You mean the rockchip driver?

I'm assuming the local Chromium copy of the DMA_BUF_IOCTL_SYNC ioctl (the headers are only exposed >= 4.4) is correct.  Quickly scanning the code (dma-buf.c), the ioctl is a no-op when the driver specific hoot not defined (the rockchip driver uses the default drm_gem_prime_dmabuf_ops):

if (WARN_ON(!dmabuf))
    return -EINVAL;

if (dmabuf->ops->begin_cpu_access)
    ret = dmabuf->ops->begin_cpu_access(dmabuf, direction);

I would check the dmesg if the kernel spews any error when this happens cause it may signal you don't have a valid fd. Anyways, I have a few patches in works that implement cached mappings when DMA_BUF_SYNC_READ is specified on rockchip, write combined otherwise, so I'll assign this to myself.

Yes, I meant rockchip. This is returning a value that results in log output each time we map/unmap a buffer. We know that mapping works as this is currently used for low-latency stylus updates so the problem is not a bad fd. The ioctl is not returning a value that results in logging on Intel devices.

https://chromium-review.googlesource.com/c/461287/

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/a7215699b8815b4aa6ff6d069414054f44065dbe

commit a7215699b8815b4aa6ff6d069414054f44065dbe
Author: Marek Szyprowski <m.szyprowski@samsung.com>
Date: Wed Mar 29 00:16:53 2017

UPSTREAM: dma-buf: add support for compat ioctl

Add compat ioctl support to dma-buf. This lets one to use DMA_BUF_IOCTL_SYNC
ioctl from 32bit application on 64bit kernel. Data structures for both 32
and 64bit modes are same, so there is no need for additional translation
layer.

Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Christian Knig <christian.koenig@amd.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Link: http://patchwork.freedesktop.org/patch/msgid/1487683261-2655-1-git-send-email-m.szyprowski@samsung.com
(cherry picked from commit 888022c0473d079bff9b47fb50434b1f20f8f37f)
Signed-off-by: Gurchetan Singh <gurchetansingh@chromium.org>

BUG= chromium:705189 
TEST=local testing with SYNC ioctl

Change-Id: Id83676d8ac8291a29602bad5e7cf135df2a691b1
Reviewed-on: https://chromium-review.googlesource.com/461287
Commit-Ready: Gurchetan Singh <gurchetansingh@chromium.org>
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Tomasz Figa <tfiga@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>

[modify] https://crrev.com/a7215699b8815b4aa6ff6d069414054f44065dbe/drivers/dma-buf/dma-buf.c