<div class="__aliyun_email_body_block"><div  style="line-height:1.7;font-family:Tahoma,Arial,STHeiti,SimSun;font-size:14.0px;color:#000000;"><div  style="clear:both;">Hi Felix,</div><div  style="clear:both;"><div  style="clear:both;"><br ></div><div  style="clear:both;">Thank you for mentioning this. Concisely, I have written something about it in a new thread, and we can discuss it there at: <a  href="https://mail.openjdk.org/pipermail/riscv-port-dev/2022-September/000615.html" target="_blank">https://mail.openjdk.org/pipermail/riscv-port-dev/2022-September/000615.html</a>.</div><div  style="clear:both;">There seems confusion about the format, generated from my mailbox. Please forgive him.<br ></div><div  style="clear:both;"><br ></div><div  style="clear:both;">Best,</div><span >Xiaolin</span></div><div  style="clear:both;"><br ></div><blockquote  style="margin-right:.0px;margin-top:.0px;margin-bottom:.0px;font-family:Tahoma,Arial,STHeiti,SimSun;font-size:14.0px;color:#000000;"><div  style="clear:both;">------------------------------------------------------------------</div><div  style="clear:both;">From:yangfei <yangfei@iscas.ac.cn></div><div  style="clear:both;">Send Time:2022年9月14日(星期三) 15:33</div><div  style="clear:both;">To:郑孝林(云矅) <yunyao.zxl@alibaba-inc.com></div><div  style="clear:both;">Cc:riscv-port-dev <riscv-port-dev-retn@openjdk.org>; Aleksey Shipilev <shade@redhat.com>; riscv-port-dev@openjdk.org <riscv-port-dev@openjdk.org>; Vladimir Kempik <vladimir.kempik@gmail.com></div><div  style="clear:both;">Subject:Re: Re: RVC by default?</div><div  style="clear:both;"><br ></div><p  style="margin-top:14.0px;margin-bottom:14.0px;">
        Hi Xiaolin,
</p><p  style="margin-top:14.0px;margin-bottom:14.0px;">
        <br ></p><p  style="margin-top:14.0px;margin-bottom:14.0px;">
            I am interested in your new proposal for supporting the RVC extension. 
</p><p  style="margin-top:14.0px;margin-bottom:14.0px;">
            Can you provide a simple description of how it works and maybe the new interfaces? 
</p><p  style="margin-top:14.0px;margin-bottom:14.0px;">
            I guess developers will need to be aware of this when working on this port. 
</p><p  style="margin-top:14.0px;margin-bottom:14.0px;">
        <br ></p><p  style="margin-top:14.0px;margin-bottom:14.0px;">
        Thanks,
</p><p  style="margin-top:14.0px;margin-bottom:14.0px;">
        Fei
</p><p  style="margin-top:14.0px;margin-bottom:14.0px;">
        <br ></p><div  style="margin:14.0px 40.0px;">
        -----Original Messages-----<br ><b >From:</b><span  id="rc_from">"Xiaolin Zheng" <yunyao.zxl@alibaba-inc.com></span><br ><b >Sent Time:</b><span  id="rc_senttime">2022-09-08 20:09:59 (Thursday)</span><br ><b >To:</b> riscv-port-dev <riscv-port-dev-retn@openjdk.org>, "Aleksey Shipilev" <shade@redhat.com>, "riscv-port-dev@openjdk.org" <riscv-port-dev@openjdk.org>, "Vladimir Kempik" <vladimir.kempik@gmail.com><br ><b >Cc:</b> <br ><b >Subject:</b> Re: RVC by default?<br ><br >
        
                <div  class=" __aliyun_node_has_color" style="line-height:1.7;font-family:Tahoma,Arial,STHeiti,SimSun;font-size:14.0px;color:#000000;">
                        <div  style="clear:both;">
                                <span >Hi Aleksey and Vladimir,</span> 
                        </div>
                        <div  style="clear:both;">
                                <br >
                        </div>
                        <div  style="clear:both;">
                                The current RVC support is okay but not complete: it only covers ~10% of total instructions emitted (mostly C2 code, including some part of Stub code), and we might want to transform instructions into the compressed counterparts as much as possible, so maybe the design will change from a whitelist mode (the class CompressibleRegion) to a black list mode. There is one implementation at my local branch <a  href="https://github.com/zhengxiaolinX/jdk/commits/REBASE-rvc-beautify" target="_blank">https://github.com/zhengxiaolinX/jdk/commits/REBASE-rvc-beautify</a> (might not be stable yet, I have not gotten enough time to give it a sufficient test on jtregs and specjbb2015/other benchmarks yet). There are plans reserved to commit them (which cover ~20% of instructions under some tests) after reviewing, but this is currently WIP and waiting loom port to merge first.
                        </div>
                </div>
        
</div></blockquote></div></div>