From ludovic at rivosinc.com  Mon Jan 16 10:04:22 2023
From: ludovic at rivosinc.com (Ludovic Henry)
Date: Mon, 16 Jan 2023 11:04:22 +0100
Subject: Backports to jdk11u and jdk17u
Message-ID: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>

Hello,

As we are seeing more and more adoption of RISC-V, there is a growing need
for users to have access to more versions of Java available on RISC-V. Many
organizations are still using Java 11 and Java 17 in production and it
would be great to have RISC-V support available for them too. This would
allow them to take advantage of the benefits that RISC-V offers by making
their codebase more portable.

Taking a similar approach to AArch64 backports to JDK8u [1], we would want
to create the openjdk/riscv-port-jdk11u and openjdk/riscv-port-jdk17u
repositories on GitHub. Additionally, we can work with the maintainers of
the JDK 11u [2] and JDK 17u [3] projects to ensure there is a possible path
to getting it merged.

Some questions which I'm not sure about yet:
- Is there a need to backport to JDK 8u?
- What is the engineering cost of the divergence between JDK8u and JDK head
for the backporting effort?

Looking forward to your feedback,

Thank you,
Ludovic

[1] https://hg.openjdk.java.net/aarch64-port/jdk8u
[2]
https://wiki.openjdk.org/display/JDKUpdates/JDK11u#JDK11u-maintainersMaintainers
[3] https://wiki.openjdk.org/display/JDKUpdates/JDK+17u#JDK17u-maintainers
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230116/47aa2050/attachment.htm>

From vladimir.kempik at gmail.com  Mon Jan 16 11:02:13 2023
From: vladimir.kempik at gmail.com (Vladimir Kempik)
Date: Mon, 16 Jan 2023 14:02:13 +0300
Subject: Backports to jdk11u and jdk17u
In-Reply-To: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
Message-ID: <1215A048-282E-4E43-BCDF-D39BBDF1C396@gmail.com>

Hello

I know a few independent developers maintaining their own jdk17 risc-v backport.

Combining it in a one riscv-port-jdk17u will allow to save some resources.

However there is one big difference between general jdk17u and riscv secific one: while jdk17u accepts only bug fixes and minor perf improvements, the riscv-port-jdk17u will have to accept larger patches: as risc-v port is far from being perfect so far, performance-wise is worse than arm64 port.

Regards, Vladimir

> 16 ???. 2023 ?., ? 13:04, Ludovic Henry <ludovic at rivosinc.com> ???????(?):
> 
> Hello,
> 
> As we are seeing more and more adoption of RISC-V, there is a growing need for users to have access to more versions of Java available on RISC-V. Many organizations are still using Java 11 and Java 17 in production and it would be great to have RISC-V support available for them too. This would allow them to take advantage of the benefits that RISC-V offers by making their codebase more portable.
> 
> Taking a similar approach to AArch64 backports to JDK8u [1], we would want to create the openjdk/riscv-port-jdk11u and openjdk/riscv-port-jdk17u repositories on GitHub. Additionally, we can work with the maintainers of the JDK 11u [2] and JDK 17u [3] projects to ensure there is a possible path to getting it merged.
> 
> Some questions which I'm not sure about yet:
> - Is there a need to backport to JDK 8u?
> - What is the engineering cost of the divergence between JDK8u and JDK head for the backporting effort?
> 
> Looking forward to your feedback,
> 
> Thank you,
> Ludovic
> 
> [1] https://hg.openjdk.java.net/aarch64-port/jdk8u
> [2] https://wiki.openjdk.org/display/JDKUpdates/JDK11u#JDK11u-maintainersMaintainers
> [3] https://wiki.openjdk.org/display/JDKUpdates/JDK+17u#JDK17u-maintainers
> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230116/9ce0891c/attachment.htm>

From yangfei at iscas.ac.cn  Mon Jan 16 11:31:29 2023
From: yangfei at iscas.ac.cn (yangfei at iscas.ac.cn)
Date: Mon, 16 Jan 2023 19:31:29 +0800 (GMT+08:00)
Subject: Backports to jdk11u and jdk17u
In-Reply-To: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
Message-ID: <79ab2542.1f08.185ba593e97.Coremail.yangfei@iscas.ac.cn>

Hi,

In fact, there were some discussions before on the list about the backporting work [1].
AFAIK, developers from Huawei already have workable versions of jdk11u &amp; jdk17u on Linux/RISCV platform.
So I guess the rest of the work should be upstreaming. I think Yadong Wang from Huawei should have more details about that?

Regarding backports to jdk8u, I also wonder whether users are interested to deploy jdk8u on riscv.
Xiang Zhang from ISCAS (Institute of Software, Chinese Academy of Sciences) is currently working on backporting to jdk8u
and has achieved some progress. But let me try to get the people involved in this discussion and see. 

Thanks,
Fei

[1] https://mail.openjdk.org/pipermail/jdk-updates-dev/2022-May/014391.html


-----Original Messages-----
From:"Ludovic Henry" <ludovic at rivosinc.com>
Sent Time:2023-01-16 18:04:22 (Monday)
To: riscv-port-dev at openjdk.org
Cc: 
Subject: Backports to jdk11u and jdk17u

Hello,  
  As we are seeing more and more adoption of RISC-V, there is a growing need for users to have access to more versions of Java available on RISC-V. Many organizations are still using Java 11 and Java 17 in production and it would be great to have RISC-V support available for them too. This would allow them to take advantage of the benefits that RISC-V offers by making their codebase more portable.   
  
  
  Taking a similar approach to AArch64 backports to JDK8u [1], we would want to create the openjdk/riscv-port-jdk11u and openjdk/riscv-port-jdk17u repositories on GitHub. Additionally, we can work with the maintainers of the JDK 11u [2] and JDK 17u [3] projects to ensure there is a possible path to getting it merged.  
  
  Some questions which I'm not sure about yet:  
  - Is there a need to backport to JDK 8u?  
  - What is the engineering cost of the divergence between JDK8u and JDK head for the backporting effort?  
  
  Looking forward to your feedback,  
  
  Thank you,  
  Ludovic  
  
  [1] https://hg.openjdk.java.net/aarch64-port/jdk8u   
  [2] https://wiki.openjdk.org/display/JDKUpdates/JDK11u#JDK11u-maintainersMaintainers   
  [3] https://wiki.openjdk.org/display/JDKUpdates/JDK+17u#JDK17u-maintainers


</ludovic at rivosinc.com>

From thomas.stuefe at gmail.com  Mon Jan 16 12:42:13 2023
From: thomas.stuefe at gmail.com (=?UTF-8?Q?Thomas_St=C3=BCfe?=)
Date: Mon, 16 Jan 2023 13:42:13 +0100
Subject: Risc V Backports to jdk11u and jdk17u
In-Reply-To: <1215A048-282E-4E43-BCDF-D39BBDF1C396@gmail.com>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
 <1215A048-282E-4E43-BCDF-D39BBDF1C396@gmail.com>
Message-ID: <CAA-vtUwEhhS+h+g5NDCN=Jv8bkWbPbr3M4paXvFtQY+wz=5Edw@mail.gmail.com>

Adding jdk-updates-dev.

On Mon, Jan 16, 2023 at 12:02 PM Vladimir Kempik <vladimir.kempik at gmail.com>
wrote:

> Hello
>
> I know a few independent developers maintaining their own jdk17 risc-v
> backport.
>
> Combining it in a one riscv-port-jdk17u will allow to save some resources.
>
> However there is one big difference between general jdk17u and riscv
> secific one: while jdk17u accepts only bug fixes and minor perf
> improvements, the riscv-port-jdk17u will have to accept larger patches: as
> risc-v port is far from being perfect so far, performance-wise is worse
> than arm64 port.
>
> Regards, Vladimir
>
> 16 ???. 2023 ?., ? 13:04, Ludovic Henry <ludovic at rivosinc.com> ???????(?):
>
> Hello,
>
> As we are seeing more and more adoption of RISC-V, there is a growing need
> for users to have access to more versions of Java available on RISC-V. Many
> organizations are still using Java 11 and Java 17 in production and it
> would be great to have RISC-V support available for them too. This would
> allow them to take advantage of the benefits that RISC-V offers by making
> their codebase more portable.
>
> Taking a similar approach to AArch64 backports to JDK8u [1], we would want
> to create the openjdk/riscv-port-jdk11u and openjdk/riscv-port-jdk17u
> repositories on GitHub. Additionally, we can work with the maintainers of
> the JDK 11u [2] and JDK 17u [3] projects to ensure there is a possible path
> to getting it merged.
>
> Some questions which I'm not sure about yet:
> - Is there a need to backport to JDK 8u?
> - What is the engineering cost of the divergence between JDK8u and JDK
> head for the backporting effort?
>
> Looking forward to your feedback,
>
> Thank you,
> Ludovic
>
> [1] https://hg.openjdk.java.net/aarch64-port/jdk8u
> [2]
> https://wiki.openjdk.org/display/JDKUpdates/JDK11u#JDK11u-maintainersMaintainers
> [3] https://wiki.openjdk.org/display/JDKUpdates/JDK+17u#JDK17u-maintainers
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230116/c48db9d8/attachment-0001.htm>

From thomas.stuefe at gmail.com  Mon Jan 16 12:44:01 2023
From: thomas.stuefe at gmail.com (=?UTF-8?Q?Thomas_St=C3=BCfe?=)
Date: Mon, 16 Jan 2023 13:44:01 +0100
Subject: Risc V Backports to jdk11u and jdk17u
In-Reply-To: <CAA-vtUwEhhS+h+g5NDCN=Jv8bkWbPbr3M4paXvFtQY+wz=5Edw@mail.gmail.com>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
 <1215A048-282E-4E43-BCDF-D39BBDF1C396@gmail.com>
 <CAA-vtUwEhhS+h+g5NDCN=Jv8bkWbPbr3M4paXvFtQY+wz=5Edw@mail.gmail.com>
Message-ID: <CAA-vtUyeQ-r00=p2pBpzCaXGoi6585WVMeysbNnU_5e71Xr0MA@mail.gmail.com>

Guess I should add it, then...

