summary refs log tree commit diff stats
path: root/results/classifier/108/other/1349277
diff options
context:
space:
mode:
Diffstat (limited to 'results/classifier/108/other/1349277')
-rw-r--r--results/classifier/108/other/134927777
1 files changed, 77 insertions, 0 deletions
diff --git a/results/classifier/108/other/1349277 b/results/classifier/108/other/1349277
new file mode 100644
index 000000000..7b5adbc11
--- /dev/null
+++ b/results/classifier/108/other/1349277
@@ -0,0 +1,77 @@
+graphic: 0.870
+performance: 0.856
+vnc: 0.820
+device: 0.810
+permissions: 0.804
+PID: 0.804
+semantic: 0.747
+socket: 0.732
+files: 0.691
+debug: 0.686
+KVM: 0.680
+network: 0.654
+boot: 0.651
+other: 0.563
+
+AArch64 emulation ignores SPSel=0 when taking (or returning from) an exception at EL1 or greater
+
+The AArch64 emulation ignores SPSel=0 when:
+
+(1) taking an interrupt from an exception level greater than EL0 (e.g., EL1t),
+
+(2) returning from an exception (via ERET) to an exception level greater than EL0 (e.g., EL1t), with SPSR_ELx[SPSel]=0.
+
+The attached patch fixes the problem in my application.
+
+Background:
+
+I'm running a standalone application (toy OS) that is performing preemptive multithreading between threads running at EL1t, with exception handling / context switching occurring at EL1h.  This bug causes the stack pointer to be corrupted in the threads running at EL1t (they end up with a version of the EL1h stack pointer (SP_EL1)).
+
+Occurs in:
+	qemu-2.1.0-rc1 (found in)
+	commit c60a57ff497667780132a3fcdc1500c83af5d5c0 (current master)
+
+
+
+On Mon, Jul 28, 2014 at 07:40:56AM -0000, T McIntosh wrote:
+> Public bug reported:
+> 
+> The AArch64 emulation ignores SPSel=0 when:
+> 
+> (1) taking an interrupt from an exception level greater than EL0 (e.g.,
+> EL1t),
+> 
+> (2) returning from an exception (via ERET) to an exception level greater
+> than EL0 (e.g., EL1t), with SPSR_ELx[SPSel]=0.
+> 
+> The attached patch fixes the problem in my application.
+
+Hi,
+
+Patches 1-3 in the following series fix the this problem.
+http://lists.gnu.org/archive/html/qemu-devel/2014-06/msg03675.html
+
+Cheers,
+Edgar
+
+
+The attachment "Proposed fix" seems to be a patch.  If it isn't, please remove the "patch" flag from the attachment, remove the "patch" tag, and if you are a member of the ~ubuntu-reviewers, unsubscribe the team.
+
+[This is an automated message performed by a Launchpad user owned by ~brian-murray, for any issues please contact him.]
+
+Uploaded fixed package for Vivid: https://launchpad.net/ubuntu/+source/qemu/2.1+dfsg-7ubuntu3
+
+Please let me know if this fixes the issue.
+
+This bug was fixed in the package qemu - 2.1+dfsg-7ubuntu3
+
+---------------
+qemu (2.1+dfsg-7ubuntu3) vivid; urgency=medium
+
+  * d/p/target-arm-A64-Break-out-aarch64_save-restore_sp.patch
+    d/p/target-arm-A64-Respect-SPSEL-in-ERET-SP-restore.patch
+    d/p/target-arm-A64-Respect-SPSEL-when-taking-exceptions.patch:
+    Cherry-pick of upstream patches in order to fix AArch64 emulation ignoring
+    SPSel=0 in certain conditions. (LP: #1349277)
+ -- Chris J Arges <email address hidden>   Thu, 04 Dec 2014 14:17:01 -0600
+