I''m currently working on moving storage services into their own domain and I''ve been looking at blktap2. I''ve been trying to get an image mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging instead of quitting at the end. I haven''t removed any of the storage startup code at this point so everything should be as it normally is in xen-unstable. Is this a known bug? Are there some requirements on blktap2 that I''m possibly not meeting? Is there anything in particular that I should be looking at to try to debug the problem? Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Thu, Jan 21, 2010 at 05:27:45PM -0500, David P. Quigley wrote:> I''m currently working on moving storage services into their own domain > and I''ve been looking at blktap2. I''ve been trying to get an image > mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging > instead of quitting at the end. I haven''t removed any of the storage > startup code at this point so everything should be as it normally is in > xen-unstable. Is this a known bug? Are there some requirements on > blktap2 that I''m possibly not meeting? Is there anything in particular > that I should be looking at to try to debug the problem? >What kernel are you using? pv_ops dom0 kernel doesn''t have working blktap2 yet.. linux-2.6.18-xen has working blktap2 support. -- Pasi _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Fri, 2010-01-22 at 00:38 +0200, Pasi Kärkkäinen wrote:> On Thu, Jan 21, 2010 at 05:27:45PM -0500, David P. Quigley wrote: > > I''m currently working on moving storage services into their own domain > > and I''ve been looking at blktap2. I''ve been trying to get an image > > mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging > > instead of quitting at the end. I haven''t removed any of the storage > > startup code at this point so everything should be as it normally is in > > xen-unstable. Is this a known bug? Are there some requirements on > > blktap2 that I''m possibly not meeting? Is there anything in particular > > that I should be looking at to try to debug the problem? > > > > What kernel are you using? pv_ops dom0 kernel doesn''t have working blktap2 yet.. > linux-2.6.18-xen has working blktap2 support. > > -- PasiI''m using the patches maintained by Andrew Lyon that are applied to 2.6.31.10. These are supposedly based on the OpenSUSE patches and can be found at http://code.google.com/p/gentoo-xen-kernel/downloads/list. I need a more recent kernel version so the 2.6.18 dom0 kernel isn''t an option for me. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Thu, 2010-01-21 at 14:54 -0800, Daniel Stodden wrote:> On Thu, 2010-01-21 at 17:27 -0500, David P. Quigley wrote: > > I''m currently working on moving storage services into their own domain > > and I''ve been looking at blktap2. I''ve been trying to get an image > > mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging > > instead of quitting at the end. I haven''t removed any of the storage > > startup code at this point so everything should be as it normally is in > > xen-unstable. Is this a known bug? Are there some requirements on > > blktap2 that I''m possibly not meeting? Is there anything in particular > > that I should be looking at to try to debug the problem? > > Are you running pvops? > Cheer up, after 2.6.23 it''s not you but just broken. > > The hang you encounter is likely blktap trying to remove a sysfs node > from within its own callback. That could be circumvented, but it''s > likely not the only issue. > > Daniel >Well if its just broken in general any suggestions on where I should look to try and fix it? blktap2 seems to be my best bet for implementing my component so if I need to fix it to get it working I''m fine with that. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Thu, 2010-01-21 at 17:27 -0500, David P. Quigley wrote:> I''m currently working on moving storage services into their own domain > and I''ve been looking at blktap2. I''ve been trying to get an image > mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging > instead of quitting at the end. I haven''t removed any of the storage > startup code at this point so everything should be as it normally is in > xen-unstable. Is this a known bug? Are there some requirements on > blktap2 that I''m possibly not meeting? Is there anything in particular > that I should be looking at to try to debug the problem?Are you running pvops? Cheer up, after 2.6.23 it''s not you but just broken. The hang you encounter is likely blktap trying to remove a sysfs node from within its own callback. That could be circumvented, but it''s likely not the only issue. Daniel _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Thu, Jan 21, 2010 at 05:46:05PM -0500, David P. Quigley wrote:> On Fri, 2010-01-22 at 00:38 +0200, Pasi Kärkkäinen wrote: > > On Thu, Jan 21, 2010 at 05:27:45PM -0500, David P. Quigley wrote: > > > I''m currently working on moving storage services into their own domain > > > and I''ve been looking at blktap2. I''ve been trying to get an image > > > mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging > > > instead of quitting at the end. I haven''t removed any of the storage > > > startup code at this point so everything should be as it normally is in > > > xen-unstable. Is this a known bug? Are there some requirements on > > > blktap2 that I''m possibly not meeting? Is there anything in particular > > > that I should be looking at to try to debug the problem? > > > > > > > What kernel are you using? pv_ops dom0 kernel doesn''t have working blktap2 yet.. > > linux-2.6.18-xen has working blktap2 support. > > > > -- Pasi > > I''m using the patches maintained by Andrew Lyon that are applied to > 2.6.31.10. These are supposedly based on the OpenSUSE patches and can be > found at http://code.google.com/p/gentoo-xen-kernel/downloads/list. I > need a more recent kernel version so the 2.6.18 dom0 kernel isn''t an > option for me. >Ok. I''m not sure if those OpenSUSE patches have blktap2 support included? I know that at least Citrix XenServer, XCP and XCI 2.6.27 kernels do have blktap2. See: http://wiki.xensource.com/xenwiki/XenDom0Kernels -- Pasi _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Fri, Jan 22, 2010 at 12:55:52AM +0200, Pasi Kärkkäinen wrote:> On Thu, Jan 21, 2010 at 05:46:05PM -0500, David P. Quigley wrote: > > On Fri, 2010-01-22 at 00:38 +0200, Pasi Kärkkäinen wrote: > > > On Thu, Jan 21, 2010 at 05:27:45PM -0500, David P. Quigley wrote: > > > > I''m currently working on moving storage services into their own domain > > > > and I''ve been looking at blktap2. I''ve been trying to get an image > > > > mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging > > > > instead of quitting at the end. I haven''t removed any of the storage > > > > startup code at this point so everything should be as it normally is in > > > > xen-unstable. Is this a known bug? Are there some requirements on > > > > blktap2 that I''m possibly not meeting? Is there anything in particular > > > > that I should be looking at to try to debug the problem? > > > > > > > > > > What kernel are you using? pv_ops dom0 kernel doesn''t have working blktap2 yet.. > > > linux-2.6.18-xen has working blktap2 support. > > > > > > -- Pasi > > > > I''m using the patches maintained by Andrew Lyon that are applied to > > 2.6.31.10. These are supposedly based on the OpenSUSE patches and can be > > found at http://code.google.com/p/gentoo-xen-kernel/downloads/list. I > > need a more recent kernel version so the 2.6.18 dom0 kernel isn''t an > > option for me. > > > > Ok. I''m not sure if those OpenSUSE patches have blktap2 support included? > I know that at least Citrix XenServer, XCP and XCI 2.6.27 kernels do have blktap2. > > See: > http://wiki.xensource.com/xenwiki/XenDom0Kernels >More details about the pv_ops dom0 kernel (2.6.31.6 currently): It has initial version of blktap2 patches, but they''d need some debugging, because they''re not yet working. -- Pasi _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Fri, 2010-01-22 at 00:55 +0200, Pasi Kärkkäinen wrote:> On Thu, Jan 21, 2010 at 05:46:05PM -0500, David P. Quigley wrote: > > On Fri, 2010-01-22 at 00:38 +0200, Pasi Kärkkäinen wrote: > > > On Thu, Jan 21, 2010 at 05:27:45PM -0500, David P. Quigley wrote: > > > > I''m currently working on moving storage services into their own domain > > > > and I''ve been looking at blktap2. I''ve been trying to get an image > > > > mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging > > > > instead of quitting at the end. I haven''t removed any of the storage > > > > startup code at this point so everything should be as it normally is in > > > > xen-unstable. Is this a known bug? Are there some requirements on > > > > blktap2 that I''m possibly not meeting? Is there anything in particular > > > > that I should be looking at to try to debug the problem? > > > > > > > > > > What kernel are you using? pv_ops dom0 kernel doesn''t have working blktap2 yet.. > > > linux-2.6.18-xen has working blktap2 support. > > > > > > -- Pasi > > > > I''m using the patches maintained by Andrew Lyon that are applied to > > 2.6.31.10. These are supposedly based on the OpenSUSE patches and can be > > found at http://code.google.com/p/gentoo-xen-kernel/downloads/list. I > > need a more recent kernel version so the 2.6.18 dom0 kernel isn''t an > > option for me. > > > > Ok. I''m not sure if those OpenSUSE patches have blktap2 support included? > I know that at least Citrix XenServer, XCP and XCI 2.6.27 kernels do have blktap2. > > See: > http://wiki.xensource.com/xenwiki/XenDom0Kernels > > -- Pasi$ grep BLKTAP2 * 60011_xen3-auto-xen-drivers.patch1:+#define BLKTAP2_IOCTL_KICK_FE 1 60011_xen3-auto-xen-drivers.patch1:+#define BLKTAP2_IOCTL_ALLOC_TAP 200 60011_xen3-auto-xen-drivers.patch1:+#define BLKTAP2_IOCTL_FREE_TAP 201 60011_xen3-auto-xen-drivers.patch1:+#define BLKTAP2_IOCTL_CREATE_DEVICE 202 60011_xen3-auto-xen-drivers.patch1:+#define BLKTAP2_IOCTL_SET_PARAMS 203 60011_xen3-auto-xen-drivers.patch1:+#define BLKTAP2_IOCTL_PAUSE 204 60011_xen3-auto-xen-drivers.patch1:+#define BLKTAP2_IOCTL_REOPEN 205 60011_xen3-auto-xen-drivers.patch1:+#define BLKTAP2_IOCTL_RESUME 206 60011_xen3-auto-xen-drivers.patch1:+#define BLKTAP2_MAX_MESSAGE_LEN 256 60011_xen3-auto-xen-drivers.patch1:+#define BLKTAP2_RING_MESSAGE_PAUSE 1 60011_xen3-auto-xen-drivers.patch1:+#define BLKTAP2_RING_MESSAGE_RESUME 2 60011_xen3-auto-xen-drivers.patch1:+#define BLKTAP2_RING_MESSAGE_CLOSE 3 60011_xen3-auto-xen-drivers.patch1:+ char name[BLKTAP2_MAX_MESSAGE_LEN]; 60011_xen3-auto-xen-drivers.patch1:+ case BLKTAP2_IOCTL_ALLOC_TAP: { 60011_xen3-auto-xen-drivers.patch1:+ case BLKTAP2_IOCTL_FREE_TAP: 60011_xen3-auto-xen-drivers.patch1:+ case BLKTAP2_IOCTL_KICK_FE: 60011_xen3-auto-xen-drivers.patch1:+ case BLKTAP2_IOCTL_CREATE_DEVICE: 60011_xen3-auto-xen-drivers.patch1:+ case BLKTAP2_IOCTL_SET_PARAMS: 60011_xen3-auto-xen-drivers.patch1:+ case BLKTAP2_IOCTL_PAUSE: 60011_xen3-auto-xen-drivers.patch1:+ case BLKTAP2_IOCTL_REOPEN: 60011_xen3-auto-xen-drivers.patch1:+ case BLKTAP2_IOCTL_RESUME: 60011_xen3-auto-xen-drivers.patch1:+ blktap_ring_set_message(tap, BLKTAP2_RING_MESSAGE_RESUME); 60011_xen3-auto-xen-drivers.patch1:+ blktap_ring_set_message(tap, BLKTAP2_RING_MESSAGE_PAUSE); 60011_xen3-auto-xen-drivers.patch1:+ blktap_ring_set_message(tap, BLKTAP2_RING_MESSAGE_CLOSE); 60011_xen3-auto-xen-drivers.patch1:+ if (size > BLKTAP2_MAX_MESSAGE_LEN) { 60011_xen3-auto-xen-drivers.patch1:+ if (strnlen(buf, BLKTAP2_MAX_MESSAGE_LEN) >= BLKTAP2_MAX_MESSAGE_LEN) { Based on that I think it has some level of blktap2 integration. Unfortunately its not clear if it has it all and if so at what point the code was taken from. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Fri, 2010-01-22 at 00:59 +0200, Pasi Kärkkäinen wrote:> On Fri, Jan 22, 2010 at 12:55:52AM +0200, Pasi Kärkkäinen wrote: > > On Thu, Jan 21, 2010 at 05:46:05PM -0500, David P. Quigley wrote: > > > On Fri, 2010-01-22 at 00:38 +0200, Pasi Kärkkäinen wrote: > > > > On Thu, Jan 21, 2010 at 05:27:45PM -0500, David P. Quigley wrote: > > > > > I''m currently working on moving storage services into their own domain > > > > > and I''ve been looking at blktap2. I''ve been trying to get an image > > > > > mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging > > > > > instead of quitting at the end. I haven''t removed any of the storage > > > > > startup code at this point so everything should be as it normally is in > > > > > xen-unstable. Is this a known bug? Are there some requirements on > > > > > blktap2 that I''m possibly not meeting? Is there anything in particular > > > > > that I should be looking at to try to debug the problem? > > > > > > > > > > > > > What kernel are you using? pv_ops dom0 kernel doesn''t have working blktap2 yet.. > > > > linux-2.6.18-xen has working blktap2 support. > > > > > > > > -- Pasi > > > > > > I''m using the patches maintained by Andrew Lyon that are applied to > > > 2.6.31.10. These are supposedly based on the OpenSUSE patches and can be > > > found at http://code.google.com/p/gentoo-xen-kernel/downloads/list. I > > > need a more recent kernel version so the 2.6.18 dom0 kernel isn''t an > > > option for me. > > > > > > > Ok. I''m not sure if those OpenSUSE patches have blktap2 support included? > > I know that at least Citrix XenServer, XCP and XCI 2.6.27 kernels do have blktap2. > > > > See: > > http://wiki.xensource.com/xenwiki/XenDom0Kernels > > > > More details about the pv_ops dom0 kernel (2.6.31.6 currently): It has initial version > of blktap2 patches, but they''d need some debugging, because they''re not yet working. > > -- PasiI haven''t had much success with pv_ops as a dom0 kernel in the past but it would be more beneficial to try fixing blktap2 in the context of that kernel then the Novell patches that I''m currently using. I''ll try to build the pv_ops kernel as a dom0 and if I can run it as a dom0 successfully I''ll see what I can try to do to get blktap2 working on it. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Fri, 2010-01-22 at 00:59 +0200, Pasi Kärkkäinen wrote:> On Fri, Jan 22, 2010 at 12:55:52AM +0200, Pasi Kärkkäinen wrote: > > On Thu, Jan 21, 2010 at 05:46:05PM -0500, David P. Quigley wrote: > > > On Fri, 2010-01-22 at 00:38 +0200, Pasi Kärkkäinen wrote: > > > > On Thu, Jan 21, 2010 at 05:27:45PM -0500, David P. Quigley wrote: > > > > > I''m currently working on moving storage services into their own domain > > > > > and I''ve been looking at blktap2. I''ve been trying to get an image > > > > > mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging > > > > > instead of quitting at the end. I haven''t removed any of the storage > > > > > startup code at this point so everything should be as it normally is in > > > > > xen-unstable. Is this a known bug? Are there some requirements on > > > > > blktap2 that I''m possibly not meeting? Is there anything in particular > > > > > that I should be looking at to try to debug the problem? > > > > > > > > > > > > > What kernel are you using? pv_ops dom0 kernel doesn''t have working blktap2 yet.. > > > > linux-2.6.18-xen has working blktap2 support. > > > > > > > > -- Pasi > > > > > > I''m using the patches maintained by Andrew Lyon that are applied to > > > 2.6.31.10. These are supposedly based on the OpenSUSE patches and can be > > > found at http://code.google.com/p/gentoo-xen-kernel/downloads/list. I > > > need a more recent kernel version so the 2.6.18 dom0 kernel isn''t an > > > option for me. > > > > > > > Ok. I''m not sure if those OpenSUSE patches have blktap2 support included? > > I know that at least Citrix XenServer, XCP and XCI 2.6.27 kernels do have blktap2. > > > > See: > > http://wiki.xensource.com/xenwiki/XenDom0Kernels > > > > More details about the pv_ops dom0 kernel (2.6.31.6 currently): It has initial version > of blktap2 patches, but they''d need some debugging, because they''re not yet working. > > -- PasiLooking under drivers through a make menuconfig on the pvops kernel I don''t see an entry for blktap2. I just see the entry for blktap. I know that blktap2 was supposed to replace blktap but in the Novell patches they are two separate config options and I don''t seem to be seeing the second on in the pvops kernel. Do I just enable blktap under drivers and use tap:tapdisk:<type> in the domU config to use blktap2? Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Thu, Jan 21, 2010 at 06:13:46PM -0500, David P. Quigley wrote:> On Fri, 2010-01-22 at 00:59 +0200, Pasi Kärkkäinen wrote: > > On Fri, Jan 22, 2010 at 12:55:52AM +0200, Pasi Kärkkäinen wrote: > > > On Thu, Jan 21, 2010 at 05:46:05PM -0500, David P. Quigley wrote: > > > > On Fri, 2010-01-22 at 00:38 +0200, Pasi Kärkkäinen wrote: > > > > > On Thu, Jan 21, 2010 at 05:27:45PM -0500, David P. Quigley wrote: > > > > > > I''m currently working on moving storage services into their own domain > > > > > > and I''ve been looking at blktap2. I''ve been trying to get an image > > > > > > mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging > > > > > > instead of quitting at the end. I haven''t removed any of the storage > > > > > > startup code at this point so everything should be as it normally is in > > > > > > xen-unstable. Is this a known bug? Are there some requirements on > > > > > > blktap2 that I''m possibly not meeting? Is there anything in particular > > > > > > that I should be looking at to try to debug the problem? > > > > > > > > > > > > > > > > What kernel are you using? pv_ops dom0 kernel doesn''t have working blktap2 yet.. > > > > > linux-2.6.18-xen has working blktap2 support. > > > > > > > > > > -- Pasi > > > > > > > > I''m using the patches maintained by Andrew Lyon that are applied to > > > > 2.6.31.10. These are supposedly based on the OpenSUSE patches and can be > > > > found at http://code.google.com/p/gentoo-xen-kernel/downloads/list. I > > > > need a more recent kernel version so the 2.6.18 dom0 kernel isn''t an > > > > option for me. > > > > > > > > > > Ok. I''m not sure if those OpenSUSE patches have blktap2 support included? > > > I know that at least Citrix XenServer, XCP and XCI 2.6.27 kernels do have blktap2. > > > > > > See: > > > http://wiki.xensource.com/xenwiki/XenDom0Kernels > > > > > > > More details about the pv_ops dom0 kernel (2.6.31.6 currently): It has initial version > > of blktap2 patches, but they''d need some debugging, because they''re not yet working. > > > > -- Pasi > > > Looking under drivers through a make menuconfig on the pvops kernel I > don''t see an entry for blktap2. I just see the entry for blktap. I know > that blktap2 was supposed to replace blktap but in the Novell patches > they are two separate config options and I don''t seem to be seeing the > second on in the pvops kernel. Do I just enable blktap under drivers and > use tap:tapdisk:<type> in the domU config to use blktap2? >See: http://git.kernel.org/?p=linux/kernel/git/jeremy/xen.git;a=shortlog;h=xen/master Commits like: 2009-12-10 Jeremy Fitzhardinge Merge branch ''xen/dom0/backend/blktap2'' into xen/master 2009-12-10 Jeremy Fitzhardinge xen/blktap2: don''t init on non-Xen system 2009-12-04 Ian Campbell blktap: cleanup usage of sysfs 2009-12-04 Ian Campbell blktap: wean off of use of the BlkBack pageflag. 2009-12-04 Ian Campbell Port blktap2 from 2.6.18 to tip. 2009-12-04 Ian Campbell Add build infrastrucuture for blktap2. 2009-12-04 Ian Campbell patch 03-blktap2-patch 2009-12-04 Ian Campbell patch 02-blkback-pagemap-patch 2009-12-04 Ian Campbell patch 01-blktap-patch I haven''t tried it myself yet.. -- Pasi _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On 01/21/2010 11:14 PM, Pasi Kärkkäinen wrote:> On Thu, Jan 21, 2010 at 06:13:46PM -0500, David P. Quigley wrote: > >> On Fri, 2010-01-22 at 00:59 +0200, Pasi Kärkkäinen wrote: >> >>> On Fri, Jan 22, 2010 at 12:55:52AM +0200, Pasi Kärkkäinen wrote: >>> >>>> On Thu, Jan 21, 2010 at 05:46:05PM -0500, David P. Quigley wrote: >>>> >>>>> On Fri, 2010-01-22 at 00:38 +0200, Pasi Kärkkäinen wrote: >>>>> >>>>>> On Thu, Jan 21, 2010 at 05:27:45PM -0500, David P. Quigley wrote: >>>>>> >>>>>>> I''m currently working on moving storage services into their own domain >>>>>>> and I''ve been looking at blktap2. I''ve been trying to get an image >>>>>>> mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging >>>>>>> instead of quitting at the end. I haven''t removed any of the storage >>>>>>> startup code at this point so everything should be as it normally is in >>>>>>> xen-unstable. Is this a known bug? Are there some requirements on >>>>>>> blktap2 that I''m possibly not meeting? Is there anything in particular >>>>>>> that I should be looking at to try to debug the problem? >>>>>>> >>>>>>> >>>>>> What kernel are you using? pv_ops dom0 kernel doesn''t have working blktap2 yet.. >>>>>> linux-2.6.18-xen has working blktap2 support. >>>>>> >>>>>> -- Pasi >>>>>> >>>>> I''m using the patches maintained by Andrew Lyon that are applied to >>>>> 2.6.31.10. These are supposedly based on the OpenSUSE patches and can be >>>>> found at http://code.google.com/p/gentoo-xen-kernel/downloads/list. I >>>>> need a more recent kernel version so the 2.6.18 dom0 kernel isn''t an >>>>> option for me. >>>>> >>>>> >>>> Ok. I''m not sure if those OpenSUSE patches have blktap2 support included? >>>> I know that at least Citrix XenServer, XCP and XCI 2.6.27 kernels do have blktap2. >>>> >>>> See: >>>> http://wiki.xensource.com/xenwiki/XenDom0Kernels >>>> >>>> >>> More details about the pv_ops dom0 kernel (2.6.31.6 currently): It has initial version >>> of blktap2 patches, but they''d need some debugging, because they''re not yet working. >>> >>> -- Pasi >>> >> >> Looking under drivers through a make menuconfig on the pvops kernel I >> don''t see an entry for blktap2. I just see the entry for blktap. I know >> that blktap2 was supposed to replace blktap but in the Novell patches >> they are two separate config options and I don''t seem to be seeing the >> second on in the pvops kernel. Do I just enable blktap under drivers and >> use tap:tapdisk:<type> in the domU config to use blktap2? >> >> > See: > http://git.kernel.org/?p=linux/kernel/git/jeremy/xen.git;a=shortlog;h=xen/master > > Commits like: > > 2009-12-10 Jeremy Fitzhardinge Merge branch ''xen/dom0/backend/blktap2'' into xen/master > 2009-12-10 Jeremy Fitzhardinge xen/blktap2: don''t init on non-Xen system > 2009-12-04 Ian Campbell blktap: cleanup usage of sysfs > 2009-12-04 Ian Campbell blktap: wean off of use of the BlkBack pageflag. > 2009-12-04 Ian Campbell Port blktap2 from 2.6.18 to tip. > 2009-12-04 Ian Campbell Add build infrastrucuture for blktap2. > 2009-12-04 Ian Campbell patch 03-blktap2-patch > 2009-12-04 Ian Campbell patch 02-blkback-pagemap-patch > 2009-12-04 Ian Campbell patch 01-blktap-patch > > I haven''t tried it myself yet.. >It is pretty broken at present, and I''ll probably commit a change to disable it for now. Fortunately it looks like Daniel Stodden is going to have a crack at fixing it up. J _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Fri, 2010-01-22 at 10:43 -0800, Jeremy Fitzhardinge wrote:> On 01/21/2010 11:14 PM, Pasi Kärkkäinen wrote: > > On Thu, Jan 21, 2010 at 06:13:46PM -0500, David P. Quigley wrote: > > > >> On Fri, 2010-01-22 at 00:59 +0200, Pasi Kärkkäinen wrote: > >> > >>> On Fri, Jan 22, 2010 at 12:55:52AM +0200, Pasi Kärkkäinen wrote: > >>> > >>>> On Thu, Jan 21, 2010 at 05:46:05PM -0500, David P. Quigley wrote: > >>>> > >>>>> On Fri, 2010-01-22 at 00:38 +0200, Pasi Kärkkäinen wrote: > >>>>> > >>>>>> On Thu, Jan 21, 2010 at 05:27:45PM -0500, David P. Quigley wrote: > >>>>>> > >>>>>>> I''m currently working on moving storage services into their own domain > >>>>>>> and I''ve been looking at blktap2. I''ve been trying to get an image > >>>>>>> mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging > >>>>>>> instead of quitting at the end. I haven''t removed any of the storage > >>>>>>> startup code at this point so everything should be as it normally is in > >>>>>>> xen-unstable. Is this a known bug? Are there some requirements on > >>>>>>> blktap2 that I''m possibly not meeting? Is there anything in particular > >>>>>>> that I should be looking at to try to debug the problem? > >>>>>>> > >>>>>>> > >>>>>> What kernel are you using? pv_ops dom0 kernel doesn''t have working blktap2 yet.. > >>>>>> linux-2.6.18-xen has working blktap2 support. > >>>>>> > >>>>>> -- Pasi > >>>>>> > >>>>> I''m using the patches maintained by Andrew Lyon that are applied to > >>>>> 2.6.31.10. These are supposedly based on the OpenSUSE patches and can be > >>>>> found at http://code.google.com/p/gentoo-xen-kernel/downloads/list. I > >>>>> need a more recent kernel version so the 2.6.18 dom0 kernel isn''t an > >>>>> option for me. > >>>>> > >>>>> > >>>> Ok. I''m not sure if those OpenSUSE patches have blktap2 support included? > >>>> I know that at least Citrix XenServer, XCP and XCI 2.6.27 kernels do have blktap2. > >>>> > >>>> See: > >>>> http://wiki.xensource.com/xenwiki/XenDom0Kernels > >>>> > >>>> > >>> More details about the pv_ops dom0 kernel (2.6.31.6 currently): It has initial version > >>> of blktap2 patches, but they''d need some debugging, because they''re not yet working. > >>> > >>> -- Pasi > >>> > >> > >> Looking under drivers through a make menuconfig on the pvops kernel I > >> don''t see an entry for blktap2. I just see the entry for blktap. I know > >> that blktap2 was supposed to replace blktap but in the Novell patches > >> they are two separate config options and I don''t seem to be seeing the > >> second on in the pvops kernel. Do I just enable blktap under drivers and > >> use tap:tapdisk:<type> in the domU config to use blktap2? > >> > >> > > See: > > http://git.kernel.org/?p=linux/kernel/git/jeremy/xen.git;a=shortlog;h=xen/master > > > > Commits like: > > > > 2009-12-10 Jeremy Fitzhardinge Merge branch ''xen/dom0/backend/blktap2'' into xen/master > > 2009-12-10 Jeremy Fitzhardinge xen/blktap2: don''t init on non-Xen system > > 2009-12-04 Ian Campbell blktap: cleanup usage of sysfs > > 2009-12-04 Ian Campbell blktap: wean off of use of the BlkBack pageflag. > > 2009-12-04 Ian Campbell Port blktap2 from 2.6.18 to tip. > > 2009-12-04 Ian Campbell Add build infrastrucuture for blktap2. > > 2009-12-04 Ian Campbell patch 03-blktap2-patch > > 2009-12-04 Ian Campbell patch 02-blkback-pagemap-patch > > 2009-12-04 Ian Campbell patch 01-blktap-patch > > > > I haven''t tried it myself yet.. > > > > It is pretty broken at present, and I''ll probably commit a change to > disable it for now. Fortunately it looks like Daniel Stodden is going > to have a crack at fixing it up. > > JWell if he needs any help feel free to ask. I definitely can justify spending time on working on this. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Thu, 2010-01-21 at 14:54 -0800, Daniel Stodden wrote:> On Thu, 2010-01-21 at 17:27 -0500, David P. Quigley wrote: > > I''m currently working on moving storage services into their own domain > > and I''ve been looking at blktap2. I''ve been trying to get an image > > mounted with blktap2 and for some odd reason and tapdisk2 keeps hanging > > instead of quitting at the end. I haven''t removed any of the storage > > startup code at this point so everything should be as it normally is in > > xen-unstable. Is this a known bug? Are there some requirements on > > blktap2 that I''m possibly not meeting? Is there anything in particular > > that I should be looking at to try to debug the problem? > > Are you running pvops? > Cheer up, after 2.6.23 it''s not you but just broken. > > The hang you encounter is likely blktap trying to remove a sysfs node > from within its own callback. That could be circumvented, but it''s > likely not the only issue. > > Daniel >So I tried getting the pvops kernel working and it works fine as a standard Linux kernel but when I try to run it on top of Xen once I get to Linux I don''t have any keyboard input. I''m using the grub conf on the pvops page so that part of the config should be correct. I''m going to try to update to the latest version of xen-unstable and of the pvops tree to see if that fixes anything. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
> So I tried getting the pvops kernel working and it works fine as a > standard Linux kernel but when I try to run it on top of Xen once I get > to Linux I don''t have any keyboard input. I''m using the grub conf on theThat smells like inittab..> pvops page so that part of the config should be correct. I''m going to > try to update to the latest version of xen-unstable and of the pvops > tree to see if that fixes anything.Here is a little script that fixes most common problems. grep -q xenfs /etc/fstab || echo "xenfs /proc/xen xenfs defaults 0 0" >> /etc/fstab if grep -q ata_piix /etc/modprobe.conf ; then NR=`cat /etc/modprobe.conf | grep scsi | wc -l` grep -q pata_oldpiix /etc/modprobe.conf || echo "alias scsi_hostadapter$NR pata_oldpiix" >> /etc/modprobe.conf fi grep -q hvc /etc/inittab || echo "co:2345:respawn:/sbin/agetty hvc0 9600 vt100-nav" >> /etc/inittab grep -q hvc0 /etc/securetty || echo "hvc0" >> /etc/securetty _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Fri, 2010-01-22 at 17:12 -0500, Konrad Rzeszutek Wilk wrote:> > So I tried getting the pvops kernel working and it works fine as a > > standard Linux kernel but when I try to run it on top of Xen once I get > > to Linux I don''t have any keyboard input. I''m using the grub conf on the > > That smells like inittab.. > > pvops page so that part of the config should be correct. I''m going to > > try to update to the latest version of xen-unstable and of the pvops > > tree to see if that fixes anything. > > > > Here is a little script that fixes most common problems. > > grep -q xenfs /etc/fstab || echo "xenfs /proc/xen xenfs defaults 0 0" >> /etc/fstab > if grep -q ata_piix /etc/modprobe.conf ; then > NR=`cat /etc/modprobe.conf | grep scsi | wc -l` > grep -q pata_oldpiix /etc/modprobe.conf || echo "alias scsi_hostadapter$NR pata_oldpiix" >> /etc/modprobe.conf > fi > grep -q hvc /etc/inittab || echo "co:2345:respawn:/sbin/agetty hvc0 9600 vt100-nav" >> /etc/inittab > grep -q hvc0 /etc/securetty || echo "hvc0" >> /etc/securetty >Is this for a domU or a dom0 pvops kernel? I didn''t think that the terminal for dom0 would be changed to hvc but I''m new to this so I could be wrong. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
> > Here is a little script that fixes most common problems. > > > > grep -q xenfs /etc/fstab || echo "xenfs /proc/xen xenfs defaults 0 0" >> /etc/fstab > > if grep -q ata_piix /etc/modprobe.conf ; then > > NR=`cat /etc/modprobe.conf | grep scsi | wc -l` > > grep -q pata_oldpiix /etc/modprobe.conf || echo "alias scsi_hostadapter$NR pata_oldpiix" >> /etc/modprobe.conf > > fi > > grep -q hvc /etc/inittab || echo "co:2345:respawn:/sbin/agetty hvc0 9600 vt100-nav" >> /etc/inittab > > grep -q hvc0 /etc/securetty || echo "hvc0" >> /etc/securetty > > > > > Is this for a domU or a dom0 pvops kernel? I didn''t think that theDom0. Thought the sections aobut /etc/inittab and seceretty can be used on DomU too.> terminal for dom0 would be changed to hvc but I''m new to this so I couldIt has. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Fri, 2010-01-22 at 17:43 -0500, Konrad Rzeszutek Wilk wrote:> > > Here is a little script that fixes most common problems. > > > > > > grep -q xenfs /etc/fstab || echo "xenfs /proc/xen xenfs defaults 0 0" >> /etc/fstab > > > if grep -q ata_piix /etc/modprobe.conf ; then > > > NR=`cat /etc/modprobe.conf | grep scsi | wc -l` > > > grep -q pata_oldpiix /etc/modprobe.conf || echo "alias scsi_hostadapter$NR pata_oldpiix" >> /etc/modprobe.conf > > > fi > > > grep -q hvc /etc/inittab || echo "co:2345:respawn:/sbin/agetty hvc0 9600 vt100-nav" >> /etc/inittab > > > grep -q hvc0 /etc/securetty || echo "hvc0" >> /etc/securetty > > > > > > > > > Is this for a domU or a dom0 pvops kernel? I didn''t think that the > > Dom0. Thought the sections aobut /etc/inittab and seceretty can be used > on DomU too. > > terminal for dom0 would be changed to hvc but I''m new to this so I could > > It has. > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-develI''ve been trying for the better part of today to get this working and I seem to keep hitting a wall. I finally managed to get the kernel booted but only with a serial console. I found out that some modules were missing so that might have caused the lack of console support but now the kernel won''t boot to a prompt at all. I keep running into a null dereference in the uhci driver. I can try to rebase Jeremy''s patches onto a newer kernel.org version but I''d first like to check to make sure this isn''t already a known problem with a solution. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Mon, Jan 25, 2010 at 06:23:55PM -0500, David P. Quigley wrote:> On Fri, 2010-01-22 at 17:43 -0500, Konrad Rzeszutek Wilk wrote: > > > > Here is a little script that fixes most common problems. > > > > > > > > grep -q xenfs /etc/fstab || echo "xenfs /proc/xen xenfs defaults 0 0" >> /etc/fstab > > > > if grep -q ata_piix /etc/modprobe.conf ; then > > > > NR=`cat /etc/modprobe.conf | grep scsi | wc -l` > > > > grep -q pata_oldpiix /etc/modprobe.conf || echo "alias scsi_hostadapter$NR pata_oldpiix" >> /etc/modprobe.conf > > > > fi > > > > grep -q hvc /etc/inittab || echo "co:2345:respawn:/sbin/agetty hvc0 9600 vt100-nav" >> /etc/inittab > > > > grep -q hvc0 /etc/securetty || echo "hvc0" >> /etc/securetty > > > > > > > > > > > > > Is this for a domU or a dom0 pvops kernel? I didn''t think that the > > > > Dom0. Thought the sections aobut /etc/inittab and seceretty can be used > > on DomU too. > > > terminal for dom0 would be changed to hvc but I''m new to this so I could > > > > It has. > > > > _______________________________________________ > > Xen-devel mailing list > > Xen-devel@lists.xensource.com > > http://lists.xensource.com/xen-devel > > I''ve been trying for the better part of today to get this working and I > seem to keep hitting a wall. I finally managed to get the kernel booted > but only with a serial console. I found out that some modules were > missing so that might have caused the lack of console support but now > the kernel won''t boot to a prompt at all. I keep running into a null > dereference in the uhci driver. I can try to rebase Jeremy''s patches > onto a newer kernel.org version but I''d first like to check to make sure > this isn''t already a known problem with a solution. >You can try my .config files for pv_ops dom0 kernel, available from here: http://wiki.xensource.com/xenwiki/XenParavirtOps in the "Troubleshooting" section.. just use "make oldconfig" with them. -- Pasi _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
