Package Details: linux-macbook 5.9.9.arch1-1

Git Clone URL: https://aur.archlinux.org/linux-macbook.git (read-only, click to copy)
Package Base: linux-macbook
Description: The Linux-macbook kernel and modules
Upstream URL: https://git.archlinux.org/linux.git/log/?h=v5.9.9-arch1
Keywords: kernel macbook mbp suspend
Licenses: GPL2
Submitter: m3thodic
Maintainer: greg2010
Last Packager: greg2010
Votes: 46
Popularity: 0.026191
First Submitted: 2016-07-05 23:22 (UTC)
Last Updated: 2020-11-19 06:48 (UTC)

Pinned Comments

Latest Comments

bitjockey commented on 2022-05-02 07:55 (UTC)

Ok, I finally got this to build using the steps here: https://github.com/bitjockey42/archlinux-macbook-air-71#linux-macbook-kernel. Hope that helps someone.

Even with the kernel though I still cannot get suspend to RAM to work so I've just been using hibernate...

trey.merkley commented on 2021-11-21 22:38 (UTC)

@greg2010 can you point me to how I can update this myself? I'm not particularly enthused by the idea of being responsible for a package, but I think I can get this limping enough with my limited kernel knowledge.

trey.merkley commented on 2021-10-03 23:54 (UTC)

If you edit the build files and go to PKGBUILD, you can tweak the URL for git pulling to reference kernel.org's repo by basing it off of this PKGBUILD:

https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=linux-mainline

Keep in mind you'll need to modify the valid GPG keys section so they reference the keys in the linux-mainline PKGBUILD as well, and import the GPG keys.

Moseuglen commented on 2021-09-08 10:11 (UTC)

@linux-lover23

Running into this same problem, crossing fingers it'l get fixed soon. It looks like this kernel might just fix a few problems i've been dealing with. Took a short look at the build-files, but i'm not tech savvy enough to know where to make edits to fix it ¯\(ツ)

linux-lover23 commented on 2021-08-19 17:36 (UTC) (edited on 2021-08-19 17:39 (UTC) by linux-lover23)

After cloning and running makepkg I get the following error: fatal: unable to update url base from redirection: asked for: https://git.archlinux.org/linux.git/info/refs?service=git-upload-pack redirect: https://gitlab.archlinux.org/explore/groups ==> ERROR: Failure while downloading archlinux-linux git repo

pika02 commented on 2021-04-18 00:34 (UTC)

Thanks for your works. However I cannot resume from the swapfile (following the archwiki guide) after hibernate. There is a white underline flashing on the top left of my screen and it freeze. I must delete the resume options from rEFInd. My model is MBP 11,5

TomCatT commented on 2021-02-07 01:04 (UTC)

I get a peer reset error trying to download this. I don't know why it goes so slow.

teacher4711 commented on 2021-02-03 14:10 (UTC)

Thanks so much for this kernel. I had a weird issue and I do not know if you addressed it when modifying this kernel to MacBook needs. My headphones would play garbled sound with an echo. The "options snd-hda-intel model=mba42" for my MacBook Air 4,2 would not fix it. I took the model=mba42 from here: https://www.kernel.org/doc/html/v4.14/sound/hd-audio/models.html#cirrus-logic-cs4206-4207 I also found out that if you separated the audio-channels and set left and right output further apart, the issue would be fixed. The output balance on both, right and left, stayed the same, however you adjusted the controls.

Now, with this kernel (linux-macbook 5.9.9), the issue is resolved, even after a startup. Thanks so much. Michael

greg2010 commented on 2020-12-08 19:22 (UTC)

@ohell happy to hear your issues were resolved. The resume hook, AFAIK, is required if you want to have suspend to disk (hibernation) functionality. Suspend to RAM should work fine without it.

@jso8910: this package is a fork of upstream Arch kernel, with a patch addressing Apple hardware wart that would cause the OS to crash on shutdown. In addition to that, it provides a systemd unit to prevent spurious wakeups.

If you're not experiencing issues with the shutdown or suspend to ram, you probably do not need this package. As to why this package is so large, it builds the kernel from source.

jso8910 commented on 2020-12-08 13:27 (UTC)

Just a quick question: What do I need this package for? I am going to dual boot linux on my macbook pro 14,1, and I am wondering what issues this fixes. Plus, I am wondering why it takes so long to install and takes so much storage. I had a 32 GB Manjaro install a while back and this package took up (at its peak during the build) at least 10-15 GB of storage. I am reluctant to install this based on its large size, so I need some actual benefits to it because none of the comments nor the description is very clear.

ohell commented on 2020-11-27 16:43 (UTC)

5.9.9 update is looking good. Suspend and hibernate working as expected, and the intermittent issues with Bluetooth also look to be resolved. Thanks @greg2010!

When I was trying to track down the hibernation issues, I noticed that Arch wiki says initramfs needs to be configured with resume hook unless it is based on systemd hook. However, both these hooks are missing in my /etc/mkinitcpio.conf:

HOOKS="base udev autodetect modconf block keyboard keymap filesystems"

But hibernate still works. Just curious why this is? Would adding the resume hook and rebuilding intramfs be a good idea?

greg2010 commented on 2020-11-19 07:07 (UTC)

@ohell I do not know what could've possibly caused this. My guess it's an upstream change, but it could also be one of the patches I inherited with this package in conjunction with an upstream change. I'm not experiencing this with 11.2, and do not have access to 11.1, which makes it infeasible to debug.

If you would like to investigate further, here's a helpful article on kernel.org: https://www.kernel.org/doc/html/latest/power/basic-pm-debugging.html Note that you will need to rebuild the kernel with CONFIG_PM_DEBUG flag.

The usb error message usually indicates a malfunctioning USB device, or a hardware issue with the port, although I have seen this happen in the past with no visible consequences.

5.9.9 is now published. As usual, please let me know if there are any issues.

ohell commented on 2020-11-15 14:41 (UTC)

Hello again,

I have an update on suspend failing, journal output said swap not found. So I looked at fstab, and there was no line to mount the swap partition. Add that seems to have fixed the problem (at least I haven't had a reboot in in last 3 lid opens). I am not sure how this could have happened as part of the kernel update. Maybe you have an idea @greg2010?

Apart from this, only other suspicious thing is I have strted these messages in dmesg error log:

usb usb2-port3: Cannot enable. Maybe the USB cable is bad?

No idea if they they are serious or not.

ohell commented on 2020-11-13 10:35 (UTC) (edited on 2020-11-13 10:37 (UTC) by ohell)

Hi @greg2010.

I use pamac and yay. Neither seem to have built aur packages in cache. Anyhow, minor.

I resolved the webcam issue - was due to manually built bcwc-pcie driver for facetimehd (because of kernel incompatibility mentioned here earlier).

But suspend is completely broken - lid open always reboots. My mbp is 11.1

With previous kernel I didn't need to reboot it in weeks :)

Thank you for maintaining this amazing package :)

greg2010 commented on 2020-11-13 08:13 (UTC)

Been busy couple of weeks. I'll see if I can push out 5.9.X some time soon.

@ohell: That depends on what tools you use to interact with AUR.

I'm surprised to hear you're having issues with suspend - what generation of MBP do you have?

ohell commented on 2020-11-09 19:02 (UTC)

so ... I installed 5.83 over the weekend. Doesn't play nice with my MBP - Facetime HD camera is not detected and suspend is broken - machine reboots when opening lid.

Whereas 5.61 was rock solid. So I want to roll back. Question is whether I can find the built binary cached on my machine somewhere, or is rebuilding from source required? Usually I use downgrade to pin packages, but it just errors out with linux-macbook.

Thanks

xristian commented on 2020-09-14 03:32 (UTC)

@greg2010 thanks for the tip! running the code below fixed the public key issue curl -s https://keybase.io/heftig/pgp_keys.asc/\?fingerprint\=a2ff3a36aaa56654109064ab19802f8b0d70fc30 | gpg --with-colons --import-options import-show --import | grep 3B94A80E50A477C7

jso8910 commented on 2020-09-13 15:42 (UTC) (edited on 2020-09-13 15:43 (UTC) by jso8910)

How can I download this? The GPG keys aren't working to validate it. I have managed to download the files with yay, but I can't get any farther than that. I get this error

==> ERROR: One or more PGP signatures could not be verified!
error downloading sources: linux-macbook

dilyn commented on 2020-09-09 07:23 (UTC)

I've been meaning to ask this for a while now; has anybody noticed issues relates to this one? Errors related to applesmc would be reported in dmesg for anybody wishing to check; you might have to change loglevels to see the errors. https://bbs.archlinux.org/viewtopic.php?id=258303

greg2010 commented on 2020-09-09 07:18 (UTC)

@xristian heftig's gpg key has changed. See https://bbs.archlinux.org/viewtopic.php?id=255968 for help.

xristian commented on 2020-09-08 06:35 (UTC)

Errors when trying to install, any suggestions?

==> Verifying source file signatures with gpg...
    archlinux-linux git repo ... FAILED (unknown public key 3B94A80E50A477C7)
==> ERROR: One or more PGP signatures could not be verified!
error downloading sources: linux-macbook

greg2010 commented on 2020-09-04 18:07 (UTC)

@m.aesophor:

Assuming you are having the same issue (the last line in the log before freeze is fb0: switching to inteldrmfb from EFI VGA) this might help: https://bbs.archlinux.org/viewtopic.php?pid=1913977#p1913977

m.aesophor commented on 2020-09-04 11:14 (UTC)

Off-topic question here:

Has the i915 driver freezing issue been resolved? My a1398 still freezes upon boot. (linux-5.8.5)

greg2010 commented on 2020-08-26 04:40 (UTC)

5.8.x is now out. Note that Intel IOMMU is enabled by default (CONFIG_INTEL_IOMMU_DEFAULT_ON=y). This is a requirement for the kernel to boot with some (most?) older intel GPUs. Use kernel parameter if you need it disabled for any reason.

greg2010 commented on 2020-08-02 06:19 (UTC)

Can confirm the package build fails with gcc 10.1.0. There are reports that the video driver issue has been resolved in 5.7.7. I will investigate this some time next week, and if that's indeed the case will finally push 5.7.x. Stay tuned.

