October 2018 - Arm-dev - lists.centos.org

rng-tools missing and /dev/hwrng without driver
by SW＠EU 31 Oct '25

31 Oct '25

Hi, today I will report a problem that is released to ipa-server. This server contains a certificate authority and such service need many entropy. The default on CentOS 7 on a Banana PI is not enough, i.e. $(cat /proc/sys/kernel/random/entropy_avail) is less than 1000. I have solved this in meantime by installing and enabling of haveged from the EPEL repository. Normally it would be done by installing the rng-tools. But there are two problems: 1. The rng-tools was not in the repositories, so I have downloaded rng-tools-5-8.fc24.armv7hl.rpm because this are the same version which is included in CentOS 7.3 for x86_64. 2. This rng-tools are usable but the daemon starts and stops immediately with the following error message: # rngd -v /dev/hwrng: No such device /dev/tpm0: No such file or directory No entropy sources found, exiting This is not the problem of this binary it is a problem of the Kernel. /dev/hwrng exists and if I remove it then it reappears after reboot, but # ls -l /dev/hwrng crw-------. 1 root root 10, 183 1. Jan 1970 /dev/hwrng # udevadm info -a -n /dev/hwrng Udevadm info starts with the device specified by the devpath and then walks up the chain of parent devices. It prints for every device found, all possible attributes in the udev rules key format. A rule to match, can be composed by the attributes of the device and the attributes from one single parent device. looking at device '/devices/virtual/misc/hw_random': KERNEL=="hw_random" SUBSYSTEM=="misc" DRIVER=="" ATTR{rng_current}=="none" ATTR{rng_available}=="" there is no driver for this device. I have searched and found this link http://forum.lemaker.org/thread-23618-1-1.html which includes a link to the full story. If I read all right then on bananian /dev/hwrng appears only if the adapted or a more actual sun4i-ss.ko module is loaded (there is written: "module author has indicated this will be going into the mainline kernel shortly“). This module is also loaded on a Banana PI with current CentOS 7. So does the kernel of CentOS 7.3 for ARM32 include this patch and if yes why it does not work or otherwise why this device appears but has no driver? TIA, Silvio Sent with [ProtonMail](https://protonmail.com) Secure Email.

3 4

EPEL repo not available?
by Boyd Ako 31 Oct '25

31 Oct '25

Can anybody verify that the EPEL repo for ARMHFP isn’t available? I keep getting an error saying the REPOMD file isn’t there when running yum. It works if I use —disablerepo=epel\* when using yum. But, there might be some stuff in the EPEL that I’m looking for. Running it on a Raspberry Pi 3 and trying to get Nagios installed for home use. Domo, Boyd H. Ako boyd.hanalei.ako(a)gmail.com (424) 244-9653 https://www.boydhanaleiako.me “Coming together is a beginning. Keeping together is progress. Working together is success.” -Henry Ford PGP/GPG Public Key: https://sks-keyservers.net/pks/lookup?op=get&search=0xC58073B21618F134

4 3

Gigabyte MP30-AR0
by Gordan Bobic 31 Oct '25

31 Oct '25

Hi, I am toying with the idea of getting a Gigabyte MP30-AR0 board (despite the insanely exorbitant price tag). Has anyone got the aarch64 CentOS working on one of those? Any special instructions? Also, if I get the ISO, is it bootable on aarch64? Does the installer work? Or does the image have to be used like on 32-bit ARM? Thanks. Gordan

13 203

epel-release
by Robert Moskowitz 30 Jun '20

30 Jun '20

Ok, I see this epel-release in the epel1 repo that I installed via the wiki. So I decide to install it, and it created an epel.repo.rpmnew file. I swicth around the files and try a 'yum update' and get: epel/armhfp/metalink | 47 kB 00:00 Could not parse metalink https://mirrors.fedoraproject.org/metalink?repo=epel-7&arch=armhfp error was No repomd file One of the configured repositories failed (Unknown), and yum doesn't have enough cached data to continue. At this point the only safe thing yum can do is fail. There are a few ways to work "fix" this: .... File /var/cache/yum/armhfp/7/epel/metalink.xml does not exist So I put everything back.

3 7

Re: [Arm-dev] Add Marvell Armada7k8k NIC (mvpp2) support patches
by Marcin Wojtas 26 May '20

26 May '20

Hi Jim, It's been a while, since I sent the patch. Any objections about merging? Is v4.11 kernel still a valid baseline or for the next releases or we should use v4.14? When is that transition and next release supposed to happen? Best regards, Marcinsob., 23 cze 2018 o 08:18 Marcin Wojtas <mw(a)semihalf.com> napisał(a): > > Hi, > > I'm sending a patch that is applicable on top of the kernel > sig-altarch7-aarch64 branch. It enables ACPI support for the Marvell > Armada7k8k NIC - this required backporting very big amount of patches > (all merged upstream), hence I decided to attach the single commit, > instead of issueing 'git send-email' of it. > > Despite many attempts I wasn't able to build entire rpm package with > the mock build utility, however I managed to test the clean kernel > v4.11 with applied all patches and built with updated .config. Please > let know if this patch is sufficient and can be merged. > > Thanks, > Marcin

5 21

Different issues with "DTOVERLAY".
by Stephan Guilloux 26 Feb '19

26 Feb '19

Hello, Small issues, today, with dtoverlay gpio-shutdown/poweroff [root@mypi ~]# dtoverlay gpio-shutdown su: user pi does not exist * Failed to apply overlay '1_gpio-shutdown' (kernel) su: user pi does not exist [root@mypi ~]# dtoverlay gpio-poweroff su: user pi does not exist su: user pi does not exist [root@mypi ~]# Even if I add a fake user "PI", this does not work better. [root@mypi ~]# useradd -r pi -p pi [root@mypi ~]# [root@mypi ~]# [root@mypi ~]# dtoverlay gpio-poweroff * Failed to apply overlay '2_gpio-poweroff' (kernel) [root@mypi ~]# dtoverlay gpio-shutdown * Failed to apply overlay '2_gpio-shutdown' (kernel) [root@mypi ~]# But when I check the overlay list, I get: [root@mypi ~]# dtoverlay -l Overlays (in load order): 0: gpio-shutdown 1: gpio-poweroff [root@mypi ~]# At least, something is done :-) Last problems: [root@mypi ~]# dtoverlay -h gpio-shutdown * Help file not found [root@mypi ~]# dtoverlay -h gpio-poweroff * Help file not found [root@mypi ~]# dtoverlay -h rpi-rtc * Help file not found [root@mypi ~]# dtoverlay -h rpi-display * Help file not found [root@mypi ~]# In case ... [root@mypi ~]# cat /proc/version Linux version 4.14.65-v7.1.el7 (mockbuild(a)armhfp-03.bsys.centos.org) (gcc version 4.8.5 20150623 (Red Hat 4.8.5-28) (GCC)) #1 SMP Mon Aug 20 19:04:20 UTC 2018 [root@mypi ~]# [root@mypi ~]# rpm -qa 'raspberrypi-vc*' raspberrypi-vc-demo-source-20170705-502.gitaa39775.el7.noarch raspberrypi-vc-libs-20170705-502.gitaa39775.el7.armv7hl raspberrypi-vc-utils-20170705-502.gitaa39775.el7.armv7hl [root@mypi ~]# Stephan.

