Page MenuHomePhabricator

honggyu.kim (Honggyu Kim)
User

Projects

User does not belong to any projects.

User Details

User Since
Jun 7 2015, 5:59 PM (227 w, 3 d)

Recent Activity

Sep 1 2016

honggyu.kim added a comment to D22825: Add a counter-function insertion pass.

I've also committed related patches. I really appreciate your help, Hal!

Sep 1 2016, 5:08 AM
honggyu.kim committed rL280357: [IR] Properly handle escape characters in Attribute::getAsString().
[IR] Properly handle escape characters in Attribute::getAsString()
Sep 1 2016, 4:52 AM
honggyu.kim closed D23792: IR: Properly handle escape characters in Attribute::getAsString() by committing rL280357: [IR] Properly handle escape characters in Attribute::getAsString().
Sep 1 2016, 4:52 AM
honggyu.kim removed a parent revision for D23792: IR: Properly handle escape characters in Attribute::getAsString(): D22825: Add a counter-function insertion pass.
Sep 1 2016, 4:49 AM
honggyu.kim removed a child revision for D22825: Add a counter-function insertion pass: D23792: IR: Properly handle escape characters in Attribute::getAsString().
Sep 1 2016, 4:49 AM
honggyu.kim committed rL280355: [Frontend] Fix mcount inlining bug.
[Frontend] Fix mcount inlining bug
Sep 1 2016, 4:37 AM
honggyu.kim closed D22666: Frontend: Fix mcount inlining bug by committing rL280355: [Frontend] Fix mcount inlining bug.
Sep 1 2016, 4:37 AM

Aug 31 2016

honggyu.kim added a comment to D22825: Add a counter-function insertion pass.

Hi Hal,

Aug 31 2016, 11:27 PM
honggyu.kim committed rL280337: Remove whitespace to test commit access.
Remove whitespace to test commit access
Aug 31 2016, 11:23 PM

Aug 29 2016

honggyu.kim added a comment to D23792: IR: Properly handle escape characters in Attribute::getAsString().

Since I don't have write permission, it would be grateful if anyone can give me commit permission on the repository.
Otherwise, can anyone please commit this patch and related ones? - D22825, D22666

Aug 29 2016, 5:07 AM

Aug 26 2016

honggyu.kim added a comment to D23792: IR: Properly handle escape characters in Attribute::getAsString().

Thank you, this looks better. A minor suggestions and then this LGTM.

Aug 26 2016, 8:29 PM
honggyu.kim updated the diff for D23792: IR: Properly handle escape characters in Attribute::getAsString().
Aug 26 2016, 8:27 PM

Aug 25 2016

honggyu.kim added a comment to D23792: IR: Properly handle escape characters in Attribute::getAsString().

These should have the same behavior; only, the existing code should be slightly more efficient (without optimization) because it knows that it's only adding a single character rather than possible an array of them. Are you actually seeing different behavior at runtime?

John.

Aug 25 2016, 6:08 AM
honggyu.kim updated the diff for D23792: IR: Properly handle escape characters in Attribute::getAsString().
Aug 25 2016, 5:58 AM

Aug 24 2016

honggyu.kim added a comment to D23792: IR: Properly handle escape characters in Attribute::getAsString().

In Attribute::getAsString(), it is expected to return std::string as the name describes. I think getAsString() function should not print the string directly with PrintEscapedString(). If I directly print attribute string here, we have to modify many functions in the execution paths above. The only problem here is printing Val. That's why I added a new function GetEscapedString(). Please correct me if there's a better approach.

PrintEscapedString prints to a raw_ostream that's passed in as a parameter, and there's a raw_string_ostream that will just append to a std::string.

John.

Aug 24 2016, 5:03 PM

Aug 23 2016

honggyu.kim updated D23792: IR: Properly handle escape characters in Attribute::getAsString().
Aug 23 2016, 6:48 PM
honggyu.kim updated D23792: IR: Properly handle escape characters in Attribute::getAsString().
Aug 23 2016, 6:48 PM

Aug 22 2016

honggyu.kim added a comment to D23792: IR: Properly handle escape characters in Attribute::getAsString().

