WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
207677
Add options for debugging WASM code.
https://bugs.webkit.org/show_bug.cgi?id=207677
Summary
Add options for debugging WASM code.
Mark Lam
Reported
2020-02-12 19:50:13 PST
Specifically ... JSC_useBBQJIT - allows the BBQ JIT to be used if true JSC_useOMGJIT - allows the OMG JIT to be used if true JSC_useWasmLLIntPrologueOSR - allows prologue OSR from Wasm LLInt if true JSC_useWasmLLIntLoopOSR - allows loop OSR from Wasm LLInt if true JSC_useWasmLLIntEpilogueOSR - allows epilogue OSR from Wasm LLInt if true JSC_wasmFunctionIndexRangeToCompile=N:M - wasm function index range to allow compilation on, e.g. 1:100
Attachments
proposed patch.
(15.26 KB, patch)
2020-02-12 19:57 PST
,
Mark Lam
ysuzuki
: review+
Details
Formatted Diff
Diff
View All
Add attachment
proposed patch, testcase, etc.
Radar WebKit Bug Importer
Comment 1
2020-02-12 19:51:16 PST
<
rdar://problem/59411390
>
Mark Lam
Comment 2
2020-02-12 19:57:59 PST
Created
attachment 390611
[details]
proposed patch.
Yusuke Suzuki
Comment 3
2020-02-12 20:08:50 PST
Comment on
attachment 390611
[details]
proposed patch. View in context:
https://bugs.webkit.org/attachment.cgi?id=390611&action=review
r=me with one comment.
> Source/JavaScriptCore/wasm/WasmSlowPaths.cpp:98 > + if (!Options::wasmFunctionIndexRangeToCompile().isInRange(codeBlock->functionIndex()))
This functionIndex is per-wasm-module thing. If we have two wasm modules, they can have the same function index for different functions. Currently, I think this is OK. But maybe, should we introduce dfgWhitelist like mechanism instead?
Mark Lam
Comment 4
2020-02-12 20:10:09 PST
(In reply to Yusuke Suzuki from
comment #3
)
> > Source/JavaScriptCore/wasm/WasmSlowPaths.cpp:98 > > + if (!Options::wasmFunctionIndexRangeToCompile().isInRange(codeBlock->functionIndex())) > > This functionIndex is per-wasm-module thing. If we have two wasm modules, > they can have the same function index for different functions. > Currently, I think this is OK. But maybe, should we introduce dfgWhitelist > like mechanism instead?
Yes, I understood this to be the case. It will do for now. Thanks.
Mark Lam
Comment 5
2020-02-12 20:12:30 PST
Thanks for the review. Landed in
r256500
: <
http://trac.webkit.org/r256500
>.
Tadeu Zagallo
Comment 6
2020-02-13 08:03:19 PST
These should all have been wrapped in UNLIKELY
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug