<!-- Some styling for better description lists --><style type='text/css'>dt { font-weight: bold;float: left;display:inline;margin-right: 1em} dd { display:block; margin-left: 2em}</style>
***: digetx has joined #linux-media
<br> mrpops2ko has quit IRC ()
<br> mrpops2ko has joined #linux-media
<br> nas_chung has joined #linux-media
<br> mvaittin has joined #linux-media
<br> nas_chung has quit IRC ()
<br> ten1572377432463050661821 has quit IRC (Ping timeout: 480 seconds)
<br> djrscally has joined #linux-media
<br> ao2 has joined #linux-media
<br> xroumegue has quit IRC (Ping timeout: 480 seconds)
<br> xroumegue has joined #linux-media
<br> mvaittin has quit IRC (Ping timeout: 480 seconds)
<br> mvaittin has joined #linux-media
hverkuil: <u>jmondi</u>: was the pisp_be driver ever checked with v4l2-compliance? I noticed that pispbe_node_queue_ops didn't set the wait_prepare/finish ops, and I believe v4l2-compliance would fail on that.
***: frieder has joined #linux-media
jmondi: <u>hverkuil</u>: I did (at least according to the cover letter I sent)
<br> <u>hverkuil</u>: I'm working on it to address Laurent's comments so I can re-test
hverkuil: It's only seen if you run with the -s option. That's probably the root cause.
***: prabhakalad has quit IRC (Quit: Konversation terminated!)
jmondi: ah maybe I didn't do that
hverkuil: Although thinking it over, that particular test can probably be run without the streaming tests.
pinchartl: I doubt v4l2-compliance will be able to test streaming on the pispbe driver
hverkuil: That specific test doesn't actually queue buffers.
<br> It allocates two buffers and calls STREAMON. Then it starts two threads: one calls DQBUF (will wait indefinitely since no buffers are queued), and the other calls STREAMOFF. That streamoff should succeed if the locking is handled correctly. If wait_prepare/finish isn't set, then that test will fail.
<br> <u>svarbanov</u>: I noticed that the venus driver also doesn't set the wait_prepare/finish ops. I suspect that the venus driver has the same problem as the pisp_be driver (discussed above).
<br> I have changed v4l2-compliance so that it performs the blocking wait test even if the -s option wasn't specified. Can you verify that this is indeed a venus driver bug?
***: prabhakalad has joined #linux-media
pinchartl: <u>hverkuil</u>: there's a single driver that has custom handlers for those two operations
<br> drivers/media/dvb-core/dvb_vb2.c
<br> could we fix that and drop the ops ?
hverkuil: That's OK, that's the dvb core vb2 driver.
pinchartl: or at least default them to vb2_ops_wait_prepare and vb2_ops_wait_finish ?
<br> so that drivers won't have to to it manually and we can avoid these bugs in the future ?
hverkuil: There are three more drivers that do not set wait_prepare/finish: omap3isp, omap4iss and atomisp.
<br> And yes, I believe this should be moved away from drivers. Which is why I am now looking at this.
***: sgruszka has joined #linux-media
pinchartl: :-)
hverkuil: I think omap3isp will fail this test as well: isp_video_dqbuf locks video->queue_lock, but without a custom wait_prepare/finish this will block isp_video_streamoff.
<br> Probably the same is true for omap4
pinchartl: I think Sakari can test omap3isp
<br> I can't test omap4iss, I doubt anyone still uses that driver
hverkuil: Is it time to drop omap4iss?
pinchartl: probably
***: mripard has joined #linux-media
-: pinchartl sheds a tear
hverkuil: <u>pinchartl</u>: would it make sense to drop omap4iss immediately, or move it to drivers/staging/media/deprecated/ and remove it 1-2 kernel cycles later?
pinchartl: I think it could be removed immediately
hverkuil: I'll prepare a patch.
javierm: pinchartl, hverkuil: I've a OMAP4 pandaboard in a drawer but haven't booted in a decade...
<br> FYI u-boot recently removed support for all OMAP4 boards
<br> https://lists.denx.de/pipermail/u-boot/2024-July/558846.html
pinchartl: <u>javierm</u>: I also have one in a drawer :-)
hverkuil: me too :-)
javierm: :) there was a time that every Linux developer involved with ARM had an OMAP board over their desk
***: ten1572377432463050661821 has joined #linux-media
prabhakalad: <u>pinchartl</u>: with introduction to streams api, for virtual channels on sensors should we be creating pads for each channel? If I remember previously during the CRU work the streams api was WIP there was a comment saying we dont need pads for virtual channel.
pinchartl: <u>prabhakalad</u>: no, you don't need multiple pads
<br> you will have a single source pad, with one stream per VC
<br> there will be "internal pads" too, which is a concept we're introducting, it's not in mainline yet
***: mvaittin has quit IRC (Ping timeout: 480 seconds)
prabhakalad: thanks, I was going through the latest patch series from Tomi for RPI which has streams API where csi support 4 channels there are 4 pads created.
