RESOLVED FIXED 168629
Air should have a disassembly mode that dumps IR and assembly intermixed 
https://bugs.webkit.org/show_bug.cgi?id=168629
Summary Air should have a disassembly mode that dumps IR and assembly intermixed
Saam Barati
Reported 2017-02-20 18:46:27 PST
Similar to the output you get from dumpDFGDisassembly, but for Air. I'm going to be spending a lot of time looking at generated code, and it will make job easier to know where generated code came from.
Attachments
WIP (17.33 KB, patch)
2017-02-21 12:58 PST, Saam Barati 		no flags
DetailsFormatted DiffDiff
patch (23.18 KB, patch)
2017-02-21 15:28 PST, Saam Barati 		fpizlo: review+
DetailsFormatted DiffDiff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.
Saam Barati
Comment 1 2017-02-21 12:58:26 PST
Created attachment 302296 [details] WIP It produces output like: BB#0: ; frequency = 1.000000 0x30b448402e00: push %rbp 0x30b448402e01: mov %rsp, %rbp 0x30b448402e04: add $0xffffffffffffffd0, %rsp Move $0x11456c960, %rax, $4636199264(@16) 0x30b448402e08: mov $0x11456c960, %rax Move %rax, 16(%rbp), @19 0x30b448402e12: mov %rax, 0x10(%rbp) Patch &Patchpoint2, %rbp, %rax, @20 0x30b448402e16: lea -0x50(%rbp), %rax 0x30b448402e1a: mov $0x1112081e0, %r11 0x30b448402e24: cmp %rax, (%r11) 0x30b448402e27: ja 0x30b448402e9a Move 56(%rbp), %rdx, @23 0x30b448402e2d: mov 0x38(%rbp), %rdx Move $0xffff000000000002, %rax, $-281474976710654(@15) 0x30b448402e31: mov $0xffff000000000002, %rax Patch &BranchTest64(3,SameAsRep)1, NonZero, %rdx, %rax, %rdx, @26 0x30b448402e3b: test %rdx, %rax 0x30b448402e3e: jnz 0x30b448402f08 Move 48(%rbp), %rax, @29 0x30b448402e44: mov 0x30(%rbp), %rax Move %rax, %rcx, @31 0x30b448402e48: mov %rax, %rcx Xor64 $6, %rcx, @31 0x30b448402e4b: xor $0x6, %rcx Patch &BranchTest64(3,SameAsRep)1, NonZero, %rcx, $-2, %rax, @35 0x30b448402e4f: test $0xfffffffffffffffe, %rcx 0x30b448402e56: jnz 0x30b448402f12 Patch &Branch32(3,SameAsRep)0, NotEqual, (%rdx), $266, %rdx, @45 0x30b448402e5c: cmp $0x10a, (%rdx) 0x30b448402e62: jnz 0x30b448402f1c BranchTest32 NonZero, %rax, $1, @49 0x30b448402e68: test $0x1, %al 0x30b448402e6a: jnz 0x30b448402e91 BB#1: ; frequency = 1.000000 Predecessors: #0 Move $0, %rcx, @65 0x30b448402e70: xor %rcx, %rcx BB#2: ; frequency = 1.000000 Predecessors: #1, #3 Move 24(%rdx), %rax, @58 0x30b448402e73: mov 0x18(%rdx), %rax Patch &BranchAdd32(4,ForceLateUseUnlessRecoverable)3, Overflow, %rcx, %rax, %rcx, %rcx, %rax, @60 0x30b448402e77: add %eax, %ecx 0x30b448402e79: jo 0x30b448402f26 Move $0xffff000000000000, %rax, $-281474976710656(@14) 0x30b448402e7f: mov $0xffff000000000000, %rax Add64 %rcx, %rax, %rax, @62 0x30b448402e89: add %rcx, %rax Ret64 %rax, @63 0x30b448402e8c: mov %rbp, %rsp 0x30b448402e8f: pop %rbp 0x30b448402e90: ret
Caio Lima
Comment 2 2017-02-21 14:46:12 PST
This is really good!
