8266609: AArch64: include FP/LR space in LIR_Assembler::initial_frame_size_in_bytes() #3898

nick-arm · 2021-05-06T10:03:14Z

LIR_Assembler::initial_frame_size_in_bytes() returns the frame size
without the additional 2*wordSize needed to store FP/LR (i.e. just the
space required for the C1 locals). When we pass this value to
MacroAssembler build_frame and remove_frame we need to remember to add
back the 2*wordSize. This patch changes initial_frame_size_in_bytes()
to return the full frame size including the space for FP/LR. The PPC
and S390 ports already do this, and it also matches the behaviour of
C->output()->frame_size_in_bytes() in C2.

This change may seem a bit trivial in mainline but the Valhalla lworld
branch makes more calls to build_frame/remove_frame in C1 and keeping
track of whether "framesize" is with or without FP/LR quickly becomes
confusing.

Tested tier1 on AArch64. The only use of this function is as input to
C1_MacroAssembler build_frame() and remove_frame() so seems safe.

Progress

Change must not contain extraneous whitespace
Commit message must refer to an issue
Change must be properly reviewed

Issue

JDK-8266609: AArch64: include FP/LR space in LIR_Assembler::initial_frame_size_in_bytes()

Reviewers

Andrew Haley (@theRealAph - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.java.net/jdk pull/3898/head:pull/3898
$ git checkout pull/3898

Update a local copy of the PR:
$ git checkout pull/3898
$ git pull https://git.openjdk.java.net/jdk pull/3898/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 3898

View PR using the GUI difftool:
$ git pr show -t 3898

Using diff file

Download this PR as a diff file:
https://git.openjdk.java.net/jdk/pull/3898.diff

…_size_in_bytes() LIR_Assembler::initial_frame_size_in_bytes() returns the frame size without the additional 2*wordSize needed to store FP/LR (i.e. just the space required for the C1 locals). When we pass this value to MacroAssembler build_frame and remove_frame we need to remember to add back the 2*wordSize. This patch changes initial_frame_size_in_bytes() to return the full frame size including the space for FP/LR. The PPC and S390 ports already do this, and it also matches the behaviour of C->output()->frame_size_in_bytes() in C2. This change may seem a bit trivial in mainline but the Valhalla lworld branch makes more calls to build_frame/remove_frame in C1 and keeping track of whether "framesize" is with or without FP/LR quickly becomes confusing. Tested tier1 on AArch64. The only use of this function is as input to C1_MacroAssembler build_frame() and remove_frame() so seems safe.

bridgekeeper · 2021-05-06T10:03:39Z

👋 Welcome back ngasson! A progress list of the required criteria for merging this PR into master will be added to the body of your pull request. There are additional pull request commands available for use with this pull request.

openjdk · 2021-05-06T10:05:54Z

@nick-arm The following label will be automatically applied to this pull request:

hotspot

When this pull request is ready to be reviewed, an "RFR" email will be sent to the corresponding mailing list. If you would like to change these labels, use the /label pull request command.

mlbridge · 2021-05-06T10:09:08Z

Webrevs

00: Full (912ea364)

theRealAph

This calculation is the same as x86. Can you explain why it has to be changed on AArch64, but should not be changed on x86?

theRealAph · 2021-05-06T10:28:32Z

src/hotspot/cpu/aarch64/c1_LIRAssembler_aarch64.cpp

-
-  // subtract two words to account for return address and link
-  return (frame_map()->framesize() - (2*VMRegImpl::slots_per_word))  * VMRegImpl::stack_slot_size;
+  return in_bytes(frame_map()->framesize_in_bytes());


Umm, really? framesize_in_bytes() returns a value that has to be converted to bytes?

framesize_in_bytes() returns a struct ByteSize that needs to be converted to an int with in_bytes(). The extra _in_bytes seems a bit redundant but it's not my name :-)

nick-arm · 2021-05-07T01:48:43Z

This calculation is the same as x86. Can you explain why it has to be changed on AArch64, but should not be changed on x86?

On AArch64 C1_MacroAssembler::build_frame() delegates to MacroAssembler::build_frame() which expects the total frame size. On x86 we build the frame directly in C1_MacroAssembler by doing roughly __ push(rbp); __ decrement(rsp, frame_size_in_bytes); so it's natural that frame_size_in_bytes shouldn't include the space for the frame pointer and the return address (which was already pushed by the call instruction). If we made this change on x86 we'd need to subtract 2*wordSize from each increment/decrement of rsp which seems like a net loss.

theRealAph · 2021-05-07T08:32:41Z

This calculation is the same as x86. Can you explain why it has to be changed on AArch64, but should not be changed on x86?

On AArch64 C1_MacroAssembler::build_frame() delegates to MacroAssembler::build_frame() which expects the total frame size. On x86 we build the frame directly in C1_MacroAssembler by doing roughly __ push(rbp); __ decrement(rsp, frame_size_in_bytes); so it's natural that frame_size_in_bytes shouldn't include the space for the frame pointer and the return address (which was already pushed by the call instruction). If we made this change on x86 we'd need to subtract 2*wordSize from each increment/decrement of rsp which seems like a net loss.

OK, makes snse.

openjdk · 2021-05-07T08:34:14Z

@nick-arm This change now passes all automated pre-integration checks.

ℹ️ This project also has non-automated pre-integration requirements. Please see the file CONTRIBUTING.md for details.

After integration, the commit message for the final commit will be:

8266609: AArch64: include FP/LR space in LIR_Assembler::initial_frame_size_in_bytes()

Reviewed-by: aph

You can use pull request commands such as /summary, /contributor and /issue to adjust it as needed.

