1. Forum
  2. Usenet
  3. LINUX.DEBIAN.DEVEL

  • how to convey package porting details?

    From Paul Wise@21:1/5 to All on Mon Jun 6 04:10:01 2022
    Hi all,

    There are lots of packages that need porting to every new architecture
    that comes along. There are others that don't require porting but
    benefit in some way from porting to some aspect of the architecture.

    In order to help porters to prioritise those packages and quickly add
    support for their architecture, it would be nice to have a standard
    mechanism for source packages to indicate that they potentially require
    porting for each new architecture, the nature of the porting required
    and where in the source package the changes are required.

    For example packages like autotools-dev, linux, gcc or linuxinfo could
    state that they require porting to all new arches and have a pointer to
    Debian and or upstream porting info. Packages containing a JIT but with interpreter fallback could list themselves as having optional porting.

    Once we have a mechanism for this, the documentation for creating new
    Debian ports could provide a way to find the list of packages that need
    porter attention; via codesearch.d.n, apt, apt-file, grep-available etc.

    https://wiki.debian.org/PortsDocs/New

    --
    bye,
    pabs

    https://wiki.debian.org/PaulWise

    -----BEGIN PGP SIGNATURE-----

    iQIzBAABCgAdFiEEYQsotVz8/kXqG1Y7MRa6Xp/6aaMFAmKdYDoACgkQMRa6Xp/6 aaOIeg/9Hs+eznfPx3Fy08H7PWqQW7/77Isz6SSdVWGkMbO5Q/r2dCNcIf8HGPgn 1/5ntmrNpQutMpwkGY2rNiYCJkHrxnDd+SAEF7UiA3Xz2+1IQIhFeHTmHJbgwTm0 VwCmc/PIxSXCoQ+ZpyP9bOGOiPHH4cjuy0lCTmVDELQUF3diuLYAcXMrMEzpj2le JASHKF/FS1RcHedNhi8SnWJe68Bi+RmvKE/pwzAfkyaY0pcVQS5a/azMRpIESx8j Juil2UfYJeh9qC1k8PCmrj1PCOIQoye2ZQrwRP1L2a73w8bHRkWBy8LV7ZHOsh3y gFQuxnnEIk7J5dU/7HWCpdVyzXmu44GEgjvO1S8ATBnb/f758Dly51CobUJ5oi4F yKqYxLq+Q/M3lanoRovt/T9qqyiN7yWc17WFI+h2Wn4LUs9XCKySsF/w23uOAFcj V9KV4uRwK4gDjimZnWJLgXYLyMqpRrGVfy7KTlRAgAGhL6x502YWFELJJpr/AxVK WSZenHtFJQZ+KETtd2QfczIQC/i2x5sXdarKcJ7Dzmly8S3pdaJ9KhYSadBXRGII CC9ubWN3JV3BlLYOHFlLvNsx1UthzEC3YJCH0ZyU4tfLYuEYtP9R1hk9PzlblH9j nu24alOh5sFBQ7sdKFsbKMSKCtGNKScIqYcmXm9v74KZAiHVfIA=
    =NVRF
    -----END PGP SIGNATURE-----

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From M. Zhou@21:1/5 to Paul Wise on Mon Jun 6 04:50:01 2022
    I like this idea. I would personally recommend adding negative priority
    as well. You know, it is completely meaningless to port some high performance scientific computing software to archs like armel...

    Meanwhile, different packages varies in the difficulty to port as well.
    A software that heavily leverages architecture specific intrisics is not
    likely easy to port. Packages that does not require hardware performance,
    and simply misses several arch-specific C macros should be easy to port.

    Since package maintainers should have somehow skimmed the whole codebase,
    they could provide an accurate hint about this, as long as such hints
    are useful for porters.

    On Mon, 2022-06-06 at 10:02 +0800, Paul Wise wrote:

    There are lots of packages that need porting to every new architecture
    that comes along. There are others that don't require porting but
    benefit in some way from porting to some aspect of the architecture.

    In order to help porters to prioritise those packages and quickly add
    support for their architecture, it would be nice to have a standard
    mechanism for source packages to indicate that they potentially require porting for each new architecture, the nature of the porting required
    and where in the source package the changes are required.

