Re: Any hppa porters interested in getting MariaDB working?
Soumyajyotii Ssarkar <[email protected]> Wed, 14 Jan 2026 23:21:48 +0530
| Newsgroups | gmane.linux.debian.ports.hppa |
|---|---|
| Message-ID | <CAGkZZ+umpfCbhP7uAvMxEvAJD8Q=DzQku8NfY4VzA1DecvzDLQ@mail.gmail.com> |
--000000000000d1708106485cc392 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hey Otto, I am actually currently working on it. I was last able to produce this error which I am currently trying to fix with mariadb codebase, but I think its progress perhaps? *# ERROR LOG 1 (initially I faced this):* Server version: 11.8.5-MariaDB-4 from Debian source revision: 2bb18a8cf1170b7c506b9c57640d8798a3b74298 The information page at https://mariadb.com/kb/en/how-to-produce-a-full-stack-trace-for-mariadbd/ contains instructions to obtain a better version of the backtrace below. Following these instructions will help MariaDB developers provide a fix quicker. Attempting backtrace. Include this in the bug report. (note: Retrieving this information may fail) Thread pointer: 0x0 stack_bottom =3D 0x0 thread_stack 0x49000 /home/soumyajyotii/build/mariadb-server/builddir/sql/mariadbd(my_print_stac= ktrace-0x36cf6292) [0x421a4dac] Writing a core file... Working directory at /home/soumyajyotii/build/mariadb-server/builddir/mysql-test/var/install.db Resource Limits (excludes unlimited resources): Limit Soft Limit Hard Limit Units Max stack size 8388608 unlimited bytes Max processes 257230 257230 processes Max open files 1024 1024 files Max locked memory 8388608 8388608 bytes Max pending signals 257230 257230 signals Max msgqueue size 819200 819200 bytes Max nice priority 0 0 Max realtime priority 0 0 Core pattern: core Kernel version: Linux version 6.12.57+deb13-amd64 ( [email protected]) (x86_64-linux-gnu-gcc-14 (Debian 14.2.0-19) 14.2.0, GNU ld (GNU Binutils for Debian) 2.44) #1 SMP PREEMPT_DYNAMIC Debian 6.12.57-1 (2025-11-05) make[1]: *** [debian/rules:162: override_dh_auto_test] Error 1 make[1]: Leaving directory '/home/soumyajyotii/build/mariadb-server' make: *** [debian/rules:253: binary] Error 2 dpkg-buildpackage: error: debian/rules binary subprocess failed with exit status 2 Currently the build is failing at the CMake *configure* stage because MariaDB still expects a required core dependency to be present, even after all related optional features have been explicitly disabled. So I am trying to figure it out and fix it. Your advice and guidance would be appreciated on how I could work more closely with the MariaDB Community for the porting. Thank you, Soumyajyotii Ssarkar. On Wed, Jan 14, 2026 at 11:01=E2=80=AFPM Otto Kek=C3=A4l=C3=A4inen <otto@de= bian.org> wrote: > Hi! > > In December many expressed interest in porting MariaDB to hppa, but I > didn't notice any comments on any of the bug report or MR or any other > follow-up, so I guess perhaps people didn't have time for this after > all..? > > > > >>>> I am happy to take your patches, do additional testing and do > uploads. > > >>>> Unfortunately I can't do the actual port fixing myself as there is > too > > >>>> much work and the domain is something I am not strong in. My attem= pt > > >>>> at fixes at > https://salsa.debian.org/mariadb-team/mariadb-server/-/merge_requests/146 > > >>>> did get some progress, so if you want to take on this challenge, y= ou > > >>>> don't need to start from zero. > > >>>> > > >>>> I also have good experiences of getting upstream to accept porting > > >>>> related patches (e.g. https://github.com/MariaDB/server/pull/4480) > so > > >>>> if you fix this in Debian and submit patches upstream, you most > likely > > >>>> will get those fixes into MariaDB permanently for all users > globally. > > >>> > > >>> The first step should be disabling Java support on hppa. Since > OpenJDK currently > > >>> does not support hppa due to the stack growing from upwards, any > packages that > > >>> make use of Java won't work on hppa. > > >>> > > >>> There is