diff options
Diffstat (limited to 'results/classifier/118/kernel/1119686')
| -rw-r--r-- | results/classifier/118/kernel/1119686 | 76 |
1 files changed, 0 insertions, 76 deletions
diff --git a/results/classifier/118/kernel/1119686 b/results/classifier/118/kernel/1119686 deleted file mode 100644 index fdba220c..00000000 --- a/results/classifier/118/kernel/1119686 +++ /dev/null @@ -1,76 +0,0 @@ -kernel: 0.857 -architecture: 0.855 -debug: 0.815 -x86: 0.802 -files: 0.785 -virtual: 0.760 -hypervisor: 0.754 -semantic: 0.745 -graphic: 0.740 -device: 0.726 -ppc: 0.720 -permissions: 0.706 -network: 0.696 -performance: 0.687 -user-level: 0.680 -socket: 0.664 -i386: 0.654 -vnc: 0.624 -peripherals: 0.620 -risc-v: 0.599 -register: 0.598 -arm: 0.594 -KVM: 0.589 -VMM: 0.559 -TCG: 0.554 -PID: 0.547 -boot: 0.508 -assembly: 0.381 -mistranslation: 0.327 - -Incorrect handling of icebp - -Wine conformance suite tests the behavior of various low-level Windows API functions. One of the tests involves checking the interaction of breakpoints and exceptions, and in particular the 'icebp' breakpoint. This test works on a Windows XP machine running either on the metal or in VMware ESX but fails when run in QEmu. - -To reproduce the issue grab the attached 'exception.exe' file and run it. If you get 'Test failed' lines like below then it means the problem is still present: - - exception.c:202: exception 0: 80000004 flags:0 addr:003F0000 - exception.c:208: Test failed: 0: Wrong exception address 003F0000/003F0001 - exception.c:214: this is the last test seen before the exception - exception: unhandled exception 80000004 at 003F0000 - exception.c:202: exception 0: c0000027 flags:2 addr:7C80E0B9 - exception.c:205: Test failed: 0: Wrong exception code c0000027/80000004 - exception.c:208: Test failed: 0: Wrong exception address 7C80E0B9/003F0001 - -Note that this bug was not present in QEmu 1.1.2+dfsg-5 (Debian Testing) but is now present in 1.4.0~rc0+dfsg-1exp (Debian Experimental). - - - -This bug is still present in QEMU 1.6.0 (as per Debian's qemu-system-x86 1.6.0+dfsg-1 package). - - -This bug is still present in QEMU 1.7.0 (as per Debian's qemu-system-x86 1.7.0+dfsg-3 package). - -The patch submitted upstream was for the kernel. Is this also a bug in QEMU when TCG is disabled? - -s/TCG/KVM/ - Is this also a bug when KVM is disabled? - -[Expired for QEMU because there has been no activity for 60 days.] - -Actually this got fixed by the following Linux kernel commit: - -https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=fd2a445a94d2ab6b39fb623dc02fee48d01a565a - -commit fd2a445a94d2ab6b39fb623dc02fee48d01a565a (patch) - -KVM: VMX: Advance rip to after an ICEBP instruction -When entering an exception after an ICEBP, the saved instruction -pointer should point to after the instruction. - -This fixes the bug here: https://bugs.launchpad.net/qemu/+bug/1119686 - -Signed-off-by: Huw Davies <email address hidden> -Reviewed-by: Jan Kiszka <email address hidden> -Signed-off-by: Marcelo Tosatti <email address hidden> - - |