Hi John,

Aug 22 2016, 9:56 PM
honggyu.kim updated the diff for D22666: Frontend: Fix mcount inlining bug.

I have fixed SOH character problem by D23792. I've verified that mcount.c and gnu-mcount.c tests are passed now.

$ llvm-lit tools/clang/test/CodeGen/mcount.c
-- Testing: 1 tests, 1 threads --
PASS: Clang :: CodeGen/mcount.c (1 of 1)
Testing Time: 0.81s
  Expected Passes    : 1
Aug 22 2016, 9:15 PM
honggyu.kim added a child revision for D22825: Add a counter-function insertion pass: D23792: IR: Properly handle escape characters in Attribute::getAsString().
Aug 22 2016, 9:08 PM
honggyu.kim added a parent revision for D23792: IR: Properly handle escape characters in Attribute::getAsString(): D22825: Add a counter-function insertion pass.
Aug 22 2016, 9:08 PM
honggyu.kim retitled D23792: IR: Properly handle escape characters in Attribute::getAsString() from to IR: Properly handle escape characters in Attribute::getAsString().
Aug 22 2016, 9:07 PM

Aug 12 2016

honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

No, the inserted character is a literal SOH, not the string "\01".

Aug 12 2016, 3:06 AM

Aug 11 2016

honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

Theres a frontend mangler and a backend mangler. You should try this on Windows GNU environments :-).

Aug 11 2016, 2:15 AM

Aug 7 2016

honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

The \01 is to prevent the mangler from touching the function name. If you noticed the check tags in the quoted test, some targets expect it to be decorated and others do not. What exactly do you mean that lit has a strings matching with \01? Its not a string \01, its the octal character 1.

Aug 7 2016, 8:02 PM

Aug 5 2016

honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

I see the difference now.
Below is the originally generated IR without this patch. It shows \01 clearly in call void @"\01__gnu_mcount_nc"().

$ clang -target armv7-unknown-none-eabi -pg -meabi gnu -S -emit-llvm -o - test-mcount.c
; ModuleID = 'mcount.c'
source_filename = "mcount.c"
target datalayout = "e-m:e-p:32:32-i64:64-v128:64:128-a:0:32-n32-S64"
target triple = "armv7-unknown-none-eabi"
Aug 5 2016, 7:51 PM
honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

The \01 is to prevent the mangler from touching the function name. If you noticed the check tags in the quoted test, some targets expect it to be decorated and others do not. What exactly do you mean that lit has a strings matching with \01? Its not a string \01, its the octal character 1.

Aug 5 2016, 5:33 PM

Aug 4 2016

honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

Hi Renato and Saleem,

Aug 4 2016, 11:03 PM
honggyu.kim added a reviewer for D22666: Frontend: Fix mcount inlining bug: rengolin.
Aug 4 2016, 10:51 PM

Aug 2 2016

honggyu.kim accepted D22825: Add a counter-function insertion pass.

It works fine with D22666 now. Thanks for your help :)

Aug 2 2016, 5:55 PM

Jul 31 2016

honggyu.kim updated the diff for D22666: Frontend: Fix mcount inlining bug.

I've updated the testcase again it adds 2 more tests for -O2 and without -pg.
Please have a look at it.

$ llvm-lit mcount.c
-- Testing: 1 tests, 1 threads --
PASS: Clang :: CodeGen/mcount.c (1 of 1)
Testing Time: 32.61s
  Expected Passes    : 1
Jul 31 2016, 11:22 PM

Jul 28 2016

honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

In this case, I think that making a simple test (changing the current test to be) like test/CodeGen/stackrealign.c would be fine. If you have any questions, please feel free to ask.

Jul 28 2016, 8:58 PM
honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

Should we also modify clang/test/CodeGen/mcount.c as well? I'm not actually familiar with test infra.

Yes, you'll need to modify it to test for the attribute instead.

Jul 28 2016, 8:11 PM
honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

Should we also modify clang/test/CodeGen/mcount.c as well? I'm not actually familiar with test infra.