some work-in-progress port for OpenJDK on hppa that I > worked on some > > >>> years ago but I never finished the actual work. It might be a good > idea to > > >>> complete that work in the future though. > > >> > > >> > > >> Sure, sounds like a great place to start? I will try to get it setup > and disable the java functionality first, perhaps to try and get it runni= ng. > > > > > > Please start by reading the > > > > https://salsa.debian.org/mariadb-team/mariadb-server/-/merge_requests/146 > > > I mentioned in my original email, and the notes and comments in it. I= t > > > will help you avoid doing things I already did, and there are also > > > commands linked you can just copy-paste to get the build and test > > > suite going on a porter box (or anywhere). > > > > > > Thanks for taking a stab at making this package build and pass tests > in hppa! > > > > --000000000000d1708106485cc392 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr">Hey Otto,<div>I am actually currently working on it.</div>= <div>I was last able to produce this error which I am currently trying to f= ix with mariadb codebase, but I think its progress perhaps?</div><div><br><= /div><div><b># ERROR LOG 1 (initially I faced this):</b><br>Server version:= 11.8.5-MariaDB-4 from Debian source revision: 2bb18a8cf1170b7c506b9c57640d= 8798a3b74298<br>The information page at <a href=3D"https://mariadb.com/kb/e= n/how-to-produce-a-full-stack-trace-for-mariadbd/">https://mariadb.com/kb/e= n/how-to-produce-a-full-stack-trace-for-mariadbd/</a><br>contains instructi= ons to obtain a better version of the backtrace below.<br>Following these i= nstructions will help MariaDB developers provide a fix quicker.<br>Attempti= ng backtrace. Include this in the bug report.<br>(note: Retrieving this inf= ormation may fail)<br>Thread pointer: 0x0<br>stack_bottom =3D 0x0 thread_st= ack 0x49000<br>/home/soumyajyotii/build/mariadb-server/builddir/sql/mariadb= d(my_print_stacktrace-0x36cf6292) [0x421a4dac]<br>Writing a core file...<br= >Working directory at /home/soumyajyotii/build/mariadb-server/builddir/mysq= l-test/var/install.db<br>Resource Limits (excludes unlimited resources):<br= >Limit =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 Soft Limit =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Hard Limit =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 Units<br>Max stack size =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A08388608 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0unlimi= ted =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0bytes<br>Max processes =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 257230 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 257230 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 p= rocesses<br>Max open files =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A01024 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 1024 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 files<br>Max locked memory =C2=A0= =C2=A0 =C2=A0 =C2=A0 8388608 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A08388608 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0bytes<br>Max pen= ding signals =C2=A0 =C2=A0 =C2=A0 257230 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0= =C2=A0 =C2=A0 257230 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 sign= als<br>Max msgqueue size =C2=A0 =C2=A0 =C2=A0 =C2=A0 819200 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 819200 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 bytes<br>Max nice priority =C2=A0 =C2=A0 =C2=A0 =C2=A0 0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A00<br>M= ax realtime priority =C2=A0 =C2=A0 0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 =C2=A0 =C2=A00<br>Core pattern: core<br>Kernel version: L= inux version 6.12.57+deb13-amd64 (<a href=3D"mailto:[email protected]= ian.org">[email protected]</a>) (x86_64-linux-gnu-gcc-14 (Debi= an 14.2.0-19) 14.2.0, GNU