Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix reinterpret performance #28707

Merged
merged 1 commit into from
Aug 17, 2018
Merged

Fix reinterpret performance #28707

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
34 changes: 13 additions & 21 deletions base/reinterpretarray.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -104,6 +104,8 @@ end
_getindex_ra(a, inds[1], tail(inds))
end

@inline _memcpy!(dst, src, n) = ccall(:memcpy, Cvoid, (Ptr{UInt8}, Ptr{UInt8}, Csize_t), dst, src, n)

@inline @propagate_inbounds function _getindex_ra(a::ReinterpretArray{T,N,S}, i1::Int, tailinds::TT) where {T,N,S,TT}
# Make sure to match the scalar reinterpret if that is applicable
if sizeof(T) == sizeof(S) && (fieldcount(T) + fieldcount(S)) == 0
Expand All @@ -123,11 +125,9 @@ end
# once it knows the data layout
while nbytes_copied < sizeof(T)
s[] = a.parent[ind_start + i, tailinds...]
while nbytes_copied < sizeof(T) && sidx < sizeof(S)
unsafe_store!(tptr, unsafe_load(sptr, sidx + 1), nbytes_copied + 1)
sidx += 1
nbytes_copied += 1
end
nb = min(sizeof(S) - sidx, sizeof(T)-nbytes_copied)
_memcpy!(tptr + nbytes_copied, sptr + sidx, nb)
nbytes_copied += nb
sidx = 0
i += 1
end
Expand Down Expand Up @@ -173,34 +173,26 @@ end
# element from the original array and overwrite the relevant parts
if sidx != 0
s[] = a.parent[ind_start + i, tailinds...]
while nbytes_copied < sizeof(T) && sidx < sizeof(S)
unsafe_store!(sptr, unsafe_load(tptr, nbytes_copied + 1), sidx + 1)
sidx += 1
nbytes_copied += 1
end
nb = min(sizeof(S) - sidx, sizeof(T))
_memcpy!(sptr + sidx, tptr, nb)
nbytes_copied += nb
a.parent[ind_start + i, tailinds...] = s[]
i += 1
sidx = 0
end
# Deal with the main body of elements
while nbytes_copied < sizeof(T) && (sizeof(T) - nbytes_copied) > sizeof(S)
while nbytes_copied < sizeof(T) && sidx < sizeof(S)
unsafe_store!(sptr, unsafe_load(tptr, nbytes_copied + 1), sidx + 1)
sidx += 1
nbytes_copied += 1
end
nb = min(sizeof(S), sizeof(T) - nbytes_copied)
_memcpy!(sptr, tptr + nbytes_copied, nb)
nbytes_copied += nb
a.parent[ind_start + i, tailinds...] = s[]
i += 1
sidx = 0
end
# Deal with trailing partial elements
if nbytes_copied < sizeof(T)
s[] = a.parent[ind_start + i, tailinds...]
while nbytes_copied < sizeof(T) && sidx < sizeof(S)
unsafe_store!(sptr, unsafe_load(tptr, nbytes_copied + 1), sidx + 1)
sidx += 1
nbytes_copied += 1
end
nb = min(sizeof(S), sizeof(T) - nbytes_copied)
_memcpy!(sptr, tptr + nbytes_copied, nb)
a.parent[ind_start + i, tailinds...] = s[]
end
end
Expand Down
2 changes: 1 addition & 1 deletion base/reshapedarray.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -222,7 +222,7 @@ end
I = ind2sub_rs(axes(A.parent), A.mi, i)
_unsafe_getindex_rs(parent(A), I)
end
_unsafe_getindex_rs(A, i::Integer) = (@inbounds ret = A[i]; ret)
@inline _unsafe_getindex_rs(A, i::Integer) = (@inbounds ret = A[i]; ret)
@inline _unsafe_getindex_rs(A, I) = (@inbounds ret = A[I...]; ret)

@inline function setindex!(A::ReshapedArrayLF, val, index::Int)
Expand Down
14 changes: 14 additions & 0 deletions src/ccall.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1820,6 +1820,20 @@ static jl_cgval_t emit_ccall(jl_codectx_t &ctx, jl_value_t **args, size_t nargs)
JL_GC_POP();
return mark_or_box_ccall_result(ctx, strp, retboxed, rt, unionall, static_rt);
}
else if (is_libjulia_func(memcpy)) {
const jl_cgval_t &dst = argv[0];
const jl_cgval_t &src = argv[1];
const jl_cgval_t &n = argv[2];
ctx.builder.CreateMemCpy(
ctx.builder.CreateIntToPtr(
emit_unbox(ctx, T_size, dst, (jl_value_t*)jl_voidpointer_type), T_pint8),
ctx.builder.CreateIntToPtr(
emit_unbox(ctx, T_size, src, (jl_value_t*)jl_voidpointer_type), T_pint8),
emit_unbox(ctx, T_size, n, (jl_value_t*)jl_ulong_type), 1,
false);
JL_GC_POP();
return ghostValue(jl_void_type);
}

jl_cgval_t retval = sig.emit_a_ccall(
ctx,
Expand Down