2 3

MacchiatoBin support int the Centos7
by Marcin Wojtas 30 Oct '18

30 Oct '18

Hi Jim, Since the newest Centos7 works out-of-the box with the latest MacchiatoBin firmware and also the integrated NICs are supported (beggining from kernel-alt-4.14.0-49.13.1.el7a), can you please shift MacchiatoBin to 'Supported Hardware' section in https://wiki.centos.org/SpecialInterestGroup/AltArch/AArch64 ? Thanks, Marcin

3 3

Enterprise distros and the two faces of 'reliability'
by Fred Gleason 28 Oct '18

28 Oct '18

Howdy Folks: Our discussion of the distinctive elements of CentOS’s Aarch64 variant has gotten me pondering. What does an ‘enterprise’ distro bring to the table that makes it desirable over the alternatives available? (By ‘enterprise distro’, I here mean the stream of data originated by the Upstream Provider, as well as those modified and distributed by all the various downstream projects, including CentOS). To my mind, I see two basic categories that are relevant here: 1) ‘Enterprise’ features, by which I mean things like: supports ‘big iron’ systems, supports fault tolerance and scalability, interoperates well with all of the common ‘non-native’ protocols and features found in a typical large computing environment. 2) ‘Stability’ features, by which I mean: has a well-defined infrastructure for distributing code fixes and security updates, has a long support lifetime, manages the ABI carefully so as to ensure that system updates can be applied with good confidence that they will not break existing locally-installed applications. These two categories are largely orthogonal. I say ‘largely’ because there is a hidden implicit commonality: reliability. Yet the presence of one need not necessarily imply the presence (or even desirability) of the other. For example, a project that uses an ‘embedded’ board (BeagleBoard, etc) to make small, autonomous ‘widget’ type devices really does not care about about the presence of 1), while 2) can be a highly desirable attribute. Category 1) is essentially an *operational* attribute, desirable in certain use cases but superfluous (or even counterproductive) in others, while category 2) addresses *maintenance* issues, critical for long-term service and support of a system but largely irrelevant to the typical day-to-day operation of applications. All of this to pose the question: is an ‘enterprise’ distro (in the specific sense meant here) an appropriate long-term choice for an ‘embedded’ project? Given the stated intention of the Upstream Provider to support only ARM systems that integrate APCI and comply with SBSA [Server Base System Architecture] standards in future major releases (see https://lists.centos.org/pipermail/arm-dev/2017-October/003120.html) is such a distro an appropriate long-term choice for an ‘embedded’ project? Cheers! |----------------------------------------------------------------------| | Frederick F. Gleason, Jr. | Chief Developer | | | Paravel Systems | |----------------------------------------------------------------------| | A good plan today is better than a perfect plan tomorrow. | | -- General George Patton | |----------------------------------------------------------------------|

6 10

Hardware Giveaway
by Gordan Bobic 27 Oct '18

27 Oct '18

I've been meaning to get back to some of my ARM hardware, but inbetween all the work commitments it doesn't look like that's feasible before these ARMs will only be good for the landfill. So, if anyone here is interested, here is what is up for grabs, for free: - Toshiba AC-100 x3 (at least one is in it's original box) - Genesi Efika MX Smartbook (in original box) - Compulab SB-A510+CM-A510 (yup, in original box) - Compulab TrimSlice (you guessed it - in original box) All have been used, but are in condition ranging from excellent to mint. If anyone wants any of these for a purpose better than it gathering dust in their cupboard rather than mine, email me off list with a shipping label for posting from mainland UK, and I'll post it to you. Best regards. Gordan

1 1

hikey 960?
by Skorpeo Skorpeo 26 Oct '18

26 Oct '18

Has anyone been able to install centos using iso on a Hikey 960? The wiki says "The Hikey from 96boards mostly works, but setup is complex. The installer currently cannot be used as one would traditionally expect." Any guides out there on how to get it to work? I am only able to get it to the initial installation menu but no further, I tried playing with various kernel parameters but to no avail. Any guidance or suggestions would be appreciated!

4 7

Re: [Arm-dev] Hardware Giveaway
by Mark Verlinde 25 Oct '18

25 Oct '18

-----Oorspronkelijk bericht----- Afzender: Gordan Bobic<gordan(a)redsleeve.org> Verstuurd: Donderdag 25 Oktober 2018 21:01 Aan: redsleeve-users(a)googlegroups.com Cc: Conversations around CentOS on ARM hardware <arm-dev(a)centos.org> Onderwerp: [Arm-dev] Hardware Giveaway Ive been meaning to get back to some of my ARM hardware, but inbetween all the work commitments it doesnt look like thats feasible before these ARMs will only be good for the landfill. So, if anyone here is interested, here is what is up for grabs, for free: - Toshiba AC-100 x3 (at least one is in its original box) - Genesi Efika MX Smartbook (in original box) - Compulab SB-A510+CM-A510 (yup, in original box) - Compulab TrimSlice (you guessed it - in original box) All have been used, but are in condition ranging from excellent to mint. If anyone wants any of these for a purpose better than it gathering dust in their cupboard rather than mine, email me off list with a shipping label for posting from mainland UK, and Ill post it to you. Best regards. Gordan The SB-A510+CM-A510 is very appealing, but I an afraid it would get the same destiny in my possession... I'll sleep over it a couple of nights,; if it still buzzes in my head i might take on your offer. Grtz Mark

1 0

Odroid UX4/HC1,2 Community kernel 4.18.x
by Mark Verlinde 25 Oct '18

25 Oct '18

-----Oorspronkelijk bericht----- > Afzender:arm_ml@rirasoft.de <arm_ml(a)rirasoft.de> > Verstuurd: Donderdag 25 Oktober 2018 9:14 > Aan: arm-dev(a)centos.org > Onderwerp: Re: [Arm-dev] Community kernel 4.18.x > > > Am 2018-10-23 17:49, schrieb Mark Verlinde: > > -----Oorspronkelijk bericht----- > > >> > >> Mark: how do you get CentOS with this actual kernel on the Odroid HC1 > >> ? > >> I've 2 of them running Fedora as a Mailserver and Nextcloudserver. > > > ---schnipp-- > > > > > > _OT_ > > For personal use and on request from others I made a flash-able img > > file with 4.18.16 installed. > > It is a bit different from the default userland-generic though: 2 > > partitions starting @4096 (1) boot (2) rootfs; compressed swap-space > > in ram. > > NOTE: this link may be short lived as the setup is experimental > > https://drive.google.com/open?id=1PvLxWVwvK9H2zJTnBF2rKlA7IDmbpa7s > > and my rough notes on the HC1: > > https://gist.github.com/markVnl/7adf25cebfdce5d5009d93b7b4caae96 > > _OT_ > > > > Greetings Mark > >> > >> Greetings > >> Andreas > >> _______________________________________________ > >> Arm-dev mailing list > >> Arm-dev(a)centos.org > >> https://lists.centos.org/mailman/listinfo/arm-dev > >> > > Hello Mark, > > excellent work ! I've a running Odroid-XU4 with Cloudshell 2 running > with CentOS !! > I've done this working from a Cubietruck Plus with CentOS. > > Thank you. > > Andreas Thank you for your feedback! Probably unnecessary to note it is still in a testing condition; So if you experiencing troubles please share them with us! Gtz Mark

