3 ms·

Why does this merge also delete all sorts of (seemingly) unrelated files? Like the 'rocket' and 'Cyclades' drivers in this case.

by LeonM 6y ago

kelnos 6y ago
I noticed that too. If you scroll through the entire diff, it looks like most of it is the removal of those seemingly unrelated things. I also see some changes to what looks like common arm64 code, which isn't guarded to check for Apple or non-Apple; it seemed strange to me that those changes would be ok. I wonder if both of those things are just changes from an unrelated merge that are showing up in this diff for no reason.
- jychang 6y ago
  See here https://asahilinux.org/2021/03/progress-report-january-february-2021/ https://asahilinux.org/2021/03/progress-report-january-febru... Mostly the section on IRQ.
  - Ericson2314 6y ago
    Yeah all this is good not bad. This is an extremely high quality port in that general-purpose yak shaves were taking as necessary. This is the exact opposite of your e.g. android phone "fork and abandon" where someone does the absolute minimal and ugliest hacks that upstream would never accept.
azernik 6y ago
Lots of the ARM-related changes are cleanups/modernization of code that the Apple support uses, e.g. https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/commit/?h=for-next&id=b10eb2d50911f98a8f1cacf00b1b677339593f4c https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/... Cyclades removal was about dead, unmaintained code: https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/commit/Documentation/driver-api/serial/cyclades_z.rst?h=for-next&id=f76edd8f7ce06cdff2fe5b6b39a49644c684a161 https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/... This is a 60-commit branch, which had to decouple a lot of different bits of ARM-platform code. Specifically, they use a totally different boot method, have a weird and special power-management system, and other stuff I don't know. See this article: https://asahilinux.org/2021/03/progress-report-january-february-2021/ https://asahilinux.org/2021/03/progress-report-january-febru...
aulin 6y ago
most probably the m1 stuff introduced some conflict in some dead and unmaintained driver and it was better to ditch them than work on a fix
- duskwuff 6y ago
  Actually, it looks as those changes were unrelated! They were merged in from the tty/tty-next tree, which included some changes required to get the M1 UART working. Here are the original commits: https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/commit/?h=for-next&id=f76edd8f7ce06cdff2fe5b6b39a49644c684a161 https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/... https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/commit/?h=for-next&id=3b00b6af7a5bd7fd7e5189ccaad0e0cfb7dc7785 https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/...
arnd 6y ago
Some of the Apple M1 changes went into the tty/serial driver tree that GregKH maintains. The actual platform code requires those changes, so we based the branch on top of that, and the top-level merge commit contains all of these. https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/commit/?h=for-next&id=1bb2fd3880d479a7 https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/... is the better commit to look at for the stuff that I just merged that was not already part of linux-next.

Why does this merge also delete all sorts of (seemingly) unrelated files? Like the 'rocket' and 'Cyclades' drivers in this case.

by LeonM 6y ago

Why does this merge also delete all sorts of (seemingly) unrelated files? Like the 'rocket' and 'Cyclades' drivers in this case.

kelnos 6y ago

I noticed that too. If you scroll through the entire diff, it looks like most of it is the removal of those seemingly unrelated things. I also see some changes to what looks like common arm64 code, which isn't guarded to check for Apple or non-Apple; it seemed strange to me that those changes would be ok. I wonder if both of those things are just changes from an unrelated merge that are showing up in this diff for no reason.

jychang 6y ago
See here https://asahilinux.org/2021/03/progress-report-january-february-2021/ https://asahilinux.org/2021/03/progress-report-january-febru... Mostly the section on IRQ.
- Ericson2314 6y ago
  Yeah all this is good not bad. This is an extremely high quality port in that general-purpose yak shaves were taking as necessary. This is the exact opposite of your e.g. android phone "fork and abandon" where someone does the absolute minimal and ugliest hacks that upstream would never accept.

azernik 6y ago

Lots of the ARM-related changes are cleanups/modernization of code that the Apple support uses, e.g. https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/commit/?h=for-next&id=b10eb2d50911f98a8f1cacf00b1b677339593f4c https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/... Cyclades removal was about dead, unmaintained code: https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/commit/Documentation/driver-api/serial/cyclades_z.rst?h=for-next&id=f76edd8f7ce06cdff2fe5b6b39a49644c684a161 https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/... This is a 60-commit branch, which had to decouple a lot of different bits of ARM-platform code. Specifically, they use a totally different boot method, have a weird and special power-management system, and other stuff I don't know. See this article: https://asahilinux.org/2021/03/progress-report-january-february-2021/ https://asahilinux.org/2021/03/progress-report-january-febru...

aulin 6y ago

most probably the m1 stuff introduced some conflict in some dead and unmaintained driver and it was better to ditch them than work on a fix

duskwuff 6y ago
Actually, it looks as those changes were unrelated! They were merged in from the tty/tty-next tree, which included some changes required to get the M1 UART working. Here are the original commits: https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/commit/?h=for-next&id=f76edd8f7ce06cdff2fe5b6b39a49644c684a161 https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/... https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/commit/?h=for-next&id=3b00b6af7a5bd7fd7e5189ccaad0e0cfb7dc7785 https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/...

arnd 6y ago

Some of the Apple M1 changes went into the tty/serial driver tree that GregKH maintains. The actual platform code requires those changes, so we based the branch on top of that, and the top-level merge commit contains all of these. https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/commit/?h=for-next&id=1bb2fd3880d479a7 https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/... is the better commit to look at for the stuff that I just merged that was not already part of linux-next.