- 16 Nov, 2015 2 commits
-
-
ahaas authored
The TiesEven rounding mode rounds float64 numbers to the nearest integer. If there are two nearest integers, then the number is rounded to the even one. This is the default rounding mode according to IEEE~754. I implemented the operator on ia32, x64, arm, arm64, mips, and mips64. I think there is a bug in the current implementation of the ppc simulator, which kept me from implementing the operator on ppc. According to my understanding of the ppc instruction manual, the FRIN instruction provides the right behavior for Float64RoundTiesEven. In the simulator, however, FRIN provides a different semantics. If there are two nearest integers, then the simulator returns the one which is further away form 0. Review URL: https://codereview.chromium.org/1440293002 Cr-Commit-Position: refs/heads/master@{#32005}
-
ahaas authored
I implemented it on x64, ia32, arm, arm64, mips, mips64, and ppc. Review URL: https://codereview.chromium.org/1444583002 Cr-Commit-Position: refs/heads/master@{#32004}
-
- 13 Nov, 2015 1 commit
-
-
danno authored
* Limit triggering of tail calls to explicit use of a new inline runtime function %_TailCall. %_TailCall works just like %_Call except for using tail-calling mechanics (currently only in TF). * Remove hack that recognized some specific usages of %_Call and converted them into tail calls. * Support tail calls for all calls where the number of callee stack parameters is less than or equal to the number of caller stack parameters. * Use the gap resolver to swizzle parameters and registers to tail calls. BUG=v8:4076 LOG=n Review URL: https://codereview.chromium.org/1439613003 Cr-Commit-Position: refs/heads/master@{#31987}
-
- 10 Nov, 2015 1 commit
-
-
bmeurer authored
Avoid write barriers when storing values in the root set, and use cheaper write barriers for storing maps or tagged pointers. Also improve the generated code for write barriers, utilizing the out of line code mechanism that is available to TurboFan backends, which moves the unlikely case out of the hot path. R=jarin@chromium.org, mstarzinger@chromium.org Review URL: https://codereview.chromium.org/1414183006 Cr-Commit-Position: refs/heads/master@{#31914}
-
- 26 Oct, 2015 2 commits
-
-
jarin authored
Review URL: https://codereview.chromium.org/1414223004 Cr-Commit-Position: refs/heads/master@{#31561}
-
jarin authored
Review URL: https://codereview.chromium.org/1411093006 Cr-Commit-Position: refs/heads/master@{#31547}
-
- 19 Oct, 2015 1 commit
-
-
jarin authored
This introduces an explicit lazy bailout. It is wrapped in the call node, mostly because the lazy deoptimization processing is married to the call processing in the instruction selector and the code generator. It is still a terrible hack. R=bmeurer@chromium.org,mstarzinger@chromium.org BUG=chromium:543994,v8:4195 LOG=n Review URL: https://codereview.chromium.org/1412443003 Cr-Commit-Position: refs/heads/master@{#31353}
-
- 16 Oct, 2015 1 commit
-
-
ahaas authored
R=titzer@google.com Review URL: https://codereview.chromium.org/1407933002 Cr-Commit-Position: refs/heads/master@{#31319}
-
- 15 Oct, 2015 1 commit
-
-
ahaas authored
Review URL: https://codereview.chromium.org/1405453003 Cr-Commit-Position: refs/heads/master@{#31313}
-
- 29 Sep, 2015 1 commit
-
-
jacob.bramley authored
It is not safe to access memory below sp, and this is also forbidden by the ABI. When pushing, we must either use an atomic operation (such as vstm+db_w) or move sp before writing the data. This patch fixes one stack access, and also adds vpush and vpop helpers to simplify similar code. BUG= Review URL: https://codereview.chromium.org/1378563002 Cr-Commit-Position: refs/heads/master@{#31008}
-
- 24 Sep, 2015 2 commits
-
-
pierre.langlois authored
