Description of problem: On boot up "pnpacpi: exceeded the max number of mem resources" Version-Release number of selected component (if applicable): Current kernel 2.6.24.3-12 How reproducible: Absolutely - everytime you boot you see the alert Steps to Reproduce: 1. Reboot 2. Observe the warning 3. Actual results: 2.6.24.3-12 Expected results: No error or warning Additional info:
I see this too, on a Dell Inspiron 9300 laptop if it makes any difference.
I also get this, and what's more the machine (an IBM thinkpad X30, 26724XG) subsequently hangs at "Starting udev:". Booting kernel-2.6.23.15-137.fc8, the machine runs fine. Clearly this is a big issue for me.
I get this message too: pnpacpi: exceeded the max nubmer of mem resources: 12 Uncompressing Linux... OK, booting the kernel. But everything works fine. My mainbord is: ASUS A8N SLI Premium. I get this message since the update to kernel 2.6.24.3-12.fc8.
The same message for me, and I have no network with 2.6.24.3-12.fc8. 2.6.23.15-137.fc8 works fine.
The same message for me, too. Thinkpad T42. Everything seems fine.
Same here on a "Dell Precision M6300 Workstation": pnpacpi: exceeded the max number of mem resources: 12 - see also: http://lkml.org/lkml/2007/12/5/301
Oh and see http://lkml.org/lkml/2007/12/13/77 as well...
i got the same message too, Thinkpad T43, 2.6.24.3-12 Everything seems fine.
I got the same message on Dell optiplex gx280, Dell Latitude 520, and Dell Inspiron 1420.
When I unplug a SATA-drive my SYSTEM FREEZES (it was unmounted before). When SATA drives were mounted, the drive was attached and then unmounted (but not unplugged, this would cause system freeze) autofs fails halting while shutting down automount on system shutdown. This behavior appears since kernel 2.6.24.3-12.fc8 and only on the machine which shows that pnpacpi message. My other machines work fine.
I just added a patch to get some more debugging info. It'll show up in builds starting at version kernel-2.6.24.3-31.fc8 If all of you seeing the overflow messages could run this when it shows up and paste the updated version of the message that might give us some idea of how large a range we're expecting to see. Thanks.
*** Bug 425837 has been marked as a duplicate of this bug. ***
Dave, Any idea when an updated kernel will make it into either testing or main repo? Only reason I ask is there are clear network adapter name bugs with this kernel as well (https://bugzilla.redhat.com/show_bug.cgi?id=436456) so it would be good to try one updated kernel fixing both these issues rather than potentially two sets of kernels being made for different issues?
here is the output from kernel-2.6.24.3-33.fc8 Linux Plug and Play Support v0.97 (c) Adam Belay pnp: PnP ACPI init ACPI: bus type pnp registered pnpacpi: exceeded the max number of mem resources. Max:12 Found:12 pnp: PnP ACPI: found 10 devices ACPI: ACPI bus type pnp unregistered
ACPI: bus type pci registered PCI: PCI BIOS revision 3.00 entry at 0xfaf90, last bus=3 PCI: Using configuration type 1 Setting up standard PCI resources ACPI: EC: Look up EC in DSDT ACPI: Interpreter enabled ACPI: (supports S0 S1 S4 S5) ACPI: Using IOAPIC for interrupt routing ACPI: PCI Root Bridge [PCI0] (0000:00) Force enabled HPET at base address 0xfed00000 PCI quirk: region 0400-047f claimed by ICH6 ACPI/GPIO/TCO PCI quirk: region 0480-04bf claimed by ICH6 GPIO PCI: Transparent bridge - 0000:00:1e.0 ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.PEX0._PRT] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.HUB0._PRT] ACPI: PCI Interrupt Link [LNKA] (IRQs 3 4 *5 6 7 9 10 11 12 14 15) ACPI: PCI Interrupt Link [LNKB] (IRQs 3 4 5 6 7 9 10 11 12 14 15) *0, disabled. ACPI: PCI Interrupt Link [LNKC] (IRQs 3 4 5 6 7 *9 10 11 12 14 15) ACPI: PCI Interrupt Link [LNKD] (IRQs 3 4 5 6 7 9 10 *11 12 14 15) ACPI: PCI Interrupt Link [LNKE] (IRQs 3 4 5 6 7 9 *10 11 12 14 15) ACPI: PCI Interrupt Link [LNKF] (IRQs 3 4 5 6 7 9 10 11 12 14 15) *0, disabled. ACPI: PCI Interrupt Link [LNK0] (IRQs 3 4 5 6 7 9 10 11 12 14 15) *0, disabled. ACPI: PCI Interrupt Link [LNK1] (IRQs *3 4 5 6 7 9 10 11 12 14 15) Linux Plug and Play Support v0.97 (c) Adam Belay pnp: PnP ACPI init ACPI: bus type pnp registered pnpacpi: exceeded the max number of mem resources: 12 pnp: PnP ACPI: found 13 devices ACPI: ACPI bus type pnp unregistered
