On Mon, 8 Jan 2024 09:04:57 GMT, Alan Bateman wrote:
>> In preparation for when virtual threads can unmount while holding a monitor
>> or unmount when blocking on monitorenter, the implementation of
>> VirtualThread's interrupt method is refactored to avoid parking/blocking
>> while holding th
On Mon, 8 Jan 2024 10:20:33 GMT, Quan Anh Mai wrote:
>>> Thanks for the updates!
>>>
>>> One more idea: Your AVX2 solution has a lot of cost for converting the mask
>>> to a permutation. Might it make sense to split this off into a separate
>>> vector-node, so that it can float out of a loop i
On Fri, 15 Dec 2023 11:15:30 GMT, Lance Andersen wrote:
>> Seeing that `ReadAfterClose.java` uses a binary test vector `crash.jar`, I
>> think it makes sense to include it in this PR, convert it to JUnit and move
>> it into `ReadZip.readAfterClose`.
>>
>> This removes the last remaining binary
> This PR suggests we retire the binary test vectors `ZipFile/input.zip`,
> `ZipFile/input.jar` and `ZipFile/crash.jar`
>
> Binary test vectors are harder to analyze, and sharing test vectors across
> unrelated tests increases maintenance burden. It would be better to have each
> test produce
On Mon, 18 Dec 2023 19:38:55 GMT, Lance Andersen wrote:
>> Eirik Bjørsnøs has updated the pull request incrementally with one
>> additional commit since the last revision:
>>
>> Merge the ReadAfterClose test into ReadZip, converting it to JUnit.
>
> test/jdk/java/util/zip/ZipFile/ReadZip.java
On Mon, 8 Jan 2024 17:30:51 GMT, Lance Andersen wrote:
> Changes look fine. Probably need to change the copyright from 2023 -> 2024
No IP was produced in 2024 for most of these files, but I've updated them
anyhow.
-
PR Comment: https://git.openjdk.org/jdk/pull/17038#issuecomment-1
On Tue, 2 Jan 2024 10:21:11 GMT, Alan Bateman wrote:
> In preparation for when virtual threads can unmount while holding a monitor
> or unmount when blocking on monitorenter, the implementation of
> VirtualThread's interrupt method is refactored to avoid parking/blocking
> while holding the Th
> As recently discussed on core libs, sealed-ness information could be included
> in the Class.toGenericString() output, analagous to how "modifiers" that also
> correspond to JVM access flags are handled.
>
> This is the initial spec, implementation, and test updated needed for that
> change.
On Mon, 8 Jan 2024 07:55:00 GMT, Emanuel Peter wrote:
>>> You are using `VectorMask pred = VectorMask.fromLong(ispecies,
>>> maskctr++);`. That basically systematically iterates over all masks, which
>>> is nice for a correctness test. But that would use different density inside
>>> one test r
> As recently discussed on core libs, sealed-ness information could be included
> in the Class.toGenericString() output, analagous to how "modifiers" that also
> correspond to JVM access flags are handled.
>
> This is the initial spec, implementation, and test updated needed for that
> change.
On Mon, 8 Jan 2024 13:24:55 GMT, Athijegannathan Sundararajan
wrote:
> Adding missing "@ since 9" in javadoc comment of the public classes,
> interfaces and packages of the jdk.dynalink module.
This pull request has now been integrated.
Changeset: 176606d0
Author:Athijegannathan Sundarara
On Mon, 8 Jan 2024 21:13:25 GMT, Viktor Klang wrote:
> As per conversation in the issue.
Hello Viktor, the skara bot very helpfully has noted that this test has been
problemlisted in a couple of files:
> 8309218 is used in problem lists: [test/jdk/ProblemList-zgc.txt,
> test/jdk/ProblemList-g
On Wed, 13 Dec 2023 18:16:56 GMT, Weibing Xiao wrote:
> Better Error Handling for Jar Tool Processing of "@File"
>
> This is a new PR for this PR since the original developer left the team. See
> all of the review history at https://github.com/openjdk/jdk/pull/16423.
>
> Thank you.