RollerGAME12 commented on 2020-08-01 12:16 (UTC)

Is anyone running into build errors? I think it is related to my GCC version being beyond what this was originally built on, based on quick research.


 CC [M]  arch/x86/kvm/../../../virt/kvm/kvm_main.o
arch/x86/kvm/../../../virt/kvm/kvm_main.c: In function ‘__kvm_gfn_to_hva_cache_init’:
arch/x86/kvm/../../../virt/kvm/kvm_main.c:2236:42: error: ‘nr_pages_avail’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
 2236 |  for ( ; start_gfn <= end_gfn; start_gfn += nr_pages_avail) {
      |                                ~~~~~~~~~~^~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
make[2]: *** [scripts/Makefile.build:268: arch/x86/kvm/../../../virt/kvm/kvm_main.o] Error 1
make[1]: *** [scripts/Makefile.build:505: arch/x86/kvm] Error 2
make: *** [Makefile:1683: arch/x86] Error 2
==> ERROR: A failure occurred in build().
    Aborting...
error making: %!s(func() string=0x55ab7dcd7bb0)

greg2010 commented on 2020-06-23 10:47 (UTC)

There is a bug in 5.7.x (5.7.4) i915 driver that causes freeze on boot with some intel integrated gpus (a1398 is affected). Not updating this package until the upstream issue is resolved.

TomCatT commented on 2020-06-08 18:12 (UTC)

Maybe my connection is bad but when I try to download the doc, the update fails because the conntion is too slow. Is there another way to download all the files?

ohell commented on 2020-04-14 21:24 (UTC) (edited on 2020-04-15 12:03 (UTC) by ohell)

another bug report: 5.6 kernel introduced incompatibility with bcwc_pcie package, so no webcam available, at least until the driver maintainer issues a new version.

In the meanwhile it is possible to update the driver manually after changing code, following the instructions at https://aur.archlinux.org/packages/bcwc-pcie-git/#comment-737146.

ohell commented on 2020-04-10 14:43 (UTC)

Found the problem! hid-apple-patched-git-dkms update to support Apple Magic keyboards does some funky auto detect of kb layout based on country code. Probably not your problem :)

ohell commented on 2020-04-10 11:12 (UTC)

I don't use any remapping. The defaults used to map the modifier keys as described in my previous post. But since the update, I've had to remap keys to get the old behaviour back.

My MBP model is Retina, 13-inch, Early 2015.

I guess if you are not aware of any changes to the keyboard driver then I need to go through the latest Manjaro update and see if any keyboard related components got updated.

greg2010 commented on 2020-04-10 04:26 (UTC)

@ohell Unfortunately I cannot reproduce - the stock behaviour (without the remaps) hasn't changed, at least for me. What software are you using to remap the keys?

ohell commented on 2020-04-09 23:09 (UTC) (edited on 2020-04-09 23:11 (UTC) by ohell)

@greg2010 The last update has changed the keyboard layout for me - previously I has the apple cmd key mapped as ctrl, left alt was win and ctrl was super.

But now the mapping of these keys is jumbled up: ctrl on the keypad seems to have become function, fn is win, cmd is alt.

greg2010 commented on 2020-03-24 01:09 (UTC)

Thank you for maintaining this package, @annguyenfoss. I'll try to keep this package up-to-date, time permitting. I've done a bit of testing and didn't encounter any issues with my macbook, but since it's the first release of this package with 5.5.x kernel, there might be broken drivers/etc. Please do post here if that's the case.

ohell commented on 2020-03-03 16:25 (UTC)

Thanks a lot for keeping it going while you could, @annguyenfoss!

I am tempted to take your place, but I have no experience in maintaining packages! I guess we can wait a few weeks for someone more qualified to step up, else I will try :).

This is too useful a package to be left behind, IMO.

annguyenfoss commented on 2020-03-01 16:05 (UTC)

My apologies. I can't maintain this package anymore.

tag_enk commented on 2020-01-24 03:29 (UTC)

@ohell @albalajba can confirm this happens on my Macbook 11,1 as well. It's not just you.

annguyenfoss commented on 2019-12-03 07:24 (UTC)

@bionade24 I disagree. You should be able to install mainstream kernel along linux-macbook. There isn't any conflict between the two.

bionade24 commented on 2019-12-01 09:30 (UTC) (edited on 2019-12-01 09:30 (UTC) by bionade24)

You should have his in the PKGBUILD:

provides = ("linux")

ohell commented on 2019-11-07 22:51 (UTC)

Thanks, @albalajba. I will try to narrow this down myself - perhaps something to do with pstate.

albalajba commented on 2019-11-07 16:10 (UTC)

@ohell I just tried replicating your problem on my MacBook Air and didn't get the same error.

ohell commented on 2019-11-07 15:45 (UTC)

Guys, please can you check if you can reproduce one issue that I notice on my MBP?

After closing the lid everything gets suspended fine. However, if I plug in the power cable to the sleeping laptop, it wakes up from sleep.

Thanks!

albalajba commented on 2019-11-07 15:42 (UTC)

@dilyn Thanks for help, installing broadcom-wl-dkms package fixed all the problems with interface, also I did enable wakeup service and now everything works as intented, once again thanks for help.

dilyn commented on 2019-11-07 15:30 (UTC) (edited on 2019-11-07 15:30 (UTC) by dilyn)

@albalajba you'll need the broadcom-wl-dkms package for exactly the reason you specified: you aren't using the regular kernel. Make sure you install the headers as well. As for sleep mode, make sure you've enabled the wakeup service. I don't know if all of your interface troubles are because of you not using the dkms package but it's a good start.

albalajba commented on 2019-11-07 15:25 (UTC) (edited on 2019-11-07 15:30 (UTC) by albalajba)

I switched to this kernel on my Arch installed on Macbook Air and weird things happen, I seem to have no network interfaces (both wired and wireless) and also computer won't enter sleep mode after closing the lid but instead wake up 2 times while lid is closed (I can tell by apple logo being lit while lid is closed). Do you know how to remedy this? Also normal linux kernel is a dependency of broadcom-wl package (network driver for macbooks) so I can't even get rid of normal kernel.

Edit - Of course I was gettting weird lid closing behaviour - I didn't enable macbook-wakeup.service... This is fixed, still the problem remains with network.

annguyenfoss commented on 2019-11-02 16:37 (UTC) (edited on 2019-11-02 16:47 (UTC) by annguyenfoss)

@darthnoward: linux-macbook doesn't have /usr/lib/firmware/brcm/brcmfmac43602-pcie.bin at all. You may have misunderstood something there.

ohell commented on 2019-10-29 10:49 (UTC)

@t3harvinator I tried again after reading your comment, and all goes well so far, after a week of running this kernel. No problems with wakeup at all. And noticeably faster start from off, sleep and suspend.

Thank you @annguyenfoss :)

darthnoward commented on 2019-10-29 09:36 (UTC)

I get:

failed to commit transaction (conflicting files) linux-firmware: /usr/lib/firmware/brcm/brcmfmac43602-pcie.bin exists in filesystem

how to solve this?

t3harvinator commented on 2019-10-14 02:54 (UTC)

i havent had any problems on a mbp late 2011.

thanks @annguyenfoss for taking this project over btw.

dilyn commented on 2019-09-16 15:34 (UTC)

@ohell @annguyenfoss I haven't had any issues with the macbook-wakeup service for the last few releases.

annguyenfoss commented on 2019-09-10 07:35 (UTC)

@ohell: Sorry for late response. I have been debugging that issue for months here also but so far I'm pretty clueless...

ohell commented on 2019-08-15 15:32 (UTC) (edited on 2019-08-15 15:39 (UTC) by ohell)

Thanks for the quick fix! I just did a clean build and install. However, it causes my MBP to not wake up after lid close/open at all, even after pressing Space/power etc. I have to hard boot the machine by holding down power for 10 seconds. I did enable the macbook-wakeup service as per the message.

Previously the behaviour was better - had to press space after lid open to wake up. There was no wifi after wake-up, but I had a hacky service running to restart adapter & wifi.

$ uname -a

Linux bert 5.2.8-arch1-2-macbook #1 SMP PREEMPT Thu Aug 15 12:28:37 BST 2019 x86_64 GNU/Linux

annguyenfoss commented on 2019-08-15 06:40 (UTC)

@dilyn: Right! Thanks a lot! I kept committing the wrong file due to a small change recently in my local workspace. And the difference are just one character.

Fixing now.