With kernel-2.6.24.3-34: Mar 14 12:35:09 localhost kernel: pnp: PnP ACPI init Mar 14 12:35:09 localhost kernel: ACPI: bus type pnp registered Mar 14 12:35:09 localhost kernel: pnpacpi: exceeded the max number of mem resources. Max:12 Found:12 Mar 14 12:35:09 localhost kernel: pnp: PnP ACPI: found 14 devices Mar 14 12:35:09 localhost kernel: ACPI: ACPI bus type pnp unregistered Mar 14 12:35:09 localhost kernel: usbcore: registered new interface driver usbfs
With 2.6.24.3-22.fc8 (ThinkPad X30) Mar 14 17:21:27 localhost kernel: pnp: PnP ACPI init Mar 14 17:21:27 localhost kernel: ACPI: bus type pnp registered Mar 14 17:21:27 localhost kernel: pnpacpi: exceeded the max number of mem resources: 12 Mar 14 17:21:27 localhost kernel: pnp: PnP ACPI: found 12 devices And it turns out that it doesn't actually die when starting udev, just takes a *very* long time.
I made the following temporary patch to /etc/yum/yum.conf to stop the update notifications from popping up: =========== Start temp.patch ================================ *** yum-updatesd.conf 2007-12-17 23:41:33.000000000 -0500 --- yum-updatesd.conf.new 2008-03-15 10:46:20.000000000 -0400 *************** *** 16,18 **** --- 16,21 ---- do_download = no # automatically download deps of updates do_download_deps = no + + # Temporarily ignore bad kernel update + kernel-2.6.24.3-12.fc8 =========== End temp.patch ================================ I tried to restart the yum-updatesd service, but got a subsys lock mesage. I rebooted to get a clean restart. Now, no update popup appears. The kernel update is still visible in Software Updater. Hope this helps, until the bad update is pulled from the repositories and/or a better kernel update is posted.
Sorry, please ignore the previous patch. It contains a syntax error: I had forgotten to include the "exclude" command. I tried fixing that, and putting wildcard *'s in, but I still get the popup. I'll go back to quitting the update process after each boot.
This patch worked for me to remove this update from being offered by the popup software updater: ========= Start of patch ===================================== *** /etc/yum.repos.d/fedora-updates.repo 2007-12-12 16:30:15.000000000 -0500 --- /etc/yum.repos.d/fedora-updates.repo.patch 2008-03-16 06:01:30.000000000 -0400 *************** *** 6,11 **** --- 6,12 ---- enabled=1 gpgcheck=1 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-fedora + exclude = kernel-2.6.24.3-12.fc8.* [updates-debuginfo] name=Fedora $releasever - $basearch - Updates - Debug~ ========= End of patch ===================================== Of course, subsequent updates should still appear.
Created attachment 298202 [details] dmesg showing error messages for 2.6.23.15-137.fc8
Created attachment 298205 [details] dmesg showing error messages for experimental version 2.6.24.3-22.fc8
Created attachment 298206 [details] /proc/iomem for working kernel 2.6.23.15-137.fc8
Created attachment 298207 [details] /proc/iomem for test kernel 2.6.24.3-22.fc8
Downloaded kernel-2.6.24.3-12.fc8.i586, not offered by updater. Same behavior. Deleted that kernel, and installed kernel-2.6.24.3-22.fc8.i686. Still behavior, and also changed versions of the message reported in bug #285251. In connection with the latter, I supplied a copy of /proc/iomem. While waiting for the debugging patch promised in comment #11, I have posted dmesg and iomem output for both the working kernel-2.6.23.15-137.fc8.i686 and the test version, kernel-2.6.24.3-22.fc8.i686. I hope this information helps speed resolution of this bug.