1 0

Re: [Arm-dev] Community kernel 4.18.x
by Mark Verlinde 25 Oct '18

25 Oct '18

-----Oorspronkelijk bericht----- > Afzender:arm_ml@rirasoft.de <arm_ml(a)rirasoft.de> > Verstuurd: Dinsdag 23 Oktober 2018 9:39 > Aan: arm-dev(a)centos.org > Onderwerp: Re: [Arm-dev] Community kernel 4.18.x > > Am 2018-10-22 19:01, schrieb Mark Verlinde: > > -----Oorspronkelijk bericht----- > >> Afzender:Mark Verlinde <mark(a)havak.nl> > >> Verstuurd: Maandag 22 Oktober 2018 14:27 > >> Aan: Conversations around CentOS on ARM hardware <arm-dev(a)centos.org> > >> Onderwerp: Re: [Arm-dev] Community kernel 4.18.x > > > >> > >> Thank you for the hints, it Led to a booting setup! > >> > >> Still need to fine tine, some customization may not be needed > >> howerever this configuration led to the booting odroid HC1. > >> > >> Appended cpuidle.off=1 console=tty1 console=ttySAC2,115200n8 to the > >> cmdline: > >> > >> # cat /boot/extlinux/extlinux.conf > >> # extlinux.conf generated by appliance-creator > >> ui menu.c32 > >> menu autoboot Welcome to Centos-Uboot-Community-kernel. Automatic boot > >> in # second{,s}. Press a key for options. > >> menu title Centos-Uboot-Community-kernel Boot Options. > >> menu hidden > >> timeout 20 > >> totaltimeout 600 > >> > >> default=Centos-Uboot-Community-kernel (4.18.14-200.el7.armv7hl) > >> label CentOS Linux (4.18.16-200.el7.armv7hl) 7 (AltArch) > >> kernel /vmlinuz-4.18.16-200.el7.armv7hl > >> append cpuidle.off=1 console=tty1 console=ttySAC2,115200n8 ro > >> root=UUID=a306d630-4957-4da0-8dd6-79f943147d61 LANG=en_US.UTF-8 > >> fdtdir /dtb-4.18.16-200.el7.armv7hl > >> initrd /initramfs-4.18.16-200.el7.armv7hl.img > >> > >> label Centos-Uboot-Community-kernel (4.18.14-200.el7.armv7hl) > >> kernel /vmlinuz-4.18.14-200.el7.armv7hl > >> append cpuidle.off=1 console=tty1 console=ttySAC2,115200n8 ro > >> root=UUID=a306d630-4957-4da0-8dd6-79f943147d61 > >> fdtdir /dtb-4.18.14-200.el7.armv7hl > >> initrd /initramfs-4.18.14-200.el7.armv7hl.img > >> > >> In an effort to mimic the FC28 intenramfs added/omitted some modules > >> and > >> added _all_ exynos and samsung divers loaded by the booting FC28 > >> setup: > >> > >> #cat /etc/dracut.conf.d/odroid_hc1.conf > >> add_dracutmodules+="dm" > >> omit_dracutmodules+="network ifcfg" > >> add_drivers+="exynosdrm exynos_bus exynos_trng exynos_nocp exynos_rng > >> phy_exynos_dp_video phy_exynos_mipi_video dwc3_exynos clk_exynos_audss > >> i2c_exynos5 dw_mmc_exynos dw_mmc_pltfm dw_mmc phy_exynos_usb2 > >> phy_exynos5_usbdrd pwm_samsung " > >> > >> (AGAIN) NOTE: some may be necessary others may be ommited.. > > > > Turns out one additional driver - dwc3 - is needed to boot the odroid > > HC1 with a rootfs residing on a SATA device. > > My best guess is dwc3_exynos actually depends on dwc3 and/or (one of) > > the dependencies of dwc3. > > > > > > #cat /etc/dracut.conf.d/odroid_hc1.conf > > add_dracutmodules+="dm" > > omit_dracutmodules+="network ifcfg" > > add_drivers+=" exynosdrm exynos_bus exynos_trng exynos_nocp exynos_rng > > phy_exynos_dp_video phy_exynos_mipi_video dwc3_exynos clk_exynos_audss > > i2c_exynos5 dw_mmc_exynos dw_mmc_pltfm dw_mmc phy_exynos_usb2 > > phy_exynos5_usbdrd pwm_samsung " > > add_drivers+=" dwc3" > > > > Mark: how do you get CentOS with this actual kernel on the Odroid HC1 ? > I've 2 of them running Fedora as a Mailserver and Nextcloudserver. The kernel are kindly build by Pablo and can be found at: https://armv7.dev.centos.org/repodir/community-kernel-latest/ On a running armhfp device (distro or kernel should not matter); You may mount the sd-card and chroot in to the card to install / update the kernel. Mount the rootfs partition (usually part4) of your sd card and the boot partition (usually part2); mount proc and sys of the "host" too (ie): mount /dev/sdX4 /mnt mount /dev/sdX2 /mount/boot mount --bind /proc /mnt/proc mount --bind /sys /mnt/sys then chroot /mnt You may need to copy the /etc/resolve.conf from your host system to kickstart networking. _OT_ For personal use and on request from others I made a flash-able img file with 4.18.16 installed. It is a bit different from the default userland-generic though: 2 partitions starting @4096 (1) boot (2) rootfs; compressed swap-space in ram. NOTE: this link may be short lived as the setup is experimental https://drive.google.com/open?id=1PvLxWVwvK9H2zJTnBF2rKlA7IDmbpa7s and my rough notes on the HC1: https://gist.github.com/markVnl/7adf25cebfdce5d5009d93b7b4caae96 _OT_ Greetings Mark > > Greetings > Andreas > _______________________________________________ > Arm-dev mailing list > Arm-dev(a)centos.org > https://lists.centos.org/mailman/listinfo/arm-dev >

2 1

More notes on Xfce
by Robert Moskowitz 23 Oct '18

23 Oct '18

I am now connected via vnc. I am using the following for my ~/.vnc/xstartup #!/bin/sh unset SESSION_MANAGER unset DBUS_SESSION_BUS_ADDRESS # exec /etc/X11/xinit/xinitrc exec /bin/sh /etc/xdg/xfce4/xinitrc xrdb $HOME/.Xresources startxfce4 & No avahi-daemon Oct 23 11:39:29 ct-c7 yum[1950]: Installed: avahi-libs-0.6.31-19.el7.armv7hl Oct 23 11:40:09 ct-c7 yum[1950]: Installed: avahi-glib-0.6.31-19.el7.armv7hl Oct 23 11:51:35 ct-c7 journal: Failed to connect to avahi service: Daemon not running Oct 23 11:51:51 ct-c7 journal: Failed to connect to avahi service: Daemon not running But is it really needed on a server? No cups installed. That would be a separate step, I assume. No auto media mount. I insert a usb drive. I see it in /dev as sdb, but it is not showing on the desktop to mount. Seems I had this when I first installed gnome DE and then installed Xfce DE, and selected Xfce DE. Only ~/Desktop directory created. None of the other directories you tend to see for a user account. I am going to put this drive to the side and work on other stuff for now. Unless someone recommends testing some addition. Meanwhile the install of the Xfce group needs to install Base and "X Windows System" it would seem. thanks

1 1

PHP7x SCL for aarch64
by Mark Verlinde 23 Oct '18

23 Oct '18

Hi, Are there rumors / tasks / plans for SCL packages targeted @aarch64? Found devtoolsets in buildlogs.centos.org, but {rh,sclo}PHP7x repositories are frightening empty. Thanks in advance, Mark Verlinde

1 0

Attempt to install Xfce by hand
by Robert Moskowitz 23 Oct '18

23 Oct '18

You may have seen my posts around on my challenges to build a couple groups to do a group install. That got me to realize, that I probably know which rpms to install and I could 'do it by hand' for now. So first step was: yum install glx-utils initial-setup-gui mesa-dri-drivers plymouth-system-theme spice-vdagent xorg-x11-drivers xorg-x11-server-Xorg xorg-x11-utils xorg-x11-xauth xorg-x11-xinit xvattr tigervnc-server wayland-protocols-devel xorg-x11-drv-keyboard xorg-x11-drv-libinput xorg-x11-drv-mouse xorg-x11-drv-openchrome Where xorg-x11-drv-openchrome was not available, but is flagged optional in the group. This installed 16 packages and 263 dependents. Next the actual Xfce: yum install Thunar xfce-utils xfce4-panel xfce4-session xfce4-settings xfconf xfdesktop xfwm4 gdm leafpad openssh-askpass orage polkit-gnome thunar-archive-plugin thunar-volman tumbler xfce4-appfinder xfce4-icon-theme xfce4-power-manager xfce4-session-engines xfce4-terminal xfwm4-theme-nodoka xfwm4-themes pinentry First this showed: No package xfce-utils available. No package leafpad available. No package thunar-archive-plugin available. No package xfce4-icon-theme available. No package xfwm4-theme-nodoka available. No leafpad is OK, as a check shows that mousepad is available and this is what is in Fedora28. No xfce-utils is not good. It got worst with critical missing dependencies that need to be fixed before I could get this to work: Error: Package: xfce4-power-manager-1.4.1-4.el7.armv7hl (epel) Requires: libxfce4util.so.6 Available: libxfce4util-4.10.1-2.el7.armv7hl (epel) libxfce4util.so.6 Installing: libxfce4util-4.12.1-2.el7.armv7hl (epel) ~libxfce4util.so.7 Error: Package: xfdesktop-4.10.3-1.el7.armv7hl (epel) Requires: libxfce4util.so.6 Available: libxfce4util-4.10.1-2.el7.armv7hl (epel) libxfce4util.so.6 Installing: libxfce4util-4.12.1-2.el7.armv7hl (epel) ~libxfce4util.so.7 Error: Package: xfce4-session-4.10.1-8.el7.armv7hl (epel) Requires: libxfce4util.so.6 Available: libxfce4util-4.10.1-2.el7.armv7hl (epel) libxfce4util.so.6 Installing: libxfce4util-4.12.1-2.el7.armv7hl (epel) ~libxfce4util.so.7 Error: Package: orage-4.10.0-5.el7.armv7hl (epel) Requires: libxfce4util.so.6 Available: libxfce4util-4.10.1-2.el7.armv7hl (epel) libxfce4util.so.6 Installing: libxfce4util-4.12.1-2.el7.armv7hl (epel) ~libxfce4util.so.7 Error: Package: Thunar-1.6.3-2.el7.armv7hl (epel) Requires: libxfce4util.so.6 Available: libxfce4util-4.10.1-2.el7.armv7hl (epel) libxfce4util.so.6 Installing: libxfce4util-4.12.1-2.el7.armv7hl (epel) ~libxfce4util.so.7 Error: Package: xfce4-appfinder-4.10.1-2.el7.armv7hl (epel) Requires: libxfce4util.so.6 Available: libxfce4util-4.10.1-2.el7.armv7hl (epel) libxfce4util.so.6 Installing: libxfce4util-4.12.1-2.el7.armv7hl (epel) ~libxfce4util.so.7 Error: Package: xfce4-session-engines-4.10.1-8.el7.armv7hl (epel) Requires: libxfce4util.so.6 Available: libxfce4util-4.10.1-2.el7.armv7hl (epel) libxfce4util.so.6 Installing: libxfce4util-4.12.1-2.el7.armv7hl (epel) ~libxfce4util.so.7 Error: Package: thunar-volman-0.8.0-6.el7.armv7hl (epel) Requires: libxfce4util.so.6 Available: libxfce4util-4.10.1-2.el7.armv7hl (epel) libxfce4util.so.6 Installing: libxfce4util-4.12.1-2.el7.armv7hl (epel) ~libxfce4util.so.7 You could try using --skip-broken to work around the problem You could try running: rpm -Va --nofiles --nodigest I am going to have to put this on the back burner until someone can fix these. I do not have the skill sets to do it. But having Xcfe would be great for an armv7 SOC over Gnome!