On Mon, Jan 16, 2023 at 1:42 PM Thomas St?fe <thomas.stuefe at gmail.com>
wrote:

> Adding jdk-updates-dev.
>
> On Mon, Jan 16, 2023 at 12:02 PM Vladimir Kempik <
> vladimir.kempik at gmail.com> wrote:
>
>> Hello
>>
>> I know a few independent developers maintaining their own jdk17 risc-v
>> backport.
>>
>> Combining it in a one riscv-port-jdk17u will allow to save some resources.
>>
>> However there is one big difference between general jdk17u and riscv
>> secific one: while jdk17u accepts only bug fixes and minor perf
>> improvements, the riscv-port-jdk17u will have to accept larger patches: as
>> risc-v port is far from being perfect so far, performance-wise is worse
>> than arm64 port.
>>
>> Regards, Vladimir
>>
>> 16 ???. 2023 ?., ? 13:04, Ludovic Henry <ludovic at rivosinc.com>
>> ???????(?):
>>
>> Hello,
>>
>> As we are seeing more and more adoption of RISC-V, there is a growing
>> need for users to have access to more versions of Java available on RISC-V.
>> Many organizations are still using Java 11 and Java 17 in production and it
>> would be great to have RISC-V support available for them too. This would
>> allow them to take advantage of the benefits that RISC-V offers by making
>> their codebase more portable.
>>
>> Taking a similar approach to AArch64 backports to JDK8u [1], we would
>> want to create the openjdk/riscv-port-jdk11u and openjdk/riscv-port-jdk17u
>> repositories on GitHub. Additionally, we can work with the maintainers of
>> the JDK 11u [2] and JDK 17u [3] projects to ensure there is a possible path
>> to getting it merged.
>>
>> Some questions which I'm not sure about yet:
>> - Is there a need to backport to JDK 8u?
>> - What is the engineering cost of the divergence between JDK8u and JDK
>> head for the backporting effort?
>>
>> Looking forward to your feedback,
>>
>> Thank you,
>> Ludovic
>>
>> [1] https://hg.openjdk.java.net/aarch64-port/jdk8u
>> [2]
>> https://wiki.openjdk.org/display/JDKUpdates/JDK11u#JDK11u-maintainersMaintainers
>> [3]
>> https://wiki.openjdk.org/display/JDKUpdates/JDK+17u#JDK17u-maintainers
>>
>>
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230116/ff85829e/attachment.htm>

From zhangxiang at iscas.ac.cn  Mon Jan 16 13:33:24 2023
From: zhangxiang at iscas.ac.cn (=?UTF-8?B?56ug57+U?=)
Date: Mon, 16 Jan 2023 21:33:24 +0800 (GMT+08:00)
Subject: Backports to jdk11u and jdk17u
In-Reply-To: <79ab2542.1f08.185ba593e97.Coremail.yangfei@iscas.ac.cn>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
 <79ab2542.1f08.185ba593e97.Coremail.yangfei@iscas.ac.cn>
Message-ID: <2b9d63dc.2ad6.185bac8df55.Coremail.zhangxiang@iscas.ac.cn>

Hello,
I'm Xiang Zhang from the Institute of Software, Chinese Academy of Sciences. I'm backporting jdk8u for RISC-V support. Here is my development repository:
https://github.com/zhangxiang-plct/jdk8u
The current branch for template interpreter backporting is:
https://github.com/zhangxiang-plct/jdk8u/tree/Fix_javac-temp
As far as the current progress is concerned, the interpreter has been successfully ported.
For example, I can successfully run "Hello World" under -Xint.
I am currently testing benchmark workloads like dacapo and SPECjvm2008. 
My next step is to backport C1&amp;C2 and I'm debugging related code for porting C1&amp;C2.

Thank you,
Xiang Zhang


&gt; -----????-----
&gt; ???: yangfei at iscas.ac.cn
&gt; ????: 2023-01-16 19:31:29 (???)
&gt; ???: "Ludovic Henry" <ludovic at rivosinc.com>
&gt; ??: riscv-port-dev at openjdk.org, yadonn.wang at huawei.com, zhangxiang at iscas.ac.cn
&gt; ??: Re: Backports to jdk11u and jdk17u
&gt; 
&gt; Hi,
&gt; 
&gt; In fact, there were some discussions before on the list about the backporting work [1].
&gt; AFAIK, developers from Huawei already have workable versions of jdk11u &amp; jdk17u on Linux/RISCV platform.
&gt; So I guess the rest of the work should be upstreaming. I think Yadong Wang from Huawei should have more details about that?
&gt; 
&gt; Regarding backports to jdk8u, I also wonder whether users are interested to deploy jdk8u on riscv.
&gt; Xiang Zhang from ISCAS (Institute of Software, Chinese Academy of Sciences) is currently working on backporting to jdk8u
&gt; and has achieved some progress. But let me try to get the people involved in this discussion and see. 
&gt; 
&gt; Thanks,
&gt; Fei
&gt; 
&gt; [1] https://mail.openjdk.org/pipermail/jdk-updates-dev/2022-May/014391.html
&gt; 
&gt; 
&gt; -----Original Messages-----
&gt; From:"Ludovic Henry" <ludovic at rivosinc.com>
&gt; Sent Time:2023-01-16 18:04:22 (Monday)
&gt; To: riscv-port-dev at openjdk.org
&gt; Cc: 
&gt; Subject: Backports to jdk11u and jdk17u
&gt; 
&gt; Hello,  
&gt;   As we are seeing more and more adoption of RISC-V, there is a growing need for users to have access to more versions of Java available on RISC-V. Many organizations are still using Java 11 and Java 17 in production and it would be great to have RISC-V support available for them too. This would allow them to take advantage of the benefits that RISC-V offers by making their codebase more portable.   
&gt;   
&gt;   
&gt;   Taking a similar approach to AArch64 backports to JDK8u [1], we would want to create the openjdk/riscv-port-jdk11u and openjdk/riscv-port-jdk17u repositories on GitHub. Additionally, we can work with the maintainers of the JDK 11u [2] and JDK 17u [3] projects to ensure there is a possible path to getting it merged.  
&gt;   
&gt;   Some questions which I'm not sure about yet:  
&gt;   - Is there a need to backport to JDK 8u?  
&gt;   - What is the engineering cost of the divergence between JDK8u and JDK head for the backporting effort?  
&gt;   
&gt;   Looking forward to your feedback,  
&gt;   
&gt;   Thank you,  
&gt;   Ludovic  
&gt;   
&gt;   [1] https://hg.openjdk.java.net/aarch64-port/jdk8u   
&gt;   [2] https://wiki.openjdk.org/display/JDKUpdates/JDK11u#JDK11u-maintainersMaintainers   
&gt;   [3] https://wiki.openjdk.org/display/JDKUpdates/JDK+17u#JDK17u-maintainers
&gt; 
&gt; 
&gt; </ludovic at rivosinc.com>
</ludovic at rivosinc.com>

From ludovic at rivosinc.com  Mon Jan 16 15:36:33 2023
From: ludovic at rivosinc.com (Ludovic Henry)
Date: Mon, 16 Jan 2023 16:36:33 +0100
Subject: Backports to jdk11u and jdk17u
In-Reply-To: <2b9d63dc.2ad6.185bac8df55.Coremail.zhangxiang@iscas.ac.cn>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
 <79ab2542.1f08.185ba593e97.Coremail.yangfei@iscas.ac.cn>
 <2b9d63dc.2ad6.185bac8df55.Coremail.zhangxiang@iscas.ac.cn>
Message-ID: <CADT88DvmwLiKZDxw4BX-zrN0p5aUYeKqTr6TORXXAz+jjRS8dw@mail.gmail.com>

> However there is one big difference between general jdk17u and riscv
secific one: while jdk17u accepts only bug fixes and minor perf
improvements, the riscv-port-jdk17u will have to accept larger patches: as
risc-v port is far from being perfect so far, performance-wise is worse
than arm64 port.

I expect to initiate the backport in a dedicated openjdk/riscv-port-jdk17u
repository for exactly that. Until the riscv port is considered very stable
and performant, we likely want to keep it in a separate repository. When
stable, we can work with the relevant maintainers of jdk8u, jdk11u and
jdk17u for possible integrations.

> In fact, there were some discussions before on the list about the
backporting work [1].

I missed this one, thanks for pointing it out.

> I'm Xiang Zhang from the Institute of Software, Chinese Academy of
Sciences. I'm backporting jdk8u for RISC-V support. Here is my development
repository:
> https://github.com/zhangxiang-plct/jdk8u

If there is a riscv-port-jdk8u project, would you be interested to
contribute there directly?

In the meantime, I'll contact skara-dev to figure out what it would take to
create the riscv-port-jdk11u and riscv-port-jdk17u repositories.

