So was finally able to compile v8. I went back to basics on the args and it
compiled so i went through. my version of the build args and stripped some
of more specific ones out for issues i've had in the past getting it to
compile and at least one of them was the result of it.
One interesting caveat of having v8gen.py generate the args it set my host
to x64 to build the target for arm64 which caused a number of failures
related to the embedded snapshot it looked like, after correcting the host
to arm64 since i'm on an M3 Max chip it compiled.
in particular these had been added for issues related to thin archives
being built, an issue with trig functions and turning off the pointer
compression shared cage.
'use_custom_libcxx': 'false',
'v8_static_library': 'true',
'v8_enable_pointer_compression_shared_cage ': 'false',
"v8_use_libm_trig_functions": "false",
"icu_disable_thin_archive": "true",
I suspect the pointer compression one may be the culprit that caused
mksnapshot to crash since it deals with memory.
Hope it helps anyone else in the future.
On Monday, December 16, 2024 at 12:32:46 AM UTC+8 Ronald Fenner wrote:
> SO after a little more digging I dropped an output stmt to print the
> address, size and access variable being sent for both 12.5.97 and 12.5.96
> and there is a difference.
> 12.5.96 outputs
> 0x39ec00000000 134217728 0
> and a whole bunch of other addresses before it none with 3 all the others
> are 0,1,2 or 4
>
> 12.5.97 outputs
> 0x39ec00000000 134217728 3
> unlike 12.5.96 it's the only line to be output.
>
> So 12.5.97 is trying to set the memory to ReadWriteExecute where as
> 12.5.96 it was NoAccess.
>
> It looks like something in my OS doesn't like that access being set for
> that memory address.
> On Saturday, December 14, 2024 at 6:57:16 PM UTC+8 Ronald Fenner wrote:
>
>> I tracked it back to 12.5.97 tag but the commit that has code changes
>> is 536571f811e9a51756b4b71f471dd3436b97a6bf.
>>
>> Since it's been many months ago I'm wondering if this is some security
>> change or bug in Mac OSX 15.x as thats what i'm building on. I just updated
>> to 15.2 from 15.1.1 and it occurs on both versions.
>>
>> On Friday, December 13, 2024 at 7:14:17 PM UTC+8 Clemens Backes wrote:
>>
>>> We expect that if setting permissions on memory fails, it must be due to
>>> an out of memory situation (the ENOMEM error code is the integer 12).
>>> What you get is EACCES (13). This is unexpected.
>>>
>>> It's hard to tell why this fails from just that stack trace. You could
>>> help narrowing it down by running a bisection to figure out which change
>>> introduced this (see https://git-scm.com/docs/git-bisect).
>>>
>>> On Thu, Dec 12, 2024 at 4:22 PM 'Ronald Fenner' via v8-dev <
>>> v8-...@googlegroups.com> wrote:
>>>
>>>> I was updating my version of v8 lib since in relation to my previous
>>>> bug with cppgc was on a 12.4.254.15 and it was pointed out that there have
>>>> been changes like the wrapper descriptor being removed and new functions
>>>> to
>>>> wrap/unwrap a cppgc instance added to v8:Object.
>>>> During the build process when it hits the snapshot creation it throws
>>>> an error and prints the stack trace
>>>> [2123/2127] ACTION
>>>> //:run_mksnapshot_default(//build/toolchain/mac:clang_arm64)
>>>> FAILED: gen/embedded.S snapshot_blob.bin
>>>> python3 ../../tools/run.py ./mksnapshot --turbo_instruction_scheduling
>>>> --stress-turbo-late-spilling --target_os=mac --target_arch=arm64
>>>> --embedded_src gen/embedded.S --predictable --no-use-ic --embedded_variant
>>>> Default --random-seed 314159265 --startup_blob snapshot_blob.bin
>>>> --no-native-code-counters
>>>>
>>>>
>>>> #
>>>> # Fatal error in , line 0
>>>> # Check failed: 12 == (*__error()).