$ cat llvm/tools/clang/test/CodeGen/mcount.c
// RUN: %clang_cc1 -pg -triple i386-unknown-unknown -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -pg -triple powerpc-unknown-gnu-linux -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
// RUN: %clang_cc1 -pg -triple powerpc64-unknown-gnu-linux -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
// RUN: %clang_cc1 -pg -triple powerpc64le-unknown-gnu-linux -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
// RUN: %clang_cc1 -pg -triple i386-netbsd -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
// RUN: %clang_cc1 -pg -triple x86_64-netbsd -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
// RUN: %clang_cc1 -pg -triple arm-netbsd-eabi -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
// RUN: %clang_cc1 -pg -triple aarch64-netbsd -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
// RUN: %clang_cc1 -pg -triple mips-netbsd -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
// RUN: %clang_cc1 -pg -triple powerpc-netbsd -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
// RUN: %clang_cc1 -pg -triple powerpc64-netbsd -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
// RUN: %clang_cc1 -pg -triple powerpc64le-netbsd -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
// RUN: %clang_cc1 -pg -triple sparc-netbsd -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
// RUN: %clang_cc1 -pg -triple sparc64-netbsd -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-PREFIXED %s
void foo(void) {
// CHECK: call void @mcount()
// CHECK-PREFIXED: call void @_mcount()
}
Jul 28 2016, 7:47 PM
honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

Comments about the comment, but otherwise, LGTM.

Jul 28 2016, 7:20 PM
honggyu.kim updated the diff for D22666: Frontend: Fix mcount inlining bug.
Jul 28 2016, 7:19 PM
honggyu.kim removed a parent revision for D22666: Frontend: Fix mcount inlining bug: D22825: Add a counter-function insertion pass.
Jul 28 2016, 6:33 PM
honggyu.kim removed a child revision for D22825: Add a counter-function insertion pass: D22666: Frontend: Fix mcount inlining bug.
Jul 28 2016, 6:33 PM
honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

I just wrote how I tested this in the bugzilla report page below:
https://llvm.org/bugs/show_bug.cgi?id=28660

Jul 28 2016, 12:02 AM

Jul 27 2016

honggyu.kim updated D22666: Frontend: Fix mcount inlining bug.
Jul 27 2016, 11:40 PM
honggyu.kim added a child revision for D22825: Add a counter-function insertion pass: D22666: Frontend: Fix mcount inlining bug.
Jul 27 2016, 11:38 PM
honggyu.kim added a parent revision for D22666: Frontend: Fix mcount inlining bug: D22825: Add a counter-function insertion pass.
Jul 27 2016, 11:38 PM
honggyu.kim updated the diff for D22666: Frontend: Fix mcount inlining bug.

I've updated the diff that can be working with D22825. I appreciate your work in backend, Hal.
Please correct me if I'm wrong.

Jul 27 2016, 11:38 PM
honggyu.kim added inline comments to D22825: Add a counter-function insertion pass.
Jul 27 2016, 6:57 PM

Jul 26 2016

honggyu.kim added a comment to D22825: Add a counter-function insertion pass.

And I think we should consider -finstrument-functions option as well. It inserts two functions both at the entry and exit. We maybe better to change the name 'CountingFunction' to something else to cover both cases. 'CountingEntryFunction' and 'CountingExitFunction' seems to be okay as well.

No, those are different. They're supposed to be duplicated upon inlining. The GCC manual says:

"This instrumentation is also done for functions expanded inline in other functions. The profiling calls will indicate where, conceptually, the inline function is entered and exited. This means that addressable versions of such functions must be available. If all your uses of a function are expanded inline, this may mean an additional expansion of code size. If you use `extern inline' in your C code, an addressable version of such functions must be provided."

So these can be be inserted in the frontend as they are now.

Jul 26 2016, 7:37 PM
honggyu.kim added a comment to D22825: Add a counter-function insertion pass.

And I think we should consider -finstrument-functions option as well. It inserts two functions both at the entry and exit. We maybe better to change the name 'CountingFunction' to something else to cover both cases. 'CountingEntryFunction' and 'CountingExitFunction' seems to be okay as well.

Jul 26 2016, 7:15 PM
honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

Here's a patch for the backend part: D22825 - If this makes sense to everyone, we can change the frontend to add the corresponding attribute instead of inserting the function calls directly.

