summaryrefslogtreecommitdiff
path: root/yjit/src/backend/ir.rs
AgeCommit message (Collapse)Author
2025-03-06YJIT: Rename get_temp_regs2() back to get_temp_regs() (#12866)Takashi Kokubun
Notes: Merged-By: maximecb <[email protected]>
2024-12-09YJIT: Spill/load argument registers to reuse blocks (#12287)Takashi Kokubun
* YJIT: Spill/load argument registers to reuse blocks * Mention the immediate function name * Explain the context behind spill/load operations Notes: Merged-By: k0kubun <[email protected]>
2024-08-27YJIT: Pass method arguments using registers (#11280)Takashi Kokubun
* YJIT: Pass method arguments using registers * s/at_current_insn/at_compile_target/ * Implement register shuffle Notes: Merged-By: k0kubun <[email protected]>
2024-07-18YJIT: Allow dev_nodebug to disasm release-mode code (#11198)Takashi Kokubun
* YJIT: Allow dev_nodebug to disasm release-mode code * Revert "YJIT: Squash canary before falling back" This reverts commit f05ad373d84909da7541bd6d6ace38b48eaf24a1. The stray canary issue should have been solved by def7023ee4a3fc6eeba9d3a34c31a5bcff315fac, alleviating this codegen accommodation. * s/runtime_assertions/runtime_checks/ --------- Co-authored-by: Alan Wu <[email protected]> Notes: Merged-By: k0kubun <[email protected]>
2024-07-15YJIT: Local variable register allocation (#11157)Takashi Kokubun
* YJIT: Local variable register allocation * locals are not stack temps * Rename RegTemps to RegMappings * Rename RegMapping to RegOpnd * Rename local_size to num_locals * s/stack value/operand/ * Rename spill_temps() to spill_regs() * Clarify when num_locals becomes None * Mention that InsnOut uses different registers * Rename get_reg_mapping to get_reg_opnd * Resurrect --yjit-temp-regs capability * Use MAX_CTX_TEMPS and MAX_CTX_LOCALS
2024-07-08YJIT: `dump-disasm`: Print comments and bytes in release buildsAlan Wu
This change implements a fallback mode for the `--yjit-dump-disasm` development command-line option to make it usable in release builds. Previously, using the option with release builds of YJIT yielded only a warning asking the user to build with `--enable-yjit=dev`. While builds that use the `disasm` feature still give the best output, just having the comments is useful enough for many kinds of debugging. Having it usable in release builds is nice for new hackers, too, since this allows for tinkering without having to learn how to build YJIT in development mode. Sample output on A64: ``` # regenerate_branch # Insn: 0001 opt_send_without_block (stack_size: 1) # guard known object with singleton class 0x11f7e0034: 4b 00 00 58 03 00 00 14 08 ce 9c 04 01 00 00 0x11f7e0043: 00 3f 00 0b eb 81 06 01 54 1f 20 03 d5 # RUBY_VM_CHECK_INTS(ec) 0x11f7e0050: 8b 02 42 b8 cb 07 01 35 # stack overflow check 0x11f7e0058: ab 62 02 91 7f 02 0b eb 69 07 01 54 # save PC to CFP 0x11f7e0064: 0b 3b 9a d2 2b 2f a0 f2 0b 00 cc f2 6b 02 00 0x11f7e0073: f8 ab 82 00 91 ``` To ensure this feature doesn't incur too much cost when running without the `--yjit-dump-disasm` option, I checked that there is no significant impact to compile time and memory usage with the `compile_time_ns` and `yjit_alloc_size` entry in `RubyVM::YJIT.runtime_stats`. For each sample, I ran 3 iterations of the `lobsters` YJIT benchmark. The statistics summary and done with the `summary` function in R. Compile time, sample size of 60, lower is better: ``` Before After Min. :2.054e+09 Min. :2.028e+09 1st Qu.:2.069e+09 1st Qu.:2.044e+09 Median :2.081e+09 Median :2.060e+09 Mean :2.089e+09 Mean :2.066e+09 3rd Qu.:2.109e+09 3rd Qu.:2.085e+09 Max. :2.146e+09 Max. :2.144e+09 ``` Allocation size, sample size of 20, lower is better: ``` Before After Min. :21804742 Min. :21794082 1st Qu.:21826682 1st Qu.:21816282 Median :21844042 Median :21826814 Mean :21960664 Mean :22026291 3rd Qu.:21861228 3rd Qu.:22040439 Max. :22587426 Max. :22930614 ``` The `yjit_alloc_size` samples are noisy, but since the average increased by only 0.3%, and the median is lower, I feel safe saying that there is no significant change.