This patch checks the type of the lhs operand of a floating point comparison for ARM, and commutes the operands if it is #0.0. It allows us to optimize a comparison with zero, as the vcmp instruction accepts #0.0 as rhs operand. Code before for "0.0 < 0.123": ------------------------------ movw ip, #29360 movt ip, #37224 movw r9, #31981 movt r9, #16319 vmov d0, ip, r9 mov ip, #0 vmov d1, ip, ip vcmp.f64 d1, d0 vmrs APSR, FPSCR bcc +12 Code after: ----------- movw ip, #29360 movt ip, #37224 movw r9, #31981 movt r9, #16319 vmov d0, ip, r9 vcmp.f64 d0, #0.0 vmrs APSR, FPSCR bgt +12 BUG= Review URL: https://codereview.chromium.org/1361913003 Cr-Commit-Position: refs/heads/master@{#30911}
-
martyn.capewell authored
Reduce operations of the form f64cmp(fp32to64(x), k) to f32cmp(x, k) when k can be encoded as a 32-bit float. Review URL: https://codereview.chromium.org/1365623002 Cr-Commit-Position: refs/heads/master@{#30909}
-
- 23 Sep, 2015 1 commit
-
-
pierre.langlois authored
This patch checks the type of the lhs operand of a floating point comparison, and commutes the operands if it is #0.0. It allows us to optimize a comparison with zero, as the fcmp instruction accepts #0.0 as rhs operand. Code before for "0.0 < 0.123": ------------------------------ fmov d1, xzr ldr d0, pc+96 fcmp d1, d0 b.lo #+0xc Code after: ----------- ldr d0, pc+92 fcmp d0, #0.0 b.gt #+0xc Before this patch, we used unsigned condition codes for floating point comparisons, but the unordered case was not correctly commuted. Review URL: https://codereview.chromium.org/1356283003 Cr-Commit-Position: refs/heads/master@{#30881}
-
- 21 Sep, 2015 1 commit
-
-
titzer authored
R=bmeurer@chromium.org Review URL: https://codereview.chromium.org/1356913002 Cr-Commit-Position: refs/heads/master@{#30849}
-
- 03 Sep, 2015 1 commit
-
-
titzer authored
This is to support WebAssembly 64-bit ints in the short term, since it currently uses CheckedLoad/CheckedStore for accesses to the memory. In the long run, we'll change this to be explicit bounds checks that throw on out of bounds. R=bmeurer@chromium.org BUG= Review URL: https://codereview.chromium.org/1310323006 Cr-Commit-Position: refs/heads/master@{#30555}
-
- 24 Aug, 2015 1 commit
-
-
rmcilroy authored
Replaces all instances of the code which computed the debug name of a stub or function with calls to CompileInfo::GetDebugName instead. Also: - Removes useless parameter on CodeStub::GetMajorName - Removes FakeStubForTesting since it is no longer required - Adds CompileInfo::ShouldEnsureSpaceForLazyDeopt() to replace unclear calls to IsStub(). Review URL: https://codereview.chromium.org/1297203002 Cr-Commit-Position: refs/heads/master@{#30324}
-
- 18 Aug, 2015 1 commit
-
-
danno authored
Previously, it was not possible to specify StackSlotOperands for all slots in both the caller and callee stacks. Specifically, the region of the callee's stack including the saved return address, frame pointer, function pointer and context pointer could not be addressed by the register allocator/gap resolver. In preparation for better tail call support, which will use the gap resolver to reconcile outgoing parameters, this change makes it possible to address all slots on the stack, because slots in the previously inaccessible dead zone may become parameter slots for outgoing tail calls. All caller stack slots are accessible as they were before, with slot -1 corresponding to the last stack parameter. Stack slot indices >= 0 access the callee stack, with slot 0 corresponding to the callee's saved return address, 1 corresponding to the saved frame pointer, 2 corresponding to the current function context, 3 corresponding to the frame marker/JSFunction, and slots 4 and above corresponding to spill slots. The following changes were specifically needed: * Frame has been