At the time when this comment was updated there had been 20 new commits pushed to the master branch:

ebb68d2: 8049700: Deprecate obsolete classes and methods in javax/swing/plaf/basic
3a474d9: 8265612: revise the help info for jmap histo command
c97f56c: 8266172: -Wstringop-overflow happens in vmError.cpp
43ad24f: 8265465: jcmd VM.cds should keep already dumped archive when exception happens
66191ff: 8266193: BasicJMapTest does not include testHistoParallel methods
36e5ad6: 8263236: runtime/os/TestTracePageSizes.java fails on old kernels
0ca86da: 8266002: vmTestbase/nsk/jvmti/ClassPrepare/classprep001 should skip events for unexpected classes
52f1db6: 8262002: java/lang/instrument/VerifyLocalVariableTableOnRetransformTest.sh failed with "TestCaseScaffoldException: DummyClassWithLVT did not match .class file"
04f7112: 8266293: Key protection using PBEWithMD5AndDES fails with "java.security.InvalidAlgorithmParameterException: Salt must be 8 bytes long"
a90b33a: 8266573: Make sure blackholes are tagged for all JVMCI paths
... and 10 more: https://git.openjdk.java.net/jdk/compare/a86ee9b3f370b59caea2ae78169d13498560cd8e...master

As there are no conflicts, your changes will automatically be rebased on top of these commits when integrating. If you prefer to avoid this automatic rebasing, please check the documentation for the /integrate command for further details.

➡️ To integrate this PR with the above commit message to the master branch, type /integrate in a new comment.

nick-arm · 2021-05-07T09:18:56Z

/integrate

openjdk · 2021-05-07T09:20:40Z

@nick-arm Since your change was applied there have been 20 commits pushed to the master branch:

ebb68d2: 8049700: Deprecate obsolete classes and methods in javax/swing/plaf/basic
3a474d9: 8265612: revise the help info for jmap histo command
c97f56c: 8266172: -Wstringop-overflow happens in vmError.cpp
43ad24f: 8265465: jcmd VM.cds should keep already dumped archive when exception happens
66191ff: 8266193: BasicJMapTest does not include testHistoParallel methods
36e5ad6: 8263236: runtime/os/TestTracePageSizes.java fails on old kernels
0ca86da: 8266002: vmTestbase/nsk/jvmti/ClassPrepare/classprep001 should skip events for unexpected classes
52f1db6: 8262002: java/lang/instrument/VerifyLocalVariableTableOnRetransformTest.sh failed with "TestCaseScaffoldException: DummyClassWithLVT did not match .class file"
04f7112: 8266293: Key protection using PBEWithMD5AndDES fails with "java.security.InvalidAlgorithmParameterException: Salt must be 8 bytes long"
a90b33a: 8266573: Make sure blackholes are tagged for all JVMCI paths
... and 10 more: https://git.openjdk.java.net/jdk/compare/a86ee9b3f370b59caea2ae78169d13498560cd8e...master

Your commit was automatically rebased without conflicts.

Pushed as commit 71b8ad4.

💡 You may see a message that your pull request was closed with unmerged commits. This can be safely ignored.

openjdk bot added the rfr Pull request is ready for review label May 6, 2021

openjdk bot added the hotspot hotspot-dev@openjdk.org label May 6, 2021

theRealAph reviewed May 6, 2021

View reviewed changes

theRealAph approved these changes May 7, 2021

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label May 7, 2021

openjdk bot closed this May 7, 2021

openjdk bot added integrated Pull request has been integrated and removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels May 7, 2021

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

8266609: AArch64: include FP/LR space in LIR_Assembler::initial_frame_size_in_bytes() #3898

8266609: AArch64: include FP/LR space in LIR_Assembler::initial_frame_size_in_bytes() #3898

Uh oh!

nick-arm commented May 6, 2021 •

edited by openjdk bot

Loading

Uh oh!

bridgekeeper bot commented May 6, 2021

Uh oh!

openjdk bot commented May 6, 2021

Uh oh!

mlbridge bot commented May 6, 2021

Uh oh!

theRealAph left a comment

Uh oh!

theRealAph May 6, 2021

Uh oh!

nick-arm May 7, 2021

Uh oh!

nick-arm commented May 7, 2021

Uh oh!

theRealAph commented May 7, 2021

Uh oh!

openjdk bot commented May 7, 2021 •

edited

Loading

Uh oh!

nick-arm commented May 7, 2021

Uh oh!

openjdk bot commented May 7, 2021

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

2 participants

8266609: AArch64: include FP/LR space in LIR_Assembler::initial_frame_size_in_bytes() #3898

8266609: AArch64: include FP/LR space in LIR_Assembler::initial_frame_size_in_bytes() #3898

Uh oh!

Conversation

nick-arm commented May 6, 2021 • edited by openjdk bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Progress

Issue

Reviewers

Reviewing

Uh oh!

bridgekeeper bot commented May 6, 2021

Uh oh!

openjdk bot commented May 6, 2021

Uh oh!

mlbridge bot commented May 6, 2021

Webrevs

Uh oh!

theRealAph left a comment

Choose a reason for hiding this comment

Uh oh!

theRealAph May 6, 2021

Choose a reason for hiding this comment

Uh oh!

nick-arm May 7, 2021

Choose a reason for hiding this comment

Uh oh!

nick-arm commented May 7, 2021

Uh oh!

theRealAph commented May 7, 2021

Uh oh!

openjdk bot commented May 7, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

nick-arm commented May 7, 2021

Uh oh!

openjdk bot commented May 7, 2021

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

2 participants

nick-arm commented May 6, 2021 •

edited by openjdk bot

Loading

openjdk bot commented May 7, 2021 •

edited

Loading