https://bugs.kde.org/show_bug.cgi?id=377394
Daniel Vrátil <dvra...@kde.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Assignee|chrig...@fastmail.fm |dvra...@kde.org
CC| |dvra...@kde.org
Ever confirmed|0 |1
Status|UNCONFIRMED |CONFIRMED
--- Comment #1 from Daniel Vrátil <dvra...@kde.org> ---
Upon initial investigation, it turns out this might be really tricky to
implement properly because Qt does not really provide us with any means of
monitoring the progress of data being sent. It simply buffers the entire
payload and passes it to the kernel, which buffers it internally again.
We could maybe try to abuse the way Qt works with its internal buffers and pass
the payload to Qt in several smaller chunks, then we would get notified upon
every chunk being written to the socket and we could check how much data
remains in Qt's the buffer. To get a reasonable progress reporting we would,
however, have to pass the payload in really small chunks. The question is what
is the performance impact (in the terms of upload speed) compared to just
passing the entire payload to the kernel in one piece to deal with it
internally.
--
You are receiving this mail because:
You are watching all bug changes.