> I''ve been trying for the better part of today to get this working and I > seem to keep hitting a wall. I finally managed to get the kernel booted > but only with a serial console. I found out that some modules wereCare to send it?> missing so that might have caused the lack of console support but now > the kernel won''t boot to a prompt at all. I keep running into a null > dereference in the uhci driver. I can try to rebase Jeremy''s patchesThat is not good. First time I hear of it. You could compile most of the USB drivers in your kernel, and also the storage controller ones.> onto a newer kernel.org version but I''d first like to check to make sure > this isn''t already a known problem with a solution.What was your underlaying distro? Fedora? Debian?> > Dave >_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Tue, 2010-01-26 at 09:10 +0200, Pasi Kärkkäinen wrote:> On Mon, Jan 25, 2010 at 06:23:55PM -0500, David P. Quigley wrote: > > On Fri, 2010-01-22 at 17:43 -0500, Konrad Rzeszutek Wilk wrote: > > > > > Here is a little script that fixes most common problems. > > > > > > > > > > grep -q xenfs /etc/fstab || echo "xenfs /proc/xen xenfs defaults 0 0" >> /etc/fstab > > > > > if grep -q ata_piix /etc/modprobe.conf ; then > > > > > NR=`cat /etc/modprobe.conf | grep scsi | wc -l` > > > > > grep -q pata_oldpiix /etc/modprobe.conf || echo "alias scsi_hostadapter$NR pata_oldpiix" >> /etc/modprobe.conf > > > > > fi > > > > > grep -q hvc /etc/inittab || echo "co:2345:respawn:/sbin/agetty hvc0 9600 vt100-nav" >> /etc/inittab > > > > > grep -q hvc0 /etc/securetty || echo "hvc0" >> /etc/securetty > > > > > > > > > > > > > > > > > Is this for a domU or a dom0 pvops kernel? I didn''t think that the > > > > > > Dom0. Thought the sections aobut /etc/inittab and seceretty can be used > > > on DomU too. > > > > terminal for dom0 would be changed to hvc but I''m new to this so I could > > > > > > It has. > > > > > > _______________________________________________ > > > Xen-devel mailing list > > > Xen-devel@lists.xensource.com > > > http://lists.xensource.com/xen-devel > > > > I''ve been trying for the better part of today to get this working and I > > seem to keep hitting a wall. I finally managed to get the kernel booted > > but only with a serial console. I found out that some modules were > > missing so that might have caused the lack of console support but now > > the kernel won''t boot to a prompt at all. I keep running into a null > > dereference in the uhci driver. I can try to rebase Jeremy''s patches > > onto a newer kernel.org version but I''d first like to check to make sure > > this isn''t already a known problem with a solution. > > > > You can try my .config files for pv_ops dom0 kernel, available from here: > http://wiki.xensource.com/xenwiki/XenParavirtOps > > in the "Troubleshooting" section.. just use "make oldconfig" with them. > > -- PasiThanks for the suggestion. I am already using your config though so its not clear to me if its a config issue. Once I got that working and booting I keep getting null pointer dereferences in some other code thats not xen related. The only reason it might be affected by xen is because its complaining about something with the PCI space. The error text is below. uhci_hcd 0000:00:1a.1: PCI INT B -> GSI 17 (level, low) -> IRQ 17 uhci_hcd 0000:00:1a.1: UHCI Host Controller uhci_hcd 0000:00:1a.1: new USB bus registered, assigned bus number 3 uhci_hcd 0000:00:1a.1: irq 17, io base 0x0000ff00 uhci_hcd 0000:00:1a.1: host system error, PCI problems? uhci_hcd 0000:00:1a.1: host controller halted, very bad! uhci_hcd 0000:00:1a.1: HC died; cleaning up _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Tue, 2010-01-26 at 09:42 -0500, Konrad Rzeszutek Wilk wrote:> > I''ve been trying for the better part of today to get this working and I > > seem to keep hitting a wall. I finally managed to get the kernel booted > > but only with a serial console. I found out that some modules were > > Care to send it?The kernel or the config? I''m using the config given on the pvops wiki page and I haven''t made any modifications. Although I''m not sure if SELinux is enabled so I might have to change that.> > missing so that might have caused the lack of console support but now > > the kernel won''t boot to a prompt at all. I keep running into a null > > dereference in the uhci driver. I can try to rebase Jeremy''s patches > > That is not good. First time I hear of it. You could compile most of the > USB drivers in your kernel, and also the storage controller ones.Yea I found this to be a little weird considering on other kernels the uhci driver has no problems at all. Someone here thinks it might be an issue with the ACPI aperture with things not lining up. I''m running an optiplex 960 for my xen development box.> > > onto a newer kernel.org version but I''d first like to check to make sure > > this isn''t already a known problem with a solution. > > What was your underlaying distro? Fedora? Debian? > > > > Dave > >So I''m running on an updated Fedora 11 box. I''ve grabbed Jeremy''s pvops kernel from kernel.org and I''m using the latest xen-unstable tree as of yesterday at 5pm. Several of the uhci host controller sections in dmesg are reporting the same error but I have copied some of the error text with the kernel oops below. uhci_hcd 0000:00:1a.1: PCI INT B -> GSI 17 (level, low) -> IRQ 17 uhci_hcd 0000:00:1a.1: UHCI Host Controller uhci_hcd 0000:00:1a.1: new USB bus registered, assigned bus number 3 uhci_hcd 0000:00:1a.1: irq 17, io base 0x0000ff00 uhci_hcd 0000:00:1a.1: host system error, PCI problems? uhci_hcd 0000:00:1a.1: host controller halted, very bad! uhci_hcd 0000:00:1a.1: HC died; cleaning up usb usb3: unable to read config index 0 descriptor/start: -108 usb usb3: chopping to 0 config(s) usb usb3: string descriptor 0 read error: -108 ata3: SATA link down (SStatus 4 SControl 300) ata6: SATA link down (SStatus 0 SControl 300) usb usb3: New USB device found, idVendor=1d6b, idProduct=0001 usb usb3: New USB device strings: Mfr=3, Product=2, SerialNumber=1 usb usb3: no configuration chosen from 0 choices uhci_hcd 0000:00:1a.1: HC died; cleaning up BUG: unable to handle kernel NULL pointer dereference at 0000000000000098 IP: [<ffffffff813be3d2>] usb_kick_khubd+0x23/0x4c PGD 0 Oops: 0000 [#1] SMP last sysfs file: CPU 0 Modules linked in: Pid: 112, comm: work_for_cpu Not tainted 2.6.31.6-pvops-dom0 #6 OptiPlex 960 RIP: e030:[<ffffffff813be3d2>] [<ffffffff813be3d2>] usb_kick_khubd+0x23/0x4c RSP: e02b:ffff8801dd071d40 EFLAGS: 00010046 RAX: 0000000000000000 RBX: ffff8801dd1a0000 RCX: 00000000000001af RDX: ffff8801dd7d4880 RSI: ffff8801dd1a0848 RDI: ffff8801dd1a0848 RBP: ffff8801dd071d50 R08: 000000000000000c R09: ffff8801dd071bf0 R10: 00000002c6caa903 R11: 0000000000000144 R12: 0000000000000200 R13: ffff8801dd1a0848 R14: ffff8801dd1a0848 R15: 0000000000000011 FS: 0000000000000000(0000) GS:ffffc90000000000(0000) knlGS:0000000000000000 CS: e033 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000000098 CR3: 0000000001001000 CR4: 0000000000002660 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process work_for_cpu (pid: 112, threadinfo ffff8801dd070000, task ffff8801dd7d4880) Stack: 0000000000000200 00000000035b2d49 ffff8801dd071d80 ffffffff813c422b <0> ffff8801dd071d80 00000000035b2d49 ffff8801dd1a0000 0000000000000000 <0> ffff8801dd071df0 ffffffff813c55f6 40ff8801dd071df0 000000000000ff00 Call Trace: [<ffffffff813c422b>] usb_hc_died+0x78/0xa3 [<ffffffff813c55f6>] usb_add_hcd+0x544/0x6a8 [<ffffffff813d5206>] usb_hcd_pci_probe+0x263/0x3bd [<ffffffff8107e91c>] ? do_work_for_cpu+0x0/0x50 [<ffffffff8129658f>] local_pci_probe+0x2a/0x42 [<ffffffff8107e943>] do_work_for_cpu+0x27/0x50 [<ffffffff81083c80>] kthread+0xac/0xb4 [<ffffffff810161aa>] child_rip+0xa/0x20 [<ffffffff81015b10>] ? restore_args+0x0/0x30 [<ffffffff810161a0>] ? child_rip+0x0/0x20 Code: 18 5b 41 5c 41 5d c9 c3 55 48 89 e5 48 83 ec 10 0f 1f 44 00 00 65 48 8b 04 25 28 00 00 00 48 89 45 f8 31 c0 48 8b 87 88 02 00 00 <48> 8b 80 98 00 00 00 48 8b b8 f8 00 00 00 e8 6d fe ff ff 48 8b RIP [<ffffffff813be3d2>] usb_kick_khubd+0x23/0x4c RSP <ffff8801dd071d40> CR2: 0000000000000098 ---[ end trace cf0e97f02af8794f ]--- _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Tue, Jan 26, 2010 at 11:32:58AM -0500, David P. Quigley wrote:> On Tue, 2010-01-26 at 09:10 +0200, Pasi Kärkkäinen wrote: > > On Mon, Jan 25, 2010 at 06:23:55PM -0500, David P. Quigley wrote: > > > On Fri, 2010-01-22 at 17:43 -0500, Konrad Rzeszutek Wilk wrote: > > > > > > Here is a little script that fixes most common problems. > > > > > > > > > > > > grep -q xenfs /etc/fstab || echo "xenfs /proc/xen xenfs defaults 0 0" >> /etc/fstab > > > > > > if grep -q ata_piix /etc/modprobe.conf ; then > > > > > > NR=`cat /etc/modprobe.conf | grep scsi | wc -l` > > > > > > grep -q pata_oldpiix /etc/modprobe.conf || echo "alias scsi_hostadapter$NR pata_oldpiix" >> /etc/modprobe.conf > > > > > > fi > > > > > > grep -q hvc /etc/inittab || echo "co:2345:respawn:/sbin/agetty hvc0 9600 vt100-nav" >> /etc/inittab > > > > > > grep -q hvc0 /etc/securetty || echo "hvc0" >> /etc/securetty > > > > > > > > > > > > > > > > > > > > > Is this for a domU or a dom0 pvops kernel? I didn''t think that the > > > > > > > > Dom0. Thought the sections aobut /etc/inittab and seceretty can be used > > > > on DomU too. > > > > > terminal for dom0 would be changed to hvc but I''m new to this so I could > > > > > > > > It has. > > > > > > > > _______________________________________________ > > > > Xen-devel mailing list > > > > Xen-devel@lists.xensource.com > > > > http://lists.xensource.com/xen-devel > > > > > > I''ve been trying for the better part of today to get this working and I > > > seem to keep hitting a wall. I finally managed to get the kernel booted > > > but only with a serial console. I found out that some modules were > > > missing so that might have caused the lack of console support but now > > > the kernel won''t boot to a prompt at all. I keep running into a null > > > dereference in the uhci driver. I can try to rebase Jeremy''s patches > > > onto a newer kernel.org version but I''d first like to check to make sure > > > this isn''t already a known problem with a solution. > > > > > > > You can try my .config files for pv_ops dom0 kernel, available from here: > > http://wiki.xensource.com/xenwiki/XenParavirtOps > > > > in the "Troubleshooting" section.. just use "make oldconfig" with them. > > > > -- Pasi > > Thanks for the suggestion. I am already using your config though so its > not clear to me if its a config issue. Once I got that working and > booting I keep getting null pointer dereferences in some other code > thats not xen related. The only reason it might be affected by xen is > because its complaining about something with the PCI space. The error > text is below. > > uhci_hcd 0000:00:1a.1: PCI INT B -> GSI 17 (level, low) -> IRQ 17 > uhci_hcd 0000:00:1a.1: UHCI Host Controller > uhci_hcd 0000:00:1a.1: new USB bus registered, assigned bus number 3 > uhci_hcd 0000:00:1a.1: irq 17, io base 0x0000ff00 > uhci_hcd 0000:00:1a.1: host system error, PCI problems? > uhci_hcd 0000:00:1a.1: host controller halted, very bad! > uhci_hcd 0000:00:1a.1: HC died; cleaning up >Uh oh :) That looks bad.. I haven''t seen that on my testboxes.. so dunno really what would help. Is Xen complaining something when that happens? Check "xm dmesg". Do you have "loglvl=all guest_loglvl=all" options for xen.gz in grub.conf? -- Pasi _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Tue, 2010-01-26 at 18:47 +0200, Pasi Kärkkäinen wrote:> On Tue, Jan 26, 2010 at 11:32:58AM -0500, David P. Quigley wrote: > > On Tue, 2010-01-26 at 09:10 +0200, Pasi Kärkkäinen wrote: > > > On Mon, Jan 25, 2010 at 06:23:55PM -0500, David P. Quigley wrote: > > > > On Fri, 2010-01-22 at 17:43 -0500, Konrad Rzeszutek Wilk wrote: > > > > > > > Here is a little script that fixes most common problems. > > > > > > > > > > > > > > grep -q xenfs /etc/fstab || echo "xenfs /proc/xen xenfs defaults 0 0" >> /etc/fstab > > > > > > > if grep -q ata_piix /etc/modprobe.conf ; then > > > > > > > NR=`cat /etc/modprobe.conf | grep scsi | wc -l` > > > > > > > grep -q pata_oldpiix /etc/modprobe.conf || echo "alias scsi_hostadapter$NR pata_oldpiix" >> /etc/modprobe.conf > > > > > > > fi > > > > > > > grep -q hvc /etc/inittab || echo "co:2345:respawn:/sbin/agetty hvc0 9600 vt100-nav" >> /etc/inittab > > > > > > > grep -q hvc0 /etc/securetty || echo "hvc0" >> /etc/securetty > > > > > > > > > > > > > > > > > > > > > > > > > Is this for a domU or a dom0 pvops kernel? I didn''t think that the > > > > > > > > > > Dom0. Thought the sections aobut /etc/inittab and seceretty can be used > > > > > on DomU too. > > > > > > terminal for dom0 would be changed to hvc but I''m new to this so I could > > > > > > > > > > It has. > > > > > > > > > > _______________________________________________ > > > > > Xen-devel mailing list > > > > > Xen-devel@lists.xensource.com > > > > > http://lists.xensource.com/xen-devel > > > > > > > > I''ve been trying for the better part of today to get this working and I > > > > seem to keep hitting a wall. I finally managed to get the kernel booted > > > > but only with a serial console. I found out that some modules were > > > > missing so that might have caused the lack of console support but now > > > > the kernel won''t boot to a prompt at all. I keep running into a null > > > > dereference in the uhci driver. I can try to rebase Jeremy''s patches > > > > onto a newer kernel.org version but I''d first like to check to make sure > > > > this isn''t already a known problem with a solution. > > > > > > > > > > You can try my .config files for pv_ops dom0 kernel, available from here: > > > http://wiki.xensource.com/xenwiki/XenParavirtOps > > > > > > in the "Troubleshooting" section.. just use "make oldconfig" with them. > > > > > > -- Pasi > > > > Thanks for the suggestion. I am already using your config though so its > > not clear to me if its a config