Thanks,
Ludovic


On Mon, Jan 16, 2023 at 2:33 PM ?? <zhangxiang at iscas.ac.cn> wrote:

> Hello,
> I'm Xiang Zhang from the Institute of Software, Chinese Academy of
> Sciences. I'm backporting jdk8u for RISC-V support. Here is my development
> repository:
> https://github.com/zhangxiang-plct/jdk8u
> The current branch for template interpreter backporting is:
> https://github.com/zhangxiang-plct/jdk8u/tree/Fix_javac-temp
> As far as the current progress is concerned, the interpreter has been
> successfully ported.
> For example, I can successfully run "Hello World" under -Xint.
> I am currently testing benchmark workloads like dacapo and SPECjvm2008.
> My next step is to backport C1&amp;C2 and I'm debugging related code for
> porting C1&amp;C2.
>
> Thank you,
> Xiang Zhang
>
>
> &gt; -----????-----
> &gt; ???: yangfei at iscas.ac.cn
> &gt; ????: 2023-01-16 19:31:29 (???)
> &gt; ???: "Ludovic Henry" <ludovic at rivosinc.com>
> &gt; ??: riscv-port-dev at openjdk.org, yadonn.wang at huawei.com,
> zhangxiang at iscas.ac.cn
> &gt; ??: Re: Backports to jdk11u and jdk17u
> &gt;
> &gt; Hi,
> &gt;
> &gt; In fact, there were some discussions before on the list about the
> backporting work [1].
> &gt; AFAIK, developers from Huawei already have workable versions of
> jdk11u &amp; jdk17u on Linux/RISCV platform.
> &gt; So I guess the rest of the work should be upstreaming. I think Yadong
> Wang from Huawei should have more details about that?
> &gt;
> &gt; Regarding backports to jdk8u, I also wonder whether users are
> interested to deploy jdk8u on riscv.
> &gt; Xiang Zhang from ISCAS (Institute of Software, Chinese Academy of
> Sciences) is currently working on backporting to jdk8u
> &gt; and has achieved some progress. But let me try to get the people
> involved in this discussion and see.
> &gt;
> &gt; Thanks,
> &gt; Fei
> &gt;
> &gt; [1]
> https://mail.openjdk.org/pipermail/jdk-updates-dev/2022-May/014391.html
> &gt;
> &gt;
> &gt; -----Original Messages-----
> &gt; From:"Ludovic Henry" <ludovic at rivosinc.com>
> &gt; Sent Time:2023-01-16 18:04:22 (Monday)
> &gt; To: riscv-port-dev at openjdk.org
> &gt; Cc:
> &gt; Subject: Backports to jdk11u and jdk17u
> &gt;
> &gt; Hello,
> &gt;   As we are seeing more and more adoption of RISC-V, there is a
> growing need for users to have access to more versions of Java available on
> RISC-V. Many organizations are still using Java 11 and Java 17 in
> production and it would be great to have RISC-V support available for them
> too. This would allow them to take advantage of the benefits that RISC-V
> offers by making their codebase more portable.
> &gt;
> &gt;
> &gt;   Taking a similar approach to AArch64 backports to JDK8u [1], we
> would want to create the openjdk/riscv-port-jdk11u and
> openjdk/riscv-port-jdk17u repositories on GitHub. Additionally, we can work
> with the maintainers of the JDK 11u [2] and JDK 17u [3] projects to ensure
> there is a possible path to getting it merged.
> &gt;
> &gt;   Some questions which I'm not sure about yet:
> &gt;   - Is there a need to backport to JDK 8u?
> &gt;   - What is the engineering cost of the divergence between JDK8u and
> JDK head for the backporting effort?
> &gt;
> &gt;   Looking forward to your feedback,
> &gt;
> &gt;   Thank you,
> &gt;   Ludovic
> &gt;
> &gt;   [1] https://hg.openjdk.java.net/aarch64-port/jdk8u
> &gt;   [2]
> https://wiki.openjdk.org/display/JDKUpdates/JDK11u#JDK11u-maintainersMaintainers
>
> &gt;   [3]
> https://wiki.openjdk.org/display/JDKUpdates/JDK+17u#JDK17u-maintainers
> &gt
> <https://wiki.openjdk.org/display/JDKUpdates/JDK+17u#JDK17u-maintainers&gt>;
>
> &gt;
> &gt; </ludovic at rivosinc.com>
> </ludovic at rivosinc.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230116/5339936d/attachment-0001.htm>

From yadonn.wang at huawei.com  Tue Jan 17 01:34:11 2023
From: yadonn.wang at huawei.com (wangyadong (E))
Date: Tue, 17 Jan 2023 01:34:11 +0000
Subject: Backports to jdk11u and jdk17u
In-Reply-To: <CADT88DvmwLiKZDxw4BX-zrN0p5aUYeKqTr6TORXXAz+jjRS8dw@mail.gmail.com>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
 <79ab2542.1f08.185ba593e97.Coremail.yangfei@iscas.ac.cn>
 <2b9d63dc.2ad6.185bac8df55.Coremail.zhangxiang@iscas.ac.cn>
 <CADT88DvmwLiKZDxw4BX-zrN0p5aUYeKqTr6TORXXAz+jjRS8dw@mail.gmail.com>
Message-ID: <1d1bb8984eda4ae88b749e51909dfc1b@huawei.com>

>I expect to initiate the backport in a dedicated openjdk/riscv-port-jdk17u repository for exactly that. Until the riscv port is considered very stable and performant, we likely want to keep it in a separate repository. When stable, we can work with the relevant maintainers of jdk8u, jdk11u and jdk17u for possible integrations.

Agree. It?s a good idea to make a stable backport before upstream to jdk17u.
We discussed with Goetz this May [1], and I have an backport version [2] with more changes in shared code protected by #ifdef RISCV that would meet Goetz?s concerns.
I can contribute the initial backport when the riscv-port-jdk17u repo is ready, and we can backport patches we think are necessary until it is ready to go to jdk17u-dev.

Do you think that?s okay?

[1] https://mail.openjdk.org/pipermail/jdk-updates-dev/2022-May/014396.html
[2] https://github.com/yadongw/jdk17u-dev/tree/backport-riscv-port-v5

From: Ludovic Henry [mailto:ludovic at rivosinc.com]
Sent: Monday, January 16, 2023 11:37 PM
To: ?? <zhangxiang at iscas.ac.cn>
Cc: yangfei at iscas.ac.cn; riscv-port-dev at openjdk.org; wangyadong (E) <yadonn.wang at huawei.com>; jdk-updates-dev at openjdk.java.net
Subject: Re: Re: Backports to jdk11u and jdk17u

> However there is one big difference between general jdk17u and riscv secific one: while jdk17u accepts only bug fixes and minor perf improvements, the riscv-port-jdk17u will have to accept larger patches: as risc-v port is far from being perfect so far, performance-wise is worse than arm64 port.

I expect to initiate the backport in a dedicated openjdk/riscv-port-jdk17u repository for exactly that. Until the riscv port is considered very stable and performant, we likely want to keep it in a separate repository. When stable, we can work with the relevant maintainers of jdk8u, jdk11u and jdk17u for possible integrations.

> In fact, there were some discussions before on the list about the backporting work [1].

I missed this one, thanks for pointing it out.

> I'm Xiang Zhang from the Institute of Software, Chinese Academy of Sciences. I'm backporting jdk8u for RISC-V support. Here is my development repository:
> https://github.com/zhangxiang-plct/jdk8u

If there is a riscv-port-jdk8u project, would you be interested to contribute there directly?

In the meantime, I'll contact skara-dev to figure out what it would take to create the riscv-port-jdk11u and riscv-port-jdk17u repositories.

Thanks,
Ludovic


On Mon, Jan 16, 2023 at 2:33 PM ?? <zhangxiang at iscas.ac.cn<mailto:zhangxiang at iscas.ac.cn>> wrote:
Hello,
I'm Xiang Zhang from the Institute of Software, Chinese Academy of Sciences. I'm backporting jdk8u for RISC-V support. Here is my development repository:
https://github.com/zhangxiang-plct/jdk8u
The current branch for template interpreter backporting is:
https://github.com/zhangxiang-plct/jdk8u/tree/Fix_javac-temp
As far as the current progress is concerned, the interpreter has been successfully ported.
For example, I can successfully run "Hello World" under -Xint.
I am currently testing benchmark workloads like dacapo and SPECjvm2008.
My next step is to backport C1&amp;C2 and I'm debugging related code for porting C1&amp;C2.

Thank you,
Xiang Zhang


&gt; -----????-----
&gt; ???: yangfei at iscas.ac.cn<mailto:yangfei at iscas.ac.cn>
&gt; ????: 2023-01-16 19:31:29 (???)
&gt; ???: "Ludovic Henry" <ludovic at rivosinc.com<mailto:ludovic at rivosinc.com>>
&gt; ??: riscv-port-dev at openjdk.org<mailto:riscv-port-dev at openjdk.org>, yadonn.wang at huawei.com<mailto:yadonn.wang at huawei.com>, zhangxiang at iscas.ac.cn<mailto:zhangxiang at iscas.ac.cn>
&gt; ??: Re: Backports to jdk11u and jdk17u
&gt;
&gt; Hi,
&gt;
&gt; In fact, there were some discussions before on the list about the backporting work [1].
&gt; AFAIK, developers from Huawei already have workable versions of jdk11u &amp; jdk17u on Linux/RISCV platform.
&gt; So I guess the rest of the work should be upstreaming. I think Yadong Wang from Huawei should have more details about that?
&gt;
&gt; Regarding backports to jdk8u, I also wonder whether users are interested to deploy jdk8u on riscv.
&gt; Xiang Zhang from ISCAS (Institute of Software, Chinese Academy of Sciences) is currently working on backporting to jdk8u
&gt; and has achieved some progress. But let me try to get the people involved in this discussion and see.
&gt;
&gt; Thanks,
&gt; Fei
&gt;
&gt; [1] https://mail.openjdk.org/pipermail/jdk-updates-dev/2022-May/014391.html
&gt;
&gt;
&gt; -----Original Messages-----
&gt; From:"Ludovic Henry" <ludovic at rivosinc.com<mailto:ludovic at rivosinc.com>>
&gt; Sent Time:2023-01-16 18:04:22 (Monday)
&gt; To: riscv-port-dev at openjdk.org<mailto:riscv-port-dev at openjdk.org>
&gt; Cc:
&gt; Subject: Backports to jdk11u and jdk17u
&gt;
&gt; Hello,
&gt;   As we are seeing more and more adoption of RISC-V, there is a growing need for users to have access to more versions of Java available on RISC-V. Many organizations are still using Java 11 and Java 17 in production and it would be great to have RISC-V support available for them too. This would allow them to take advantage of the benefits that RISC-V offers by making their codebase more portable.
&gt;
&gt;
&gt;   Taking a similar approach to AArch64 backports to JDK8u [1], we would want to create the openjdk/riscv-port-jdk11u and openjdk/riscv-port-jdk17u repositories on GitHub. Additionally, we can work with the maintainers of the JDK 11u [2] and JDK 17u [3] projects to ensure there is a possible path to getting it merged.
&gt;
&gt;   Some questions which I'm not sure about yet:
&gt;   - Is there a need to backport to JDK 8u?
&gt;   - What is the engineering cost of the divergence between JDK8u and JDK head for the backporting effort?
&gt;
&gt;   Looking forward to your feedback,
&gt;
&gt;   Thank you,
&gt;   Ludovic
&gt;
&gt;   [1] https://hg.openjdk.java.net/aarch64-port/jdk8u
&gt;   [2] https://wiki.openjdk.org/display/JDKUpdates/JDK11u#JDK11u-maintainersMaintainers
&gt;   [3] https://wiki.openjdk.org/display/JDKUpdates/JDK+17u#JDK17u-maintainers
&gt;
&gt;
&gt; </ludovic at rivosinc.com<mailto:ludovic at rivosinc.com>>
</ludovic at rivosinc.com<mailto:ludovic at rivosinc.com>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230117/b27b9164/attachment-0001.htm>

From zhangxiang at iscas.ac.cn  Tue Jan 17 02:00:11 2023
From: zhangxiang at iscas.ac.cn (zhangxiang at iscas.ac.cn)
Date: Tue, 17 Jan 2023 10:00:11 +0800
Subject: Backports to jdk11u and jdk17u
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>, 
 <79ab2542.1f08.185ba593e97.Coremail.yangfei@iscas.ac.cn>, 
 <2b9d63dc.2ad6.185bac8df55.Coremail.zhangxiang@iscas.ac.cn>, 
 <CADT88DvmwLiKZDxw4BX-zrN0p5aUYeKqTr6TORXXAz+jjRS8dw@mail.gmail.com>
Message-ID: <202301171000104762353@iscas.ac.cn>

>If there is a riscv-port-jdk8u project, would you be interested to contribute there directly?

