[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[PATCH] target/i386: ensure EXCP0D_GPF is propagated back to the guest
From: |
Mark Cave-Ayland |
Subject: |
[PATCH] target/i386: ensure EXCP0D_GPF is propagated back to the guest |
Date: |
Sat, 30 Oct 2021 14:29:43 +0100 |
In the case where mmu_translate() returns EXCP0D_GPF ensure that
handle_mmu_fault()
returns immediately to propagate the fault back to the guest instead of
returning
EXCP0E_PAGE.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Fixes: 661ff4879e ("target/i386: extract mmu_translate")
Closes: https://gitlab.com/qemu-project/qemu/-/issues/394
---
[Paolo: this appears to fix the regression booting Windows 7 on TCG that
appeared in 6.1
as per the above Gitlab issue. Unfortunately as I'm not really familiar with
x86 it will
probably need a better implementation/description but it should at least
indicate what
the problem is.]
target/i386/tcg/sysemu/excp_helper.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/target/i386/tcg/sysemu/excp_helper.c
b/target/i386/tcg/sysemu/excp_helper.c
index 7af887be4d..0170f7f791 100644
--- a/target/i386/tcg/sysemu/excp_helper.c
+++ b/target/i386/tcg/sysemu/excp_helper.c
@@ -439,6 +439,10 @@ static int handle_mmu_fault(CPUState *cs, vaddr addr, int
size,
prot, mmu_idx, page_size);
return 0;
} else {
+ if (cs->exception_index == EXCP0D_GPF) {
+ return 1;
+ }
+
if (env->intercept_exceptions & (1 << EXCP0E_PAGE)) {
/* cr2 is not modified in case of exceptions */
x86_stq_phys(cs,
--
2.20.1
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- [PATCH] target/i386: ensure EXCP0D_GPF is propagated back to the guest,
Mark Cave-Ayland <=