5 33

Re: [Arm-dev] Community kernel 4.18.x
by Mark Verlinde 23 Oct '18

23 Oct '18

-----Oorspronkelijk bericht----- > Afzender:Mark Verlinde <mark(a)havak.nl> > Verstuurd: Woensdag 17 Oktober 2018 18:49 > Aan: Conversations around CentOS on ARM hardware <arm-dev(a)centos.org> > Onderwerp: Re: [Arm-dev] Community kernel 4.18.x > > > -----Oorspronkelijk bericht----- > > Afzender:Pablo Sebastián Greco <pablo(a)fliagreco.com.ar> > > Verstuurd: Woensdag 17 Oktober 2018 18:09 > > Aan: Conversations around CentOS on ARM hardware <arm-dev(a)centos.org> > > Onderwerp: Re: [Arm-dev] Community kernel 4.18.x > > > > > > > > El 17/10/18 a las 11:47, Mark Verlinde escribió: > > > Hi, > > > > > > Kernel 4.18.x brings nice features, CPUFreq for Allwinner H3 SOC's being among them. > > > > > > Therefore I had a go on updating to the community-kernel-latest version 4.18.14-200.el7 which failed; > > > Dracut could not (re)build a intitramfs-4.18.14-.. and (probably because of this) uInitrd-4.18.14-.. was missing to. > > > Unfortunately did not write the exact error message down; next time will do so. > > Does /boot have enough space?, I usually try to have 3 kernels at most. > > I've been using that kernel in 3 different SoCs without problems, > > including Bananapi M3 (same SoC as the cubieboard5) > > I do not doubt it had because it was the second kernel update done (i.e would be the 3rd installed) > > However : just successfully installed with a more conservative approach 4.17.17 > 4.18.9 > 4.18.14 > Side note is: this is on a home brew image without RPI bits and peaces, 786MB boot partition with ext4 file-system. > In time I will try again with default Centos-Generic image. > > i'm happy now:) > cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor > ondemand > > > > > > > BTW: Update to Kernel 4.17.17-200.el7 was successful on my oragngepi-plus2e > > > > > > In my own interest i'd really want to help to debug this; > > > but to be able to do this i need to understand in more detail how we boot Centos on our u-boot based SBC's. > > > I got the broad picture: u-boot >> extlinux bootloader >> kernel >> > > Correct > > > Getting lost however looking at extlinux/extlinux.conf , grub/grub.conf and not knowing the content of boot.scr > > You can ignore grub.conf and boot.scr, extlinux.conf is all you need > > Sorry to persist, I'd really like to understand in more detail; > Questions circling in my head (among others) are: > Why do we have a uInitrd _and_ initramfs ? > Why do we have a uimage _and_ vmlinux ? > Why does my odroid HC1 boot with a Fedora image but not with Centos (kernel 4.14.71 nor 4.17.17)? (despite specific USB3 patches are backported to 4.14) > > > > > > > Can somebody enlighten me on this, or show me some further reading. > > > > > > (I did search fedora's arm documentation without getting smarter..) > > > > > > Actually in response to "Cubieboard5 performance" > > > (https://lists.centos.org/pipermail/arm-dev/2018-October/003515.html) > > > > > > Just a heads up, update to 4.18.16 works flawless on the above described setup on a oragngepi-plus2e. Did not find the time to test it with a default img as I'm still in a fight with the odroid HC1. For the latter the good new is: it's not an issue with the kernel but with the initramfs.

5 8

Re: [Arm-dev] Community kernel 4.18.x
by Mark Verlinde 23 Oct '18

23 Oct '18

-----Oorspronkelijk bericht----- > Afzender:Mark Verlinde <mark(a)havak.nl> > Verstuurd: Maandag 22 Oktober 2018 14:27 > Aan: Conversations around CentOS on ARM hardware <arm-dev(a)centos.org> > Onderwerp: Re: [Arm-dev] Community kernel 4.18.x > > Thank you for the hints, it Led to a booting setup! > > Still need to fine tine, some customization may not be needed howerever this configuration led to the booting odroid HC1. > > Appended cpuidle.off=1 console=tty1 console=ttySAC2,115200n8 to the cmdline: > > # cat /boot/extlinux/extlinux.conf > # extlinux.conf generated by appliance-creator > ui menu.c32 > menu autoboot Welcome to Centos-Uboot-Community-kernel. Automatic boot in # second{,s}. Press a key for options. > menu title Centos-Uboot-Community-kernel Boot Options. > menu hidden > timeout 20 > totaltimeout 600 > > default=Centos-Uboot-Community-kernel (4.18.14-200.el7.armv7hl) > label CentOS Linux (4.18.16-200.el7.armv7hl) 7 (AltArch) > kernel /vmlinuz-4.18.16-200.el7.armv7hl > append cpuidle.off=1 console=tty1 console=ttySAC2,115200n8 ro root=UUID=a306d630-4957-4da0-8dd6-79f943147d61 LANG=en_US.UTF-8 > fdtdir /dtb-4.18.16-200.el7.armv7hl > initrd /initramfs-4.18.16-200.el7.armv7hl.img > > label Centos-Uboot-Community-kernel (4.18.14-200.el7.armv7hl) > kernel /vmlinuz-4.18.14-200.el7.armv7hl > append cpuidle.off=1 console=tty1 console=ttySAC2,115200n8 ro root=UUID=a306d630-4957-4da0-8dd6-79f943147d61 > fdtdir /dtb-4.18.14-200.el7.armv7hl > initrd /initramfs-4.18.14-200.el7.armv7hl.img > > In an effort to mimic the FC28 intenramfs added/omitted some modules and > added _all_ exynos and samsung divers loaded by the booting FC28 setup: > > #cat /etc/dracut.conf.d/odroid_hc1.conf > add_dracutmodules+="dm" > omit_dracutmodules+="network ifcfg" > add_drivers+="exynosdrm exynos_bus exynos_trng exynos_nocp exynos_rng phy_exynos_dp_video phy_exynos_mipi_video dwc3_exynos clk_exynos_audss i2c_exynos5 dw_mmc_exynos dw_mmc_pltfm dw_mmc phy_exynos_usb2 phy_exynos5_usbdrd pwm_samsung " > > (AGAIN) NOTE: some may be necessary others may be ommited.. Turns out one additional driver - dwc3 - is needed to boot the odroid HC1 with a rootfs residing on a SATA device. My best guess is dwc3_exynos actually depends on dwc3 and/or (one of) the dependencies of dwc3. #cat /etc/dracut.conf.d/odroid_hc1.conf add_dracutmodules+="dm" omit_dracutmodules+="network ifcfg" add_drivers+=" exynosdrm exynos_bus exynos_trng exynos_nocp exynos_rng phy_exynos_dp_video phy_exynos_mipi_video dwc3_exynos clk_exynos_audss i2c_exynos5 dw_mmc_exynos dw_mmc_pltfm dw_mmc phy_exynos_usb2 phy_exynos5_usbdrd pwm_samsung " add_drivers+=" dwc3"

