Re: [PATCH 0/9] bus: mhi: ep: Add async read/write support

[Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>]

On Wed, Dec 13, 2023 at 01:31:03PM -0600, Bjorn Helgaas wrote:
> On Mon, Nov 27, 2023 at 06:15:20PM +0530, Manivannan Sadhasivam wrote:
> > Hi,
> > 
> > This series add async read/write support for the MHI endpoint stack by
> > modifying the MHI ep stack and the MHI EPF (controller) driver.
> > 
> > Currently, only sync read/write operations are supported by the stack,
> > this resulting in poor data throughput as the transfer is halted until
> > receiving the DMA completion. So this series adds async support such
> > that the MHI transfers can continue without waiting for the transfer
> > completion. And once the completion happens, host is notified by sending
> > the transfer completion event.
> > 
> > This series brings iperf throughput of ~4Gbps on SM8450 based dev platform,
> > where previously 1.6Gbps was achieved with sync operation.
> > 
> > - Mani
> > 
> > Manivannan Sadhasivam (9):
> >   bus: mhi: ep: Pass mhi_ep_buf_info struct to read/write APIs
> >   bus: mhi: ep: Rename read_from_host() and write_to_host() APIs
> >   bus: mhi: ep: Introduce async read/write callbacks
> >   PCI: epf-mhi: Simulate async read/write using iATU
> >   PCI: epf-mhi: Add support for DMA async read/write operation
> >   PCI: epf-mhi: Enable MHI async read/write support
> >   bus: mhi: ep: Add support for async DMA write operation
> >   bus: mhi: ep: Add support for async DMA read operation
> >   bus: mhi: ep: Add checks for read/write callbacks while registering
> >     controllers
> > 
> >  drivers/bus/mhi/ep/internal.h                |   1 +
> >  drivers/bus/mhi/ep/main.c                    | 256 +++++++++------
> >  drivers/bus/mhi/ep/ring.c                    |  41 +--
> >  drivers/pci/endpoint/functions/pci-epf-mhi.c | 314 ++++++++++++++++---
> >  include/linux/mhi_ep.h                       |  33 +-
> >  5 files changed, 485 insertions(+), 160 deletions(-)
> 
> Mani, do you want to merge this via your MHI tree?  If so, you can
> include Krzysztof's Reviewed-by tags and my:
> 
> Acked-by: Bjorn Helgaas <bhelgaas@google.com>
> 
> If you think it'd be better via the PCI tree, let me know and we can
> do that, too.
> 

Thanks Bjorn! Yes, to avoid possible conflicts with other MHI patches, I need to
take this series via MHI tree.

- Mani

> Bjorn

-- 
மணிவண்ணன் சதாசிவம்