Created attachment 298229 [details] dmesg showing error messages for experimental version 2.6.24.3-34.fc8
Created attachment 298230 [details] /proc/iomem for test kernel-2.6.24.3-34.fc8
Much improved. I have to start the network manually, using System/Administration/Network Device Control/Activate. My network devices now have .bak suffixes.
Significant diffs of the dmesg and iomem files are as follows: diff iomem.2.6.24.3-22.fc8 iomem.2.6.24.3-34.fc8 9,10c9,10 < 00400000-0062f688 : Kernel code < 0062f689-0074a723 : Kernel data --- > 00400000-0062f698 : Kernel code > 0062f699-0074a723 : Kernel data and diff dmesg.2.6.24.3-22.fc8 dmesg.2.6.24.3-34.fc8 2c2 < Linux version 2.6.24.3-22.fc8 (mockbuild.redhat.com) (gcc version 4.1.2 20070925 (Red Hat 4.1.2-33)) #1 SMP Fri Mar 7 10:08:30 EST 2008 --- > Linux version 2.6.24.3-34.fc8 (mockbuild.redhat.com) (gcc version 4.1.2 20070925 (Red Hat 4.1.2-33)) #1 SMP Wed Mar 12 18:17:20 EDT 2008 75c75 < Kernel command line: ro root=LABEL=/ rhgb quiet nohz=off highres=off --- > Kernel command line: ro root=LABEL=/ rhgb quiet ... 167,168c167,168 < pnpacpi: exceeded the max number of mem resources: 12 < pnpacpi: exceeded the max number of IO resources: 40 --- > pnpacpi: exceeded the max number of mem resources. Max:12 Found:12 > pnpacpi: exceeded the max number of IO resources. Max:40 Found:40 ... 417,418c418,421 < iTCO_vendor_support: vendor-support=0 < input: Power Button (FF) as /devices/virtual/input/input4 --- > sd 0:0:0:0: Attached scsi generic sg0 type 0 > sr 1:0:0:0: Attached scsi generic sg1 type 5 > input: PC Speaker as /devices/platform/pcspkr/input/input4 > input: Power Button (FF) as /devices/virtual/input/input5 420c423,428 < input: Lid Switch as /devices/virtual/input/input5 --- > input: Lid Switch as /devices/virtual/input/input6 > ACPI: Lid Switch [LID] > input: Power Button (CM) as /devices/virtual/input/input7 > iTCO_vendor_support: vendor-support=0 > ACPI: Power Button (CM) [PWRB] > ACPI: Battery Slot [BAT1] (battery present) ... more diffs regarding hardware assigned differently > warning: process `system-control-' used the deprecated sysctl system call with 1.23. > warning: process `/usr/sbin/syste' used the deprecated sysctl system call with 1.23. > ADDRCONF(NETDEV_UP): eth1: link is not ready > e100: eth0: e100_watchdog: link up, 100Mbps, full-duplex > ADDRCONF(NETDEV_UP): eth0: link is not ready > ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready > eth0: no IPv6 routers present
Created attachment 298232 [details] output from yum/grubby when the kernel update fails
Comment on attachment 298232 [details] output from yum/grubby when the kernel update fails Hi. I'm also seeing this, but interestingly the kernel fails to even install properly (kernel-2.6.24.3-34). When I try to update via YUM, something is failing and /boot/grub/menu.1st doesn't get updated properly. The entry for the new kernel isn't getting put in. I'm attaching the file with the output I get when it fails. If I manually update /boot/grub/menu.1st I can get my box to boot, but I see the "pnpacpi: exceeded the max number of mem resources" and the machine behaves somewhat erratically. I can explain "somewhat erratically" in more detail if it's helpful, but I'm guessing the attached file will be more useful. If I'm wrong let me know and do my best to explain what I see.
