WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
199882
Force useLLInt to true on arm64_32
https://bugs.webkit.org/show_bug.cgi?id=199882
Summary
Force useLLInt to true on arm64_32
Keith Miller
Reported
2019-07-17 14:23:26 PDT
Force useLLInt to true on arm64_32
Attachments
Patch
(1.53 KB, patch)
2019-07-17 14:26 PDT
,
Keith Miller
no flags
Details
Formatted Diff
Diff
View All
Add attachment
proposed patch, testcase, etc.
Keith Miller
Comment 1
2019-07-17 14:26:35 PDT
Created
attachment 374331
[details]
Patch
Keith Miller
Comment 2
2019-07-17 14:27:15 PDT
rdar://problem/53207586
Yusuke Suzuki
Comment 3
2019-07-17 14:35:05 PDT
Comment on
attachment 374331
[details]
Patch r=me
Mark Lam
Comment 4
2019-07-17 14:39:47 PDT
Comment on
attachment 374331
[details]
Patch The reason we went with an assertion failure before instead of auto-forcing the LLInt on is so that we don’t end up silently running misconfigured tests, and thinking everything is dandy when we didn’t actually run with the intended config. The better fix is to skip those tests if the JIT is not available for a certain target. I’m ok with landing this patch as a workaround until we can implement the better fix. If you want to do that, please #if the change on arm64_32, and add a FiXME so that it can be properly fixed later. r=me conditional on the above.
Keith Miller
Comment 5
2019-07-17 14:46:55 PDT
(In reply to Mark Lam from
comment #4
)
> Comment on
attachment 374331
[details]
> Patch > > The reason we went with an assertion failure before instead of auto-forcing > the LLInt on is so that we don’t end up silently running misconfigured > tests, and thinking everything is dandy when we didn’t actually run with the > intended config. The better fix is to skip those tests if the JIT is not > available for a certain target. I’m ok with landing this patch as a > workaround until we can implement the better fix. If you want to do that, > please #if the change on arm64_32, and add a FiXME so that it can be > properly fixed later.
We already do similar things for !ENABLE(JIT) where we force the JIT flag to false. The upside of this change is that if we enable JIT on arm64_32 the "fix" will stop working.
> > r=me conditional on the above.
WebKit Commit Bot
Comment 6
2019-07-17 16:12:08 PDT
Comment on
attachment 374331
[details]
Patch Clearing flags on attachment: 374331 Committed
r247543
: <
https://trac.webkit.org/changeset/247543
>
WebKit Commit Bot
Comment 7
2019-07-17 16:12:10 PDT
All reviewed patches have been landed. Closing bug.
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