summary refs log tree commit diff stats
path: root/results/classifier/gemma3:12b/device/1164
diff options
context:
space:
mode:
Diffstat (limited to 'results/classifier/gemma3:12b/device/1164')
-rw-r--r--results/classifier/gemma3:12b/device/116418
1 files changed, 18 insertions, 0 deletions
diff --git a/results/classifier/gemma3:12b/device/1164 b/results/classifier/gemma3:12b/device/1164
new file mode 100644
index 00000000..4f7b5d91
--- /dev/null
+++ b/results/classifier/gemma3:12b/device/1164
@@ -0,0 +1,18 @@
+
+q35: incorrect values for PCIEXBAR masks
+Description of problem:
+https://lore.kernel.org/all/1fded151ce5ecbf7010427871b908000b2aba9ee.1520867956.git.x1917x@gmail.com/
+
+In function [mch_update_pciexbar](https://gitlab.com/qemu-project/qemu/-/blob/master/hw/pci-host/q35.c#L295)
+
+There are two small issues in PCIEXBAR address mask handling:
+- wrong bit positions for address mask bits (see PCIEXBAR description
+  in Q35 datasheet)
+- incorrect usage of 64ADR_MASK
+
+Due to this, attempting to write a valid PCIEXBAR address may cause it to
+shift to another address, causing memory layout corruption where emulated
+MMIO regions may overlap real (passed through) MMIO ranges. Fix this
+by providing correct values.
+Additional information:
+Q35 datasheet: https://www.intel.com/Assets/PDF/datasheet/316966.pdf  ( 5.1.16 PCIEXBAR—PCI Express* Register Range Base Address )