2 1

Re: [Arm-dev] Community kernel 4.18.x
by Mark Verlinde 22 Oct '18

22 Oct '18

-----Oorspronkelijk bericht----- > Afzender:Pablo Sebastián Greco <pablo(a)fliagreco.com.ar> > Verstuurd: Zondag 21 Oktober 2018 23:58 > Aan: Conversations around CentOS on ARM hardware <arm-dev(a)centos.org> > Onderwerp: Re: [Arm-dev] Community kernel 4.18.x > > > El 21/10/18 a las 17:06, Mark Verlinde escribió: > > -----Oorspronkelijk bericht----- > >> Afzender:Mark Verlinde <mark(a)havak.nl> > >> Verstuurd: Woensdag 17 Oktober 2018 18:49 > >> Aan: Conversations around CentOS on ARM hardware <arm-dev(a)centos.org> > >> Onderwerp: Re: [Arm-dev] Community kernel 4.18.x > >> > >> > >> -----Oorspronkelijk bericht----- > >>> Afzender:Pablo Sebastián Greco <pablo(a)fliagreco.com.ar> > >>> Verstuurd: Woensdag 17 Oktober 2018 18:09 > >>> Aan: Conversations around CentOS on ARM hardware <arm-dev(a)centos.org> > >>> Onderwerp: Re: [Arm-dev] Community kernel 4.18.x > >>> > >>> > >>> > >>> El 17/10/18 a las 11:47, Mark Verlinde escribió: > >>>> Hi, > >>>> > >>>> Kernel 4.18.x brings nice features, CPUFreq for Allwinner H3 SOC's being among them. > >>>> > >>>> Therefore I had a go on updating to the community-kernel-latest version 4.18.14-200.el7 which failed; > >>>> Dracut could not (re)build a intitramfs-4.18.14-.. and (probably because of this) uInitrd-4.18.14-.. was missing to. > >>>> Unfortunately did not write the exact error message down; next time will do so. > >>> Does /boot have enough space?, I usually try to have 3 kernels at most. > >>> I've been using that kernel in 3 different SoCs without problems, > >>> including Bananapi M3 (same SoC as the cubieboard5) > >> I do not doubt it had because it was the second kernel update done (i.e would be the 3rd installed) > >> > >> However : just successfully installed with a more conservative approach 4.17.17 > 4.18.9 > 4.18.14 > >> Side note is: this is on a home brew image without RPI bits and peaces, 786MB boot partition with ext4 file-system. > >> In time I will try again with default Centos-Generic image. > >> > >> i'm happy now:) > >> cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor > >> ondemand > >> > >>>> BTW: Update to Kernel 4.17.17-200.el7 was successful on my oragngepi-plus2e > >>>> > >>>> In my own interest i'd really want to help to debug this; > >>>> but to be able to do this i need to understand in more detail how we boot Centos on our u-boot based SBC's. > >>>> I got the broad picture: u-boot >> extlinux bootloader >> kernel >> > >>> Correct > >>>> Getting lost however looking at extlinux/extlinux.conf , grub/grub.conf and not knowing the content of boot.scr > >>> You can ignore grub.conf and boot.scr, extlinux.conf is all you need > >> Sorry to persist, I'd really like to understand in more detail; > >> Questions circling in my head (among others) are: > >> Why do we have a uInitrd _and_ initramfs ? > >> Why do we have a uimage _and_ vmlinux ? > >> Why does my odroid HC1 boot with a Fedora image but not with Centos (kernel 4.14.71 nor 4.17.17)? (despite specific USB3 patches are backported to 4.14) > >> > >>>> Can somebody enlighten me on this, or show me some further reading. > >>>> > >>>> (I did search fedora's arm documentation without getting smarter..) > >>>> > >>>> Actually in response to "Cubieboard5 performance" > >>>> (https://lists.centos.org/pipermail/arm-dev/2018-October/003515.html) > >>>> > >>>> > > Just a heads up, update to 4.18.16 works flawless on the above described setup on a oragngepi-plus2e. > > Did not find the time to test it with a default img as I'm still in a fight with the odroid HC1. > > For the latter the good new is: it's not an issue with the kernel but with the initramfs. > > > Thanks for testing!! > WRT the odroid, do you have an error log? > I've had my share of boot problems with supported devices, and they end > up being solved by something like what I said in this mail > https://lists.centos.org/pipermail/arm-dev/2018-August/003399.html Thank you for the hints, it Led to a booting setup! Still need to fine tine, some customization may not be needed howerever this configuration led to the booting odroid HC1. Appended cpuidle.off=1 console=tty1 console=ttySAC2,115200n8 to the cmdline: # cat /boot/extlinux/extlinux.conf # extlinux.conf generated by appliance-creator ui menu.c32 menu autoboot Welcome to Centos-Uboot-Community-kernel. Automatic boot in # second{,s}. Press a key for options. menu title Centos-Uboot-Community-kernel Boot Options. menu hidden timeout 20 totaltimeout 600 default=Centos-Uboot-Community-kernel (4.18.14-200.el7.armv7hl) label CentOS Linux (4.18.16-200.el7.armv7hl) 7 (AltArch) kernel /vmlinuz-4.18.16-200.el7.armv7hl append cpuidle.off=1 console=tty1 console=ttySAC2,115200n8 ro root=UUID=a306d630-4957-4da0-8dd6-79f943147d61 LANG=en_US.UTF-8 fdtdir /dtb-4.18.16-200.el7.armv7hl initrd /initramfs-4.18.16-200.el7.armv7hl.img label Centos-Uboot-Community-kernel (4.18.14-200.el7.armv7hl) kernel /vmlinuz-4.18.14-200.el7.armv7hl append cpuidle.off=1 console=tty1 console=ttySAC2,115200n8 ro root=UUID=a306d630-4957-4da0-8dd6-79f943147d61 fdtdir /dtb-4.18.14-200.el7.armv7hl initrd /initramfs-4.18.14-200.el7.armv7hl.img In an effort to mimic the FC28 intenramfs added/omitted some modules and added _all_ exynos and samsung divers loaded by the booting FC28 setup: #cat /etc/dracut.conf.d/odroid_hc1.conf add_dracutmodules+="dm" omit_dracutmodules+="network ifcfg" add_drivers+="exynosdrm exynos_bus exynos_trng exynos_nocp exynos_rng phy_exynos_dp_video phy_exynos_mipi_video dwc3_exynos clk_exynos_audss i2c_exynos5 dw_mmc_exynos dw_mmc_pltfm dw_mmc phy_exynos_usb2 phy_exynos5_usbdrd pwm_samsung " (AGAIN) NOTE: some may be necessary others may be ommited.. > > > Pablo. > > _______________________________________________ > Arm-dev mailing list > Arm-dev(a)centos.org > https://lists.centos.org/mailman/listinfo/arm-dev >