This pull
One optimization of Jlink SystemModulesPlugin pre-resolves the module graph for
modules with a main class. It stores the name of the initial module and the
generated `SystemModules` class name in two arrays that can be obtained from
`SystemModulesMap::moduleNames` and `SystemModulesMap::classNa
On Sun, 22 Oct 2023 17:26:52 GMT, Laurent Bourgès wrote:
>> * improved mixed insertion sort (makes whole sorting faster)
>> * introduced Radix which sort shows several times boost of performance and
>> has linear complexity instead of n*ln(n)
>> * improved merging sort for almost sorted data
>>
> As recently discussed on core libs, sealed-ness information could be included
> in the Class.toGenericString() output, analagous to how "modifiers" that also
> correspond to JVM access flags are handled.
>
> This is the initial spec, implementation, and test updated needed for that
> change.
On Wed, 3 Jan 2024 19:44:51 GMT, Pavel Rappo wrote:
>> I think the best place, or least-bad place, to discuss the "modifier"
>> ordering of sealed/non-sealed would be an informative note on
>> Modifier.toString(int) -- "The sealed/non-sealed Java language modifiers are
>> not represented in th
On Mon, 8 Jan 2024 13:24:55 GMT, Athijegannathan Sundararajan
wrote:
> Adding missing "@ since 9" in javadoc comment of the public classes,
> interfaces and packages of the jdk.dynalink module.
Marked as reviewed by attila (Reviewer).
-
PR Review: https://git.openjdk.org/jdk/pull
As per conversation in the issue.
-
Commit messages:
- Updating copyright year for OOMEInAQS.java
- Addresses JBS issue 8309218 but pinning OOMEInAQS to always run on G1
Changes: https://git.openjdk.org/jdk/pull/17313/files
Webrev: https://webrevs.openjdk.org/?repo=jdk&pr=17313&ra
> Re-write the IndexOf code without the use of the pcmpestri instruction, only
> using AVX2 instructions. This change accelerates String.IndexOf on average
> 1.3x for AVX2. The benchmark numbers:
>
>
> BenchmarkScore
> Latest
* Glavo:
> I wish OpenJDK would parse the /etc/ld.so.conf to get the full library
> path list so it would be consistent with the behavior of ld. Can
> anyone consider this suggestion?
Parsing /etc/ld.so.conf is the wrong approach. Even glibc itself does
not look at that file at run time, it's s
On Mon, 8 Jan 2024 15:59:17 GMT, Volker Simonis wrote:
> Shouldn't this be something like tryPresize(this.size() + m.size()) to
> accommodate for the worst case where there are no common keys in this map and
> m?
Its a good callout. Not sure if it is a miss, or intentionally conservative.
> B
On Mon, 8 Jan 2024 13:48:06 GMT, Raffaello Giulietti
wrote:
>> Adds serialization misdeclaration events to JFR.
>
> Raffaello Giulietti has updated the pull request with a new target base due
> to a merge or a rebase. The incremental webrev excludes the unrelated changes
> brought in by the me
On Wed, 8 Nov 2023 11:55:22 GMT, Glavo wrote:
> This PR significantly speeds up decompressing resources in Jimage while
> significantly reducing temporary memory allocations in the process.
>
> This will improve startup speed for runtime images generated using `jlink
> --compress 1` and `jlink
On Fri, 22 Dec 2023 07:55:24 GMT, Eirik Bjørsnøs wrote:
>> ZipInputStream.readEnd currently assumes a Zip64 data descriptor if the
>> number of compressed or uncompressed bytes read from the inflater is larger
>> than the Zip64 magic value.