issue. Once I got that working and > > booting I keep getting null pointer dereferences in some other code > > thats not xen related. The only reason it might be affected by xen is > > because its complaining about something with the PCI space. The error > > text is below. > > > > uhci_hcd 0000:00:1a.1: PCI INT B -> GSI 17 (level, low) -> IRQ 17 > > uhci_hcd 0000:00:1a.1: UHCI Host Controller > > uhci_hcd 0000:00:1a.1: new USB bus registered, assigned bus number 3 > > uhci_hcd 0000:00:1a.1: irq 17, io base 0x0000ff00 > > uhci_hcd 0000:00:1a.1: host system error, PCI problems? > > uhci_hcd 0000:00:1a.1: host controller halted, very bad! > > uhci_hcd 0000:00:1a.1: HC died; cleaning up > > > > Uh oh :) That looks bad.. I haven''t seen that on my testboxes.. > so dunno really what would help. > > Is Xen complaining something when that happens? Check "xm dmesg". > > Do you have "loglvl=all guest_loglvl=all" options for xen.gz in grub.conf? > > -- PasiI don''t see anything that immediately jumps out at me but I''m new to this so I may be missing something. My grub config entry is below and I have attached the entire output from boot to the oops that I get on my serial console. title Fedora (2.6.31.6-pvops-dom0-dontedit) kernel /xen-4.0.gz console=com1,hvc0 com1=115200,8n1 loglvl=all guest_loglvl=all sync_console console_to_ring module /vmlinuz-2.6.31.6-pvops-dom0 ro root=/dev/mapper/VolGroup-lv_root console=com1,hvc0 com1=115200,8n1 earlyprintk=xen module /initrd-2.6.31.6-pvops-dom0.img module /xenpolicy.24 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Tue, Jan 26, 2010 at 11:42:52AM -0500, David P. Quigley wrote:> On Tue, 2010-01-26 at 09:42 -0500, Konrad Rzeszutek Wilk wrote: > > > I''ve been trying for the better part of today to get this working and I > > > seem to keep hitting a wall. I finally managed to get the kernel booted > > > but only with a serial console. I found out that some modules were > > > > Care to send it? > > The kernel or the config? I''m using the config given on the pvops wikiThe kernel output. Can you boot the pvops kernel without Xen and provide the ''dmesg'' and ''cat /proc/interrupts'' output please? _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Tue, Jan 26, 2010 at 11:49:32AM -0500, David P. Quigley wrote:> > > > > > > > > > I''ve been trying for the better part of today to get this working and I > > > > > seem to keep hitting a wall. I finally managed to get the kernel booted > > > > > but only with a serial console. I found out that some modules were > > > > > missing so that might have caused the lack of console support but now > > > > > the kernel won''t boot to a prompt at all. I keep running into a null > > > > > dereference in the uhci driver. I can try to rebase Jeremy''s patches > > > > > onto a newer kernel.org version but I''d first like to check to make sure > > > > > this isn''t already a known problem with a solution. > > > > > > > > > > > > > You can try my .config files for pv_ops dom0 kernel, available from here: > > > > http://wiki.xensource.com/xenwiki/XenParavirtOps > > > > > > > > in the "Troubleshooting" section.. just use "make oldconfig" with them. > > > > > > > > -- Pasi > > > > > > Thanks for the suggestion. I am already using your config though so its > > > not clear to me if its a config issue. Once I got that working and > > > booting I keep getting null pointer dereferences in some other code > > > thats not xen related. The only reason it might be affected by xen is > > > because its complaining about something with the PCI space. The error > > > text is below. > > > > > > uhci_hcd 0000:00:1a.1: PCI INT B -> GSI 17 (level, low) -> IRQ 17 > > > uhci_hcd 0000:00:1a.1: UHCI Host Controller > > > uhci_hcd 0000:00:1a.1: new USB bus registered, assigned bus number 3 > > > uhci_hcd 0000:00:1a.1: irq 17, io base 0x0000ff00 > > > uhci_hcd 0000:00:1a.1: host system error, PCI problems? > > > uhci_hcd 0000:00:1a.1: host controller halted, very bad! > > > uhci_hcd 0000:00:1a.1: HC died; cleaning up > > > > > > > Uh oh :) That looks bad.. I haven''t seen that on my testboxes.. > > so dunno really what would help. > > > > Is Xen complaining something when that happens? Check "xm dmesg". > > > > Do you have "loglvl=all guest_loglvl=all" options for xen.gz in grub.conf? > > > > -- Pasi > > > I don''t see anything that immediately jumps out at me but I''m new to > this so I may be missing something. My grub config entry is below and I > have attached the entire output from boot to the oops that I get on my > serial console. > > title Fedora (2.6.31.6-pvops-dom0-dontedit) > kernel /xen-4.0.gz console=com1,hvc0 com1=115200,8n1 loglvl=all guest_loglvl=all sync_console console_to_ring > module /vmlinuz-2.6.31.6-pvops-dom0 ro root=/dev/mapper/VolGroup-lv_root console=com1,hvc0 com1=115200,8n1 earlyprintk=xen > module /initrd-2.6.31.6-pvops-dom0.img > module /xenpolicy.24 >There are some extra options, but those shouldn''t be the problem. Cleaned up grub.conf: title Fedora (2.6.31.6-pvops-dom0-dontedit) kernel /xen-4.0.gz console=com1 com1=115200,8n1 loglvl=all guest_loglvl=all sync_console console_to_ring module /vmlinuz-2.6.31.6-pvops-dom0 ro root=/dev/mapper/VolGroup-lv_root console=hvc0 earlyprintk=xen module /initrd-2.6.31.6-pvops-dom0.img module /xenpolicy.24 Maybe try adding dom0_mem=1024M to the xen-4.0.gz line.. that might have some effect.> ehci_hcd: USB 2.0 ''Enhanced'' Host Controller (EHCI) Driver > xen_set_ioapic_routing: irq 22 gsi 22 vector 22 ioapic 0 pin 22 triggering 1 polarity 1 > ehci_hcd 0000:00:1a.7: PCI INT C -> GSI 22 (level, low) -> IRQ 22 > ehci_hcd 0000:00:1a.7: EHCI Host Controller > ehci_hcd 0000:00:1a.7: new USB bus registered, assigned bus number 1 > ehci_hcd 0000:00:1a.7: debug port 1 > ehci_hcd 0000:00:1a.7: irq 22, io mem 0xfebda000 > ehci_hcd 0000:00:1a.7: USB 2.0 started, EHCI 1.00 > usb usb1: New USB device found, idVendor=1d6b, idProduct=0002 > usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1 > usb usb1: Product: EHCI Host Controller > usb usb1: Manufacturer: Linux 2.6.31.6-pvops-dom0 ehci_hcd > usb usb1: SerialNumber: 0000:00:1a.7 > usb usb1: configuration #1 chosen from 1 choice > hub 1-0:1.0: USB hub found > hub 1-0:1.0: 6 ports detected > work_for_cpu used greatest stack depth: 3888 bytes left > xen_set_ioapic_routing: irq 23 gsi 23 vector 23 ioapic 0 pin 23 triggering 1 polarity 1 > ehci_hcd 0000:00:1d.7: PCI INT A -> GSI 23 (level, low) -> IRQ 23 > ehci_hcd 0000:00:1d.7: EHCI Host Controller > ehci_hcd 0000:00:1d.7: new USB bus registered, assigned bus number 2 > ehci_hcd 0000:00:1d.7: debug port 1 > ehci_hcd 0000:00:1d.7: irq 23, io mem 0xff980000 > ehci_hcd 0000:00:1d.7: USB 2.0 started, EHCI 1.00 > usb usb2: New USB device found, idVendor=1d6b, idProduct=0002 > usb usb2: New USB device strings: Mfr=3, Product=2, SerialNumber=1 > usb usb2: Product: EHCI Host Controller > usb usb2: Manufacturer: Linux 2.6.31.6-pvops-dom0 ehci_hcd > usb usb2: SerialNumber: 0000:00:1d.7 > usb usb2: configuration #1 chosen from 1 choice > hub 2-0:1.0: USB hub found > hub 2-0:1.0: 6 ports detected > ohci_hcd: USB 1.1 ''Open'' Host Controller (OHCI) Driver > uhci_hcd: USB Universal Host Controller Interface driver > xen_allocate_pirq: returning irq 16 for gsi 16 > xen_set_ioapic_routing: irq 16 gsi 16 vector 16 ioapic 0 pin 16 triggering 1 polarity 1 > uhci_hcd 0000:00:1a.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16 > uhci_hcd 0000:00:1a.0: UHCI Host Controller > uhci_hcd 0000:00:1a.0: new USB bus registered, assigned bus number 3 > uhci_hcd 0000:00:1a.0: irq 16, io base 0x0000ff20 > uhci_hcd 0000:00:1a.0: host system error, PCI problems? > uhci_hcd 0000:00:1a.0: host controller halted, very bad! > uhci_hcd 0000:00:1a.0: HC died; cleaning up > uhci_hcd 0000:00:1a.0: USB bus 3 deregistered > uhci_hcd 0000:00:1a.0: PCI INT A disabled > uhci_hcd 0000:00:1a.0: init 0000:00:1a.0 fail, -108 > uhci_hcd: probe of 0000:00:1a.0 failed with error -108 > xen_allocate_pirq: returning irq 17 for gsi 17 > xen_set_ioapic_routing: irq 17 gsi 17 vector 17 ioapic 0 pin 17 triggering 1 polarity 1 > uhci_hcd 0000:00:1a.1: PCI INT B -> GSI 17 (level, low) -> IRQ 17 > uhci_hcd 0000:00:1a.1: UHCI Host Controller > uhci_hcd 0000:00:1a.1: new USB bus registered, assigned bus number 3 > uhci_hcd 0000:00:1a.1: irq 17, io base 0x0000ff00 > uhci_hcd 0000:00:1a.1: host system error, PCI problems? > uhci_hcd 0000:00:1a.1: host controller halted, very bad! > uhci_hcd 0000:00:1a.1: HC died; cleaning up > usb usb3: unable to read config index 0 descriptor/start: -108 > usb usb3: chopping to 0 config(s) > usb usb3: string descriptor 0 read error: -108 > ata3: SATA link down (SStatus 4 SControl 300) > ata6: SATA link down (SStatus 0 SControl 300) > usb usb3: New USB device found, idVendor=1d6b, idProduct=0001 > usb usb3: New USB device strings: Mfr=3, Product=2, SerialNumber=1 > usb usb3: no configuration chosen from 0 choices > uhci_hcd 0000:00:1a.1: HC died; cleaning up > BUG: unable to handle kernel NULL pointer dereference at 0000000000000098 > IP: [<ffffffff813be3d2>] usb_kick_khubd+0x23/0x4c > PGD 0 > Oops: 0000 [#1] SMP > last sysfs file: > CPU 0 > Modules linked in: > Pid: 112, comm: work_for_cpu Not tainted 2.6.31.6-pvops-dom0 #6 OptiPlex 960 > RIP: e030:[<ffffffff813be3d2>] [<ffffffff813be3d2>] usb_kick_khubd+0x23/0x4c > RSP: e02b:ffff8801dd071d40 EFLAGS: 00010046 > RAX: 0000000000000000 RBX: ffff8801dd1a0000 RCX: 00000000000001af > RDX: ffff8801dd7d4880 RSI: ffff8801dd1a0848 RDI: ffff8801dd1a0848 > RBP: ffff8801dd071d50 R08: 000000000000000c R09: ffff8801dd071bf0 > R10: 00000002c6caa903 R11: 0000000000000144 R12: 0000000000000200 > R13: ffff8801dd1a0848 R14: ffff8801dd1a0848 R15: 0000000000000011 > FS: 0000000000000000(0000) GS:ffffc90000000000(0000) knlGS:0000000000000000 > CS: e033 DS: 0000 ES: 0000 CR0: 000000008005003b > CR2: 0000000000000098 CR3: 0000000001001000 CR4: 0000000000002660 > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 > DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 > Process work_for_cpu (pid: 112, threadinfo ffff8801dd070000, task ffff8801dd7d4880) > Stack: > 0000000000000200 00000000035b2d49 ffff8801dd071d80 ffffffff813c422b > <0> ffff8801dd071d80 00000000035b2d49 ffff8801dd1a0000 0000000000000000 > <0> ffff8801dd071df0 ffffffff813c55f6 40ff8801dd071df0 000000000000ff00 > Call Trace: > [<ffffffff813c422b>] usb_hc_died+0x78/0xa3 > [<ffffffff813c55f6>] usb_add_hcd+0x544/0x6a8 > [<ffffffff813d5206>] usb_hcd_pci_probe+0x263/0x3bd > [<ffffffff8107e91c>] ? do_work_for_cpu+0x0/0x50 > [<ffffffff8129658f>] local_pci_probe+0x2a/0x42 > [<ffffffff8107e943>] do_work_for_cpu+0x27/0x50 > [<ffffffff81083c80>] kthread+0xac/0xb4 > [<ffffffff810161aa>] child_rip+0xa/0x20 > [<ffffffff81015b10>] ? restore_args+0x0/0x30 > [<ffffffff810161a0>] ? child_rip+0x0/0x20 > Code: 18 5b 41 5c 41 5d c9 c3 55 48 89 e5 48 83 ec 10 0f 1f 44 00 00 65 48 8b 04 25 28 00 00 00 48 89 45 f8 31 c0 48 8b 87 88 02 00 00 <48> 8b 80 98 00 00 00 48 8b b8 f8 00 00 00 e8 6d fe ff ff 48 8b > RIP [<ffffffff813be3d2>] usb_kick_khubd+0x23/0x4c > RSP <ffff8801dd071d40> > CR2: 0000000000000098 > ---[ end trace cf0e97f02af8794f ]--- > BUG: spinlock lockup on CPU#1, khubd/42, ffffffff817fa050 > Pid: 42, comm: khubd Tainted: G D 2.6.31.6-pvops-dom0 #6 > Call Trace: > [<ffffffff812852f4>] _raw_spin_lock+0x118/0x156 > [<ffffffff81518854>] _spin_lock_irq+0x71/0x9a > [<ffffffff813c5dc9>] ? usb_hcd_submit_urb+0x304/0xa07 > [<ffffffff813c6c60>] ? usb_get_urb+0x2a/0x48 > [<ffffffff813c5dc9>] usb_hcd_submit_urb+0x304/0xa07 > [<ffffffff8100eb68>] ? xen_force_evtchn_callback+0x20/0x36 > [<ffffffff81074a4e>] ? lock_timer_base+0x3f/0x7d > [<ffffffff81074a4e>] ? lock_timer_base+0x3f/0x7d > [<ffffffff8100eb68>] ? xen_force_evtchn_callback+0x20/0x36 > [<ffffffff81097b93>] ? lockdep_init_map+0xad/0x138 > [<ffffffff813c6ba3>] usb_submit_urb+0x25a/0x2ed > [<ffffffff81084577>] ? __init_waitqueue_head+0x4d/0x76 > [<ffffffff813c8509>] usb_start_wait_urb+0x71/0x1d4 > [<ffffffff813c8907>] usb_control_msg+0x10c/0x144 > [<ffffffff813bd9c4>] hub_port_status+0x99/0x137 > [<ffffffff813bdd21>] hub_port_reset+0xa9/0x2f9 > [<ffffffff813bf346>] hub_port_init+0x100/0x7ee > [<ffffffff8100f61f>] ? xen_restore_fl_direct_end+0x0/0x1 > [<ffffffff81518489>] ? _spin_unlock_irqrestore+0x64/0x7f > [<ffffffff813c2bae>] hub_events+0x94d/0x11e6 > [<ffffffff8100f632>] ? check_events+0x12/0x20 > [<ffffffff8100eb68>] ? xen_force_evtchn_callback+0x20/0x36 > [<ffffffff813c348d>] hub_thread+0x46/0x1d1 > [<ffffffff810840d3>] ? autoremove_wake_function+0x0/0x5f > [<ffffffff813c3447>] ? hub_thread+0x0/0x1d1 > [<ffffffff81083c80>] kthread+0xac/0xb4 > [<ffffffff810161aa>] child_rip+0xa/0x20 > [<ffffffff81015b10>] ? restore_args+0x0/0x30 > [<ffffffff810161a0>] ? child_rip+0x0/0x20 >Uh oh.. hopefully someone else has more ideas. I can only suggest trying to limit dom0 memory (dom0_mem=1024M) and see if that helps. -- Pasi _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Tue, 2010-01-26 at 11:57 -0500, Konrad Rzeszutek Wilk wrote:> On Tue, Jan 26, 2010 at 11:42:52AM -0500, David P. Quigley wrote: > > On Tue, 2010-01-26 at 09:42 -0500, Konrad Rzeszutek Wilk wrote: > > > > I''ve been trying for the better part of today to get this working and I > > > > seem to keep hitting a wall. I finally managed to get the kernel booted > > > > but only with a serial console. I found out that some modules were > > > > > > Care to send it? > > > > The kernel or the config? I''m using the config given on the pvops wiki > > The kernel output. Can you boot the pvops kernel without Xen and provide > the ''dmesg'' and ''cat /proc/interrupts'' output please?I attached two files with the output. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Konrad Rzeszutek Wilk