@all: I'm very sorry for all the issues you have experienced recently. This will not happen again.

dilyn commented on 2019-08-14 15:11 (UTC) (edited on 2019-08-14 15:11 (UTC) by dilyn)

Also had an issue building the last two updates - the checksum error given by @ohell, as well as an issue with the source url. It seems that the srcver in the PKGBUILD was erroneously(?) changed from 5.x.x-arch1 to 5.x.x.arch1, which is not a valid branch.

annguyenfoss commented on 2019-08-12 10:08 (UTC)

@ohell: Thanks. I will check and comeback within this week.

ohell commented on 2019-08-10 09:07 (UTC)

I am getting an error in package integrity (downloaded multiple times). Thanks

Resolving deltas: 100% (5735280/5735280), done. -> Found config -> Found 60-linux.hook -> Found 90-linux.hook -> Found linux.preset -> Found macbook-wakeup.service -> Found RFC-PCI-Workaround-to-enable-poweroff-on-Mac-Pro-11.patch ==> Validating source files with sha256sums... archlinux-linux ... Skipped config ... FAILED 60-linux.hook ... Passed 90-linux.hook ... Passed linux.preset ... Passed macbook-wakeup.service ... Passed RFC-PCI-Workaround-to-enable-poweroff-on-Mac-Pro-11.patch ... Passed ==> ERROR: One or more files did not pass the validity check! panic: runtime error: makeslice: cap out of range

System: Linux maclin 4.19.62-1-MANJARO #1 SMP PREEMPT Sun Jul 28 09:14:58 UTC 2019 x86_64 GNU/Linux

annguyenfoss commented on 2019-07-16 05:00 (UTC)

Built and tested 5.2.0-arch2-1. Released.

annguyenfoss commented on 2019-07-06 17:47 (UTC)

Built and tested 5.1.15.arch1-1. Released. Updated config to match stock kernel. Fixed issue with post install message

annguyenfoss commented on 2019-06-24 11:29 (UTC)

@hcl31415: Ah nevermind, it was a mistake of my part during the build. I removed that comment of mine immediately before releasing 5.1.5.arch1-1 build.

@jketcham: As above.

@dilyn: I'm setting up a new build server with Jenkins. Stay tuned.

dilyn commented on 2019-06-19 22:01 (UTC)

Been a longtime user of this kernel and noticed it recently changed hands; wondering if there's anything I can do to help the process of keeping this up to date?

jketcham commented on 2019-06-16 08:21 (UTC)

I'm seeing the same error and typo as @hcl31415, that patch would be appreciated. I'm new to contributing to AUR packages; I see the read-only git url, how would one submit the patch themselves?

And in case anyone is curious, this is the message that is trying to be printed:

    To get suspend working properly you must do the following:

    Enable the macbook-wakeup service for suspend/resume on LID events:
    systemctl enable --now macbook-wakeup.service

    Optionally, include the kernel boot option processor.ignore_ppc=1
    to allow frequency changing on battery power.

hcl31415 commented on 2019-05-29 09:20 (UTC) (edited on 2019-05-29 09:30 (UTC) by hcl31415)

@anngyenfoss: As far as I know since considerable time, when installing the package an error message occurs because of a spelling error in a function name. Could you maybe include the following diff:

diff --git a/linux.install b/linux.install index ab733fb..dc80fa2 100644 --- a/linux.install +++ b/linux.install @@ -1,4 +1,4 @@ -display_display_message () { +display_message () {

Sorry for the styling... I am new to this.

annguyenfoss commented on 2019-05-29 03:33 (UTC)

Built and tested 5.1.5.arch1-1. Released. However, there are differences between this build's config and official kernel build's. I'll review and address those differences in next builds.

saulocastelo commented on 2019-05-20 02:34 (UTC) (edited on 2019-05-20 02:34 (UTC) by saulocastelo)

Hi ...

I can run the tests on my MacbookPro retina 11,1 for us before you launch

If one owner of each Macbook(Pro) version do this will you be able to continue this amazing project?

We could create an repo exclusive for the tests and I'm able to maintain this for some time on my server (Google Cloud)

Big thank you for all your work

annguyenfoss commented on 2019-05-10 04:44 (UTC)

Built and tested 5.0.13.arch1-1. Released.

m3thodic commented on 2019-05-07 19:36 (UTC)

@annguyenfoss Thank you for the thank you's, greatly appreciated!

Most of the patches came from mailing lists -- I wanna say they were titled something like Macbook RFC that had 3 different versions I had to port to newer kernels but nothing super heavy handed. It mostly had to do with using the same code just having to massage it a bit for when newer releases added things that screwed up the line number(s) relating to the patches.

Sorry for orphaning this, but at the end of the day I don't have any way of doing any sort of quality control/assurance since the place I work for has moved from Macbook's to Dell XPS hardware. I'm still subscribed to receiving alerts to this package so I'll continue to keep an eye out in case anyone gets stuck, but of course ultimately I'd like to see the mainline kernel codebase Just Work(tm) with Apple hardware.

annguyenfoss commented on 2019-05-06 06:29 (UTC)

@m3thodic Thanks a lot! I've built every single version of this kernel in the last 3+ years. It has been really helpful for me all these years.

Now, I'm a python dev and my C/C++/makefile writing are quite rustee. Is it very difficult from a programming perspective to maintain this build? Did you have to write the patch yourself and fix bug/improve the patch's codes all these years while maintaining this package?

I'm trying to figure out what would be required if I'm to adopt this package.

Thanks!

m3thodic commented on 2019-05-05 22:36 (UTC)

@annguyenfoss Yeah I orphaned it, I haven't had a macbook in almost 2 years now I've just been maintaining it without really being able to test new releases.

annguyenfoss commented on 2019-05-05 13:15 (UTC)

Anyone aware how this package is now orphaned? The maintainer had no time for this anymore or he no longer uses macbook to provide maintenance?

portabello commented on 2019-04-28 21:31 (UTC)

I have installed the latest version 5.0.9 from the binary repo but neither the lighting controls or the suspend work. I have an early 2015 macbook pro with 13" retina display. My computer won't resume from suspend mode and I am forced to hold the power button down and restart. This happens regardless of whether macbook-wakeup.service is enabled. Additionally closing the lid only enters suspend mode when I am using a wm. When in tty mode I need to run syatemctl suspend. Brightness controls don't work either. The behavior seems to be identical to using the normal arch kernel.

architekt commented on 2019-01-05 04:54 (UTC)

I can't seem to find any information which devices are supported with this? I'm running a Macbook Pro 10,x (15" Retina Early 2013), would this be compatible?

FiniteParadox commented on 2018-12-28 12:03 (UTC) (edited on 2018-12-28 12:07 (UTC) by FiniteParadox)

Can provides=("linux=${pkgver}") and provides=("linux-headers=${pkgver}") fields be added in PKGBUILD (sections _package() and _package-headers() respectively) so that packages that depend on linux are satisfied by linux-macbook?

kflak commented on 2018-11-30 19:09 (UTC)

Thanks for this one! Unfortunately I am still not getting good enough audio performance from this, and have to use the rt-lts kernel, as this is the only kernel I have found that 1) gives rock solid audio performance and 2) wakes up after lid close-induced suspend... However, the package seems not to be maintained much, so I'd be very interested in a macbook-centric version of the realtime patch. Any hope for such a thing? Alternatively, any pointers as to how I could compile such a beast myself?

LawnGnome commented on 2018-11-27 21:11 (UTC)

@m3thodic Thanks! The new package installs fine for me. I didn't get it via pacman -Syu, though: it seems like the repo metadata still points to -1.

m3thodic commented on 2018-11-27 01:15 (UTC)

@LawnGnome Should be all set, thanks again!

m3thodic commented on 2018-11-27 00:42 (UTC)

@LawnGnome Hmm, yeah was afraid I had overlooked something -- I was coming from the 4.18.16 package to 4.19.4 and ran into a little snag. Thanks for the heads-up I'll build a new package straight from the AUR source to be safe and I'll push it up shortly.

LawnGnome commented on 2018-11-26 20:35 (UTC) (edited on 2018-11-26 20:35 (UTC) by LawnGnome)

I'm getting a file conflict with the 4.19.4.arch1-1 binary package:

error: failed to commit transaction (conflicting files)
linux-macbook: /usr/lib/systemd/system exists in filesystem
Errors occurred, no packages were upgraded.

Building from source did not result in the same conflict, interestingly.

espenaf commented on 2018-10-23 17:02 (UTC)

Works now, tnx.

m3thodic commented on 2018-10-17 20:47 (UTC)

@espenaf I just updated to 4.18.14 -- can you please try now? Thanks!

espenaf commented on 2018-10-15 11:01 (UTC) (edited on 2018-10-15 11:01 (UTC) by espenaf)

Seems like libpcap.net does not have the latest version, I am getting

error: failed retrieving file 'linux-macbook-4.18.13.arch1-1-x86_64.pkg.tar.xz' from libpcap.net : The requested URL returned error: 404

zsd commented on 2018-08-13 12:08 (UTC)

About having to hit the power button when opening the lid after S2R:

I didn't have to do this with 4.10.14, 4.11.8, 4.11.9, 4.14.4, 4.15.15, but starting with 4.17.11 (haven't tried 4.17.13 yet) I also have to hit the power button.

This is with a MacBookPro11,4.

deimosfr commented on 2018-06-11 15:21 (UTC)

I confirm @rjl6789 comment. The issue disappeared with 4.16.X version and is now back on 4.17.0-1 :(

rjl6789 commented on 2018-06-11 11:19 (UTC) (edited on 2018-06-11 11:23 (UTC) by rjl6789)

As with @deimosfr I'm having the issue where I need to push the power button to wake it up after a lid close. Macbook 11.2, linux macbook=4.17.0-1-macbook. Enabled + Disabled macbook-wakeup service and also implemented service modules as per mac arch wiki, power management, referenced forum post ( https://wiki.archlinux.org/index.php/mac ). This issue doesn't present itself with linux or linux-lts

dmidecode gist here: https://gist.github.com/rjl6789/02e8e6e992c421aeb5ec6b5ac60a0a94

Thanks!

aguitel commented on 2018-05-12 22:30 (UTC) (edited on 2018-05-12 23:03 (UTC) by aguitel)

is this kernel safe to install in Macbook Air model 2017 ( come with sierra os) ? what this Mac gain from the "normal" kernel

m3thodic commented on 2018-03-24 18:49 (UTC)

Hey everyone, sorry about the lack of updates as of recent. The stable release schedule got super aggressive within the past six months, so it's been a challenge to keep on top it; on a related note some of the diffs weren't cleanly applying and the last thing I want to do is push out a package that uses a fudge factor for patching kernel code.

