05/08/2020 10:40, Bruce Richardson: > On Wed, Aug 05, 2020 at 10:27:27AM +0200, Thomas Monjalon wrote: > > 05/08/2020 01:39, Dmitry Kozlyuk: > > > On Tue, 4 Aug 2020 09:29:46 +0300, Fady Bader wrote: > > > > Telemetry didn't compile under Windows. > > > > Empty stubs implementation was added for Windows. > > > > > > > > Signed-off-by: Fady Bader <f...@mellanox.com> > > > > --- > > > > lib/librte_telemetry/rte_telemetry.h | 4 +++ > > > > lib/librte_telemetry/telemetry.c | 51 > > > > ++++++++++++++++++++++++++++++++- > > > > lib/librte_telemetry/telemetry_legacy.c | 26 ++++++++++++++++- > > > > 3 files changed, 79 insertions(+), 2 deletions(-) > > > > > > You could #ifdef code in librte_ethdev that uses librte_telemetry and not > > > build librte_telemetry at all. This approach is already taken in > > > eal_common_options.c and it avoids spreading #ifdef throughout telemetry > > > code. > > > > The problem is that telemetry can be used anywhere, not only in ethdev. > > I feel it is better to #ifdef telemetry than every telemetry calls. > > > Given that the majority of telemetry has no external dependencies (jansson > is only used for the older compatibility part), and it uses sockets for > communication, is there a reason why it can't just be made to build and > work on windows? The unix domain socket could be converted to a standard > UDP socket on localhost, perhaps. Is there anything unix-specific beyond > that?
Yes of course telemetry should build on Windows. But it seems nobody committed yet to work on it. Feel free :)
