Date: Sat, 13 Jul 2024 18:49:27 +0200 From: Daniel Engberg <daniel.engberg.lists@pyret.net> To: Mathieu Arnold <mat@freebsd.org> Cc: ports-committers@freebsd.org, dev-commits-ports-all@freebsd.org, dev-commits-ports-main@freebsd.org, Baptiste Daroussin <bapt@freebsd.org>, "arrowd@FreeBSD.org" <arrowd@freebsd.org> Subject: Re: git: 06601897e5cd - main - framework: reintroduce the feature enabling code Message-ID: <d66b209388c08a537cef775a5d5d7891@mail.infomaniak.com> In-Reply-To: <4cvwsu7vebcrzorvjygkhzefbnjq2cegtobxbybxpglltsla3k@637dblgngbon> References: <202404120754.43C7slbr026326@gitrepo.freebsd.org> <12af8e210220224883a7856115f61be9@mail.infomaniak.com> <6kivt3yarts23vqyv277vqrw6dhswo4hilbdkspvnaz544mtvc@yiyqgdtnp6y7> <457bab3436f42fd12d80beb16e067cf0@mail.infomaniak.com> <4cvwsu7vebcrzorvjygkhzefbnjq2cegtobxbybxpglltsla3k@637dblgngbon>
next in thread | previous in thread | raw e-mail | index | archive | help
--_=_swift_1720889367_b53d715a89f570dc5586e0b6fcddbe74_=_ Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable On 2024-07-13T16:30:31.000+02:00, Mathieu Arnold <mat@freebsd.org> wrote:= >=C2=A0On=C2=A0Sat,=C2=A0Jul=C2=A013,=C2=A02024=C2=A0at=C2=A002:04:48= PM=C2=A0GMT,=C2=A0Daniel=C2=A0Engberg=C2=A0wrote: >>=C2=A0=C2=A0On=C2= =A02024-07-13T08:47:47.000+02:00,=C2=A0Mathieu=C2=A0Arnold >>=C2=A0=C2= =A0<mat@freebsd.org> >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0wrote: >>= =C2=A0=C2=A0 >>>=C2=A0=C2=A0=C2=A0On=C2=A0Sat,=C2=A0Jul=C2=A013,=C2=A0202= 4=C2=A0at=C2=A006:11:04AM=C2=A0GMT,=C2=A0Daniel=C2=A0Engberg=C2=A0wrote: = >>>=C2=A0=C2=A0=C2=A0 >>>>=C2=A0=C2=A0=C2=A0=C2=A0Hi, >>>>=C2=A0=C2= =A0=C2=A0=C2=A0 >>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0This=C2=A0changes=C2= =A0so=C2=A0LTO=C2=A0option=C2=A0is=C2=A0no=C2=A0longer=C2=A0applied=C2= =A0to=C2=A0Rust >>>>=C2=A0=C2=A0=C2=A0=C2=A0(cargo) >>>>=C2=A0=C2=A0= =C2=A0=C2=A0 >>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0ports >>>>=C2=A0=C2= =A0=C2=A0=C2=A0 >>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0BY=C2=A0DEFAULT=C2= =A0causing=C2=A0a=C2=A0regresssion,=C2=A0please=C2=A0fix. >>>=C2=A0=C2= =A0=C2=A0 >>>=C2=A0=C2=A0=C2=A0=C2=A0As=C2=A0it=C2=A0has=C2=A0been=C2= =A0three=C2=A0months,=C2=A0nobody=C2=A0complained=C2=A0something=C2=A0was= >>>=C2=A0=C2=A0=C2=A0broken >>>=C2=A0=C2=A0=C2=A0 >>>=C2=A0=C2=A0= =C2=A0=C2=A0so, >>>=C2=A0=C2=A0=C2=A0 >>>=C2=A0=C2=A0=C2=A0=C2=A0I= =C2=A0don't=C2=A0think=C2=A0anything=C2=A0is=C2=A0actually=C2=A0broken. >= >>=C2=A0=C2=A0=C2=A0 >>>=C2=A0=C2=A0=C2=A0=C2=A0LTO=C2=A0as=C2=A0are= =C2=A0a=C2=A0few=C2=A0other=C2=A0features=C2=A0like=C2=A0SSP=C2=A0are=C2= =A0user=C2=A0facing >>>=C2=A0=C2=A0=C2=A0features, >>>=C2=A0=C2=A0= =C2=A0 >>>=C2=A0=C2=A0=C2=A0=C2=A0not=C2=A0a >>>=C2=A0=C2=A0=C2=A0 >>= >=C2=A0=C2=A0=C2=A0=C2=A0porters=C2=A0facing=C2=A0options,=C2=A0it=C2=A0mea= ns,=C2=A0it's=C2=A0up=C2=A0to=C2=A0the=C2=A0person=C2=A0doing >>>=C2= =A0=C2=A0=C2=A0the >>>=C2=A0=C2=A0=C2=A0 >>>=C2=A0=C2=A0=C2=A0=C2=A0bui= lding=C2=A0to=C2=A0choose=C2=A0wether=C2=A0to=C2=A0enable=C2=A0it=C2=A0or= =C2=A0not,=C2=A0it=C2=A0is=C2=A0**not**=C2=A0up >>>=C2=A0=C2=A0=C2=A0to= >>>=C2=A0=C2=A0=C2=A0 >>>=C2=A0=C2=A0=C2=A0=C2=A0the >>>=C2=A0=C2= =A0=C2=A0 >>>=C2=A0=C2=A0=C2=A0=C2=A0person=C2=A0porting=C2=A0the=C2= =A0software=C2=A0to=C2=A0forcefully=C2=A0enable=C2=A0it. >>>=C2=A0=C2= =A0=C2=A0 >>>=C2=A0=C2=A0=C2=A0=C2=A0--=C2=A0 >>>=C2=A0=C2=A0=C2=A0 >= >>=C2=A0=C2=A0=C2=A0=C2=A0Mathieu=C2=A0Arnold >>=C2=A0=C2=A0 >>=C2= =A0=C2=A0=C2=A0Hi, >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0Likely=C2=A0becau= se=C2=A0this=C2=A0pretty=C2=A0much=C2=A0silently=C2=A0went=C2=A0by=C2=A0bec= ause=C2=A0it=C2=A0was >>=C2=A0=C2=A0posted >>=C2=A0=C2=A0 >>=C2=A0= =C2=A0=C2=A0on=C2=A0Phab=C2=A0and=C2=A0you=C2=A0only=C2=A0CCed=C2=A0bapt.= =C2=A0It's=C2=A0been=C2=A0enabled=C2=A0since=C2=A0Jan=C2=A02023 >>=C2= =A0=C2=A0 >>=C2=A0=C2=A0=C2=A0(https://cgit.freebsd.org/ports/commit/Mk/U= ses/cargo.mk?id=3D967022fd812cf67dec264ee4e53bd016b69e7a2b) >>=C2=A0= =C2=A0 >>=C2=A0=C2=A0=C2=A0and=C2=A0tested/discussed=C2=A0here=C2=A0https= ://reviews.freebsd.org/D36736 >>=C2=A0=C2=A0before >>=C2=A0=C2=A0 >>= =C2=A0=C2=A0=C2=A0being=C2=A0enabled/committed.=C2=A0I=C2=A0noticed=C2= =A0it=C2=A0now=C2=A0while=C2=A0updating=C2=A0a=C2=A0Rust >>=C2=A0=C2= =A0 >>=C2=A0=C2=A0=C2=A0(cargo-based)=C2=A0port. >=C2=A0 >=C2=A0Mmmm,= =C2=A0yes,=C2=A0I=C2=A0know=C2=A0about=C2=A0that,=C2=A0and=C2=A0I=C2=A0agre= e,=C2=A0this=C2=A0commits=C2=A0reverts=C2=A0this >=C2=A0 >=C2=A0behavio= r. >=C2=A0 >=C2=A0Because=C2=A0choosing=C2=A0to=C2=A0build=C2=A0with= =C2=A0or=C2=A0without=C2=A0LTO=C2=A0is=C2=A0a=C2=A0user=C2=A0facing >= =C2=A0feature, >=C2=A0 >=C2=A0not=C2=A0a=C2=A0porter=C2=A0facing=C2= =A0feature,=C2=A0so,=C2=A0it=C2=A0has=C2=A0to=C2=A0be=C2=A0set=C2=A0by= =C2=A0people=C2=A0building >=C2=A0the >=C2=A0 >=C2=A0things,=C2=A0not= =C2=A0by=C2=A0the=C2=A0framework=C2=A0or=C2=A0a=C2=A0port. >=C2=A0 >= =C2=A0--=C2=A0 >=C2=A0 >=C2=A0Mathieu=C2=A0Arnold Hi, Given t= hat it was commonly agreed upon as a sane default I don't see how your op= inion in this case weights more. That being said, any option could be def= ined as user facing option some of which are already agreed upon as sane = such as DOCS and EXAMPLES. Best regards, Daniel --_=_swift_1720889367_b53d715a89f570dc5586e0b6fcddbe74_=_ Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable <html><body><div>On 2024-07-13T16:30:31.000+02:00, Mathieu Arnold <mat@f= reebsd.org> wrote:<br></div><div class=3D"ik_mail_quote answerContentMes= sage"><blockquote class=3D"ws-ng-quote"><pre style=3D"white-space: normal;"= ><div>On Sat, Jul 13, 2024 at 02:04:48PM GMT, Daniel Engberg wrote:<br></di= v><blockquote class=3D"ws-ng-quote"><div> On 2024-07-13T08:47:47.000+02:00= , Mathieu Arnold <<a href=3D"mailto:mat@freebsd.org" class=3D"defaultMai= lLink">mat@freebsd.org</a>><br></div><div> wrote:<br></div><div> <br></d= iv><blockquote class=3D"ws-ng-quote"><div> On Sat, Jul 13, 2024 at 06:11:= 04AM GMT, Daniel Engberg wrote:<br></div><blockquote class=3D"ws-ng-quote">= <div> Hi,<br></div><div> <br></div><div> This changes so LTO option= is no longer applied to Rust (cargo)<br></div><div> ports<br></div><div>= <br></div><div> BY DEFAULT causing a regresssion, please fix.