changed to explicitly manage three areas of the callee frame, the fixed header, the spill slot area, and the callee-saved register area. * Conversions from stack slot indices to fp offsets all now go through a common bottleneck: OptimizedFrame::StackSlotOffsetRelativeToFp * The generation of deoptimization translation tables has been changed to support the new stack slot indexing scheme. Crankshaft, which doesn't support the new slot numbering in its register allocator, must adapt the indexes when creating translation tables. * Callee-saved parameters are now kept below spill slots, not above, to support saving only the optimal set of used registers, which is only known after register allocation is finished and spill slots have been allocated. Review URL: https://codereview.chromium.org/1261923007 Cr-Commit-Position: refs/heads/master@{#30224}
-
- 17 Aug, 2015 1 commit
-
-
titzer authored
R=jarin@chromium.org BUG= Review URL: https://codereview.chromium.org/1291113003 Cr-Commit-Position: refs/heads/master@{#30203}
-
- 11 Aug, 2015 2 commits
-
-
titzer authored
Reland: [turbofan] Various fixes to allow unboxed doubles as arguments in registers and on the stack. OCL: https://codereview.chromium.org/1263033004/ R=mstarzinger@chromium.org BUG= Review URL: https://codereview.chromium.org/1284893002 Cr-Commit-Position: refs/heads/master@{#30115}
-
yangguo authored
Revert of [turbofan] Various fixes to allow unboxed doubles as arguments in registers and on the stack. (patchset #7 id:120001 of https://codereview.chromium.org/1263033004/ ) Reason for revert: This CL breaks MIPS (roll blocker). https://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20mipsel%20-%20sim/builds/2061/steps/Check/logs/Run_Int32_Select_1 Original issue's description: > [turbofan] Various fixes to allow unboxed doubles as arguments in registers and on the stack. > > R=jarin@chromium.org > BUG= > > Committed: https://crrev.com/71409be5395f867bbca0f6998bf6caa175cd8192 > Cr-Commit-Position: refs/heads/master@{#30091} TBR=jarin@chromium.org,titzer@chromium.org NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true BUG= Review URL: https://codereview.chromium.org/1284853002 Cr-Commit-Position: refs/heads/master@{#30101}
-
- 10 Aug, 2015 1 commit
-
-
titzer authored
R=jarin@chromium.org BUG= Review URL: https://codereview.chromium.org/1263033004 Cr-Commit-Position: refs/heads/master@{#30091}
-
- 07 Aug, 2015 2 commits
-
-
titzer authored
Rationale: The {kind} of a call descriptor describes what the {target} being called is--i.e. a JSFunction, code object, or address. That kind materially dictates the instruction(s) generated for an outgoing call. The other flags on a call descriptor should describe specific properties (like whether a roots register is valid or not) so that backend logic doesn't have to switch over the kind, but is informed directly of what it wants to know. R=mstarzinger@chromium.org BUG= Review URL: https://codereview.chromium.org/1268273003 Cr-Commit-Position: refs/heads/master@{#30065}
-
titzer authored
[turbofan] Remove architecture-specific linkage files and LinkageTraits. Use macro-assembler-defined constants. R=mstarzinger@chromium.org BUG= Review URL: https://codereview.chromium.org/1272883003 Cr-Commit-Position: refs/heads/master@{#30063}
-
- 31 Jul, 2015 1 commit
-
-
titzer authored
R=danno@chromium.org BUG= Review URL: https://codereview.chromium.org/1269913002 Cr-Commit-Position: refs/heads/master@{#29957}
-
- 30 Jul, 2015 3 commits
-
-
titzer authored