After fighting with a weird hardware/BIOS issue, I started looking into this error message on reboot. kernel-2.6.24.3-12.fc8 dmesg output. Linux Plug and Play Support v0.97 (c) Adam Belay pnp: PnP ACPI init ACPI: bus type pnp registered pnpacpi: exceeded the max number of mem resources: 12 pnp: PnP ACPI: found 14 devices ACPI: ACPI bus type pnp unregistered As it shows, I have 14 devices but only 12 are accepted. I don't know how this affects my machine but I will have to wait and see. I did search and found that this ia a kernel level decision and some of it is waiting for dynamic code. But in the mean time, something has to be done now. This is what I found. http://www.mail-archive.com/linux-kernel@vger.kernel.org/msg236413.html http://kerneltrap.org/mailarchive/linux-kernel/2007/11/30/464546 Having an error message like this can take time away from finding and repairing real problems.
IIRC, there was a message about the resources being exceeded with the last 2.6.23 kernel made available for F8. It just didn't give any stats. With the 2.6.24 kernels it would be appear to be using >= instead > when doing a check. If 12 slots are needed and 12 are available there is no problem and it has not exceeded the available resources.
My comments on this issue are here http://lkml.org/lkml/2008/3/12/255 As it seems we'll be living with these hard-coded, but sometimes insufficient limits for some time, I'm inclined to reduce KERN_ERR to something less severe. If Red Hat deleted this message entirely from Fedora to stop alarming users, I'd be okay with that too.
Created attachment 298469 [details] diff showing /etc/sysconfig/hwconf change After seeing that the pnpapci message is probably inoccuous, I have decided to go ahead and try kernel 2.6.24.3-34.fc8. On reboot, I found my network was dead. In the System/Administration/Network, menu, I changed the eth0.bak device name to eth0. Also, I made sure to make this device become active after booting. Did a diff -r <previous /etc/hwconfig> /etc/hwconfig. Several differences. The result is it Attachment 7 [details].
Well, I meant the 7th in this posting, not Attachment 7 [details]. The diff is here: https://bugzilla.redhat.com/attachment.cgi?id=298469 .
(In reply to comment #34) > As it seems we'll be living with these hard-coded, > but sometimes insufficient limits for some time, > I'm inclined to reduce KERN_ERR to something less severe. > If Red Hat deleted this message entirely from Fedora > to stop alarming users, I'd be okay with that too. Changed to KERN_DEBUG in 2.6.24.3-50
Well w.r.t. hard coded limit I also get "pnpacpi: exceeded the max number of I/O resources 40:40" on a rather standard Dell Dimension 9150 when booting 2.6.24.3-34.fc8 x86_64 Daniel
upstream I'll change this to KERN_WARNING. Fedora boots with "quiet" by default, and so KERN_WARNING will not be printed on the default boot. for users that delete the "quiet", they'll get this message along with a zillion others, which is appropriate. going to KERN_DEBUG and thus requiring "debug" on the cmdline to see this message seems to be burying it a little too deep, as it is a valid warning after all. also, note also that 2.6.25-rc7 made this message less likely: 03c086a747d0b242878eb881971ec61c1555869d include/linux/pnp.h -#define PNP_MAX_MEM 12 +#define PNP_MAX_MEM 24
The consequence of this seems to be that devices don't work/activate. Symptoms on my machine are that the sound system drops alsa, oss or arts. I noticed that switching xine, mms or amarok to a different sound module causes the external USB hard disk to unmount and disappear, remounting the USB hard disk then causes sound to stop working. From history it seems others are having similar issue but with different devices, network cards etc.
I think comment 40 is correct. See my comments 20 and 21 (and attachment(id=299389)) on bug#436583. Udev times out during boot and I lose network and sound.
The "exceeded the number of resources" message was just a warning. So many people, all with different unrelated bugs, have piled onto this report that there is no way to make any sense of it now. The severity of the warning has been lowered so it will no longer show during normal (quiet) boot. Anyone who is still having a problem please file a new separate bug (after trying 2.6.24.4-64).