>>
>> While the ZIP format mandates that the data de
On Sat, 6 Jan 2024 18:01:01 GMT, fabioromano1 wrote:
> The method `MutableBigInteger.divWord(long, int)` can use the algorithm of
> Hacker's Delight (2nd ed), section 9.3, the same used in
> `Long.divideUnsigned(long, long)` and `Long.remainderUnsigned(long, long)`,
> to get the computation fa
On Wed, 20 Dec 2023 18:51:08 GMT, Eirik Bjørsnøs wrote:
> This PR suggests that `Files.setPosixPermissions`as implemented by
> `ZipFileSystem` should preserve the leading seven bits of the 'external file
> attributes' field. These bits contain the 'file type', 'setuid', 'setgid',
> and 'sticky
On Tue, 2 Jan 2024 13:31:01 GMT, Eirik Bjørsnøs wrote:
>> This PR suggests we retire the binary test vectors `ZipFile/input.zip`,
>> `ZipFile/input.jar` and `ZipFile/crash.jar`
>>
>> Binary test vectors are harder to analyze, and sharing test vectors across
>> unrelated tests increases mainte
On Mon, 8 Jan 2024 16:16:50 GMT, Per Minborg wrote:
> This PR proposes to add a clarification that an `Arena` always returns
> zeroed-out segments for `Arena::allocate` methods.
>
> Note that other overloaded methods refer to the abstract `Arena::allocate`
> method via implementation notes.
M
On Thu, 14 Dec 2023 14:55:08 GMT, Eirik Bjørsnøs wrote:
>> This PR suggests we retire the binary test vectors `ZipFile/input.zip`,
>> `ZipFile/input.jar` and `ZipFile/crash.jar`
>>
>> Binary test vectors are harder to analyze, and sharing test vectors across
>> unrelated tests increases maint
On Mon, 8 Jan 2024 13:24:55 GMT, Athijegannathan Sundararajan
wrote:
> Adding missing "@ since 9" in javadoc comment of the public classes,
> interfaces and packages of the jdk.dynalink module.
Marked as reviewed by iris (Reviewer).
-
PR Review: https://git.openjdk.org/jdk/pull/1
On Mon, 8 Jan 2024 16:17:47 GMT, Adam Sotona wrote:
>> ClassFile API performance related improvements have been separated from
>> #17121 into this PR.
>>
>> These improvements are important to minimize performance regression of
>> 8294961: Convert java.base/java.lang.reflect.ProxyGenerator to u
On Mon, 8 Jan 2024 14:43:58 GMT, Chen Liang wrote:
> You need to update the slot counting from `DirectCodeBuilder` and
> `StackMapDecoder` to fully avoid creating any MethodTypeDesc.
It would be good to avoid all bottlenecks, however not all of them have equal
effect.
This patch avoids MTD con
This PR proposes to add a clarification that an `Arena` always returns
zeroed-out segments for `Arena::allocate` methods.
Note that other overloaded methods refer to the abstract `Arena::allocate`
method via implementation notes.
-
Commit messages:
- Add in @implSpec that segments
On Mon, 8 Jan 2024 14:19:12 GMT, Maurizio Cimadamore
wrote:
>> Adam Sotona has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> applied suggested changes
>
> src/java.base/share/classes/jdk/internal/classfile/impl/StackCounter.java
> line 3
> ClassFile API performance related improvements have been separated from
> #17121 into this PR.
>
> These improvements are important to minimize performance regression of
> 8294961: Convert java.base/java.lang.reflect.ProxyGenerator to use the
> Classfile API to generate proxy classes #17121
>
> ClassFile API performance related improvements have been separated from
> #17121 into this PR.
>
> These improvements are important to minimize performance regression of
> 8294961: Convert java.base/java.lang.reflect.ProxyGenerator to use the
> Classfile API to generate proxy classes #17121
>
On Mon, 8 Jan 2024 14:20:17 GMT, Maurizio Cimadamore
wrote:
>> Adam Sotona has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> applied suggested changes
>
> src/java.base/share/classes/jdk/internal/classfile/impl/StackCounter.java
> line 4
> ClassFile API performance related improvements have been separated from
> #17121 into this PR.
