[compiler] Allow for StackSlots of arbitrary size

[compiler] Allow for StackSlots of arbitrary size

This will be used to pass parameters of wasm functions to the wasm
interpreter. All of them need to be packed into one buffer, which is
then passed to the interpreter.

R=ahaas@chromium.org, titzer@chromium.org
BUG=v8:5822
Review-Url: https://codereview.chromium.org/2624183002
Cr-Commit-Position: refs/heads/master@{#42239}
Committed: https://chromium.googlesource.com/v8/v8/+/7a3366fbcd556ba6e5a714b34e9ae2085446d987

[Clemens Hammacher, 2017-01-11 12:33:14]

The CQ bit was checked by clemensh@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-01-11 12:33:16]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-01-11 12:58:47]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-01-11 12:58:48]

Dry run: This issue passed the CQ dry run.

[ahaas, 2017-01-11 15:10:28]

lgtm

[titzer, 2017-01-11 15:13:40]

https://codereview.chromium.org/2624183002/diff/1/src/compiler/frame.h
File src/compiler/frame.h (right):

https://codereview.chromium.org/2624183002/diff/1/src/compiler/frame.h#newcod...
src/compiler/frame.h:137: if (kPointerSize == 4 && width >= 8 && (width & 7) ==
0 &&
Why not just alignSize = (width & 7) == 0 ? 8 : kPointerSize?

And actually, you probably want to align to 16 bytes 16-byte aligned sizes too
(SIMD).

[titzer, 2017-01-11 15:15:05]

lgtm with comment

https://codereview.chromium.org/2624183002/diff/1/src/compiler/machine-operat...
File src/compiler/machine-operator.cc (right):

https://codereview.chromium.org/2624183002/diff/1/src/compiler/machine-operat...
src/compiler/machine-operator.cc:461: #define STACK_SLOT_CACHED_SIZES_LIST(V)
V(1) V(2) V(4) V(8) V(16)
I'd just cache 4, 8, and 16, because smaller sized slots are probably rare.

[Clemens Hammacher, 2017-01-11 15:41:12]

The CQ bit was checked by clemensh@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-01-11 15:41:17]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Clemens Hammacher, 2017-01-11 15:41:34]

https://codereview.chromium.org/2624183002/diff/1/src/compiler/frame.h
File src/compiler/frame.h (right):

https://codereview.chromium.org/2624183002/diff/1/src/compiler/frame.h#newcod...
src/compiler/frame.h:137: if (kPointerSize == 4 && width >= 8 && (width & 7) ==
0 &&
On 2017/01/11 at 15:13:40, titzer wrote:
> Why not just alignSize = (width & 7) == 0 ? 8 : kPointerSize?
> 
> And actually, you probably want to align to 16 bytes 16-byte aligned sizes too
(SIMD).

It wasn't there before, but sounds like a good idea. Done.
PTAL.

https://codereview.chromium.org/2624183002/diff/1/src/compiler/machine-operat...
File src/compiler/machine-operator.cc (right):

https://codereview.chromium.org/2624183002/diff/1/src/compiler/machine-operat...
src/compiler/machine-operator.cc:461: #define STACK_SLOT_CACHED_SIZES_LIST(V)
V(1) V(2) V(4) V(8) V(16)
On 2017/01/11 at 15:15:05, titzer wrote:
> I'd just cache 4, 8, and 16, because smaller sized slots are probably rare.

Done.

[titzer, 2017-01-11 15:44:39]

On 2017/01/11 15:41:34, Clemens Hammacher wrote:
> https://codereview.chromium.org/2624183002/diff/1/src/compiler/frame.h
> File src/compiler/frame.h (right):
> 
>
https://codereview.chromium.org/2624183002/diff/1/src/compiler/frame.h#newcod...
> src/compiler/frame.h:137: if (kPointerSize == 4 && width >= 8 && (width & 7)
==
> 0 &&
> On 2017/01/11 at 15:13:40, titzer wrote:
> > Why not just alignSize = (width & 7) == 0 ? 8 : kPointerSize?
> > 
> > And actually, you probably want to align to 16 bytes 16-byte aligned sizes
too
> (SIMD).
> 
> It wasn't there before, but sounds like a good idea. Done.
> PTAL.
> 
>
https://codereview.chromium.org/2624183002/diff/1/src/compiler/machine-operat...
> File src/compiler/machine-operator.cc (right):
> 
>
https://codereview.chromium.org/2624183002/diff/1/src/compiler/machine-operat...
> src/compiler/machine-operator.cc:461: #define STACK_SLOT_CACHED_SIZES_LIST(V)
> V(1) V(2) V(4) V(8) V(16)
> On 2017/01/11 at 15:15:05, titzer wrote:
> > I'd just cache 4, 8, and 16, because smaller sized slots are probably rare.
> 
> Done.

lgtm

[commit-bot: I haz the power, 2017-01-11 16:08:54]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-01-11 16:08:55]

Dry run: This issue passed the CQ dry run.

[Clemens Hammacher, 2017-01-11 16:09:11]

The CQ bit was checked by clemensh@chromium.org

[Clemens Hammacher, 2017-01-11 16:09:11]

The patchset sent to the CQ was uploaded after l-g-t-m from ahaas@chromium.org
Link to the patchset: https://codereview.chromium.org/2624183002/#ps20001
(title: "Address comments")

[commit-bot: I haz the power, 2017-01-11 16:09:14]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-01-11 16:10:52]

CQ is committing da patch.

Bot data: {"patchset_id": 20001, "attempt_start_ts": 1484150951495750,
"parent_rev": "e904894915ff88c5f500fbc1bd816250885ae2d3", "commit_rev":
"7a3366fbcd556ba6e5a714b34e9ae2085446d987"}

[commit-bot: I haz the power, 2017-01-11 16:10:57]

Description was changed from

==========
[compiler] Allow for StackSlots of arbitrary size

This will be used to pass parameters of wasm functions to the wasm
interpreter. All of them need to be packed into one buffer, which is
then passed to the interpreter.

R=ahaas@chromium.org, titzer@chromium.org
BUG=v8:5822
==========

to

==========
[compiler] Allow for StackSlots of arbitrary size

This will be used to pass parameters of wasm functions to the wasm
interpreter. All of them need to be packed into one buffer, which is
then passed to the interpreter.

R=ahaas@chromium.org, titzer@chromium.org
BUG=v8:5822

Review-Url: https://codereview.chromium.org/2624183002
Cr-Commit-Position: refs/heads/master@{#42239}
Committed:
https://chromium.googlesource.com/v8/v8/+/7a3366fbcd556ba6e5a714b34e9ae208544...
==========

[commit-bot: I haz the power, 2017-01-11 16:10:59]

Committed patchset #2 (id:20001) as
https://chromium.googlesource.com/v8/v8/+/7a3366fbcd556ba6e5a714b34e9ae208544...