On 11/17/2017 02:58 PM, Richard Sandiford wrote:
> This patch adds support for SVE gather loads. It uses the basically
> the same analysis code as the AVX gather support, but after that
> there are two major differences:
>
> - It uses new internal functions rather than target built-ins.
> The interface is:
>
> IFN_GATHER_LOAD (base, offsets, scale)
> IFN_MASK_GATHER_LOAD (base, offsets, scale, mask)
>
> which should be reasonably generic. One of the advantages of
> using internal functions is that other passes can understand what
> the functions do, but a more immediate advantage is that we can
> query the underlying target pattern to see which scales it supports.
>
> - It uses pattern recognition to convert the offset to the right width,
> if it was originally narrower than that. This avoids having to do
> a widening operation as part of the gather expansion itself.
>
> Tested on aarch64-linux-gnu (with and without SVE), x86_64-linux-gnu
> and powerpc64le-linux-gnu. OK to install?
>
> Richard
>
>
> 2017-11-17 Richard Sandiford <richard.sandif...@linaro.org>
> Alan Hayward <alan.hayw...@arm.com>
> David Sherwood <david.sherw...@arm.com>
>
> gcc/
> * doc/md.texi (gather_load@var{m}): Document.
> (mask_gather_load@var{m}): Likewise.
> * genopinit.c (main): Add supports_vec_gather_load and
> supports_vec_gather_load_cached to target_optabs.
> * optabs-tree.c (init_tree_optimization_optabs): Use
> ggc_cleared_alloc to allocate target_optabs.
> * optabs.def (gather_load_optab, mask_gather_laod_optab): New optabs.
> * internal-fn.def (GATHER_LOAD, MASK_GATHER_LOAD): New internal
> functions.
> * internal-fn.h (internal_load_fn_p): Declare.
> (internal_gather_scatter_fn_p): Likewise.
> (internal_fn_mask_index): Likewise.
> (internal_gather_scatter_fn_supported_p): Likewise.
> * internal-fn.c (gather_load_direct): New macro.
> (expand_gather_load_optab_fn): New function.
> (direct_gather_load_optab_supported_p): New macro.
> (direct_internal_fn_optab): New function.
> (internal_load_fn_p): Likewise.
> (internal_gather_scatter_fn_p): Likewise.
> (internal_fn_mask_index): Likewise.
> (internal_gather_scatter_fn_supported_p): Likewise.
> * optabs-query.c (supports_at_least_one_mode_p): New function.
> (supports_vec_gather_load_p): Likewise.
> * optabs-query.h (supports_vec_gather_load_p): Declare.
> * tree-vectorizer.h (gather_scatter_info): Add ifn, element_type
> and memory_type field.
> (NUM_PATTERNS): Bump to 15.
> * tree-vect-data-refs.c (vect_gather_scatter_fn_p): New function.
> (vect_describe_gather_scatter_call): Likewise.
> (vect_check_gather_scatter): Try using internal functions for
> gather loads. Recognize existing calls to a gather load function.
> (vect_analyze_data_refs): Consider using gather loads if
> supports_vec_gather_load_p.
> * tree-vect-patterns.c (vect_get_load_store_mask): New function.
> (vect_get_gather_scatter_offset_type): Likewise.
> (vect_convert_mask_for_vectype): Likewise.
> (vect_add_conversion_to_patterm): Likewise.
> (vect_try_gather_scatter_pattern): Likewise.
> (vect_recog_gather_scatter_pattern): New pattern recognizer.
> (vect_vect_recog_func_ptrs): Add it.
> * tree-vect-stmts.c (exist_non_indexing_operands_for_use_p): Use
> internal_fn_mask_index and internal_gather_scatter_fn_p.
> (check_load_store_masking): Take the gather_scatter_info as an
> argument and handle gather loads.
> (vect_get_gather_scatter_ops): New function.
> (vectorizable_call): Check internal_load_fn_p.
> (vectorizable_load): Likewise. Handle gather load internal
> functions.
> (vectorizable_store): Update call to check_load_store_masking.
> * config/aarch64/aarch64.md (UNSPEC_LD1_GATHER): New unspec.
> * config/aarch64/iterators.md (SVE_S, SVE_D): New mode iterators.
> * config/aarch64/predicates.md (aarch64_gather_scale_operand_w)
> (aarch64_gather_scale_operand_d): New predicates.
> * config/aarch64/aarch64-sve.md (gather_load<mode>): New expander.
> (mask_gather_load<mode>): New insns.
>
> gcc/testsuite/
> * gcc.target/aarch64/sve_gather_load_1.c: New test.
> * gcc.target/aarch64/sve_gather_load_2.c: Likewise.
> * gcc.target/aarch64/sve_gather_load_3.c: Likewise.
> * gcc.target/aarch64/sve_gather_load_4.c: Likewise.
> * gcc.target/aarch64/sve_gather_load_5.c: Likewise.
> * gcc.target/aarch64/sve_gather_load_6.c: Likewise.
> * gcc.target/aarch64/sve_gather_load_7.c: Likewise.
> * gcc.target/aarch64/sve_mask_gather_load_1.c: Likewise.
> * gcc.target/aarch64/sve_mask_gather_load_2.c: Likewise.
> * gcc.target/aarch64/sve_mask_gather_load_3.c: Likewise.
> * gcc.target/aarch64/sve_mask_gather_load_4.c: Likewise.
> * gcc.target/aarch64/sve_mask_gather_load_5.c: Likewise.
> * gcc.target/aarch64/sve_mask_gather_load_6.c: Likewise.
> * gcc.target/aarch64/sve_mask_gather_load_7.c: Likewise.
As with other patches that had a target component, I didn't review those
bits. The generic bits are OK for the trunk.
After doing all this work, any thoughts on if we'd be better off
modeling the avx bits as internal functions vs target builtins?
jeff
