- 17 Mar, 2011 40 commits
-
-
Xiao Guangrong authored
These macros are not used, so removed Signed-off-by:
Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Signed-off-by:
Avi Kivity <avi@redhat.com>
-
Xiao Guangrong authored
Using __get_free_page instead of alloc_page and page_address, using free_page instead of __free_page and virt_to_page Signed-off-by:
-
Xiao Guangrong authored
No need to record the gfn to verifier the pte has the same mode as current vcpu, it's because we only speculatively update the pte only if the pte and vcpu have the same mode Signed-off-by:
-
Xiao Guangrong authored
kvm_mmu_calculate_mmu_pages need to walk all memslots and it's protected by kvm->slots_lock, so move it out of mmu spinlock Signed-off-by:
-
Xiao Guangrong authored
Set spte accessed bit only if guest_initiated == 1 that means the really accessed Signed-off-by:
-
Xiao Guangrong authored
Only remove write access in the last sptes. Signed-off-by:
-
Jan Kiszka authored
The goal of this document shall be - overview of all locks used in KVM core - provide details on the scope of each lock - explain the lock type, specifically of a raw spin locks - provide a lock ordering guide Signed-off-by:
Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by:
-
Lai Jiangshan authored
use EFER_SCE, EFER_LME and EFER_LMA instead of magic numbers. Signed-off-by:
Lai Jiangshan <laijs@cn.fujitsu.com> Signed-off-by:
-
Lai Jiangshan authored
The hash array of async gfns may still contain some left gfns after kvm_clear_async_pf_completion_queue() called, need to clear them. Signed-off-by:
Gleb Natapov <gleb@redhat.com> Signed-off-by:
-
Gleb Natapov authored
Currently vm86 task is initialized on each real mode entry and vcpu reset. Initialization is done by zeroing TSS and updating relevant fields. But since all vcpus are using the same TSS there is a race where one vcpu may use TSS while other vcpu is initializing it, so the vcpu that uses TSS will see wrong TSS content and will behave incorrectly. Fix that by initializing TSS only once. Signed-off-by:
-
Gleb Natapov authored
When rmode.vm86 is active TR descriptor is updated with vm86 task values, but selector is left intact. vmx_set_segment() makes sure that if TR register is written into while vm86 is active the new values are saved for use after vm86 is deactivated, but since selector is not updated on vm86 activation/deactivation new value is lost. Fix this by writing new selector into vmcs immediately. Signed-off-by:
-
Lai Jiangshan authored
The changelog of 104f226b said "adds the __noclone attribute", but it was missing in its patch. I think it is still needed. Signed-off-by:
Andi Kleen <ak@linux.intel.com> Signed-off-by:
Marcelo Tosatti <mtosatti@redhat.com>
-
Jan Kiszka authored
Code under this lock requires non-preemptibility. Ensure this also over -rt by converting it to raw spinlock. Signed-off-by:
-
Gleb Natapov authored
isr_ack logic was added by e4825800 to avoid unnecessary IPIs. Back then it made sense, but now the code checks that vcpu is ready to accept interrupt before sending IPI, so this logic is no longer needed. The patch removes it. Fixes a regression with Debian/Hurd. Signed-off-by:
Jonathan Nieder <jrnieder@gmail.com> Signed-off-by:
-
Joseph Cihula authored
This patch fixes the logic used to detect whether BIOS has disabled VMX, for the case where VMX is enabled only under SMX, but tboot is not active. Signed-off-by:
Joseph Cihula <joseph.cihula@intel.com> Signed-off-by:
-
Jan Kiszka authored
Code under this lock requires non-preemptibility. Ensure this also over -rt by converting it to raw spinlock. Signed-off-by:
-
Avi Kivity authored
When we enable an NMI window, we ask for an IRET intercept, since the IRET re-enables NMIs. However, the IRET intercept happens before the instruction executes, while the NMI window architecturally opens afterwards. To compensate for this mismatch, we only open the NMI window in the following exit, assuming that the IRET has by then executed; however, this assumption is not always correct; we may exit due to a host interrupt or page fault, without having executed the instruction. Fix by checking for forward progress by recording and comparing the IRET's rip. This is somewhat of a hack, since an unchaging rip does not mean that no forward progress has been made, but is the simplest fix for now. Signed-off-by: -
Avi Kivity authored
