Hi all,
In this patch (v5), I've removed the redundant call to mark_vs_dirty(s)
within the gen_ldst_stride_main_loop() function.
The reason for this change is that mark_vs_dirty(s) is already being called
at a higher level, making the call inside gen_ldst_stride_main_loop()
unnecessary.
static void gen_ldst_stride_main_loop(...)
{
...
- mark_vs_dirty(s);
...
}
static bool ldst_stride_trans(...)
{
....
mark_vs_dirty(s);
gen_ldst_stride_main_loop(s, dest, rs1, rs2, vm, nf, ld_fn, st_fn,
is_load);
}
patch v4 changes:
- Use ctz32() replace to for-loop
https://lore.kernel.org/qemu-devel/cover.1755333616.git.chao....@yeah.net/
patch v3 changes:
- Fix the get_log2() function:
https://lore.kernel.org/qemu-riscv/cover.1755287531.git.chao....@yeah.net/T/#t
- Add test for vlsseg8e32 instruction.
- Rebase on top of the latest master.
patch v2 changes:
- Split the TCG node emulation of the complex strided load/store operation into
two separate functions to simplify the implementation:
https://lore.kernel.org/qemu-riscv/20250312155547.289642-1-paolo.sav...@embecosm.com/
Best regards,
Chao
Chao Liu (2):
Generate strided vector loads/stores with tcg nodes.
tests/tcg/riscv64: Add test for vlsseg8e32 instruction
target/riscv/insn_trans/trans_rvv.c.inc | 317 ++++++++++++++++++----
tests/tcg/riscv64/Makefile.softmmu-target | 8 +-
tests/tcg/riscv64/test-vlsseg8e32.S | 107 ++++++++
3 files changed, 380 insertions(+), 52 deletions(-)
create mode 100644 tests/tcg/riscv64/test-vlsseg8e32.S
--
2.50.1
