On Fri, Nov 20, 2015 at 01:59:02PM -0500, Ben Romer wrote: > On 11/17/2015 05:18 PM, Greg KH wrote: > >On Tue, Nov 17, 2015 at 09:57:58AM -0500, Benjamin Romer wrote: > >>This patch series adds a centralized infrastructure and device support > >>for channel interrupts sent to s-Par virtual devices. With these changes, > >>the visorhba device is ~80% faster than with only polling, and visornic > >>receives a speedup of over 3500% (from ~9Mb/s to between 360Mb/s and > >>390Mb/s). > > > >Why are you adding new functions to the code before it is merged out of > >staging? Please don't do that, please get the code out of staging > >first. > > > >thanks, > > > >greg k-h > > > > Hi Greg, > > Interrupt support was originally part of the driver set when we submitted > it, but we removed it when we renamed the drivers and restructured the set. > The work was listed in the TODO so I thought it would be okay to patch it in > staging, but if we should wait until the drivers are out, then we will wait. > :) Well, it might be, but as you didn't say that in the changelog here, how was I supposed to remember that? :) > Do you think our drivers are clean enough to be moved out soon? None of the > files generate checkpatch errors anymore (the one in visorhba_main was > discussed before and decided to be okay). There are 7 files that generate > warnings still, controlvmchannel.h being the biggest offender, with the rest > all in single digits. > > I'll get those warnings cleaned up - are there other things you'd like us to > fix? That's a great first-step, after that, we can do an audit of the code, but please do all of the "easy things" first. thanks, greg k-h _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