Thank you for your proposal! And it would be my honor to contribute to this project.
My next step is to port C1 & C2, I would be happy if more people join this project.

Thanks,
Xiang Zhang




zhangxiang at iscas.ac.cn
 
From: Ludovic Henry
Date: 2023-01-16 23:36
To: ??
CC: yangfei; riscv-port-dev; yadonn.wang; jdk-updates-dev
Subject: Re: Re: Backports to jdk11u and jdk17u
> However there is one big difference between general jdk17u and riscv secific one: while jdk17u accepts only bug fixes and minor perf improvements, the riscv-port-jdk17u will have to accept larger patches: as risc-v port is far from being perfect so far, performance-wise is worse than arm64 port.

I expect to initiate the backport in a dedicated openjdk/riscv-port-jdk17u repository for exactly that. Until the riscv port is considered very stable and performant, we likely want to keep it in a separate repository. When stable, we can work with the relevant maintainers of jdk8u, jdk11u and jdk17u for possible integrations.

> In fact, there were some discussions before on the list about the backporting work [1].

I missed this one, thanks for pointing it out. 

> I'm Xiang Zhang from the Institute of Software, Chinese Academy of Sciences. I'm backporting jdk8u for RISC-V support. Here is my development repository:
> https://github.com/zhangxiang-plct/jdk8u

If there is a riscv-port-jdk8u project, would you be interested to contribute there directly?

In the meantime, I'll contact skara-dev to figure out what it would take to create the riscv-port-jdk11u and riscv-port-jdk17u repositories.

Thanks,
Ludovic


On Mon, Jan 16, 2023 at 2:33 PM ?? <zhangxiang at iscas.ac.cn> wrote:
Hello,
I'm Xiang Zhang from the Institute of Software, Chinese Academy of Sciences. I'm backporting jdk8u for RISC-V support. Here is my development repository:
https://github.com/zhangxiang-plct/jdk8u
The current branch for template interpreter backporting is:
https://github.com/zhangxiang-plct/jdk8u/tree/Fix_javac-temp
As far as the current progress is concerned, the interpreter has been successfully ported.
For example, I can successfully run "Hello World" under -Xint.
I am currently testing benchmark workloads like dacapo and SPECjvm2008. 
My next step is to backport C1&amp;C2 and I'm debugging related code for porting C1&amp;C2.

Thank you,
Xiang Zhang


&gt; -----????-----
&gt; ???: yangfei at iscas.ac.cn
&gt; ????: 2023-01-16 19:31:29 (???)
&gt; ???: "Ludovic Henry" <ludovic at rivosinc.com>
&gt; ??: riscv-port-dev at openjdk.org, yadonn.wang at huawei.com, zhangxiang at iscas.ac.cn
&gt; ??: Re: Backports to jdk11u and jdk17u
&gt; 
&gt; Hi,
&gt; 
&gt; In fact, there were some discussions before on the list about the backporting work [1].
&gt; AFAIK, developers from Huawei already have workable versions of jdk11u &amp; jdk17u on Linux/RISCV platform.
&gt; So I guess the rest of the work should be upstreaming. I think Yadong Wang from Huawei should have more details about that?
&gt; 
&gt; Regarding backports to jdk8u, I also wonder whether users are interested to deploy jdk8u on riscv.
&gt; Xiang Zhang from ISCAS (Institute of Software, Chinese Academy of Sciences) is currently working on backporting to jdk8u
&gt; and has achieved some progress. But let me try to get the people involved in this discussion and see. 
&gt; 
&gt; Thanks,
&gt; Fei
&gt; 
&gt; [1] https://mail.openjdk.org/pipermail/jdk-updates-dev/2022-May/014391.html
&gt; 
&gt; 
&gt; -----Original Messages-----
&gt; From:"Ludovic Henry" <ludovic at rivosinc.com>
&gt; Sent Time:2023-01-16 18:04:22 (Monday)
&gt; To: riscv-port-dev at openjdk.org
&gt; Cc: 
&gt; Subject: Backports to jdk11u and jdk17u
&gt; 
&gt; Hello,  
&gt;   As we are seeing more and more adoption of RISC-V, there is a growing need for users to have access to more versions of Java available on RISC-V. Many organizations are still using Java 11 and Java 17 in production and it would be great to have RISC-V support available for them too. This would allow them to take advantage of the benefits that RISC-V offers by making their codebase more portable.   
&gt;   
&gt;   
&gt;   Taking a similar approach to AArch64 backports to JDK8u [1], we would want to create the openjdk/riscv-port-jdk11u and openjdk/riscv-port-jdk17u repositories on GitHub. Additionally, we can work with the maintainers of the JDK 11u [2] and JDK 17u [3] projects to ensure there is a possible path to getting it merged.  
&gt;   
&gt;   Some questions which I'm not sure about yet:  
&gt;   - Is there a need to backport to JDK 8u?  
&gt;   - What is the engineering cost of the divergence between JDK8u and JDK head for the backporting effort?  
&gt;   
&gt;   Looking forward to your feedback,  
&gt;   
&gt;   Thank you,  
&gt;   Ludovic  
&gt;   
&gt;   [1] https://hg.openjdk.java.net/aarch64-port/jdk8u   
&gt;   [2] https://wiki.openjdk.org/display/JDKUpdates/JDK11u#JDK11u-maintainersMaintainers   
&gt;   [3] https://wiki.openjdk.org/display/JDKUpdates/JDK+17u#JDK17u-maintainers
&gt; 
&gt; 
&gt; </ludovic at rivosinc.com>
</ludovic at rivosinc.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230117/67d37a68/attachment.htm>

From yangfei at iscas.ac.cn  Tue Jan 17 08:23:06 2023
From: yangfei at iscas.ac.cn (yangfei at iscas.ac.cn)
Date: Tue, 17 Jan 2023 16:23:06 +0800 (GMT+08:00)
Subject: Backports to jdk11u and jdk17u
In-Reply-To: <CADT88DvmwLiKZDxw4BX-zrN0p5aUYeKqTr6TORXXAz+jjRS8dw@mail.gmail.com>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
 <79ab2542.1f08.185ba593e97.Coremail.yangfei@iscas.ac.cn>
 <2b9d63dc.2ad6.185bac8df55.Coremail.zhangxiang@iscas.ac.cn>
 <CADT88DvmwLiKZDxw4BX-zrN0p5aUYeKqTr6TORXXAz+jjRS8dw@mail.gmail.com>
Message-ID: <5e09da27.4add.185bed323a6.Coremail.yangfei@iscas.ac.cn>

Hi,

-----Original Messages-----
From:"Ludovic Henry" <ludovic at rivosinc.com>
Sent Time:2023-01-16 23:36:33 (Monday)
To: "??" <zhangxiang at iscas.ac.cn>
Cc: yangfei at iscas.ac.cn, riscv-port-dev at openjdk.org, yadonn.wang at huawei.com, jdk-updates-dev at openjdk.java.net
Subject: Re: Re: Backports to jdk11u and jdk17u

&gt;&gt; However there is one big difference between general jdk17u and riscv secific one: while jdk17u accepts only bug fixes and minor perf improvements, the riscv-port-jdk17u will have to accept larger patches: as risc-v port is far from being perfect so far, performance-wise is worse than arm64 port.  
&gt;  I expect to initiate the backport in a dedicated openjdk/riscv-port-jdk17u repository for exactly that. Until the riscv port is considered very stable and performant, we likely want to keep it in a separate repository. When stable, we can work with the relevant maintainers of jdk8u, jdk11u and jdk17u for possible integrations.  

That sounds reasonable to me as we are able to find more time to look at the jvm performance issues on RISCV now. 

&gt;  &gt; In fact, there were some discussions before on the list about the backporting work [1].  
&gt; 
&gt;  I missed this one, thanks for pointing it out.   
&gt; 
&gt;  &gt; I'm Xiang Zhang from the Institute of Software, Chinese Academy of Sciences. I'm backporting jdk8u for RISC-V support. Here is my development repository:  
&gt;&gt; https://github.com/zhangxiang-plct/jdk8u   
&gt;  If there is a riscv-port-jdk8u project, would you be interested to contribute there directly?  
&gt; 
&gt;  In the meantime, I'll contact skara-dev to figure out what it would take to create the riscv-port-jdk11u and riscv-port-jdk17u repositories.  

Besides the repositories needed, I also wonder the correct workflow for future reviewing &amp; merging of backporting PRs on those repos. 
I would like to hear comments/suggestions from the jdk update maintainers. 

Thanks,
Fei </zhangxiang at iscas.ac.cn></ludovic at rivosinc.com>

From goetz.lindenmaier at sap.com  Tue Jan 17 15:06:35 2023
From: goetz.lindenmaier at sap.com (Lindenmaier, Goetz)
Date: Tue, 17 Jan 2023 15:06:35 +0000
Subject: Backports to jdk11u and jdk17u
In-Reply-To: <5e09da27.4add.185bed323a6.Coremail.yangfei@iscas.ac.cn>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
 <79ab2542.1f08.185ba593e97.Coremail.yangfei@iscas.ac.cn>
 <2b9d63dc.2ad6.185bac8df55.Coremail.zhangxiang@iscas.ac.cn>
 <CADT88DvmwLiKZDxw4BX-zrN0p5aUYeKqTr6TORXXAz+jjRS8dw@mail.gmail.com>
 <5e09da27.4add.185bed323a6.Coremail.yangfei@iscas.ac.cn>
Message-ID: <PAXPR02MB8245971774A3A617415AA6B8ECC69@PAXPR02MB8245.eurprd02.prod.outlook.com>

Hi,

As pointed out in the mail you cited below, not breaking 
existing code is the major concern of such a backport in 
the view of the maintainers.  

Shared changes should be kept to a minimum.
Especially in 11, only enabling shared changes are
eligible, not changes that aim to improve the performance.

It would be great to see a complete list of 
changes that need to be backported. These should 
make up a functioning port. A project repo can serve
the purpose to collect these changes and prove
they are a complete set.
We can then look at this list of changes and give a 
basic ok for these to be integrated in 17u/11u. Then
you can run the normal backport process for them.

Best regards, 
  Goetz.








> -----Original Message-----
> From: jdk-updates-dev <jdk-updates-dev-retn at openjdk.org> On Behalf Of
> yangfei at iscas.ac.cn
> Sent: Tuesday, January 17, 2023 9:23 AM
> To: Ludovic Henry <ludovic at rivosinc.com>
> Cc: ?? <zhangxiang at iscas.ac.cn>; riscv-port-dev at openjdk.org;
> yadonn.wang at huawei.com; jdk-updates-dev at openjdk.java.net
> Subject: Re: Re: Re: Backports to jdk11u and jdk17u
> 
> [Some people who received this message don't often get email from
> yangfei at iscas.ac.cn. Learn why this is important at
> https://aka.ms/LearnAboutSenderIdentification ]
> 
> Hi,
> 
> -----Original Messages-----
> From:"Ludovic Henry" <ludovic at rivosinc.com>
> Sent Time:2023-01-16 23:36:33 (Monday)
> To: "??" <zhangxiang at iscas.ac.cn>
> Cc: yangfei at iscas.ac.cn, riscv-port-dev at openjdk.org,
> yadonn.wang at huawei.com, jdk-updates-dev at openjdk.java.net
> Subject: Re: Re: Backports to jdk11u and jdk17u
> 
> &gt;&gt; However there is one big difference between general jdk17u and
> riscv secific one: while jdk17u accepts only bug fixes and minor perf
> improvements, the riscv-port-jdk17u will have to accept larger patches: as
> risc-v port is far from being perfect so far, performance-wise is worse than
> arm64 port.
> &gt;  I expect to initiate the backport in a dedicated openjdk/riscv-port-
> jdk17u repository for exactly that. Until the riscv port is considered very
> stable and performant, we likely want to keep it in a separate repository.
> When stable, we can work with the relevant maintainers of jdk8u, jdk11u
> and jdk17u for possible integrations.
> 
> That sounds reasonable to me as we are able to find more time to look at the
> jvm performance issues on RISCV now.
> 
> &gt;  &gt; In fact, there were some discussions before on the list about the
> backporting work [1].
> &gt;
> &gt;  I missed this one, thanks for pointing it out.
> &gt;
> &gt;  &gt; I'm Xiang Zhang from the Institute of Software, Chinese Academy
> of Sciences. I'm backporting jdk8u for RISC-V support. Here is my
> development repository:
> &gt;&gt; https://github.com/zhangxiang-plct/jdk8u
> &gt;  If there is a riscv-port-jdk8u project, would you be interested to
> contribute there directly?
> &gt;
> &gt;  In the meantime, I'll contact skara-dev to figure out what it would take
> to create the riscv-port-jdk11u and riscv-port-jdk17u repositories.
> 
> Besides the repositories needed, I also wonder the correct workflow for
> future reviewing &amp; merging of backporting PRs on those repos.
> I would like to hear comments/suggestions from the jdk update maintainers.
> 
> Thanks,
> Fei </zhangxiang at iscas.ac.cn></ludovic at rivosinc.com>

From ludovic at rivosinc.com  Wed Jan 18 09:05:58 2023
From: ludovic at rivosinc.com (Ludovic Henry)
Date: Wed, 18 Jan 2023 10:05:58 +0100
Subject: Backports to jdk11u and jdk17u
In-Reply-To: <PAXPR02MB8245971774A3A617415AA6B8ECC69@PAXPR02MB8245.eurprd02.prod.outlook.com>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
 <79ab2542.1f08.185ba593e97.Coremail.yangfei@iscas.ac.cn>
 <2b9d63dc.2ad6.185bac8df55.Coremail.zhangxiang@iscas.ac.cn>
 <CADT88DvmwLiKZDxw4BX-zrN0p5aUYeKqTr6TORXXAz+jjRS8dw@mail.gmail.com>
 <5e09da27.4add.185bed323a6.Coremail.yangfei@iscas.ac.cn>
 <PAXPR02MB8245971774A3A617415AA6B8ECC69@PAXPR02MB8245.eurprd02.prod.outlook.com>
Message-ID: <CADT88DsT6Hu_t=rXVCYTvWzKTYBBFGhNQDVrSKkSzOoVoEf=hw@mail.gmail.com>

Hi,

> As pointed out in the mail you cited below, not breaking
> existing code is the major concern of such a backport in
> the view of the maintainers.

> Shared changes should be kept to a minimum.
> Especially in 11, only enabling shared changes are
> eligible, not changes that aim to improve the performance.

Thank you for the clarification, that makes a lot of sense, especially the
focus on not breaking users of other platforms.

> It would be great to see a complete list of
> changes that need to be backported.

Let's make sure we can bake the backports in a dedicated repository so that
we can have a better sense on what the changes would entail.

After reaching to skara-dev, the only thing needed to create these
repositories is for the riscv-project to contact ops at openjdk dot org.
@Fei Yang, I am assuming you'd need to contact them being the riscv-port
lead.

Thank you,
Ludovic

On Tue, Jan 17, 2023 at 4:06 PM Lindenmaier, Goetz <
goetz.lindenmaier at sap.com> wrote:

> Hi,
>
> As pointed out in the mail you cited below, not breaking
> existing code is the major concern of such a backport in
> the view of the maintainers.
>
> Shared changes should be kept to a minimum.
> Especially in 11, only enabling shared changes are
> eligible, not changes that aim to improve the performance.
>
> It would be great to see a complete list of
> changes that need to be backported. These should
> make up a functioning port. A project repo can serve
> the purpose to collect these changes and prove
> they are a complete set.
> We can then look at this list of changes and give a
> basic ok for these to be integrated in 17u/11u. Then
> you can run the normal backport process for them.
>
> Best regards,
>   Goetz.
>
>
>
>
>
>
>
>
> > -----Original Message-----
> > From: jdk-updates-dev <jdk-updates-dev-retn at openjdk.org> On Behalf Of
> > yangfei at iscas.ac.cn
> > Sent: Tuesday, January 17, 2023 9:23 AM
> > To: Ludovic Henry <ludovic at rivosinc.com>
> > Cc: ?? <zhangxiang at iscas.ac.cn>; riscv-port-dev at openjdk.org;
> > yadonn.wang at huawei.com; jdk-updates-dev at openjdk.java.net
> > Subject: Re: Re: Re: Backports to jdk11u and jdk17u
> >
> > [Some people who received this message don't often get email from
> > yangfei at iscas.ac.cn. Learn why this is important at
> > https://aka.ms/LearnAboutSenderIdentification ]
> >
> > Hi,
> >
> > -----Original Messages-----
> > From:"Ludovic Henry" <ludovic at rivosinc.com>
> > Sent Time:2023-01-16 23:36:33 (Monday)
> > To: "??" <zhangxiang at iscas.ac.cn>
> > Cc: yangfei at iscas.ac.cn, riscv-port-dev at openjdk.org,
> > yadonn.wang at huawei.com, jdk-updates-dev at openjdk.java.net
> > Subject: Re: Re: Backports to jdk11u and jdk17u
> >
> > &gt;&gt; However there is one big difference between general jdk17u and
> > riscv secific one: while jdk17u accepts only bug fixes and minor perf
> > improvements, the riscv-port-jdk17u will have to accept larger patches:
> as
> > risc-v port is far from being perfect so far, performance-wise is worse
> than
> > arm64 port.
> > &gt;  I expect to initiate the backport in a dedicated
> openjdk/riscv-port-
> > jdk17u repository for exactly that. Until the riscv port is considered
> very
> > stable and performant, we likely want to keep it in a separate
> repository.
> > When stable, we can work with the relevant maintainers of jdk8u, jdk11u
> > and jdk17u for possible integrations.
> >
> > That sounds reasonable to me as we are able to find more time to look at
> the
> > jvm performance issues on RISCV now.
> >
> > &gt;  &gt; In fact, there were some discussions before on the list about
> the
> > backporting work [1].
> > &gt;
> > &gt;  I missed this one, thanks for pointing it out.
> > &gt;
> > &gt;  &gt; I'm Xiang Zhang from the Institute of Software, Chinese
> Academy
> > of Sciences. I'm backporting jdk8u for RISC-V support. Here is my
> > development repository:
> > &gt;&gt; https://github.com/zhangxiang-plct/jdk8u
> > &gt;  If there is a riscv-port-jdk8u project, would you be interested to
> > contribute there directly?
> > &gt;
> > &gt;  In the meantime, I'll contact skara-dev to figure out what it
> would take
> > to create the riscv-port-jdk11u and riscv-port-jdk17u repositories.
> >
> > Besides the repositories needed, I also wonder the correct workflow for
> > future reviewing &amp; merging of backporting PRs on those repos.
> > I would like to hear comments/suggestions from the jdk update
> maintainers.
> >
> > Thanks,
> > Fei </zhangxiang at iscas.ac.cn></ludovic at rivosinc.com>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230118/67448c96/attachment-0001.htm>

