Page MenuHomePhabricator

Orlando (Orlando Cazalet-Hyams)
User

Projects

User does not belong to any projects.

User Details

User Since
Mar 28 2019, 8:58 AM (94 w, 4 d)

Recent Activity

Fri, Jan 15

Orlando requested review of D94761: [DebugInfo][dexter] Add dexter tests for merged values.
Fri, Jan 15, 3:19 AM · debug-info, Restricted Project

Thu, Jan 14

Orlando updated subscribers of D91424: [DebugInfo] Improve debug info accuracy for locals after inlining alloca uses [2/3].

As @bjope points out on D91425, it should be mentioned that while solution will improve the debugging experience in a number of cases it isn't bulletproof. For example a pass could insert a non-debug instruction between the dbg.value+deref and the to-be-inlined call.

Thu, Jan 14, 2:58 AM · debug-info, Restricted Project
Orlando updated the diff for D91425: [DebugInfo] Skip dbg.value+derefs in RemoveRedundantDbgInstrs forward scan [3/3].

Address feedback from @bjope:
+ Change comment for dbg.value+derefs in ShouldRemove lambda.
+ Update the test to check that the backward scan removes adjacent duplicate dbg.value+derefs still.

Thu, Jan 14, 2:51 AM · debug-info, Restricted Project
Orlando updated the diff for D91425: [DebugInfo] Skip dbg.value+derefs in RemoveRedundantDbgInstrs forward scan [3/3].

Rebase

Thu, Jan 14, 2:49 AM · debug-info, Restricted Project
Orlando updated the diff for D91424: [DebugInfo] Improve debug info accuracy for locals after inlining alloca uses [2/3].

Rebase and fix a typo in a comment.

Thu, Jan 14, 2:00 AM · debug-info, Restricted Project
Orlando updated the diff for D91423: [NFC][DebugInfo] Factor LowerDbgDeclare into two functions [1/3].

Updated based on @aprantl's comments. I removed LowerDbgDeclare's doxygen comment from Local.cpp (it still exists in the header Local.h). I didn't add a doxygen comment to the new function ConvertIndirectDbgIntrinsicToDbgValues because it is added in the header in the next patch D91424.

Thu, Jan 14, 1:07 AM · debug-info, Restricted Project

Nov 24 2020

Orlando added a project to D92013: [DebugInfo] Don't use DW_OP_implicit_value for fragments: debug-info.
Nov 24 2020, 7:48 AM · debug-info, Restricted Project

Nov 23 2020

Orlando added a comment to D91423: [NFC][DebugInfo] Factor LowerDbgDeclare into two functions [1/3].

Thanks @aprantl for looking at this. I meant to state in the description that this NFC change is set up for D91424 to help highlight the functional change there.

Nov 23 2020, 2:13 AM · debug-info, Restricted Project
Orlando updated the summary of D91423: [NFC][DebugInfo] Factor LowerDbgDeclare into two functions [1/3].
Nov 23 2020, 2:13 AM · debug-info, Restricted Project

Nov 18 2020

Orlando added a comment to D91425: [DebugInfo] Skip dbg.value+derefs in RemoveRedundantDbgInstrs forward scan [3/3].

Thank you @bjope for taking a look at this.

Nov 18 2020, 9:53 AM · debug-info, Restricted Project

Nov 13 2020

Orlando added a reviewer for D91425: [DebugInfo] Skip dbg.value+derefs in RemoveRedundantDbgInstrs forward scan [3/3]: bjope.
Nov 13 2020, 7:23 AM · debug-info, Restricted Project
Orlando requested review of D91425: [DebugInfo] Skip dbg.value+derefs in RemoveRedundantDbgInstrs forward scan [3/3].
Nov 13 2020, 7:17 AM · debug-info, Restricted Project
Orlando requested review of D91424: [DebugInfo] Improve debug info accuracy for locals after inlining alloca uses [2/3].
Nov 13 2020, 7:17 AM · debug-info, Restricted Project
Orlando requested review of D91423: [NFC][DebugInfo] Factor LowerDbgDeclare into two functions [1/3].
Nov 13 2020, 7:16 AM · debug-info, Restricted Project

Nov 9 2020

Orlando added a comment to D90772: [Coroutines] Add missing llvm.dbg.declare's to cover more allocas.

