[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Stable-8.1.2 22/45] accel/tcg: mttcg remove false-negative halted asser
From: |
Michael Tokarev |
Subject: |
[Stable-8.1.2 22/45] accel/tcg: mttcg remove false-negative halted assertion |
Date: |
Wed, 4 Oct 2023 11:01:43 +0300 |
From: Nicholas Piggin <npiggin@gmail.com>
mttcg asserts that an execution ending with EXCP_HALTED must have
cpu->halted. However between the event or instruction that sets
cpu->halted and requests exit and the assertion here, an
asynchronous event could clear cpu->halted.
This leads to crashes running AIX on ppc/pseries because it uses
H_CEDE/H_PROD hcalls, where H_CEDE sets self->halted = 1 and
H_PROD sets other cpu->halted = 0 and kicks it.
H_PROD could be turned into an interrupt to wake, but several other
places in ppc, sparc, and semihosting follow what looks like a similar
pattern setting halted = 0 directly. So remove this assertion.
Reported-by: Ivan Warren <ivan@vmfacility.fr>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Message-Id: <20230829010658.8252-1-npiggin@gmail.com>
[rth: Keep the case label and adjust the comment.]
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
(cherry picked from commit 0e5903436de712844b0e6cdd862b499c767e09e9)
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
diff --git a/accel/tcg/tcg-accel-ops-mttcg.c b/accel/tcg/tcg-accel-ops-mttcg.c
index b276262007..4b0dfb4be7 100644
--- a/accel/tcg/tcg-accel-ops-mttcg.c
+++ b/accel/tcg/tcg-accel-ops-mttcg.c
@@ -100,14 +100,9 @@ static void *mttcg_cpu_thread_fn(void *arg)
break;
case EXCP_HALTED:
/*
- * during start-up the vCPU is reset and the thread is
- * kicked several times. If we don't ensure we go back
- * to sleep in the halted state we won't cleanly
- * start-up when the vCPU is enabled.
- *
- * cpu->halted should ensure we sleep in wait_io_event
+ * Usually cpu->halted is set, but may have already been
+ * reset by another thread by the time we arrive here.
*/
- g_assert(cpu->halted);
break;
case EXCP_ATOMIC:
qemu_mutex_unlock_iothread();
--
2.39.2
- [Stable-8.1.2 24/45] ui/vnc: fix debug output for invalid audio message, (continued)
- [Stable-8.1.2 24/45] ui/vnc: fix debug output for invalid audio message, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 27/45] migration: Fix possible race when setting rp_state.error, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 19/45] hw/arm/boot: Set SCR_EL3.FGTEn when booting kernel, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 20/45] target/arm: Don't skip MTE checks for LDRT/STRT at EL0, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 21/45] meson.build: Make keyutils independent from keyring, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 23/45] hw/scsi/scsi-disk: Disallow block sizes smaller than 512 [CVE-2023-42467], Michael Tokarev, 2023/10/04
- [Stable-8.1.2 18/45] include/exec: Widen tlb_hit/tlb_hit_page(), Michael Tokarev, 2023/10/04
- [Stable-8.1.2 25/45] ui/vnc: fix handling of VNC_FEATURE_XVP, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 26/45] migration: Fix race that dest preempt thread close too early, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 17/45] tests/file-io-error: New test, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 22/45] accel/tcg: mttcg remove false-negative halted assertion,
Michael Tokarev <=
- [Stable-8.1.2 29/45] migration: Fix possible race when shutting down to_dst_file, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 30/45] migration: Remove redundant cleanup of postcopy_qemufile_src, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 28/45] migration: Fix possible races when shutting down the return path, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 31/45] migration: Consolidate return path closing code, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 33/45] migration: Move return path cleanup to main migration thread, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 32/45] migration: Replace the return path retry logic, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 34/45] softmmu: Use async_run_on_cpu in tcg_commit, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 35/45] accel/tcg: Avoid load of icount_decr if unused, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 39/45] accel/tcg: Always set CF_LAST_IO with CF_NOIRQ, Michael Tokarev, 2023/10/04
- [Stable-8.1.2 40/45] accel/tcg: Always require can_do_io, Michael Tokarev, 2023/10/04