This is the first step in cutting the Gordian linkage/linkage-impl knot. This basically changes the axis along which we organize call descriptor building logic from having platform-specific files dedicated to all call descriptor types to having call-descriptor-type-specific files that have The next step is to factor the JS, code stub, and runtime call descriptors similarly, dumping them into: compiler/js-linkage.cc compiler/runtime-linkage.cc compiler/code-stub-linkage.cc or, alternatively, all of them just into compiler/js-linkage.cc. This also anticipates a wasm-linkage.cc file in the future. R=bmeurer@chromium.org,danno@chromium.org BUG= Review URL: https://codereview.chromium.org/1266603002 Cr-Commit-Position: refs/heads/master@{#29931}
-
rmcilroy authored
Adds interpreter entry and exit trampoline builtins. Also implements the Return bytecode handler and fixes a few bugs in InterpreterAssembler highlighted by running on other architectures. MIPS and MIPS64 port contributed by Paul Lind (paul.lind@imgtec.com) BUG=v8:4280 LOG=N Review URL: https://codereview.chromium.org/1245133002 Cr-Commit-Position: refs/heads/master@{#29929}
-
rmcilroy authored
Changes the interpreter to use a BytecodeArray pointer and an offset to avoid having an inner pointer to a BytecodeArray object in registers during dispatch. BUG=v8:4280 LOG=N Review URL: https://codereview.chromium.org/1254293006 Cr-Commit-Position: refs/heads/master@{#29910}
-
- 23 Jul, 2015 1 commit
-
-
rmcilroy authored
Adds basic support for generation of interpreter bytecode handler code snippets. The InterpreterAssembler class exposes a set of low level, interpreter specific operations which can be used to build a Turbofan graph. The Interpreter class generates a bytecode handler snippet for each bytecode by assembling operations using an InterpreterAssembler. Currently only two simple bytecodes are supported: LoadLiteral0 and Return. BUG=v8:4280 LOG=N Review URL: https://codereview.chromium.org/1239793002 Cr-Commit-Position: refs/heads/master@{#29814}
-
- 13 Jul, 2015 2 commits
-
-
rmcilroy authored
BUG=v8:4280 LOG=N Review URL: https://codereview.chromium.org/1234443004 Cr-Commit-Position: refs/heads/master@{#29591}
-
bmeurer authored
Some TurboFan instructions block the literal pool for their entire duration. If a long enough sequence of these instructions is encountered, the literal pool can be blocked until it goes out of range. Patch from issue 1236603002 at patchset 1 (http://crrev.com/1236603002#ps1). BUG=v8:4292 LOG=y R=jarin@chromium.org Review URL: https://codereview.chromium.org/1232343002 Cr-Commit-Position: refs/heads/master@{#29587}
-
- 08 Jul, 2015 1 commit
-
-
bmeurer authored
We actually need round to zero truncation to implement the counterpart of LDoubleToI in TurboFan, which tries to convert a double to an integer as required for keyed load/store optimizations. Drive-by-cleanup: Reduce some code duplication in the InstructionSelector implementations. R=jarin@chromium.org Review URL: https://codereview.chromium.org/1225993002 Cr-Commit-Position: refs/heads/master@{#29527}
-
- 02 Jul, 2015 1 commit
-
-
danno authored
This involves: - Enabling the tail call optimization reducer in all cases. - Adding an addition flag to CallFunctionParameters to mark call sites that can be tail-called enabled. - Only set the tail-call flag for %_CallFunction. R=bmeurer@chromium.org Review URL: https://codereview.chromium.org/1216933011 Cr-Commit-Position: refs/heads/master@{#29436}
-
- 30 Jun, 2015 1 commit
-
-
akos.palfi authored
This change makes possible to save and restore the FP registers in the Prologue and Return parts for the CallAddress kind functions. TEST=test-simplified-lowering/RunNumberDivide_2_TruncatingToUint32, test-simplified-lowering/RunNumberMultiply_TruncatingToUint32 Review URL: https://codereview.chromium.org/1191513003 Cr-Commit-Position: refs/heads/master@{#29378}
-
- 26 Jun, 2015 1 commit
-
-
bmeurer authored
