-
-
Notifications
You must be signed in to change notification settings - Fork 5.5k
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.
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
Avoid generic call in most cases for getproperty #50523
Conversation
src/codegen.cpp
Outdated
@@ -3839,9 +3851,17 @@ static bool emit_builtin_call(jl_codectx_t &ctx, jl_cgval_t *ret, jl_value_t *f, | |||
*ret = emit_getfield_knownidx(ctx, obj, 0, utt, order); | |||
return true; | |||
} | |||
else { | |||
Value *index = ctx.builder.CreateCall(prepare_call(jlfieldindex_func), | |||
{emit_typeof(ctx, obj, false, false), boxed(ctx, fld), ConstantInt::get(getInt32Ty(ctx.builder.getContext()), 0)}); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
boxed(ctx, fld)
Does LLVM then unbox the fld value, before this actually reaches real code? This PR is meant to remove an allocation here right?
Can you also leave a comment explaining that?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
boxed
checks if the object needs a box before being passed into julia, if it was an Int or something without a type tag. In this case fld is known to be a Symbol
so it's a no op. The boxing that used to happen here is because we used the generic jlcall
convention.
I can add a comment if you think it's useful though.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yeah I think anything where the reason for the core isn't obvious or straightforward should have a comment. 👍 Thanks
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM pending my requests for comments and a test. Thanks for fixing this, @gbaraldi!! We really appreciate it
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It seems odd to delete most of the optimization for single-argument functions while adding this
Line 2353 in dcca46b
emit_getfield_unknownidx , we just weren't hitting it before, the generic one lowers to the same thing as before bar not doing the literal value for the symbol compare, but I'm not sure how much we are saving there?I can just add it back if we want. |
Agree with @vtjnash. This will be slower for the 1-arg case. I can see this being useful for types with interesting homogeneities, where unknownindex actually does something interesting. However, for the general case, I would much prefer a variant of |
I will readd the special case we had, I guess saving the field_index call (which doesn't need boxing often anyway) doesn't cost us anything. |
Backported PRs: - [x] #50411 <!-- Fix weird dispatch of * with zero arguments --> - [x] #50202 <!-- Remove dynamic dispatch from _wait/wait2 --> - [x] #50064 <!-- Fix numbered prompt with input only with comment --> - [x] #50026 <!-- Store heapsnapshot files in tempdir() instead of current directory --> - [x] #50402 <!-- Add CPU feature helper function --> - [x] #50387 <!-- update newpages pointer after actually sweeping pages --> - [x] #50424 <!-- avoid potential type-instability in _replace_(str, ...) --> - [x] #50444 <!-- Optimize getfield lowering to avoid boxing in some cases --> - [x] #50474 <!-- docs: Fix a `!!! note` which was miscapitalized --> - [x] #50466 <!-- relax assertion involving pg->nold to reflect that it may be a bit in… --> - [x] #50490 <!-- Fix compat annotation for italic printstyled --> - [x] #50488 <!-- fix typo in `Base.isassigned` with `Tridiagonal` --> - [x] #50476 <!-- Profile: Add specifying dir for `take_heap_snapshot` and handling if current dir is unwritable --> - [x] #50461 <!-- fix typo in the --gcthreads argument description --> - [x] #50528 <!-- ssair: Correctly handle stmt insertion at end of basic block --> - [x] #50533 <!