Saam Barati
Comment 3 2017-02-21 15:22:28 PST
Ok, I revised the output a bit. It'll look like this when using dumpFTLDisassembly: Generated FTL JIT code for foo#CUaFiQ:[0x10b76c960->0x10b76c2d0->0x10b7b6da0, FTLFunctionCall, 40 (NeverInline)], instruction count = 40: BB#0: ; frequency = 1.000000 0x469004e02e00: push %rbp 0x469004e02e01: mov %rsp, %rbp 0x469004e02e04: add $0xffffffffffffffd0, %rsp Move $0x10b76c960, %rax, $4487301472(@16) 0x469004e02e08: mov $0x10b76c960, %rax Move %rax, 16(%rbp), @19 0x469004e02e12: mov %rax, 0x10(%rbp) Patch &Patchpoint2, %rbp, %rax, @20 0x469004e02e16: lea -0x50(%rbp), %rax 0x469004e02e1a: mov $0x1084081e0, %r11 0x469004e02e24: cmp %rax, (%r11) 0x469004e02e27: ja 0x469004e02e9a Move 56(%rbp), %rdx, @23 0x469004e02e2d: mov 0x38(%rbp), %rdx Move $0xffff000000000002, %rax, $-281474976710654(@15) 0x469004e02e31: mov $0xffff000000000002, %rax Patch &BranchTest64(3,SameAsRep)1, NonZero, %rdx, %rax, %rdx, @26 0x469004e02e3b: test %rdx, %rax 0x469004e02e3e: jnz 0x469004e02f08 Move 48(%rbp), %rax, @29 0x469004e02e44: mov 0x30(%rbp), %rax Move %rax, %rcx, @31 0x469004e02e48: mov %rax, %rcx Xor64 $6, %rcx, @31 0x469004e02e4b: xor $0x6, %rcx Patch &BranchTest64(3,SameAsRep)1, NonZero, %rcx, $-2, %rax, @35 0x469004e02e4f: test $0xfffffffffffffffe, %rcx 0x469004e02e56: jnz 0x469004e02f12 Patch &Branch32(3,SameAsRep)0, NotEqual, (%rdx), $266, %rdx, @45 0x469004e02e5c: cmp $0x10a, (%rdx) 0x469004e02e62: jnz 0x469004e02f1c BranchTest32 NonZero, %rax, $1, @49 0x469004e02e68: test $0x1, %al 0x469004e02e6a: jnz 0x469004e02e91 Successors: #3, #1 BB#1: ; frequency = 1.000000 Predecessors: #0 Move $0, %rcx, @65 0x469004e02e70: xor %rcx, %rcx Jump @66 Successors: #2 BB#2: ; frequency = 1.000000 Predecessors: #1, #3 Move 24(%rdx), %rax, @58 0x469004e02e73: mov 0x18(%rdx), %rax Patch &BranchAdd32(4,ForceLateUseUnlessRecoverable)3, Overflow, %rcx, %rax, %rcx, %rcx, %rax, @60 0x469004e02e77: add %eax, %ecx 0x469004e02e79: jo 0x469004e02f26 Move $0xffff000000000000, %rax, $-281474976710656(@14) 0x469004e02e7f: mov $0xffff000000000000, %rax Add64 %rcx, %rax, %rax, @62 0x469004e02e89: add %rcx, %rax Ret64 %rax, @63 0x469004e02e8c: mov %rbp, %rsp 0x469004e02e8f: pop %rbp 0x469004e02e90: ret BB#3: ; frequency = 1.000000 Predecessors: #0 Move 16(%rdx), %rcx, @52 0x469004e02e91: mov 0x10(%rdx), %rcx Jump @55 0x469004e02e95: jmp 0x469004e02e73 Successors: #2 # Late paths 0x469004e02e9a: mov $0x0, 0x24(%rbp) 0x469004e02ea1: mov $0x108406618, %r9 0x469004e02eab: mov (%r9), %r9 0x469004e02eae: add $0xffffffffffffffd8, %r9 0x469004e02eb2: mov %rbx, (%r9) 0x469004e02eb5: mov %r12, 0x8(%r9) 0x469004e02eb9: mov %r13, 0x10(%r9) 0x469004e02ebd: mov %r14, 0x18(%r9) 0x469004e02ec1: mov %r15, 0x20(%r9) 0x469004e02ec5: mov %rbp, %rdi 0x469004e02ec8: mov $0x10b76c960, %rsi 0x469004e02ed2: mov $0x0, %r11 0x469004e02edc: call *%r11 0x469004e02edf: mov $0x108400000, %rdi 0x469004e02ee9: mov %rbp, %rsi 0x469004e02eec: mov $0x0, %r11 0x469004e02ef6: call *%r11 0x469004e02ef9: mov $0x1084080f0, %rsi 0x469004e02f03: mov (%rsi), %rsi 0x469004e02f06: jmp *%rsi 0x469004e02f08: push $0x0 0x469004e02f0d: jmp 0x469004e02f12 0x469004e02f12: push $0x1 0x469004e02f17: jmp 0x469004e02f1c 0x469004e02f1c: push $0x2 0x469004e02f21: jmp 0x469004e02f26 0x469004e02f26: sub %eax, %ecx 0x469004e02f28: push $0x3 0x469004e02f2d: jmp 0x469004e02f32
Saam Barati
Comment 4 2017-02-21 15:28:22 PST
Created attachment 302320 [details] patch
WebKit Commit Bot
Comment 5 2017-02-21 15:31:36 PST
Attachment 302320 [details] did not pass style-queue: ERROR: Source/JavaScriptCore/b3/air/AirGenerate.cpp:209: Weird number of spaces at line-start. Are you using a 4-space indent? [whitespace/indent] [3] ERROR: Source/JavaScriptCore/b3/air/AirGenerate.cpp:213: Weird number of spaces at line-start. Are you using a 4-space indent? [whitespace/indent] [3] ERROR: Source/JavaScriptCore/b3/air/AirGenerate.cpp:227: Weird number of spaces at line-start. Are you using a 4-space indent? [whitespace/indent] [3] Total errors found: 3 in 8 files If any of these errors are false positives, please file a bug against check-webkit-style.
Saam Barati
Comment 6 2017-02-21 15:59:03 PST
landed in: https://trac.webkit.org/changeset/212775
Note You need to log in before you can comment on or make changes to this bug.