trap keypress on bootup, during initrd
Hi all, I am developing an embedded OS, and at times, i would to boot the OS different, lets say in a less secure mode when a user press or hold down, CTRL + ESC, for example. I am not sure what is the best way to do it. please advise, Thanks Toan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: ttyUSB0 and GPS
Add this udev rule to /etc/udev/rules.d/50-gps.rules BUS=="usb", KERNEL="ttyUSB*", NAME="ttyUSB0" On Fri, Dec 11, 2009 at 10:26 AM, Jon Dowland > wrote: > On Sat, Dec 12, 2009 at 12:12:44AM +1300, Chris Bannister > wrote: > > Set up a udev rule for that particular device. > > I'd second this: hard-coding to ttyUSB8 or similar could > potentially fail in the unlikely event you use up to that > number prior to plugging in the GPS. YOu can use udev to > provide a unique name, safe from clashes, such as > /dev/my_gps. > > -BEGIN PGP SIGNATURE- > Version: GnuPG v1.4.10 (GNU/Linux) > > iEYEARECAAYFAksiZLwACgkQFotOcXAy8jghCgCgu+G7WBef1WuZIsQ+q+7U4vv9 > jcMAnRrExSNrKVx9EdCIRi5Soy6LHSka > =FcpK > -END PGP SIGNATURE- > >
Re: ttyUSB0 and GPS
sorry, typo on the last email.. BUS=="usb", KERNEL=="ttyUSB*", NAME="ttyUSB0" On Fri, Dec 11, 2009 at 12:06 PM, Toan Pham wrote: > > Add this udev rule to /etc/udev/rules.d/50-gps.rules > > BUS=="usb", KERNEL="ttyUSB*", NAME="ttyUSB0" > > On Fri, Dec 11, 2009 at 10:26 AM, Jon Dowland > > > wrote: > >> On Sat, Dec 12, 2009 at 12:12:44AM +1300, Chris Bannister >> wrote: >> > Set up a udev rule for that particular device. >> >> I'd second this: hard-coding to ttyUSB8 or similar could >> potentially fail in the unlikely event you use up to that >> number prior to plugging in the GPS. YOu can use udev to >> provide a unique name, safe from clashes, such as >> /dev/my_gps. >> >> -BEGIN PGP SIGNATURE- >> Version: GnuPG v1.4.10 (GNU/Linux) >> >> iEYEARECAAYFAksiZLwACgkQFotOcXAy8jghCgCgu+G7WBef1WuZIsQ+q+7U4vv9 >> jcMAnRrExSNrKVx9EdCIRi5Soy6LHSka >> =FcpK >> -END PGP SIGNATURE- >> >> >
Re: building initramfs also as root filesystem problem
Hi all, I got this resolved, the trick when building ramfs as rootfs os is to: 1. Use busybox (symlink all neccessary tools and package all busybox dependencies with ramfs) 2. bootup into busybox bash, which should be /bin/sh symlinked to /bin/busybox 3. You may get nothing (no stardard ouput) when you try to run an executable, or bash error no such file found (but yet it is there). 4. Try to execute the executable with /lib/ld-2.5.90.so . Using this method, the linker actually tells you what's wrong with the executable if it can not be executed. -toan On 11/24/10, Toan Pham wrote: > Hi, > > I am building a linux os by incorporating in-kernel initramfs, which > also runs as root-filesystem. > The result of this work would be booting/distributing linux with only > one file, which is the kernel + linked in initramfs + rootfs. > I am having a problem booting up the os when everything is linked in. > Here is the two errors I've been getting when: > > 1. root-filesystem is prepared in initramfs and linked in with kernel > (total size is about 5 megs) - > a. Kernel Panic - not syncing: Attempted to kill init! > b. Also tried to boot directly to bash instead (rdinit=/bin/init > and/or init=/bin/sh), same error. > > 2. Moved root-filesystem to boot-drive (/rootfs), boot kernel with > linked in initramfs (total size about 2 megs) - > a. Boot OK to bash (still within initramfs shell). > b. Manually and successfully mount: rootfs on boot-drive located > at /rootfs to /tmp/rootfs as vfat filesystem-type > c. I can cat all mounted files at /tmp/rootfs. For example, cat > /tmp/rootfs/bin/clear works find. > d. However, I can not execute that file, for example execute > /tmp/rootfs/bin/clear. Bash would display an error like clear: no > such file or directory. > e. Permission setting is correct of-course. > > > This is really an effort of building linux distribution from scratch > and compact it into one single file, and not related to Debian at all. > If anyone has experience on this, please give me some advices. > > > Thank you. > > -toan > -- Toan Pham Atronix Engineering, Inc. UPS Advanced Technology Center (321) 277-3336 -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/aanlktint-ob2ovoeu+ecgdo2m=shzgxdc+ttphdgv...@mail.gmail.com
