For what you describe, building a release DLL via MSVC should work fine 
(however Debug builds still have issues).

See https://groups.google.com/forum/#!topic/v8-users/cvFGONOg_BY and 
https://groups.google.com/forum/#!topic/v8-users/mmwWxpb64_I for prior 
discussions.

 - Bill

On Tuesday, February 25, 2020 at 9:37:55 PM UTC-8, Ben Ernst wrote:
>
> Clemens, Dimitar,
> This use case is very important to me also. I think there are a few 
> threads in this group on the topic. For my part, I have complex application 
> with many third party dependencies. Those dependencies do not support 
> clang. So I can't compile my application with clang. I also call V8 code 
> from several different first-party dynamically linked libraries within my 
> application. Using a static linked build of V8 doesn't work in this 
> situation. There is static data in V8 that ends up in multiple DLLs, and 
> messes things up in horrible ways. I can re-architect to only link to V8 
> from one DLL to fix this. Since dynamic-linked V8 is no longer supported, I 
> am slowly working on this, but it's a complex and painful task.
> Cheers,
> Ben Ernst
>
> On Wednesday, 26 February 2020 05:17:47 UTC+10:30, Clemens Backes wrote:
>>
>> On Tue, Feb 25, 2020 at 7:01 PM <[email protected]> wrote:
>>
>>> Hi Clemens,
>>>
>>> I don't see any discussion in the bug  https://crbug.com/v8/9555 about 
>>> the support for MSVC.
>>> Is there a place, where we can vote or explain our use case for not 
>>> breaking the usage of v8 from MSVC?
>>>
>>
>> You are right, the issue only raises the question but does not answer it.
>> This is a product decision that would best be made by +Michael Hablich.
>>
>> I think this mailing list would actually be a good place to explain your 
>> use case. Alternatively, a v8 issue would also do, since depending on the 
>> outcome there will be infrastructure work to do on our side.
>>
>> Cheers,
>> Clemens
>>
>>
>>> Best regards,
>>>  Dimitar
>>>
>>> On Monday, July 29, 2019 at 1:43:30 PM UTC+3, Clemens Backes wrote:
>>>>
>>>> Note: This is being tracked as https://crbug.com/v8/9555 now. I am 
>>>> just landing a fix for this <https://crrev.com/c/1720669>, and on the 
>>>> bug we are discussing whether we care enough about MSVC debug to add a bot 
>>>> to the waterfall.
>>>>
>>>> On Mon, Jul 29, 2019 at 12:36 PM Jakob Kummerow <[email protected]> 
>>>> wrote:
>>>>
>>>>> if someone has the magic args.gn needed, I'd appreciate it.
>>>>>
>>>>>
>>>>> For the record, you can easily get any build bot's args.gn: click 
>>>>> through to a specific build (in this case, e.g. 
>>>>> https://ci.chromium.org/p/v8/builders/try/v8_win64_msvc_compile_rel/b8906608002392908144),
>>>>>  
>>>>> and in the "Steps and Logs" section, look for "gn_args".
>>>>>
>>>>> Is this still supported?
>>>>>
>>>>>
>>>>> I'm not sure what the official policy is at the moment. I think at the 
>>>>> very least, we would happily accept patches that maintain the MSVC build. 
>>>>> As you have figured out, we have Release-mode buildbot coverage for it; I 
>>>>> guess we assume that MSVC Debug builds are used and/or break rarely 
>>>>> enough 
>>>>> that spending bot resources on them is hard to justify.
>>>>>
>>>>>
>>>>> On Fri, Jul 26, 2019 at 1:05 AM billti via v8-dev <
>>>>> [email protected]> wrote:
>>>>>
>>>>>> A colleague tracked it down to the CSA_ASSERT macro in 
>>>>>> code-stub-assembler.h failing in debug builds (looks like the tryjob 
>>>>>> only 
>>>>>> builds rel for MSVC).
>>>>>>
>>>>>> This change 16 days ago seems to have broken it: 
>>>>>> https://chromium-review.googlesource.com/c/v8/v8/+/1693002 . If I 
>>>>>> check out the parent commit (0fd3766) I can build debug fine in MSVC 
>>>>>> (after 
>>>>>> one more trivial MSVC fix in a DCHECK in wasm-module-builder.cc). It 
>>>>>> still 
>>>>>> gives the "/Zc:twoPhase" compiler option warning, but hopefully that's 
>>>>>> benign.
>>>>>>
