[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[PULL 14/14] tcg: Allow top bit of SIMD_DATA_BITS to be set in simd_desc
From: |
Richard Henderson |
Subject: |
[PULL 14/14] tcg: Allow top bit of SIMD_DATA_BITS to be set in simd_desc() |
Date: |
Fri, 15 Nov 2024 12:58:49 -0800 |
From: Peter Maydell <peter.maydell@linaro.org>
In simd_desc() we create a SIMD descriptor from various pieces
including an arbitrary data value from the caller. We try to
sanitize these to make sure everything will fit: the 'data' value
needs to fit in the SIMD_DATA_BITS (== 22) sized field. However we
do that sanitizing with:
tcg_debug_assert(data == sextract32(data, 0, SIMD_DATA_BITS));
This works for the case where the data is supposed to be considered
as a signed integer (which can then be returned via simd_data()).
However, some callers want to treat the data value as unsigned.
Specifically, for the Arm SVE operations, make_svemte_desc()
assembles a data value as a collection of fields, and it needs to use
all 22 bits. Currently if MTE is enabled then its MTEDESC SIZEM1
field may have the most significant bit set, and then it will trip
this assertion.
Loosen the assertion so that we only check that the data value will
fit into the field in some way, either as a signed or as an unsigned
value. This means we will fail to detect some kinds of bug in the
callers, but we won't spuriously assert for intentional use of the
data field as unsigned.
Cc: qemu-stable@nongnu.org
Fixes: db432672dc50e ("tcg: Add generic vector expanders")
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2601
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-ID: <20241115172515.1229393-1-peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
tcg/tcg-op-gvec.c | 15 ++++++++++++++-
1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/tcg/tcg-op-gvec.c b/tcg/tcg-op-gvec.c
index 78ee1ced80..97e4df221a 100644
--- a/tcg/tcg-op-gvec.c
+++ b/tcg/tcg-op-gvec.c
@@ -88,7 +88,20 @@ uint32_t simd_desc(uint32_t oprsz, uint32_t maxsz, int32_t
data)
uint32_t desc = 0;
check_size_align(oprsz, maxsz, 0);
- tcg_debug_assert(data == sextract32(data, 0, SIMD_DATA_BITS));
+
+ /*
+ * We want to check that 'data' will fit into SIMD_DATA_BITS.
+ * However, some callers want to treat the data as a signed
+ * value (which they can later get back with simd_data())
+ * and some want to treat it as an unsigned value.
+ * So here we assert only that the data will fit into the
+ * field in at least one way. This means that some invalid
+ * values from the caller will not be detected, e.g. if the
+ * caller wants to handle the value as a signed integer but
+ * incorrectly passes us 1 << (SIMD_DATA_BITS - 1).
+ */
+ tcg_debug_assert(data == sextract32(data, 0, SIMD_DATA_BITS) ||
+ data == extract32(data, 0, SIMD_DATA_BITS));
oprsz = (oprsz / 8) - 1;
maxsz = (maxsz / 8) - 1;
--
2.43.0
- [PULL 03/14] linux-user: Tolerate CONFIG_LSM_MMAP_MIN_ADDR, (continued)
- [PULL 03/14] linux-user: Tolerate CONFIG_LSM_MMAP_MIN_ADDR, Richard Henderson, 2024/11/15
- [PULL 04/14] tests/tcg: Test that sigreturn() does not corrupt the signal mask, Richard Henderson, 2024/11/15
- [PULL 05/14] target/i386: fix hang when using slow path for ptw_setl, Richard Henderson, 2024/11/15
- [PULL 08/14] linux-user: Drop image_info.alignment, Richard Henderson, 2024/11/15
- [PULL 06/14] cpu: ensure we don't call start_exclusive from cpu_exec, Richard Henderson, 2024/11/15
- [PULL 09/14] linux-user/aarch64: Reduce vdso alignment to 4k, Richard Henderson, 2024/11/15
- [PULL 10/14] linux-user/arm: Reduce vdso alignment to 4k, Richard Henderson, 2024/11/15
- [PULL 11/14] linux-user/loongarch64: Reduce vdso alignment to 4k, Richard Henderson, 2024/11/15
- [PULL 12/14] linux-user/ppc: Reduce vdso alignment to 4k, Richard Henderson, 2024/11/15
- [PULL 07/14] linux-user: Honor elf alignment when placing images, Richard Henderson, 2024/11/15
- [PULL 14/14] tcg: Allow top bit of SIMD_DATA_BITS to be set in simd_desc(),
Richard Henderson <=
- [PULL 13/14] linux-user/arm: Select vdso for be8 and be32 modes, Richard Henderson, 2024/11/15
- Re: [PULL 00/14] tcg + linux-user patch queue, Peter Maydell, 2024/11/16