<br> > +#define CSI2_NUM_CHANNELS 4
<br> > +
<br> > +#define CSI2_PAD_SINK 0
<br> > +#define CSI2_PAD_FIRST_SOURCE 1
<br> > +#define CSI2_PAD_NUM_SOURCES 4
<br> > +#define CSI2_NUM_PADS 5
<br> https://www.spinics.net/lists/linux-media/msg260931.html
pinchartl: those are the source pads of the CSI-2 receiver
<br> the CSI-2 RX has a single sink pad, on which it receives all VCs
<br> and that particular device has 4 outputs
prabhakalad: ahh got you, sorry for the confusion.
pinchartl: no worries
prabhakalad: thank you.
***: tmerciai1 has quit IRC (Remote host closed the connection)
<br> andrzej_p32 has quit IRC (Quit: The Lounge - https://thelounge.chat)
<br> digetx is now known as Guest2214
<br> digetx has joined #linux-media
<br> Guest2214 has quit IRC (Ping timeout: 480 seconds)
<br> sgruszka has quit IRC (Ping timeout: 480 seconds)
<br> eelstrebor has joined #linux-media
<br> eelstrebor has quit IRC (Quit: Ex-Chat)
<br> nickware has quit IRC (Remote host closed the connection)
<br> nickware has joined #linux-media
prabhakalad: just trying my luck, anyone know a 22pin mipi camera which I can buy off the shelf?
pinchartl: 22-pin as in compatible with the Raspberry Pi pinout ?
prabhakalad: <u>pinchartl</u>: Yes with RPI pinout.
pinchartl: get any rpi 15-pin camera module, and add a https://www.adafruit.com/product/5818 cable
<br> or use https://thepihut.com/products/camera-cable-joiner-for-raspberry-pi-15-pin-to-22-pin as an adapter board
prabhakalad: <u>pinchartl</u>: thanks :)
***: sgruszka has joined #linux-media
<br> frieder has quit IRC (Remote host closed the connection)
<br> sgruszka has quit IRC (Quit: Leaving)
<br> digetx has quit IRC (Ping timeout: 480 seconds)
<br> digetx has joined #linux-media
<br> ao2 has quit IRC (Quit: Leaving)
<br> fedora has joined #linux-media
<br> fedora has quit IRC ()
<br> gouchi has joined #linux-media
<br> darkapex has quit IRC (Remote host closed the connection)
<br> darkapex has joined #linux-media
<br> jmassot_ has joined #linux-media
<br> Moarc_ has joined #linux-media
<br> digetx_ has joined #linux-media
<br> paulk-bis has joined #linux-media
<br> digetx has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> ten1572377432463050661821 has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> mripard has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> prabhakalad has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> xroumegue has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> djrscally has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> jmassot has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> Moarc has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> paulk has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> Kwiboo has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> AndyCap has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> tomba has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> z3ntu has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> djrscally[m] has quit IRC (reticulum.oftc.net helix.oftc.net)
<br> prabhakalad has joined #linux-media
<br> Kwiboo has joined #linux-media
<br> AndyCap has joined #linux-media
<br> digetx has joined #linux-media
<br> ten1572377432463050661821 has joined #linux-media
<br> mripard has joined #linux-media
<br> xroumegue has joined #linux-media
<br> djrscally has joined #linux-media
<br> jmassot has joined #linux-media
<br> paulk has joined #linux-media
<br> tomba has joined #linux-media
<br> z3ntu has joined #linux-media
<br> djrscally[m] has joined #linux-media
<br> z3ntu has quit IRC (Ping timeout: 480 seconds)
<br> ten15723774324630506618216 has joined #linux-media
<br> paulk has quit IRC (Read error: Network is unreachable)
<br> ten1572377432463050661821 has quit IRC (Ping timeout: 480 seconds)
<br> ten15723774324630506618216 is now known as ten1572377432463050661821
<br> digetx has quit IRC (Ping timeout: 480 seconds)
<br> djrscally has quit IRC (Ping timeout: 480 seconds)
<br> jmassot has quit IRC (Ping timeout: 480 seconds)
<br> mripard has quit IRC (Ping timeout: 480 seconds)
<br> z3ntu has joined #linux-media
<br> gouchi has quit IRC (Remote host closed the connection)