The latest kernel version apparently finally brought the gmux/backlight code into the official tree. I tested it on my MacBook Air and was able to dial down my brightness using the native GNOME hotkeys. All that's left is the suspend/poweroff patches, so I reverted my last attempt at consolidating all the patches since the latest changes I made appeared to cause issues for some people.

Post here if something isn't working, but equally as important please post if everything is working great -- the more feedback you can give the better the chances the rest of these patches can finally be merged into mainline.

Thanks again for your patience!

rockon999 commented on 2018-03-13 00:04 (UTC)

@m3thodic I am unable to download the linux-macbook-headers package from your repository (it attempts but fails because no data is received in time). The linux-macbook package, however, is fine. Is the repo not supposed to supply the headers package?

crencom commented on 2018-03-11 18:45 (UTC)

Figured it out for those with the same question / problem. I can get the GPU switch to work from Intel back to the AMD GPU by reversing the Command that shuts off the AMD GPU while on Intel video-

from - echo OFF > /sys/kernel/debug/vgaswitcheroo/switch (on Intel video)

to - echo ON > /sys/kernel/debug/vgaswitcheroo/switch

then run your gpu-switch -d command

reboot.

This has worked fro me so far.

crencom commented on 2018-03-08 15:48 (UTC)

Running this kernel on a MBPr 11,5 - AMD GPU AND INTEL video. I am having issues swithing from the Intel video back to the AMD GPU. I use the gpu-switch -d command just prior to reboot. My refind.conf is set up correct as well. When I reboot, I get a black screen and I have to then login to the terminal to delete the 20-intel.conf file to get it to boot back with the AMD GPU. Any help is appreciated.

thanks

deimosfr commented on 2018-02-28 08:28 (UTC)

Hi,

Here is the gist: https://gist.github.com/deimosfr/f277253b2310474af20e66ccdac25b73

It's a MacBookPro11,5.

Thanks

m3thodic commented on 2018-02-25 10:43 (UTC)

@deimosfr what version of Macbook do you have? It would be helpful if you could run dmidecode with sudo or as root and post the output to a gist.

Thanks.

deimosfr commented on 2018-02-24 23:18 (UTC)

Hi, with 4.15.4-1, when the lid I need to push the power button to wake it up. I didn't have this behavior in previous version, I mean it automatically waked up when I opened the lid. Is there any fix for it ? By the way, I enabled macbook-wakeup.service. Thanks

Ram-Z commented on 2018-01-08 08:40 (UTC) (edited on 2018-01-08 13:27 (UTC) by Ram-Z)

Here's a patch that syncs with upstream and updates to 4.14.12.

https://gist.github.com/Ram-Z/99fd6e1dfd5c2b65f817d7b41c46559d

zsd commented on 2017-12-19 21:35 (UTC)

My bad... for anyone else reading this who doesn't know, the PKGBUILD file shows which of the patches were used, and in what order.

zsd commented on 2017-12-10 03:36 (UTC)

There are 3 patches dealing with suspend and poweroff. Can anyone clarify which ones were actually applied to the kernel source to do The Right Thing?

Thanks.

Rob070511 commented on 2017-11-28 19:59 (UTC)

@m3thodic Ever since the update past 4.13.7-1 my virtualbox module will not load, this issue does not happen on the main linux kernel. Any advice?

bcc commented on 2017-10-21 12:41 (UTC)

I keep getting: Oct 21 13:39:43 hotty kernel: bcm5974: bad trackpad package, length: 8 Oct 21 13:39:43 hotty kernel: bcm5974: bad trackpad package, length: 8 Oct 21 13:39:43 hotty kernel: bcm5974: bad trackpad package, length: 8 I fix it by rmmod/modprobe. There is a script/bug report here https://bugzilla.redhat.com/show_bug.cgi?id=465677#c2

NitroOxid commented on 2017-10-17 12:49 (UTC) (edited on 2017-10-17 12:50 (UTC) by NitroOxid)

This good kernel! Please, add broadcom wl support(i have Broadcom BCM4360 in MacBook Air 13(6.2, 2014 year)

valentjedi commented on 2017-10-15 07:37 (UTC)

With stock kernel v4.13.5 backlight now works out of the box for me (mpb pro mid-2015). With the help of gmux_backlight of course.

JeremyTiki commented on 2017-10-10 21:19 (UTC) (edited on 2017-10-11 03:02 (UTC) by JeremyTiki)

@m3thodic Freeze happened again today, tried to SSH in to no avail. I also thought it might be CPU/memory related, but both were under 10% when the freeze happened. I don't know whether this will help or not but I've also put up my dmesg on pastebin: https://pastebin.com/01PC4A7N

JeremyTiki commented on 2017-10-07 22:17 (UTC)

@m3thodic Absolutely. I've been trying to get the freeze to happen today, but can't seem to reproduce it now. I did an -Syu last night maybe one of the updated packages fixed the issue? Anyway if I can get the issue to happen again today I'll try SSHing into it.

m3thodic commented on 2017-10-07 19:16 (UTC)

@Jack12816 Good looking out, I overlooked it porting the package to the 4.13 release. I'm gonna have a 4.13.5 build out in about an hour, probably less. @JeremyTiki So many variables at play here, I wouldn't know where to start. Can you confirm it's a hard freeze by trying to switch to a virtual terminal or logging in over SSH?

Jack12816 commented on 2017-10-07 17:42 (UTC)

Looks like the systemd macbook-wakeup.service file is not packaged anymore? Sounds like a bug to me. Can someone confirm this? (tested on latest build 4.13.4-1)

JeremyTiki commented on 2017-10-07 00:19 (UTC)

Getting a weird freezing issue when playing Youtube through Firefox on the latest update. I'm using a MBP 11,5 any advice?

valentjedi commented on 2017-08-10 17:06 (UTC)

Well, backlight will work with stock kernel if I use apple_set_os.efi by 0xbb (https://github.com/0xbb/apple_set_os.efi). But I have to run it on every reboot. I don't reboot that often but it's quite tedious.

valentjedi commented on 2017-08-10 15:51 (UTC)

For me stock kernel works for almost anything except backlight. Backlight works only for radeon_b10/brightness and only in a off/on way. So 0 means screen is off and any other number means max out backlight. With this patched kernel all works great. I have mid-2016 macbook pro (idk if it's 11,4 or 11,5 but it has discrette Radeon graphics). So I guess it's a bit early for deprecation :)

m3thodic commented on 2017-08-07 19:05 (UTC)

@enrique Thanks for the feedback! Can anyone else confirm on the functionality of 4.12.4? If everything looks fixed I'll deprecate this package. Thanks!

enrique commented on 2017-08-05 09:08 (UTC)

Yes it looks like 4.12.4-1-ARCH works on my MacBookPro11,4 @m3thodic brigthness keys for keyboard and screen works for me

m3thodic commented on 2017-07-31 19:24 (UTC)

@jbg What about the brightness keys, do they work out of the box as well?

jbg commented on 2017-07-31 09:10 (UTC)

FYI, as of the 4.12.4-1 release of the standard "linux" package (normal Arch kernel) my Macbook powers off, reboots and suspends without any issue without needing this patched kernel any more.

m3thodic commented on 2017-07-13 19:05 (UTC)

@rossome Please don't flag this package out-of-date unless it's older than what is being shipped with Arch Linux. I understand kernel.org says 4.12.1 is out, but this package is updated only when Arch Linux updates their kernel package (especially for major version bumps). Thanks.

zsd commented on 2017-07-09 18:44 (UTC)

Question for anyone using this package with a MacBookPro 11,4: if you have a thunderbird-ethernet dongle plugged in when you suspend, does your computer resume properly? Mine wakes up enough to show me what was on the screen when I suspended, but it seems to be otherwise completely dead... no reaction to the mouse or the keyboard. (For that matter, if anyone with any other type of Mac wants to comment, it might be useful information.) Thanks.

zsd commented on 2017-07-09 18:40 (UTC)

supertuba: rather than installing the packages, I built the kernel by hand I noticed that the intel-pstate-backport.patch and fix_race_condition_in_packet_set_ring.diff patches are in the mainline kernels for some time now. Could that be part of your problem? Maintainers: given that you are targeting this package to 4.11.9, should those patches be sent to the history department?

supertuba commented on 2017-07-09 02:11 (UTC)

Is anyone having trouble installing the packages? I'm currently getting the message that linux-macbook package has failed to install, please check .SRCINFO I'm new to arch and find little guidance when it comes to troubleshooting macbook specific issues.

petris commented on 2017-07-08 14:24 (UTC)

It appears that you started signing these packages yesterday. Much appreciated!

ephemient commented on 2017-07-03 15:03 (UTC)

There is an updated suspend/poweroff patch proposed to upstream. https://patchwork.kernel.org/patch/9821775/ Also, the .config in this package is missing a few changes in the linux package. https://bugs.archlinux.org/task/54197 https://bugs.archlinux.org/task/54268 https://bugs.archlinux.org/task/53281

maze commented on 2017-06-15 07:46 (UTC)

Thanks a lot for these excellent packages. Would you mind signing the binary builds please?

m3thodic commented on 2017-06-15 02:06 (UTC)

@Tom_B Thanks for the feedback and tips!

