The change also affects vectorizer in avx case which could be seen for gcc.dg/tree-ssa/loop-19.c test.
After the change report says loop-19_bad.c:16: note: === vect_analyze_data_refs_alignment === loop-19_bad.c:16: note: vect_compute_data_ref_alignment: loop-19_bad.c:16: note: can't force alignment of ref: a[j_9] loop-19_bad.c:16: note: vect_compute_data_ref_alignment: loop-19_bad.c:16: note: can't force alignment of ref: c[j_9] AFAICS first condition in ix86_data_alignment was true before the change so 256 was a return value. Do we need to tweak this test also? Thanks, Igor > Hi! > > This PR is about DATA_ALIGNMENT macro increasing alignment of some decls for > optimization purposes beyond ABI mandated levels. It is fine to emit the > vars aligned as much as we want for optimization purposes, but if we can't be > sure that references to that decl bind to the definition we increased the > alignment on (e.g. common variables, or -fpic code without hidden visibility, > weak vars etc.), we can't assume that alignment. > As DECL_ALIGN is used for both the alignment emitted for the definitions and > alignment assumed on code referring to it, this patch increases DECL_ALIGN > only on decls where decl_binds_to_current_def_p is true, and otherwise the > optimization part on top of that emits only when aligning definition. > On x86_64, DATA_ALIGNMENT macro was partly an optimization, partly ABI > mandated alignment increase, so I've introduced a new macro, > DATA_ABI_ALIGNMENT, which is the ABI mandated increase only (on x86-64 I > think the only one is that arrays with size 16 bytes or more (and VLAs, but > that is not handled by DATA*ALIGNMENT) are at least 16 byte aligned). > > Bootstrapped/regtested on x86_64-linux and i686-linux. No idea about other > targets, I've kept them all using DATA_ALIGNMENT, which is considered > optimization increase only now, if there is some ABI mandated alignment > increase on other targets, that should be done in DATA_ABI_ALIGNMENT as well > as DATA_ALIGNMENT. The patch causes some vectorization regressions (tweaked > in the testsuite), especially for common vars where we used to align say > common arrays to 256 bits rather than the ABI mandated 128 bits, or for -fpic > code, but I'm afraid we need to live with that, if you compile another file > with say icc or some other compiler which doesn't increase alignment beyond > ABI mandated level and that other file defines the var say as non-common, we > have wrong-code. > > 2013-06-07 Jakub Jelinek <ja...@redhat.com> > > PR target/56564 > * varasm.c (align_variable): Don't use DATA_ALIGNMENT or > CONSTANT_ALIGNMENT if !decl_binds_to_current_def_p (decl). > Use DATA_ABI_ALIGNMENT for that case instead if defined. > (get_variable_align): New function. > (get_variable_section, emit_bss, emit_common, > assemble_variable_contents, place_block_symbol): Use > get_variable_align instead of DECL_ALIGN. > (assemble_noswitch_variable): Add align argument, use it > instead of DECL_ALIGN. > (assemble_variable): Adjust caller. Use get_variable_align > instead of DECL_ALIGN. > * config/i386/i386.h (DATA_ALIGNMENT): Adjust x86_data_alignment > caller. > (DATA_ABI_ALIGNMENT): Define. > * config/i386/i386-protos.h (x86_data_alignment): Adjust prototype. > * config/i386/i386.c (x86_data_alignment): Add opt argument. If > opt is false, only return the psABI mandated alignment increase. > * doc/tm.texi.in (DATA_ABI_ALIGNMENT): Document. > * doc/tm.texi: Regenerated. > > * gcc.target/i386/pr56564-1.c: New test. > * gcc.target/i386/pr56564-2.c: New test. > * gcc.target/i386/pr56564-3.c: New test. > * gcc.target/i386/pr56564-4.c: New test. > * gcc.target/i386/avx256-unaligned-load-4.c: Add -fno-common. > * gcc.target/i386/avx256-unaligned-store-1.c: Likewise. > * gcc.target/i386/avx256-unaligned-store-3.c: Likewise. > * gcc.target/i386/avx256-unaligned-store-4.c: Likewise. > * gcc.target/i386/vect-sizes-1.c: Likewise. > * gcc.target/i386/memcpy-1.c: Likewise. > * gcc.dg/vect/costmodel/i386/costmodel-vect-31.c (tmp): Initialize. > * gcc.dg/vect/costmodel/x86_64/costmodel-vect-31.c (tmp): Likewise. >
