D4975: Prevent misdetection of EOF on Linux

[Peter Wu]

Lekensteyn added a comment.


  In https://phabricator.kde.org/D4975#95727, @ossi wrote:
  
  > please verify that this is in fact a kernel bug (include ml references), not something that should be expected for some weird reasons (compare the solaris path).
  
  
  I suspect it is a kernel bug, but have no ml references. It seems to be a race condition for which it is hard to prove that it does not exist.
  See https://bugs.kde.org/show_bug.cgi?id=372991#c10, poll returned but FIONREAD outputted 0 bytes.
  
  > if this is a bug, does it actually affect released versions which are not being "upgraded away" on short notice? i wouldn't want a workaround for a bug that had a very short life span.
  
  This issue has been there for several months (maybe even a year) and I regularly (multiple times a week) hit this condition when interrupting a program (today it even happened when interrupting a `git push` which was waiting for the SSH passphrase).
  
  I have tried to pinpoint the root cause (but failed even after spending hours), what else should I do or can I try?

INLINE COMMENTS

> ossi wrote in kptydevice.cpp:307
> this should be constrained more - affected versions, and the fact that it's a bug.

Constrained in what sense? Should the first known broken kernel version be noted here?

REPOSITORY
  R291 KPty

REVISION DETAIL
  https://phabricator.kde.org/D4975

To: Lekensteyn, #konsole, hindenburg, ossi
Cc: #frameworks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-frameworks-devel/attachments/20170317/625fd3d9/attachment.html>