[compiler] Remove CodeStub from CompilationInfo

[compiler] Remove CodeStub from CompilationInfo

The motivation for this is that CompilationInfo really shouldn't
explicitly know anything about CodeStubs. This is evident in
the TurboFan stubs pipeline, which only needs to pass down
information about Code::Flags to the code generator and not
any of the CallInterfaceDescriptor silliness that Hydrogen has
to push around, since TF has the Linkage class that
encapsulates everything that is needed for the stub ABI. So,
instead of threading CodeStub machinery through the TF stub
pipeline, it is now removed from CompilationInfo and replaced
by only the explicit bits needed both by the Crankshaft and
TF pipelines in code generation.

Committed: https://crrev.com/d1d0196473e6919b222bf98a9fd9093805a5c5bc
Cr-Commit-Position: refs/heads/master@{#33410}

[danno, 2016-01-18 16:03:29]

Description was changed from

==========
[compiler] Remove CodeStub from CompilationInfo
==========

to

==========
[compiler] Remove CodeStub from CompilationInfo

The motivation for this is that CompilationInfo really shouldn't
explicitly know anything about CodeStubs. This is evident in 
the TurboFan stubs pipeline, which only needs to pass down 
information about Code::Flags to the code generator and not
any of the CallInterfaceDescriptor silliness that Hydrogen has 
to push around, since TF has the Linkage class that
encapsulates everything that is needed for the stub ABI. So,
instead of CodeStub machinery through the TF stub pipeline, 
it is now removed from CompilationInfo and replaced only
but the explicit bits needed both by the Crankshaft and TF
pipelines in code generation.
==========

[danno, 2016-01-18 16:04:24]

Description was changed from

==========
[compiler] Remove CodeStub from CompilationInfo

The motivation for this is that CompilationInfo really shouldn't
explicitly know anything about CodeStubs. This is evident in 
the TurboFan stubs pipeline, which only needs to pass down 
information about Code::Flags to the code generator and not
any of the CallInterfaceDescriptor silliness that Hydrogen has 
to push around, since TF has the Linkage class that
encapsulates everything that is needed for the stub ABI. So,
instead of CodeStub machinery through the TF stub pipeline, 
it is now removed from CompilationInfo and replaced only
but the explicit bits needed both by the Crankshaft and TF
pipelines in code generation.
==========

to

==========
[compiler] Remove CodeStub from CompilationInfo

The motivation for this is that CompilationInfo really shouldn't
explicitly know anything about CodeStubs. This is evident in 
the TurboFan stubs pipeline, which only needs to pass down 
information about Code::Flags to the code generator and not
any of the CallInterfaceDescriptor silliness that Hydrogen has 
to push around, since TF has the Linkage class that
encapsulates everything that is needed for the stub ABI. So,
instead of threading CodeStub machinery through the TF stub
pipeline, it is now removed from CompilationInfo and replaced
by only the explicit bits needed both by the Crankshaft and
TF pipelines in code generation.
==========

[danno, 2016-01-18 16:04:58]

danno@chromium.org changed reviewers:
+ bmeurer@chromium.org, mstarzinger@chromium.org

[danno, 2016-01-18 16:04:58]

PTAL

[Benedikt Meurer, 2016-01-19 06:09:32]

bmeurer@chromium.org changed required reviewers:
+ mstarzinger@chromium.org

[Benedikt Meurer, 2016-01-19 06:09:32]

Nice, LGTM.

[Michael Starzinger, 2016-01-19 09:37:03]

LGTM, I like it, just nits.

https://codereview.chromium.org/1604543002/diff/120001/src/compiler/wasm-comp...
File src/compiler/wasm-compiler.cc (right):

https://codereview.chromium.org/1604543002/diff/120001/src/compiler/wasm-comp...
src/compiler/wasm-compiler.cc:1854: // TODO(titzer): this is technically a WASM
wrapper, not a wasm function.
nit: The TODO probably belongs to the flags computation, let's move it up two
lines.

https://codereview.chromium.org/1604543002/diff/120001/src/compiler/wasm-comp...
src/compiler/wasm-compiler.cc:1929: // TODO(titzer): this is technically a WASM
wrapper, not a wasm function.
nit: Likewise.

https://codereview.chromium.org/1604543002/diff/120001/test/cctest/compiler/t...
File test/cctest/compiler/test-linkage.cc (right):

https://codereview.chromium.org/1604543002/diff/120001/test/cctest/compiler/t...
test/cctest/compiler/test-linkage.cc:74: TEST(TestLinkageCodeStubIncoming) {
nit: This test no longer tests the "incoming" descriptor. Could we move the
content of this test down to "TestLinkageStubCall" a few lines below and then
just drop this test?