2024-04-17YJIT: A64: Use CBZ/CBNZ to check for zeroAlan Wu
* YJIT: A64: Add CBZ and CBNZ encoding functions * YJIT: A64: Use CBZ/CBNZ to check for zero Instead of emitting `cmp x0, #0` plus `b.z #target`, A64 offers Compare and Branch on Zero for us to just do `cbz x0, #target`. This commit utilizes that and the related CBNZ instruction when appropriate. We check for zero most commonly in interrupt checks: ```diff # Insn: 0003 leave (stack_size: 1) # RUBY_VM_CHECK_INTS(ec) ldur w11, [x20, #0x20] -tst w11, w11 -b.ne #0x109002164 +cbnz w11, #0x1049021d0 ``` * fix copy paste error Co-authored-by: Randy Stauner <[email protected]> --------- Co-authored-by: Randy Stauner <[email protected]>
2024-02-21YJIT: Assert Opnd::Stack's SP expectation (#10061)Takashi Kokubun
2024-02-21YJIT: Allow non-leaf calls on opt_* insns (#10033)Takashi Kokubun
* YJIT: Allow non-leaf calls on opt_* insns * s/on_send_insn/is_sendish/ * Repeat known_cfunc_codegen
2024-02-20YJIT: Verify the assumption of leaf C calls (#10002)Takashi Kokubun
2024-02-08YJIT: Allow tracing a counted exit (#9890)Takashi Kokubun
* YJIT: Allow tracing a counted exit * Avoid clobbering caller-saved registers
2024-02-02YJIT: add missing jge comparison instruction (#9819)Maxime Chevalier-Boisvert
I ran into this while trying to implement setbyte, was surprised to find out we hadn't implemented it yet.
2024-01-29YJIT: add asm comment when we clear local types (#9713)Maxime Chevalier-Boisvert
Small PR to add a comment when we clear local variable types, so we can be aware that it's happening when looking at the disasm.
2024-01-18YJIT: Stop incrementing chain_depth on defer_compilation (#9597)Takashi Kokubun
2023-11-28YJIT: Assert no patch overlap on pos_marker (#9048)Takashi Kokubun
2023-11-10YJIT: Auto fix for clippy::clone_on_copyAlan Wu
2023-11-07YJIT: Use u32 for CodePtr to save 4 bytes eachAlan Wu
We've long had a size restriction on the code memory region such that a u32 could refer to everything. This commit capitalizes on this restriction by shrinking the size of `CodePtr` to be 4 bytes from 8. To derive a full raw pointer from a `CodePtr`, one needs a base pointer. Both `CodeBlock` and `VirtualMemory` can be used for this purpose. The base pointer is readily available everywhere, except for in the case of the `jit_return` "branch". Generalize lea_label() to lea_jump_target() in the IR to delay deriving the `jit_return` address until `compile()`, when the base pointer is available. On railsbench, this yields roughly a 1% reduction to `yjit_alloc_size` (58,397,765 to 57,742,248).
2023-10-19YJIT: Return Option from asm.compile() for has_dropped_bytes()Alan Wu
So that we get a reminder to check CodeBlock::has_dropped_bytes(). Internally, asm.compile() already checks it, and this patch just propagates it out to the caller with a `#[must_use]`. Code GC logic moved out one level in entry_stub_hit(), so the body can freely use `?`
2023-10-19YJIT: Enable the dead_code lint and delete some dead codeAlan Wu
2023-10-17YJIT: Fix clippy::redundant_localsAlan Wu
> note: `#[deny(clippy::redundant_locals)]` on by default On Rust 1.73.0.
