Revision 313f636d5c490c9741d3f750dc8da33029edbc6b authored by David Matlack on 09 March 2016, 00:19:44 UTC, committed by Paolo Bonzini on 09 March 2016, 10:54:14 UTC
When growing halt-polling, there is no check that the poll time exceeds the limit. It's possible for vcpu->halt_poll_ns grow once past halt_poll_ns, and stay there until a halt which takes longer than vcpu->halt_poll_ns. For example, booting a Linux guest with halt_poll_ns=11000: ... kvm:kvm_halt_poll_ns: vcpu 0: halt_poll_ns 0 (shrink 10000) ... kvm:kvm_halt_poll_ns: vcpu 0: halt_poll_ns 10000 (grow 0) ... kvm:kvm_halt_poll_ns: vcpu 0: halt_poll_ns 20000 (grow 10000) Signed-off-by: David Matlack <dmatlack@google.com> Fixes: aca6ff29c4063a8d467cdee241e6b3bf7dc4a171 Cc: stable@vger.kernel.org Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
1 parent 9522b37
| File | Mode | Size |
|---|---|---|
| broadcom | ||
| efi | ||
| Kconfig | -rw-r--r-- | 6.0 KB |
| Makefile | -rw-r--r-- | 908 bytes |
| arm_scpi.c | -rw-r--r-- | 19.8 KB |
| dcdbas.c | -rw-r--r-- | 15.6 KB |
| dcdbas.h | -rw-r--r-- | 2.8 KB |
| dell_rbu.c | -rw-r--r-- | 19.1 KB |
| dmi-id.c | -rw-r--r-- | 6.5 KB |
| dmi-sysfs.c | -rw-r--r-- | 16.8 KB |
| dmi_scan.c | -rw-r--r-- | 25.5 KB |
| edd.c | -rw-r--r-- | 20.0 KB |
| iscsi_ibft.c | -rw-r--r-- | 19.4 KB |
| iscsi_ibft_find.c | -rw-r--r-- | 2.8 KB |
| memmap.c | -rw-r--r-- | 12.0 KB |
| pcdp.c | -rw-r--r-- | 3.3 KB |
| pcdp.h | -rw-r--r-- | 2.7 KB |
| psci.c | -rw-r--r-- | 11.1 KB |
| qcom_scm-32.c | -rw-r--r-- | 12.7 KB |
| qcom_scm-64.c | -rw-r--r-- | 1.9 KB |
| qcom_scm.c | -rw-r--r-- | 3.0 KB |
| qcom_scm.h | -rw-r--r-- | 1.6 KB |
| raspberrypi.c | -rw-r--r-- | 6.6 KB |
Computing file changes ...