-- ensure internal_obj_base_ptr checks whether objects past freelist pointer are in freelist --> - [x] #49322 <!-- improve cat design / performance --> - [x] #50540 <!-- gc: remove over-eager assertion --> - [x] #50542 <!-- gf: remove unnecessary assert cycle==depth --> - [x] #50559 <!-- Expand kwcall lowering positional default check to vararg --> - [x] #50058 <!-- Add unwrapping mechanism for triangular mul and solves --> - [x] #50551 <!-- typeintersect: also record chained `innervars` --> - [x] #50552 <!-- read(io, Char): fix read with too many leading ones --> - [x] #50541 <!-- precompile: ensure globals are not accidentally created where disallowed --> - [x] #50576 <!-- use atomic compare exchange when setting the GC mark-bit --> - [x] #50578 <!-- gf: make method overwrite/delete an error during precompile --> - [x] #50516 <!-- Fix visibility of assert on GCC12/13 --> - [x] #50597 <!-- Fix memory corruption if task is launched inside finalizer --> - [x] #50591 <!-- build: fix various makefile bugs --> - [x] #50599 <!-- faster invalid object lookup in conservative gc --> - [x] #50634 <!-- 🤖 [master] Bump the SparseArrays stdlib from b4b0e72 to 99c99b4 --> - [x] #50639 <!-- Backport LLVM patches to fix various issues. --> - [x] #50546 <!-- Revert storage of method instance in LineInfoNode --> - [x] #50631 <!-- Shift DCE pass to optimize imaging mode code better --> - [x] #50525 <!-- only check that values are finite in `generic_lufact` when `check=true` --> - [x] #50587 <!-- isassigned for ranges with BigInt indices --> - [x] #50144 <!-- Page based heap size heuristics --> Need manual backport: - [ ] #50595 <!-- Rename ENV variable `JULIA_USE_NEW_PARSER` -> `JULIA_USE_FLISP_PARSER` --> Non-merged PRs with backport label: - [ ] #50637 <!-- Remove SparseArrays legacy code --> - [ ] #50618 <!-- inference: continue const-prop' when concrete-eval returns non-inlineable --> - [ ] #50598 <!-- only limit types in stack traces in the REPL --> - [ ] #50594 <!-- Disallow non-index Integer types in isassigned --> - [ ] #50568 <!-- `Array(::AbstractRange)` should return an `Array` --> - [ ] #50523 <!-- Avoid generic call in most cases for getproperty --> - [ ] #50172 <!-- print feature flags used for matching pkgimage -->
Can we merge this @vtjnash |
Will/can this be backported to 1.9.X as well? |
(cherry picked from commit 462ac49)
Backported PRs: - [x] #50637 <!-- Remove SparseArrays legacy code --> - [x] #50665 <!-- print `@time` msg into print buffer --> - [x] #50523 <!-- Avoid generic call in most cases for getproperty --> - [x] #50635 <!-- `versioninfo()`: include build info and unofficial warning --> - [x] #50670 <!-- Make reinterpret specialize fully. --> - [x] #50666 <!-- include `--pkgimage=no` caches for stdlibs --> - [x] #50765 - [x] #50764 - [x] #50768 - [x] #50767 - [x] #50618 <!-- inference: continue const-prop' when concrete-eval returns non-inlineable --> - [x] #50689 <!-- Attach `tanpi` docstring to method --> - [x] #50671 <!-- Fix rdiv of complex lhs by real factorizations --> - [x] #50598 <!-- only limit types in stack traces in the REPL --> - [x] #50766 <!-- Don't partition alwaysinline functions --> - [x] #50771 <!-- re-allow non-string values in ENV `get!` --> - [x] #50682 <!-- Add fallback if we have make a weird GC decision. --> - [x] #50781 <!-- fix `bit_map!` with aliasing --> - [x] #50172 <!-- print feature flags used for matching pkgimage --> - [x] #50844 <!-- Bump OpenBLAS binaries to use the new GEMM multithreading threshold --> - [x] #50826 <!-- Update dependency builds --> - [x] #50845 <!-- fix #50438, use default pool for at-threads --> - [x] #50568 <!-- `Array(::AbstractRange)` should return an `Array` --> - [x] #50655 <!