The interrupt injection logic looks something like if an nmi is pending, and nmi injection allowed inject nmi if an nmi is pending request exit on nmi window the problem is that "nmi is pending" can be set asynchronously by the PIT; if it happens to fire between the two if statements, we will request an nmi window even though nmi injection is allowed. On SVM, this has disasterous results, since it causes eflags.TF to be set in random guest code. The fix is simple; make nmi_pending synchronous using the standard vcpu->requests mechanism; this ensures the code above is completely synchronous wrt nmi_pending. Signed-off-by: -
Rik van Riel authored
Instead of sleeping in kvm_vcpu_on_spin, which can cause gigantic slowdowns of certain workloads, we instead use yield_to to get another VCPU in the same KVM guest to run sooner. This seems to give a 10-15% speedup in certain workloads. Signed-off-by:
Rik van Riel <riel@redhat.com> Signed-off-by:
-
Rik van Riel authored
Keep track of which task is running a KVM vcpu. This helps us figure out later what task to wake up if we want to boost a vcpu that got preempted. Unfortunately there are no guarantees that the same task always keeps the same vcpu, so we can only track the task across a single "run" of the vcpu. Signed-off-by:
-
Rik van Riel authored
Export the symbols required for a race-free kvm_vcpu_on_spin. Signed-off-by:
-
Avi Kivity authored
Use the new support in the emulator, and drop the ad-hoc code in x86.c. Signed-off-by:
-
Avi Kivity authored
Mark some instructions as vendor specific, and allow the caller to request emulation only of vendor specific instructions. This is useful in some circumstances (responding to a #UD fault). Signed-off-by:
-
Avi Kivity authored
x86_decode_insn() doesn't return X86EMUL_* values, so the check for X86EMUL_PROPOGATE_FAULT will always fail. There is a proper check later on, so there is no need for a replacement for this code. Signed-off-by:
-
Jan Kiszka authored
This warning was once used for debugging QEMU user space. Though uncommon, it is actually possible to send an INIT request to a running VCPU. So better drop this warning before someone misuses it to flood kernel logs this way. Signed-off-by:
-
Glauber Costa authored
When a vcpu is reset, kvmclock page keeps being written to this days. This is wrong and inconsistent: a cpu reset should take it to its initial state. Signed-off-by:
Glauber Costa <glommer@redhat.com> CC: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by:
-
Huang Ying authored
Unused. Signed-off-by:
Huang Ying <ying.huang@intel.com> Signed-off-by:
-
Huang Ying authored
is_hwpoison_address only checks whether the page table entry is hwpoisoned, regardless the memory page mapped. While __get_user_pages will check both. QEMU will clear the poisoned page table entry (via unmap/map) to make it possible to allocate a new memory page for the virtual address across guest rebooting. But it is also possible that the underlying memory page is kept poisoned even after the corresponding page table entry is cleared, that is, a new memory page can not be allocated. __get_user_pages can catch these situations. Signed-off-by:
-
Huang Ying authored
Make __get_user_pages return -EHWPOISON for HWPOISON page only if FOLL_HWPOISON is specified. With this patch, the interested callers can distinguish HWPOISON pages from general FAULT pages, while other callers will still get -EFAULT for all these pages, so the user space interface need not to be changed. This feature is needed by KVM, where UCR MCE should be relayed to guest for HWPOISON page, while instruction emulation and MMIO will be tried for general FAULT page. The idea comes from Andrew Morton. Signed-off-by:
-
Huang Ying authored
In most cases, get_user_pages and get_user_pages_fast should be used to pin user pages in memory. But sometimes, some special flags except FOLL_GET, FOLL_WRITE and FOLL_FORCE are needed, for example in following patch, KVM needs FOLL_HWPOISON. To support these users, __get_user_pages is exported directly. There are some symbol name conflicts in infiniband driver, fixed them too. Signed-off-by:
-
john cooper authored