Jeremy said:

In theory this kind of behaviour is exactly what dbg.addr is for (variable lives in memory, maybe changes due to control flow). However moving everything to use dbg.addr stalled a long time ago, I believe @Orlando found that it's often unexpectedly dropped by optimisations, any opinion on whether it's usable @Orlando? We're hoping to use it for things someday soon, but we're definitely not there yet.

Nov 9 2020, 2:10 AM · Restricted Project

Oct 26 2020

Orlando added a comment to D89543: [DebugInfo][dexter] Add dexter tests for escaped locals.

Thanks. I tried running these on Windows but ran into issues with the dexter args so I landed these with UNSUPPORTED: system-windows.

Oct 26 2020, 10:36 AM · Restricted Project, debug-info
Orlando updated the summary of D89543: [DebugInfo][dexter] Add dexter tests for escaped locals.
Oct 26 2020, 10:34 AM · Restricted Project, debug-info
Orlando committed rG66d03af88c2e: [DebugInfo][dexter] Add dexter tests for escaped locals (authored by Orlando).
[DebugInfo][dexter] Add dexter tests for escaped locals
Oct 26 2020, 10:30 AM
Orlando closed D89543: [DebugInfo][dexter] Add dexter tests for escaped locals.
Oct 26 2020, 10:30 AM · Restricted Project, debug-info

Oct 23 2020

Orlando updated the diff for D89543: [DebugInfo][dexter] Add dexter tests for escaped locals.

Addressed Reid's comments:
+ Converted all tests to use attributes to control inlining behaviour (as opposed to a mix of this and clang flags).
+ Improved comments in some tests.

Oct 23 2020, 9:13 AM · Restricted Project, debug-info
Orlando closed D89810: [mem2reg] Remove dbg.values describing contents of dead allocas.
Oct 23 2020, 2:20 AM · debug-info, Restricted Project
Orlando added a comment to D89810: [mem2reg] Remove dbg.values describing contents of dead allocas.

Apologies, in shortening the commit message I carelessly removed the review number. This was committed in fea067bdfde4.

Oct 23 2020, 2:19 AM · debug-info, Restricted Project
Orlando committed rGfea067bdfde4: [mem2reg] Remove dbg.values describing contents of dead allocas (authored by Orlando).
[mem2reg] Remove dbg.values describing contents of dead allocas
Oct 23 2020, 2:01 AM

Oct 21 2020

Orlando added inline comments to D89810: [mem2reg] Remove dbg.values describing contents of dead allocas.
Oct 21 2020, 2:44 AM · debug-info, Restricted Project
Orlando updated the diff for D89810: [mem2reg] Remove dbg.values describing contents of dead allocas.

Addressed @vsk's comments:
+ Resize SmallVector<...,8> -> SmallVector<...,1>
+ Rename Declares -> DbgUsers.

Oct 21 2020, 2:20 AM · debug-info, Restricted Project
Orlando added a comment to D89810: [mem2reg] Remove dbg.values describing contents of dead allocas.

Hi @vsk, thanks for taking a look, I'll address your inline comments shortly.

Oct 21 2020, 1:42 AM · debug-info, Restricted Project

Oct 20 2020

Orlando requested review of D89810: [mem2reg] Remove dbg.values describing contents of dead allocas.
Oct 20 2020, 10:31 AM · debug-info, Restricted Project
Orlando added a comment to D85555: [InstCombine] Remove dbg.values describing contents of dead allocas.

FWIW, ignoring the test style discussion, the change LGTM.
AFAICT, prior to D80264 these dbg.value+derefs would've been silently dropped instead of made undef, so we're really just returning to previous behaviour with this patch. I've got a patch brewing which applies this fix to mem2reg which I'll upload shortly.

