On Fri, Aug 09, 2024 at 08:56:29AM +0200, Rick Wertenbroek wrote: > On Tue, Aug 6, 2024 at 9:15 PM Bjorn Helgaas <helgaas@xxxxxxxxxx> wrote: > > On Tue, Aug 06, 2024 at 06:27:54PM +0200, Rick Wertenbroek wrote: > > > The test for a DMA transfer was done in pci_epf_test_cmd_handler, which > > > if not supported would lead to the endpoint function just printing an > > > error message and waiting for further commands. This would leave the > > > > I guess it's the *test* that prints the error message? Is this the > > "Cannot transfer data using DMA" message? > > That is the error message, the error message is printed by the > endpoint function, on the endpoint device. On the host side, nothing > happens; the test program just hangs because the driver waits > indefinitely. With the change I proposed, the test program completes > the test and will display "NOT OKAY" as normal when a test fails. Thanks for this; something like this would be helpful in the commit log because it makes the two kernels involved more explicit. Bjorn
