ChanServ changed the topic of #linux-msm to:
marvin24_ has joined #linux-msm
marvin24 has quit [Ping timeout: 480 seconds]
jhovold has joined #linux-msm
pevik_ has joined #linux-msm
jn has quit [Remote host closed the connection]
jn has joined #linux-msm
<z3ntu> It kind of looks like what's written here is supposed to replace it but I haven't seen anything happen there either.. https://lore.kernel.org/linux-arm-msm/SJ0PR02MB7135746D204F13550E9BAE77F8B29@SJ0PR02MB7135.namprd02.prod.outlook.com/
staticfinalpengu has joined #linux-msm
staticfinalpengu has left #linux-msm [#linux-msm]
svarbanov has quit [Ping timeout: 480 seconds]
srinik has quit [Killed (NickServ (Too many failed password attempts.))]
srinik has joined #linux-msm
<minecrell> z3ntu: perhaps with 7/29 they actually meant July 2029
pespin has joined #linux-msm
<arnd> mani_s, bamse: I don't understand the "arm64: dts: qcom: sc8280xp-x13s: Update firmware location" patch. Does that mean one has to now update the dtb in lockstep with the linux-firmware package?
<arnd> In general, the firmware names are generated by the driver specifically to deal with changes like this, and have fallback locations.
<arnd> I suppose the driver already needs a lookup table to match up old "firmware-name" properties with newer filenames when the driver gets updated to support new firmware, so maybe the same place should also check both the old and new location for known firmware files
<mani_s> arnd, bamse knows the history better than me but afaik after some previous discussions, we have settled to extract the firmware names from dtb
<mani_s> and it is not straightforward to use a generic name for firmware files anyway
<arnd> but then the filename can never change
<arnd> you should fix the firmware package then
<mani_s> I'll leave it up to bamse to speak on this ;)
<bamse> arnd: the firmware is not generic for the driver, because it's signed by the device vendor :/
<arnd> I understand it's not generic, but it's still broken to require a dtb update when the firmware package changes, or when the interface between firmware and driver changes
<bamse> arnd: i agree
<bamse> arnd: but the problem here wasn't so much that the firmware package changed, as us not putting in the agreed upon value in the first place
<bamse> arnd: so with the fixes, the name is now stable
<arnd> bamse: ok fair enough. I've now merged it.
<bamse> arnd: thanks
<bamse> arnd: i've not forgotten our previous discussions about interface versioning; it's remains an unsolved problem, but it's a limited problem as long as we deal with stable firmware...we definitely need a solution when that changes
pevik_ has quit [Ping timeout: 480 seconds]
pespin has quit [Remote host closed the connection]
jhovold has quit [Ping timeout: 480 seconds]