and constantly having to ask users which version of the code they are
>>>> running?  The maintenance cost outweighs the advantages, IMHO.
>>>> Personally I'd go for option 3.
>>>
>>> Personally, I feel a tinge of sadness to discard the current code,
>>> however, your opinion sounds reasonable.
>>>
>>> I will submit a new patch in which all the codes specific to the
>>> current implementation are removed.
>>
>> Attached is the patch in git format-patch format.
>> All the codes specific to the current implementation are removed.
>>
>> Despite the utmost care, the changes are relatively large, so some
>> degradation may exist.
>>
>> I will appreciate if you could test.
> 
> I built a new DLL with this patch and this looks pretty good to me so
> far.  I'm a bad tester for this stuff, though, using just basic Cygwin
> tools usually.
> 
> At one point we should probably just apply this patch and create
> developer snapshots to broaden the tester base, but for now, maybe
> I should upload my test DLL somewhere?  Is anybody besides Thomas
> willing to test this new pseudo console stuff?

I don't have any specific tests for the pseudo console, but I'll install 
Takashi's patch and do some things that have often caught bugs in the past 
(e.g., running through my FIFO test suite).

Ken


More information about the Cygwin-developers mailing list