qemu/target
Ilya Leoshkevich 066b9de4b3 target/s390x: Fix MVC not always invalidating translation blocks
Node.js crashes in qemu-system-s390x with random SIGSEGVs / SIGILLs.

The v8 JIT used by Node.js can garbage collect and overwrite unused
code. Overwriting is performed by WritableJitAllocation::CopyCode(),
which ultimately calls memcpy(). For certain sizes, memcpy() uses the
MVC instruction.

QEMU implements MVC and other similar instructions using helpers. While
TCG store ops invalidate affected translation blocks automatically,
helpers must do this manually by calling probe_access_flags(). The MVC
helper does this using the access_prepare() -> access_prepare_nf() ->
s390_probe_access() -> probe_access_flags() call chain.

At the last step of this chain, the store size is replaced with 0. This
causes the probe_access_flags() -> notdirty_write() ->
tb_invalidate_phys_range_fast() chain to miss some translation blocks.

When this happens, QEMU executes a mix of old and new code. This
quickly leads to either a SIGSEGV or a SIGILL in case the old code
ends in the middle of a new instruction.

Fix by passing the true size.

Reported-by: Berthold Gunreben <azouhr@opensuse.org>
Cc: Sarah Kriesch <ada.lovelace@gmx.de>
Cc: qemu-stable@nongnu.org
Closes: https://bugzilla.opensuse.org/show_bug.cgi?id=1235709
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Fixes: e2faabee78 ("accel/tcg: Forward probe size on to notdirty_write")
Message-ID: <20250128001338.11474-1-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit e43ced8be1)
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
2025-02-01 11:59:14 +03:00
..
alpha target/alpha: Explicitly set 2-NaN propagation rule 2024-11-05 10:09:56 +00:00
arm target/arm: arm_reset_sve_state() should set FPSR, not FPCR 2025-01-29 22:25:22 +03:00
avr target/avr: Use explicit little-endian LD/ST API 2024-10-15 12:13:59 -03:00
hexagon target/hexagon: Use explicit little-endian LD/ST API 2024-10-15 11:55:09 -03:00
hppa target/hppa: Explicitly set 2-NaN propagation rule 2024-11-05 10:09:54 +00:00
i386 i386/cpu: Mark avx10_version filtered when prefix is NULL 2025-01-17 21:54:51 +03:00
loongarch target/loongarch: Use actual operand size with vbsrl check 2024-12-29 01:50:54 +03:00
m68k target/m68k: Initialize float_status fields in gdb set/get functions 2024-11-05 10:09:54 +00:00
microblaze Misc HW patch queue 2024-11-06 17:28:45 +00:00
mips target-arm queue: 2024-11-05 21:27:18 +00:00
openrisc target/openrisc: Explicitly set 2-NaN propagation rule 2024-11-05 10:09:57 +00:00
ppc target/ppc: Fix THREAD_SIBLING_FOREACH for multi-socket 2024-11-27 02:49:31 +10:00
riscv target/riscv: Avoid bad shift in riscv_cpu_do_interrupt() 2024-12-03 12:26:24 +01:00
rx target/rx: Explicitly set 2-NaN propagation rule 2024-11-05 10:09:57 +00:00
s390x target/s390x: Fix MVC not always invalidating translation blocks 2025-02-01 11:59:14 +03:00
sh4 license: Update deprecated SPDX tag LGPL-2.0+ to LGPL-2.0-or-later 2024-09-20 10:11:59 +03:00
sparc target/sparc: Explicitly set 2-NaN propagation rule 2024-11-05 10:09:55 +00:00
tricore target/tricore: Use tcg_constant_tl() instead of tcg_gen_movi_tl() 2024-10-15 12:13:59 -03:00
xtensa target/xtensa: Explicitly set 2-NaN propagation rule 2024-11-05 10:09:55 +00:00
Kconfig target/cris: Remove the deprecated CRIS target 2024-09-13 20:11:13 +02:00
meson.build target/cris: Remove the deprecated CRIS target 2024-09-13 20:11:13 +02:00