3 3

SELinux warnings?
by Robert Moskowitz 18 Oct '18

18 Oct '18

I am assuming that none of these are a problem, other than they occur anytime you do anything with SELinux, like get an update: Updating : selinux-policy-targeted-3.13.1-192.el7_5.6.noarch 52/163 [ 864.238915] SELinux: Permission getrlimit in class process not defined in policy. [ 864.255572] SELinux: Permission map in class file not defined in policy. [ 864.271347] SELinux: Permission map in class dir not defined in policy. [ 864.287009] SELinux: Permission map in class lnk_file not defined in policy. [ 864.302952] SELinux: Permission map in class chr_file not defined in policy. [ 864.318679] SELinux: Permission map in class blk_file not defined in policy. [ 864.334072] SELinux: Permission map in class sock_file not defined in policy. [ 864.349585] SELinux: Permission map in class fifo_file not defined in policy. [ 864.364986] SELinux: Permission map in class socket not defined in policy. [ 864.380106] SELinux: Permission map in class tcp_socket not defined in policy. [ 864.395447] SELinux: Permission map in class udp_socket not defined in policy. [ 864.410657] SELinux: Permission map in class rawip_socket not defined in policy. [ 864.425984] SELinux: Permission map in class netlink_socket not defined in policy. [ 864.441531] SELinux: Permission map in class packet_socket not defined in policy. [ 864.457032] SELinux: Permission map in class key_socket not defined in policy. [ 864.472253] SELinux: Permission map in class unix_stream_socket not defined in policy. [ 864.488097] SELinux: Permission map in class unix_dgram_socket not defined in policy. [ 864.503933] SELinux: Permission map in class netlink_route_socket not defined in policy. [ 864.519918] SELinux: Permission map in class netlink_tcpdiag_socket not defined in policy. [ 864.536181] SELinux: Permission map in class netlink_nflog_socket not defined in policy. [ 864.552161] SELinux: Permission map in class netlink_xfrm_socket not defined in policy. [ 864.568076] SELinux: Permission map in class netlink_selinux_socket not defined in policy. [ 864.584204] SELinux: Permission map in class netlink_iscsi_socket not defined in policy. [ 864.600178] SELinux: Permission map in class netlink_audit_socket not defined in policy. [ 864.616037] SELinux: Permission map in class netlink_fib_lookup_socket not defined in policy. [ 864.632237] SELinux: Permission map in class netlink_connector_socket not defined in policy. [ 864.648096] SELinux: Permission map in class netlink_netfilter_socket not defined in policy. [ 864.663754] SELinux: Permission map in class netlink_dnrt_socket not defined in policy. [ 864.678752] SELinux: Permission map in class netlink_kobject_uevent_socket not defined in policy. [ 864.694514] SELinux: Permission map in class netlink_generic_socket not defined in policy. [ 864.709433] SELinux: Permission map in class netlink_scsitransport_socket not defined in policy. [ 864.724784] SELinux: Permission map in class netlink_rdma_socket not defined in policy. [ 864.739110] SELinux: Permission map in class netlink_crypto_socket not defined in policy. [ 864.753607] SELinux: Permission map in class appletalk_socket not defined in policy. [ 864.767591] SELinux: Permission map in class dccp_socket not defined in policy. [ 864.781189] SELinux: Permission map in class tun_socket not defined in policy. [ 864.794607] SELinux: Class sctp_socket not defined in policy. [ 864.806570] SELinux: Class icmp_socket not defined in policy. [ 864.818225] SELinux: Class ax25_socket not defined in policy. [ 864.829649] SELinux: Class ipx_socket not defined in policy. [ 864.840752] SELinux: Class netrom_socket not defined in policy. [ 864.851923] SELinux: Class atmpvc_socket not defined in policy. [ 864.862911] SELinux: Class x25_socket not defined in policy. [ 864.873393] SELinux: Class rose_socket not defined in policy. [ 864.883767] SELinux: Class decnet_socket not defined in policy. [ 864.894133] SELinux: Class atmsvc_socket not defined in policy. [ 864.904385] SELinux: Class rds_socket not defined in policy. [ 864.914444] SELinux: Class irda_socket not defined in policy. [ 864.924608] SELinux: Class pppox_socket not defined in policy. [ 864.934862] SELinux: Class llc_socket not defined in policy. [ 864.944876] SELinux: Class can_socket not defined in policy. [ 864.954877] SELinux: Class tipc_socket not defined in policy. [ 864.964974] SELinux: Class bluetooth_socket not defined in policy. [ 864.975659] SELinux: Class iucv_socket not defined in policy. [ 864.985843] SELinux: Class rxrpc_socket not defined in policy. [ 864.996237] SELinux: Class isdn_socket not defined in policy. [ 865.006492] SELinux: Class phonet_socket not defined in policy. [ 865.017007] SELinux: Class ieee802154_socket not defined in policy. [ 865.027974] SELinux: Class caif_socket not defined in policy. [ 865.038332] SELinux: Class alg_socket not defined in policy. [ 865.048641] SELinux: Class nfc_socket not defined in policy. [ 865.058824] SELinux: Class vsock_socket not defined in policy. [ 865.069225] SELinux: Class kcm_socket not defined in policy. [ 865.079405] SELinux: Class qipcrtr_socket not defined in policy. [ 865.090009] SELinux: Class smc_socket not defined in policy. [ 865.100225] SELinux: the above unknown classes and permissions will be allowed [ 870.760142] SELinux: Context unconfined_u:unconfined_r:sandbox_t:s0-s0:c0.c1023 became invalid (unmapped). [ 871.187593] SELinux: Context system_u:unconfined_r:sandbox_t:s0-s0:c0.c1023 became invalid (unmapped). [ 872.635092] SELinux: policy capability network_peer_controls=1 [ 872.645801] SELinux: policy capability open_perms=1 [ 872.655404] SELinux: policy capability extended_socket_class=0 [ 872.666023] SELinux: policy capability always_check_network=0 [ 872.676507] SELinux: policy capability cgroup_seclabel=1 [ 872.686511] SELinux: policy capability nnp_nosuid_transition=1 that is quite an extensive list...

1 0

Re: [Arm-dev] Community kernel 4.18.x
by Mark Verlinde 17 Oct '18

17 Oct '18

-----Oorspronkelijk bericht----- > Afzender:Pablo Sebastián Greco <pablo(a)fliagreco.com.ar> > Verstuurd: Woensdag 17 Oktober 2018 18:09 > Aan: Conversations around CentOS on ARM hardware <arm-dev(a)centos.org> > Onderwerp: Re: [Arm-dev] Community kernel 4.18.x > > > > El 17/10/18 a las 11:47, Mark Verlinde escribió: > > Hi, > > > > Kernel 4.18.x brings nice features, CPUFreq for Allwinner H3 SOC's being among them. > > > > Therefore I had a go on updating to the community-kernel-latest version 4.18.14-200.el7 which failed; > > Dracut could not (re)build a intitramfs-4.18.14-.. and (probably because of this) uInitrd-4.18.14-.. was missing to. > > Unfortunately did not write the exact error message down; next time will do so. > Does /boot have enough space?, I usually try to have 3 kernels at most. > I've been using that kernel in 3 different SoCs without problems, > including Bananapi M3 (same SoC as the cubieboard5) I do not doubt it had because it was the second kernel update done (i.e would be the 3rd installed) However : just successfully installed with a more conservative approach 4.17.17 > 4.18.9 > 4.18.14 Side note is: this is on a home brew image without RPI bits and peaces, 786MB boot partition with ext4 file-system. In time I will try again with default Centos-Generic image. i'm happy now:) cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor ondemand > > > > BTW: Update to Kernel 4.17.17-200.el7 was successful on my oragngepi-plus2e > > > > In my own interest i'd really want to help to debug this; > > but to be able to do this i need to understand in more detail how we boot Centos on our u-boot based SBC's. > > I got the broad picture: u-boot >> extlinux bootloader >> kernel >> > Correct > > Getting lost however looking at extlinux/extlinux.conf , grub/grub.conf and not knowing the content of boot.scr > You can ignore grub.conf and boot.scr, extlinux.conf is all you need Sorry to persist, I'd really like to understand in more detail; Questions circling in my head (among others) are: Why do we have a uInitrd _and_ initramfs ? Why do we have a uimage _and_ vmlinux ? Why does my odroid HC1 boot with a Fedora image but not with Centos (kernel 4.14.71 nor 4.17.17)? (despite specific USB3 patches are backported to 4.14) > > > > Can somebody enlighten me on this, or show me some further reading. > > > > (I did search fedora's arm documentation without getting smarter..) > > > > Actually in response to "Cubieboard5 performance" > > (https://lists.centos.org/pipermail/arm-dev/2018-October/003515.html) > > > > > > _______________________________________________ > > Arm-dev mailing list > > Arm-dev(a)centos.org > > https://lists.centos.org/mailman/listinfo/arm-dev > > > _______________________________________________ > Arm-dev mailing list > Arm-dev(a)centos.org > https://lists.centos.org/mailman/listinfo/arm-dev >

