On Fri, Jan 31, 2020 at 06:37:10PM +0100, Hans Verkuil wrote: > On 1/31/20 6:03 PM, Thierry Reding wrote: > > On Fri, Jan 31, 2020 at 03:29:52PM +0100, Hans Verkuil wrote: > >> On 1/30/20 4:42 PM, Thierry Reding wrote: > >>> On Thu, Jan 30, 2020 at 03:41:50PM +0100, Hans Verkuil wrote: > >>>> Hi Sowjanya, > >>>> > >>>> On 1/28/20 7:23 PM, Sowjanya Komatineni wrote: > >>>>> This series adds Tegra210 VI and CSI driver for built-in test pattern > >>>>> generator (TPG) capture. > >>>>> > >>>>> Tegra210 supports max 6 channels on VI and 6 ports on CSI where each > >>>>> CSI port is one-to-one mapped to VI channel for video capture. > >>>>> > >>>>> This series has TPG support only where it creates hard media links > >>>>> between CSI subdevice and VI video device without device graphs. > >>>>> > >>>>> v4l2-compliance results are available below the patch diff. > >>>>> > >>>>> [v0]: Includes, > >>>>> - Adds CSI TPG clock to Tegra210 clock driver > >>>>> - Host1x video driver with VI and CSI clients. > >>>>> - Support for Tegra210 only. > >>>>> - VI CSI TPG support with hard media links in driver. > >>>>> - Video formats supported by Tegra210 VI > >>>>> - CSI TPG supported video formats > >>>> > >>>> I'm trying to compile this patch series using the media_tree master > >>>> branch (https://git.linuxtv.org//media_tree.git), but it fails: > >>>> > >>>> drivers/staging/media/tegra/tegra-channel.c: In function ‘tegra_channel_queue_setup’: > >>>> drivers/staging/media/tegra/tegra-channel.c:71:15: warning: unused variable ‘count’ [-Wunused-variable] > >>>> 71 | unsigned int count = *nbuffers; > >>>> | ^~~~~ > >>>> drivers/staging/media/tegra/tegra-channel.c: In function ‘tegra_channel_init’: > >>>> drivers/staging/media/tegra/tegra-channel.c:518:55: error: ‘struct host1x_client’ has no member named ‘host’ > >>>> 518 | struct tegra_camera *cam = dev_get_drvdata(vi->client.host); > >>>> | ^ > >>>> make[4]: *** [scripts/Makefile.build:265: drivers/staging/media/tegra/tegra-channel.o] Error 1 > >>>> make[4]: *** Waiting for unfinished jobs.... > >>>> drivers/staging/media/tegra/tegra-vi.c: In function ‘tegra_vi_tpg_graph_init’: > >>>> drivers/staging/media/tegra/tegra-vi.c:157:55: error: ‘struct host1x_client’ has no member named ‘host’ > >>>> 157 | struct tegra_camera *cam = dev_get_drvdata(vi->client.host); > >>>> | ^ > >>>> drivers/staging/media/tegra/tegra-vi.c: In function ‘tegra_vi_init’: > >>>> drivers/staging/media/tegra/tegra-csi.c: In function ‘tegra_csi_init’: > >>>> drivers/staging/media/tegra/tegra-vi.c:213:51: error: ‘struct host1x_client’ has no member named ‘host’ > >>>> 213 | struct tegra_camera *cam = dev_get_drvdata(client->host); > >>>> | ^~ > >>>> drivers/staging/media/tegra/tegra-csi.c:259:51: error: ‘struct host1x_client’ has no member named ‘host’ > >>>> 259 | struct tegra_camera *cam = dev_get_drvdata(client->host); > >>>> | ^~ > >>>> drivers/staging/media/tegra/tegra-vi.c: In function ‘tegra_vi_exit’: > >>>> drivers/staging/media/tegra/tegra-vi.c:246:51: error: ‘struct host1x_client’ has no member named ‘host’ > >>>> 246 | struct tegra_camera *cam = dev_get_drvdata(client->host); > >>>> | ^~ > >>>> drivers/staging/media/tegra/tegra-csi.c: In function ‘tegra_csi_exit’: > >>>> drivers/staging/media/tegra/tegra-csi.c:286:51: error: ‘struct host1x_client’ has no member named ‘host’ > >>>> 286 | struct tegra_camera *cam = dev_get_drvdata(client->host); > >>>> | ^~ > >>>> > >>>> And indeed, struct host1x_client as defined in include/linux/host1x.h doesn't > >>>> have a 'host' field. > >>>> > >>>> Does this series depend on another patch that's not yet in mainline? > >>> > >>> Sowjanya's been working on top of linux-next, so, yes, this patch > >>> depends on a change that's been merged into the DRM tree for v5.6-rc1. > >>> > >>> Thierry > >>> > >> > >> Is there a specific linux-next tag that works? I tried next-20200131 but that > >> failed to boot. Same problem with the mainline repo since the host1x patches > >> were merged yesterday. It compiles fine, but the boot just stops. Or am I > >> missing some kernel config that is now important to have? > > > > linux-next and mainline are currently regressing on Tegra210 (and some > > Tegra124) boards. I just sent out a series that fixes the regression for > > me: > > > > http://patchwork.ozlabs.org/project/linux-tegra/list/?series=156215 > > > > Please test if this works for you. If so, I'll send this to Dave as soon > > as possible. > > I'll try it on Tuesday as I don't have access to the Jetson TX1 until then. It > looks promising since I think that the last message I saw was a PM message. Great. My local testing on Jetson Nano confirms that this fixes boot on top of linux-next and I've also run it through our internal test farm with success. I'll push this to my drm/tegra/for-next branch, so it should show up in linux-next on Monday. Thierry