Re: sleeping the system vs hibernate or suspend
Just a side note. if you want to do S4, make sure that you have swap enabled. And that your swap partition is recommended to be twice the size of your physical ram. On some systems with nvidia or ATI video cards wont allow system to go to suspend/hibernation. You can try to unload xorg server unload video driver module, and then suspend/hibernate. also, make sure ACPI is enabled in your bios. -toan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/aanlktimcc6-jx5cogeg=gd_b+dnfy4q+syps8nrza...@mail.gmail.com
Re: sleeping the system vs hibernate or suspend
> Twice? Why? I thought it should be at least the same size :-) > > I think nowadays you can even hibernate with no swap partition at all but > using a swap file. The state of a running system is not just RAM. It is what is in the current swapped filesystem + ram content + video ram. yes you can use swap file or swap partition. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/aanlkti=022spjcx8erf2-vgbapwvw2wkpme7hbh8c...@mail.gmail.com
Re: sleeping the system vs hibernate or suspend
> "Suspend to memory (S3) works on my system under both 2.6.32-5-686 and 2.6.36-2.dmz.5-liquorix-686 kernels. Hibernate to drive (S4) doesn't even try under 2.6.36-2.dmz.5-liquorix-686, as expected. However, Hibernate executes under 2.6.32-5-686, only to abort and return to a running system. There is no hibernate option on my GDM & Gnome menus or in Gnome Power Manager." On my system running kernel 2.6.32-21, both S3 and S4 work perfectly. However, on my other system running kernel 2.6.36-x, both S3 and S4 failed, giving results similar to your log. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/aanlkti=4wzxtisq6_o5txfsmuzwu2ctxgtnt=405u...@mail.gmail.com
Re: sleeping the system vs hibernate or suspend
camaleon, FYI, that is a recommended setting. Optimal values depends on multiple factors such as: 1. total amount of physical ram, 2. percentage of actual utilized ram, 3. average cached size to swap partition etc. -toan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/aanlktimcobrg34-8h8lpz-1ryo5esh8=8ek9_uyr5...@mail.gmail.com
Re: sleeping the system vs hibernate or suspend
Here is the ref. Please read paragraph at section "how much swap do i need?". This article is pretty good for those who want to understand swap space. https://help.ubuntu.com/community/SwapFaq Sorry that i reference a Ubuntu source on this Debian mailing list. -toan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/aanlktimkhxe_to963e-xwgttcskznmm2=7kel_5an...@mail.gmail.com
Re: Broadcom TG3 network drops, cannot recover without reboot
Justin, Have you found a solution to the NIC issue yet? FYI, I was working with Broadcom's test team but they pretty much dropped the ball on me So it is all up to us to find a solution! Please share if you know how to get the NIC not to drop out intermitently. thank you, TP On Thu, May 28, 2015 at 3:48 AM, Justin Catterall wrote: > > > On 27 May 2015, at 17:06, Toan Pham wrote: > > > > Justin, > > > > > > I've observed a similar symptom on the bcm5762 chip, not the 5720, and > > not sure if the bugs they are related. I've filed a bug report > > (https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1447664), and > > actively working with Broadcom's engineering team to get this bug > > resolved. They are running multiple test cases but could not get this > > bug to surface in a short amount of time. > > > >> In the mean time I've made a bash-script to rmmod and modprobe as > appropriate. I'll set a cron job to ping a couple of other servers on the > LAN and execute the script and restart networking should the pings fail. > > > > This is a patch, not a fix. Have you tested on kernel 4.0? > > I've not tested with 4.0. This machine needs to be rock solid, it will be > Debian stable all the way, 4.x will only get on this