1 0

Community kernel 4.18.x
by Mark Verlinde 17 Oct '18

17 Oct '18

Hi, Kernel 4.18.x brings nice features, CPUFreq for Allwinner H3 SOC's being among them. Therefore I had a go on updating to the community-kernel-latest version 4.18.14-200.el7 which failed; Dracut could not (re)build a intitramfs-4.18.14-.. and (probably because of this) uInitrd-4.18.14-.. was missing to. Unfortunately did not write the exact error message down; next time will do so. BTW: Update to Kernel 4.17.17-200.el7 was successful on my oragngepi-plus2e In my own interest i'd really want to help to debug this; but to be able to do this i need to understand in more detail how we boot Centos on our u-boot based SBC's. I got the broad picture: u-boot >> extlinux bootloader >> kernel >> Getting lost however looking at extlinux/extlinux.conf , grub/grub.conf and not knowing the content of boot.scr Can somebody enlighten me on this, or show me some further reading. (I did search fedora's arm documentation without getting smarter..) Actually in response to "Cubieboard5 performance" (https://lists.centos.org/pipermail/arm-dev/2018-October/003515.html)

2 1

Qemu Arm Emulator in CentOS host(x86)
by karthi mohanraj 17 Oct '18

17 Oct '18

Hi All, I’m trying to install fedora on QEMU emulator for the ARM. I have followed this documentation using virt-manager in this link: https://fedoraproject.org/wiki/Architectures/ARM/F26/Installation#For_Versa… <https://fedoraproject.org/wiki/Architectures/ARM/F26/Installation#For_Versa…> I was getting USB PCI device error as below. 2018-10-17 12:18:35.914+0000: starting up libvirt version: 3.9.0, package: 14.el7_5.8 (CentOS BuildSystem <http://bugs.centos.org <http://bugs.centos.org/>>, 2018-09-26-18:30:11, x86-01.bsys.centos.org <http://x86-01.bsys.centos.org/>), qemu version: 2.0.0, hostname: localhost.localdomain LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin QEMU_AUDIO_DRV=none /usr/bin/qemu-system-arm -name guest=fedora22-arm,debug-threads=on -S -machine virt,accel=tcg,usb=off,dump-guest-core=off -m 2048 -realtime mlock=off -smp 1,sockets=1,cores=1,threads=1 -uuid fc26e72b-3aaa-407b-a4fa-8896bbb8c580 -display none -no-user-config -nodefaults -chardev socket,id=charmonitor,path=/var/lib/libvirt/qemu/domain-30-fedora22-arm/monitor.sock,server,nowait -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc -no-shutdown -boot strict=on -kernel /var/lib/libvirt/images/arm/vmlinuz-4.8.6-300.fc25.armv7hl -initrd /var/lib/libvirt/images/arm/initramfs-4.8.6-300.fc25.armv7hl.img -append 'console=ttyAMA0 rw root=LABEL=_/ rootwait' -device ich9-usb-ehci1,id=usb -device ich9-usb-uhci1,masterbus=usb.0,firstport=0 -device ich9-usb-uhci2,masterbus=usb.0,firstport=2 -device ich9-usb-uhci3,masterbus=usb.0,firstport=4 -device virtio-scsi-device,id=scsi0 -device virtio-serial-device,id=virtio-serial0 -drive file=/var/lib/libvirt/images/arm/Fedora-Minimal-armhfp-25-1.3-sda.raw,format=raw,if=none,id=drive-scsi0-0-0-0 -device scsi-hd,bus=scsi0.0,channel=0,scsi-id=0,lun=0,drive=drive-scsi0-0-0-0,id=scsi0-0-0-0,bootindex=1 -netdev tap,fd=27,id=hostnet0 -device virtio-net-device,netdev=hostnet0,id=net0,mac=52:54:00:46:70:cb -chardev pty,id=charserial0 -serial chardev:charserial0 -chardev socket,id=charchannel0,path=/var/lib/libvirt/qemu/channel/target/domain-30-fedora22-arm/org.qemu.guest_agent.0,server,nowait -device virtserialport,bus=virtio-serial0.0,nr=1,chardev=charchannel0,id=channel0,name=org.qemu.guest_agent.0 -object rng-random,id=objrng0,filename=/dev/urandom -device virtio-rng-device,rng=objrng0,id=rng0 -msg timestamp=on (process:12978): GLib-WARNING **: gmem.c:483: custom memory allocation vtable not supported char device redirected to /dev/pts/2 (label charserial0) 2018-10-17T12:18:36.052249Z qemu-system-arm: -device ich9-usb-ehci1,id=usb: No 'PCI' bus found for device 'ich9-usb-ehci1' 2018-10-17 12:18:36.282+0000: shutting down, reason=failed Thanks, Karthi

1 0

wlan0: link is not ready
by Robert Moskowitz 15 Oct '18

15 Oct '18

On my Cubietruck, I keep getting: [ 716.571958] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready How can I stop this, as I am not using the wifi? thanks

2 3

Re: [Arm-dev] Cubieboard5 performance
by Pablo Sebastián Greco 13 Oct '18

13 Oct '18

El 9/10/18 a las 14:41, Andreas Reschke escribió: > Am 21.08.18 um 12:14 schrieb arm_ml(a)rirasoft.de: >> El 20/8/18 a las 15:57, Iván Chavero escribió: >>>>>>> >>>>>> I had to apply some patches to the sunxi kernel that were pending >>>>>> to >>>>>> be merged upstream in order to have SMP and HDMI running. >>>>>> I also had to patch the kernel's dtb for the HDMI running (i've >>>>>> been >>>>>> to lazy to send the patch to the kernel list), if you need it i >>>>>> can >>>>>> upload this code to github or i can pack you a kernel+initramfs >>>>>> for >>>>>> you to use it with your cubieboard5 >>>>>> Cheers, >>>>>> Ivan >>>>>> >>>>> Ivan, that kernel has both smp and your patch included ;-). >>>>> Either way, I'll try to update to 4.18 in a few days. >>>>> >>>> Let me know if i can help, i have my cubieboard5 clusters running if >>>> there's need for testing and other stuff. >>>> >>> Iván, I didn't have time today to build 4.18 today, but I pushed >>> 4.17.17, and it should appear here >>> https://armv7.dev.centos.org/repodir/community-kernel-latest/kernel/ >>> between today and tomorrow. >>> It has smp for a83t (bananapi-m3 and cubieborard5) >>> Your hdmi patch for cubieboard5. >>> Some other fixes for R40. >>> >>> It would be great if you could test it. >> >> Hello Pablo, Iván, >> >> thank you very much for your work. With this kernel everything seems >> to work: 8CPUs, Ethernet, USBtoSATA, not tested: HDMI >> >> I'm now in the process to copy everything from SD-Card to the onboard >> EMMC, so I can boot from it. >> >> >> With your work I have a powerfull device. >> >> Thanks >> Andreas >> >> (thinking what can I do with the cubietruck plus: Time-Server (DCF77) >> ? DNS-Server? Pi-Hole? ...?) >> >> PS: sometimes I get these messages in the syslog: >> [ 286.918766] usb usb2-port1: attempt power cycle >> [ 287.295335] ehci-platform 1c1b000.usb: port 1 reset error -110 >> [ 288.193314] ehci-platform 1c1b000.usb: port 1 reset error -110 >> [ 289.033322] ehci-platform 1c1b000.usb: port 1 reset error -110 >> [ 289.873370] ehci-platform 1c1b000.usb: port 1 reset error -110 >> [ 290.713280] ehci-platform 1c1b000.usb: port 1 reset error -110 >> [ 291.344084] usb usb2-port1: Cannot enable. Maybe the USB cable is >> bad? >> >> No USB-Device plugged in. >> _______________________________________________ >> Arm-dev mailing list >> Arm-dev(a)centos.org >> https://lists.centos.org/mailman/listinfo/arm-dev > > Hello Pablo, > > I'm getting more and more failures at the syslog concerning USB (see > above). Is there a solution? Kernel-Update? Sorry, no news about that, have you tested the latest fedora kernel (4.19.x based)?, just to see if anything changes? > > BTW: did the HDD-RAID board work with CentOS and this kernel > (4.17.17-200.el7.armv7hl) ? Haven't tested, but if it doesn't work, let me know and we'll investigate. BTW, I'll be building a test 4.18.14, it should appear tomorrow. (takes a while to build :-)) > > Thanks > > Andreas > > > Pablo

