https://bugs.dragonflybsd.org/https://bugs.dragonflybsd.org/favicon.ico?16293952082007-01-30T12:37:00ZDragonFlyBSD bugtrackerDragonFlyBSD - Bug #547: 1.8 release cd ACPI message loophttps://bugs.dragonflybsd.org/issues/547?journal_id=22302007-01-30T12:37:00Zdillon
<ul></ul><p>:Booting 1.8 using VMWare Fusion on OS X works fine.<br />:<br />:Both systems showed messages like this at shutdown of the LiveCD:<br />:(this one's from the VMWare environment)<br />:unmount(/dev): Cannot unmount: 0 namecache references still present<br />:unmount(/dev): Cannot unmount: 5 process references still present<br />:unmount(/var): Cannot unmount: 0 namecache references still present<br />:unmount(/var): Cannot unmount: 5 process references still present<br />:<br />:and after the disk sync:<br />:unmount(0xc12d57c0): Forced unmount: 2 namecache references still present<br />:unmount(0xc12d57c0): Forced unmount: 4 process references still present<br />:EXDEV case 1 0xc10c4238</p>
<pre><code>Yah, those can be ignored. The CD is read-only anyhow. Getting rid<br /> of them means cleaning up the installer a bit.</code></pre>
<pre><code>-Matt<br /> Matthew Dillon <br /> &lt;<a class="email" href="mailto:dillon@backplane.com">dillon@backplane.com</a>&gt;</code></pre> DragonFlyBSD - Bug #547: 1.8 release cd ACPI message loophttps://bugs.dragonflybsd.org/issues/547?journal_id=22342007-01-31T08:38:02Zqhwt+dfly
<ul></ul><p>Does this message show up very quickly? If not, can you try to drop<br />into the boot loader(press "6" on Fred menu):<br /> set debug.acpi_semaphore_debug=1<br />then boot, and write down some messages beginning with "AcpiOsWaitSemaphore" <br />and "AcpiOsSignalSemaphore".</p>
<p>I couldn't find the DSDT for 600E, but only for 600 on the Internet:<br /> <a class="external" href="http://acpi.sourceforge.net/dsdt/view.php?id=457">http://acpi.sourceforge.net/dsdt/view.php?id=457</a></p>
<p>according to this one, the trace is something like:<br /> \_TZ.THM0._TMP<br /> --> \_SB.PCI0,ISA0,EC0.UPDT<br /> --> \_SB.PCI0,ISA0,EC0.I2RB<br /> --> (access to one of members of ECOR, trying to acquire a semaphore)</p>
<p>The only place where AE_TIME is returned is AcpiOsWaitSemaphore(),<br />so it's either<br />- there's four or more threads waiting for the same semaphore,<br />- msleep(9) failed for some reason, or<br />- AcpiOsWaitSemaphore() was instructed to return immediately<br /> (but it's only in AcpiExSystemWaitSemaphore() and no warnings<br /> should show up)</p>
<p>There's a mechanism to show you extra debugging information, but<br />unfortunately acpi.ko gets compiled without ACPI_DEBUG defined,<br />so it's not available via release ISO (but in this case,<br />debug.acpi_semaphore_debug should help). I think making ACPI_DEBUG<br />enabled for snapshot ISOs is a good idea.</p>
<p>I think debug.acpi.disabled="thermal" (or "ec") in the boot loader can be<br />used as a workaround for now.</p>
<p>Cheers.</p> DragonFlyBSD - Bug #547: 1.8 release cd ACPI message loophttps://bugs.dragonflybsd.org/issues/547?journal_id=28772007-06-04T03:54:20Zjustin
