Re: [linux-sunxi] Re: [PATCH] dma: sun4i: expose block size and wait cycle configuration to DMA users

From	Vinod Koul <vinod.koul@intel.com>
Newsgroups	linux.kernel
Subject	Re: [linux-sunxi] Re: [PATCH] dma: sun4i: expose block size and wait cycle configuration to DMA users
Date	2016-03-08 11:10 +0100
Message-ID	<ramKv-7SB-37@gated-at.bofh.it> (permalink)
References	(2 earlier) <ra4Xf-4ss-5@gated-at.bofh.it> <raa6C-7JQ-5@gated-at.bofh.it> <rag2m-38j-21@gated-at.bofh.it> <rakII-6lY-39@gated-at.bofh.it> <ralv5-6V9-15@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Tue, Mar 08, 2016 at 09:42:31AM +0100, Hans de Goede wrote:
> <wild speculation>
> 
> I see 2 possible reasons why waiting till checking for drq can help:
> 
> 1) A lot of devices have an internal fifo hooked up to a single mmio data
> register which gets read using the general purpose dma-engine, it allows
> this fifo to fill, and thus do burst transfers
> (We've seen similar issues with the scanout engine for the display which
>  has its own dma engine, and doing larger transfers helps a lot).
> 
> 2) Physical memory on the sunxi SoCs is (often) divided into banks
> with a shared data / address bus doing bank-switches is expensive, so
> this wait cycles may introduce latency which allows a user of another
> bank to complete its RAM accesses before the dma engine forces a
> bank switch, which ends up avoiding a lot of (interleaved) bank switches
> while both try to access a different banj and thus waiting makes things
> (much) faster in the end (again a known problem with the display
> scanout engine).
> 
> </wild speculation>
> 
> Note the differences these kinda tweaks make can be quite dramatic,
> when using a 1920x1080p60 hdmi output on the A10 SoC with a 16 bit
> memory bus (real world worst case scenario), the memory bandwidth
> left for userspace processes (measured through memset) almost doubles
> from 48 MB/s to 85 MB/s, source:
> http://ssvb.github.io/2014/11/11/revisiting-fullhd-x11-desktop-performance-of-the-allwinner-a10.html
> 
> TL;DR: Waiting before starting DMA allows for doing larger burst
> transfers which ends up making things more efficient.
> 
> Given this, I really expect there to be other dma-engines which
> have some option to wait a bit before starting/unpausing a transfer
> instead of starting it as soon as (more) data is available, so I think
> this would make a good addition to dma_slave_config.

I tend to agree but before we do that I would like this hypothesis to be
confirmed :)

-- 
~Vinod

Thread

[PATCH] dma: sun4i: expose block size and wait cycle configuration to DMA users Boris Brezillon <boris.brezillon@free-electrons.com> - 2016-03-07 11:30 +0100
  Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-07 16:00 +0100
    Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Boris Brezillon <boris.brezillon@free-electrons.com> - 2016-03-07 16:10 +0100
      Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Maxime Ripard <maxime.ripard@free-electrons.com> - 2016-03-07 21:40 +0100
        Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-08 04:00 +0100
        Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-08 04:00 +0100
          Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Maxime Ripard <maxime.ripard@free-electrons.com> - 2016-03-08 09:00 +0100
            Re: [linux-sunxi] Re: [PATCH] dma: sun4i: expose block size and wait  cycle configuration to DMA users Hans de Goede <hdegoede@redhat.com> - 2016-03-08 09:50 +0100
              Re: [linux-sunxi] Re: [PATCH] dma: sun4i: expose block size and wait  cycle configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-08 11:10 +0100
                Re: [linux-sunxi] Re: [PATCH] dma: sun4i: expose block size and wait  cycle configuration to DMA users Maxime Ripard <maxime.ripard@free-electrons.com> - 2016-03-09 12:20 +0100
            Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Boris Brezillon <boris.brezillon@free-electrons.com> - 2016-03-08 09:50 +0100
              Re: [linux-sunxi] Re: [PATCH] dma: sun4i: expose block size and  wait cycle configuration to DMA users Priit Laes <plaes@plaes.org> - 2016-03-08 10:20 +0100
              Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-08 11:10 +0100
              Re: [linux-sunxi] Re: [PATCH] dma: sun4i: expose block size and wait  cycle configuration to DMA users LABBE Corentin <clabbe.montjoie@gmail.com> - 2016-03-09 11:10 +0100
            Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-08 11:00 +0100
          Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Boris Brezillon <boris.brezillon@free-electrons.com> - 2016-03-09 11:20 +0100
            Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-11 07:30 +0100
              Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Boris Brezillon <boris.brezillon@free-electrons.com> - 2016-03-11 10:50 +0100
                Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-11 11:10 +0100
                Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Boris Brezillon <boris.brezillon@free-electrons.com> - 2016-03-11 11:30 +0100
                Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-11 12:20 +0100
          Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Boris Brezillon <boris.brezillon@free-electrons.com> - 2016-03-09 12:30 +0100
            Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-11 07:30 +0100
              Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Boris Brezillon <boris.brezillon@free-electrons.com> - 2016-03-11 10:50 +0100
                Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-11 11:10 +0100
                Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Maxime Ripard <maxime.ripard@free-electrons.com> - 2016-03-11 12:00 +0100
                Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-11 12:20 +0100
                Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Maxime Ripard <maxime.ripard@free-electrons.com> - 2016-03-14 12:50 +0100
                Re: [PATCH] dma: sun4i: expose block size and wait cycle  configuration to DMA users Vinod Koul <vinod.koul@intel.com> - 2016-03-16 04:20 +0100
  Re: [linux-sunxi] [PATCH] dma: sun4i: expose block size and wait  cycle configuration to DMA users Priit Laes <plaes@plaes.org> - 2016-03-07 16:40 +0100
    Re: [linux-sunxi] [PATCH] dma: sun4i: expose block size and wait  cycle configuration to DMA users Boris Brezillon <boris.brezillon@free-electrons.com> - 2016-03-07 17:00 +0100
      Re: [linux-sunxi] [PATCH] dma: sun4i: expose block size and wait  cycle configuration to DMA users Emilio López <emilio@elopez.com.ar> - 2016-03-07 18:20 +0100

csiph-web