1 0

Re: [Arm-dev] Using zram-swap
by Mark Verlinde 12 Oct '18

12 Oct '18

Being the hidden source of this discussion it's only fair to try to explain the origin. ;) First objective was to get rid of the swap partition: - we do not know upfront the amount of ram and the appropriate swap size for a particular SBC/system - being a hardware guy it does not make sense to me to provide a sap partition while in realty data is written all over the medium (SD card) to level out wear. NOTE this is from my low level hardware perspective there may be compelling arguments from a (high level) system perspective. And yes the most commly offerd 1G is a little comes a little to short for us: We would love to offer a mail-server with antivirus capabilities utilizing Rspamd and ClamAV, and unfortunately CalmAV has a memory foot print of about 650 MB. You may argue this is not feasible. My vredo is you can't be sure before you tried and and squeeze every drop of performance out of out out ar boards. That's why compressed swap-space in memory came up the table. Reading on it got me convinced it's a solution wort to persuit and we are moving to it until proven faulty. here the kickstart files we currently using to make images: https://github.com/markVnl/nethserver-createimg/tree/ks_wip/ks Would love to benchmark; ie craft a test to get some numbers on the table Unfortunaly my knowlage comes to sort to provode this. What can say is my compile time of big demanding packages on a odroid HC1 drop about 20-25%; especially "linking" speeds up remarkable! How can we craft a test? -----Oorspronkelijk bericht----- Afzender: Robert Moskowitz<rgm(a)htt-consult.com> Verstuurd: Vrijdag 12 Oktober 2018 19:36 Aan: Conversations around CentOS on ARM hardware <arm-dev(a)centos.org>; Fred Gleason <fredg(a)paravelsystems.com> Onderwerp: Re: [Arm-dev] Using zram-swap On 10/12/18 1:24 PM, Fred Gleason wrote: On Oct 12, 2018, at 13:15, Gordan Bobic <gordan(a)redsleeve.org <mailto:gordan@redsleeve.org>> wrote: It depends on what you are running on that system. If you have 1GB of RAM and want to run a desktop environment web browser, that'll go pretty poorly without swap. So how does putting swap on RAM help that situation? You’re not increasing the overall memory space by doing so, but merely partitioning it into two different buckets (while also adding the complexity and performance overhead of accessing the part that is now ‘swap’). See Gordon's earlier comment on Firefox's bad memory allocation. Zram is a big help there; I see this on my F29 system. # free total used free shared buff/cache available Mem: 1022148 655520 107204 22284 259424 327636 Swap: 487304 253952 233352 # zramctl NAME ALGORITHM DISKSIZE DATA COMPR TOTAL STREAMS MOUNTPOINT /dev/zram0 lz4 475.9M 246.4M 109.1M 114.2M 2 [SWAP] My mail server can be all over the place on memory and swap usage (it uses real swap file on sata) depending on what all the scanning software is doing and who is using roundcubemail. So YMMV, and this should not hurt if your system does not need to swap. # free total used free shared buff/cache available Mem: 2060396 189332 1189972 26432 681092 1766392 Swap: 982420 0 982420 # zramctl NAME ALGORITHM DISKSIZE DATA COMPR TOTAL STREAMS MOUNTPOINT /dev/zram0 lz4 959.4M 4K 64B 4K 2 <swap> I may switch this setup to a Cubeboard2 (1GB mem) from the Cubietruck. I have a number of C2 and only a couple CT, so if it does not need the memory... Cheers! |----------------------------------------------------------------------| | Frederick F. Gleason, Jr. | Chief Developer | | | Paravel Systems | |----------------------------------------------------------------------| | A room without books is like a body without a soul. | | -- Cicero | |----------------------------------------------------------------------| _______________________________________________ Arm-dev mailing list Arm-dev(a)centos.org <mailto:Arm-dev@centos.org> https://lists.centos.org/mailman/listinfo/arm-dev <https://lists.centos.org/mailman/listinfo/arm-dev> _______________________________________________ Arm-dev mailing list Arm-dev(a)centos.org https://lists.centos.org/mailman/listinfo/arm-dev

2 1

Using zram-swap
by Robert Moskowitz 12 Oct '18

12 Oct '18

I have very recently learned about zram-swap. Fedora-29-arm is using it. It supposedly makes a considerable difference even over HD swap. Since many (potential) users of Centos-arm do not have a ready HD, perhaps we can switch the image from a physical partition to zram? Or at least instructions to change to this on the wiki. Perhaps target the 7.6 image? (is there any information as to when 7.6 may be available?) thanks

4 18

ffmpeg for CentOS 7 armv7hl
by arm_ml＠rirasoft.de 08 Oct '18

08 Oct '18

Hello I search for ffmpeg for CentOS 7 ARM and can't find anything. Only for Fedora. But all RPMs from Fedora I tried have missing dependency. I want to build a converter for my Car-Audio-Player that only accept MP3 and I have over 2TB of FLAC. Thanks Andreas

4 3

Arm Development Using QEMU
by karthi mohanraj 04 Oct '18

04 Oct '18

Hi Marcin & Johnny, Thanks for the reply. My requirement is to setup a VM with aarch64(64 bit arm architecture) on x86_64 host. I have tried installing through virt-manager but i’m getting crash when starting the VM. I have attached the error below. In this, centos-7.2.img is built using virt-builder with lentos-7.2 for the aarch64. ##qemu-system-arm -m 1024 -machine vexpress-a9 -drive file=centos-7.2.img,format=raw,if=ide (process:12672): GLib-WARNING **: gmem.c:483: custom memory allocation vtable not supported pulseaudio: set_sink_input_volume() failed pulseaudio: Reason: Invalid argument pulseaudio: set_sink_input_mute() failed pulseaudio: Reason: Invalid argument qemu: fatal: Trying to execute code outside RAM or ROM at 0x04000000 R00=00000000 R01=00000000 R02=00000000 R03=00000000 R04=00000000 R05=00000000 R06=00000000 R07=00000000 R08=00000000 R09=00000000 R10=00000000 R11=00000000 R12=00000000 R13=00000000 R14=00000000 R15=04000000 PSR=400001d3 -Z-- A svc32 s00=00000000 s01=00000000 d00=0000000000000000 s02=00000000 s03=00000000 d01=0000000000000000 s04=00000000 s05=00000000 d02=0000000000000000 s06=00000000 s07=00000000 d03=0000000000000000 s08=00000000 s09=00000000 d04=0000000000000000 s10=00000000 s11=00000000 d05=0000000000000000 s12=00000000 s13=00000000 d06=0000000000000000 s14=00000000 s15=00000000 d07=0000000000000000 if i use virtio for the drive interface. i’m getting below error. ##qemu-system-arm -m 1024 -machine vexpress-a9 -drive file=centos-7.2.img,format=raw,if=virtio (process:14847): GLib-WARNING **: gmem.c:483: custom memory allocation vtable not supported pulseaudio: set_sink_input_volume() failed pulseaudio: Reason: Invalid argument pulseaudio: set_sink_input_mute() failed pulseaudio: Reason: Invalid argument qemu-system-arm: -drive file=centos-7.2.img,format=raw,if=virtio: No 'PCI' bus found for device 'virtio-blk-pci' Regards, karthi On 09/30/2018 08:08 AM, Marcin Juszkiewicz wrote: > W dniu 28.09.2018 o 12:42, karthi mohanraj pisze: >> Hi All, >> >> I’m a newbie to arm architecture. I want to setup a development VM >> with ARMv8 board using QEMU/KVM on CentOS based host. Is there any >> documentation available to try out? >> >> Or is there any other easy way to achieve this as i need to have a >> Development setup for CentOS? >> >> Please suggest me if any other better way to achieve this. > Easiest? Install AAVMF (uefi firmware for armv8), virt-manager and > qemu-system-arm. > > Run Virt-manager, choose 'new VM', open architecture chooser, select > aarch64. Give it hdd, install iso and go. You can also follow this guide to install a 32 bit VM on an aarch64 host, if 32 bit arm is what you want to dev on: https://www.arrfab.net/posts/2017/Sep/29/using-centos-7-armhfp-vm-on-centos… <https://www.arrfab.net/posts/2017/Sep/29/using-centos-7-armhfp-vm-on-centos…> Thanks, Johnny Hughes

2 1

Arm Development Using QEMU
by karthi mohanraj 02 Oct '18

02 Oct '18

Hi All, I’m a newbie to arm architecture. I want to setup a development VM with ARMv8 board using QEMU/KVM on CentOS based host. Is there any documentation available to try out? Or is there any other easy way to achieve this as i need to have a Development setup for CentOS? Please suggest me if any other better way to achieve this. Thanks, Karthi

3 2

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

Arm-dev October 2018