2010-Jan-26 18:19 UTC
[Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
> > The kernel output. Can you boot the pvops kernel without Xen and provide > > the ''dmesg'' and ''cat /proc/interrupts'' output please? > > I attached two files with the output.> Initializing cgroup subsys cpuset > Initializing cgroup subsys cpu > Linux version 2.6.31.6-pvops-dom0 (dpquigl@moss-wolverine.epoch.ncsc.mil) (gcc version 4.4.1 20090725 (Red Hat 4.4.1-2) (GCC) ) #7 SMP Tue Jan 26 12:09:38 EST 2010 > Command line: ro root=/dev/mapper/VolGroup-lv_root console=tty0 console=ttyS0,115200n8 > KERNEL supported cpus: > Intel GenuineIntel > AMD AuthenticAMD > Centaur CentaurHauls > BIOS-provided physical RAM map:... snip ..> pci 0000:00:1a.0: reg 20 io port: [0xff20-0xff3f] > pci 0000:00:1a.1: reg 20 io port: [0xff00-0xff1f] > pci 0000:00:1a.2: reg 20 io port: [0xfc00-0xfc1f] > pci 0000:00:1a.7: reg 10 32bit mmio: [0xfebda000-0xfebda3ff] > pci 0000:00:1a.7: PME# supported from D0 D3hot D3cold > pci 0000:00:1a.7: PME# disabled... snip ..> IOMMU: Setting identity map for device 0000:00:1a.0 [0xbfe58000 - 0xbfe70000] > IOMMU: Setting identity map for device 0000:00:1a.1 [0xbfe58000 - 0xbfe70000] > IOMMU: Setting identity map for device 0000:00:1a.2 [0xbfe58000 - 0xbfe70000] > IOMMU: Setting identity map for device 0000:00:1a.7 [0xbfe58000 - 0xbfe70000]... snip ...> ehci_hcd 0000:00:1a.7: PCI INT C -> GSI 22 (level, low) -> IRQ 22 > ehci_hcd 0000:00:1a.7: setting latency timer to 64 > ehci_hcd 0000:00:1a.7: EHCI Host Controller > ehci_hcd 0000:00:1a.7: new USB bus registered, assigned bus number 1 > ehci_hcd 0000:00:1a.7: debug port 1 > ehci_hcd 0000:00:1a.7: cache line size of 32 is not supported > ehci_hcd 0000:00:1a.7: irq 22, io mem 0xfebda000 > ehci_hcd 0000:00:1a.7: USB 2.0 started, EHCI 1.00 > usb usb1: New USB device found, idVendor=1d6b, idProduct=0002 > usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1 > usb usb1: Product: EHCI Host Controller > usb usb1: Manufacturer: Linux 2.6.31.6-pvops-dom0 ehci_hcd > usb usb1: SerialNumber: 0000:00:1a.7 > usb usb1: configuration #1 chosen from 1 choice > hub 1-0:1.0: USB hub found > hub 1-0:1.0: 6 ports detected > work_for_cpu used greatest stack depth: 3888 bytes left > alloc irq_desc for 23 on node 0 > alloc kstat_irqs on node 0 > ehci_hcd 0000:00:1d.7: PCI INT A -> GSI 23 (level, low) -> IRQ 23 > ehci_hcd 0000:00:1d.7: setting latency timer to 64 > ehci_hcd 0000:00:1d.7: EHCI Host Controller > ehci_hcd 0000:00:1d.7: new USB bus registered, assigned bus number 2 > ehci_hcd 0000:00:1d.7: debug port 1 > ehci_hcd 0000:00:1d.7: cache line size of 32 is not supported > ehci_hcd 0000:00:1d.7: irq 23, io mem 0xff980000 > ehci_hcd 0000:00:1d.7: USB 2.0 started, EHCI 1.00 > usb usb2: New USB device found, idVendor=1d6b, idProduct=0002 > usb usb2: New USB device strings: Mfr=3, Product=2, SerialNumber=1 > usb usb2: Product: EHCI Host Controller > usb usb2: Manufacturer: Linux 2.6.31.6-pvops-dom0 ehci_hcd > usb usb2: SerialNumber: 0000:00:1d.7 > usb usb2: configuration #1 chosen from 1 choice > hub 2-0:1.0: USB hub found > hub 2-0:1.0: 6 ports detected > ohci_hcd: USB 1.1 ''Open'' Host Controller (OHCI) DriverSo what happend to 0000:00:1a.0 and 0000:00:1a.1? In the Xen boot those devices were the ones that failed: uhci_hcd 0000:00:1a.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16 uhci_hcd 0000:00:1a.0: UHCI Host Controller uhci_hcd 0000:00:1a.0: new USB bus registered, assigned bus number 3 uhci_hcd 0000:00:1a.0: irq 16, io base 0x0000ff20 uhci_hcd 0000:00:1a.0: host system error, PCI problems? uhci_hcd 0000:00:1a.0: host controller halted, very bad! uhci_hcd 0000:00:1a.0: HC died; cleaning up uhci_hcd 0000:00:1a.0: USB bus 3 deregistered uhci_hcd 0000:00:1a.0: PCI INT A disabled uhci_hcd 0000:00:1a.0: init 0000:00:1a.0 fail, -108 uhci_hcd: probe of 0000:00:1a.0 failed with error -108 But they don''t show up in your bootup here? So did somethinh change? Did they get initialized later? Can you provide the ''lspci -vvvv'' from your bare-metal? Wonder what the 1a.0 and 1a.1 show as. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
David P. Quigley
2010-Jan-26 18:44 UTC
[Xen-devel] Re: Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Tue, 2010-01-26 at 13:19 -0500, Konrad Rzeszutek Wilk wrote:> > > The kernel output. Can you boot the pvops kernel without Xen and provide > > > the ''dmesg'' and ''cat /proc/interrupts'' output please? > > > > I attached two files with the output. > > > Initializing cgroup subsys cpuset > > Initializing cgroup subsys cpu > > Linux version 2.6.31.6-pvops-dom0 (dpquigl@moss-wolverine.epoch.ncsc.mil) (gcc version 4.4.1 20090725 (Red Hat 4.4.1-2) (GCC) ) #7 SMP Tue Jan 26 12:09:38 EST 2010 > > Command line: ro root=/dev/mapper/VolGroup-lv_root console=tty0 console=ttyS0,115200n8 > > KERNEL supported cpus: > > Intel GenuineIntel > > AMD AuthenticAMD > > Centaur CentaurHauls > > BIOS-provided physical RAM map: > > ... snip .. > > pci 0000:00:1a.0: reg 20 io port: [0xff20-0xff3f] > > pci 0000:00:1a.1: reg 20 io port: [0xff00-0xff1f] > > pci 0000:00:1a.2: reg 20 io port: [0xfc00-0xfc1f] > > pci 0000:00:1a.7: reg 10 32bit mmio: [0xfebda000-0xfebda3ff] > > pci 0000:00:1a.7: PME# supported from D0 D3hot D3cold > > pci 0000:00:1a.7: PME# disabled > > ... snip .. > > IOMMU: Setting identity map for device 0000:00:1a.0 [0xbfe58000 - 0xbfe70000] > > IOMMU: Setting identity map for device 0000:00:1a.1 [0xbfe58000 - 0xbfe70000] > > IOMMU: Setting identity map for device 0000:00:1a.2 [0xbfe58000 - 0xbfe70000] > > IOMMU: Setting identity map for device 0000:00:1a.7 [0xbfe58000 - 0xbfe70000] > > ... snip ... > > ehci_hcd 0000:00:1a.7: PCI INT C -> GSI 22 (level, low) -> IRQ 22 > > ehci_hcd 0000:00:1a.7: setting latency timer to 64 > > ehci_hcd 0000:00:1a.7: EHCI Host Controller > > ehci_hcd 0000:00:1a.7: new USB bus registered, assigned bus number 1 > > ehci_hcd 0000:00:1a.7: debug port 1 > > ehci_hcd 0000:00:1a.7: cache line size of 32 is not supported > > ehci_hcd 0000:00:1a.7: irq 22, io mem 0xfebda000 > > ehci_hcd 0000:00:1a.7: USB 2.0 started, EHCI 1.00 > > usb usb1: New USB device found, idVendor=1d6b, idProduct=0002 > > usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1 > > usb usb1: Product: EHCI Host Controller > > usb usb1: Manufacturer: Linux 2.6.31.6-pvops-dom0 ehci_hcd > > usb usb1: SerialNumber: 0000:00:1a.7 > > usb usb1: configuration #1 chosen from 1 choice > > hub 1-0:1.0: USB hub found > > hub 1-0:1.0: 6 ports detected > > work_for_cpu used greatest stack depth: 3888 bytes left > > alloc irq_desc for 23 on node 0 > > alloc kstat_irqs on node 0 > > ehci_hcd 0000:00:1d.7: PCI INT A -> GSI 23 (level, low) -> IRQ 23 > > ehci_hcd 0000:00:1d.7: setting latency timer to 64 > > ehci_hcd 0000:00:1d.7: EHCI Host Controller > > ehci_hcd 0000:00:1d.7: new USB bus registered, assigned bus number 2 > > ehci_hcd 0000:00:1d.7: debug port 1 > > ehci_hcd 0000:00:1d.7: cache line size of 32 is not supported > > ehci_hcd 0000:00:1d.7: irq 23, io mem 0xff980000 > > ehci_hcd 0000:00:1d.7: USB 2.0 started, EHCI 1.00 > > usb usb2: New USB device found, idVendor=1d6b, idProduct=0002 > > usb usb2: New USB device strings: Mfr=3, Product=2, SerialNumber=1 > > usb usb2: Product: EHCI Host Controller > > usb usb2: Manufacturer: Linux 2.6.31.6-pvops-dom0 ehci_hcd > > usb usb2: SerialNumber: 0000:00:1d.7 > > usb usb2: configuration #1 chosen from 1 choice > > hub 2-0:1.0: USB hub found > > hub 2-0:1.0: 6 ports detected > > ohci_hcd: USB 1.1 ''Open'' Host Controller (OHCI) Driver > > So what happend to 0000:00:1a.0 and 0000:00:1a.1? In the Xen boot > those devices were the ones that failed: > > uhci_hcd 0000:00:1a.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16 > uhci_hcd 0000:00:1a.0: UHCI Host Controller > uhci_hcd 0000:00:1a.0: new USB bus registered, assigned bus number 3 > uhci_hcd 0000:00:1a.0: irq 16, io base 0x0000ff20 > uhci_hcd 0000:00:1a.0: host system error, PCI problems? > uhci_hcd 0000:00:1a.0: host controller halted, very bad! > uhci_hcd 0000:00:1a.0: HC died; cleaning up > uhci_hcd 0000:00:1a.0: USB bus 3 deregistered > uhci_hcd 0000:00:1a.0: PCI INT A disabled > uhci_hcd 0000:00:1a.0: init 0000:00:1a.0 fail, -108 > uhci_hcd: probe of 0000:00:1a.0 failed with error -108 > > But they don''t show up in your bootup here? So did somethinh change? Did > they get initialized later? > > Can you provide the ''lspci -vvvv'' from your bare-metal? Wonder what the > 1a.0 and 1a.1 show as.Ahh sorry about that. I removed the uhci module from the kernel build to see if I could get the pvops kernel to finish booting. I''ve rebuilt it with uhci back in and here is the updated dmesg and interrupts. Something that was odd when I took the uhci driver out was that it hung just after the failure to initialize the radeon driver which both dmesg outputs have in them so I don''t think thats the problem. Just after that it hangs. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
David P. Quigley