From yangfei at iscas.ac.cn  Wed Jan 18 12:47:15 2023
From: yangfei at iscas.ac.cn (yangfei at iscas.ac.cn)
Date: Wed, 18 Jan 2023 20:47:15 +0800 (GMT+08:00)
Subject: Backports to jdk11u and jdk17u
In-Reply-To: <CADT88DsT6Hu_t=rXVCYTvWzKTYBBFGhNQDVrSKkSzOoVoEf=hw@mail.gmail.com>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
 <79ab2542.1f08.185ba593e97.Coremail.yangfei@iscas.ac.cn>
 <2b9d63dc.2ad6.185bac8df55.Coremail.zhangxiang@iscas.ac.cn>
 <CADT88DvmwLiKZDxw4BX-zrN0p5aUYeKqTr6TORXXAz+jjRS8dw@mail.gmail.com>
 <5e09da27.4add.185bed323a6.Coremail.yangfei@iscas.ac.cn>
 <PAXPR02MB8245971774A3A617415AA6B8ECC69@PAXPR02MB8245.eurprd02.prod.outlook.com>
 <CADT88DsT6Hu_t=rXVCYTvWzKTYBBFGhNQDVrSKkSzOoVoEf=hw@mail.gmail.com>
Message-ID: <1bb15352.1af1.185c4eb5693.Coremail.yangfei@iscas.ac.cn>

-----Original Messages-----
From:"Ludovic Henry" <ludovic at rivosinc.com>
Sent Time:2023-01-18 17:05:58 (Wednesday)
To: "Lindenmaier, Goetz" <goetz.lindenmaier at sap.com>
Cc: "yangfei at iscas.ac.cn" <yangfei at iscas.ac.cn>, "??" <zhangxiang at iscas.ac.cn>, "riscv-port-dev at openjdk.org" <riscv-port-dev at openjdk.org>, "yadonn.wang at huawei.com" <yadonn.wang at huawei.com>, "jdk-updates-dev at openjdk.java.net" <jdk-updates-dev at openjdk.java.net>
Subject: Re: Re: Re: Backports to jdk11u and jdk17u

Hi,  

&gt;&gt; It would be great to see a complete list of
&gt;&gt; changes that need to be backported.
&gt; Let's make sure we can bake the backports in a dedicated repository so that we can have a better sense on what the changes would entail.  
&gt; 
&gt;  After reaching to skara-dev, the only thing needed to create these repositories is for the riscv-project to contact ops at openjdk dot org. @Fei Yang, I am assuming you'd need to contact them being the riscv-port lead.

 Sure! I just sent an email to ops at openjdk.java.net requesting creation of following three new repos:

  1. https://github.com/openjdk/riscv-port-jdk17u
  2. https://github.com/openjdk/riscv-port-jdk11u
  3. https://github.com/openjdk/riscv-port-jdk8u

Let's see :-)

Thanks,
Fei
</jdk-updates-dev at openjdk.java.net></yadonn.wang at huawei.com></riscv-port-dev at openjdk.org></zhangxiang at iscas.ac.cn></yangfei at iscas.ac.cn></goetz.lindenmaier at sap.com></ludovic at rivosinc.com>

From vladimir.kempik at gmail.com  Wed Jan 18 13:09:19 2023
From: vladimir.kempik at gmail.com (Vladimir Kempik)
Date: Wed, 18 Jan 2023 16:09:19 +0300
Subject: Backports to jdk11u and jdk17u
In-Reply-To: <1bb15352.1af1.185c4eb5693.Coremail.yangfei@iscas.ac.cn>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
 <79ab2542.1f08.185ba593e97.Coremail.yangfei@iscas.ac.cn>
 <2b9d63dc.2ad6.185bac8df55.Coremail.zhangxiang@iscas.ac.cn>
 <CADT88DvmwLiKZDxw4BX-zrN0p5aUYeKqTr6TORXXAz+jjRS8dw@mail.gmail.com>
 <5e09da27.4add.185bed323a6.Coremail.yangfei@iscas.ac.cn>
 <PAXPR02MB8245971774A3A617415AA6B8ECC69@PAXPR02MB8245.eurprd02.prod.outlook.com>
 <CADT88DsT6Hu_t=rXVCYTvWzKTYBBFGhNQDVrSKkSzOoVoEf=hw@mail.gmail.com>
 <1bb15352.1af1.185c4eb5693.Coremail.yangfei@iscas.ac.cn>
Message-ID: <72D74CBB-B0CF-416C-B602-13013E4FD587@gmail.com>

Hello
After the initial jep backport is done, I can contribute a good number of backports I have done so far. I will prepare a list first.

Regards, Vladimir

> 18 ???. 2023 ?., ? 15:47, yangfei at iscas.ac.cn ???????(?):
> 
> -----Original Messages-----
> From:"Ludovic Henry" <ludovic at rivosinc.com>
> Sent Time:2023-01-18 17:05:58 (Wednesday)
> To: "Lindenmaier, Goetz" <goetz.lindenmaier at sap.com>
> Cc: "yangfei at iscas.ac.cn" <yangfei at iscas.ac.cn>, "??" <zhangxiang at iscas.ac.cn>, "riscv-port-dev at openjdk.org" <riscv-port-dev at openjdk.org>, "yadonn.wang at huawei.com" <yadonn.wang at huawei.com>, "jdk-updates-dev at openjdk.java.net" <jdk-updates-dev at openjdk.java.net>
> Subject: Re: Re: Re: Backports to jdk11u and jdk17u
> 
> Hi,  
> 
> &gt;&gt; It would be great to see a complete list of
> &gt;&gt; changes that need to be backported.
> &gt; Let's make sure we can bake the backports in a dedicated repository so that we can have a better sense on what the changes would entail.  
> &gt; 
> &gt;  After reaching to skara-dev, the only thing needed to create these repositories is for the riscv-project to contact ops at openjdk dot org. @Fei Yang, I am assuming you'd need to contact them being the riscv-port lead.
> 
> Sure! I just sent an email to ops at openjdk.java.net requesting creation of following three new repos:
> 
>  1. https://github.com/openjdk/riscv-port-jdk17u
>  2. https://github.com/openjdk/riscv-port-jdk11u
>  3. https://github.com/openjdk/riscv-port-jdk8u
> 
> Let's see :-)
> 
> Thanks,
> Fei
> </jdk-updates-dev at openjdk.java.net></yadonn.wang at huawei.com></riscv-port-dev at openjdk.org></zhangxiang at iscas.ac.cn></yangfei at iscas.ac.cn></goetz.lindenmaier at sap.com></ludovic at rivosinc.com>


From goetz.lindenmaier at sap.com  Wed Jan 18 13:13:52 2023
From: goetz.lindenmaier at sap.com (Lindenmaier, Goetz)
Date: Wed, 18 Jan 2023 13:13:52 +0000
Subject: Backports to jdk11u and jdk17u
In-Reply-To: <72D74CBB-B0CF-416C-B602-13013E4FD587@gmail.com>
References: <CADT88DvbXvzMx5h6kDCnmrp2rvKy8iDDPpWWSiZuCT8_hkTGig@mail.gmail.com>
 <79ab2542.1f08.185ba593e97.Coremail.yangfei@iscas.ac.cn>
 <2b9d63dc.2ad6.185bac8df55.Coremail.zhangxiang@iscas.ac.cn>
 <CADT88DvmwLiKZDxw4BX-zrN0p5aUYeKqTr6TORXXAz+jjRS8dw@mail.gmail.com>
 <5e09da27.4add.185bed323a6.Coremail.yangfei@iscas.ac.cn>
 <PAXPR02MB8245971774A3A617415AA6B8ECC69@PAXPR02MB8245.eurprd02.prod.outlook.com>
 <CADT88DsT6Hu_t=rXVCYTvWzKTYBBFGhNQDVrSKkSzOoVoEf=hw@mail.gmail.com>
 <1bb15352.1af1.185c4eb5693.Coremail.yangfei@iscas.ac.cn>
 <72D74CBB-B0CF-416C-B602-13013E4FD587@gmail.com>
Message-ID: <PAXPR02MB824538D96D60FE01548ADA17ECC79@PAXPR02MB8245.eurprd02.prod.outlook.com>

Hi,

Please keep the list for the initial backport restricted to
the changes that are definitely needed to get it working.

Best regards,
  Goetz.

> -----Original Message-----
> From: Vladimir Kempik <vladimir.kempik at gmail.com>
> Sent: Wednesday, January 18, 2023 2:09 PM
> To: yangfei at iscas.ac.cn
> Cc: Ludovic Henry <ludovic at rivosinc.com>; Lindenmaier, Goetz
> <goetz.lindenmaier at sap.com>; ?? <zhangxiang at iscas.ac.cn>; riscv-port-
> dev at openjdk.org; yadonn.wang at huawei.com; jdk-updates-
> dev at openjdk.java.net
> Subject: Re: Backports to jdk11u and jdk17u
> 
> Hello
> After the initial jep backport is done, I can contribute a good number of
> backports I have done so far. I will prepare a list first.
> 
> Regards, Vladimir
> 
> > 18 ???. 2023 ?., ? 15:47, yangfei at iscas.ac.cn ???????(?):
> >
> > -----Original Messages-----
> > From:"Ludovic Henry" <ludovic at rivosinc.com>
> > Sent Time:2023-01-18 17:05:58 (Wednesday)
> > To: "Lindenmaier, Goetz" <goetz.lindenmaier at sap.com>
> > Cc: "yangfei at iscas.ac.cn" <yangfei at iscas.ac.cn>, "??"
> <zhangxiang at iscas.ac.cn>, "riscv-port-dev at openjdk.org" <riscv-port-
> dev at openjdk.org>, "yadonn.wang at huawei.com"
> <yadonn.wang at huawei.com>, "jdk-updates-dev at openjdk.java.net" <jdk-
> updates-dev at openjdk.java.net>
> > Subject: Re: Re: Re: Backports to jdk11u and jdk17u
> >
> > Hi,
> >
> > &gt;&gt; It would be great to see a complete list of
> > &gt;&gt; changes that need to be backported.
> > &gt; Let's make sure we can bake the backports in a dedicated repository
> so that we can have a better sense on what the changes would entail.
> > &gt;
> > &gt;  After reaching to skara-dev, the only thing needed to create these
> repositories is for the riscv-project to contact ops at openjdk dot org. @Fei
> Yang, I am assuming you'd need to contact them being the riscv-port lead.
> >
> > Sure! I just sent an email to ops at openjdk.java.net requesting creation of
> following three new repos:
> >
> >  1. https://github.com/openjdk/riscv-port-jdk17u
> >  2. https://github.com/openjdk/riscv-port-jdk11u
> >  3. https://github.com/openjdk/riscv-port-jdk8u
> >
> > Let's see :-)
> >
> > Thanks,
> > Fei
> > </jdk-updates-
> dev at openjdk.java.net></yadonn.wang at huawei.com></riscv-port-
> dev at openjdk.org></zhangxiang at iscas.ac.cn></yangfei at iscas.ac.cn></goetz
> .lindenmaier at sap.com></ludovic at rivosinc.com>


