On 12/12/18 8:48 PM, Oliver Hartkopp wrote: > Hi Marc, > > On 12/12/18 3:57 PM, Marc Kleine-Budde wrote: >> On 12/5/18 7:07 AM, Oleksij Rempel wrote: >>> Currently, (E)TP transfers will be aborted, as soon application will >>> call close() or exit, as the socket will be automatically closed by the >>> kernel. >> >> With this patch, the application hangs on the close() call until all >> data has been send out. This might take a while. What happens if you >> attach a debugger? I think the wait_event_interruptible() will return as >> attaching a debugger involves signals. > > I'm not sure whether this could be solved differently. The solution is correct IMHO, just the signal related error handling is missing. > It might be that we're currently running into a similar problem with > isotp.c, see > > https://github.com/linux-can/can-utils/issues/113 > > But I think UDP/IP should have an analogue requirement. Will take a look > into it. Marc -- Pengutronix e.K. | Marc Kleine-Budde | Industrial Linux Solutions | Phone: +49-231-2826-924 | Vertretung West/Dortmund | Fax: +49-5121-206917-5555 | Amtsgericht Hildesheim, HRA 2686 | http://www.pengutronix.de |
Attachment:
signature.asc
Description: OpenPGP digital signature