2010-Jan-26 19:24 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Tue, 2010-01-26 at 13:19 -0500, Konrad Rzeszutek Wilk wrote:> > > The kernel output. Can you boot the pvops kernel without Xen and provide > > > the ''dmesg'' and ''cat /proc/interrupts'' output please? > > > > I attached two files with the output. > > > Initializing cgroup subsys cpuset > > Initializing cgroup subsys cpu > > Linux version 2.6.31.6-pvops-dom0 (dpquigl@moss-wolverine.epoch.ncsc.mil) (gcc version 4.4.1 20090725 (Red Hat 4.4.1-2) (GCC) ) #7 SMP Tue Jan 26 12:09:38 EST 2010 > > Command line: ro root=/dev/mapper/VolGroup-lv_root console=tty0 console=ttyS0,115200n8 > > KERNEL supported cpus: > > Intel GenuineIntel > > AMD AuthenticAMD > > Centaur CentaurHauls > > BIOS-provided physical RAM map: > > ... snip .. > > pci 0000:00:1a.0: reg 20 io port: [0xff20-0xff3f] > > pci 0000:00:1a.1: reg 20 io port: [0xff00-0xff1f] > > pci 0000:00:1a.2: reg 20 io port: [0xfc00-0xfc1f] > > pci 0000:00:1a.7: reg 10 32bit mmio: [0xfebda000-0xfebda3ff] > > pci 0000:00:1a.7: PME# supported from D0 D3hot D3cold > > pci 0000:00:1a.7: PME# disabled > > ... snip .. > > IOMMU: Setting identity map for device 0000:00:1a.0 [0xbfe58000 - 0xbfe70000] > > IOMMU: Setting identity map for device 0000:00:1a.1 [0xbfe58000 - 0xbfe70000] > > IOMMU: Setting identity map for device 0000:00:1a.2 [0xbfe58000 - 0xbfe70000] > > IOMMU: Setting identity map for device 0000:00:1a.7 [0xbfe58000 - 0xbfe70000] > > ... snip ... > > ehci_hcd 0000:00:1a.7: PCI INT C -> GSI 22 (level, low) -> IRQ 22 > > ehci_hcd 0000:00:1a.7: setting latency timer to 64 > > ehci_hcd 0000:00:1a.7: EHCI Host Controller > > ehci_hcd 0000:00:1a.7: new USB bus registered, assigned bus number 1 > > ehci_hcd 0000:00:1a.7: debug port 1 > > ehci_hcd 0000:00:1a.7: cache line size of 32 is not supported > > ehci_hcd 0000:00:1a.7: irq 22, io mem 0xfebda000 > > ehci_hcd 0000:00:1a.7: USB 2.0 started, EHCI 1.00 > > usb usb1: New USB device found, idVendor=1d6b, idProduct=0002 > > usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1 > > usb usb1: Product: EHCI Host Controller > > usb usb1: Manufacturer: Linux 2.6.31.6-pvops-dom0 ehci_hcd > > usb usb1: SerialNumber: 0000:00:1a.7 > > usb usb1: configuration #1 chosen from 1 choice > > hub 1-0:1.0: USB hub found > > hub 1-0:1.0: 6 ports detected > > work_for_cpu used greatest stack depth: 3888 bytes left > > alloc irq_desc for 23 on node 0 > > alloc kstat_irqs on node 0 > > ehci_hcd 0000:00:1d.7: PCI INT A -> GSI 23 (level, low) -> IRQ 23 > > ehci_hcd 0000:00:1d.7: setting latency timer to 64 > > ehci_hcd 0000:00:1d.7: EHCI Host Controller > > ehci_hcd 0000:00:1d.7: new USB bus registered, assigned bus number 2 > > ehci_hcd 0000:00:1d.7: debug port 1 > > ehci_hcd 0000:00:1d.7: cache line size of 32 is not supported > > ehci_hcd 0000:00:1d.7: irq 23, io mem 0xff980000 > > ehci_hcd 0000:00:1d.7: USB 2.0 started, EHCI 1.00 > > usb usb2: New USB device found, idVendor=1d6b, idProduct=0002 > > usb usb2: New USB device strings: Mfr=3, Product=2, SerialNumber=1 > > usb usb2: Product: EHCI Host Controller > > usb usb2: Manufacturer: Linux 2.6.31.6-pvops-dom0 ehci_hcd > > usb usb2: SerialNumber: 0000:00:1d.7 > > usb usb2: configuration #1 chosen from 1 choice > > hub 2-0:1.0: USB hub found > > hub 2-0:1.0: 6 ports detected > > ohci_hcd: USB 1.1 ''Open'' Host Controller (OHCI) Driver > > So what happend to 0000:00:1a.0 and 0000:00:1a.1? In the Xen boot > those devices were the ones that failed: > > uhci_hcd 0000:00:1a.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16 > uhci_hcd 0000:00:1a.0: UHCI Host Controller > uhci_hcd 0000:00:1a.0: new USB bus registered, assigned bus number 3 > uhci_hcd 0000:00:1a.0: irq 16, io base 0x0000ff20 > uhci_hcd 0000:00:1a.0: host system error, PCI problems? > uhci_hcd 0000:00:1a.0: host controller halted, very bad! > uhci_hcd 0000:00:1a.0: HC died; cleaning up > uhci_hcd 0000:00:1a.0: USB bus 3 deregistered > uhci_hcd 0000:00:1a.0: PCI INT A disabled > uhci_hcd 0000:00:1a.0: init 0000:00:1a.0 fail, -108 > uhci_hcd: probe of 0000:00:1a.0 failed with error -108 > > But they don''t show up in your bootup here? So did somethinh change? Did > they get initialized later? > > Can you provide the ''lspci -vvvv'' from your bare-metal? Wonder what the > 1a.0 and 1a.1 show as. > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-develI forgot to attach the lspci output that you asked for as well. This is with the UHCI module built back into the kernel as well. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
David P. Quigley
2010-Jan-26 19:30 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
I have a small update on the problem. It does seem like it is tied to VT-D because I just turned it off in the bios and the dom0 manages to boot on top of xen and I can log in. Are there any options for VT-D that you would like me to try to help pin point the problem? Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Pasi Kärkkäinen
2010-Jan-26 19:46 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Tue, Jan 26, 2010 at 02:30:09PM -0500, David P. Quigley wrote:> I have a small update on the problem. It does seem like it is tied to > VT-D because I just turned it off in the bios and the dom0 manages to > boot on top of xen and I can log in. Are there any options for VT-D that > you would like me to try to help pin point the problem? >At least update to the latest xen-unstable tree, since -rc2 got just released, and it has some VT-d fixes/changes. -- Pasi _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
David P. Quigley
2010-Jan-26 20:21 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Tue, 2010-01-26 at 21:46 +0200, Pasi Kärkkäinen wrote:> On Tue, Jan 26, 2010 at 02:30:09PM -0500, David P. Quigley wrote: > > I have a small update on the problem. It does seem like it is tied to > > VT-D because I just turned it off in the bios and the dom0 manages to > > boot on top of xen and I can log in. Are there any options for VT-D that > > you would like me to try to help pin point the problem? > > > > At least update to the latest xen-unstable tree, since -rc2 got just released, > and it has some VT-d fixes/changes. > > -- PasiI updated my xen-unstable tree and tried to turn VT-D back on and it breaks booting again so we still have a bug on our hands. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Han, Weidong
2010-Jan-27 02:49 UTC
RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
Can you pls post the serial output? It should have warning messages if it is caused by VT-d. Regards, Weidong -----Original Message----- From: xen-devel-bounces@lists.xensource.com [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of David P. Quigley Sent: Wednesday, January 27, 2010 4:21 AM To: Pasi Kärkkäinen Cc: Daniel Stodden; xen-devel@lists.xensource.com; Konrad Rzeszutek Wilk Subject: Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG() On Tue, 2010-01-26 at 21:46 +0200, Pasi Kärkkäinen wrote:> On Tue, Jan 26, 2010 at 02:30:09PM -0500, David P. Quigley wrote: > > I have a small update on the problem. It does seem like it is tied to > > VT-D because I just turned it off in the bios and the dom0 manages to > > boot on top of xen and I can log in. Are there any options for VT-D that > > you would like me to try to help pin point the problem? > > > > At least update to the latest xen-unstable tree, since -rc2 got just released, > and it has some VT-d fixes/changes. > > -- PasiI updated my xen-unstable tree and tried to turn VT-D back on and it breaks booting again so we still have a bug on our hands. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Sander Eikelenboom
2010-Jan-27 09:08 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
Hello David, From LKML I remember some suggestions about disabeling "Legacy USB" when using VT-d to me able to boot linux. Perhaps this also works for you with xen+linux ? http://www.mail-archive.com/kvm@vger.kernel.org/msg05313.html -- Sander Tuesday, January 26, 2010, 9:21:22 PM, you wrote:> On Tue, 2010-01-26 at 21:46 +0200, Pasi Kärkkäinen wrote: >> On Tue, Jan 26, 2010 at 02:30:09PM -0500, David P. Quigley wrote: >> > I have a small update on the problem. It does seem like it is tied to >> > VT-D because I just turned it off in the bios and the dom0 manages to >> > boot on top of xen and I can log in. Are there any options for VT-D that >> > you would like me to try to help pin point the problem? >> > >> >> At least update to the latest xen-unstable tree, since -rc2 got just released, >> and it has some VT-d fixes/changes. >> >> -- Pasi> I updated my xen-unstable tree and tried to turn VT-D back on and it > breaks booting again so we still have a bug on our hands.> Dave-- Best regards, Sander mailto:linux@eikelenboom.it _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
David P. Quigley
2010-Jan-27 18:48 UTC
RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Wed, 2010-01-27 at 10:49 +0800, Han, Weidong wrote:> Can you pls post the serial output? It should have warning messages if it is caused by VT-d. > > Regards, > Weidong > > -----Original Message----- > From: xen-devel-bounces@lists.xensource.com [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of David P. Quigley > Sent: Wednesday, January 27, 2010 4:21 AM > To: Pasi Kärkkäinen > Cc: Daniel Stodden; xen-devel@lists.xensource.com; Konrad Rzeszutek Wilk > Subject: Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG() > > On Tue, 2010-01-26 at 21:46 +0200, Pasi Kärkkäinen wrote: > > On Tue, Jan 26, 2010 at 02:30:09PM -0500, David P. Quigley wrote: > > > I have a small update on the problem. It does seem like it is tied to > > > VT-D because I just turned it off in the bios and the dom0 manages to > > > boot on top of xen and I can log in. Are there any options for VT-D that > > > you would like me to try to help pin point the problem? > > > > > > > At least update to the latest xen-unstable tree, since -rc2 got just released, > > and it has some VT-d fixes/changes. > > > > -- Pasi > > I updated my xen-unstable tree and tried to turn VT-D back on and it > breaks booting again so we still have a bug on our hands. > > Dave > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-develHello, Attached is the full serial console output with vt-d enabled from the xen 4.0-pre ascii art to the kernel oops in dom0. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
David P. Quigley
2010-Jan-27 20:22 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Tue, 2010-01-26 at 14:30 -0500, David P. Quigley wrote:> I have a small update on the problem. It does seem like it is tied to > VT-D because I just turned it off in the bios and the dom0 manages to > boot on top of xen and I can log in. Are there any options for VT-D that > you would like me to try to help pin point the problem? > > Dave > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-develSo I''ve managed to get my dom0 booting on top of xen however now I''m running into a problem where my domU is kernel oopsing on startup. The traces I''ve gotten seem to be in the memory allocation functions. Normally this is a sign of memory corruption. Considering this VM was working perfectly with the 2.6.31.10 dom0 based on the Novell patches and xen 4.0 I don''t think it is the VM that is the problem (however I can try starting it under kvm to be certain. Any suggestions on what I can do to try to hunt down the memory corruption? Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Han, Weidong
2010-Jan-28 01:50 UTC
RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
Hi Dave,
I found the DMAR (VT-d) was also enabled in pvops dom0 in the log. It should not
happen because Xen zaps ACPI DMAR signature to prevent dom0 using VT-d. Could
you apply below small patch to confirm that xen really zaps ACPI DMAR?
diff -r 2f98b700605f xen/drivers/passthrough/vtd/dmar.c
--- a/xen/drivers/passthrough/vtd/dmar.c Wed Jan 27 08:59:47 2010 +0000
+++ b/xen/drivers/passthrough/vtd/dmar.c Thu Jan 28 09:44:31 2010 +0800
@@ -673,6 +673,8 @@ out:
out:
/* Zap ACPI DMAR signature to prevent dom0 using vt-d HW. */
dmar->header.signature[0] = '\0';
+ printk(XENLOG_WARNING
+ "Zap ACPI DMAR signature to prevent dom0 using VT-d
HW.\n");
return ret;
}
-----Original Message-----
From: David P. Quigley [mailto:dpquigl@tycho.nsa.gov]
Sent: Thursday, January 28, 2010 2:48 AM
To: Han, Weidong
Cc: Pasi Kärkkäinen; Daniel Stodden; xen-devel@lists.xensource.com; Konrad
Rzeszutek Wilk
Subject: RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Wed, 2010-01-27 at 10:49 +0800, Han, Weidong wrote:> Can you pls post the serial output? It should have warning messages if it
is caused by VT-d.
>
> Regards,
> Weidong
>
> -----Original Message-----
> From: xen-devel-bounces@lists.xensource.com
> [mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of David P.
> Quigley
> Sent: Wednesday, January 27, 2010 4:21 AM
> To: Pasi Kärkkäinen
> Cc: Daniel Stodden; xen-devel@lists.xensource.com; Konrad Rzeszutek
> Wilk
> Subject: Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI =
> BUG()
>
> On Tue, 2010-01-26 at 21:46 +0200, Pasi Kärkkäinen wrote:
> > On Tue, Jan 26, 2010 at 02:30:09PM -0500, David P. Quigley wrote:
> > > I have a small update on the problem. It does seem like it is
tied
> > > to VT-D because I just turned it off in the bios and the dom0
> > > manages to boot on top of xen and I can log in. Are there any
> > > options for VT-D that you would like me to try to help pin point
the problem?
> > >
> >
> > At least update to the latest xen-unstable tree, since -rc2 got just
> > released, and it has some VT-d fixes/changes.
> >
> > -- Pasi
>
> I updated my xen-unstable tree and tried to turn VT-D back on and it
> breaks booting again so we still have a bug on our hands.
>
> Dave
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel
Hello,
Attached is the full serial console output with vt-d enabled from the xen
4.0-pre ascii art to the kernel oops in dom0.