    For example packages like autotools-dev, linux, gcc or linuxinfo could
    state that they require porting to all new arches and have a pointer to Debian and or upstream porting info. Packages containing a JIT but with interpreter fallback could list themselves as having optional porting.

    Once we have a mechanism for this, the documentation for creating new
    Debian ports could provide a way to find the list of packages that need porter attention; via codesearch.d.n, apt, apt-file, grep-available etc.

    https://wiki.debian.org/PortsDocs/New


    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Marco d'Itri@21:1/5 to Paul Wise on Mon Jun 6 10:50:01 2022
    On Jun 06, Paul Wise <[email protected]> wrote:

    There are lots of packages that need porting to every new architecture
    that comes along. There are others that don't require porting but
    benefit in some way from porting to some aspect of the architecture.
    Is this really worth the effort, considering that probably RISC-V is
    going to be our last port for a very long time?

    --
    ciao,
    Marco

    -----BEGIN PGP SIGNATURE-----

    iHUEABYIAB0WIQQnKUXNg20437dCfobLPsM64d7XgQUCYp2/KgAKCRDLPsM64d7X gWkGAP4ifNptO+OJtsJdkbITn9s1+t5fyVaUm1URJl+j4ecbXAEA0QuMHAnDFck/ x1+3lu9DkWEya4f5ms0r0I0kjR8jegg=
    =ulrb
    -----END PGP SIGNATURE-----

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Adam Borowski@21:1/5 to Marco d'Itri on Mon Jun 6 12:40:01 2022
    On Mon, Jun 06, 2022 at 10:47:38AM +0200, Marco d'Itri wrote:
    On Jun 06, Paul Wise <[email protected]> wrote:

    There are lots of packages that need porting to every new architecture
    that comes along. There are others that don't require porting but
    benefit in some way from porting to some aspect of the architecture.
    Is this really worth the effort, considering that probably RISC-V is
    going to be our last port for a very long time?

    loongarch has just been merged into mainline kernel, is in binutils 2.38
    and gcc 12, and those guys seems to be doing some serious work to get
    into distributions.

    Besides CPU archs, it looks worthwhile to investigate replacing glibc
    with musl, which would represented as a new arch.

    We may also want to raise ISA baseline in eg. amd64 -- with the amount
    of old hardware in the wild, it'd certainly be done like armel->armhf
    rather than i386->(really i686).


    Meow!
    --
    ⢀⣴⠾⠻⢶⣦⠀
    ⣾⠁⢠⠒⠀⣿⡁ Loongarch's name is loong.
    ⢿⡄⠘⠷⠚⠋⠀
    ⠈⠳⣄⠀⠀⠀⠀

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From =?UTF-8?B?eGlhbyBzaGVuZyB3ZW4o6IKW5@21:1/5 to All on Mon Jun 6 12:30:01 2022
    This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --------------9w0l3pt69K034oNtmZLXkxi0
    Content-Type: text/plain; charset=UTF-8
    Content-Transfer-Encoding: base64

    5ZyoIDIwMjIvNi82IDE2OjQ3LCBNYXJjbyBkJ0l0cmkg5YaZ6YGTOg0KDQo+IElzIHRoaXMg cmVhbGx5IHdvcnRoIHRoZSBlZmZvcnQsIGNvbnNpZGVyaW5nIHRoYXQgcHJvYmFibHkgUklT Qy1WIGlzIA0KPiBnb2luZyB0byBiZSBvdXIgbGFzdCBwb3J0IGZvciBhIHZlcnkgbG9uZyB0 aW1lPw0KDQpQZXJoYXBzIGxvb25nYXJjaDY0IHdpbGwgdGhlIG5leHQgcG9ydC4NCg0KaHR0 cHM6Ly93aWtpLmRlYmlhbi5vcmcvUG9ydHMvbG9vbmdhcmNoNjQNCg0KDQpTb21lIGNvZGVz IG9mIGxvb25nYXJjaDY0IGhhZCBiZWVuIGFkZGVkIHRvIHVwc3RyZWFtIGtlcm5lbDoNCg0K aHR0cHM6Ly9naXQua2VybmVsLm9yZy9wdWIvc2NtL2xpbnV4L2tlcm5lbC9naXQvdG9ydmFs ZHMvbGludXguZ2l0L2NvbW1pdC8/aD12NS4xOS1yYzEmaWQ9OGJlNDQ5MzExOWIwYWVkZjdk ZDYxZTFjYTUyMGZiMzk4NTM3YjUzZQ0KDQoNCi0tIA0K6IKW55ub5paHIHhpYW8gc2hlbmcg d2VuIEZhcmlzIFhpYW8gDQrlvq7kv6Eod2VjaGF0Ke+8mmF0emxpbnV4DQrjgIrpk5zosYzo sYYgTGludXjjgItodHRwczovL3d3dy5hdHpsaW51eC5jb20NCuWfuuS6jiBEZWJpYW4g55qE IExpbnV4IOS4reaWhyDmoYzpnaIg5pON5L2c57O757ufDQpEZWJpYW4gUUEgcGFnZTogaHR0 cHM6Ly9xYS5kZWJpYW4ub3JnL2RldmVsb3Blci5waHA/bG9naW49YXR6bGludXglNDBzaW5h LmNvbQ0KR251UEcgUHVibGljIEtleTogMHgwMDE4NjYwMjMzOTI0MENCDQoNCg==

    --------------9w0l3pt69K034oNtmZLXkxi0--

    -----BEGIN PGP SIGNATURE-----

    wsF5BAABCAAjFiEEvGv7H5NUQYeSuhtTJ2Egg8PSprAFAmKd08wFAwAAAAAACgkQJ2Egg8PSprDb Tw/8D2NaHBsl0AmWE2CUmtFJhpegcw0lCLyqshUszNf647AmeDlHKHrFNMIbxSj4v2PisldfIuQ9 3u+jjdC8MWkoRl5ASsjjuqtX9wspw3bD17qdZpsqGylnqjqvxe2WPslq/TuiaugTF76cXRaLKX6I RzQf86Q//Pxl3NwVMUPdX6NSkCgclJ4cIWRo0wDoORdRDNAucRx43wnBo4AQbt/a0VhWtSJbPZXV VJb5ixzJEC/+fiRjjUpvIVEncQutC8YFrlPLBvo9eb/I2kC2aSuLbarTZMLGAiszO6tHjOHFC+8M klk3730vX4c0BzFzml6vCgvFz5wbcFEkKz6I89dQgLx7KWVi8038Xl+pAcp5fmSVxj7ILRR/FjkL O69QWy7uQ1vVrnuRa6x0tx56KBDT4/mVQn9oVFocWQX3L1Z4chLNtOyX/xW64ymaEogLW06vsY40 sC2fMKPzY7qRiA7pkYdwj7TWdIqqegLeCFpnIMBiosz2cdFLSCaAVpKGNx1XzFtzTUAYzc703fRD 9CM1W4wsqrAJodMUXbhnj2qW3Bp2KtrwdAtYoK7uv7CQ0qoZFqBjVwa1091ahgfTzImKR/TWNkbe KcHa9MjxSCfYblm78jqMdxnA7NraTv7Kg4Oqv6PQBzi+zYEJCqvtDixracXNtV9XmKu8y7jW2VVU JiM=
    =wX7C
    -----END PGP SIGNATURE-----

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Holger Levsen@21:1/5 to Marco d'Itri on Mon Jun 6 13:10:01 2022
    On Mon, Jun 06, 2022 at 10:47:38AM +0200, Marco d'Itri wrote:
    Is this really worth the effort, considering that probably RISC-V is
    going to be our last port for a very long time?

    you mean like 640kb should be enough for everyone? :)


    --
    cheers,
    Holger

    ⢀⣴⠾⠻⢶⣦⠀
    ⣾⠁⢠⠒⠀⣿⡁ holger@(debian|reproducible-builds|layer-acht).org
    ⢿⡄⠘⠷⠚⠋⠀ OpenPGP: B8BF54137B09D35CF026FE9D 091AB856069AAA1C
    ⠈⠳⣄

    Words may inspire but only action creates change.

