8282578: AIOOBE in javax.sound.sampled.Clip #9016

azuev-java · 2022-06-03T15:43:39Z

Add try/catch clause to ignore an exception since it is harmless for we isolated
the massge data before passing it ro processor.
Add test case.

Progress

Change must be properly reviewed (1 review required, with at least 1 Reviewer)
Change must not contain extraneous whitespace
Commit message must refer to an issue

Issue

JDK-8282578: AIOOBE in javax.sound.sampled.Clip

Reviewers

Phil Race (@prrace - Reviewer) ⚠️ Review applies to d085cfb1
Alexey Ivanov (@aivanov-jdk - Reviewer) ⚠️ Review applies to d085cfb1
Alexander Zvegintsev (@azvegint - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk pull/9016/head:pull/9016
$ git checkout pull/9016

Update a local copy of the PR:
$ git checkout pull/9016
$ git pull https://git.openjdk.org/jdk pull/9016/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 9016

View PR using the GUI difftool:
$ git pr show -t 9016

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/9016.diff

Surround SysEx message processing block with try/catch allowing MIDI subsystem to attempt to ingest the rest of the file. Added test case.

bridgekeeper · 2022-06-03T15:44:56Z

👋 Welcome back kizune! 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 · 2022-06-03T15:47:04Z

@azuev-java The following label will be automatically applied to this pull request:

client

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 · 2022-06-03T15:51:04Z

Webrevs

victordyakov · 2022-06-03T16:43:15Z

@azvegint @prrace please review

mrserb · 2022-06-03T21:15:21Z

Probably it is better to check the data length for each sys message and discard it if the data is too small? Ignoring all possible AIOOBE from this large method which calls many other large methods from SoftVoice/SoftTuning may hide some other bugs.

openjdk · 2022-06-03T21:20:39Z

@azuev-java 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:

8282578: AIOOBE in javax.sound.sampled.Clip

Reviewed-by: prr, aivanov, azvegint

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 201 new commits pushed to the master branch:

8b69a2e: 8298099: [JVMCI] decouple libgraal from JVMCI module at runtime
8a9911e: 8295803: Console should be usable in jshell and other environments
5d4c71c: 8281236: (D)TLS key exchange named groups
10356e7: 8298303: (fs) temporarily remove Path.getExtension
3b8c7ef: 8157023: Integrate NMT with JFR
e86f31b: 8298301: C2: assert(main_cmp->in(2)->Opcode() == Op_Opaque1) failed: main loop has no opaque node?
8edb98d: 8165943: LineBreakMeasurer does not measure correctly if TextAttribute.TRACKING is set.
3934484: 8298205: Prefer Member Initialization Lists for JFR classes in os_perf.hpp
389b8f4: 8297298: SequenceInputStream should override transferTo
dd7385d: 8298202: [AIX] Dead code elimination removed jfr constructor used by AIX
... and 191 more: https://git.openjdk.org/jdk/compare/105d9d75e84a46400f52fafda2ea00c99c14eaf0...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.

azuev-java · 2022-06-04T13:46:10Z

Probably it is better to check the data length for each sys message and discard it if the data is too small? Ignoring all possible AIOOBE from this large method which calls many other large methods from SoftVoice/SoftTuning may hide some other bugs.

No, because determining the correct length of the message will require basically to parse it all, the correct length could be a 2 bytes or hundreds of bytes - in order to determine we have to process the message completely.

mrserb · 2022-06-04T20:58:37Z

No, because determining the correct length of the message will require basically to parse it all, the correct length could be a 2 bytes or hundreds of bytes - in order to determine we have to process the message completely.

There are 64 usages of data in that method most of them use constants like 1/2/3 and other usages are in the loops, both can be easily checked. There is also code like int[] destinations = new int[(data.length - 7) / 2]; which could cause the NegativeArrayException, or probably some others. It s better to check the out of bands access before access than to have a try/catch block for 300 lines of code.

bridgekeeper · 2022-07-03T04:48:19Z

@azuev-java This pull request has been inactive for more than 4 weeks and will be automatically closed if another 4 weeks passes without any activity. To avoid this, simply add a new comment to the pull request. Feel free to ask for assistance if you need help with progressing this pull request towards integration!

AJ1062910 · 2022-07-04T15:06:01Z

pre submit tests failed*

bridgekeeper · 2022-08-11T11:31:50Z

@azuev-java This pull request has been inactive for more than 4 weeks and will be automatically closed if another 4 weeks passes without any activity. To avoid this, simply add a new comment to the pull request. Feel free to ask for assistance if you need help with progressing this pull request towards integration!

bridgekeeper · 2022-09-08T12:42:44Z

@azuev-java This pull request has been inactive for more than 8 weeks and will now be automatically closed. If you would like to continue working on this pull request in the future, feel free to reopen it! This can be done using the /open pull request command.