[danno, 2016-01-20 14:56:13]

Feedback addressed, landing.

https://codereview.chromium.org/1604543002/diff/120001/src/compiler/wasm-comp...
File src/compiler/wasm-compiler.cc (right):

https://codereview.chromium.org/1604543002/diff/120001/src/compiler/wasm-comp...
src/compiler/wasm-compiler.cc:1854: // TODO(titzer): this is technically a WASM
wrapper, not a wasm function.
On 2016/01/19 09:37:03, Michael Starzinger wrote:
> nit: The TODO probably belongs to the flags computation, let's move it up two
> lines.

Done.

https://codereview.chromium.org/1604543002/diff/120001/src/compiler/wasm-comp...
src/compiler/wasm-compiler.cc:1929: // TODO(titzer): this is technically a WASM
wrapper, not a wasm function.
On 2016/01/19 09:37:03, Michael Starzinger wrote:
> nit: Likewise.

Done.

https://codereview.chromium.org/1604543002/diff/120001/test/cctest/compiler/t...
File test/cctest/compiler/test-linkage.cc (right):

https://codereview.chromium.org/1604543002/diff/120001/test/cctest/compiler/t...
test/cctest/compiler/test-linkage.cc:74: TEST(TestLinkageCodeStubIncoming) {
On 2016/01/19 09:37:03, Michael Starzinger wrote:
> nit: This test no longer tests the "incoming" descriptor. Could we move the
> content of this test down to "TestLinkageStubCall" a few lines below and then
> just drop this test?

Done.

[danno, 2016-01-20 14:56:21]

The CQ bit was checked by danno@chromium.org

[danno, 2016-01-20 14:56:21]

The patchset sent to the CQ was uploaded after l-g-t-m from
bmeurer@chromium.org, mstarzinger@chromium.org
Link to the patchset: https://codereview.chromium.org/1604543002/#ps140001
(title: "Review feedback")

[commit-bot: I haz the power, 2016-01-20 14:56:28]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1604543002/140001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1604543002/140001

[commit-bot: I haz the power, 2016-01-20 15:17:46]

Description was changed from

==========
[compiler] Remove CodeStub from CompilationInfo

The motivation for this is that CompilationInfo really shouldn't
explicitly know anything about CodeStubs. This is evident in 
the TurboFan stubs pipeline, which only needs to pass down 
information about Code::Flags to the code generator and not
any of the CallInterfaceDescriptor silliness that Hydrogen has 
to push around, since TF has the Linkage class that
encapsulates everything that is needed for the stub ABI. So,
instead of threading CodeStub machinery through the TF stub
pipeline, it is now removed from CompilationInfo and replaced
by only the explicit bits needed both by the Crankshaft and
TF pipelines in code generation.
==========

to

==========
[compiler] Remove CodeStub from CompilationInfo

The motivation for this is that CompilationInfo really shouldn't
explicitly know anything about CodeStubs. This is evident in 
the TurboFan stubs pipeline, which only needs to pass down 
information about Code::Flags to the code generator and not
any of the CallInterfaceDescriptor silliness that Hydrogen has 
to push around, since TF has the Linkage class that
encapsulates everything that is needed for the stub ABI. So,
instead of threading CodeStub machinery through the TF stub
pipeline, it is now removed from CompilationInfo and replaced
by only the explicit bits needed both by the Crankshaft and
TF pipelines in code generation.
==========

[commit-bot: I haz the power, 2016-01-20 15:17:47]

Committed patchset #8 (id:140001)

[commit-bot: I haz the power, 2016-01-20 15:18:18]

Description was changed from

==========
[compiler] Remove CodeStub from CompilationInfo

The motivation for this is that CompilationInfo really shouldn't
explicitly know anything about CodeStubs. This is evident in 
the TurboFan stubs pipeline, which only needs to pass down 
information about Code::Flags to the code generator and not
any of the CallInterfaceDescriptor silliness that Hydrogen has 
to push around, since TF has the Linkage class that
encapsulates everything that is needed for the stub ABI. So,
instead of threading CodeStub machinery through the TF stub
pipeline, it is now removed from CompilationInfo and replaced
by only the explicit bits needed both by the Crankshaft and
TF pipelines in code generation.
==========

to

==========
[compiler] Remove CodeStub from CompilationInfo

The motivation for this is that CompilationInfo really shouldn't
explicitly know anything about CodeStubs. This is evident in
the TurboFan stubs pipeline, which only needs to pass down
information about Code::Flags to the code generator and not
any of the CallInterfaceDescriptor silliness that Hydrogen has
to push around, since TF has the Linkage class that
encapsulates everything that is needed for the stub ABI. So,
instead of threading CodeStub machinery through the TF stub
pipeline, it is now removed from CompilationInfo and replaced
by only the explicit bits needed both by the Crankshaft and
TF pipelines in code generation.

Committed: https://crrev.com/d1d0196473e6919b222bf98a9fd9093805a5c5bc
Cr-Commit-Position: refs/heads/master@{#33410}
==========

[commit-bot: I haz the power, 2016-01-20 15:18:19]

Patchset 8 (id:??) landed as
https://crrev.com/d1d0196473e6919b222bf98a9fd9093805a5c5bc
Cr-Commit-Position: refs/heads/master@{#33410}