Bug 214986 - Compute number of PAC bits from what the OS says its address space is
Summary: Compute number of PAC bits from what the OS says its address space is
Status: REOPENED
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: WebKit Nightly Build
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Keith Miller
URL:
Keywords: InRadar
Depends on: 215065
Blocks:
  Show dependency tree / graph
 
Reported: 2020-07-30 13:41 PDT by Keith Miller
Modified: 2020-08-01 14:47 PDT (History)
10 users (show)

See Also:

Attachments
Patch (3.45 KB, patch)
2020-07-30 13:45 PDT, Keith Miller 		no flags Details | Formatted Diff | Diff
Patch for landing (3.33 KB, patch)
2020-07-30 14:39 PDT, Keith Miller 		no flags Details | Formatted Diff | Diff
Show Obsolete (1) View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Keith Miller 2020-07-30 13:41:36 PDT 
Compute number of PAC bits from what the OS says its address space is
Comment 1 Keith Miller 2020-07-30 13:45:09 PDT 
Created attachment 405609 [details]
Patch
Comment 2 Saam Barati 2020-07-30 14:11:57 PDT 
Comment on attachment 405609 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=405609&action=review

> Source/WTF/wtf/PtrTag.h:34
> +#if CPU(ARM64E) && __has_include(<mach/arm/vm_param.h>)

I feel like this should be DARWIN based, w/ an assert we have the header
Comment 3 Saam Barati 2020-07-30 14:15:43 PDT 
Comment on attachment 405609 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=405609&action=review

> Source/WTF/wtf/PtrTag.h:36
> +constexpr unsigned maximumNumberOfPointerAuthenticationBits = 64 - getMSBSetConstexpr(MACH_VM_MAX_ADDRESS_RAW);

is this off by 1?
Comment 4 Saam Barati 2020-07-30 14:36:21 PDT 
I suggested to Keith to just do 64 - OS_CONSTANT(EFFECTIVE_ADDRESS_WIDTH)
Comment 5 Keith Miller 2020-07-30 14:39:58 PDT 
Created attachment 405622 [details]
Patch for landing
Comment 6 Keith Miller 2020-07-30 14:40:05 PDT 
(In reply to Saam Barati from comment #4)
> I suggested to Keith to just do 64 - OS_CONSTANT(EFFECTIVE_ADDRESS_WIDTH)

Done.
Comment 7 EWS 2020-07-30 15:26:06 PDT 
Committed r265113: <https://trac.webkit.org/changeset/265113>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 405622 [details].
Comment 8 Radar WebKit Bug Importer 2020-07-30 15:27:24 PDT 
<rdar://problem/66350659>
Comment 9 Ryan Haddad 2020-07-30 16:24:22 PDT 
This change broke the Windows build, as EWS predicted:
C:\cygwin\worker\win10-release\build\WebKitBuild\Release\DerivedSources\ForwardingHeaders\wtf/CagedPtr.h(43,1): error C2220: the following warning is treated as an error [C:\cygwin\worker\win10-release\build\WebKitBuild\Release\Source\JavaScriptCore\LLIntOffsetsExtractor.vcxproj]
C:\cygwin\worker\win10-release\build\Source\JavaScriptCore\runtime\ArrayBuffer.h(61): message : see reference to class template instantiation 'WTF::CagedPtr<Gigacage::Primitive,void,true,WTF::DumbPtrTraits<T>>' being compiled [C:\cygwin\worker\win10-release\build\WebKitBuild\Release\Source\JavaScriptCore\LLIntOffsetsExtractor.vcxproj]
C:\cygwin\worker\win10-release\build\Source\JavaScriptCore\runtime\ArrayBuffer.h(61): message :         with [C:\cygwin\worker\win10-release\build\WebKitBuild\Release\Source\JavaScriptCore\LLIntOffsetsExtractor.vcxproj]
C:\cygwin\worker\win10-release\build\Source\JavaScriptCore\runtime\ArrayBuffer.h(61): message :         [ [C:\cygwin\worker\win10-release\build\WebKitBuild\Release\Source\JavaScriptCore\LLIntOffsetsExtractor.vcxproj]
C:\cygwin\worker\win10-release\build\Source\JavaScriptCore\runtime\ArrayBuffer.h(61): message :             T=void [C:\cygwin\worker\win10-release\build\WebKitBuild\Release\Source\JavaScriptCore\LLIntOffsetsExtractor.vcxproj]
C:\cygwin\worker\win10-release\build\Source\JavaScriptCore\runtime\ArrayBuffer.h(61): message :         ] [C:\cygwin\worker\win10-release\build\WebKitBuild\Release\Source\JavaScriptCore\LLIntOffsetsExtractor.vcxproj]
Comment 10 Keith Miller 2020-07-30 17:19:12 PDT 
(In reply to Ryan Haddad from comment #9)
> This change broke the Windows build, as EWS predicted:
> C:\cygwin\worker\win10-
> release\build\WebKitBuild\Release\DerivedSources\ForwardingHeaders\wtf/
> CagedPtr.h(43,1): error C2220: the following warning is treated as an error
> [C:\cygwin\worker\win10-
> release\build\WebKitBuild\Release\Source\JavaScriptCore\LLIntOffsetsExtractor
> .vcxproj]
> C:\cygwin\worker\win10-
> release\build\Source\JavaScriptCore\runtime\ArrayBuffer.h(61): message : see
> reference to class template instantiation
> 'WTF::CagedPtr<Gigacage::Primitive,void,true,WTF::DumbPtrTraits<T>>' being
> compiled
> [C:\cygwin\worker\win10-
> release\build\WebKitBuild\Release\Source\JavaScriptCore\LLIntOffsetsExtractor
> .vcxproj]
> C:\cygwin\worker\win10-
> release\build\Source\JavaScriptCore\runtime\ArrayBuffer.h(61): message :    
> with
> [C:\cygwin\worker\win10-
> release\build\WebKitBuild\Release\Source\JavaScriptCore\LLIntOffsetsExtractor
> .vcxproj]
> C:\cygwin\worker\win10-
> release\build\Source\JavaScriptCore\runtime\ArrayBuffer.h(61): message :    
> [
> [C:\cygwin\worker\win10-
> release\build\WebKitBuild\Release\Source\JavaScriptCore\LLIntOffsetsExtractor
> .vcxproj]
> C:\cygwin\worker\win10-
> release\build\Source\JavaScriptCore\runtime\ArrayBuffer.h(61): message :    
> T=void
> [C:\cygwin\worker\win10-
> release\build\WebKitBuild\Release\Source\JavaScriptCore\LLIntOffsetsExtractor
> .vcxproj]
> C:\cygwin\worker\win10-
> release\build\Source\JavaScriptCore\runtime\ArrayBuffer.h(61): message :    
> ]
> [C:\cygwin\worker\win10-
> release\build\WebKitBuild\Release\Source\JavaScriptCore\LLIntOffsetsExtractor
> .vcxproj]

Should be fixed with r265122.
Comment 11 WebKit Commit Bot 2020-08-01 14:47:03 PDT 
Re-opened since this is blocked by bug 215065