Jul 26 2016, 7:01 PM
honggyu.kim added a comment to D22825: Add a counter-function insertion pass.

This seems like the right approach to me; I won't review the pass parts, though. Should we do this + the change to actually start setting the flags in one patch?

Do you prefer that? I generally prefer when people split the Clang and LLVM patches into separate reviews (especially when they're functionally separable). On other other hand, if we move everything to one git repository, single patches will be easier...

Ah, no, doing it separately is fine, then.

Hi, I have tested it and it works great for the backend. Thanks for the patch!

Jul 26 2016, 7:00 PM
honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

One more problem is that -finstrument-functions has the same inlining problem. The code in llvm/tools/clang/lib/CodeGen/CodeGenFunction.cpp shows that.

void CodeGenFunction::StartFunction(GlobalDecl GD,
                                    QualType RetTy,
                                    llvm::Function *Fn,
                                    const CGFunctionInfo &FnInfo,
                                    const FunctionArgList &Args,
                                    SourceLocation Loc,
                                    SourceLocation StartLoc) {
    ...
  if (ShouldInstrumentFunction())
    EmitFunctionInstrumentation("__cyg_profile_func_enter");
Jul 26 2016, 6:48 AM
honggyu.kim added a reviewer for D22666: Frontend: Fix mcount inlining bug: hfinkel.
Jul 26 2016, 6:39 AM
honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

Note that the presence of the mcount call itself will significantly impact inlining and, really, the entire optimization pipeline. Having this be a late pass seems more in keeping with what I assume is a goal that this instrumentation doesn't drastically change the generated code.

Jul 26 2016, 6:39 AM

Jul 22 2016

honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

What's the actual desired behavior here? Should the inliner strip out calls to mcount() as it inlines?

I think they basically just want a late pass (as in immediately prior to codegen) to insert the mcount calls.

Jul 22 2016, 7:10 PM
honggyu.kim added a comment to D22666: Frontend: Fix mcount inlining bug.

Hi John and Hans,

Jul 22 2016, 1:40 AM
honggyu.kim retitled D22666: Frontend: Fix mcount inlining bug from to Frontend: Fix mcount inlining bug.
Jul 22 2016, 1:39 AM

Oct 28 2015

honggyu.kim added a comment to D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.

Hi,

You are right the diff is is based on the hash. We already tried to
use an earlier hash generator (before the patch was introduced), which
generates a slightly different plist, that is why the current version
does not work with the patch.
We will fix CodeChecker to use new hash tag introduced in the final
patch (we did not change it so far because we didn't know what will be
the accepted naming convention and plist format).

Hi, thanks for the explanation. I will try again when it is modified based on the final patch.

Oct 28 2015, 5:59 AM

Oct 25 2015

honggyu.kim updated subscribers of D14014: Checker of proper vfork usage.
Oct 25 2015, 7:31 AM

Oct 23 2015

honggyu.kim added a comment to D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.

By default it is disabled for security reasons. You can enable by passing a command line option to the code checker, something like --not-host-only.

Oct 23 2015, 5:25 AM

Oct 22 2015

honggyu.kim added a comment to D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.

Hi Anna & Kim,

Hi Daniel,

Oct 22 2015, 6:07 AM
honggyu.kim added a comment to D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.

Hi Anna,

I think you misunderstood my comment. I am not talking about using the existing HTML files here but rather having an HTML viewer, which you could use to browse source code. This viewer would be extended to read the bug reports from the plist files and display them. Currently, we create an html file with source code + report info for each bug report. This does not scale when you have a lot of reports on a single large file (ex: sqlite).

I know you already mentioned about the scalability issue but I was thinking to use the existing implementation since I thought we didn't have the HTML viewer you mentioned as of now. (before I saw Daniel's comment)

Oct 22 2015, 6:05 AM

Oct 21 2015

honggyu.kim added a comment to D10305: [Clang Static Analyzer] Bug identification.

Hi Gabor,

Oct 21 2015, 4:26 AM
honggyu.kim abandoned D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.