…282578

analysis.

azuev-java · 2022-12-04T22:10:25Z

Just uploaded the new version with proper length analysis so would be thankful for the continuation of the review.

azuev-java · 2022-12-04T22:12:00Z

/open

openjdk · 2022-12-04T22:13:21Z

@azuev-java This pull request is now open

azuev-java · 2022-12-07T06:48:22Z

@azvegint Can you please take a look?

azvegint · 2022-12-07T11:49:51Z

src/java.desktop/share/classes/com/sun/media/sound/SoftTuning.java

                int subid2 = data[4] & 0xFF;
                switch (subid2) {
                case 0x01: // BULK TUNING DUMP (NON-REAL-TIME)
                {
                    // http://www.midi.org/about-midi/tuning.shtml
                    //if (!checksumOK2(data))
                    //    break;
+                    if (data.length < 406) {


At first glance it looks like a magic numbers to me.

for better readability we could declare r before this check and use 128*3 + r

Makes sense, fixed.

azvegint · 2022-12-07T13:51:39Z

src/java.desktop/share/classes/com/sun/media/sound/SoftTuning.java

                    int ll = data[6] & 0xFF;
+                    if (data.length < ll * 4 + 8) {


Shouldn't it be ll * 4 + 7 == ll * 4 + r?

azvegint · 2022-12-07T14:10:58Z

src/java.desktop/share/classes/com/sun/media/sound/SoftTuning.java

@@ -193,6 +228,9 @@ public void load(byte[] data) {
                           // Real-Time/REAL-TIME)
                {
                    // http://www.midi.org/about-midi/tuning-scale.shtml
+                    if (data.length < 21) {


< 20?

int[] data = new int[20]; for (int i = 0; i < 12; i++) { System.out.println(data[i + 8]); }

Yes. Math is hard.

azvegint · 2022-12-07T15:15:34Z

src/java.desktop/share/classes/com/sun/media/sound/SoftMainMixer.java

+                                    int[] ranges = new int[(data.length - 7) / 2];
+                                    int ix = 0;
+                                    for (int j = 6; j < data.length - 1; j += 2) {
+                                        destinations[ix] = data[j] & 0xFF;


Here is another possible AIOOOBE

e.g. if data length is 8, we will have destination and ranges length of 0:

int[] data = new int[8]; System.out.println("data len " + data.length); if (data.length < 7) { System.out.println("Prevent"); return; } int newSize = (data.length - 7) / 2; System.out.println("new size " + newSize); int[] destinations = new int[newSize]; int[] ranges = new int[newSize]; int ix = 0; for (int j = 6; j < data.length - 1; j += 2) { System.out.println("index %d %d".formatted(j, ix) ); destinations[ix] = data[j] & 0xFF; System.out.println("index " + (j + 1)); ranges[ix] = data[j + 1] & 0xFF; ix++; }

Same applies to similar cases below.

Ok, i think there are 3 places total with this possibility when increment goes by 2 so fixed them all.

Length check won't help here:

int[] data = new int[100]; if (data.length < 8) { return; } int[] destinations = new int[(data.length - 7) / 2]; int[] ranges = new int[(data.length - 7) / 2]; int ix = 0; for (int j = 6; j < data.length - 1; j += 2) { destinations[ix] = data[j] & 0xFF; ranges[ix] = data[j + 1] & 0xFF; ix++; }

Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: Index 46 out of bounds for length 46

We might want to add more test cases to the test.

Length check won't help here:

I think the problem here is in the original code: the array allocation was reused across 3 places and it reserved buffers for ranges and destinations that should load up the remaining of data from the offset 7, but in first two places are trying to load data from offset 6, which causes buffer overflow. The idea of my new fix here is to create a proper buffers.

We might want to add more test cases to the test.

That would be an extensive task, there is a set of the tests being run by third party - and that's them who created the initial issue. I do not think it is practical to bring all the cases they test here. That testing takes a very long time.

azvegint · 2022-12-07T15:24:41Z

src/java.desktop/share/classes/com/sun/media/sound/SoftMainMixer.java

+
+                        case 0x0A:  // Key Based Instrument Control
+                        {
+                            if (data.length < 7 || (data[4] & 0xFF) != 01) {


Neat: 0x01

azvegint · 2022-12-07T15:34:48Z

src/java.desktop/share/classes/com/sun/media/sound/SoftMainMixer.java

                                        break;
                                    }
-                                    default:
-                                        break;
+                                    // http://www.midi.org/about-midi/tuning_extens.shtml


Unrelated to change, but looks like those links become unavailable years ago.

Yes. That's the problem referencing the external site for information. Unfortunately the documentation layout on midi site changed and one needs to read trough all of it to figure out the correct path that both relates to the midi-1.0 standard that we support and is not overwritten by some later addendum making it non-informative.

