Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 17 Jul 2019 04:02:01 +0000
From:      Mark Linimon <linimon@lonesome.com>
To:        Dennis Clarke <dclarke@blastwave.org>
Cc:        FreeBSD PowerPC ML <freebsd-ppc@freebsd.org>
Subject:   Re: FreeBSD current on ppc64 and RISC-V rv64imafdc
Message-ID:  <20190717040200.GC1700@lonesome.com>
In-Reply-To: <cc52126e-7263-7ab2-5f46-969271c44734@blastwave.org>
References:  <20190713152404.GJ1503@dendrobates> <b6876ce7-5147-5f1a-d712-3515e124cd36@gmail.com> <6b3dcec02b425a8559605b4e6c5ada19ee642728.camel@freebsd.org> <20190713194256.GK1503@dendrobates> <85e35455-793a-84fe-c8cf-39b89ade7c11@gmail.com> <26864b34-624a-075f-4a10-859314649d2f@blastwave.org> <20190714172816.GA26897@lonesome.com> <3e91570f-69b4-1d29-ada9-bd13d42e236a@blastwave.org> <20190716003844.GB19462@lonesome.com> <cc52126e-7263-7ab2-5f46-969271c44734@blastwave.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Jul 15, 2019 at 10:38:42PM -0400, Dennis Clarke wrote:
> >>Essentially one may checkout the sources and do a clean build only to
> >>see the kernel panic on ppc64 hardware.

OK.  So ... IIUC, you are reporting:

  "-CURRENT has regressions on older hardware."

If so, this, to me, seems much less alarming than the original message
that "FreeBSD/powerpc64 is unusable".

All the new development work is being done on new hardware.  From time
to time, the developers do regression testing on older machines.  This
is to be expected -- it's the most efficient way for them to proceed;
progress would be much slower if the requirement was to test everywhere.

I'd be much more concerned to hear that "-STABLE has regressions on
older hardware" or even "releases have regressions on older hardware".
Sometimes these things happen but we do try to prevent it -- the
'regression' tag in Bugzilla is designed for this very purpose.

But regressions on -CURRENT ... that's to be expected on -CURRENT.
It's primarily for developers, not users.  Anyone using -CURRENT should
consider the tradeoffs, and be willing to help debug and analyze.  FreeBSD
has always operated this way.

mcl



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20190717040200.GC1700>