<ul></ul><p>(Resurrecting ancient issue)</p>
<p>Tested this with a 1.9 CD from yesterday, with debug.acpi_semaphore_debug set:</p>
<p>AcpiOsWaitSemaphore: Acquire 1, units 0, pending 0, sem 0xc1080b28, thread 1<br />AcpiOsWaitSemaphore: Sleep 1000, pending 1, semaphore 0xc1080b28, thread 1<br />AcpiOsWaitSemaphore: Timeout 1000, pending 0, semaphore 0xc1080b28<br />ACPI Exception (evregion-0529): AE_TIME, Returned by Handler for<br />[EmbeddedControl] [20061109]<br />AcpiOsSignalSemaphore: Release 1, units 1, pending 0, semaphore 0xc1080b28, thread 1<br />ACPI Error (psparse-0638): Method parse/execution failed<br />[\_SB_.PCI0.ISA0.EC0_.I2RB] (Node 0xc1084e78), AE_TIME<br />ACPI Error (psparse-0638): Method parse/execution failed<br />[\_SB_.PCI0.ISA0.EC0_.UPDT] (Node 0xc1084dd0), AE_TIME<br />ACPI Error (psparse-0638): Method parse/execution failed [\_TZ_.THM0._TMP]<br />(Node 0xc1084a10), AE_TIME</p>
<p>I do notice early on a number of ACPI messages about "cannot release global<br />lock" - they scroll by fast enough I haven't been able to capture them. Setting <br />debug.acpi.disabled="thermal" (or "ec") gave a slightly new set of errors, which<br />I can write out if they're important.</p> DragonFlyBSD - Bug #547: 1.8 release cd ACPI message loophttps://bugs.dragonflybsd.org/issues/547?journal_id=31772007-07-09T12:19:00Zqhwt+dfly
<ul></ul><p>I uploaded an ISO image built from yesterday's source code<br />on my leaf account, with ACPI_DEBUG=yes defined:<br /> <a class="external" href="http://leaf.dragonflybsd.org/~y0netan1/dfly-20070708.iso.bz2">http://leaf.dragonflybsd.org/~y0netan1/dfly-20070708.iso.bz2</a><br />Please drop into boot loader, and type following commands:<br /> set debug.acpi.layer=ACPI_OS_SERVICES<br /> set debug.acpi.level=ACPI_LV_MUTEX<br />This should enable some more debugging messages which may be useful.</p>
<p>Besides that, can you send me the acpidump output on your computer?</p>
<p>I'd like to know whether setting debug.acpi.disabled="thermal" (or "ec")<br />fixes the first (looping) message?</p>
<p>Cheers.</p> DragonFlyBSD - Bug #547: 1.8 release cd ACPI message loophttps://bugs.dragonflybsd.org/issues/547?journal_id=36852007-09-16T10:17:56Zjustin
<ul></ul><p>I finally got a chance to test this ACPI issue. Setting<br />debug.acpi.disabled="thermal" lets it boot, using the dfly-20070708.iso image.</p>
<p>Output of acpidump -t is below:</p>
<p>/*<br /> RSD PTR: OEM=IBM, ACPI_Rev=1.0x (0)<br /> RSDT=0x11fd0000, cksum=155<br /> <strong>/<br />/</strong><br /> RSDT: Length=44, Revision=1, Checksum=52,<br /> OEMID=IBM, OEM Table ID=TP600R, OEM Revision=0x1,<br /> Creator ID=, Creator Revision=0x0<br /> Entries={ 0x11fd0100, 0x11fd0040 }<br /> <strong>/<br />/</strong><br /> FACP: Length=116, Revision=1, Checksum=231,<br /> OEMID=IBM, OEM Table ID=TP600R, OEM Revision=0x1,<br /> Creator ID=, Creator Revision=0x0<br /> FACS=0x11fd0100, DSDT=0x11fd0200<br /> INT_MODEL=PIC<br /> Preferred_PM_Profile=Unspecified (0)<br /> SCI_INT=9<br /> SMI_CMD=0xb2, ACPI_ENABLE=0xa0, ACPI_DISABLE=0xa1, S4BIOS_REQ=0xa2<br /> PSTATE_CNT=0x0<br /> PM1a_EVT_BLK=0xef00-0xef03<br /> PM1a_CNT_BLK=0xef04-0xef05<br /> PM2_CNT_BLK=0x22-0x22<br /> PM_TMR_BLK=0xef08-0xef0b<br /> GPE0_BLK=0xef0c-0xef0f<br /> P_LVL2_LAT=1 us, P_LVL3_LAT=65 us<br /> FLUSH_SIZE=32768, FLUSH_STRIDE=32<br /> DUTY_OFFSET=1, DUTY_WIDTH=3<br /> DAY_ALRM=13, MON_ALRM=0, CENTURY=50<br /> IAPC_BOOT_ARCH=<br /> Flags={WBINVD,PROC_C1,P_LVL2_UP,SLP_BUTTON,RTC_S4,DCK_CAP}<br /> <strong>/<br />/</strong><br /> FACS: Length=64, HwSig=0x00000106, Firm_Wake_Vec=0x00000000<br /> Global_Lock=<br /> Flags=<br /> Version=0<br /> <strong>/<br />/</strong><br /> DSDT: Length=50233, Revision=1, Checksum=131,<br /> OEMID=IBM, OEM Table ID=TP600R, OEM Revision=0x106,<br /> Creator ID=MSFT, Creator Revision=0x100000c<br /> <strong>/<br />/</strong><br /> BOOT: Length=40, Revision=1, Checksum=107,<br /> OEMID=IBM, OEM Table ID=TP600R, OEM Revision=0x1,<br /> Creator ID=, Creator Revision=0x0<br /> */</p> DragonFlyBSD - Bug #547: 1.8 release cd ACPI message loophttps://bugs.dragonflybsd.org/issues/547?journal_id=36862007-09-16T22:05:02Zqhwt+dfly
<ul></ul><p>So ... does it boot with more recent official snapshot?<br />debug.acpi.disabled="thermal" works on an ACPI module compiled<br />without ACPI_DEBUG, so it's either because turning on ACPI_DEBUG<br />changes something in ACPI-CA code, or the recent changes up to 20070708<br />have fixed something for your laptop PC to be able to boot.</p>
<p>And if the network interface works, can you send me the dmesg.boot<br />file, and the DSDT output (you need to specify "-d" or "-o file" <br />to acpidump to obtain it).</p>
<p>Cheers.</p> DragonFlyBSD - Bug #547: 1.8 release cd ACPI message loophttps://bugs.dragonflybsd.org/issues/547?journal_id=36872007-09-17T04:58:06Zjustin
<ul></ul><p>It boots under 1.10.1 the same - with thermal off, it's fine.</p>
<p>It's not online, so it'll take a while to obtain the other data.</p> DragonFlyBSD - Bug #547: 1.8 release cd ACPI message loophttps://bugs.dragonflybsd.org/issues/547?journal_id=36902007-09-17T09:40:04Zqhwt+dfly
<ul></ul><p>Ah, that's OK, it narrowed down the thing a bit. I'll try to pick up<br />some recent(well, almost three years ago) changes which look relavant<br />and port them over for us.</p>
<p>Cheers.</p> DragonFlyBSD - Bug #547: 1.8 release cd ACPI message loophttps://bugs.dragonflybsd.org/issues/547?journal_id=71202009-08-23T18:16:28Ztuxillo
<ul></ul><p>Justin,<br />Do you have that old Thinkpad 600E to give it a try with current HEAD?</p> DragonFlyBSD - Bug #547: 1.8 release cd ACPI message loophttps://bugs.dragonflybsd.org/issues/547?journal_id=71242009-08-24T05:45:50Zjustin
<ul></ul><p>Marking resolved; it's being resolved by age, not by actually having it work. <br />The laptop is so old I practically have to shovel coal in to get it started;<br />it's not worth worrying about when faster, more compatible models exist for <$200.</p>