We might want to track this under some issue, if not done already.

Created JDK-8298319 to track this issue.

azuev-java · 2022-12-08T00:30:12Z

/integrate

openjdk · 2022-12-08T00:32:34Z

Going to push as commit af8fb7e.
Since your change was applied there have been 201 commits pushed to the master branch:

8b69a2e: 8298099: [JVMCI] decouple libgraal from JVMCI module at runtime
8a9911e: 8295803: Console should be usable in jshell and other environments
5d4c71c: 8281236: (D)TLS key exchange named groups
10356e7: 8298303: (fs) temporarily remove Path.getExtension
3b8c7ef: 8157023: Integrate NMT with JFR
e86f31b: 8298301: C2: assert(main_cmp->in(2)->Opcode() == Op_Opaque1) failed: main loop has no opaque node?
8edb98d: 8165943: LineBreakMeasurer does not measure correctly if TextAttribute.TRACKING is set.
3934484: 8298205: Prefer Member Initialization Lists for JFR classes in os_perf.hpp
389b8f4: 8297298: SequenceInputStream should override transferTo
dd7385d: 8298202: [AIX] Dead code elimination removed jfr constructor used by AIX
... and 191 more: https://git.openjdk.org/jdk/compare/105d9d75e84a46400f52fafda2ea00c99c14eaf0...master

Your commit was automatically rebased without conflicts.

openjdk · 2022-12-08T00:32:50Z

@azuev-java Pushed as commit af8fb7e.

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

mrserb · 2022-12-08T22:49:01Z

Thank you for removing that large try/catch block. Looks much better.

aivanov-jdk · 2022-12-12T17:22:47Z

Thank you for removing that large try/catch block. Looks much better.

I agree, the code is clearer now.

I started reviewing the updated code but I haven't finished it. It's integrated now.

8282578: AIOOBE in javax.sound.sampled.Clip

d085cfb

Surround SysEx message processing block with try/catch allowing MIDI subsystem to attempt to ingest the rest of the file. Added test case.

openjdk bot added the rfr Pull request is ready for review label Jun 3, 2022

openjdk bot added the client client-libs-dev@openjdk.org label Jun 3, 2022

prrace approved these changes Jun 3, 2022

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Jun 3, 2022

aivanov-jdk approved these changes Jul 4, 2022

View reviewed changes

Merge branch 'master'

71eb226

bridgekeeper bot closed this Sep 8, 2022

azuev-java added 3 commits September 12, 2022 11:09

Merge branch 'master' of https://github.com/azuev-java/jdk into JDK-8…

c7524e1

…282578

Merge branch 'master' into JDK-8282578

f7ed070

Got rid of the try/catch and using the proper per-case data length

f2ff2d1

analysis.

openjdk bot reopened this Dec 4, 2022

azvegint reviewed Dec 7, 2022

View reviewed changes

Fixed length calculations in the array access checks

85c27f6

Fix the buffers length

6486883

azvegint approved these changes Dec 7, 2022

View reviewed changes

openjdk bot added the integrated Pull request has been integrated label Dec 8, 2022

openjdk bot closed this Dec 8, 2022

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Dec 8, 2022

8282578: AIOOBE in javax.sound.sampled.Clip #9016

8282578: AIOOBE in javax.sound.sampled.Clip #9016

Conversation

azuev-java commented Jun 3, 2022 • edited by openjdk bot Loading

Progress

Issue

Reviewers

Reviewing

bridgekeeper bot commented Jun 3, 2022

openjdk bot commented Jun 3, 2022

mlbridge bot commented Jun 3, 2022 • edited Loading

Webrevs

victordyakov commented Jun 3, 2022

mrserb commented Jun 3, 2022

openjdk bot commented Jun 3, 2022 • edited Loading

azuev-java commented Jun 4, 2022

mrserb commented Jun 4, 2022

bridgekeeper bot commented Jul 3, 2022

AJ1062910 commented Jul 4, 2022 • edited by bridgekeeper bot Loading

bridgekeeper bot commented Aug 11, 2022

bridgekeeper bot commented Sep 8, 2022

azuev-java commented Dec 4, 2022

azuev-java commented Dec 4, 2022

openjdk bot commented Dec 4, 2022

azuev-java commented Dec 7, 2022

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

azuev-java commented Dec 8, 2022

openjdk bot commented Dec 8, 2022

openjdk bot commented Dec 8, 2022

mrserb commented Dec 8, 2022

aivanov-jdk commented Dec 12, 2022

azuev-java commented Jun 3, 2022 •

edited by openjdk bot

Loading

mlbridge bot commented Jun 3, 2022 •

edited

Loading

openjdk bot commented Jun 3, 2022 •

edited

Loading

AJ1062910 commented Jul 4, 2022 •

edited by bridgekeeper bot

Loading