Oct 20 2020, 8:31 AM · Restricted Project

Oct 19 2020

Orlando added inline comments to D89543: [DebugInfo][dexter] Add dexter tests for escaped locals.
Oct 19 2020, 7:49 AM · Restricted Project, debug-info
Orlando added a comment to D89543: [DebugInfo][dexter] Add dexter tests for escaped locals.

Thanks for taking a look @rnk. Your comment on struct-dse.c reminds me that I left out some important info in the description. All of these except ctrl-flow.c "fail" (score < 1.0) at the moment. Should they be marked XFAIL?

Oct 19 2020, 1:26 AM · Restricted Project, debug-info

Oct 16 2020

Orlando requested review of D89543: [DebugInfo][dexter] Add dexter tests for escaped locals.
Oct 16 2020, 4:52 AM · Restricted Project, debug-info

Sep 23 2020

Orlando added inline comments to D87357: [SelectionDAG][DebugInfo] Use entry-values to recover variables values.
Sep 23 2020, 4:12 AM · Restricted Project, debug-info

Sep 22 2020

Orlando added a comment to D68620: DebugInfo: Use base address selection entries for debug_loc.

I guess all of these measurements were done without Split DWARF (shouldn't make things better/worse overall (across .o and .dwo), really, compared to DWARFv5 non-split - but means when only looking at .o files the difference of avoiding more debug_addr entries is more significant because there's fewer remaining .o debug bytes to begin with) and without compression (-gz) enabled?

Sep 22 2020, 9:48 AM · Restricted Project
Orlando added inline comments to D87233: [POC][DebugInfo] Use entry values within IR.
Sep 22 2020, 9:42 AM · Restricted Project, debug-info
Orlando added inline comments to D87357: [SelectionDAG][DebugInfo] Use entry-values to recover variables values.
Sep 22 2020, 9:38 AM · Restricted Project, debug-info
Orlando added a comment to D68620: DebugInfo: Use base address selection entries for debug_loc.

Just to clarify: when I say "elf" here I'm talking about the linked executable file, and "object files" are the pre-link .o files.

Sep 22 2020, 6:25 AM · Restricted Project
Orlando added a comment to D68620: DebugInfo: Use base address selection entries for debug_loc.

The following builds are with clang @ 485e6db8729 (3rd September) targeting x86.

+---------------------------------------------------------------+
| File size (bytes) of clang-3.4 built with -O2 -gdwarf-4       |
*===============================================================*
|                    | base addr    | no base addr | % change   |
+--------------------+--------------+--------------+------------+
| Accumulated object | 1874653208   | 1924003152   | +2.63      |
| file sizes         |              |              |            |
+--------------------+--------------+--------------+------------+
| Elf size           | 527591064    | 513946184    | -2.57      |
+--------------------+--------------+--------------+------------+
Sep 22 2020, 5:38 AM · Restricted Project

Sep 21 2020

Orlando added a comment to D68620: DebugInfo: Use base address selection entries for debug_loc.

Out of curiosity I also did a clang-3.4 build too using master @ 485e6db8729 (3rd September) with "-O2 -gdwarf-4". It is smaller when disabling base address specifiers (and emitting DWARFv 4) too:
With base addresses (default): Total File Size: 527591064
Without base addresses: Total File Size: 513946184 (-2.59 %)

Sep 21 2020, 7:57 AM · Restricted Project
Orlando updated subscribers of D87233: [POC][DebugInfo] Use entry values within IR.
Sep 21 2020, 7:20 AM · Restricted Project, debug-info
Orlando added inline comments to D87233: [POC][DebugInfo] Use entry values within IR.
Sep 21 2020, 5:38 AM · Restricted Project, debug-info

Sep 17 2020

Orlando added a comment to D68620: DebugInfo: Use base address selection entries for debug_loc.

I have here a copy of the table I shared earlier, with a new row "ShouldUseBaseAddress=false". The stats for this row are taken at 57d8acac64b (D86153) with the changes mentioned in my previous comment (disabling base address specifiers).

