similar to: [PATCH] FLush pending softirqs when cpu offline

Hi, Keir, In debugging the issue "system hang when doing cpu offline", I identified a situation that could cause a dead lock. The scenario is: mwait_idle_with_hint inside play_dead will access per cpu variable, which causes #PF. The #PF handler will use printk, which will schedule a tasklet. In scheduling a tasklet, per cpu variables are needed, otherwise, there will be another #PF.

This patch implements cpu offline feature. Best Regards Haitao Shan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

Currently, Xen does not support injecting an NMI to HVM guest OS. Adding this feature is necessary for those softwares which depend on NMI to function correctly, such as KDB and oprofile. The attached patch allows NMI to be injected to guest OS in NMIP capable platforms. It also enables to queue an NMI and then inject it as soon as possible. Signed-off-by: Haitao Shan

flight 10626 xen-unstable real [real] http://www.chiark.greenend.org.uk/~xensrcts/logs/10626/ Regressions :-( Tests which did not succeed and are blocking, including tests which could not be run: test-amd64-i386-xl-multivcpu 16 guest-start.2 fail in 10623 REGR. vs. 10621 Tests which are failing intermittently (not blocking): test-amd64-i386-rhel6hvm-intel 4 xen-install fail

This patch add MSI support to passthrough HVM domain. Currently it only inercept access to MSI config space, no MSI-x support. Signed-off-by: Jiang Yunhong <yunhong.jiang@intel.com> Signed-off-by: Shan Haitao <haitao.shan@intel. <<msi_passthrough.patch>> com> Best Regards Haitao Shan _______________________________________________ Xen-devel mailing list

flight 8091 xen-unstable real [real] http://www.chiark.greenend.org.uk/~xensrcts/logs/8091/ Regressions :-( Tests which did not succeed and are blocking: test-amd64-i386-rhel6hvm-intel 12 guest-localmigrate/x10 fail REGR. vs. 8071 Tests which did not succeed, but are not blocking, including regressions (tests previously passed) regarded as allowable: test-amd64-amd64-xl-pcipt-intel 9

MSI-x may have multiple vectors, however in current interrupt remapping code, one device only has one entry in interrupt remapping table. This patch adds ''remap_index'' in msi_desc structure to track its index in interrupt remapping table. Signed-off-by: Haitao Shan <haitao.shan@intel.com> Signed-off-by: Weidong Han <weidong.han@intel.com>

Hi, Keir, This patch will add the missing MSI-X support to HVM domain. All code is added to qemu. Can I know your comments on this? Best Regards Haitao Shan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

This patch changes the pirq to be per-domain in xen tree. Signed-off-by: Jiang Yunhong <yunhong.jiang@intel.com> Signed-off-by: Shan Haitao <haitao.shan@intel.com> Best Regards Shan Haitao _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

Hi, Keir, Jan, This patch is a bugfix pointed by Jan. Fix mask_base(actually MSI-X table base, copy name from native) to be a virtual address rather than a physical address. And remove wrong printk in pci_disable_msix. Jan, the error message you saw is wrong output from kernel''s MSI code. Really sorry for my dirty code there. Could you please review the patch and give me feedback?

Hi, Keir, The following patches refines Xen support for CPU Xsave/Xrestore support. There are four patches included. Patch 1/4: Cleaning up existing Xsave code in Xen. Replace xfeature_low and xfeature_high with a u64 variable xfeature_mask. In structure hvm_vcpu, rename xfeature_mask to xcr0 Provide EDX:EAX with all bits set to 1 for XSAVE and XRSTOR as spec recommends

__napi_schedule might raise softirq but nothing causes do_softirq to trigger, so it does not in fact run. As a result, the error message "NOHZ: local_softirq_pending 08" sometimes occurs during boot of a KVM guest when the network service is started and we are oom: ... Bringing up loopback interface: [ OK ] Bringing up interface eth0: Determining IP information for

__napi_schedule might raise softirq but nothing causes do_softirq to trigger, so it does not in fact run. As a result, the error message "NOHZ: local_softirq_pending 08" sometimes occurs during boot of a KVM guest when the network service is started and we are oom: ... Bringing up loopback interface: [ OK ] Bringing up interface eth0: Determining IP information for

flight 10413 xen-unstable real [real] http://www.chiark.greenend.org.uk/~xensrcts/logs/10413/ Regressions :-( Tests which did not succeed and are blocking: test-amd64-i386-rhel6hvm-intel 3 host-install(3) broken test-amd64-i386-win 8 guest-saverestore fail REGR. vs. 10201 test-i386-i386-win 8 guest-saverestore fail REGR. vs. 10201

Hi, Keir, Currently, HVM guests do not have access to performance counters. So it is not possible to use performance analyzer software such as vtune in HVM guest to analyze programme performance. Other usage of performance counters , for example, the NMI watchdog, won''t function either. This patch will enable performance counters in HVM guest. Currently, only Core 2 Duo is implemented.

tools/firmware: remove "_PS0/3" Method Do not expose the ACPI power management "_PS0/3" Method to guest firmware. According to section 3.4 of the APCI specification 4.0, PCI device control the device power through its own specification but not through APCI. Qemu pushes "_PS0/3" to guest will cause a mess between ACPI PM and PCI PM as a result of incorrect ACPI table

Hi, Keir, This patch will move reading and writing of PCI configuration spaces from dom0 to Xen. It also changes VTD code, so that they can touch the PCI configuration spaces with proper lock. This will also benefit MSI support in Xen. Can you give some comments? Thanks! <<pci_conf_xen.patch>> Best Regards Haitao Shan _______________________________________________ Xen-devel

Hi, Keir, This patch is a bugfix for saving and restoring MSI/MSI-X across S3. Currently, Dom0''s PCI layer unmaps MSI when S3 and maps them back when resuming. However, this triggers unexpected behaviors. For example, if the drivers still holds that irq at the point of unmapping MSI, Xen will force unbind that pirq. But after resume, we have no mechanism to rebind that pirq. The device

The early "do we need to do anything" check needs adjustment, too. Thanks to Haitao Shan for pointing this out. Signed-off-by: Jan Beulich <jbeulich@novell.com> --- a/xen/arch/x86/cpu/intel.c +++ b/xen/arch/x86/cpu/intel.c @@ -61,7 +61,8 @@ static void __devinit set_cpuidmask(cons const char *extra = ""; if (!~(opt_cpuid_mask_ecx & opt_cpuid_mask_edx & -

In "Understanding the Linux Kernel" 3rd edition, section 4.7 "Softirqs and Tasklets" it states: "Activation and execution [of defferable functions] are bound together: a deferrable function that has been activated by a given CPU must be executed on the same CPU. There is no self-evident reason suggesting that this rule is beneficial for system performance. Binding the