Tom_B commented on 2017-06-14 14:56 (UTC)

For what it's worth, I struggled with Grub like others. I switched to systemd-boot and it's a lot easier. Add the relevant macbook.conf ``` title Arch Linux (Macbook kernel) linux /vmlinuz-linux-macbook initrd /initramfs-linux-macbook.img options root=PARTUUID=d2722f04-6165-489e-aba6-edeeb8e38634 rw quiet splash video.use_native_backlight=1 processor.ignore_ppc=1 ``` (change your uuid to your / partition) and upgrading the kernel via yaourt just works without any extra hassle.

aranel commented on 2017-05-31 05:21 (UTC)

Thanks for this kernel, i'm very new to arch and such things make me like it even more. I have couple of issues though, here's the output of them. May 31 08:11:44 aranel kernel: ACPI Error: Needed type [Reference], found [Integer] ffff880264868f78 (20170119/exresop-103) May 31 08:11:44 aranel kernel: ACPI Exception: AE_AML_OPERAND_TYPE, While resolving operands for [OpcodeName unavailable] (20170119/dswexec-461) May 31 08:11:44 aranel kernel: ACPI Error: Method parse/execution failed [\_PR.CPU0._PDC] (Node ffff8802660ee118), AE_AML_OPERAND_TYPE (20170119/psp May 31 08:11:45 aranel kernel: brcmfmac: brcmf_c_preinit_dcmds: Firmware version = wl0: Nov 10 2015 06:38:10 version 7.35.177.61 (r598657) FWID 01-e May 31 08:11:45 aranel kernel: brcmfmac: brcmf_cfg80211_reg_notifier: not a ISO3166 code (0x30 0x30) May 31 08:11:45 aranel systemd-vconsole-setup[593]: Virtual console /dev/tty0 is not in K_XLATE or K_UNICODE. May 31 08:11:46 aranel kernel: brcmfmac: brcmf_p2p_create_p2pdev: set p2p_disc error May 31 08:11:46 aranel kernel: brcmfmac: brcmf_cfg80211_add_iface: add iface p2p-dev-wlp3s0 type 10 failed: err=-16 May 31 08:11:52 aranel kernel: brcmfmac: brcmf_inetaddr_changed: fail to get arp ip table err:-23 And wifi stopped working after i updated kernel. Do you guys have any idea what could be a problem?

notjacktall commented on 2017-05-31 03:45 (UTC)

update: it was because i was booting from a boot.efi file and the boot.efi doesn't regenerate every time you run grub-mkconfig. So to fix it all i had to do was regenerate the boot.efi file.

notjacktall commented on 2017-05-30 13:43 (UTC) (edited on 2017-05-30 13:46 (UTC) by notjacktall)

@m3thodic Yeah thats what I was thinking and I do dual boot. Its just arch and mac linux. I am using efi. Just looked into refind that does look like it will solve my problems. I have never used refind though, is there any more additional set up other than installing it using the arch package manager?

m3thodic commented on 2017-05-30 04:40 (UTC)

@notjacktall I'm guessing it has something to do with your partition setup re: EFI. Do you dual-boot at all? I personally install refind on all my EFI-based systems (refind-efi-bin in the AUR). For me it's the quickest and easiest solution that supports every type of EFI setup (or lack there-of) I've come across.

notjacktall commented on 2017-05-29 22:45 (UTC)

I did that and the two are identical. I've pasted the grub.cgf below if you want to take a look at it. But the good news is the kernel dix fix the suspend for me and made the laptop a lot more usable. So, thanks a ton for that :) [code] # # DO NOT EDIT THIS FILE # # It is automatically generated by grub-mkconfig using templates # from /etc/grub.d and settings from /etc/default/grub # ### BEGIN /etc/grub.d/00_header ### insmod part_gpt insmod part_msdos if [ -s $prefix/grubenv ]; then load_env fi if [ "${next_entry}" ] ; then set default="${next_entry}" set next_entry= save_env next_entry set boot_once=true else set default="0" fi if [ x"${feature_menuentry_id}" = xy ]; then menuentry_id_option="--id" else menuentry_id_option="" fi export menuentry_id_option if [ "${prev_saved_entry}" ]; then set saved_entry="${prev_saved_entry}" save_env saved_entry set prev_saved_entry= save_env prev_saved_entry set boot_once=true fi function savedefault { if [ -z "${boot_once}" ]; then saved_entry="${chosen}" save_env saved_entry fi } function load_video { if [ x$feature_all_video_module = xy ]; then insmod all_video else insmod efi_gop insmod efi_uga insmod ieee1275_fb insmod vbe insmod vga insmod video_bochs insmod video_cirrus fi } set menu_color_normal=light-blue/black set menu_color_highlight=light-cyan/blue if [ x$feature_default_font_path = xy ] ; then font=unicode else insmod part_gpt insmod ext2 set root='hd0,gpt6' if [ x$feature_platform_search_hint = xy ]; then search --no-floppy --fs-uuid --set=root --hint-bios=hd0,gpt6 --hint-efi=hd0,gpt6 --hint-baremetal=ahci0,gpt6 c113c01e-6b64-40fe-a8e9-1d43524fb01f else search --no-floppy --fs-uuid --set=root c113c01e-6b64-40fe-a8e9-1d43524fb01f fi font="/usr/share/grub/unicode.pf2" fi if loadfont $font ; then set gfxmode=auto load_video insmod gfxterm set locale_dir=$prefix/locale set lang=en_US insmod gettext fi terminal_input console terminal_output gfxterm if [ x$feature_timeout_style = xy ] ; then set timeout_style=menu set timeout=5 # Fallback normal timeout code in case the timeout_style feature is # unavailable. else set timeout=5 fi ### END /etc/grub.d/00_header ### ### BEGIN /etc/grub.d/10_linux ### menuentry 'Arch Linux' --class arch --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-simple-c113c01e-6b64-40fe-a8e9-1d43524fb01f' { load_video set gfxpayload=keep insmod gzio insmod part_gpt insmod ext2 set root='hd0,gpt5' if [ x$feature_platform_search_hint = xy ]; then search --no-floppy --fs-uuid --set=root --hint-bios=hd0,gpt5 --hint-efi=hd0,gpt5 --hint-baremetal=ahci0,gpt5 76ed38c7-2c83-42c9-ad99-2e78614710df else search --no-floppy --fs-uuid --set=root 76ed38c7-2c83-42c9-ad99-2e78614710df fi echo 'Loading Linux linux-macbook ...' linux /vmlinuz-linux-macbook root=UUID=c113c01e-6b64-40fe-a8e9-1d43524fb01f rw BOOT_IMAGE=/vmlinuz-linux-macbook disable_mode=1 quiet splash rootflags=data=writeback ipv6.disable=1 echo 'Loading initial ramdisk ...' initrd /initramfs-linux-macbook.img } submenu 'Advanced options for Arch Linux' $menuentry_id_option 'gnulinux-advanced-c113c01e-6b64-40fe-a8e9-1d43524fb01f' { menuentry 'Arch Linux, with Linux linux-macbook' --class arch --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-linux-macbook-advanced-c113c01e-6b64-40fe-a8e9-1d43524fb01f' { load_video set gfxpayload=keep insmod gzio insmod part_gpt insmod ext2 set root='hd0,gpt5' if [ x$feature_platform_search_hint = xy ]; then search --no-floppy --fs-uuid --set=root --hint-bios=hd0,gpt5 --hint-efi=hd0,gpt5 --hint-baremetal=ahci0,gpt5 76ed38c7-2c83-42c9-ad99-2e78614710df else search --no-floppy --fs-uuid --set=root 76ed38c7-2c83-42c9-ad99-2e78614710df fi echo 'Loading Linux linux-macbook ...' linux /vmlinuz-linux-macbook root=UUID=c113c01e-6b64-40fe-a8e9-1d43524fb01f rw BOOT_IMAGE=/vmlinuz-linux-macbook disable_mode=1 quiet splash rootflags=data=writeback ipv6.disable=1 echo 'Loading initial ramdisk ...' initrd /initramfs-linux-macbook.img } menuentry 'Arch Linux, with Linux linux-macbook (fallback initramfs)' --class arch --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-linux-macbook-fallback-c113c01e-6b64-40fe-a8e9-1d43524fb01f' { load_video set gfxpayload=keep insmod gzio insmod part_gpt insmod ext2 set root='hd0,gpt5' if [ x$feature_platform_search_hint = xy ]; then search --no-floppy --fs-uuid --set=root --hint-bios=hd0,gpt5 --hint-efi=hd0,gpt5 --hint-baremetal=ahci0,gpt5 76ed38c7-2c83-42c9-ad99-2e78614710df else search --no-floppy --fs-uuid --set=root 76ed38c7-2c83-42c9-ad99-2e78614710df fi echo 'Loading Linux linux-macbook ...' linux /vmlinuz-linux-macbook root=UUID=c113c01e-6b64-40fe-a8e9-1d43524fb01f rw BOOT_IMAGE=/vmlinuz-linux-macbook disable_mode=1 quiet splash rootflags=data=writeback ipv6.disable=1 echo 'Loading initial ramdisk ...' initrd /initramfs-linux-macbook-fallback.img } menuentry 'Arch Linux, with Linux linux' --class arch --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-linux-advanced-c113c01e-6b64-40fe-a8e9-1d43524fb01f' { load_video set gfxpayload=keep insmod gzio insmod part_gpt insmod ext2 set root='hd0,gpt5' if [ x$feature_platform_search_hint = xy ]; then search --no-floppy --fs-uuid --set=root --hint-bios=hd0,gpt5 --hint-efi=hd0,gpt5 --hint-baremetal=ahci0,gpt5 76ed38c7-2c83-42c9-ad99-2e78614710df else search --no-floppy --fs-uuid --set=root 76ed38c7-2c83-42c9-ad99-2e78614710df fi echo 'Loading Linux linux ...' linux /vmlinuz-linux root=UUID=c113c01e-6b64-40fe-a8e9-1d43524fb01f rw BOOT_IMAGE=/vmlinuz-linux-macbook disable_mode=1 quiet splash rootflags=data=writeback ipv6.disable=1 echo 'Loading initial ramdisk ...' initrd /initramfs-linux.img } menuentry 'Arch Linux, with Linux linux (fallback initramfs)' --class arch --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-linux-fallback-c113c01e-6b64-40fe-a8e9-1d43524fb01f' { load_video set gfxpayload=keep insmod gzio insmod part_gpt insmod ext2 set root='hd0,gpt5' if [ x$feature_platform_search_hint = xy ]; then search --no-floppy --fs-uuid --set=root --hint-bios=hd0,gpt5 --hint-efi=hd0,gpt5 --hint-baremetal=ahci0,gpt5 76ed38c7-2c83-42c9-ad99-2e78614710df else search --no-floppy --fs-uuid --set=root 76ed38c7-2c83-42c9-ad99-2e78614710df fi echo 'Loading Linux linux ...' linux /vmlinuz-linux root=UUID=c113c01e-6b64-40fe-a8e9-1d43524fb01f rw BOOT_IMAGE=/vmlinuz-linux-macbook disable_mode=1 quiet splash rootflags=data=writeback ipv6.disable=1 echo 'Loading initial ramdisk ...' initrd /initramfs-linux-fallback.img } } ### END /etc/grub.d/10_linux ### ### BEGIN /etc/grub.d/20_linux_xen ### ### END /etc/grub.d/20_linux_xen ### ### BEGIN /etc/grub.d/30_os-prober ### ### END /etc/grub.d/30_os-prober ### ### BEGIN /etc/grub.d/40_custom ### # This file provides an easy way to add custom menu entries. Simply type the # menu entries you want to add after this comment. Be careful not to change # the 'exec tail' line above. ### END /etc/grub.d/40_custom ### ### BEGIN /etc/grub.d/41_custom ### if [ -f ${config_directory}/custom.cfg ]; then source ${config_directory}/custom.cfg elif [ -z "${config_directory}" -a -f $prefix/custom.cfg ]; then source $prefix/custom.cfg; fi ### END /etc/grub.d/41_custom ### [/code]

