Message ID | 20230626014940.36902-1-juzhe.zhong@rivai.ai |
---|---|
State | New |
Headers | show |
Series | GIMPLE_FOLD: Fix gimple fold for LEN_MASK_{LOAD,STORE} | expand |
On Mon, 26 Jun 2023, juzhe.zhong@rivai.ai wrote: > From: Ju-Zhe Zhong <juzhe.zhong@rivai.ai> > > Hi, previous I made a mistake on GIMPLE_FOLD of LEN_MASK_{LOAD,STORE}. > > We should fold LEN_MASK_{LOAD,STORE} (bias+len) == vf (nunits instead of bytesize) && mask = all trues mask > > into: > MEM_REF [...]. > > This patch added testcase to test gimple fold of LEN_MASK_{LOAD,STORE}. > > Ok for trunk ? Err, but then the IFN_LEN_LOAD case is wrong as well - we surely do not want differing behavior here. So please fix them both? > gcc/ChangeLog: > > * gimple-fold.cc (gimple_fold_partial_load_store_mem_ref): Fix LEN_MASK_{LOAD,STORE}. > > gcc/testsuite/ChangeLog: > > * gcc.target/riscv/rvv/autovec/partial/gimple_fold-1.c: New test. > > --- > gcc/gimple-fold.cc | 16 ++++--- > .../riscv/rvv/autovec/partial/gimple_fold-1.c | 43 +++++++++++++++++++ > 2 files changed, 54 insertions(+), 5 deletions(-) > create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/partial/gimple_fold-1.c > > diff --git a/gcc/gimple-fold.cc b/gcc/gimple-fold.cc > index 3d46b76edeb..94e79c32218 100644 > --- a/gcc/gimple-fold.cc > +++ b/gcc/gimple-fold.cc > @@ -5397,19 +5397,25 @@ gimple_fold_partial_load_store_mem_ref (gcall *call, tree vectype, bool mask_p) > unsigned int nargs = gimple_call_num_args (call); > tree bias = gimple_call_arg (call, nargs - 1); > gcc_assert (TREE_CODE (bias) == INTEGER_CST); > - if (maybe_ne (wi::to_poly_widest (basic_len) - wi::to_widest (bias), > - GET_MODE_SIZE (TYPE_MODE (vectype)))) > - return NULL_TREE; > > - /* For LEN_MASK_{LOAD,STORE}, we should also check whether > - the mask is all ones mask. */ > internal_fn ifn = gimple_call_internal_fn (call); > if (ifn == IFN_LEN_MASK_LOAD || ifn == IFN_LEN_MASK_STORE) > { > + /* We fold LEN_MASK_{LOAD,STORE} into MEM_REF when (len + bias) == VF > + and mask = all trues mask. */ > + if (maybe_ne (wi::to_poly_widest (basic_len) + wi::to_widest (bias), > + GET_MODE_NUNITS (TYPE_MODE (vectype)))) > + return NULL_TREE; > tree mask = gimple_call_arg (call, internal_fn_mask_index (ifn)); > if (!integer_all_onesp (mask)) > return NULL_TREE; > } > + else > + { > + if (maybe_ne (wi::to_poly_widest (basic_len) - wi::to_widest (bias), > + GET_MODE_SIZE (TYPE_MODE (vectype)))) > + return NULL_TREE; > + } > } > > unsigned HOST_WIDE_INT align = tree_to_uhwi (alias_align); > diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/partial/gimple_fold-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/partial/gimple_fold-1.c > new file mode 100644 > index 00000000000..23407a2d3f4 > --- /dev/null > +++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/partial/gimple_fold-1.c > @@ -0,0 +1,43 @@ > +/* { dg-do compile } */ > +/* { dg-options "-march=rv32gcv -mabi=ilp32d --param riscv-autovec-preference=fixed-vlmax --param riscv-autovec-lmul=m8 -O3 -fdump-tree-optimized-details" } */ > + > +#include <stdint-gcc.h> > + > +#define SZ 255 > + > +#define DEF(TYPE) void fn_##TYPE (TYPE *__restrict a); > + > +#define RUN(TYPE) \ > + TYPE a##TYPE[SZ]; \ > + for (int i = 0; i < SZ; i++) \ > + { \ > + a##TYPE[i] = 127; \ > + } \ > + fn_##TYPE (a##TYPE); > + > +#define RUN_ALL() \ > + RUN (int8_t) \ > + RUN (int16_t) \ > + RUN (int32_t) \ > + RUN (int64_t) \ > + RUN (uint8_t) \ > + RUN (uint16_t) \ > + RUN (uint32_t) \ > + RUN (uint64_t) > + > +DEF (int8_t) > +DEF (int16_t) > +DEF (int32_t) > +DEF (int64_t) > +DEF (uint8_t) > +DEF (uint16_t) > +DEF (uint32_t) > +DEF (uint64_t) > + > +int > +main () > +{ > + RUN_ALL () > +} > + > +/* { dg-final { scan-tree-dump-times "\.LEN_MASK_STORE" 6 "optimized" } } */ >