<br></di= v></blockquote><div> <br></div><div> As it has been three months, nobody= complained something was broken<br></div><div> so,<br></div><div> <br></= div><div> I don't think anything is actually broken.<br></div><div> <br><= /div><div> LTO as are a few other features like SSP are user facing featur= es,<br></div><div> not a<br></div><div> <br></div><div> porters facing o= ptions, it means, it's up to the person doing the<br></div><div> <br></div= ><div> building to choose wether to enable it or not, it is **not** up to<= br></div><div> the<br></div><div> <br></div><div> person porting the sof= tware to forcefully enable it.<br></div><div> <br></div><div> -- <br></di= v><div> <br></div><div> Mathieu Arnold<br></div></blockquote><div> <br><= /div><div> Hi,<br></div><div> <br></div><div> Likely because this pretty mu= ch silently went by because it was posted<br></div><div> on Phab and you on= ly CCed bapt. It's been enabled since Jan 2023<br></div><div> (<a href=3D"h= ttps://cgit.freebsd.org/ports/commit/Mk/Uses/cargo.mk?id=3D967022fd812cf67d= ec264ee4e53bd016b69e7a2b" class=3D"defaultMailLink" target=3D"_blank" rel= =3D"noopener noreferrer" data-ik=3D"ik-secure">https://cgit.freebsd.org/por= ts/commit/Mk/Uses/cargo.mk?id=3D967022fd812cf67dec264ee4e53bd016b69e7a2b</a= >)<br></div><div> and tested/discussed here <a href=3D"https://reviews.free= bsd.org/D36736" class=3D"defaultMailLink" target=3D"_blank" rel=3D"noopener= noreferrer" data-ik=3D"ik-secure">https://reviews.freebsd.org/D36736</a>; b= efore<br></div><div> being enabled/committed. I noticed it now while updati= ng a Rust<br></div><div> (cargo-based) port.<br></div></blockquote><div> <b= r></div><div>Mmmm, yes, I know about that, and I agree, this commits revert= s this<br></div><div>behavior.<br></div><div><br></div><div>Because choosin= g to build with or without LTO is a user facing feature,<br></div><div>not = a porter facing feature, so, it has to be set by people building the<br></d= iv><div>things, not by the framework or a port.<br></div><div><br></div><di= v>-- <br></div><div>Mathieu Arnold<br></div></pre></blockquote></div><div>H= i,<br></div><div><br></div><div>Given that it was commonly agreed upon as a= sane default I don't see how your opinion in this case weights more. That = being said, any option could be defined as user facing option some of which= are already agreed upon as sane such as DOCS and EXAMPLES.<br></div><div><= br></div><div>Best regards,<br></div><div>Daniel<br></div></body></html> --_=_swift_1720889367_b53d715a89f570dc5586e0b6fcddbe74_=_--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?d66b209388c08a537cef775a5d5d7891>