m3thodic commented on 2017-05-29 22:20 (UTC)

@notjacktall disable_mode isn't required anymore, but wouldn't hurt anything if you still had it in your grub config.

m3thodic commented on 2017-05-29 22:19 (UTC)

@notjacktall try running grub-mkconfig -o /tmp/grub.cfg and diff the two configs, but before that it wouldn't hurt running the following as root also: mkinitcpio -p linux-macbook

notjacktall commented on 2017-05-29 22:19 (UTC)

Am I adding the wrong thing in the grub file or is there something else that I'm missing? I tried doing the disable_mode=1, but that had no effect also.

notjacktall commented on 2017-05-29 21:54 (UTC)

@m3thodic yeah i've tried that, but it didn't work. I checked the /boot/grub/grub.cgf and it looks like it is writing to it, but I just can't choose it in the GRUB on boot

m3thodic commented on 2017-05-29 21:45 (UTC)

@notjacktall after changing /etc/default/grub you have to run the following command as root: grub-mkconfig -o /boot/grub/grub.cfg

notjacktall commented on 2017-05-29 21:36 (UTC)

@m3thodic for some weird reason I am never able to boot straight into the kernel. I have to manually edit the GRUB every boot. I check my grub.cgf and it seems fine, but it just won't boot or give me the option to boot into the linux-macbook kernel. I change my GRUB_CMDLINE_DEFAULT in /etc/default/grub to BOOT_IMAGE=vmlinuz-linux-macbook, but that doesn't seem to be working.

Vokinloksar33 commented on 2017-05-28 17:23 (UTC)

Many thanks to m3thodic! I'm finally able to work on my mid-2015 mbp with arch linux. Just the wake-up from a suspend status needs a few extra seconds. would be perfect to fix that. not a big problem tho.

m3thodic commented on 2017-05-26 18:36 (UTC)

Thanks @jdz1214 if the updated wake-from-sleep is working for you I will commit the changes. I had issues with systemd and target=multi-user regarding cyclical dependencies or something along those lines but maybe this has been fixed. Also to everyone, working on syncing linux-macbook with the latest Arch kernel release. Should have something posted in a few hours or so.

jdz1214 commented on 2017-05-05 13:15 (UTC) (edited on 2017-05-05 13:26 (UTC) by jdz1214)

I created a fix to the intermittent wake-from-sleep issue by adding two lines to the macbook-wakeup.service file. The complete file is below. The modified lines are "After=suspend.target" and "WantedBy=multi-user.target sleep.target". The modifications cause the service to run whenever the computer wakes from sleep (in addition to running at boot). My mac likes to re-enable some acpi wakeups when it has gone to sleep, so running the service upon waking reverses any unwanted enabling. Steps to modify: 1) sudo nano /usr/lib/systemd/system/macbook-wakeup.service 2) add After=suspend.target and WantedBy=multi-user.target sleep.target as shown below 3) ctrl + o followed by Enter to save, ctrl + x to exit nano. 4) sudo systemctl disable macbook-wakeup.service 5) sudo systemctl enable macbook-wakeup.service 6) sudo systemctl start macbook-wakeup.service ----------macbook-wakeup.service--------------- [Unit] Description=Configure device events for linux-macbook ConditionPathExists=/proc/acpi/wakeup After=suspend.target [Service] User=root Type=oneshot ExecStart=/bin/sh -c "awk '$1 !~ /^LID/ && $3 ~ /enabled/ {print $1}' /proc/acpi/wakeup | xargs -I{} echo '{}' > /proc/acpi/wakeup" ExecStart=/bin/sh -c "awk '$1 ~ /^LID/ && $3 ~ /disabled/ {print $1}' /proc/acpi/wakeup | xargs -I{} echo '{}' > /proc/acpi/wakeup" TimeoutSec=0 [Install] WantedBy=multi-user.target sleep.target

m3thodic commented on 2017-04-18 00:55 (UTC)

@Synthead I'll have to look at PKGBUILDS for other kernels. I have no problem adding provides=() where it makes sense (ala linux-ck), I just don't know what (if any) repercussions providing the default linux kernel will have.

Synthead commented on 2017-04-17 20:13 (UTC)

Could you add provides=() to the packages in this PKGBUILD? I.e. linux-macbook would have provides=(linux), etc. This should allow other packages to use this -macbook flavored kernel as a substitute to the upstream Arch kernel. Without this, linux, linux-headers, etc. still need to be installed, but the kernel is not used.

jimnutt commented on 2017-04-17 19:45 (UTC)

Boot up problem was due to reFind using the wrong initrd. Fixing the reFind configuration solves the problem.

jimnutt commented on 2017-04-17 15:52 (UTC)

@m3thodic I'm not using grub at all, booting directly from reFind. I'll try adding splash to the kernel arguments and see what happens. I've tried running mkinitcpio by hand a few times to no avail. I'll give this newer version a shot.

Tom_B commented on 2017-04-10 19:47 (UTC)

Hi I'm getting odd issues with this new version. Occasionally suspend will work, other times it will come back on after ~90 seconds. After sudo systemctl suspend it turns back on. I'm not sure why as it was working prior to 4.10.

enrique commented on 2017-04-08 14:34 (UTC)

Hi This kernel fixed poweroff and suspend on my MacBookPro11,4, thanks!

m3thodic commented on 2017-04-04 00:38 (UTC)

@jimnutt which version were you attempting to boot? Your hooks appear correct, I'm going to venture a guess and say something isn't synced properly on your host. Try running `mkinitcpio -p linux-macbook` manually, and also in /etc/default/grub add the boot argument "splash" after "quiet" to GRUB_CMDLINE_LINUX_DEFAULT in /etc/default/grub and then run `grub-mkconfig -o /boot/grub/grub.cfg`. I had some weird issues regarding EFI and being able to properly boot a 4.10 kernel when upgrading from 4.9 and adding that splash boot argument fixed both my macbook and a custom-built PC. I tried asking about it in the arch forums but some forum crank admin moved it to the Newbie Corner.

jimnutt commented on 2017-04-03 16:47 (UTC)

I'm having trouble with this booting from LVM2. It waits the 10 seconds for the LVM devices, then drops into an emergency shell. Unfortunately, there doesn't appear to be any keyboard drivers or something, as I can't actually do anything in the emergency shell (no keystrokes appear [from either keyboard, laptop or USB] and no commands will execute). I have the following HOOKS set in mkinitcpio.conf: HOOKS="base udev autodetect keyboard modconf block lvm2 filesystems fsck" Any suggestions would be appreciated. I can fall back to the regular kernel, which works just fine, it's something to do with the macbook specific kernel.

Kisuke-CZE commented on 2017-03-15 17:00 (UTC)

just do gpg --recv-keys 38DBBDC86092693E gpg --recv-keys 79BE3E4300411886

Desdic commented on 2017-03-14 20:27 (UTC)

I'm trying to get my suspend to work and noticed this package. But I get ==> Verifying source file signatures with gpg... linux-4.9.tar ... FAILED (unknown public key 79BE3E4300411886) patch-4.9.14 ... FAILED (unknown public key 38DBBDC86092693E) ==> ERROR: One or more PGP signatures could not be verified! :: failed to verify linux-macbook integrity when installing .. what do I need ?

Kisuke-CZE commented on 2017-03-13 07:59 (UTC)

Package does not build. Something with "signature failure". Seems hashes of kernel sources has changed. Does it work for you?

deimosfr commented on 2017-02-19 21:39 (UTC)

Hi, thanks a lot for this patched kernel ! That really works better with it :-). I've another request, it looks like acpi -b can't get battery information with this patched kernel compared to the current one on Arch. Does anyone get an idea on how to get it? May be there's another way to get it? Thanks

GuineaPiet commented on 2017-02-19 12:57 (UTC)

Okay posting solution here for the unwary traveler that may come across this thread in search of answers: Chances are that after installing the linux-macbook kernel from the AUR you may have forgotten to rebuild your grub.efi file that is placed on the hard drive you are booting from. More instructions about this can be found here: http://zanshin.net/2015/02/05/arch-linux-on-a-macbook-pro-part-3-base-installation/ More specifically, the section that says Bootloader configuration and setup Chances are that this is actually posted somewhere in some document that I am just unable to find. Good luck

GuineaPiet commented on 2017-02-19 10:59 (UTC)

So if I build using makepkg, will the poweroff patches automatically be applied?

