[Webkit-unassigned] [Bug 232265] Refactoring and PutByVal cleanup
bugzilla-daemon at webkit.org
bugzilla-daemon at webkit.org
Wed Nov 3 11:33:26 PDT 2021
https://bugs.webkit.org/show_bug.cgi?id=232265
--- Comment #5 from Saam Barati <sbarati at apple.com> ---
Comment on attachment 442524
--> https://bugs.webkit.org/attachment.cgi?id=442524
Patch
View in context: https://bugs.webkit.org/attachment.cgi?id=442524&action=review
> Source/JavaScriptCore/dfg/DFGSpeculativeJIT.cpp:2513
> +#if USE(JSVALUE32_64)
> + m_jit.store32(valueTagReg, MacroAssembler::BaseIndex(storageReg, propertyReg, MacroAssembler::TimesEight, OBJECT_OFFSETOF(JSValue, u.asBits.tag)));
> + m_jit.store32(valuePayloadReg, MacroAssembler::BaseIndex(storageReg, propertyReg, MacroAssembler::TimesEight, OBJECT_OFFSETOF(JSValue, u.asBits.payload)));
> +#else
> + m_jit.store64(valueReg, MacroAssembler::BaseIndex(storageReg, propertyReg, MacroAssembler::TimesEight));
> +#endif
Is it worth making this code JSValueRegs based?
> Source/JavaScriptCore/dfg/DFGSpeculativeJIT.cpp:2547
> +#if USE(JSVALUE32_64)
> + m_jit.store32(valueTagReg, MacroAssembler::BaseIndex(storageReg, propertyReg, MacroAssembler::TimesEight, OBJECT_OFFSETOF(JSValue, u.asBits.tag)));
> + m_jit.store32(valuePayloadReg, MacroAssembler::BaseIndex(storageReg, propertyReg, MacroAssembler::TimesEight, OBJECT_OFFSETOF(JSValue, u.asBits.payload)));
> +#else
> + m_jit.store64(valueReg, MacroAssembler::BaseIndex(storageReg, propertyReg, MacroAssembler::TimesEight));
> +#endif
ditto
> Source/JavaScriptCore/dfg/DFGSpeculativeJIT64.cpp:3555
> + compilePutByVal(node);
why? This seems to be just moving code around within the same file. Is the goal just to have compilePutByVal functions on both 32-bit and 64-bit, even though they don't share an implementation?
--
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20211103/b8f3f4f6/attachment.htm>
More information about the webkit-unassigned
mailing list