>
> These improvements are important to minimize performance regression of
> 8294961: Convert java.base/java.lang.reflect.ProxyGenerator to use the
> Classfile API to generate proxy classes #17121
>
On Wed, 3 Jan 2024 21:29:57 GMT, Joshua Cao wrote:
>> ConcurrentHashMap's copy constructor calls `putAll()` -> `tryPresize()` ->
>> `transfer()`. When coming from the copy constructor, the Map is empty, so
>> there is nothing to transfer. But `transfer()` will still copy all the empty
>> nodes
On Mon, 8 Jan 2024 14:47:48 GMT, Jaikiran Pai wrote:
> With the current proposed change, we not only rely on the Inflater for this
> decision but also rely on ZipEntry itself to tell us whether to read 8 bytes
> or 4 bytes each.
The proposed change resides solely in one single internal/privat
On Sun, 7 Jan 2024 18:24:02 GMT, Chen Liang wrote:
> Also `StackMapDecoder::initFrameFromLocals(ClassEntry, ...)` can benefit from
> 2 changes:
>
> 1. Iterate MTD by index instead of creating copies
I'll add this fix to the performance improvements PR, thanks!
> 2. Pass ClassEntry from reader
On Fri, 5 Jan 2024 23:55:53 GMT, Mandy Chung wrote:
> > Profiling of the benchmarks revealed several slowdowns:
> > ```
> > * many expensive conversions from `Class` to `ClassDesc` to
> > `ClassEntry`, or even more expensive `MethodTypeDesc`
> >
> > * building proxy class from scratch from symb
On Fri, 5 Jan 2024 21:03:28 GMT, Mandy Chung wrote:
>> Adam Sotona has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> StackCounter fix
>
> src/java.base/share/classes/java/lang/reflect/ProxyGenerator.java line 148:
>
>> 146: ge
> java.base java.lang.reflect.ProxyGenerator uses ASM to generate proxy classes.
>
> This patch converts it to use Classfile API.
>
> It is continuation of https://github.com/openjdk/jdk/pull/10991
>
> Any comments and suggestions are welcome.
>
> Please review.
>
> Thank you,
> Adam
Adam Sot
On Fri, 22 Dec 2023 07:55:24 GMT, Eirik Bjørsnøs wrote:
>> ZipInputStream.readEnd currently assumes a Zip64 data descriptor if the
>> number of compressed or uncompressed bytes read from the inflater is larger
>> than the Zip64 magic value.
>>
>> While the ZIP format mandates that the data de
On Fri, 22 Dec 2023 07:55:24 GMT, Eirik Bjørsnøs wrote:
>> ZipInputStream.readEnd currently assumes a Zip64 data descriptor if the
>> number of compressed or uncompressed bytes read from the inflater is larger
>> than the Zip64 magic value.
>>
>> While the ZIP format mandates that the data de
The method `MutableBigInteger.divWord(long, int)` can use the algorithm of
Hacker's Delight (2nd ed), section 9.3, the same used in
`Long.divideUnsigned(long, long)` and `Long.remainderUnsigned(long, long)`, to
get the computation faster.
-
Commit messages:
- Merge branch 'openjdk
On Fri, 22 Dec 2023 07:55:24 GMT, Eirik Bjørsnøs wrote:
>> ZipInputStream.readEnd currently assumes a Zip64 data descriptor if the
>> number of compressed or uncompressed bytes read from the inflater is larger
>> than the Zip64 magic value.
>>
>> While the ZIP format mandates that the data de
On Fri, 22 Dec 2023 07:55:24 GMT, Eirik Bjørsnøs wrote:
>> ZipInputStream.readEnd currently assumes a Zip64 data descriptor if the
>> number of compressed or uncompressed bytes read from the inflater is larger
>> than the Zip64 magic value.
>>
>> While the ZIP format mandates that the data de
On Fri, 22 Dec 2023 07:55:24 GMT, Eirik Bjørsnøs wrote:
>> ZipInputStream.readEnd currently assumes a Zip64 data descriptor if the
>> number of compressed or uncompressed bytes read from the inflater is larger
>> than the Zip64 magic value.
>>
>> While the ZIP format mandates that the data de
On Mon, 8 Jan 2024 13:38:16 GMT, Adam Sotona wrote:
> ClassFile API performance related improvements have been separated from
> #17121 into this PR.