GuineaPiet commented on 2017-02-18 08:25 (UTC)

Sorry if I am being an idiot, but for the life of me I cannot find the instructions to get the power off on the Retina Macbook Pro Mid 2015 to work correctly. Can someone point me in the right direction please?

m3thodic commented on 2017-02-11 00:48 (UTC)

Hey everyone -- first off thanks for all the feedback and getting involved, very appreciated! So I was able to finish updating this port to 4.9 with the necessary patches required. I was able to adjust the brightness, suspend/resume and power-off the machine without any issues. Hopefully these updates help others that had issues with the 4.8 branch! Thanks again, please report back on any successes or failures. NOTE: if you were running this kernel since 4.8, you no longer need disable_mode=1 in your boot options.

m3thodic commented on 2017-02-10 04:10 (UTC)

Just a quick update everyone -- I got 4.9.9 to compile, but on reboot it said it couldn't find initrd or something to that effect (the text on the screen was too small for me to read). Anyway, I probably just missed something in the PKGBUILD but I should have this done and committed in a day or so. That is, unless they push out 4.9.10 -- anyone know why the kernel releases have been super aggressive recently (in the past year or so)?

m3thodic commented on 2017-02-01 19:04 (UTC) (edited on 2017-02-01 19:04 (UTC) by m3thodic)

@crencom I have been actively working on getting this up to date with 4.9 -- I've ported over all the files, hooks and configs from the base kernel since it went from testing to base but I haven't cracked it yet. I'm hoping to get something working in a few days tho, getting super close.

crencom commented on 2017-01-28 01:00 (UTC)

I have the 2015 MBPr 11,5 and have successfully installed this kernel version. Has anyone attempted to build a newer kernel version - 4.9, 4.10? I am trying to build the 4.9.6, but am erroring out at the PCI Quirk driver. Which seems to be related to the poweroff quirk patch. Any ideas?

captainju commented on 2017-01-25 13:03 (UTC)

People who experience regular freezes, have you found a solution ? I get myself freezes, especially when watching videos on youtube or netflix.

crobbins commented on 2017-01-22 05:05 (UTC)

@jjhoughton - In retrospect that should've been obvious. Thanks for your help! I can confirm that on a 2015 MacBook Pro this allows you to control the screen backlight with gmux_backlight.

jjhoughton commented on 2017-01-22 01:32 (UTC)

@crobbins You just install it using pacman sudo pacman -U linux-macbook-4.8.16-1-x86_64.pkg.tar.xz linux-macbook-headers-4.8.16-1-x86_64.pkg.tar.xz and then just add an entry in grub as you said. You should see these two files in your boot directory initramfs-linux-macbook.img vmlinuz-linux-macbook the first being your initrd and the second the kernel.

crobbins commented on 2017-01-22 01:22 (UTC)

I used makepkg to build linux-macbook, but it's not clear how to actually install and configure it. I can add a boot entry in GRUB, and I see stuff in the pkg dir, but I doubt I'm supposed to just copy the files around manually. How do you properly install this package?

jjhoughton commented on 2017-01-21 14:25 (UTC) (edited on 2017-01-21 18:24 (UTC) by jjhoughton)

From 7724b7c799279d99f1d92efe7b65ee169518963e Mon Sep 17 00:00:00 2001 From: Joshua Houghton <josh@awful.name> Date: Sat, 21 Jan 2017 18:24:56 +0000 Subject: [PATCH] drm/radeon: drop verde dpm quirks commit 7192c54a68013f6058b1bb505645fcd07015191c upstream Fixes the flickering on 2015 macbook pros. Signed-off-by: Joshua Houghton <josh@awful.name> --- drivers/gpu/drm/radeon/si_dpm.c | 13 ------------- 1 file changed, 13 deletions(-) diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index c49934527a87..2c903dadaa7a 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -3008,19 +3008,6 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev, (rdev->pdev->device == 0x6817) || (rdev->pdev->device == 0x6806)) max_mclk = 120000; - } else if (rdev->family == CHIP_VERDE) { - if ((rdev->pdev->revision == 0x81) || - (rdev->pdev->revision == 0x83) || - (rdev->pdev->revision == 0x87) || - (rdev->pdev->device == 0x6820) || - (rdev->pdev->device == 0x6821) || - (rdev->pdev->device == 0x6822) || - (rdev->pdev->device == 0x6823) || - (rdev->pdev->device == 0x682A) || - (rdev->pdev->device == 0x682B)) { - max_sclk = 75000; - max_mclk = 80000; - } } else if (rdev->family == CHIP_OLAND) { if ((rdev->pdev->revision == 0xC7) || (rdev->pdev->revision == 0x80) || -- 2.11.0

Pilzbauer commented on 2017-01-21 10:18 (UTC) (edited on 2017-01-21 10:20 (UTC) by Pilzbauer)

To ignore PGP signature errors with yaourt use: yaourt --m-arg "--skippgpcheck" -S linux-macbook

purboo commented on 2016-12-30 16:16 (UTC)

To get rid of the signature verification errors, add the --skippgpcheck parameter to makepkg: $ makepkg --skippgpcheck --si

bytemonger commented on 2016-12-17 20:06 (UTC)

Even after adding keys as suggested per @m3thodic I am still getting signature verification errors and failed builds. Running on Macbook Pro 11,5, Latest install. Any suggestions? Thank you!

bradpitcher commented on 2016-12-05 19:34 (UTC) (edited on 2016-12-06 15:38 (UTC) by bradpitcher)

Can you change the kernel config to compile in HFS+ support (rather than have it as a module)? I'm trying to export a shared HFS+ drive with NFS and I'm getting this error: exportfs: /media/mac-drive does not support NFS export From what I've seen in forums, compiling HFS+ support into the kernal should fix the issue. EDIT: I've edited the PKGBUILD, config/config.x86_64 to test this and I still get the same issue trying to use the drive with NFS. I'll let you know if I get it working. The changes I made were to change this: CONFIG_HFS_FS=m CONFIG_HFSPLUS_FS=m to this: CONFIG_HFS_FS=y CONFIG_HFSPLUS_FS=y

toastercup commented on 2016-12-01 16:01 (UTC) (edited on 2016-12-01 16:03 (UTC) by toastercup)

I can reproduce the freezes consistently (10+ times, for good measure) when having a browser + IDE open + trying to compile the kernel. It'll kaboom after about 10-15 minutes of compilation. Could mbpfan be causing any issues with heat management? Does anyone who's experiencing the freezes have mbpfan running?

Tom_B commented on 2016-11-29 15:08 (UTC)

I haven't had any freezes but could it be a temperature issue? What do you see in `sensors`? The flickering issue may be related: An overheating GPU. Does (as root) echo battery > /sys/class/drm/card0/device/power_dpm_state have any affect on stability?

anoctopus commented on 2016-11-26 03:11 (UTC)

I have also started experiencing freezes today despite no recent changes, and I'm using 4.8.4. Audio loops several seconds of output, in case that helps with debugging.

Tom_B commented on 2016-11-22 19:34 (UTC)

This patch does not appear necessary any more. On My MBP11,5 I can suspend and reboot perfectly with the stock kernel (4.8.10) and without disable_mode=1. However, I am getting the screen flicker issue. I have replied to the thread https://bbs.archlinux.org/viewtopic.php?id=219442 with more observations.

Derin commented on 2016-11-21 14:09 (UTC)

I'm still on 4.8.6 and I, too, am experiencing random freezes (have to reboot the machine to get it to restart). They just started happening today, and I haven't installed anything, so I'm curious as to what's causing them. I'll report back, here, if I find anything.

Ram-Z commented on 2016-11-18 13:39 (UTC)

I don't see any graphical glitches with 4.8.8, but I've been experiencing some random freezes. Running AwesomeWM with radeon driver on a MacBookPro11,5.

Ram-Z commented on 2016-11-17 13:57 (UTC)

I'll be testing 4.8.8, reporting back when it's done building. Instead of using `makepkg -g >> PKGBUILD`, you should be using `updpkgsums`.

gnunn commented on 2016-10-27 13:21 (UTC)

My freezing issues seem to have abated as I haven't had one since Friday (21 October). I'm not sure what fixed it though, none of the packages updated that day seem like obvious candidates. There was the 4.8.4 kernel update on Saturday though, maybe I just got lucky on Friday?

damme commented on 2016-10-24 15:55 (UTC)

I've also had problems with freezez of system. And I now managed to crash my system manually. Its the bluetooth what crashes my system. Writing hcitool -i hci0 info [connected_mac] bamm total lock-up!

toastercup commented on 2016-10-20 23:13 (UTC)

@gnunn @m3thodic Dang. Same here - NVRAM reset didn't seem to fix the freezes. I'll see if I can't extract anything from logs across my system at the time of my freezes. Thank you kindly for the suggestion!

gnunn commented on 2016-10-19 00:15 (UTC)

And it just froze on me again, so the PRAM reset didn't help unfortunately.

gnunn commented on 2016-10-18 22:34 (UTC)

Thanks @m3thodic, I've reset the PRAM as per the instructions so hopefully this will work.

hexchain commented on 2016-10-18 13:53 (UTC) (edited on 2016-10-18 13:55 (UTC) by hexchain)

@m3thodic, that file do "exists". The kernel is a 4.8.1 one but it searched for the "4.8.0" directory - I don't think we've ever used x.y.0 version number. Anyway things are back normal in 4.8.2.