+------------------------------------------------------------------------------------------- +
| Mean binary size for benchmarks normalized as a percentage of llvm-3 builds                |
+---------------------------------+------------+------------------+-----------------+--------+
| llvm version                    | .debug_loc | other debug info | everything else | Total  |
+---------------------------------+------------+------------------+-----------------+--------+
| llvm-3                          | 13.7       | 33.2             | 53.1            | 100    |
| llvm-4                          | 12.7       | 33.8             | 53.8            | 100.3  |
| llvm-5                          | 13.4       | 35.6             | 54.6            | 103.7  |
| llvm-7                          | 18.4       | 35.6             | 54.0            | 108.0  |
| llvm-8                          | 17.5       | 37.1             | 54.5            | 109.1  |
| llvm-9                          | 19.7       | 37.2             | 54.6            | 111.5  |
| llvm-10 before dblaikie commit  | 19.8       | 37.4             | 54.9            | 112.1  |
| llvm-10 with dblaikie commit    | 25.6       | 37.4             | 54.9            | 117.9  |
| llvm-10                         | 25.8       | 37.5             | 54.8            | 118.1  |
| llvm-master before my commits   | 26.2       | 37.4             | 54.8            | 118.4  |
| llvm-master with my commits     | 18.4       | 35.5             | 55.3            | 109.3  |
| ShouldUseBaseAddress=false      | 14.9       | 35.5             | 55.3            | 105.7  |
+---------------------------------+------------+------------------+-----------------+--------+
Sep 17 2020, 2:49 PM · Restricted Project
Orlando added a comment to D68620: DebugInfo: Use base address selection entries for debug_loc.

Ah, OK - so it sounds like we're back down below the size before I added debug_loc base address specifiers? That's good to hear!

Sep 17 2020, 2:01 AM · Restricted Project

Sep 16 2020

Orlando added inline comments to D82678: [CGP] Set debug locations when optimizing phi types.
Sep 16 2020, 11:37 AM · debug-info, Restricted Project
Orlando added inline comments to D82678: [CGP] Set debug locations when optimizing phi types.
Sep 16 2020, 10:28 AM · debug-info, Restricted Project
Orlando abandoned D79949: [WIP][Example] Drop out-of-scope variable locations.

The real thing exists over at D82129 (landed).

Sep 16 2020, 1:23 AM · Restricted Project
Orlando added a comment to D68620: DebugInfo: Use base address selection entries for debug_loc.

Thanks for the data!

Sep 16 2020, 1:16 AM · Restricted Project

Sep 15 2020

Orlando added a comment to D68620: DebugInfo: Use base address selection entries for debug_loc.

@Orlando mentioned he was collecting some size data that would be relevant here, he'll post it when he's done. Basically .debug_loc sizes at various points.

Sep 15 2020, 7:39 AM · Restricted Project

Sep 3 2020

Orlando added a comment to D86937: Adding GDB PrettyPrinter for mlir::Identifier..

I've reverted this for now (485e6db8729) because one of our internal build bot is failing in the same way. I don't think many (any?) of the bots on http://lab.llvm.org:8011/console build the debuginfo-tests project!

Sep 3 2020, 2:55 AM · Restricted Project
Orlando added a reverting change for rG9e9e6e698d8e: Adding GDB PrettyPrinter for mlir::Identifier.: rG485e6db87293: Revert "Adding GDB PrettyPrinter for mlir::Identifier.".
Sep 3 2020, 2:50 AM
Orlando committed rG485e6db87293: Revert "Adding GDB PrettyPrinter for mlir::Identifier." (authored by Orlando).
Revert "Adding GDB PrettyPrinter for mlir::Identifier."
Sep 3 2020, 2:50 AM
Orlando added a reverting change for D86937: Adding GDB PrettyPrinter for mlir::Identifier.: rG485e6db87293: Revert "Adding GDB PrettyPrinter for mlir::Identifier.".
Sep 3 2020, 2:50 AM · Restricted Project

Aug 27 2020

Orlando added a comment to D86152: [DwarfDebug] Improve multi-BB single location detection in validThroughout (3/4).

Before landing I removed the XFAIL:* from llvm/test/DebugInfo/AArch64/inlined-argument.ll because it passes with this patch.

