From nobody Mon Jun 19 16:21:56 2023 X-Original-To: freebsd-arm@mlmmj.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mlmmj.nyi.freebsd.org (Postfix) with ESMTP id 4QlFQp3kh9z4f3mx for ; Mon, 19 Jun 2023 16:22:10 +0000 (UTC) (envelope-from wlosh@bsdimp.com) Received: from mail-ed1-x52a.google.com (mail-ed1-x52a.google.com [IPv6:2a00:1450:4864:20::52a]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256 client-signature RSA-PSS (2048 bits) client-digest SHA256) (Client CN "smtp.gmail.com", Issuer "GTS CA 1D4" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 4QlFQp22BJz3hSw for ; Mon, 19 Jun 2023 16:22:10 +0000 (UTC) (envelope-from wlosh@bsdimp.com) Authentication-Results: mx1.freebsd.org; none Received: by mail-ed1-x52a.google.com with SMTP id 4fb4d7f45d1cf-51a4d215e09so1726336a12.3 for ; Mon, 19 Jun 2023 09:22:10 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=bsdimp-com.20221208.gappssmtp.com; s=20221208; t=1687191728; x=1689783728; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:from:to:cc:subject:date:message-id:reply-to; bh=WaI6pYG5flwv8yR+T8uXVQSHtv4blfwG4zTvtW7ALQs=; b=qPRvCKenkDDbKYS2OwGXP1gBem9NuM+LaVwHwV6JZqJDhvkalutknRcaBUg4ZcEpyI VWNfno7JGTv4bB5SllRa28quRZuTlcuWDPOgE8eKcKoYYINWcUhgUSO9tCR1AzZa1Xhs QvUYMkeK/l0Kb04Wf2azauvyF14NMZrMkQ/EHXW9vqp7rR9Ot2hrLi6jnQ02Fsq15DPb krG9yktjmEqEdk7zTzuHkHPcRdfW35WaKB32QIlySHFEIbyierC50Trnf9Wrt/RdZIKI P3cxA5hXWUAvEWtgFt4UJSRQtCwuNrJA7p5RFHxSzf0cOttbZF2uKVVp4XoiiII7f6Qh VGOw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1687191728; x=1689783728; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=WaI6pYG5flwv8yR+T8uXVQSHtv4blfwG4zTvtW7ALQs=; b=dHHYon4ovIRmFugipwnuUhlF6RZFYp1d1uhzcMnhd/4GktK6FiYYXAMWbGumcFVNee XydFTikzAK5nMbk6eEfEYBIQ343PhEyY1NkhbrNVvwWhJbZcsOsxBSw7+rftdPXBrUak sqT3M+uDLmMUtnOzGd4X7b4v/wV0NbCBnh39MNhFHJCkMFMbCYl9S/AaxK5kvjN3641s o/cxcPnhr7i/Keiim3Al9M8m003x/SJUqzVyIsusleCa7QehsfVs+N4ixJqdz/cbtVlk DZWmTezfPHLjpRRy0Y7xRfDNELh5WV+SJObHi4taiWsZKplllrh9p2JNnIiqeqedgI6G n2/Q== X-Gm-Message-State: AC+VfDypE9+dwkIvEhRdxKxYbERQrr4JFXrytQx8NZknxZ+E0J2GZaLD oOXf6vLQbwZf12QmffV142h1Zak1f7vUFC699FR8bNx3wLWt/azY X-Google-Smtp-Source: ACHHUZ7xl0Ahg9PBWOYfunQMlLx8yhCrlycfgrbpkJ46TIGI0eJTpUiTeJcTtg2lqmdGn6VxskcxyxMkzcS/+qTr6E8= X-Received: by 2002:a05:6402:14d5:b0:51a:5c19:16f4 with SMTP id f21-20020a05640214d500b0051a5c1916f4mr1518111edx.4.1687191727859; Mon, 19 Jun 2023 09:22:07 -0700 (PDT) List-Id: Porting FreeBSD to ARM processors List-Archive: https://lists.freebsd.org/archives/freebsd-arm List-Help: List-Post: List-Subscribe: List-Unsubscribe: Sender: owner-freebsd-arm@freebsd.org MIME-Version: 1.0 References: <202306191455.35JEtuV0085343@office.dignus.com> In-Reply-To: <202306191455.35JEtuV0085343@office.dignus.com> From: Warner Losh Date: Mon, 19 Jun 2023 10:21:56 -0600 Message-ID: Subject: Re: -mfloat-abi=soft no longer viable? To: Thomas David Rivers Cc: freebsd-arm@freebsd.org Content-Type: multipart/alternative; boundary="0000000000009b81b105fe7def47" X-Rspamd-Queue-Id: 4QlFQp22BJz3hSw X-Spamd-Bar: ---- X-Spamd-Result: default: False [-4.00 / 15.00]; REPLY(-4.00)[]; ASN(0.00)[asn:15169, ipnet:2a00:1450::/32, country:US] X-Rspamd-Pre-Result: action=no action; module=replies; Message is reply to one we originated X-ThisMailContainsUnwantedMimeParts: N --0000000000009b81b105fe7def47 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Mon, Jun 19, 2023 at 8:56=E2=80=AFAM Thomas David Rivers wrote: > > Just curious - I can't seem to find an answer to this question > on-line. > > Seems in 2019 ARM (32-bit) support moved to -mfloat-abi=3Dhard > armv6 and armv7, yes. The legacy armv[45] didn't move, but they've been removed from the tree. > (unless you built your own system?) But, in the postings > I've stumbled over there was mention of /libsoft and /usr/libsoft. > > But - in my recent FreeBSD-13.2 install I don't see those. > > Has this gone away? Is there no way to build a program > with -mfloat-abi=3Dsoft and link with the math library (-lm)? > We switch the default then. For a while we had a soft compat directory as a transition, but have stopped doing that. make buildworld TARGET_ARCH=3Darmv7 CPUTYPE=3Dsoft will give you the old -mfloat=3Dsoft. However, be advised that the -mfloat=3Dsoft means 'use hard floating point code, but pass argument args in integer registers rather than floating point ones'. This is different than what most people think when they think of what -mfloat=3Dsof= t might mean w/o reading the manual. The libsoft stuff you've stumbled upon was the compat libraries that we used to transition from mfloat=3Dsoft to mfloat=3Dhard. Hope that helps. Warner --0000000000009b81b105fe7def47 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable


=
On Mon, Jun 19, 2023 at 8:56=E2=80=AF= AM Thomas David Rivers <rivers@dign= us.com> wrote:

Just curious - I can't seem to find an answer to this question
on-line.

Seems in 2019 ARM (32-bit) support moved to -mfloat-abi=3Dhard

armv6 and armv7, yes. The legacy armv[45] didn'= t move, but they've been
removed from the tree.
=C2= =A0=C2=A0
(unless you built your own system?)=C2=A0 =C2=A0But, in the postings
I've stumbled over there was mention of /libsoft and /usr/libsoft.

But - in my recent FreeBSD-13.2 install I don't see those.

Has this gone away?=C2=A0 Is there no way to build a program
with -mfloat-abi=3Dsoft and link with the math library (-lm)?

We switch the default then. For a while we had a sof= t compat directory as
a transition, but have stopped doing that.<= /div>

make buildworld TARGET_ARCH=3Darmv7 CPUTYPE=3Dsoft=

will give you the old -mfloat=3Dsoft. However, be= advised that the
-mfloat=3Dsoft means 'use hard floating poi= nt code, but pass argument
args in integer registers rather than = floating point ones'. This is
different than what most people= think when they think of what -mfloat=3Dsoft
might mean w/o read= ing the manual.=C2=A0

The libsoft stuff you've= stumbled upon was the compat libraries
that we used to transitio= n from mfloat=3Dsoft to mfloat=3Dhard.

Hope that h= elps.

Warner
--0000000000009b81b105fe7def47--