>>>> #
>>>> #
>>>> #
>>>> #FailureMessage Object: 0x16fd36658
>>>> ==== C stack trace ===============================
>>>>
>>>> 0 mksnapshot 0x0000000100a2fa8c
>>>> v8::base::debug::StackTrace::StackTrace() + 24
>>>> 1 mksnapshot 0x0000000100a339a0
>>>> v8::platform::(anonymous namespace)::PrintStackTrace() + 24
>>>> 2 mksnapshot 0x0000000100a26448
>>>> V8_Fatal(char const*, ...) + 356
>>>> 3 mksnapshot 0x0000000100a2e2b0
>>>> v8::base::OS::SetPermissions(void*, unsigned long,
>>>> v8::base::OS::MemoryPermission) + 284
>>>> 4 mksnapshot 0x0000000100a205d0
>>>> v8::base::BoundedPageAllocator::SetPermissions(void*, unsigned long,
>>>> v8::PageAllocator::Permission) + 32
>>>> 5 mksnapshot 0x00000001001ff344
>>>> v8::internal::CodeRange::InitReservation(v8::PageAllocator*, unsigned
>>>> long,
>>>> bool) + 584
>>>> 6 mksnapshot 0x0000000100317ba8
>>>> v8::internal::(anonymous
>>>> namespace)::InitCodeRangeOnce(std::__1::unique_ptr<v8::internal::CodeRange,
>>>>
>>>> std::__1::default_delete<v8::internal::CodeRange>>*, v8::PageAllocator*,
>>>> unsigned long, bool) + 128
>>>> 7 mksnapshot 0x0000000100a2abb0
>>>> v8::base::CallOnceImpl(std::__1::atomic<unsigned char>*,
>>>> std::__1::function<void ()>) + 68
>>>> 8 mksnapshot 0x0000000100317ac8
>>>> v8::internal::IsolateGroup::EnsureCodeRange(unsigned long) + 136
>>>> 9 mksnapshot 0x000000010026e990
>>>> v8::internal::Heap::SetUp(v8::internal::LocalHeap*) + 180
>>>> 10 mksnapshot 0x00000001001d6724
>>>> v8::internal::Isolate::Init(v8::internal::SnapshotData*,
>>>> v8::internal::SnapshotData*, v8::internal::SnapshotData*, bool) + 1684
>>>> 11 mksnapshot 0x00000001001d6084
>>>> v8::internal::Isolate::InitWithoutSnapshot() + 28
>>>> 12 mksnapshot 0x0000000100696418
>>>> v8::internal::SnapshotCreatorImpl::InitInternal(v8::StartupData const*) +
>>>> 96
>>>> 13 mksnapshot 0x00000001006967f4
>>>> v8::internal::SnapshotCreatorImpl::SnapshotCreatorImpl(v8::internal::Isolate*,
>>>>
>>>> v8::Isolate::CreateParams const&) + 276
>>>> 14 mksnapshot 0x00000001000bad98
>>>> v8::SnapshotCreator::SnapshotCreator(v8::Isolate*,
>>>> v8::Isolate::CreateParams const&) + 48
>>>> 15 mksnapshot 0x00000001000afb2c main +
>>>> 1392
>>>> 16 dyld 0x000000018d51c274 start +
>>>> 2840
>>>> Return code is -5
>>>>
>>>> this is for the latest stable release of 13.1.201.16 that ships in the
>>>> latest chromium version.
>>>>
>>>> I'm assuming some flags have changed and may be causing the issue.
>>>> My build args are for debug
>>>> is_debug = true
>>>> is_component_build = false
>>>> v8_monolithic = false
>>>> v8_use_external_startup_data = true
>>>> v8_enable_backtrace = true
>>>> v8_enable_slow_dchecks = true
>>>> v8_optimized_debug = false
>>>> use_custom_libcxx = false
>>>> v8_static_library = true
>>>> v8_enable_pointer_compression_shared_cage = false
>>>> enable_iterator_debugging = true
>>>> v8_use_libm_trig_functions = false
>>>> icu_disable_thin_archive = true
>>>> v8_target_cpu = "arm64"
>>>> target_cpu = "arm64"
>>>>
>>>> and release is
>>>> is_debug = false
>>>> is_component_build = false
>>>> v8_monolithic = false
>>>> v8_use_external_startup_data = true
>>>> dcheck_always_on = false
>>>> use_custom_libcxx = false
>>>> v8_static_library = true
>>>> v8_enable_pointer_compression_shared_cage = false
>>>> v8_use_libm_trig_functions = false
>>>> icu_disable_thin_archive = true
>>>> v8_target_cpu = "arm64"
>>>> target_cpu = "arm64"
>>>>
>>>> This occurs in both build of it. I also tried 13.1.201.18 which gives
>>>> the same error with a slightly more informative assert adding something
>>>> like 13 != 12
>>>>
>>>> --
>>>> --
>>>> v8-dev mailing list
>>>> v8-...@googlegroups.com
>>>> http://groups.google.com/group/v8-dev
>>>> ---
>>>> You received this message because you are subscribed to the Google
>>>> Groups "v8-dev" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>> an email to v8-dev+un...@googlegroups.com.
>>>> To view this discussion visit
>>>> https://groups.google.com/d/msgid/v8-dev/15094b1c-60e6-47e9-8e24-0367293ee326n%40googlegroups.com
>>>>
>>>> <https://groups.google.com/d/msgid/v8-dev/15094b1c-60e6-47e9-8e24-0367293ee326n%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>> .
>>>>
>>>
>>>
>>> --
>>>
>>> Clemens Backes
>>>
>>> Software Engineer
>>>
>>> clem...@google.com
>>>
>>> Google Germany GmbH
>>>
>>> Erika-Mann-Straße 33
>>>
>>> 80636 München
>>>
>>> Geschäftsführer: Paul Manicle, Liana Sebastian
>>>
>>> Registergericht und -nummer: Hamburg, HRB 86891
>>>
>>> Sitz der Gesellschaft: Hamburg
>>>
>>> Diese E-Mail ist vertraulich. Falls sie diese fälschlicherweise erhalten
>>> haben sollten, leiten Sie diese bitte nicht an jemand anderes weiter,
>>> löschen Sie alle Kopien und Anhänge davon und lassen Sie mich bitte wissen,
>>> dass die E-Mail an die falsche Person gesendet wurde.
>>>
>>>
>>> This e-mail is confidential. If you received this communication by
>>> mistake, please don't forward it to anyone else, please erase all copies
>>> and attachments, and please let me know that it has gone to the wrong
>>> person.
>>>
>>>
>>>
--
--
v8-dev mailing list
v8-dev@googlegroups.com
http://groups.google.com/group/v8-dev
---
You received this message because you are subscribed to the Google Groups
"v8-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to v8-dev+unsubscr...@googlegroups.com.
To view this discussion visit
https://groups.google.com/d/msgid/v8-dev/2fb4087d-223e-4828-b17c-62d3fb842f91n%40googlegroups.com.
