V Tue, 26 Feb 2019 13:09:30 +0100 Nicolas Saenz Julienne <nsaenzjulienne@xxxxxxx> napsáno: > Hi, > as I'm sure most of you are aware of, every URB submission triggers a small > memory allocation in the host controller. On top of that it might be run in > atomic context. This has always seemed to me as a possible source of very hard > to reproduce and potentially nasty issues. For example I'm thinking of embedded > systems driving some critical operation through USB. > > I wonder if anyone has been able to log any failure on that area. I was unable > to find anything on the relevant bugzillas, nor by talking with some > knowledgeable people. Actually, searching for "page allocation failure" and xhci_urb_enqueue finds a few posts. Among others: https://lists.gt.net/xen/devel/183060 Although the allocation failure may itself be a result of a bug, if you read on, at one point Konrad says: > Then the em28xx gets horribly confused that it could not send an URB > anymore and from there one things get worst. This makes me believe there is some value in making things at least a bit better. Just my two cents, Petr T
Attachment:
pgpEzbcfblZ7g.pgp
Description: Digitálnà podpis OpenPGP
