Closed ethomson closed 9 years ago
For additional background, see http://lists.gnu.org/archive/html/bug-gnulib/2014-09/msg00045.html.
Please note the updated patch later in the thread (http://lists.gnu.org/archive/html/bug-gnulib/2014-09/msg00070.html) which is what was cherry-picked here.
MSysGit - the development behind Git for Windows » git #277 SUCCESS This pull request looks good (what's this?)
Thank you so much!
Do you want a new 1.9.5.msysgit.1
release with this?
Nope! I'm in no hurry, I'm happy to use a locally patched build. Thanks! :)
Okay, then I'll forward port this patch to the upcoming MSys2-based Git for Windows.
MSysGit - the development behind Git for Windows » git #278 SUCCESS This pull request looks good (what's this?)
Good catch!
MSysGit - the development behind Git for Windows » git #279 SUCCESS This pull request looks good (what's this?)
@ethomson @dscho Is this patch safe to include into maint-1.9? If yes I would just cherry-pick without PR. Due to the openssl security bugs I'm currently preparing a 1.9.5.msysgit.1 release.
@t-b yes, it is safe to pick. Thank you!
@dscho Thanks. I"ve picked it.
Ensure that when passing a pipe, the gnulib poll replacement will not return 0 before the timeout has passed.
Not obeying the timeout (and merely returning 0) causes pathological behavior when preparing a packfile for a repository and taking a long time to do so. If poll were to return 0 immediately, this would cause keep-alives to get sent as quickly as possible until the packfile was created. Such deviance from the standard would cause megabytes (or more) of keep-alive packets to be sent.
GetTickCount
is used as it is efficient, stable and monotonically increasing. (NeitherGetSystemTime
norQueryPerformanceCounter
have all three of these properties.)