    -----BEGIN PGP SIGNATURE-----

    iQIzBAABCgAdFiEEuL9UE3sJ01zwJv6dCRq4VgaaqhwFAmKd35wACgkQCRq4Vgaa qhwg7w//dSldDyNCVoxH+qDQsFb8VqM8cJnWKaD3yG8zr0RW11A6srK1AuEKxp/Y toK1C07/ICVKbk5Bkt8pBsPPy72PHGk3iTZ5NS5aIqaZjiA2q4zSb4Ma2rPQK5IP 4Y04StK/HKQBRxFvbSgX9NSXVNAE0u0Q67DVvM5EolkSl7lSmFzAU1lrKoKlyGnQ Jyh3I2S6R38PAECzd4cib3oDZyTelsIcfT533gr91atEzKFTZqQY/AgLLt0DbvEM v2SBb7m/GXqn+Yx8wFZyI6jRQrJEbJV5jAR3ARGKRW8lmFQ+rV2EK8KnXNMNCVPl uSaHXJvT2RsQ/tfH0QnTo9ar22C+qpVpyJ1AnCVcQed5d4/wQbdFfHJgMUgElnfE VbY+fQcMe7xTrppqY+Fiw2phLyx/z+xBldU88gYP4lKydrSeP1LFUKsjeZ1vd2rC YXgfopRtMGvhYB66k1oyuyTgFUD6cUl/NFB9N2w9jkcH0cNZt1d9PbE7I2a0ZRiM SKjA6DJUvknbt3TaqJY7GCUsHMPBSZokQvuZbTqqOgR1R34B9ZvXPqS91cqTPJPA m8rz8KkPORGUJwJMveXroiOeWciWd4r6jqK9BBkauv/6pfEtFgu2NWrrUhLC4t0
  • From Paul Wise@21:1/5 to Marco d'Itri on Mon Jun 6 13:20:01 2022
    On Mon, 2022-06-06 at 10:47 +0200, Marco d'Itri wrote:

    Is this really worth the effort, considering that probably RISC-V is
    going to be our last port for a very long time?

    There are at least two or three new architectures in the pipeline
    already. loongarch64 from Loongson was already mentioned, but there is
    also already arc and probably arc64 from Synopsis. I'm not sure if
    CHERI and ARM Morello will require a new arch, but ISTR they will.
    If The Mill ever gets off the ground I expect there will be a Debian
    port of that. There are also other CPU arches supported by Linux like
    C-Sky as well as ideas for ports to various kernels and other things
    like CloudABI or WASM. There are also perhaps rebootstrap-only ports
    or full ports for moving baselines backwards or forwards like to
    amd64v3 or MIPSr1, which Debian no longer supports but could easily
    support and would enable Debian on hardware that currently ships.

    https://wiki.debian.org/Ports/loongarch64
    https://wiki.debian.org/Ports/arc
    https://wiki.debian.org/Ideas/Ports

    --
    bye,
    pabs

    https://wiki.debian.org/PaulWise

    -----BEGIN PGP SIGNATURE-----

    iQIzBAABCgAdFiEEYQsotVz8/kXqG1Y7MRa6Xp/6aaMFAmKd4OAACgkQMRa6Xp/6 aaO5kA//SkkeQ/B926tkhGjTysTLQRpmve2JMFHQ+v4DzAOdhIJHyFhooQxmLhYb iRx4tSuPXrt9j6CJDrEK4DxPFo9+apEIyvuA/34HDI2oHvdgr5rOGFQSCDftWHvB gxtzK12lXmYO3USI7CCkdNYeN2wkIYcv92j/W6rMPmzYZJvw9coy/LeQqEyF9znF peMIP5+GFJNFbl265C7I6h5dIMv8FFjWZeg3Ty+fsEOAESHknk/uTZegEkSTOj3g aZlUR0y4q7pNowq3iKHTa68TcQKMZWP8GtNRhpg75uADdbF+k6x5lauA/ZFgvo4w ExHb8wjVsh1ZXrRxP9FaybdD8wSb9HDjwbNRK5zD9ghg9LS3GQnCxDPr3fvgMtJy Xu3GEU6zsYV1g/spXN6A+wiG40c3OGSqz8q7FDiVj9uP5SYq6YQg3ZVUCtPF0jWq gxCz/CR5QODvAT8ct4ht9tA1mcRQdlufcAxbLwi87i7vX39JDLBKzhGmjpleMZNu yVoVSSsYuhDbV7b8GY/gI7IDT0rZNXgXFpCAl8G0LHDcVErFRhFaC/UlRQupApop TRrGBdrSV1VRJ91a6FSMsU3ZjHg5V/FYpCzd403MONaS2sYWGAb2HXbkyJZm1Ep+ YjkONrZSTR2Nh99Wup/XynzJ6qIw9fHX1h9Udkleya7YswfV0x4=
    =PD5S
    -----END PGP SIGNATURE-----

