diff options
Diffstat (limited to '')
| -rw-r--r-- | results/classifier/108/device/1854 | 33 | ||||
| -rw-r--r-- | results/classifier/108/device/1854878 | 63 |
2 files changed, 96 insertions, 0 deletions
diff --git a/results/classifier/108/device/1854 b/results/classifier/108/device/1854 new file mode 100644 index 00000000..bcd6f27b --- /dev/null +++ b/results/classifier/108/device/1854 @@ -0,0 +1,33 @@ +device: 0.934 +graphic: 0.892 +files: 0.865 +PID: 0.855 +permissions: 0.661 +vnc: 0.655 +debug: 0.634 +network: 0.626 +semantic: 0.578 +socket: 0.563 +boot: 0.363 +performance: 0.288 +other: 0.127 +KVM: 0.087 + +s390x: qemu-user: ERROR:../linux-user/elfload.c:2239:zero_bss: code should not be reached +Description of problem: +The nolibc-test program from the Linux kernel crashes since 5f4e5b34092556ab1577e25d1262bd5975b26980 . +Reverting that commit fixes the issue. +Steps to reproduce: +1. Build `nolibc-test` for s390x from Linux kernel tree. (from `tools/testing/selftests/nolibc/`). EDIT: compiled binary is uploaded below. +2. Run it under qemu-s390x. + +``` + ./qemu-s390x nolibc-test +** +ERROR:../linux-user/elfload.c:2239:zero_bss: code should not be reached +Bail out! ERROR:../linux-user/elfload.c:2239:zero_bss: code should not be reached +Aborted (core dumped) + +``` +Additional information: + diff --git a/results/classifier/108/device/1854878 b/results/classifier/108/device/1854878 new file mode 100644 index 00000000..250b55b2 --- /dev/null +++ b/results/classifier/108/device/1854878 @@ -0,0 +1,63 @@ +device: 0.964 +graphic: 0.923 +semantic: 0.751 +performance: 0.734 +other: 0.676 +PID: 0.608 +permissions: 0.568 +files: 0.468 +socket: 0.438 +debug: 0.394 +boot: 0.389 +network: 0.370 +vnc: 0.334 +KVM: 0.281 + +Physical USB thumbdrive treated as read-only + +So I have installed FreeDOS on my USB thumbdrive, by using Rufus. Everything goes as expected so far. That's good. + +When I run QEMU with this command line: +qemu-system-x86_64.exe -drive file=\\.\PhysicalDrive1 + +it of course is read-only, just like the resulting console message says: +WARNING: Image format was not specified for '\\.\PhysicalDrive1' and probing guessed raw. + Automatically detecting the format is dangerous for raw images, write operations on block 0 will be restricted. + Specify the 'raw' format explicitly to remove the restrictions. + + +So what I then did, was I ran QEMU with this command line: +qemu-system-x86_64.exe -drive file=\\.\PhysicalDrive1,format=raw + +As expected, the above mentioned console message no longer appears. +However, beyond that, QEMU doesn't behave as it should regarding read-only status. When I try any operation that involves writing to the drive, it becomes quite clear that the drive is still read-only. Any writing operations to the drive result in FreeDOS giving me the error message: +Error writing to drive C: DOS area: sector not found. + +The above situation is clearly a bug. QEMU should not be treating it as read-only once I specify format=raw. + +Note that drive C is how the guest OS refers to the USB thumbdrive (it's drive E in my host OS, and drive C in my host OS is the actual system drive). + +And yes, it is a QEMU bug. It's not a FreeDOS bug I tested it with this command line, so that all changes would be written to a temporary snapshot file: +qemu-system-x86_64.exe -drive file=\\.\PhysicalDrive1,format=raw,snapshot +That last drive option "snapshot" tells QEMU to create a temporary snapshot file, and to write all changes to that. When I do that, all write operations are successful. So it seems that there is a bug in QEMU where it keeps read-only mode in place for a physical drive, even when format=raw is specified. Please fix this bug. Thanks in advance. + +Here's my current setup. +Host OS: Windows 10 (64bit) +Guest OS: FreeDOS +QEMU version: 4.1.0 + +The QEMU project is currently considering to move its bug tracking to +another system. For this we need to know which bugs are still valid +and which could be closed already. Thus we are setting older bugs to +"Incomplete" now. + +If you still think this bug report here is valid, then please switch +the state back to "New" within the next 60 days, otherwise this report +will be marked as "Expired". Or please mark it as "Fix Released" if +the problem has been solved with a newer version of QEMU already. + +Thank you and sorry for the inconvenience. + + +[Expired for QEMU because there has been no activity for 60 days.] + |