Hi Anna,

Oct 21 2015, 4:18 AM

Sep 21 2015

honggyu.kim added a comment to D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.

I agree with Gabor and do not think we should have 2 entries in the plist file for bug identification. This is confusing and I do not see a need for this.

Any comparison script can check for 2 fields in a plist file and concatenate them. We've already discussed this during the patch review for the other patch. For example, the latest version does not include the file name.

I didn't put 2 entries in plist, I made issue_hash as it was while adding bugid to html report only.
So this patch makes issue_hash based on Babati's patch after removing redundent fields. And it also makes BugID that contains all the information so that it doesn't require post processing.
BugId is for HTML report comparison.

Sep 21 2015, 9:30 PM

Sep 18 2015

honggyu.kim added a comment to D10305: [Clang Static Analyzer] Bug identification.

Hi Babati,

Sep 18 2015, 3:26 AM
honggyu.kim added a comment to D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.

Hi Anna and Gábor,

Sep 18 2015, 3:21 AM
honggyu.kim added a child revision for D10305: [Clang Static Analyzer] Bug identification: D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.
Sep 18 2015, 3:00 AM
honggyu.kim added a parent revision for D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py: D10305: [Clang Static Analyzer] Bug identification.
Sep 18 2015, 3:00 AM
honggyu.kim updated the diff for D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.
Sep 18 2015, 2:59 AM

Sep 17 2015

honggyu.kim added a comment to D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.

You made a comment while I was writing other comment :)

Sep 17 2015, 7:48 PM
honggyu.kim added a comment to D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.

Just to make it simpler to understand, original strings of issue_hash are as below with this patch:

Sep 17 2015, 7:38 PM

Sep 16 2015

honggyu.kim added a comment to D10305: [Clang Static Analyzer] Bug identification.

honggyu.kim,

Uniquing HTML reports is out of the scope of this patch and should be discussed elsewhere (either send a design idea to cfe-dev, send a patch for review, or file a bugzilla request).

I agree that this patch is a definite improvement to issue identification; however, as I mentioned earlier, it needs tests.

Sep 16 2015, 8:06 AM
honggyu.kim added a comment to D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.

issue_hash is generated with hash value rather than line offset from function as below:

<key>issue_hash</key><string>765b04830932fef5631bf3c48c4d2db2</string>
Sep 16 2015, 8:02 AM
honggyu.kim updated D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.
Sep 16 2015, 8:00 AM
honggyu.kim added a comment to D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py.

I think we can enhance the bug comparison method one by one rather than making a big change at once.
The point of this patch is to make "issue_hash" field in plist more robust while still using existing CmpRuns.py for comparison.

Sep 16 2015, 7:57 AM
honggyu.kim retitled D12906: [RFC] Bug identification("issue_hash") change for CmpRuns.py from to [RFC] Bug identification("issue_hash") change for CmpRuns.py.
Sep 16 2015, 7:56 AM

Sep 12 2015

honggyu.kim added a comment to D12827: [analyzer] fix an error finding clang path.

Thanks, Honggyu!

Sep 12 2015, 3:16 PM
honggyu.kim updated the diff for D12827: [analyzer] fix an error finding clang path.
Sep 12 2015, 5:03 AM
honggyu.kim added a comment to D12827: [analyzer] fix an error finding clang path.

After applying rL247466, I got the following error which was not shown before.

Sep 12 2015, 2:11 AM
honggyu.kim retitled D12827: [analyzer] fix an error finding clang path from to [analyzer] fix an error finding clang path.
Sep 12 2015, 2:09 AM

Sep 8 2015

honggyu.kim updated subscribers of D12406: [Analyzer] Add -analyzer-config option for function size the inliner considers as large.
Sep 8 2015, 1:29 AM

Sep 7 2015

honggyu.kim added a comment to D12673: [analyzer] Remove whitespace in source code.

We generally don't do large whitespace changes like this because of issue
with version control (not all told are whitespace ignorant which makes
history work difficult).

I agree that it would be difficult to trace the history of svn or git.
But I wrote this patch because it would be better to have cleaner source code.
If we don't make whitespace in the future changes, it would be okay to trace the history in the long term.

