Vinod Koul <vinod.koul@xxxxxxxxx> writes: >> It's not just sound. The recommendation has always been set this >> flag unless you know why you are not setting it. So pretty much >> every DMAengine client driver sets it. > Actually there has never been a recommendation for slave cases. It came from async > API and slave users should never have used it, untill now. > > Now to solve the mess, either we can revert this patch and find a new one > for actual reuse of descriptors. Alternate would be fix users > > I am leaning towards former then follow up by fixing the users who dont want > this. Since this is going to be a tree wide update it will need one merge > cycle to complete. Once done we can bring back the usage required for actual > descriptor reuse case. This will lead only one broken driver from Robert. Is > the client for your driver merged? No Vinod, it's still under review. And I'll follow any of your directions here, either revert or tree wide update or new flag. I can help to do the sweeping too, if I had a clear idea of what you have in mind. After all, I want that feature, so I can contribute to tidy up. Cheers. -- Robert -- To unsubscribe from this list: send the line "unsubscribe dmaengine" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
