Looking at the code it seems that dmatest takes the max aligned buffer size from the actual allocated buffer size. This can make the buffer being tested much smaller than the actual specified test buffer size. When testing, the intent is to test up to N bytes of buffers right? Shouldn't we try to allocate a buffer that's large enough to accommodate the alignment restriction and then work on the aligned buffer instead? i.e. src_buf = kmalloc(buf_size+align, GFP_KERNEL); src_off += (unsigned long)PTR_ALIGN(src_buf, align) - (unsigned long)src_buf; -- 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