2023-10-05YJIT: Avoid writing return value to memory in `leave`Alan Wu
Previously, at the end of `leave` we did `*caller_cfp->sp = return_value`, like the interpreter. With future changes that leaves the SP field uninitialized for C frames, this will become problematic. For cases like returning from `rb_funcall()`, the return value was written above the stack and never read anyway (callers use the copy in the return register). Leave the return value in a register at the end of `leave` and have the code at `cfp->jit_return` decide what to do with it. This avoids the unnecessary memory write mentioned above. For JIT-to-JIT returns, it goes through `asm.stack_push()` and benefits from register allocation for stack temporaries. Mostly flat on benchmarks, with maybe some marginal speed improvements. Co-authored-by: Takashi Kokubun <[email protected]>
2023-09-29YJIT: Chain-guard opt_mult overflow (#8554)Takashi Kokubun
* YJIT: Chain-guard opt_mult overflow * YJIT: Support regenerating Jo after Mul
2023-09-15YJIT: Avoid creating a vector in get_temp_regs() (#8446)Takashi Kokubun
* YJIT: Avoid creating a vector in get_temp_regs() Co-authored-by: Alan Wu <[email protected]> * Remove unused import --------- Co-authored-by: Alan Wu <[email protected]> Co-authored-by: Alan Wu <[email protected]>
2023-09-15YJIT: Fix and enable the unused_imports warningAlan Wu
2023-09-14YJIT: Skip Insn::Comment and format! if disasm is disabled (#8441)Takashi Kokubun
* YJIT: Skip Insn::Comment and format! if disasm is disabled Co-authored-by: Alan Wu <[email protected]> * YJIT: Get rid of asm.comment --------- Co-authored-by: Alan Wu <[email protected]> Notes: Merged-By: k0kubun <[email protected]>
2023-09-14YJIT: Initialize Vec with capacity for iterators (#8439)Takashi Kokubun
Notes: Merged-By: k0kubun <[email protected]>
2023-09-14YJIT: Initialize Assembler vectors with capacity (#8437)Takashi Kokubun
Notes: Merged-By: maximecb <[email protected]>
2023-08-11YJIT: implement codegen for rb_int_lshift (#8201)Maxime Chevalier-Boisvert
* YJIT: implement codegen for rb_int_lshift * Update yjit/src/asm/x86_64/mod.rs Co-authored-by: Takashi Kokubun <[email protected]> --------- Co-authored-by: Takashi Kokubun <[email protected]> Notes: Merged-By: maximecb <[email protected]>
2023-08-10YJIT: add mul() instruction to backend IR (#8195)Maxime Chevalier-Boisvert
2023-08-04YJIT: guard for array_len >= num in expandarray (#8169)Maxime Chevalier-Boisvert
Avoid generating long dispatch chains for all array lengths seen. Notes: Merged-By: maximecb <[email protected]>
2023-08-03YJIT: add jb (unsigned less-than) instruction to backend (#8168)Maxime Chevalier-Boisvert
Notes: Merged-By: maximecb <[email protected]>
2023-07-29YJIT: Drop Copy trait from Context (#8138)Takashi Kokubun
Notes: Merged-By: maximecb <[email protected]>
2023-07-27YJIT: implement missing `asm.jg` instruction in backend (#8130)Maxime Chevalier-Boisvert
YJIT: implement missing jg instruction in backend While trying to implement a specialize integer left shift, I ran into a problem where we have no way to do a greater-than comparison at the moment. Surprising we went this far without ever needing it. Notes: Merged-By: maximecb <[email protected]>
2023-07-06YJIT: Use registers to pass stack temps to C calls (#7920)Takashi Kokubun
* YJIT: Use registers to pass stack temps to C calls * YJIT: Update comments in ccall
2023-07-04YJIT: Break register cycles for C arguments (take 2) (#8018)Takashi Kokubun
* Revert "Revert "YJIT: Break register cycles for C arguments (#7918)"" This reverts commit 78ca085785460de46bfc4851a898d525c1698ef8. * Use shfited_live_ranges for the last-insn check Notes: Merged-By: maximecb <[email protected]>
2023-06-12Revert "YJIT: Break register cycles for C arguments (#7918)"Takashi Kokubun
This reverts commit 888ba29e462075472776098f4f95eb6d3df8e730. It caused a CI failure http://ci.rvm.jp/results/trunk-yjit@ruby-sp2-docker/4598881 and I'm investigating it.