Sep 7 2015, 6:25 PM
honggyu.kim added a comment to D10305: [Clang Static Analyzer] Bug identification.

I would like to also write about bug identification methods.
As I observed the current CmpRuns.py script, the IssueIdentifier is defined as follows:

Sep 7 2015, 6:46 AM
honggyu.kim added a comment to D10305: [Clang Static Analyzer] Bug identification.

Sorry for the late answer.
I would like to write the comment for HTML report comparison first.

Sep 7 2015, 6:39 AM
honggyu.kim added a comment to D12673: [analyzer] Remove whitespace in source code.

Note: I am not the owner of this code but to me, this should be committed without a review.

Sep 7 2015, 6:23 AM
honggyu.kim updated subscribers of D12673: [analyzer] Remove whitespace in source code.
Sep 7 2015, 6:21 AM
honggyu.kim added a comment to D12673: [analyzer] Remove whitespace in source code.

If anyone thinks this patch is too big, I can submit smaller separated patches.

Sep 7 2015, 6:20 AM
honggyu.kim retitled D12673: [analyzer] Remove whitespace in source code from to [analyzer] Remove whitespace in source code.
Sep 7 2015, 5:47 AM

Aug 16 2015

honggyu.kim updated subscribers of D5102: [analyzer][Bugfix/improvement] Fix for PR16833.
Aug 16 2015, 8:50 AM

Aug 12 2015

honggyu.kim updated subscribers of D12002: Initial WebAssembly support in clang.
Aug 12 2015, 5:40 PM
honggyu.kim accepted D10356: scan-build: Add --analyzer-target option.

Committed r244400

Aug 12 2015, 4:59 PM

Aug 11 2015

honggyu.kim updated subscribers of D11968: Add sanitizer blacklists to the rules generated with -M/-MM/-MD/-MMD..
Aug 11 2015, 5:24 PM

Aug 10 2015

honggyu.kim added a comment to D10305: [Clang Static Analyzer] Bug identification.

honggyu.kim,

You are right, CmpRuns.py does not work with HTML files.

The list of HTML reports is produced by scan-build and there is no facility there to search for newly generated reports.

Thanks for the confirmation.

Aug 10 2015, 7:03 PM
honggyu.kim added a comment to D10305: [Clang Static Analyzer] Bug identification.

Hi Babati,

Aug 10 2015, 7:00 AM

Aug 9 2015

honggyu.kim added a comment to D6551: Improvements to scan-build..

Hi Anton,

Aug 9 2015, 8:17 PM

Aug 8 2015

honggyu.kim added a comment to D6551: Improvements to scan-build..

Hi all,

Aug 8 2015, 7:37 AM
honggyu.kim updated subscribers of D6551: Improvements to scan-build..
Aug 8 2015, 3:50 AM

Aug 7 2015

honggyu.kim added a comment to D10356: scan-build: Add --analyzer-target option.

I see. The problem is that the ad hoc interposition that scan-build uses does not handle the case that the underlying compiler is built as a cross-compiler, but does not take a direct target triple. That compiler only builds for the other target, unlike clang which is a cross-compiler.

That's exactly what I was going to explain. I hope that scan-build can automatically extract the target triple information from the given compiler, but there's no way to do it as of yet. That's the reason I made this option to explicitly set the target triple info for analysis.

Aug 7 2015, 10:43 PM

Aug 4 2015

honggyu.kim added a comment to D10356: scan-build: Add --analyzer-target option.

Here's one more example.

Aug 4 2015, 6:06 AM
honggyu.kim added a comment to D10356: scan-build: Add --analyzer-target option.

Hi Ted,

Aug 4 2015, 5:49 AM

Aug 2 2015

honggyu.kim added a comment to D10356: scan-build: Add --analyzer-target option.

I have updated the new patchset with more information in help message and changed the option name to —analyzer-target.
If you think the "extra flags to pass to the compiler” is still needed, I will write another patch for that.
I appreciate your comment and suggestion.

Aug 2 2015, 11:59 PM
honggyu.kim updated the diff for D10356: scan-build: Add --analyzer-target option.
Aug 2 2015, 11:58 PM