-- fix hashing regression. --> - [x] #50779 <!-- Minor refactor to image generation --> - [x] #50791 <!-- Make symbols internal in jl_create_native, and only externalize them when partitioning --> - [x] #50724 <!-- Merge opaque closure modules with the rest of the workqueue --> - [x] #50738 <!-- Add alignment to constant globals --> - [x] #50871 <!-- macOS: Don't inspect dead threadtls during exception handling. --> Need manual backport: Contains multiple commits, manual intervention needed: Non-merged PRs with backport label: - [ ] #50850 <!-- Remove weird Rational dispatch and add pi functions to list --> - [ ] #50823 <!-- Make ranges more robust with unsigned indexes. --> - [ ] #50809 <!-- Limit type-printing in MethodError --> - [ ] #50663 <!-- Fix Expr(:loopinfo) codegen --> - [ ] #50594 <!-- Disallow non-index Integer types in isassigned --> - [ ] #50385 <!-- Precompile pidlocks: add to NEWS and docs --> - [ ] #49805 <!-- Limit TimeType subtraction to AbstractDateTime -->
Backported PRs: - [x] #47782 <!-- Generalize Bool parse method to AbstractString --> - [x] #48634 <!-- Remove unused "deps" mechanism in internal sorting keywords [NFC] --> - [x] #49931 <!-- Lock finalizers' lists at exit --> - [x] #50064 <!-- Fix numbered prompt with input only with comment --> - [x] #50474 <!-- docs: Fix a `!!! note` which was miscapitalized --> - [x] #50516 <!-- Fix visibility of assert on GCC12/13 --> - [x] #50635 <!-- `versioninfo()`: include build info and unofficial warning --> - [x] #49915 <!-- Revert "Remove number / vector (#44358)" --> - [x] #50781 <!-- fix `bit_map!` with aliasing --> - [x] #50845 <!-- fix #50438, use default pool for at-threads --> - [x] #49031 <!-- Update inference.md --> - [x] #50289 <!-- Initialize prev_nold and nold in gc_reset_page --> - [x] #50559 <!-- Expand kwcall lowering positional default check to vararg --> - [x] #49582 <!-- Update HISTORY.md for `DelimitedFiles` --> - [x] #50341 <!-- invokelatest docs should say not exported before 1.9 --> - [x] #50525 <!-- only check that values are finite in `generic_lufact` when `check=true` --> - [x] #50444 <!-- Optimize getfield lowering to avoid boxing in some cases --> - [x] #50523 <!-- Avoid generic call in most cases for getproperty --> - [x] #50860 <!-- Add `Base.get_extension` to docs/API --> - [x] #50164 <!-- codegen: handle dead code with unsafe_store of FCA pointers --> - [x] #50568 <!-- `Array(::AbstractRange)` should return an `Array` --> - [x] #50871 <!-- macOS: Don't inspect dead threadtls during exception handling. --> Need manual backport: - [ ] #48542 <!-- Add docs on task-specific buffering using multithreading --> - [ ] #50591 <!-- build: fix various makefile bugs --> Non-merged PRs with backport label: - [ ] #50842 <!-- Avoid race conditions with recursive rm --> - [ ] #50823 <!-- Make ranges more robust with unsigned indexes. --> - [ ] #50663 <!-- Fix Expr(:loopinfo) codegen --> - [ ] #49716 <!-- Update varinfo() docstring signature --> - [ ] #49713 <!-- prevent REPL from erroring in numbered mode in some situations --> - [ ] #49573 <!-- Implement jl_cpu_pause on PPC64 --> - [ ] #48726 <!-- fix macro expansion of property destructuring --> - [ ] #48642 <!-- Use gc alloc instead of alloc typed in lowering --> - [ ] #48183 <!-- Don't use pkgimage for package if any includes fall in tracked path for coverage or alloc tracking --> - [ ] #48050 <!-- improve `--heap-size-hint` arg handling --> - [ ] #47615 <!-- Allow threadsafe access to buffer of type inference profiling trees -->
More generic than #50444.
Should we keep the boundscheck on the getfield since we got the index from the runtime itself?