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Marco d'Itri@21:1/5 to Holger Levsen on Mon Jun 6 17:00:01 2022
    On Jun 06, Holger Levsen <[email protected]> wrote:

    Is this really worth the effort, considering that probably RISC-V is
    going to be our last port for a very long time?
    you mean like 640kb should be enough for everyone? :)
    More like "free ABI is eating the world".

    I do not see much future for the architectures mentioned here except
    maybe loongarch64 as a geopolitical tool, but sure: no objections if the people who actually work on ports think this is worthwhile.
    Maybe the usual suspects which require significant porting work could
    start documenting instructions for porters in debian/README.source.

    --
    ciao,
    Marco

    -----BEGIN PGP SIGNATURE-----

    iHUEABYIAB0WIQQnKUXNg20437dCfobLPsM64d7XgQUCYp4UvwAKCRDLPsM64d7X gRVzAQC0ENxvhYZQAO+M80V8UH56XWrIPjqnrVt0/CUIGurRbAEA8d71Zo/8ZCf5 /ALTm+DabXChb2htJTxynP9b/e2lXAg=
    =jXY8
    -----END PGP SIGNATURE-----

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Paul Wise@21:1/5 to Marco d'Itri on Tue Jun 7 02:50:01 2022
    On Mon, 2022-06-06 at 16:52 +0200, Marco d'Itri wrote:

    Maybe the usual suspects which require significant porting work could
    start documenting instructions for porters in debian/README.source.

    Thats a good start, but doesn't provide a standard way to find out
    which packages contain such instructions. So maybe README.porting?

    --
    bye,
    pabs

    https://wiki.debian.org/PaulWise

    -----BEGIN PGP SIGNATURE-----

    iQIzBAABCgAdFiEEYQsotVz8/kXqG1Y7MRa6Xp/6aaMFAmKenjQACgkQMRa6Xp/6 aaOczBAAj6iuLJCLG6Cbsp+9O26CGlXbqpejf81EjbXSa+Wy8Or4yyMSZP+iPf4S myMiGnOPNE6EMYLoo2BraoRGyt5Zs9gjGbFWd78MjPr2jGFmCBfHzE9oS3VV9S20 LHZwM8xScNmewk9nFzgYWVXu4+zBiqT/7rbxQOEE0QeXjw0or01MjPzXz6vfj7Ue vxW0gUfe4fWa+ARNC2YOaK7YnkiSz+TOa2jWH3MT7/jLqIjLKuSKMBaXi+zfCB0n D2+iW57cHmTT4NzmuoQjeY5UOTOWyyTzXDxwP2jHhYDb0YKxLp1d9t0gaDxhR2wI 85QwovfabsxelnZwXHxJmEGLBSDoj8X20voFG0fNh0SbMjb2yRHcQh39BBfj8f1f pw5No/UJNEEXJx2eEl5py37mn2E9OkFyZ7ajv6RdM+cEyzKA7RArTCLLDXqf+X8W rIXcIt67tiCmCLTIaCPp5P7GKdt8KtBnmDJC5vO7B4QW0kjC4X33XqEP9KUMHPBx 9m49Ln8mS29p6ZE7KXoV4fpKWHDH0yRD3mNkRGgxiC+RsvIlMSILm3mwFdrbFCq+ JtQhhPGGtRQE4TurM08SXMTFeACm5vrcdeQiekOIeMfWVhTP9O7iXnnAHNFZn9iF tJGgagaWo1sqLghIOjXJ50U/rEaSaDOvGXeIgne5kQa/FwQ0EAY=
    =h0zw
    -----END PGP SIGNATURE-----

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)