A correction to Intel cpu model CPUID data (patch queued) caused winxp to BSOD when booted with a Penryn model. This was traced to the CPUID "model" field correction from 6 -> 23 (as is proper for a Penryn class of cpu). Only in this case does the problem surface. The cause for this failure is winxp accessing the BBL_CR_CTL3 MSR which is unsupported by current kvm, appears to be a legacy MSR not fully characterized yet existing in current silicon, and is apparently carried forward in MSR space to accommodate vintage code as here. It is not yet conclusive whether this MSR implements any of its legacy functionality or is just an ornamental dud for compatibility. While I found no silicon version specific documentation link to this MSR, a general description exists in Intel's developer's reference which agrees with the functional behavior of other bootloader/kernel code I've examined accessing BBL_CR_CTL3. Regrettably winxp appears to be setting bit #19 called out as "reserved" in the above document. So to minimally accommodate this MSR, kvm msr get will provide the equivalent mock data and kvm msr write will simply toss the guest passed data without interpretation. While this treatment of BBL_CR_CTL3 addresses the immediate problem, the approach may be modified pending clarification from Intel. Signed-off-by:
john cooper <john.cooper@redhat.com> Signed-off-by:
-
Xiao Guangrong authored
Now, we have 'vcpu->mode' to judge whether need to send ipi to other cpus, this way is very exact, so checking request bit is needless, then we can drop the spinlock let it's collateral Signed-off-by:
-
Xiao Guangrong authored
Currently we keep track of only two states: guest mode and host mode. This patch adds an "exiting guest mode" state that tells us that an IPI will happen soon, so unless we need to wait for the IPI, we can avoid it completely. Also 1: No need atomically to read/write ->mode in vcpu's thread 2: reorganize struct kvm_vcpu to make ->mode and ->requests in the same cache line explicitly Signed-off-by:
-
Heiko Carstens authored
Get rid of this warning: CC arch/s390/kvm/../../../virt/kvm/kvm_main.o arch/s390/kvm/../../../virt/kvm/kvm_main.c:596:12: warning: 'kvm_create_dirty_bitmap' defined but not used The only caller of the function is within a !CONFIG_S390 section, so add the same ifdef around kvm_create_dirty_bitmap() as well. Signed-off-by:
Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by:
-
Jan Kiszka authored
This case is a pure user space error we do not need to record. Moreover, it can be misused to flood the kernel log. Remove it. Signed-off-by:
-
Xiao Guangrong authored
Fix: [ 1001.499596] =================================================== [ 1001.499599] [ INFO: suspicious rcu_dereference_check() usage. ] [ 1001.499601] --------------------------------------------------- [ 1001.499604] include/linux/kvm_host.h:301 invoked rcu_dereference_check() without protection! ...... [ 1001.499636] Pid: 6035, comm: qemu-system-x86 Not tainted 2.6.37-rc6+ #62 [ 1001.499638] Call Trace: [ 1001.499644] [] lockdep_rcu_dereference+0x9d/0xa5 [ 1001.499653] [] gfn_to_memslot+0x8d/0xc8 [kvm] [ 1001.499661] [] gfn_to_hva+0x16/0x3f [kvm] [ 1001.499669] [] kvm_read_guest_page+0x1e/0x5e [kvm] [ 1001.499681] [] kvm_read_guest_page_mmu+0x53/0x5e [kvm] [ 1001.499699] [] load_pdptrs+0x3f/0x9c [kvm] [ 1001.499705] [] ? vmx_set_cr0+0x507/0x517 [kvm_intel] [ 1001.499717] [] kvm_arch_vcpu_ioctl_set_sregs+0x1f3/0x3c0 [kvm] [ 1001.499727] [] kvm_vcpu_ioctl+0x6a5/0xbc5 [kvm] Signed-off-by:
-
Avi Kivity authored
Instead of exchanging the guest and host rcx, have separate storage for each. This allows us to avoid using the xchg instruction, which is is a little slower than normal operations. Signed-off-by:
-
Avi Kivity authored
Change push top-of-stack pop guest-rcx pop dummy to pop guest-rcx which is the same thing, only simpler. Signed-off-by:
-
Rik van Riel authored
On some CPUs, a ple_gap of 41 is simply insufficient to ever trigger PLE exits, even with the minimalistic PLE test from kvm-unit-tests. http://git.kernel.org/?p=virt/kvm/kvm-unit-tests.git;a=commitdiff;h=eda71b28fa122203e316483b35f37aaacd42f545 For example, the Xeon X5670 CPU needs a ple_gap of at least 48 in order to get pause loop exits: # modprobe kvm_intel ple_gap=47 # taskset 1 /usr/local/bin/qemu-system-x86_64 \ -device testdev,chardev=log -chardev stdio,id=log \ -kernel x86/vmexit.flat -append ple-round-robin -smp 2 VNC server running on `::1:5900' enabling apic enabling apic ple-round-robin 58298446 # rmmod kvm_intel # modprobe kvm_intel ple_gap=48 # taskset 1 /usr/local/bin/qemu-system-x86_64 \ -device testdev,chardev=log -chardev stdio,id=log \ -kernel x86/vmexit.flat -append ple-round-robin -smp 2 VNC server running on `::1:5900' enabling apic enabling apic ple-round-robin 36616 Increase the ple_gap to 128 to be on the safe side. Signed-off-by:
Zhai, Edwin <edwin.zhai@intel.com> Signed-off-by:
-
Joerg Roedel authored
This patch adds the necessary code to run perf-kvm on AMD machines. Signed-off-by:
Joerg Roedel <joerg.roedel@amd.com> Signed-off-by:
-
