From git://xenbits.xensource.com/qemu-xen-unstable.git 7a11b22eab559e24540059d46c70f4feffe052d7 fix raw_aio_remove (Stefano Stabellini) f9ad69304fd52092c54fbf64b2937ebdb9980649 Call qemu_bh_delete at bdrv_aio_bh_cb a05958b6e32f1748ea70b1efca13394956c0698b block: Clean up after deleting BHs 200a4ece52fa37d6b884b52c47e3d56b9c1c7563 memory/fd leak in pt_msix_init()
Created attachment 471479 [details] Backport of 4 upstream commits mentioned in comment #0 This is the backport of upstream commit mentioned in comment #0. Tested on x86_64 RHEL-5 dom0 using the x86_64 HVM guest with PCI passthrough. Michal
This request was evaluated by Red Hat Product Management for inclusion in the current release of Red Hat Enterprise Linux. Because the affected component is not scheduled to be updated in the current release, Red Hat is unfortunately unable to address this request at this time. Red Hat invites you to ask your support representative to propose this request, if appropriate and relevant, in the next release of Red Hat Enterprise Linux.
This request was erroneously denied for the current release of Red Hat Enterprise Linux. The error has been fixed and this request has been re-proposed for the current release.
Hi Michal, Could you give more info on this bug? I do not know how to verify the fix. Regards, Pengzhen
(In reply to comment #9) > Hi Michal, > > Could you give more info on this bug? > I do not know how to verify the fix. > > Regards, > Pengzhen Hi Pengzhen, this is connected mainly to PCI passthrough using MSI-X to properly free the memory and about flushing the block device using asynchronous I/O. For testing this bug you need to test memory leaks using PCI passthrough of HVM guest. Michal
(In reply to comment #10) > (In reply to comment #9) > > Hi Michal, > > > > Could you give more info on this bug? > > I do not know how to verify the fix. > > > > Regards, > > Pengzhen > > Hi Pengzhen, > this is connected mainly to PCI passthrough using MSI-X to properly free the > memory and about flushing the block device using asynchronous I/O. For testing > this bug you need to test memory leaks using PCI passthrough of HVM guest. > > Michal I tried with "valgrind" to trace xend for memory leak but without luck. It could not trace xen's pricmd #valgrind --tool=memcheck --leak-check=yes --trace-children=yes --log-file=/var/log/valgrind.log /etc/init.d/xend --5647-- WARNING: Serious error when reading debug info --5647-- When reading debug info from /proc/xen/privcmd: --5647-- can't read file to inspect ELF header Now I will try with a patch for valgrind to handle xen's pricmd: http://lists.xensource.com/archives/html/xen-devel/2010-08/msg01246.html Am I on the right track?
Michal, I still can not find way to verify this. Could I just do some sanity test with PCI pass-through to verify this bug? Regards, Pengzhen
Run PCI pass through test with USB controller , Host version: kernel-264, xen-3.0.3-132 , x86_64bit Guest: rhel6.1 x86_64bit Host pass through the onboard usb controller to the guest, and plug a usb disk on that port, guest could use the usb disk without any issue. So I think this bug can be verified.
Created attachment 501962 [details] logs for verify this bug
Technical note added. If any revisions are required, please edit the "Technical Notes" field accordingly. All revisions will be proofread by the Engineering Content Services team. New Contents: Memory leaks were discovered in the code following the deletion of block headers and also in the following functions: raw_aio_remove(), qemu_bh_delete() and pt_msix_init(). With this update, a set of patches has been provided to fix these bugs.
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on therefore solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHBA-2011-1070.html