2023-06-12YJIT: Break register cycles for C arguments (#7918)Takashi Kokubun
Notes: Merged-By: maximecb <[email protected]>
2023-04-20YJIT: Merge lower_stack into the split pass (#7748)Takashi Kokubun
Notes: Merged-By: k0kubun <[email protected]>
2023-04-19YJIT: Remove Insn::RegTemps (#7741)Takashi Kokubun
* YJIT: Remove Insn::RegTemps * Update a comment Co-authored-by: Maxime Chevalier-Boisvert <[email protected]> --------- Co-authored-by: Maxime Chevalier-Boisvert <[email protected]> Notes: Merged-By: k0kubun <[email protected]>
2023-04-17YJIT: Spill a caller stack as late as possible (#7726)Takashi Kokubun
Notes: Merged-By: maximecb <[email protected]>
2023-04-14YJIT: Add a counter to all side exits (#7720)Takashi Kokubun
Notes: Merged-By: maximecb <[email protected]>
2023-04-14YJIT: Introduce Target::SideExit (#7712)Takashi Kokubun
* YJIT: Introduce Target::SideExit * YJIT: Obviate Insn::SideExitContext * YJIT: Avoid cloning a Context for each insn Notes: Merged-By: maximecb <[email protected]>
2023-04-13YJIT: Move stack_opnd from Context to Assembler (#7700)Takashi Kokubun
Notes: Merged-By: maximecb <[email protected]>
2023-04-12YJIT: Let Assembler own Context (#7691)Takashi Kokubun
* YJIT: Let Assembler own Context * Update a comment Co-authored-by: Maxime Chevalier-Boisvert <[email protected]> --------- Co-authored-by: Maxime Chevalier-Boisvert <[email protected]> Notes: Merged-By: k0kubun <[email protected]>
2023-04-06YJIT: Stack temp register allocation for arm64 (#7659)Takashi Kokubun
* YJIT: Stack temp register allocation for arm64 * Update a comment Co-authored-by: Maxime Chevalier-Boisvert <[email protected]> * Update comments about assertion * Update a comment Co-authored-by: Maxime Chevalier-Boisvert <[email protected]> --------- Co-authored-by: Maxime Chevalier-Boisvert <[email protected]> Notes: Merged-By: maximecb <[email protected]>
2023-04-04YJIT: Stack temp register allocation (#7651)Takashi Kokubun
Co-authored-by: Maxime Chevalier-Boisvert <[email protected]> Notes: Merged-By: k0kubun <[email protected]>
2023-03-15YJIT: Eliminate unnecessary mov for trampolines (#7537)Takashi Kokubun
Notes: Merged-By: k0kubun <[email protected]>
2023-03-07YJIT: Add comments to peek and x86_mergeTakashi Kokubun
Notes: Merged: https://github.com/ruby/ruby/pull/7453
2023-03-07YJIT: Merge lea and mov on x86_64 when possibleTakashi Kokubun
Notes: Merged: https://github.com/ruby/ruby/pull/7453
2023-03-02YJIT: shrink stack_size/sp_offet to u8/i8 (#7426)Maxime Chevalier-Boisvert
Notes: Merged-By: maximecb <[email protected]>
generated by cgit v1.2.3 (git 2.25.1) at 2025-05-02 04:57:50 +0000