On Tue, Jun 09, 2015 at 07:08:34PM +0800, anna wrote:
[...]
mgp->sram_size satisfy the condition in line 4062, the value of
status is 0. So the function myri10ge_probe() will return 0 to
its caller functions when it runs error, leading to a wrong
return value of function myri10ge_probe().
Okay, that is a bug. I will try to prepare a patch. Is this a real
problem for you (i.e. your setup is tripping this bug), or a
theoretical one you found via scripts or something? If it is a real
problem, can you give me more context (e.g. NIC model, firmware
version)?
Also, I am going to need a proper "your name <address>", unless you do
not want your name in the patch.