On Tue, Jul 13, 2021 at 03:19:39PM +0100, Ananyev, Konstantin wrote:
>
> > +#include "rte_dmadev_core.h"
> > +
> > +/**
> > + * DMA flags to augment operation preparation.
> > + * Used as the 'flags' parameter of rte_dmadev_copy/copy_sg/fill/fill_sg.
> > + */
> > +#define RTE_DMA_FLAG_FENCE (1ull << 0)
> > +/**< DMA fence flag
> > + * It means the operation with this flag must be processed only after all
> > + * previous operations are completed.
> > + *
> > + * @see rte_dmadev_copy()
> > + * @see rte_dmadev_copy_sg()
> > + * @see rte_dmadev_fill()
> > + * @see rte_dmadev_fill_sg()
> > + */
> > +
> > +/**
> > + * @warning
> > + * @b EXPERIMENTAL: this API may change without prior notice.
> > + *
> > + * Enqueue a copy operation onto the virtual DMA channel.
> > + *
> > + * This queues up a copy operation to be performed by hardware, but does
> > not
> > + * trigger hardware to begin that operation.
> > + *
> > + * @param dev_id
> > + * The identifier of the device.
> > + * @param vchan
> > + * The identifier of virtual DMA channel.
> > + * @param src
> > + * The address of the source buffer.
> > + * @param dst
> > + * The address of the destination buffer.
> > + * @param length
> > + * The length of the data to be copied.
> > + * @param flags
> > + * An flags for this operation.
> > + *
> > + * @return
> > + * - 0..UINT16_MAX: index of enqueued copy job.
> > + * - <0: Error code returned by the driver copy function.
> > + */
> > +__rte_experimental
> > +static inline int
> > +rte_dmadev_copy(uint16_t dev_id, uint16_t vchan, rte_iova_t src,
> > rte_iova_t dst,
> > + uint32_t length, uint64_t flags)
> > +{
> > + struct rte_dmadev *dev = &rte_dmadevices[dev_id];
>
> One question I have - did you guys consider hiding definitions of struct
> rte_dmadev
> and rte_dmadevices[] into .c straight from the start?
> Probably no point to repeat our famous ABI ethdev/cryptodev/... pitfalls here.
>
I considered it, but I found even moving one operation (the doorbell one)
to be non-inline made a small but noticable perf drop. Until we get all the
drivers done and more testing in various scenarios, I'd rather err on the
side of getting the best performance.