machine when stable is > updated to that kernel. > > > > Alternatively, try to get yourself an Intel NIC that works with the igb > >> driver (don't get an Intel NIC that needs the e1000e driver) to replace > >> the hardlock-prone bcm5720 + tg3 combination. > > > > I ended up with an intel NIC instead, but with the e1000e driver. > > What's wrong with the e1000e driver by the way, please update. Thank > > you > > You'll have to hope Henrique is still following this, I don't have an > answer. > > Just thinking out-loud here: I've got an identical server with FreeNAS > installed, that's never disappeared off the network. Also I've had a server > for about 5 years at home with the same on-board NIC, (different mobo), > that's never locked up either, but I was able to force it to lock up with > "/etc/init.d/networking restart". It seems that, at home at least, I've > been fortunate so far. WRT to the FreeNAS, I have no idea how that's > driving the NIC, listing the modules there doesn't show anything I > recognise, and I've minimal experience with *BSD. > > -- > Justin C, by the sea. > > -- > To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org > with a subject of "unsubscribe". Trouble? Contact > listmas...@lists.debian.org > Archive: > https://lists.debian.org/a1e820dd-50c4-4708-8816-f9506ad1c...@masonsmusic.co.uk > >
Re: Broadcom TG3 network drops, cannot recover without reboot
Justin, I've observed a similar symptom on the bcm5762 chip, not the 5720, and not sure if the bugs they are related. I've filed a bug report (https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1447664), and actively working with Broadcom's engineering team to get this bug resolved. They are running multiple test cases but could not get this bug to surface in a short amount of time. > In the mean time I've made a bash-script to rmmod and modprobe as > appropriate. I'll set a cron job to ping a couple of other servers on the LAN > and execute the script and restart networking should the pings fail. This is a patch, not a fix. Have you tested on kernel 4.0? > Alternatively, try to get yourself an Intel NIC that works with the igb > driver (don't get an Intel NIC that needs the e1000e driver) to replace > the hardlock-prone bcm5720 + tg3 combination. I ended up with an intel NIC instead, but with the e1000e driver. What's wrong with the e1000e driver by the way, please update. Thank you -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/cagnmleofs1aackvgktfhmes+e1l4z3d3pzvo1et-txqzrne...@mail.gmail.com
building initramfs also as root filesystem problem
Hi, I am building a linux os by incorporating in-kernel initramfs, which also runs as root-filesystem. The result of this work would be booting/distributing linux with only one file, which is the kernel + linked in initramfs + rootfs. I am having a problem booting up the os when everything is linked in. Here is the two errors I've been getting when: 1. root-filesystem is prepared in initramfs and linked in with kernel (total size is about 5 megs) - a. Kernel Panic - not syncing: Attempted to kill init! b. Also tried to boot directly to bash instead (rdinit=/bin/init and/or init=/bin/sh), same error. 2. Moved root-filesystem to boot-drive (/rootfs), boot kernel with linked in initramfs (total size about 2 megs) - a. Boot OK to bash (still within initramfs shell). b. Manually and successfully mount: rootfs on boot-drive located at /rootfs to /tmp/rootfs as vfat filesystem-type c. I can cat all mounted files at /tmp/rootfs. For example, cat /tmp/rootfs/bin/clear works find. d. However, I can not execute that file, for example execute /tmp/rootfs/bin/clear. Bash would display an error like clear: no such file or directory. e. Permission setting is correct of-course. This is really an effort of building linux distribution from scratch and compact it into one single file, and not related to Debian at all. If anyone has experience on this, please give me some advices. Thank you. -toan -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/aanlktikkj8blswjms2bzswifoodv6hcfmvzrpc6ef...@mail.gmail.com