This optimization is already implemented in fullcodegen, and basically makes sure that we do not unecessarily blow up the code with duplicated return sequences everywhere. R=danno@chromium.org Review URL: https://codereview.chromium.org/1211373002 Cr-Commit-Position: refs/heads/master@{#29315}
-
- 25 Jun, 2015 1 commit
-
-
Benedikt Meurer authored
This introduces some initial building blocks for calling out to C/C++ functions directly from TurboFan generated code objects. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/1205023002. Cr-Commit-Position: refs/heads/master@{#29279}
-
- 17 Jun, 2015 1 commit
-
-
mstarzinger authored
This fixes CodeGenerator::EnsureSpaceForLazyDeopt to no longer be treated as a lazy deopt site in itself. Calls mark themselves as lazy bailout sites in CodeGenerator::RecordCallPosition, which suffices. R=jarin@chromium.org Review URL: https://codereview.chromium.org/1186353003 Cr-Commit-Position: refs/heads/master@{#29069}
-
- 05 Jun, 2015 1 commit
-
-
danno authored
Only optimized for TF R=mstarzinger@chromium.org Review URL: https://codereview.chromium.org/1146963002 Cr-Commit-Position: refs/heads/master@{#28812}
-
- 04 Jun, 2015 1 commit
-
-
mbrandy authored
Embed constant pools within their corresponding Code objects. This removes support for out-of-line constant pools in favor of the new approach -- the main advantage being that it eliminates the need to allocate and manage separate constant pool array objects. Currently supported on PPC and ARM. Enabled by default on PPC only. This yields a 6% improvment in Octane on PPC64. R=bmeurer@chromium.org, rmcilroy@chromium.org, michael_dawson@ca.ibm.com BUG=chromium:478811 LOG=Y Review URL: https://codereview.chromium.org/1162993006 Cr-Commit-Position: refs/heads/master@{#28801}
-
- 03 Jun, 2015 1 commit
-
-
bmeurer authored
Revert of Embedded constant pools. (patchset #12 id:220001 of https://codereview.chromium.org/1131783003/) Reason for revert: Breaks Linux nosnap cctest/test-api/FastReturnValuesWithProfiler, see http://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20nosnap%20-%20debug%20-%202/builds/609/steps/Check/logs/FastReturnValuesWithP.. Original issue's description: > Add support for Embedded Constant Pools for PPC and Arm > > Embed constant pools within their corresponding Code > objects. > > This removes support for out-of-line constant pools in favor > of the new approach -- the main advantage being that it > eliminates the need to allocate and manage separate constant > pool array objects. > > Currently supported on PPC and ARM. Enabled by default on > PPC only. > > This yields a 6% improvment in Octane on PPC64. > > R=danno@chromium.org, svenpanne@chromium.org, bmeurer@chromium.org, rmcilroy@chromium.org, dstence@us.ibm.com, michael_dawson@ca.ibm.com > BUG=chromium:478811 > LOG=Y > > Committed: https://crrev.com/a9404029343d65f146e3443f5280c40a97e736af > Cr-Commit-Position: refs/heads/master@{#28770} TBR=rmcilroy@chromium.org,ishell@chromium.org,rodolph.perfetta@arm.com,mbrandy@us.ibm.com NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true BUG=chromium:478811 Review URL: https://codereview.chromium.org/1155703006 Cr-Commit-Position: refs/heads/master@{#28772}
-
- 02 Jun, 2015 1 commit
-
-
mbrandy authored
Embed constant pools within their corresponding Code objects. This removes support for out-of-line constant pools in favor of the new approach -- the main advantage being that it eliminates the need to allocate and manage separate constant pool array objects. Currently supported on PPC and ARM. Enabled by default on PPC only. This yields a 6% improvment in Octane on PPC64. R=danno@chromium.org, svenpanne@chromium.org, bmeurer@chromium.org, rmcilroy@chromium.org, dstence@us.ibm.com, michael_dawson@ca.ibm.com BUG=chromium:478811 LOG=Y Review URL: https://codereview.chromium.org/1131783003 Cr-Commit-Position: refs/heads/master@{#28770}
-