Aug 27 2020, 4:25 AM · debug-info, Restricted Project
Orlando added a reverting change for rGb9d977b0ca60: [DWARF] Add cuttoff guarding quadratic validThroughout behaviour: rGb6cca0ec05cf: Revert "[DWARF] Add cuttoff guarding quadratic validThroughout behaviour".
Aug 27 2020, 4:14 AM
Orlando committed rGb6cca0ec05cf: Revert "[DWARF] Add cuttoff guarding quadratic validThroughout behaviour" (authored by Orlando).
Revert "[DWARF] Add cuttoff guarding quadratic validThroughout behaviour"
Aug 27 2020, 4:14 AM
Orlando committed rG57d8acac64b8: [DwarfDebug] Improve validThroughout performance (4/4) (authored by Orlando).
[DwarfDebug] Improve validThroughout performance (4/4)
Aug 27 2020, 4:14 AM
Orlando committed rG3c491881d2ae: [DwarfDebug] Improve multi-BB single location detection in validThroughout (3/4) (authored by Orlando).
[DwarfDebug] Improve multi-BB single location detection in validThroughout (3/4)
Aug 27 2020, 4:14 AM
Orlando added a reverting change for D83236: [DWARF] Add cutoff guarding validThroughout to avoid near-quadratic behaviour: rGb6cca0ec05cf: Revert "[DWARF] Add cuttoff guarding quadratic validThroughout behaviour".
Aug 27 2020, 4:14 AM · Restricted Project
Orlando committed rG0b5a8050ea39: [DwarfDebug] Improve single location detection in validThroughout (2/4) (authored by Orlando).
[DwarfDebug] Improve single location detection in validThroughout (2/4)
Aug 27 2020, 4:14 AM
Orlando committed rGe048ea7b1a05: [NFC][DebugInfo] Create InstructionOrdering helper class (1/4) (authored by Orlando).
[NFC][DebugInfo] Create InstructionOrdering helper class (1/4)
Aug 27 2020, 4:14 AM
Orlando closed D86153: [DwarfDebug] Improve validThroughout performance (4/4).
Aug 27 2020, 4:14 AM · debug-info, Restricted Project
Orlando closed D86152: [DwarfDebug] Improve multi-BB single location detection in validThroughout (3/4).
Aug 27 2020, 4:14 AM · debug-info, Restricted Project
Orlando closed D86151: [DwarfDebug] Improve single location detection in validThroughout (2/4).
Aug 27 2020, 4:14 AM · debug-info, Restricted Project
Orlando closed D86150: [NFC][DebugInfo] Create InstructionOrdering helper class (1/4).
Aug 27 2020, 4:14 AM · debug-info, Restricted Project

Aug 26 2020

Orlando added a comment to D86152: [DwarfDebug] Improve multi-BB single location detection in validThroughout (3/4).

@djtodoro do you have any more thoughts on this patch? Copying part of my inline comment here for any other reviewers:

Aug 26 2020, 7:36 AM · debug-info, Restricted Project

Aug 25 2020

Orlando closed D85636: [llvm-dwarfdump] Fix misleading scope byte coverage statistics.
Aug 25 2020, 1:57 AM · Restricted Project, debug-info
Orlando added inline comments to D86152: [DwarfDebug] Improve multi-BB single location detection in validThroughout (3/4).
Aug 25 2020, 1:55 AM · debug-info, Restricted Project
Orlando updated the diff for D86152: [DwarfDebug] Improve multi-BB single location detection in validThroughout (3/4).

A small update to address comments from @djtodoro.

Aug 25 2020, 1:53 AM · debug-info, Restricted Project
Orlando added a comment to D85636: [llvm-dwarfdump] Fix misleading scope byte coverage statistics.

Oops, I didn't include the Phabricator revision line in the commit message.

Aug 25 2020, 1:43 AM · Restricted Project, debug-info
Orlando committed rG3ff4d75c9cbf: [llvm-dwarfdump] Fix misleading scope byte coverage statistics (authored by Orlando).
[llvm-dwarfdump] Fix misleading scope byte coverage statistics
Aug 25 2020, 1:38 AM
Orlando added a comment to D85636: [llvm-dwarfdump] Fix misleading scope byte coverage statistics.

