mlx4_core 0000:07:00.0: swiotlb buffer is full and OOM observed during stress test on reset_controller

Christoph Hellwig hch at lst.de
Wed Mar 8 07:48:15 PST 2017

Previous message (by thread): mlx4_core 0000:07:00.0: swiotlb buffer is full and OOM observed during stress test on reset_controller
Next message (by thread): mlx4_core 0000:07:00.0: swiotlb buffer is full and OOM observed during stress test on reset_controller
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Why is that system using swiotlb?  mlx4 really doesn't have any
weird addressing limits, does it?

Previous message (by thread): mlx4_core 0000:07:00.0: swiotlb buffer is full and OOM observed during stress test on reset_controller
Next message (by thread): mlx4_core 0000:07:00.0: swiotlb buffer is full and OOM observed during stress test on reset_controller
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

More information about the Linux-nvme mailing list