<PabloPL>
Is Exynos MFC working for anyone on mainline kernel (for example 4.20)? I'm getting very strange kernel oops after just enabling this module (since around 4.19.rc7 - i think it was working fine around rc4)
<PabloPL>
Unable to handle kernel NULL pointer dereference at virtual address 00000000
<PabloPL>
/ stacktrace
<PabloPL>
[ 8.992674] [<803d7e44>] (platform_match) from [<803d62f4>] (__device_attach_driver+0x38/0xcc)
<PabloPL>
[ 8.975210] [<805b8f70>] (strcmp) from [<803d7e44>] (platform_match+0xa8/0xbc)
<PabloPL>
[ 9.011590] [<803d62f4>] (__device_attach_driver) from [<803d4324>] (bus_for_each_drv+0x58/0xb8)
<PabloPL>
[ 9.030683] [<803d4324>] (bus_for_each_drv) from [<803d5c40>] (__device_attach+0xd0/0x138)
<PabloPL>
[ 9.049271] [<803d5c40>] (__device_attach) from [<803d5148>] (bus_probe_device+0x84/0x8c)
<PabloPL>
[ 9.067794] [<803d5148>] (bus_probe_device) from [<803d20e0>] (device_add+0x41c/0x624)
<PabloPL>
[ 9.086180] [<803d20e0>] (device_add) from [<7f1868b4>] (s5p_mfc_alloc_memdev+0x84/0xc8 [s5p_mfc])
<PabloPL>
[ 9.105675] [<7f1868b4>] (s5p_mfc_alloc_memdev [s5p_mfc]) from [<7f186db8>] (s5p_mfc_probe+0x4bc/0x868 [s5p_mfc])
<PabloPL>
If this errors shows, it also shows at the same time in brcmfmac (probed later), but if i disable MFC (when building), this issue is not happening and all is working fine
<mszyprow>
PabloPL: seems to be working fine, at least built-in
<mszyprow>
PabloPL: tested on v4.20 on trats2
<PabloPL>
ok, will try to built-in it. didn't tried to bisect it yet
<mszyprow>
PabloPL: this doesn't exclude the fact that it might fail if build as module
<mszyprow>
PabloPL: let me check this
<PabloPL>
mszyprow: strange that it was working, i was testing mfc (after adding missing reserved memory nodes for s5pv210). didn't tried to bisect it yet
<mszyprow>
PabloPL: reproduced
<mszyprow>
PabloPL: seems to be relatedto no-iommu support
<PabloPL>
mszyprow: ok, i remember that it was working around 4.19.rc4 and broke at 4.19.rc7 (or even i lower version)
<PabloPL>
mszyprow: should i try to bisect it?
<mszyprow>
PabloPL: no need, I will check this in a few minutes
<PabloPL>
mszyprow: don't i need to have/submit extcon driver first? there is one existing for fsa9480 (done by tom3q 4 years ago), working fine ;)
<mszyprow>
extcon support is independent from udc driver
<mszyprow>
having both is probably required for all features though
<mszyprow>
PabloPL: commit a66d972465d15b1d89281258805eb8b47d66bd36 causes this null ptr issue, no idea why yet, but I need to go now. I will investigate it in 2019 once I get back to the office
<PabloPL>
mszyprow: Ok, thanks!
PabloPL has quit [Remote host closed the connection]
mszyprow has quit [Ping timeout: 250 seconds]
PabloPL has joined #linux-exynos
PabloPL has quit [Remote host closed the connection]
PabloPL has joined #linux-exynos
_whitelogger has joined #linux-exynos
PabloPL has quit [Remote host closed the connection]
mszyprow|home has joined #linux-exynos
mszyprow|home has quit [Ping timeout: 246 seconds]
mszyprow|home has joined #linux-exynos
PabloPL has joined #linux-exynos
PabloPL has quit [Remote host closed the connection]
mszyprow|home has quit [Ping timeout: 244 seconds]
aalm has quit [Read error: Connection reset by peer]