Hi @djtodoro thanks for taking a look!

Aug 25 2020, 1:05 AM · Restricted Project, debug-info

Aug 21 2020

Orlando updated the diff for D85636: [llvm-dwarfdump] Fix misleading scope byte coverage statistics.

+ Rebase
+ Update test for latest yaml2obj with change suggested by @Higuoxing (thank you!).
+ Make the test more interesting, @dblaikie wdyt?

Aug 21 2020, 2:56 AM · Restricted Project, debug-info

Aug 20 2020

Orlando added inline comments to D85636: [llvm-dwarfdump] Fix misleading scope byte coverage statistics.
Aug 20 2020, 2:57 AM · Restricted Project, debug-info
Orlando added a comment to D86151: [DwarfDebug] Improve single location detection in validThroughout (2/4).

Just a note: This patch moves the InstructionOrdering class added in D86150 out from DbgEntityHistoryCalculator.cpp to DbgEntityHistoryCalculator.h so that we're able to use it in DebugHandlerBase.

Aug 20 2020, 2:17 AM · debug-info, Restricted Project

Aug 19 2020

Orlando added a comment to D86150: [NFC][DebugInfo] Create InstructionOrdering helper class (1/4).

Thanks for the review! I'll land the patch alongside the others in the stack when they're good to go.

Aug 19 2020, 10:21 AM · debug-info, Restricted Project
Orlando added a comment to D83895: [DebugInfo] Process DBG_VALUE_LIST in LiveDebugVariables.

Hi, sorry for leaving this for so long! I think it will take me a couple of passes to fully understand it, but for now I left some inline nits/questions. Please feel free to leave the nit changes until later revisions.

Aug 19 2020, 2:26 AM · Restricted Project, debug-info
Orlando added a comment to D85636: [llvm-dwarfdump] Fix misleading scope byte coverage statistics.

Thanks @jhenderson. I think this should have also addressed @dblaikie's concerns regarding the test, so I will aim to land this in a couple of days if there are no objections.

Aug 19 2020, 1:06 AM · Restricted Project, debug-info
Orlando added a comment to D86153: [DwarfDebug] Improve validThroughout performance (4/4).

Thank you @aprantl.

Aug 19 2020, 1:01 AM · debug-info, Restricted Project
Orlando added inline comments to D86150: [NFC][DebugInfo] Create InstructionOrdering helper class (1/4).
Aug 19 2020, 12:56 AM · debug-info, Restricted Project
Orlando updated the summary of D86150: [NFC][DebugInfo] Create InstructionOrdering helper class (1/4).
Aug 19 2020, 12:56 AM · debug-info, Restricted Project

Aug 18 2020

Orlando requested review of D86153: [DwarfDebug] Improve validThroughout performance (4/4).
Aug 18 2020, 10:17 AM · debug-info, Restricted Project
Orlando added reviewers for D86152: [DwarfDebug] Improve multi-BB single location detection in validThroughout (3/4): vsk, djtodoro, dblaikie, aprantl.
Aug 18 2020, 10:16 AM · debug-info, Restricted Project
Orlando requested review of D86152: [DwarfDebug] Improve multi-BB single location detection in validThroughout (3/4).
Aug 18 2020, 10:16 AM · debug-info, Restricted Project
Orlando requested review of D86151: [DwarfDebug] Improve single location detection in validThroughout (2/4).
Aug 18 2020, 10:12 AM · debug-info, Restricted Project
Orlando requested review of D86150: [NFC][DebugInfo] Create InstructionOrdering helper class (1/4).
Aug 18 2020, 10:11 AM · debug-info, Restricted Project
Orlando added inline comments to D86045: [llvm-dwarfdump] --statistics: switch to json::OStream. NFC.
Aug 18 2020, 2:36 AM · Restricted Project

Aug 14 2020

Orlando updated the diff for D85636: [llvm-dwarfdump] Fix misleading scope byte coverage statistics.

Thank you so much @Higuoxing, that's very kind!