Dave
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel
Pasi Kärkkäinen
2010-Jan-28 09:24 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Wed, Jan 27, 2010 at 03:22:57PM -0500, David P. Quigley wrote:> On Tue, 2010-01-26 at 14:30 -0500, David P. Quigley wrote: > > I have a small update on the problem. It does seem like it is tied to > > VT-D because I just turned it off in the bios and the dom0 manages to > > boot on top of xen and I can log in. Are there any options for VT-D that > > you would like me to try to help pin point the problem? > > > > Dave > > > > > > _______________________________________________ > > Xen-devel mailing list > > Xen-devel@lists.xensource.com > > http://lists.xensource.com/xen-devel > > > So I''ve managed to get my dom0 booting on top of xen however now I''m > running into a problem where my domU is kernel oopsing on startup. The > traces I''ve gotten seem to be in the memory allocation functions. > Normally this is a sign of memory corruption. Considering this VM was > working perfectly with the 2.6.31.10 dom0 based on the Novell patches > and xen 4.0 I don''t think it is the VM that is the problem (however I > can try starting it under kvm to be certain. Any suggestions on what I > can do to try to hunt down the memory corruption? >Hmm.. interesting. You could try this: Set on_crash="preserve" in /etc/xen/<guest> cfgfile, and when the guest crashes, run in dom0: /usr/lib/xen/bin/xenctx -s System.map-domUkernelversion <domUid> If your system is 64bit then xenctx might be under /usr/lib64/ That should give you a stack trace of the crashed guest.. -- Pasi _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
David P. Quigley
2010-Jan-28 14:49 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Thu, 2010-01-28 at 11:24 +0200, Pasi Kärkkäinen wrote:> On Wed, Jan 27, 2010 at 03:22:57PM -0500, David P. Quigley wrote: > > On Tue, 2010-01-26 at 14:30 -0500, David P. Quigley wrote: > > > I have a small update on the problem. It does seem like it is tied to > > > VT-D because I just turned it off in the bios and the dom0 manages to > > > boot on top of xen and I can log in. Are there any options for VT-D that > > > you would like me to try to help pin point the problem? > > > > > > Dave > > > > > > > > > _______________________________________________ > > > Xen-devel mailing list > > > Xen-devel@lists.xensource.com > > > http://lists.xensource.com/xen-devel > > > > > > So I''ve managed to get my dom0 booting on top of xen however now I''m > > running into a problem where my domU is kernel oopsing on startup. The > > traces I''ve gotten seem to be in the memory allocation functions. > > Normally this is a sign of memory corruption. Considering this VM was > > working perfectly with the 2.6.31.10 dom0 based on the Novell patches > > and xen 4.0 I don''t think it is the VM that is the problem (however I > > can try starting it under kvm to be certain. Any suggestions on what I > > can do to try to hunt down the memory corruption? > > > > Hmm.. interesting. > > You could try this: Set on_crash="preserve" in /etc/xen/<guest> cfgfile, > and when the guest crashes, run in dom0: > > /usr/lib/xen/bin/xenctx -s System.map-domUkernelversion <domUid> > > If your system is 64bit then xenctx might be under /usr/lib64/ > That should give you a stack trace of the crashed guest.. > > -- PasiI''ll run that when I get a chance. I managed to get the domU booting. I removed all of my debug statements off of the xen-4.0.gz line in my grub entry and it worked. I''m not sure if that code is the cause or if it just didn''t stress a race condition that happens when the system is running slower because of debug code. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Pasi Kärkkäinen
2010-Jan-28 15:02 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Thu, Jan 28, 2010 at 09:49:16AM -0500, David P. Quigley wrote:> On Thu, 2010-01-28 at 11:24 +0200, Pasi Kärkkäinen wrote: > > On Wed, Jan 27, 2010 at 03:22:57PM -0500, David P. Quigley wrote: > > > On Tue, 2010-01-26 at 14:30 -0500, David P. Quigley wrote: > > > > I have a small update on the problem. It does seem like it is tied to > > > > VT-D because I just turned it off in the bios and the dom0 manages to > > > > boot on top of xen and I can log in. Are there any options for VT-D that > > > > you would like me to try to help pin point the problem? > > > > > > > > Dave > > > > > > > > > > > > _______________________________________________ > > > > Xen-devel mailing list > > > > Xen-devel@lists.xensource.com > > > > http://lists.xensource.com/xen-devel > > > > > > > > > So I''ve managed to get my dom0 booting on top of xen however now I''m > > > running into a problem where my domU is kernel oopsing on startup. The > > > traces I''ve gotten seem to be in the memory allocation functions. > > > Normally this is a sign of memory corruption. Considering this VM was > > > working perfectly with the 2.6.31.10 dom0 based on the Novell patches > > > and xen 4.0 I don''t think it is the VM that is the problem (however I > > > can try starting it under kvm to be certain. Any suggestions on what I > > > can do to try to hunt down the memory corruption? > > > > > > > Hmm.. interesting. > > > > You could try this: Set on_crash="preserve" in /etc/xen/<guest> cfgfile, > > and when the guest crashes, run in dom0: > > > > /usr/lib/xen/bin/xenctx -s System.map-domUkernelversion <domUid> > > > > If your system is 64bit then xenctx might be under /usr/lib64/ > > That should give you a stack trace of the crashed guest.. > > > > -- Pasi > > I''ll run that when I get a chance. I managed to get the domU booting. I > removed all of my debug statements off of the xen-4.0.gz line in my grub > entry and it worked. I''m not sure if that code is the cause or if it > just didn''t stress a race condition that happens when the system is > running slower because of debug code. >Hmm.. I remember some discussions that "sync_console console_to_ring" might cause problems. Not sure about that though. -- Pasi _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
David P. Quigley
2010-Jan-28 16:10 UTC
RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
I''ve attached the output from the patched xen hypervisor. I see the message being printed out so it appears to be clearing the signature. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Han, Weidong
2010-Jan-30 13:50 UTC
RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
It''s so weird pvops dom0 still detects DMAR and enables it after zap DMAR signature in Xen. Dom0 should not detect DMAR in ACPI after the signature is zapped. I suspect that your USB issue is caused by this. Pls have a try with turn off VT-d in pvops config (# CONFIG_DMAR is not set) to see if it solves your issue. Regards, Weidong -----Original Message----- From: David P. Quigley [mailto:dpquigl@tycho.nsa.gov] Sent: Friday, January 29, 2010 12:10 AM To: Han, Weidong Cc: Pasi Kärkkäinen; Daniel Stodden; xen-devel@lists.xensource.com; Konrad Rzeszutek Wilk Subject: RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG() I''ve attached the output from the patched xen hypervisor. I see the message being printed out so it appears to be clearing the signature. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
David P. Quigley
2010-Feb-01 21:35 UTC
RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Sat, 2010-01-30 at 21:50 +0800, Han, Weidong wrote:> It''s so weird pvops dom0 still detects DMAR and enables it after zap DMAR signature in Xen. Dom0 should not detect DMAR in ACPI after the signature is zapped. > > I suspect that your USB issue is caused by this. Pls have a try with turn off VT-d in pvops config (# CONFIG_DMAR is not set) to see if it solves your issue. > > Regards, > Weidong > > -----Original Message----- > From: David P. Quigley [mailto:dpquigl@tycho.nsa.gov] > Sent: Friday, January 29, 2010 12:10 AM > To: Han, Weidong > Cc: Pasi Kärkkäinen; Daniel Stodden; xen-devel@lists.xensource.com; Konrad Rzeszutek Wilk > Subject: RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG() > > I''ve attached the output from the patched xen hypervisor. I see the message being printed out so it appears to be clearing the signature. > > DaveSo I managed to sucessfully book my pvops dom0 kernel with VT-D on in my bios when I turned off DMAR from the dom0 kernel config. Is there anything else you would like me to do to help hunt this bug down? Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Weidong Han
2010-Feb-03 03:08 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
David P. Quigley wrote:> On Sat, 2010-01-30 at 21:50 +0800, Han, Weidong wrote: > >> It''s so weird pvops dom0 still detects DMAR and enables it after zap DMAR signature in Xen. Dom0 should not detect DMAR in ACPI after the signature is zapped. >> >> I suspect that your USB issue is caused by this. Pls have a try with turn off VT-d in pvops config (# CONFIG_DMAR is not set) to see if it solves your issue. >> >> Regards, >> Weidong >> >> -----Original Message----- >> From: David P. Quigley [mailto:dpquigl@tycho.nsa.gov] >> Sent: Friday, January 29, 2010 12:10 AM >> To: Han, Weidong >> Cc: Pasi Kärkkäinen; Daniel Stodden; xen-devel@lists.xensource.com; Konrad Rzeszutek Wilk >> Subject: RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG() >> >> I''ve attached the output from the patched xen hypervisor. I see the message being printed out so it appears to be clearing the signature. >> >> Dave >> > > So I managed to sucessfully book my pvops dom0 kernel with VT-D on in my > bios when I turned off DMAR from the dom0 kernel config. Is there > anything else you would like me to do to help hunt this bug down? > > Dave > >Could you dump ACPI in native Linux and Xen dom0? Let''s see what difference between "zap" and "no zap" DMAR signature. Regards, Weidong _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
David P. Quigley
2010-Feb-03 16:38 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Wed, 2010-02-03 at 11:08 +0800, Weidong Han wrote:> David P. Quigley wrote: > > On Sat, 2010-01-30 at 21:50 +0800, Han, Weidong wrote: > > > >> It''s so weird pvops dom0 still detects DMAR and enables it after zap DMAR signature in Xen. Dom0 should not detect DMAR in ACPI after the signature is zapped. > >> > >> I suspect that your USB issue is caused by this. Pls have a try with turn off VT-d in pvops config (# CONFIG_DMAR is not set) to see if it solves your issue. > >> > >> Regards, > >> Weidong > >> > >> -----Original Message----- > >> From: David P. Quigley [mailto:dpquigl@tycho.nsa.gov] > >> Sent: Friday, January 29, 2010 12:10 AM > >> To: Han, Weidong > >> Cc: Pasi Kärkkäinen; Daniel Stodden; xen-devel@lists.xensource.com; Konrad Rzeszutek Wilk > >> Subject: RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG() > >> > >> I''ve attached the output from the patched xen hypervisor. I see the message being printed out so it appears to be clearing the signature. > >> > >> Dave > >> > > > > So I managed to sucessfully book my pvops dom0 kernel with VT-D on in my > > bios when I turned off DMAR from the dom0 kernel config. Is there > > anything else you would like me to do to help hunt this bug down? > > > > Dave > > > > > Could you dump ACPI in native Linux and Xen dom0? Let''s see what > difference between "zap" and "no zap" DMAR signature. > > Regards, > WeidongSure. How do I go about doing that? Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Weidong Han
2010-Feb-04 01:59 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
David P. Quigley wrote:> On Wed, 2010-02-03 at 11:08 +0800, Weidong Han wrote: > >> David P. Quigley wrote: >> >>> On Sat, 2010-01-30 at 21:50 +0800, Han, Weidong wrote: >>> >>> >>>> It''s so weird pvops dom0 still detects DMAR and enables it after zap DMAR signature in Xen. Dom0 should not detect DMAR in ACPI after the signature is zapped. >>>> >>>> I suspect that your USB issue is caused by this. Pls have a try with turn off VT-d in pvops config (# CONFIG_DMAR is not set) to see if it solves your issue. >>>> >>>> Regards, >>>> Weidong >>>> >>>> -----Original Message----- >>>> From: David P. Quigley [mailto:dpquigl@tycho.nsa.gov] >>>> Sent: Friday, January 29, 2010 12:10 AM >>>> To: Han, Weidong >>>> Cc: Pasi Kärkkäinen; Daniel Stodden; xen-devel@lists.xensource.com; Konrad Rzeszutek Wilk >>>> Subject: RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG() >>>> >>>> I''ve attached the output from the patched xen hypervisor. I see the message being printed out so it appears to be clearing the signature. >>>> >>>> Dave >>>> >>>> >>> So I managed to sucessfully book my pvops dom0 kernel with VT-D on in my >>> bios when I turned off DMAR from the dom0 kernel config. Is there >>> anything else you would like me to do to help hunt this bug down? >>> >>> Dave >>> >>> >>> >> Could you dump ACPI in native Linux and Xen dom0? Let''s see what >> difference between "zap" and "no zap" DMAR signature. >> >> Regards, >> Weidong >> > > Sure. How do I go about doing that? > > Dave > >You can search and download acpidump. After install it, you can use following command to dump acpi table: acpidump >acpidump.txt Dump acpi table on both native Linux and xen dom0, and post dump files. Regards, Weidong _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
David P. Quigley
2010-Feb-04 16:05 UTC
Re: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG()
On Thu, 2010-02-04 at 09:59 +0800, Weidong Han wrote:> David P. Quigley wrote: > > On Wed, 2010-02-03 at 11:08 +0800, Weidong Han wrote: > > > >> David P. Quigley wrote: > >> > >>> On Sat, 2010-01-30 at 21:50 +0800, Han, Weidong wrote: > >>> > >>> > >>>> It''s so weird pvops dom0 still detects DMAR and enables it after zap DMAR signature in Xen. Dom0 should not detect DMAR in ACPI after the signature is zapped. > >>>> > >>>> I suspect that your USB issue is caused by this. Pls have a try with turn off VT-d in pvops config (# CONFIG_DMAR is not set) to see if it solves your issue. > >>>> > >>>> Regards, > >>>> Weidong > >>>> > >>>> -----Original Message----- > >>>> From: David P. Quigley [mailto:dpquigl@tycho.nsa.gov] > >>>> Sent: Friday, January 29, 2010 12:10 AM > >>>> To: Han, Weidong > >>>> Cc: Pasi Kärkkäinen; Daniel Stodden; xen-devel@lists.xensource.com; Konrad Rzeszutek Wilk > >>>> Subject: RE: [Xen-devel] Xen 4.0 + PVOPS + Intel VTD + USB EHCI = BUG() > >>>> > >>>> I''ve attached the output from the patched xen hypervisor. I see the message being printed out so it appears to be clearing the signature. > >>>> > >>>> Dave > >>>> > >>>> > >>> So I managed to sucessfully book my pvops dom0 kernel with VT-D on in my > >>> bios when I turned off DMAR from the dom0 kernel config. Is there > >>> anything else you would like me to do to help hunt this bug down? > >>> > >>> Dave > >>> > >>> > >>> > >> Could you dump ACPI in native Linux and Xen dom0? Let''s see what > >> difference between "zap" and "no zap" DMAR signature. > >> > >> Regards, > >> Weidong > >> > > > > Sure. How do I go about doing that? > > > > Dave > > > > > You can search and download acpidump. After install it, you can use > following command to dump acpi table: > > acpidump >acpidump.txt > > Dump acpi table on both native Linux and xen dom0, and post dump files. > > Regards, > WeidongI seem to have run into a problem. My dom0 kernel no longer boots as a stand alone kernel. It will only boot on top of Xen. I''ll try to figure this out and I''ll post the results when I do. Dave _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel