mbox series

[v5,00/24] target/m68k: fpu improvements

Message ID 20250507211300.9735-1-richard.henderson@linaro.org
Headers show
Series target/m68k: fpu improvements | expand

Message

Richard Henderson May 7, 2025, 9:12 p.m. UTC 
v3: https://lore.kernel.org/qemu-devel/20240909172823.649837-1-richard.henderson@linaro.org/
v4: https://lore.kernel.org/qemu-devel/20250224171444.440135-1-richard.henderson@linaro.org/

Changes for v5:
  - Rebase from February.  :-/
  - Use ldl_be_p in m68k_fpu_gdb_set_reg (phil).

Patches needing review:
  03-target-m68k-Keep-FPSR-up-to-date.patch
  05-target-m68k-Update-FPSR-for-FMOVECR.patch
  09-target-m68k-Use-OS_UNSIZED-in-LEA-PEA-JMP.patch
  10-target-m68k-Move-pre-dec-post-inc-to-gen_lea_mode.patch
  11-target-m68k-Split-gen_ea_mode-for-load-store.patch
  21-target-m68k-Implement-packed-decimal-real-loads-a.patch


r~


Richard Henderson (24):
  target/m68k: Add FPSR exception bit defines
  target/m68k: Restore fp rounding mode on vm load
  target/m68k: Keep FPSR up-to-date
  target/m68k: Update FPSR.EXC
  target/m68k: Update FPSR for FMOVECR
  target/m68k: Introduce M68K_FEATURE_FPU_PACKED_DECIMAL
  target/m68k: Merge gen_ea into SRC_EA and DEST_EA
  target/m68k: Use g_assert_not_reached in gen_lea_mode and gen_ea_mode
  target/m68k: Use OS_UNSIZED in LEA, PEA, JMP
  target/m68k: Move pre-dec/post-inc to gen_lea_mode
  target/m68k: Split gen_ea_mode for load/store
  target/m68k: Remove env argument to gen_lea_indexed
  target/m68k: Remove env argument to gen_lea_mode
  target/m68k: Remove env argument to gen_load_mode
  target/m68k: Remove env argument to gen_store_mode
  target/m68k: Remove env argument to gen_ea_mode_fp
  target/m68k: Split gen_ea_mode_fp for load/store
  target/m68k: Move gen_addr_fault into gen_{load,store}_mode_fp
  target/m68k: Merge gen_load_fp, gen_load_mode_fp
  target/m68k: Merge gen_store_fp, gen_store_mode_fp
  target/m68k: Implement packed decimal real loads and stores
  tests/tcg/m68k: Add packed decimal tests
  target/m68k: Make vmstate variables static
  target/m68k: Implement FPIAR

 target/m68k/cpu.h                |   27 +-
 target/m68k/helper.h             |    5 +-
 target/m68k/cpu.c                |   38 +-
 target/m68k/fpu_helper.c         |  330 +-
 target/m68k/gen-floatx80-pow10.c |   33 +
 target/m68k/helper.c             |   18 +-
 target/m68k/translate.c          |  789 +++--
 tests/tcg/m68k/packeddecimal-1.c |   41 +
 tests/tcg/m68k/packeddecimal-2.c |   46 +
 target/m68k/floatx80-pow10.c.inc | 4935 ++++++++++++++++++++++++++++++
 tests/tcg/m68k/Makefile.target   |    4 +-
 11 files changed, 5746 insertions(+), 520 deletions(-)
 create mode 100644 target/m68k/gen-floatx80-pow10.c
 create mode 100644 tests/tcg/m68k/packeddecimal-1.c
 create mode 100644 tests/tcg/m68k/packeddecimal-2.c
 create mode 100644 target/m68k/floatx80-pow10.c.inc

Comments

Helge Deller May 7, 2025, 9:28 p.m. UTC | #1 
On 5/7/25 23:12, Richard Henderson wrote:
> v3: https://lore.kernel.org/qemu-devel/20240909172823.649837-1-richard.henderson@linaro.org/
> v4: https://lore.kernel.org/qemu-devel/20250224171444.440135-1-richard.henderson@linaro.org/
> 
> Changes for v5:
>    - Rebase from February.  :-/
>    - Use ldl_be_p in m68k_fpu_gdb_set_reg (phil).
> 
> Patches needing review:
>    03-target-m68k-Keep-FPSR-up-to-date.patch
>    05-target-m68k-Update-FPSR-for-FMOVECR.patch
>    09-target-m68k-Use-OS_UNSIZED-in-LEA-PEA-JMP.patch
>    10-target-m68k-Move-pre-dec-post-inc-to-gen_lea_mode.patch
>    11-target-m68k-Split-gen_ea_mode-for-load-store.patch
>    21-target-m68k-Implement-packed-decimal-real-loads-a.patch

Hi Richard,
I noticed that the FPU emulation on hppa has some issues too.
Any suggestion, how one can test the runtime FP emulation (inside the VM)?

Helge