Aug 14 2020, 8:30 AM · Restricted Project, debug-info
Orlando updated the diff for D85636: [llvm-dwarfdump] Fix misleading scope byte coverage statistics.

Thanks @aprantl and @jhenderson.

Aug 14 2020, 5:15 AM · Restricted Project, debug-info

Aug 12 2020

Orlando added inline comments to D85636: [llvm-dwarfdump] Fix misleading scope byte coverage statistics.
Aug 12 2020, 10:30 AM · Restricted Project, debug-info

Aug 11 2020

Orlando updated the diff for D85636: [llvm-dwarfdump] Fix misleading scope byte coverage statistics.

Hi @dstenb, thanks for taking a look. I've updated the diff to include a test but it feels a little unwieldy. Is there a better way to test this?

Aug 11 2020, 3:49 AM · Restricted Project, debug-info

Aug 10 2020

Orlando requested review of D85636: [llvm-dwarfdump] Fix misleading scope byte coverage statistics.
Aug 10 2020, 4:14 AM · Restricted Project, debug-info

Jul 31 2020

Orlando added inline comments to D85018: [WIP][POC][DebugInfo] Support for DW_AT_start_scope for scoped variables.
Jul 31 2020, 4:04 AM · Restricted Project

Jul 30 2020

Orlando added a comment to D84891: WIP: [Verifier] Flag dbg.declares which specify different addresses for the same fragment.

Hi @vsk, just a inline question from me. And in addition, I noticed that SourceLevelDebugging specifically states "there can only be one call to llvm.dbg.declare for a given concrete local variable". I wonder if we should update this to say something like "there can only be one call to llvm.dbg.declare for a given fragment of a local variable"?

Jul 30 2020, 2:42 AM · Restricted Project

Jul 22 2020

Orlando committed rGce6de3747bce: [DebugInfo] Drop location ranges for variables which exist entirely outside the… (authored by Orlando).
[DebugInfo] Drop location ranges for variables which exist entirely outside the…
Jul 22 2020, 5:11 AM
Orlando closed D82129: [DebugInfo] Drop location ranges for variables which exist entirely outside the variable's scope.
Jul 22 2020, 5:11 AM · debug-info, Restricted Project
Orlando added a comment to D82129: [DebugInfo] Drop location ranges for variables which exist entirely outside the variable's scope.

LGTM once Paul's comment is addressed.

Jul 22 2020, 1:43 AM · debug-info, Restricted Project

Jul 20 2020

Orlando added a comment to D82129: [DebugInfo] Drop location ranges for variables which exist entirely outside the variable's scope.

Thanks everyone for the review and discussion so far. The general idea of the patch has been okayed and it just needs a code review now. Could anyone please take a look?

Jul 20 2020, 1:57 AM · debug-info, Restricted Project

Jul 10 2020

Orlando added a comment to D83495: [DebugInfo] Add DWARF emission for DBG_VALUE_LIST.

Hey, just noticed a couple of comments to remove from the tests.

Jul 10 2020, 2:47 AM · Restricted Project, debug-info

Jul 7 2020

Orlando added a comment to D82129: [DebugInfo] Drop location ranges for variables which exist entirely outside the variable's scope.

I think I didn't fully grasp that the blocks were being (tail-)merged, which makes the scope ambiguous, and all the rest. So I withdraw the objection on that basis. DWARF is fine with multiple variables pointing to the same location, but it's less forgiving about scopes IIRC, much like it can't describe multiple source attributions for an instructions. This all makes me sad, but that's how DWARF is at the moment.

Is there still an open question about whether this wants to be a cleanup pass or a verifier check? I apologize for losing track.

Jul 7 2020, 6:33 AM · debug-info, Restricted Project

Jul 6 2020

Orlando added a comment to D82129: [DebugInfo] Drop location ranges for variables which exist entirely outside the variable's scope.

Does anyone have any concerns with this patch that they feel have not been addressed?

Jul 6 2020, 1:50 AM · debug-info, Restricted Project
Orlando updated the summary of D82129: [DebugInfo] Drop location ranges for variables which exist entirely outside the variable's scope.
Jul 6 2020, 1:50 AM · debug-info, Restricted Project