m3thodic commented on 2016-10-18 06:46 (UTC)

@gnunn @toastercup I highly suggest you reset your PRAM and see if that fixes anything: Here’s how to reset your PRAM: 1) Shut down your machine. Yes, all the way down, not sleep or logging out. 2) Press the power button and then press command-option-p-r. You have to make sure you get those keys pressed before the gray screen comes up or it won’t work. 3) Hold those keys down until your Mac reboots again and you hear the startup chime. 4) Let go of the keys when after you hear the second chime and let your Mac reboot normally. NOTE: this will reset your Mac's startup disk so you will have to boot into the recovery partition (command-r) and re-run the refind install script (or whatever boot manager you use). Let us know if that works for either of you.

m3thodic commented on 2016-10-18 06:42 (UTC)

@hexchain that file is created the depmod command which is run on install from the linux-macbook.install script. If it doesn't exist I'd suggest re-running the grub-mkconfig command I've pinned to the top of the comments here.

toastercup commented on 2016-10-18 02:48 (UTC) (edited on 2016-10-18 02:49 (UTC) by toastercup)

@gnunn @m3thodic I've also experienced the lockup issues on my Macbook Pro 12,1 for around 4-6 months now. It happens every 1-2 hours and symptoms seem to worsen as the months progress - it happened in the middle of typing this, even! In addition to linux-macbook, I experience this issue across stock kernel and linux-ck. The system will suddenly lock up, I'll have to hold the power switch, and upon reboot (even if I wait before turning it back on), it will pause for around 20 seconds with a black screen, pause for 20 seconds at the UEFI boot screen, get to the gdm login screen then pause again for 20 seconds, then finally allow me to login without anymore pauses for an hour or so. I make it a habit of rebooting my system often to lessen the bruises. How odd!

gnunn commented on 2016-10-17 17:26 (UTC)

Just an FYI, I downgraded the kernel back to 4.7.6 and linux-firmware (which was recently updated as well) and I'm still getting the freeze issue so something else is obviously going on. I'll keep investigating but thanks for your feedback m3thodic, greatly appreciated.

hexchain commented on 2016-10-17 03:09 (UTC)

I cannot boot after 4.8.1 update. It complains about missing "/lib/modules/4.8.0-macbook/modules.devname" in initramfs.

gnunn commented on 2016-10-11 19:11 (UTC)

I'm using X11 and I have the same 20-radeon.conf that you do: Section "Device" Identifier "Radeon" Driver "radeon" Option "AccelMethod" "glamor" Option "DRI" "3" Option "TearFree" "on" EndSection

m3thodic commented on 2016-10-11 16:28 (UTC) (edited on 2016-10-11 16:28 (UTC) by m3thodic)

@gnunn just out of curiosity, are you using for X drivers? Here are the contents of my /etc/X11/xorg.conf.d/20-radeon.conf Section "Device" Identifier "Radeon" Driver "radeon" Option "AccelMethod" "glamor" Option "DRI" "3" Option "TearFree" "on" EndSection Also when's the last time you booted into OSX and installed any firmware updates? I just updated the package to 4.8.1 -- any better luck using that version?

gnunn commented on 2016-10-11 12:24 (UTC) (edited on 2016-10-11 13:42 (UTC) by gnunn)

System Information Manufacturer: Apple Inc. Product Name: MacBookPro11,5 Version: 1.0 Serial Number: XXXXXXXXXXXXXX UUID: 1DF3352C-A958-5E7F-81D3-09299FBF534E Wake-up Type: Power Switch SKU Number: Family: MacBook Pro

m3thodic commented on 2016-10-10 22:30 (UTC) (edited on 2016-10-10 22:31 (UTC) by m3thodic)

@gnunn, can you run `sudo dmidecode -q` and gist the output of the following section: System Information Manufacturer: Apple Inc. Product Name: MacBookPro11,5 Version: 1.0 Serial Number: XXXXXXXXXXXX UUID: 32F43A1F-887B-5FA2-9BBF-56C5779579C5 Wake-up Type: Power Switch SKU Number: Family: MacBook Pro Thanks.

gnunn commented on 2016-10-10 22:24 (UTC)

It's the 11,5 model (Macbook Pro 15") with the AMD 370x graphics adapter.

m3thodic commented on 2016-10-10 20:40 (UTC)

I'm running this kernel on a MacBook pro 15", MacBook Air 11" and an Intel NUC (to leverage the new speed-stepping code) all without any lock-ups. What model MacBook Pro are you running?

gnunn commented on 2016-10-10 20:12 (UTC)

I just upgraded to 4.8-2 and I'm having issues with my macbook pro 15 (with AMD 370x) locking up periodically (twice in the last hour) and requiring a hard reboot. Anyone else having a similar issue?

m3thodic commented on 2016-10-10 07:00 (UTC)

For any users of the brcmsmac or brcmfmac modules, check out my latest package backports-patched-bcma package (type i for interactive mode, select defconfig and then select brcmsmac or brcmfmac).

Derin commented on 2016-10-04 13:21 (UTC) (edited on 2016-10-04 13:22 (UTC) by Derin)

@jjanzic Yep, whenever I come back from suspend my Wifi adapter won't work. I can still see any active wireless networks, but the driver crashes every time I try to connect to one. A restart always solves it, though. I haven't tried m3thodic's suspend fix involving the rc-local package, I wonder if that has anything to do with it?

jjanzic commented on 2016-09-27 09:50 (UTC)

Does anyone else have wifi issues with the latest version of this package?

dumbocove commented on 2016-09-23 03:24 (UTC)

Is this normal to get this error? ``` :: Running post-transaction hooks... (1/1) Install DKMS modules ==> dkms install -m * -v * -k 4.7.4-1-macbook Error! Could not find module source directory. Directory: /usr/src/*-* does not exist. ```

habarnam commented on 2016-09-16 09:02 (UTC)

@m3thodic, there's no need to add the two keys to pacman's key database if you're running makepkg under your user.

m3thodic commented on 2016-09-12 21:43 (UTC)

If getting GPG key errors, run the following commands: sudo pacman-key -r ABAF11C65A2970B130ABE3C479BE3E4300411886 sudo pacman-key --lsign-key ABAF11C65A2970B130ABE3C479BE3E4300411886 sudo pacman-key -r 647F28654894E3BD457199BE38DBBDC86092693E sudo pacman-key --lsign-key 647F28654894E3BD457199BE38DBBDC86092693E

magister94 commented on 2016-09-03 02:32 (UTC)

Brightness is working great for me. Suspend was as well, but recently the system resumes immediately after suspending. After powering down and starting the system, it suspends successfully and stays suspended. But after waking up from the first suspension, the system no longer stays suspended. I'm not sure which update this started happening on, but I'm currently on the latest version, 4.7.2-2. Mid 2015 Macbook Pro 15" Retina Apple Radeon R9 M370X Mac Edition Intel(R) Core(TM) i7-4980HQ CPU @ 2.80GHz I'm not sure where to begin debugging this. Any help would be appreciated.

MatejLach commented on 2016-08-24 14:58 (UTC)

@rossome You probably just don't have the proper key imported, do: $ gpg --recv-keys <FINGERPRINT>

rossome commented on 2016-08-24 05:25 (UTC)

Should I be worried about the PGP checks not passing?

m3thodic commented on 2016-08-22 18:27 (UTC)

@MatejLach latest version (4.7.2) includes the patches referenced in the link.

MatejLach commented on 2016-08-22 16:27 (UTC)

@m3thodic At least for the 2015 MBP, you should apply this kernel patch: https://bugzilla.kernel.org/show_bug.cgi?id=103211#c172

Iyyel commented on 2016-08-19 20:49 (UTC)

Greetings m3thodic. Thanks for the kernel package, just trying it out. Suspend seems to work fine, however the brightness keys are not working for me. I have a 2015 Macbook Pro, i7 4980HQ, Radeon Graphics (I believe this is the culprit?) I'd love to hear back from you on this. Thank you! (I've followed all of your instructions.)

gnunn commented on 2016-06-29 19:16 (UTC)

Thanks, I was using bcwc-pcie-git, if I switch to bcwc-pcie-dkms it works fine. Looks like the dkms package drops it in the extra folder rather then extramodules.

m3thodic commented on 2016-06-29 16:47 (UTC)

@gnunn Do you have bcwc-pcie-dkms installed as well? I can confirm the camera works fine in stuff like Google Hangouts for me.

gnunn commented on 2016-06-29 16:29 (UTC)

I've noticed that the facetimehd driver is no longer working with this kernel. I re-installed the bcwc-pcie-git package and I can see that the module is properly generated in /lib/modules/extramodules-4.6-macbook-pro/facetimehd.ko.gz however when you try to modprobe it fails: [gnunn@gnunn-macbook /]$ sudo modprobe facetimehd modprobe: FATAL: Module facetimehd not found in directory /lib/modules/4.6.3-1-macbook-pro Under the standard Arch kernel it works fine, any ideas?

farhad01 commented on 2016-06-20 04:19 (UTC) (edited on 2016-06-20 04:24 (UTC) by farhad01)

thanks a lot for providing this package, it helped me fix my backlight control and sleep/shut down problem! please let others find about this package by mentioning it on wiki for example here: https://wiki.archlinux.org/index.php/MacBookPro11,x#Screen_backlight

m3thodic commented on 2016-06-18 06:28 (UTC)

@gnunn yes I just synced abs and it looks like 4.6.2 is in core now. I'll work on updating this within the next ~24 hours.

gnunn commented on 2016-06-09 14:40 (UTC)

Any update to 4.6.1-2 to match the Arch linux package?

m3thodic commented on 2016-06-01 20:31 (UTC)

@Tom_B very welcome!

Tom_B commented on 2016-06-01 16:05 (UTC)

thanks for this and the instructions.