diff options
| author | Christian Krinitsin <mail@krinitsin.com> | 2025-05-30 16:52:07 +0200 |
|---|---|---|
| committer | Christian Krinitsin <mail@krinitsin.com> | 2025-05-30 16:52:17 +0200 |
| commit | 9260319e7411ff8281700a532caa436f40120ec4 (patch) | |
| tree | 2f6bfe5f3458dd49d328d3a9eb508595450adec0 /gitlab/issues_text/target_ppc/host_missing/accel_missing/1958 | |
| parent | 225caa38269323af1bfc2daadff5ec8bd930747f (diff) | |
| download | qemu-analysis-9260319e7411ff8281700a532caa436f40120ec4.tar.gz qemu-analysis-9260319e7411ff8281700a532caa436f40120ec4.zip | |
gitlab scraper: download in toml and text format
Diffstat (limited to 'gitlab/issues_text/target_ppc/host_missing/accel_missing/1958')
| -rw-r--r-- | gitlab/issues_text/target_ppc/host_missing/accel_missing/1958 | 21 |
1 files changed, 21 insertions, 0 deletions
diff --git a/gitlab/issues_text/target_ppc/host_missing/accel_missing/1958 b/gitlab/issues_text/target_ppc/host_missing/accel_missing/1958 new file mode 100644 index 000000000..5586e2288 --- /dev/null +++ b/gitlab/issues_text/target_ppc/host_missing/accel_missing/1958 @@ -0,0 +1,21 @@ +PPC msgsnd for DOORBELL CRITICAL masked by MSR[EE] instead of MSR[CE] +Description of problem: +When executing PPC instruction "msgsnd r3. with r3 = 0x08000001" an DOORBELL CRITICAL exception is raised on core number 1. But this exception is masked by MSR\[EE\] bit, the MSR\[EE\] should be set to 1 in core1 to get this exception. But the NXP E500MCRM.pdf reference manual indicates that MSR\[CE\] is the mask bit for DOORBELL_CRITICAL Exception. +Additional information: +In qemu-8.1.2/target/ppc/excp_helper.c i try to change in ppc_next_unmasked_interrupt_generic function: + +``` +if (FIELD_EX64(env->msr, MSR, CE)) { + /* Critical doorbell */ + if (env->pending_interrupts & PPC_INTERRUPT_CDOORBELL) { <- move this part from (async_deliver != 0) + return PPC_INTERRUPT_CDOORBELL; + } + /* External critical interrupt */ + if (env->pending_interrupts & PPC_INTERRUPT_CEXT) { + return PPC_INTERRUPT_CEXT; + } +} +``` + + +And it seems to work in my case. |