Re: mbuf trash panic turned into a printf for now

From: othermark (atkin901_at_yahoo.com)
Date: 06/29/05

  • Next message: Jens Schweikhardt: "Re: Timekeeping hosed by factor 3, high lapic[01] interrupt rates" 
    To: freebsd-current@freebsd.org
Date: Wed, 29 Jun 2005 10:25:57 -0700

    Mike Silbersack wrote:
    > Since the mbuf trash panic is being reported more often now, and we
    > haven't tracked down why yet, I've turned it into a printf for now. This
    > means that it's safe to upgrade to -current even if you might be affected.
    > :)
    >
    > And the more reports, the better, keep them coming if you see the "Memory
    > modified after free" message.

    I updated to -current after seeing this posting.

    What information do you need with each of these reports?

    [root@consrv tmp]$ dmesg -a | grep "^Memory" | uniq -u | wc -l
         430

    Attacahed is the output. Since this is dual processor machine it seems
    that two mbuf printfs can trample on each other in the output.

    Is the printf output right? Is the last address supposed to derefence
    the calling function? Right now they are mostly the same in my output.

    [root@consrv tmp]$ dmesg -a | grep "^Memory" | cut -d " " -f 8 | xargs
    addr2line -e /usr/obj/usr/src/sys/CONSRV/kernel.debug
    ??:0
    <...>

    em0@pci0:3:0: class=0x020000 card=0x10008086 chip=0x10008086 rev=0x03
    hdr=0x00
        vendor = 'Intel Corporation'
        device = '82542 Gigabit Ethernet Controller'
        class = network
        subclass = ethernet
    em1@pci0:4:0: class=0x020000 card=0x10008086 chip=0x10008086 rev=0x03
    hdr=0x00
        vendor = 'Intel Corporation'
        device = '82542 Gigabit Ethernet Controller'
        class = network
        subclass = ethernet 

    -- 
othermark
atkin901 at nospam dot yahoo dot com
(!wired)?(coffee++):(wired);

    _______________________________________________
    freebsd-current@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-current
    To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"

  • Next message: Jens Schweikhardt: "Re: Timekeeping hosed by factor 3, high lapic[01] interrupt rates"