>
> These improvements are important to minimize performance regression of
> 8294961: Convert java.base/java.lang.reflect.ProxyGenerator to use th
> java.base java.lang.reflect.ProxyGenerator uses ASM to generate proxy classes.
>
> This patch converts it to use Classfile API.
>
> It is continuation of https://github.com/openjdk/jdk/pull/10991
>
> Any comments and suggestions are welcome.
>
> Please review.
>
> Thank you,
> Adam
Adam Sot
On Fri, 5 Jan 2024 21:23:02 GMT, Mandy Chung wrote:
>> Adam Sotona has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> StackCounter fix
>
> src/java.base/share/classes/java/lang/reflect/ProxyGenerator.java line 739:
>
>> 737: va
On Mon, 8 Jan 2024 13:38:16 GMT, Adam Sotona wrote:
> ClassFile API performance related improvements have been separated from
> #17121 into this PR.
>
> These improvements are important to minimize performance regression of
> 8294961: Convert java.base/java.lang.reflect.ProxyGenerator to use th
On Mon, 8 Jan 2024 13:38:16 GMT, Adam Sotona wrote:
> ClassFile API performance related improvements have been separated from
> #17121 into this PR.
>
> These improvements are important to minimize performance regression of
> 8294961: Convert java.base/java.lang.reflect.ProxyGenerator to use th
On Fri, 5 Jan 2024 20:50:16 GMT, Mandy Chung wrote:
>> Adam Sotona has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> StackCounter fix
>
> src/java.base/share/classes/java/lang/reflect/ProxyGenerator.java line 135:
>
>> 133: var cc
On Mon, 8 Jan 2024 14:14:29 GMT, Maurizio Cimadamore
wrote:
>> ClassFile API performance related improvements have been separated from
>> #17121 into this PR.
>>
>> These improvements are important to minimize performance regression of
>> 8294961: Convert java.base/java.lang.reflect.ProxyGener
On Fri, 5 Jan 2024 20:45:22 GMT, Mandy Chung wrote:
>> Adam Sotona has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> StackCounter fix
>
> src/java.base/share/classes/java/lang/reflect/ProxyGenerator.java line 112:
>
>> 110: private st
On Sun, 7 Jan 2024 18:16:05 GMT, Chen Liang wrote:
>> Original code has been significant in profiler.
>>
>>
>> MethodTypeDesc desc = MethodTypeDesc.of(toClassDesc(returnType),
>>
>> Arrays.stream(parameterTypes).map(ProxyGenerator::toClassDesc).toArray(ClassDesc[]::new));
>
On Fri, 5 Jan 2024 20:43:00 GMT, Mandy Chung wrote:
>> Adam Sotona has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> StackCounter fix
>
> src/java.base/share/classes/java/lang/reflect/ProxyGenerator.java line 28:
>
>> 26: package java.lan
> java.base java.lang.reflect.ProxyGenerator uses ASM to generate proxy classes.
>
> This patch converts it to use Classfile API.
>
> It is continuation of https://github.com/openjdk/jdk/pull/10991
>
> Any comments and suggestions are welcome.
>
> Please review.
>
> Thank you,
> Adam
Adam Sot
ClassFile API performance related improvements have been separated from #17121
into this PR.
These improvements are important to minimize performance regression of
8294961: Convert java.base/java.lang.reflect.ProxyGenerator to use the
Classfile API to generate proxy classes #17121
Please review
> Adds serialization misdeclaration events to JFR.
Raffaello Giulietti has updated the pull request with a new target base due to
a merge or a rebase. The incremental webrev excludes the unrelated changes
brought in by the merge/rebase. The pull request contains 15 additional commits
since the
On Wed, 3 Jan 2024 12:36:26 GMT, Adam Sotona wrote:
>> java.base java.lang.reflect.ProxyGenerator uses ASM to generate proxy
>> classes.
>>
>> This patch converts it to use Classfile API.
>>
>> It is continuation of https://github.com/openjdk/jdk/pull/10991
>>
>> Any comments and suggestions
Adding missing "@ since 9" in javadoc comment of the public classes, interfaces
and packages of the jdk.dynalink module.
