Skip to content

V
V8
Switch branch/tag
  1. 12 Jan, 2016 1 commit
    • hajimehoshi's avatar
      Add Add ExternalStringResourceBase::IsCompressible · 150887a1
      hajimehoshi authored 
      This CL introduces ExternalStringResourceBase::IsCompressible.

This CL is a preparation for CompressibleString, which can
be compressed for memory reduction in Blink. We've found that
JavaScript strings account for a relatively large part of Blink
memory usage, and we are now trying to replace JavaScript String/
AtomicString with CompressibleString.

When a string is compressed, the original char data is deleted
and V8 pointer cache becomes invalid. This CL introduces
isCompressible  property and if an external string's isCompressble
return true, this is stored short_external_*_map instead of
external_*_map so that V8 always requires the char pointer whenever
V8 needs the string data.

BUG=chromium:574317
LOG=n

Review URL: https://codereview.chromium.org/1490193002

Cr-Commit-Position: refs/heads/master@{#33224}
      150887a1
  3. 07 Jan, 2016 1 commit
  5. 22 Dec, 2015 4 commits
    • cbruni's avatar
      [proxies] Expose proxies in the API · 88b58591
      cbruni authored 
      Add API-accessors for [[ProxyTarget]], [[ProxyHandler]]. Additionally
create new proxies and revoke proxies via the API.

BUG=v8:1543
LOG=n

Review URL: https://codereview.chromium.org/1542943002

Cr-Commit-Position: refs/heads/master@{#33013}
      88b58591
    • littledan's avatar
      Reland of Add web compat workarounds for ES2015 RegExp semantics (patchset #3... · 424ef009
      littledan authored 
      Reland of Add web compat workarounds for ES2015 RegExp semantics (patchset #3 id:40001 of https://codereview.chromium.org/1543723002/ )

Unexpectedly, websites depend on doing feature testing with
RegExp.prototype.sticky and browser testing with RegExp.prototype.toString().
ES2015 newly throws exceptions for both of these. In order to enable shipping
new ES2015 semantics, this patch puts in narrow workarounds for those two
cases, keeping their old behavior. UseCounters are added for how often
those particular cases come up, so we can see if it can be deprecated.

This reland replaces problematic legacy const usage with var, to
avoid issues with nosnap builds.

R=yangguo
CC=bmeurer
BUG=v8:4637,v8:4617
LOG=Y
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_chromium_rel_ng;tryserver.blink:linux_blink_rel

Review URL: https://codereview.chromium.org/1545633002

Cr-Commit-Position: refs/heads/master@{#33002}
      424ef009
    • bmeurer's avatar
      Revert of Add web compat workarounds for ES2015 RegExp semantics (patchset #3... · 08a1d1a2
      bmeurer authored 
      Revert of Add web compat workarounds for ES2015 RegExp semantics (patchset #3 id:40001 of https://codereview.chromium.org/1543723002/ )

Reason for revert:
Breaks nosnap: http://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20nosnap/builds/5883

Original issue's description:
> Add web compat workarounds for ES2015 RegExp semantics
>
> Unexpectedly, websites depend on doing feature testing with
> RegExp.prototype.sticky and browser testing with RegExp.prototype.toString().
> ES2015 newly throws exceptions for both of these. In order to enable shipping
> new ES2015 semantics, this patch puts in narrow workarounds for those two
> cases, keeping their old behavior. UseCounters are added for how often
> those particular cases come up, so we can see if it can be deprecated.
>
> R=yangguo
> BUG=v8:4637,v8:4617
> LOG=Y
> CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_chromium_rel_ng;tryserver.blink:linux_blink_rel
>
> Committed: https://crrev.com/98f819c3e0c92d54a306cdacadda73cf96d21b52
> Cr-Commit-Position: refs/heads/master@{#32997}

TBR=yangguo@google.com,yangguo@chromium.org,littledan@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=v8:4637,v8:4617

Review URL: https://codereview.chromium.org/1546493003

Cr-Commit-Position: refs/heads/master@{#32999}
      08a1d1a2
    • littledan's avatar
      Add web compat workarounds for ES2015 RegExp semantics · 98f819c3
      littledan authored 
      Unexpectedly, websites depend on doing feature testing with
RegExp.prototype.sticky and browser testing with RegExp.prototype.toString().
ES2015 newly throws exceptions for both of these. In order to enable shipping
new ES2015 semantics, this patch puts in narrow workarounds for those two
cases, keeping their old behavior. UseCounters are added for how often
those particular cases come up, so we can see if it can be deprecated.

R=yangguo
BUG=v8:4637,v8:4617
LOG=Y
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_chromium_rel_ng;tryserver.blink:linux_blink_rel

Review URL: https://codereview.chromium.org/1543723002

Cr-Commit-Position: refs/heads/master@{#32997}
      98f819c3
  7. 20 Dec, 2015 1 commit
  9. 17 Dec, 2015 1 commit
  11. 16 Dec, 2015 1 commit
    • akodat's avatar
      Add Isolate::DiscardThreadSpecificMetadata method to embedder API. · aeb8073c
      akodat authored 
      If many threads use the same Isolate (or many Isolates) and then
terminate, their PerIsolateThreadData objects are never cleaned
up, resulting in a slow memory leak and, worse, the
PerIsolateThreadData chain getting larger and larger, adversely
affecting performance.

In this situation, embedders will now be encouraged to apply
DiscardThreadSpecificMetadata against any Isolate a thread is
done with, especially if the thread is about to terminate.

Note that it is harmless to run DiscardThreadSpecificMetadata
against an Isolate for which a thread has no thread data and
per-Isolate thread data can be reestablished if a thread starts
using an Isolate again after running DiscardThreadSpecificMetadata
against it.

It is, however, an embedder error to run
DiscardThreadSpecificMetadata against an Isolate in thread with a
Locker for the Isolate in the stack or against an Entered Isolate.

This change cannot cause any change in behavior in existing apps
as the only added coded can only be reached via the new
DiscardThreadSpecificMetadata method.

R=Jakob, jochen
BUG=

Review URL: https://codereview.chromium.org/1522703002

Cr-Commit-Position: refs/heads/master@{#32909}
      aeb8073c
  13. 11 Dec, 2015 2 commits
  15. 10 Dec, 2015 5 commits
  17. 08 Dec, 2015 2 commits
  19. 07 Dec, 2015 1 commit
  21. 04 Dec, 2015 2 commits
  23. 03 Dec, 2015 3 commits
  25. 02 Dec, 2015 2 commits
  27. 01 Dec, 2015 1 commit
  29. 30 Nov, 2015 3 commits
  31. 26 Nov, 2015 4 commits
  33. 25 Nov, 2015 4 commits
  35. 24 Nov, 2015 1 commit
  37. 19 Nov, 2015 1 commit
    • kozyatinskiy's avatar
      [V8] Unify get function name for debugging purpose · 89e859fb
      kozyatinskiy authored 
      Following logic is using for getting function name in JSFunction::GetDebugName:
1. if function has displayName and its type is string then use it
2. if function has defined property Function.name as value and its type string then use it
3. otherwise use SharedFunctionInfo::DebugName as functionName.

JSFunction::GetDebugName is exposed in V8 API and in FunctionMirror interface.

BUG=chromium:17356
R=yangguo@chromium.org,mstarzinger@chromium.org
LOG=Y

Review URL: https://codereview.chromium.org/1449473005

Cr-Commit-Position: refs/heads/master@{#32124}
      89e859fb