Search Criteria
Package Details: bcwc-pcie-git 0.5.7.r1.gf85a604-1
Package Actions
Git Clone URL: | https://aur.archlinux.org/bcwc-pcie-git.git (read-only, click to copy) |
---|---|
Package Base: | bcwc-pcie-git |
Description: | Reverse engineered Linux driver for the Broadcom 1570 PCIe webcam. |
Upstream URL: | https://github.com/patjak/bcwc_pcie |
Keywords: | driver facetime macbook webcam |
Licenses: | GPL2 |
Provides: | bcwc-pcie, bcwc-pcie-dkms |
Replaces: | bcwc-pcie, bcwc-pcie-dkms |
Submitter: | whynothugo |
Maintainer: | ivladak |
Last Packager: | ivladak |
Votes: | 31 |
Popularity: | 0.021741 |
First Submitted: | 2015-12-23 14:14 (UTC) |
Last Updated: | 2021-08-13 20:58 (UTC) |
Dependencies (6)
- dkms
- facetimehd-firmware
- kmod (kmod-git, kmod-minimal-git)
- git (git-git, git-vfs, git-run-command-patch-git) (make)
- linux-headers (linux-surfacepro3-git-headers, linux-aarch64-rock64-bin-headers, linux-ec2-headers, linux-galliumos-headers, linux-zest-git-headers, linux-bootsplash-headers, linux-sumavision-q5-headers, linux-t2-wifi-headers, linux-kernel-ohio-headers, linux-rk3328-headers, linux-phicomm-n1-headers, linux-tqc-a01-headers) (optional) – Building with the stock kernel
- linux-lts-headers (optional) – Building with the lts kernel
Latest Comments
ivladak commented on 2021-08-13 21:09 (UTC)
Hi @kauron, thanks for the suggestion; done!
kauron commented on 2021-07-26 11:43 (UTC)
Hi there! Just a small suggestion for this package: the license could be set to GPL-2, according to the copyright headers in the source files. Thanks for maintaining this package!
Aetf commented on 2021-01-10 02:21 (UTC)
Disown as I no longer have this hardware
graphicmist commented on 2020-04-17 19:25 (UTC)
as mentioned by @hali this worked for me. Thanks a lot.
Note this doesn't compile on 6.0 kernel due to this bug: https://github.com/patjak/bcwc_pcie/issues/205
As mentioned there (until fixed upstream) you can change the 3 ioremap_nocache() calls to ioremap() in /usr/src/bcwc-pcie-r242.ea832ac/fthd_drv.c and run sudo dkms install bcwc-pcie/r242.ea832ac and it is OK.
joegrammer commented on 2020-04-16 15:00 (UTC) (edited on 2020-04-16 15:02 (UTC) by joegrammer)
I'm trying to build on 5.6.4-arch1-1. I changed the
ioremap_nocache
calls but I still get a number of errors, ie.I see that repeated for several uses of the
struct fthd_private *
parameter. I get this error whether I runsudo dkms install ...
or justmake
in my aur helper's cache directory (with theioremap
changes applied to the source).rberaldo commented on 2020-04-14 13:20 (UTC)
Thanks, @hali!
hali commented on 2020-04-02 16:32 (UTC) (edited on 2020-04-02 16:35 (UTC) by hali)
Note this doesn't compile on 6.0 kernel due to this bug: https://github.com/patjak/bcwc_pcie/issues/205
As mentioned there (until fixed upstream) you can change the 3
ioremap_nocache()
calls toioremap()
in/usr/src/bcwc-pcie-r242.ea832ac/fthd_drv.c
and runsudo dkms install bcwc-pcie/r242.ea832ac
and it is OK.eschwartz commented on 2019-12-15 22:48 (UTC)
Thanks for the prompt response. :)
Aetf commented on 2019-12-15 22:45 (UTC) (edited on 2019-12-16 02:38 (UTC) by Aetf)
@eschwartz thanks for pointing out the --devel flag. The last time I check it didn't work for some reason. Now it works quite well.
I'm updating all my AUR packages.
Edit: Done ;)
eschwartz commented on 2019-12-15 16:53 (UTC)
@Aetf
Speaking with my official TU hat on, you are required to stop using pkgver=0.0.0, effective immediately, on all AUR packages you maintain.
Users who wish to update AUR VCS-based packages with an AUR helper are expected to use the commonly implemented --devel flag to check VCS packages for updates. Adding an infinitely high pkgver= and relying on the pkgver() function to reset it is NOT acceptable, as this constantly spams users to update even when there is no update available. Furthermore, the pkgrel= is reset to 1 upon the pkgver() function successfully modifying the pkgver=, so it doesn't do anything.
VKen commented on 2019-12-14 02:44 (UTC)
@Aetf: Thanks for the explanation of intended behavior.
Thank you for maintaining the package and I can understand the current setup offloads the maintainer from needing to update the pkgver for every git upstream update.