diff --git a/gcc/gimple-fold.cc b/gcc/gimple-fold.cc index 3d46b76edeb..94e79c32218 100644 --- a/gcc/gimple-fold.cc +++ b/gcc/gimple-fold.cc @@ -5397,19 +5397,25 @@ gimple_fold_partial_load_store_mem_ref (gcall *call, tree vectype, bool mask_p) unsigned int nargs = gimple_call_num_args (call); tree bias = gimple_call_arg (call, nargs - 1); gcc_assert (TREE_CODE (bias) == INTEGER_CST); - if (maybe_ne (wi::to_poly_widest (basic_len) - wi::to_widest (bias), - GET_MODE_SIZE (TYPE_MODE (vectype)))) - return NULL_TREE; - /* For LEN_MASK_{LOAD,STORE}, we should also check whether - the mask is all ones mask. */ internal_fn ifn = gimple_call_internal_fn (call); if (ifn == IFN_LEN_MASK_LOAD || ifn == IFN_LEN_MASK_STORE) { + /* We fold LEN_MASK_{LOAD,STORE} into MEM_REF when (len + bias) == VF + and mask = all trues mask. */ + if (maybe_ne (wi::to_poly_widest (basic_len) + wi::to_widest (bias), + GET_MODE_NUNITS (TYPE_MODE (vectype)))) + return NULL_TREE; tree mask = gimple_call_arg (call, internal_fn_mask_index (ifn)); if (!integer_all_onesp (mask)) return NULL_TREE; } + else + { + if (maybe_ne (wi::to_poly_widest (basic_len) - wi::to_widest (bias), + GET_MODE_SIZE (TYPE_MODE (vectype)))) + return NULL_TREE; + } } unsigned HOST_WIDE_INT align = tree_to_uhwi (alias_align); diff --git a/gcc/testsuite/gcc.target/riscv/rvv/autovec/partial/gimple_fold-1.c b/gcc/testsuite/gcc.target/riscv/rvv/autovec/partial/gimple_fold-1.c new file mode 100644 index 00000000000..23407a2d3f4 --- /dev/null +++ b/gcc/testsuite/gcc.target/riscv/rvv/autovec/partial/gimple_fold-1.c @@ -0,0 +1,43 @@ +/* { dg-do compile } */ +/* { dg-options "-march=rv32gcv -mabi=ilp32d --param riscv-autovec-preference=fixed-vlmax --param riscv-autovec-lmul=m8 -O3 -fdump-tree-optimized-details" } */ + +#include <stdint-gcc.h> + +#define SZ 255 + +#define DEF(TYPE) void fn_##TYPE (TYPE *__restrict a); + +#define RUN(TYPE) \ + TYPE a##TYPE[SZ]; \ + for (int i = 0; i < SZ; i++) \ + { \ + a##TYPE[i] = 127; \ + } \ + fn_##TYPE (a##TYPE); + +#define RUN_ALL() \ + RUN (int8_t) \ + RUN (int16_t) \ + RUN (int32_t) \ + RUN (int64_t) \ + RUN (uint8_t) \ + RUN (uint16_t) \ + RUN (uint32_t) \ + RUN (uint64_t) + +DEF (int8_t) +DEF (int16_t) +DEF (int32_t) +DEF (int64_t) +DEF (uint8_t) +DEF (uint16_t) +DEF (uint32_t) +DEF (uint64_t) + +int +main () +{ + RUN_ALL () +} + +/* { dg-final { scan-tree-dump-times "\.LEN_MASK_STORE" 6 "optimized" } } */
From: Ju-Zhe Zhong <juzhe.zhong@rivai.ai> Hi, previous I made a mistake on GIMPLE_FOLD of LEN_MASK_{LOAD,STORE}. We should fold LEN_MASK_{LOAD,STORE} (bias+len) == vf (nunits instead of bytesize) && mask = all trues mask into: MEM_REF [...]. This patch added testcase to test gimple fold of LEN_MASK_{LOAD,STORE}. Ok for trunk ? gcc/ChangeLog: * gimple-fold.cc (gimple_fold_partial_load_store_mem_ref): Fix LEN_MASK_{LOAD,STORE}. gcc/testsuite/ChangeLog: * gcc.target/riscv/rvv/autovec/partial/gimple_fold-1.c: New test. --- gcc/gimple-fold.cc | 16 ++++--- .../riscv/rvv/autovec/partial/gimple_fold-1.c | 43 +++++++++++++++++++ 2 files changed, 54 insertions(+), 5 deletions(-) create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/autovec/partial/gimple_fold-1.c