-
Commit messages:
- 8310995: missing @since tags in 36 jdk.dynalink classes
Changes: https://git.openjdk.org/jdk/pull/17305/files
Webrev: https://webrevs.openj
On Mon, 8 Jan 2024 13:24:55 GMT, Athijegannathan Sundararajan
wrote:
> Adding missing "@ since 9" in javadoc comment of the public classes,
> interfaces and packages of the jdk.dynalink module.
LGTM
-
Marked as reviewed by jlaskey (Reviewer).
PR Review: https://git.openjdk.org/j
On Sun, 7 Jan 2024 18:18:32 GMT, Chen Liang wrote:
>> Adam Sotona has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> StackCounter fix
>
> src/java.base/share/classes/jdk/internal/classfile/impl/StackCounter.java
> line 306:
>
>> 304:
On Fri, 5 Jan 2024 23:06:30 GMT, Chen Liang wrote:
>> Discovered while writing a test for #16513 that
>> `ClassSignature.superclassSignature()` does not return a `ClassTypeSig`, yet
>> [JVM
>> Spec](https://docs.oracle.com/javase/specs/jvms/se21/html/jvms-4.html#jvms-4.7.9.1-4100)
>> requires
On Fri, 5 Jan 2024 09:35:34 GMT, Emanuel Peter wrote:
>> Thanks for the comment addition!
>
> Improvement suggestion:
> For a vector with 8 ints, we get `2^8 = 256` many bit patterns for the mask.
> The table has a row for each `mask` value, consisting of 8 ints, which
> provide the valid permu
On Mon, 8 Jan 2024 06:23:46 GMT, Jatin Bhateja wrote:
>> Hi,
>>
>> Patch optimizes non-subword vector compress and expand APIs for x86 AVX2
>> only targets.
>> Upcoming E-core Xeons (Sierra Forest) and Hybrid CPUs only support AVX2
>> instruction set.
>> These are very frequently used APIs in
On Mon, 8 Jan 2024 06:06:20 GMT, Jatin Bhateja wrote:
>> You are using `VectorMask pred = VectorMask.fromLong(ispecies,
>> maskctr++);`.
>> That basically systematically iterates over all masks, which is nice for a
>> correctness test.
>> But that would use different density inside one test run
On Mon, 8 Jan 2024 06:06:22 GMT, Jatin Bhateja wrote:
>> Thanks for the updates!
>>
>> One more idea: Your AVX2 solution has a lot of cost for converting the mask
>> to a permutation. Might it make sense to split this off into a separate
>> vector-node, so that it can float out of a loop if th
On Mon, 8 Jan 2024 09:04:57 GMT, Alan Bateman wrote:
>> In preparation for when virtual threads can unmount while holding a monitor
>> or unmount when blocking on monitorenter, the implementation of
>> VirtualThread's interrupt method is refactored to avoid parking/blocking
>> while holding th
> In preparation for when virtual threads can unmount while holding a monitor
> or unmount when blocking on monitorenter, the implementation of
> VirtualThread's interrupt method is refactored to avoid parking/blocking
> while holding the Thread's interrupt lock. The implementations of
> sun.ni
On Fri, 5 Jan 2024 17:28:30 GMT, Alan Bateman wrote:
>> src/java.base/share/classes/java/nio/channels/spi/AbstractInterruptibleChannel.java
>> line 180:
>>
>>> 178: Thread me = Thread.currentThread();
>>> 179: if (me.isInterrupted()) {
>>> 180: interruptor.interrupt(
On Tue, 12 Dec 2023 10:02:00 GMT, Per Minborg wrote:
> This PR proposes to change the specification for some methods that take
> `long` offsets so that they will throw an `IllegalArgumentException` rather
> than an `IndexOutOfBoundsException` for negative values.
>
> The PR also proposes to fi
The notification method `VirtualThread.notifyJvmtiDisableSuspend` should be
static.
The method disables/enables suspend of the current virtual thread, a no-op if
the current thread is a platform thread. It is confusing for this to be an
instance method, it should be static to make it clearer tha
78 matches
Mail list logo