Indeed there's no rebuild from the output below.
However, it is a matter of my tooling usage and the dissonance compared to my usual experience up to now. The AUR helper usually (with the majority of the small number of git AUR packages I use) will not be flagged for update via
yay -Syu
.I will probably need to find a way for myself to configure the AUR helper or augment my usage behavior for: 1. unable to know whether I need to update, since the AUR helper can't tell due to version difference as it overwrites the PKGBUILD with the latest
pkgver=0.0.0
to be mismatched from the current installedr242.ea832ac-1
. (intended behavior) 2. Although rebuilding is stopped, I need to stop the auto-reinstall if there's no package rebuild. (re-install is probably unintended. A tooling behavior issue to check)Aetf commented on 2019-12-13 02:03 (UTC)
@VKen That's the intended behavior. I deliberately set pkgver=0.0.0 on AUR. So AUR helpers will always remind you to update the package.
For this particular package, the PKGBUILD is unlikely to change in a long time, so the pkgver can get out-of-sync with the latest git version. (In general, I don't think it's useful to update PKGBUILD just to bump the git version)
So this workaround forces AUR helpers to download and check the latest git version. If the latest git version matches what you have locally installed, no package will be built.
VKen commented on 2019-12-13 01:37 (UTC)
@Aetf: The declared version
pkgver=0.0.0
does not coincide with the output of:During
makepkg
the execution modifiesPKGBUILD
file's declaration ofpkgver=0.0.0
to the output ofpkgver() == 'r242.ea832ac'
.On my AUR helper "yay", it continually keeps trying to update, rebuild, install, every time I run
yay -Syu
.Do you experience this update/install with every run as well?
This package depends on upstream git branch. I've taken a look at other similar git packages like "networkmanager-git"
Their declared
pkgver
coincides with the output ofpkgver()
. Their package doesn't trigger the self-update on every invocation of package sync update of the AUR helper.Aetf commented on 2019-12-10 19:36 (UTC)
@GusevAG it's the same.
master
is the default branch anyway.GusevAG commented on 2019-12-10 19:26 (UTC) (edited on 2019-12-10 19:28 (UTC) by GusevAG)
@Aetf it works perfectly! before I read your comment, I added #branch=master after the source to the PKGBUILD btw (no idea was it successful or not, but it works)
Aetf commented on 2019-12-10 18:45 (UTC)
@adbjesus thanks for pointing out. I just updated the PKGBUILD to use the master branch.
@GusevAG please try the latest version.
PS. I updated the version number on AUR to always be 0.0.0, making it in line with other git packages I manage on AUR. This way the package will always be shown as updated when using AUR helpers so you can pull the latest changes from the upstream. When building, makepkg will derive the correct version number.
GusevAG commented on 2019-12-10 15:34 (UTC) (edited on 2019-12-10 15:38 (UTC) by GusevAG)
@adbjesus wow, that is very important point! thank you for pointing at it! I had no clue about different branches and I have no idea how to change it to master in PKGBUILD could you please give me an example or a picture of what to change exactly and where? I will really appreciate it!
UPD: is it just about changing to
?
adbjesus commented on 2019-12-10 09:20 (UTC) (edited on 2019-12-10 09:21 (UTC) by adbjesus)
I was having trouble using this for linux-lts and reported it upstream, https://github.com/patjak/bcwc_pcie/issues/193
The owner of the repo suggested using the master branch instead of the mainline and also that the master branch should work fine for all kernels. I quickly tested it and it seems to work fine on both linux and linux-lts so it might be worth updating the PKGBUILD to use master instead. Or is there some reason to be using the mainline branch?
GusevAG commented on 2019-12-08 19:04 (UTC) (edited on 2019-12-08 19:06 (UTC) by GusevAG)
@eblau well, for some reason this is what i get
but
eblau commented on 2019-12-08 12:48 (UTC)
@GusevAG no, I didn't have to do anything special. The package already makes sure the modules are loaded after they are installed. facetimehd shows up as loaded for me and if I follow the guide here:
https://wiki.archlinux.org/index.php/Webcam_setup
I see the device show up:
It "just works" in any program that uses the standard access to /dev/videoX
GusevAG commented on 2019-12-07 19:02 (UTC)
@eblau wow, thats fantastic, were any additional tweaks required to make it run? because i cant manage to run it...
eblau commented on 2019-12-07 16:35 (UTC)
@GusevAG yes, the new version works for me. At least it works in zoom for web conferencing and the "lighter" tool I use for capturing ambient light levels and adjusting the screen and keyboard backlights still works fine.