From zhangze.linux at gmail.com  Sat Jan 21 04:17:42 2023
From: zhangze.linux at gmail.com (Ze Zhang)
Date: Sat, 21 Jan 2023 12:17:42 +0800
Subject: openjdk 20 will crash when run on the newest linux kernel 6.1.3
Message-ID: <CAEJU13KR7r=b2k9FE_uzERm5oxs8txGnsD=7qJ0b7u9xm2s9JA@mail.gmail.com>

hi,
     I run mill (a java program used to build scala program) on linux
riscv64 kernel 6.1.3 in qemu,
I always crash, can someone give me some idea?  here is the debug log file:
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230121/d6526f62/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: hs_err_pid461.log
Type: text/x-log
Size: 58882 bytes
Desc: not available
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230121/d6526f62/hs_err_pid461-0001.log>

From yangfei at iscas.ac.cn  Sat Jan 21 13:12:07 2023
From: yangfei at iscas.ac.cn (yangfei at iscas.ac.cn)
Date: Sat, 21 Jan 2023 21:12:07 +0800 (GMT+08:00)
Subject: openjdk 20 will crash when run on the newest linux kernel 6.1.3
In-Reply-To: <CAEJU13KR7r=b2k9FE_uzERm5oxs8txGnsD=7qJ0b7u9xm2s9JA@mail.gmail.com>
References: <CAEJU13KR7r=b2k9FE_uzERm5oxs8txGnsD=7qJ0b7u9xm2s9JA@mail.gmail.com>
Message-ID: <71ad7d2b.a499.185d4752d91.Coremail.yangfei@iscas.ac.cn>

Hi,

  I see from the log file that the crash happens when the C2 CompilerThread is doing a native clock_gettime/__vdso_clock_gettime. 
  So I wonder whether this is a problem that is QEMU-specific? Does it trigger on hardware platforms like Unmatched board? 
  Also did you try the pre-built JDK binaries for RISCV64 here at [1]? 

[1] https://builds.shipilev.net/openjdk-jdk/

Regards,
Fei Yang

-----Original Messages-----
From:"Ze Zhang" <zhangze.linux at gmail.com>
Sent Time:2023-01-21 12:17:42 (Saturday)
To: riscv-port-dev at openjdk.org
Cc: 
Subject: openjdk 20 will crash when run on the newest linux kernel 6.1.3

hi,   I run mill (a java program used to build scala program) on linux riscv64 kernel 6.1.3 in qemu,  
  I always crash, can someone give me some idea?  here is the debug log file:


</zhangze.linux at gmail.com>

From zhangze.linux at gmail.com  Sun Jan 22 00:25:54 2023
From: zhangze.linux at gmail.com (Ze Zhang)
Date: Sun, 22 Jan 2023 08:25:54 +0800
Subject: openjdk 20 will crash when run on the newest linux kernel 6.1.3
In-Reply-To: <71ad7d2b.a499.185d4752d91.Coremail.yangfei@iscas.ac.cn>
References: <CAEJU13KR7r=b2k9FE_uzERm5oxs8txGnsD=7qJ0b7u9xm2s9JA@mail.gmail.com>
 <71ad7d2b.a499.185d4752d91.Coremail.yangfei@iscas.ac.cn>
Message-ID: <CAEJU13+-BOMKtxKcF7MEJ7a9uCfUuQQF1gLSGBkwdnCp7kK4aw@mail.gmail.com>

hi,
     I tested on your daily build, but got the same result,  openjdk 20 on
linux kernel 5.19 has no this issue,
may be the newest kernel changed something, please check it, thanks a lot.

root at qemuriscv64:~/java/jdk/bin# java -version

openjdk version "21-testing" 2023-09-19
OpenJDK Runtime Environment (slowdebug build
21-testing-builds.shipilev.net-openjdk-jdk-b310-20230121)
OpenJDK 64-Bit Server VM (slowdebug build
21-testing-builds.shipilev.net-openjdk-jdk-b310-20230121, mixed mode)
root at qemuriscv64:~/java/jdk/bin#
root at qemuriscv64:~/java/jdk/bin#
root at qemuriscv64:~/java/jdk/bin#
root at qemuriscv64:~/java/jdk/bin# mill -version
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007fff8ee25916, pid=1463, tid=1464
#
# JRE version: OpenJDK Runtime Environment (21.0) (slowdebug build
21-testing-builds.shipilev.net-openjdk-jdk-b310-20230121)
# Java VM: OpenJDK 64-Bit Server VM (slowdebug
21-testing-builds.shipilev.net-openjdk-jdk-b310-20230121, mixed mode,
tiered, compressed oops, compressed class ptrs, g1 gc, linux-riscv64)
# Problematic frame:
# C  [linux-vdso.so.1+0x916]  __vdso_clock_gettime+0xe4
#
# No core dump will be written. Core dumps have been disabled. To enable
core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /home/root/java/jdk/bin/hs_err_pid1463.log
#
# If you would like to submit a bug report, please visit:
#   https://bugreport.java.com/bugreport/crash.jsp
#
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007fff9a5f3916, pid=1428, tid=1431
#
# JRE version: OpenJDK Runtime Environment (21.0) (slowdebug build
21-testing-builds.shipilev.net-openjdk-jdk-b310-20230121)
# Java VM: OpenJDK 64-Bit Server VM (slowdebug
21-testing-builds.shipilev.net-openjdk-jdk-b310-20230121, mixed mode,
tiered, compressed oops, compressed class ptrs, g1 gc, linux-riscv64)
# Problematic frame:
# C  [linux-vdso.so.1+0x916]  __vdso_clock_gettime+0xe4
#
# No core dump will be written. Core dumps have been disabled. To enable
core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /home/root/java/jdk/bin/hs_err_pid1428.log
[52.667s][warning][os] Loading hsdis library failed

=================================================================================

root at qemuriscv64:~/java/jdk/bin# java -version
openjdk version "20-internal" 2023-03-21
OpenJDK Runtime Environment (slowdebug build 20-internal-adhoc.zhangze.git)
OpenJDK 64-Bit Server VM (slowdebug build 20-internal-adhoc.zhangze.git,
mixed mode)
root at qemuriscv64:~/java/jdk/bin# mill -version
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007fffa8514916, pid=1352, tid=1355
#
# JRE version: OpenJDK Runtime Environment (20.0) (slowdebug build
20-internal-adhoc.zhangze.git)
# Java VM: OpenJDK 64-Bit Server VM (slowdebug
20-internal-adhoc.zhangze.git, mixed mode, tiered, compressed oops,
compressed class ptrs, g1 gc, linux-riscv64)
# Problematic frame:
# C  [linux-vdso.so.1+0x916]  __vdso_clock_gettime+0xe4
#
# No core dump will be written. Core dumps have been disabled. To enable
core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /home/root/java/jdk/bin/hs_err_pid1352.log
#
# If you would like to submit a bug report, please visit:
#   https://bugreport.java.com/bugreport/crash.jsp
#
Aborted



<yangfei at iscas.ac.cn> ?2023?1?21??? 21:12???

> Hi,
>
>   I see from the log file that the crash happens when the C2
> CompilerThread is doing a native clock_gettime/__vdso_clock_gettime.
>   So I wonder whether this is a problem that is QEMU-specific? Does it
> trigger on hardware platforms like Unmatched board?
>   Also did you try the pre-built JDK binaries for RISCV64 here at [1]?
>
> [1] https://builds.shipilev.net/openjdk-jdk/
>
> Regards,
> Fei Yang
>
> -----Original Messages-----
> From:"Ze Zhang" <zhangze.linux at gmail.com>
> Sent Time:2023-01-21 12:17:42 (Saturday)
> To: riscv-port-dev at openjdk.org
> Cc:
> Subject: openjdk 20 will crash when run on the newest linux kernel 6.1.3
>
> hi,   I run mill (a java program used to build scala program) on linux
> riscv64 kernel 6.1.3 in qemu,
>   I always crash, can someone give me some idea?  here is the debug log
> file:
>
>
> </zhangze.linux at gmail.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230122/8be69cb9/attachment.htm>

From zhangze.linux at gmail.com  Tue Jan 31 07:14:13 2023
From: zhangze.linux at gmail.com (Ze Zhang)
Date: Tue, 31 Jan 2023 15:14:13 +0800
Subject: openjdk crash on linux kernel 6.1.5
Message-ID: <CAEJU13K0EzJ2exUBVGTgXbc1Z325qAjFcKPRudfso_mcR=UXFA@mail.gmail.com>

hi,
on kernel 6.1.5,  java -version crashes:

Type "apropos word" to search for commands related to "word"...
Reading symbols from java...
Reading symbols from
/usr/lib/jvm/java-20+gitAUTOINC+5dfc4ec7d9-openjdk-riscv64/bin/.debug/java...
(gdb) run -version
Starting program: /usr/bin/java -version
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/libthread_db.so.1".