ld (GNU Binutils for Debian) 2.44) #1 SMP PREEMPT= _DYNAMIC Debian 6.12.57-1 (2025-11-05)<br>make[1]: *** [debian/rules:162: o= verride_dh_auto_test] Error 1<br>make[1]: Leaving directory '/home/soum= yajyotii/build/mariadb-server'<br>make: *** [debian/rules:253: binary] = Error 2<br>dpkg-buildpackage: error: debian/rules binary subprocess failed = with exit status 2<br></div><div><br></div><div><br></div><div><br></div><d= iv>Currently the build is failing at the CMake <em style=3D"">configure</em= > stage because MariaDB still expects a required core dependency to be pres= ent, even after all related optional features have been explicitly disabled= .</div><div><br></div><div>So I am trying to figure it out and fix it.</div= ><div><br></div><div>Your advice and guidance would be appreciated on how I= could work more closely with the MariaDB Community for the porting.</div><= div><br></div><div>Thank you,</div><div>Soumyajyotii Ssarkar.</div></div><b= r><div class=3D"gmail_quote gmail_quote_container"><div dir=3D"ltr" class= =3D"gmail_attr">On Wed, Jan 14, 2026 at 11:01=E2=80=AFPM Otto Kek=C3=A4l=C3= =A4inen <<a href=3D"mailto:[email protected]">[email protected]</a>> wrot= e:<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0= .8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Hi!<br> <br> In December many expressed interest in porting MariaDB to hppa, but I<br> didn't notice any comments on any of the bug report or MR or any other<= br> follow-up, so I guess perhaps people didn't have time for this after<br= > all..?<br> <br> <br> > >>>> I am happy to take your patches, do additional testin= g and do uploads.<br> > >>>> Unfortunately I can't do the actual port fixing m= yself as there is too<br> > >>>> much work and the domain is something I am not strong= in. My attempt<br> > >>>> at fixes at <a href=3D"https://salsa.debian.org/maria= db-team/mariadb-server/-/merge_requests/146" rel=3D"noreferrer" target=3D"_= blank">https://salsa.debian.org/mariadb-team/mariadb-server/-/merge_request= s/146</a><br> > >>>> did get some progress, so if you want to take on this= challenge, you<br> > >>>> don't need to start from zero.<br> > >>>><br> > >>>> I also have good experiences of getting upstream to a= ccept porting<br> > >>>> related patches (e.g. <a href=3D"https://github.com/M= ariaDB/server/pull/4480" rel=3D"noreferrer" target=3D"_blank">https://githu= b.com/MariaDB/server/pull/4480</a>) so<br> > >>>> if you fix this in Debian and submit patches upstream= , you most likely<br> > >>>> will get those fixes into MariaDB permanently for all= users globally.<br> > >>><br> > >>> The first step should be disabling Java support on hppa. = Since OpenJDK currently<br> > >>> does not support hppa due to the stack growing from upwar= ds, any packages that<br> > >>> make use of Java won't work on hppa.<br> > >>><br> > >>> There is some work-in-progress port for OpenJDK on hppa t= hat I worked on some<br> > >>> years ago but I never finished the actual work. It might = be a good idea to<br> > >>> complete that work in the future though.<br> > >><br> > >><br> > >> Sure, sounds like a great place to start? I will try to get i= t setup and disable the java functionality first, perhaps to try and get it= running.<br> > ><br> > > Please start by reading the<br> > > <a href=3D"https://salsa.debian.org/mariadb-team/mariadb-server/-= /merge_requests/146" rel=3D"noreferrer" target=3D"_blank">https://salsa.deb= ian.org/mariadb-team/mariadb-server/-/merge_requests/146</a><br> > > I mentioned in my original email, and the notes and comments in i= t. It<br> > > will help you avoid doing things I already did, and there are als= o<br> > > commands linked you can just copy-paste to get the build and test= <br> > > suite going on a porter box (or anywhere).<br> > ><br> > > Thanks for taking a stab at making this package build and pass te= sts in hppa!<br> > ><br> </blockquote></div> --000000000000d1708106485cc392--