I also have a 13" mid-2015 MacBook Pro. Here's the model:
Apple Inc. MacBookPro12,1/Mac-E43C1C25D4880AD6, BIOS 186.0.0.0.0 06/14/2019
GusevAG commented on 2019-12-07 15:48 (UTC)
@eblau yes it does builds, but does it works for you? because it doesn't on 13" 2015 bmp
eblau commented on 2019-12-07 14:36 (UTC)
My apologies. I neglected to update bcwc-pcie-git. When I rebuilt the package, it updated to r259.9d7ff828ebe9-1 and installed successfully with linux-5.4.2.arch1-1.
eblau commented on 2019-12-07 14:32 (UTC)
I updated to linux-5.4.2.arch1-1 and am getting errors when rebuilding the module:
Aetf commented on 2019-10-25 16:14 (UTC)
@GusevAG, hi, this is an error building the dependency
facetimehd-firmware
, not an error building this package. Since the error was in their makefile, I suggest you report that to the upstream: https://github.com/patjak/bcwc_pcie/issues/newPS. next time please paste the log using code format so it's easier to read: https://wiki.archlinux.org/index.php/Arch_User_Repository#Comment_syntax
GusevAG commented on 2019-10-25 13:23 (UTC) (edited on 2019-10-26 09:24 (UTC) by GusevAG)
is not working on 5.2.21-1-MANJARO KDE
[Installing linux52-headers (5.2.21-1)... Running post-transaction hooks... Updating linux52 module dependencies... Install DKMS modules Arming ConditionNeedsUpdate... Updating the info directory file...
Building facetimehd-firmware... ==> Making package: facetimehd-firmware 1.43.0-1 (Птн 25 окт 2019 16:13:22) ==> Checking runtime dependencies... ==> Checking buildtime dependencies... ==> Retrieving sources... -> Cloning facetimehd-firmware git repo... Cloning into bare repository '/var/tmp/pamac-build-andrey/facetimehd-firmware/facetimehd-firmware'... ==> Validating source files with md5sums... facetimehd-firmware ... Skipped ==> Removing existing $srcdir/ directory... ==> Extracting sources... -> Creating working copy of facetimehd-firmware git repo... Cloning into 'facetimehd-firmware'... done. ==> Starting prepare()...
Checking dependencies for driver download... /usr/bin/curl /usr/bin/xzcat /usr/bin/cpio
Downloading the driver, please wait... mv: cannot stat 'System/Library/Extensions/AppleCameraInterface.kext/Contents/MacOS/AppleCameraInterface': No such file or directory make: *** [Makefile:41: AppleCameraInterface] Error 1 ==> ERROR: A failure occurred in prepare(). Aborting...]
yan12125 commented on 2019-03-25 01:34 (UTC)
Wow thanks for the ultra fast fix!
yan12125 commented on 2019-03-24 12:47 (UTC)
Hi, could you remove the bcwc-pcie-git.install file? All dkms commands are already handled by ALPM hooks /usr/share/libalpm/hooks/70-dkms-{install,remove}.hook, so the bcwc-pcie-git.install file is not necessary. Furthermore, errors occur if the running kernel version is different from than the installed one, as commands in bcwc-pcie-git.install do not set the target kernel version:
whynothugo commented on 2018-10-02 15:55 (UTC)
I no longer own hardware which uses this driver, so I'll be disowning this package since I can't even test it any more.
Feel free to adopt is, especially if any changes are needed. Upstream has been pretty stable though, so this shouldn't have any issues.
trp commented on 2017-10-03 19:44 (UTC) (edited on 2017-10-03 19:45 (UTC) by trp)
taekyung commented on 2017-07-18 05:59 (UTC)
whynothugo commented on 2017-07-06 00:31 (UTC)
kroylar commented on 2017-07-06 00:25 (UTC)
whynothugo commented on 2017-05-24 19:56 (UTC)
khaezaid commented on 2017-05-01 22:28 (UTC)
didacog commented on 2017-04-25 17:45 (UTC)
whynothugo commented on 2017-04-25 15:37 (UTC)
whynothugo commented on 2017-04-25 15:36 (UTC)
didacog commented on 2017-04-25 10:26 (UTC)
graphicmist commented on 2017-02-03 16:15 (UTC)
bcc commented on 2016-10-20 12:34 (UTC)
m3thodic commented on 2016-10-11 07:59 (UTC)
iamazat commented on 2016-10-10 18:48 (UTC)
whynothugo commented on 2016-05-31 23:54 (UTC)
Jack12816 commented on 2016-05-31 18:20 (UTC)
gokcen commented on 2016-05-14 07:05 (UTC)
A.J commented on 2016-03-23 10:27 (UTC)
A.J commented on 2016-03-23 03:29 (UTC)
whynothugo commented on 2016-03-22 14:09 (UTC)
A.J commented on 2016-03-22 03:53 (UTC) (edited on 2016-03-22 11:52 (UTC) by A.J)
trusktr commented on 2016-02-26 01:19 (UTC)
trusktr commented on 2016-02-26 01:16 (UTC)
gsora commented on 2016-01-08 16:33 (UTC)