[New Thread 0x7ffff6086180 (LWP 543)]
[New Thread 0x7ffff4c56180 (LWP 544)]
[New Thread 0x7ffff4a54180 (LWP 545)]
[New Thread 0x7ffff4852180 (LWP 546)]
[New Thread 0x7fffbdbfb180 (LWP 547)]
[New Thread 0x7fffbd9f9180 (LWP 548)]
[New Thread 0x7fffbd776180 (LWP 549)]

Thread 2 "java" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7ffff6086180 (LWP 543)]
0x00007ffff7fdf916 in __vdso_clock_gettime ()
(gdb)
(gdb)
(gdb)
(gdb)
(gdb) bt
#0  0x00007ffff7fdf916 in __vdso_clock_gettime ()
#1  0x00007ffff7f1e712 in __GI___clock_gettime (clock_id=<optimized out>,
tp=<optimized out>) at
/usr/src/debug/glibc/2.36-r0/sysdeps/unix/sysv/linux/clock_gettime.c:42
#2  0xaa140004aa140003 in ?? ()
(gdb)
Broadcast message from systemd-journald at qemuriscv64 (Tue 2022-12-20
18:46:12 UTC):

systemd[1]: Caught <SEGV> from PID 4.


Message from syslogd at qemuriscv64 at Tue Dec 20 18:46:12 2022 ...
systemd[1]: Caught <SEGV> from PID 4.

Broadcast message from systemd-journald at qemuriscv64 (Tue 2022-12-20
18:46:12 UTC):

systemd[1]: Caught <SEGV>, dumped core as pid 554.


Broadcast message from systemd-journald at qemuriscv64 (Tue 2022-12-20
18:46:12 UTC):

systemd[1]: Freezing execution.


Message from syslogd at qemuriscv64 at Tue Dec 20 18:46:12 2022 ...
systemd[1]: Caught <SEGV>, dumped core as pid 554.

Message from syslogd at qemuriscv64 at Tue Dec 20 18:46:12 2022 ...
systemd[1]: Freezing execution.

#0  0x00007ffff7fdf916 in __vdso_clock_gettime ()
#1  0x00007ffff7f1e712 in __GI___clock_gettime (clock_id=<optimized out>,
tp=<optimized out>) at
/usr/src/debug/glibc/2.36-r0/sysdeps/unix/sysv/linux/clock_gettime.c:42
#2  0xaa140004aa140003 in ?? ()
(gdb)

#2  0xaa140004aa140003 in ?? ()
(gdb)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230131/974a17d4/attachment-0001.htm>

From yangfei at iscas.ac.cn  Tue Jan 31 07:56:30 2023
From: yangfei at iscas.ac.cn (yangfei at iscas.ac.cn)
Date: Tue, 31 Jan 2023 15:56:30 +0800 (GMT+08:00)
Subject: openjdk crash on linux kernel 6.1.5
In-Reply-To: <CAEJU13K0EzJ2exUBVGTgXbc1Z325qAjFcKPRudfso_mcR=UXFA@mail.gmail.com>
References: <CAEJU13K0EzJ2exUBVGTgXbc1Z325qAjFcKPRudfso_mcR=UXFA@mail.gmail.com>
Message-ID: <62b0ed17.12e01.18606d3cfd4.Coremail.yangfei@iscas.ac.cn>

Hi,

  The call trace looks the same as the one in your previous e-mail. 
  From the calltrace, I don't think this is an issue of the OpenJDK, but more likely an issue of the system. 
  Can you create some simple C/C++ program to exercise __vdso_clock_gettime() system call in your qemu system? 

Thanks,
Fei

-----Original Messages-----
From:"Ze Zhang" <zhangze.linux at gmail.com>
Sent Time:2023-01-31 15:14:13 (Tuesday)
To: riscv-port-dev at openjdk.org, sw-dev at groups.riscv.org
Cc: 
Subject: openjdk crash on linux kernel 6.1.5

hi,   on kernel 6.1.5,  java -version crashes:  
  
  Type "apropos word" to search for commands related to "word"...
Reading symbols from java...
Reading symbols from /usr/lib/jvm/java-20+gitAUTOINC+5dfc4ec7d9-openjdk-riscv64/bin/.debug/java...
(gdb) run -version
Starting program: /usr/bin/java -version
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/libthread_db.so.1".




[New Thread 0x7ffff6086180 (LWP 543)]
[New Thread 0x7ffff4c56180 (LWP 544)]
[New Thread 0x7ffff4a54180 (LWP 545)]
[New Thread 0x7ffff4852180 (LWP 546)]
[New Thread 0x7fffbdbfb180 (LWP 547)]
[New Thread 0x7fffbd9f9180 (LWP 548)]
[New Thread 0x7fffbd776180 (LWP 549)]

Thread 2 "java" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7ffff6086180 (LWP 543)]
0x00007ffff7fdf916 in __vdso_clock_gettime ()
(gdb) 
(gdb) 
(gdb) 
(gdb) 
(gdb) bt
#0  0x00007ffff7fdf916 in __vdso_clock_gettime ()
#1  0x00007ffff7f1e712 in __GI___clock_gettime (clock_id=<optimized out="">, tp=<optimized out="">) at /usr/src/debug/glibc/2.36-r0/sysdeps/unix/sysv/linux/clock_gettime.c:42
#2  0xaa140004aa140003 in ?? ()
(gdb) 
Broadcast message from systemd-journald at qemuriscv64 (Tue 2022-12-20 18:46:12 UTC):

systemd[1]: Caught <segv> from PID 4.


Message from syslogd at qemuriscv64 at Tue Dec 20 18:46:12 2022 ...
systemd[1]: Caught <segv> from PID 4.

Broadcast message from systemd-journald at qemuriscv64 (Tue 2022-12-20 18:46:12 UTC):

systemd[1]: Caught <segv>, dumped core as pid 554.


Broadcast message from systemd-journald at qemuriscv64 (Tue 2022-12-20 18:46:12 UTC):

systemd[1]: Freezing execution.


Message from syslogd at qemuriscv64 at Tue Dec 20 18:46:12 2022 ...
systemd[1]: Caught <segv>, dumped core as pid 554.

Message from syslogd at qemuriscv64 at Tue Dec 20 18:46:12 2022 ...
systemd[1]: Freezing execution.

#0  0x00007ffff7fdf916 in __vdso_clock_gettime ()
#1  0x00007ffff7f1e712 in __GI___clock_gettime (clock_id=<optimized out="">, tp=<optimized out="">) at /usr/src/debug/glibc/2.36-r0/sysdeps/unix/sysv/linux/clock_gettime.c:42
#2  0xaa140004aa140003 in ?? ()
(gdb) 

#2  0xaa140004aa140003 in ?? ()
(gdb)


</optimized></optimized></segv></segv></segv></segv></optimized></optimized></zhangze.linux at gmail.com>

From tommy_murphy at hotmail.com  Tue Jan 31 07:55:34 2023
From: tommy_murphy at hotmail.com (Tommy Murphy)
Date: Tue, 31 Jan 2023 07:55:34 +0000
Subject: [sw-dev] openjdk crash on linux kernel 6.1.5
In-Reply-To: <CAEJU13K0EzJ2exUBVGTgXbc1Z325qAjFcKPRudfso_mcR=UXFA@mail.gmail.com>
References: <CAEJU13K0EzJ2exUBVGTgXbc1Z325qAjFcKPRudfso_mcR=UXFA@mail.gmail.com>
Message-ID: <LO0P265MB686695260AB9933114A2FF36F9D09@LO0P265MB6866.GBRP265.PROD.OUTLOOK.COM>

What Linux, Java (I guess it's OpenJDK?), and target hardware/simulator are you using?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230131/70af6bfd/attachment.htm>

From zhangze.linux at gmail.com  Tue Jan 31 07:58:19 2023
From: zhangze.linux at gmail.com (Ze Zhang)
Date: Tue, 31 Jan 2023 15:58:19 +0800
Subject: [sw-dev] openjdk crash on linux kernel 6.1.5
In-Reply-To: <LO0P265MB686695260AB9933114A2FF36F9D09@LO0P265MB6866.GBRP265.PROD.OUTLOOK.COM>
References: <CAEJU13K0EzJ2exUBVGTgXbc1Z325qAjFcKPRudfso_mcR=UXFA@mail.gmail.com>
 <LO0P265MB686695260AB9933114A2FF36F9D09@LO0P265MB6866.GBRP265.PROD.OUTLOOK.COM>
Message-ID: <CAEJU13KfdpiQ4Fh5M29DtR1vzqc-c5TwDFNwy_hotU7Y0jceMQ@mail.gmail.com>

openjdk  20 or 21, build riscv yocto image from
https://github.com/riscv/meta-riscv/
run on qemu 7.2, linux kernel version 6.1.5

Tommy Murphy <tommy_murphy at hotmail.com> ?2023?1?31??? 15:55???

> What Linux, Java (I guess it's OpenJDK?), and target hardware/simulator
> are you using?
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230131/781eb33e/attachment.htm>

From tommy_murphy at hotmail.com  Tue Jan 31 14:01:28 2023
From: tommy_murphy at hotmail.com (Tommy Murphy)
Date: Tue, 31 Jan 2023 14:01:28 +0000
Subject: [sw-dev] openjdk crash on linux kernel 6.1.5
In-Reply-To: <CAEJU13KfdpiQ4Fh5M29DtR1vzqc-c5TwDFNwy_hotU7Y0jceMQ@mail.gmail.com>
References: <CAEJU13K0EzJ2exUBVGTgXbc1Z325qAjFcKPRudfso_mcR=UXFA@mail.gmail.com>
 <LO0P265MB686695260AB9933114A2FF36F9D09@LO0P265MB6866.GBRP265.PROD.OUTLOOK.COM>
 <CAEJU13KfdpiQ4Fh5M29DtR1vzqc-c5TwDFNwy_hotU7Y0jceMQ@mail.gmail.com>
Message-ID: <LO0P265MB68669F0625B0020B3FCDFCF4F9D09@LO0P265MB6866.GBRP265.PROD.OUTLOOK.COM>

Could it be related to this issue?

https://lore.kernel.org/all/3815b200-a9e2-a3db-c7f1-1e3f86dbab3f at cloudbear.ru/T/#rccffdbbebe5130653c80cdd5189ed25fdc609936
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/riscv-port-dev/attachments/20230131/57746e23/attachment-0001.htm>