>>>>>> I'll log a bug.
>>>>>>
>>>>>> On Thursday, July 25, 2019 at 3:38:20 PM UTC-7, [email protected] 
>>>>>> wrote:
>>>>>>>
>>>>>>> Is this still supported?
>>>>>>>
>>>>>>> Trying to build the tip of master with 'is_clang = false' in args.gn 
>>>>>>> (and various other attempted options) is giving lots of different 
>>>>>>> warnings 
>>>>>>> and failures, e.g.
>>>>>>>
>>>>>>>
>>>>>>> cl : Command line warning D9002 : ignoring unknown option 
>>>>>>> '/Zc:twoPhase'
>>>>>>> .etc.
>>>>>>> [501/1829] CXX 
>>>>>>> obj/torque_generated_initializers/proxy-set-property-tq-csa.obj 
>>>>>>> FAILED: 
>>>>>>> obj/torque_generated_initializers/proxy-set-property-tq-csa.obj
>>>>>>> .etc.
>>>>>>> C:\src\v8\v8\src/codegen/code-stub-assembler.h(493): error C2059: 
>>>>>>> syntax error: ','
>>>>>>> .etc.
>>>>>>> C:\src\v8\v8\src/objects/tagged-impl-inl.h(18): fatal error C1003: 
>>>>>>> error count exceeds 100; stopping compilation
>>>>>>> .etc.
>>>>>>>
>>>>>>> It's the first time I've tried with MSVC, so may be doing something 
>>>>>>> wrong. (I have no issues with the default clang-based builds on the 
>>>>>>> same 
>>>>>>> machine).
>>>>>>>
>>>>>>> I can see from task manager and the log output it is using the right 
>>>>>>> MSVC compiler (I tried with VS 2017 and VS 2019 installs), e.g.
>>>>>>>
>>>>>>> > ninja -t msvc -e environment.x64 -- "C:\Program Files 
>>>>>>> (x86)\Microsoft Visual 
>>>>>>> Studio\2019\Enterprise\VC\Tools\MSVC\14.22.27905\...
>>>>>>>
>>>>>>> I see some of the V8 tryjobs build with MSVC (e.g. 
>>>>>>> https://ci.chromium.org/p/v8/builders/try/v8_win64_msvc_compile_rel) 
>>>>>>> and seem to be working, so if someone has the magic args.gn needed, 
>>>>>>> I'd appreciate it.
>>>>>>>
>>>>>>> Thanks,
>>>>>>>
>>>>>>>  - Bill
>>>>>>>
>>>>>> -- 
>>>>>> -- 
>>>>>> v8-dev mailing list
>>>>>> [email protected]
>>>>>> 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 [email protected].
>>>>>> To view this discussion on the web visit 
>>>>>> https://groups.google.com/d/msgid/v8-dev/61cdf5a4-a276-4fcb-9bd7-ed37f1c40d49%40googlegroups.com
>>>>>>  
>>>>>> <https://groups.google.com/d/msgid/v8-dev/61cdf5a4-a276-4fcb-9bd7-ed37f1c40d49%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>>>> .
>>>>>>
>>>>> -- 
>>>>> -- 
>>>>> v8-dev mailing list
>>>>> [email protected]
>>>>> 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 [email protected].
>>>>> To view this discussion on the web visit 
>>>>> https://groups.google.com/d/msgid/v8-dev/CAKSzg3Q1cq4Vwzs3mS-bFjTHHKFbwSWzhhGwjqANA2V56E%3DvdQ%40mail.gmail.com
>>>>>  
>>>>> <https://groups.google.com/d/msgid/v8-dev/CAKSzg3Q1cq4Vwzs3mS-bFjTHHKFbwSWzhhGwjqANA2V56E%3DvdQ%40mail.gmail.com?utm_medium=email&utm_source=footer>
>>>>> .
>>>>>
>>>>
>>>>
>>>> -- 
>>>>
>>>> Clemens Hammacher
>>>>
>>>> Software Engineer
>>>>
>>>> [email protected]
>>>>
>>>>
>>>> Google Germany GmbH
>>>>
>>>> Erika-Mann-Straße 33
>>>>
>>>> 80636 München
>>>>
>>>> Geschäftsführer: Paul Manicle, Halimah DeLaine Prado
>>>>
>>>> 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
>>> [email protected]
>>> 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 [email protected].
>>> To view this discussion on the web visit 
>>> https://groups.google.com/d/msgid/v8-dev/04cceab7-c95c-4846-bfab-33ce84857ba0%40googlegroups.com
>>>  
>>> <https://groups.google.com/d/msgid/v8-dev/04cceab7-c95c-4846-bfab-33ce84857ba0%40googlegroups.com?utm_medium=email&utm_source=footer>
>>> .
>>>
>>
>>
>> -- 
>>
>> Clemens Backes
>>
>> Software Engineer
>>
>> [email protected]
>>
>>
>> Google Germany GmbH
>>
>> Erika-Mann-Straße 33
>>
>> 80636 München
>>
>> Geschäftsführer: Paul Manicle, Halimah DeLaine Prado
>>
>> 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
[email protected]
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 [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/v8-dev/7918eb46-8043-46f0-ab00-2b2bd8a26ba1%40googlegroups.com.

Reply via email to