Subject: Re: cvsup for i386 on amd64 using COMPAT_NETBSD32
To: Blair Sadewitz <blair.sadewitz@gmail.com>
From: Quentin Garnier <cube@cubidou.net>
List: current-users
Date: 02/16/2007 17:23:35
--DA7REkwdj/3NVgqE
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Fri, Feb 16, 2007 at 05:14:43PM +0100, Quentin Garnier wrote:
> On Fri, Feb 16, 2007 at 06:21:56AM -0500, Blair Sadewitz wrote:
> > When I try to run it, this happens:
> >  6477      1 cvsup    netbsd32_write(0x1, 0x817f00c, 0x22) =3D 34
> >       "Connecting to cvsup.us.netbsd.org\n"
> >  6477      1 cvsup    netbsd32_fcntl(0x1, 0x4, 0x2) =3D 0
> >  6477      1 cvsup    compat_30_netbsd32_socket(0x2, 0x1, 0) =3D 3
> >  6477      1 cvsup    netbsd32_setsockopt(0x3, 0xffff, 0x80,=20
> >  0xffffffffffffd590
> > , 0x8) =3D 0
> >  6477      1 cvsup    netbsd32_setsockopt(0x3, 0x6, 0x1,=20
> >  0xffffffffffffd58c, 0x
> > 4) =3D 0
> >  6477      1 cvsup    netbsd32_fcntl(0x3, 0x3, 0) =3D 2
> >  6477      1 cvsup    netbsd32_fcntl(0x3, 0x4, 0x6) =3D 0
> >  6477      1 cvsup    netbsd32_connect(0x3, 0xffffffffffffd570, 0x10)=
=20
> >  Err#36 EI
> > NPROGRESS
> >  6477      1 cvsup    netbsd32_select(0x4, 0x817c144, 0x817c154, 0x817c=
164,=20
> >  0)
> > =3D 1
> >  6477      1 cvsup    netbsd32_connect(0x3, 0xffffffffffffd5a0, 0x10)=
=20
> >  Err#56 EI
> > SCONN
> >  6477      1 cvsup    netbsd32_setsockopt(0x3, 0x6, 0x1,=20
> >  0xffffffffffffd5cc, 0x
> > 4) =3D 0
> >  6477      1 cvsup    netbsd32_fcntl(0x1, 0x3, 0) =3D 2
> >  6477      1 cvsup    netbsd32_fcntl(0x1, 0x4, 0x6) =3D 0
> >  6477      1 cvsup    netbsd32_write(0x1, 0x817f00c, 0x21) =3D 33
> >       "Connected to cvsup.us.netbsd.org\n"
> >  6477      1 cvsup    netbsd32_fcntl(0x1, 0x4, 0x2) =3D 0
> > xffffffffffffd610, 0x80e9a1c, 0x1) Err#22 EINVAL
> >  6477      1 cvsup    netbsd32_write(0x2, 0x816a0c0, 0xa1) =3D 161
> >       "\n\n***\n*** runtime error:\n***    ASSERT failed\n***    file=
=20
> >       "/usr/"
> >  6477      1 cvsup    netbsd32_write(0x2, 0x816a0c0, 0x2f) =3D 47
> >       "  use option @M3stackdump to get a stack trace\n"
> >  6477      1 cvsup    netbsd32___sigprocmask14(0x3, 0xffffffffffffd510,=
 0)=20
> >  =3D 0
> >  6477      1 cvsup    getpid()                    =3D 6477
> >  6477      1 cvsup    netbsd32_kill(0x194d, 0x6)  =3D 0
> >  6477      1 cvsup    SIGABRT SIG_DFL
> >=20
> >=20
> > I assume it's trying to set a socket option that is not supported by
> > the emulation; am I right?
>=20
> No.  The trace doesn't make it obvious what happens.  The sequence that
> leads to it printing "Connected to cvsup.us.netbsd.org" seems just fine
> to me (that's how you do a TCP connect with a non-blocking socket).

I might need more sleep.

I think you removed a part of the trace...  Something has to be missing
before the line that shows EINVAL...

--=20
Quentin Garnier - cube@cubidou.net - cube@NetBSD.org
"You could have made it, spitting out benchmarks
Owe it to yourself not to fail"
Amplifico, Spitting Out Benchmarks, Hometakes Vol. 2, 2005.

--DA7REkwdj/3NVgqE
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (NetBSD)

iQEVAwUBRdXahtgoQloHrPnoAQKVAwf/fYEXBrrDWpAfulzcKdtexpa8wcfVFL3G
OuwBHawv9KaOaLkEgojM6ctqsqj7dt+T3b5y5fxSh23rSwElfPMeYrBfd0ubBLpq
M1GBcnNcRiFw7D6VXIoG+MZ9tyOz5SrVPNwinsxELDrUAMo+XzJ5nsX3n2kq14aD
FXc3vOnF0Vz10WVFHPs0Kkx7rwea3KsrT1CjR36FFVup+EZjjiuNJCoBqv5tl5aK
JDPOnuIpT4rDzKHfm1VPh+TxIe0UKVTHwt+3kTXFoMXc0gpd+neJA8UCkORayZ6Y
bhZzY/tZ7Kr+X1SZ38sy7Hajvh69m8CnRij4HShYHQUFsydp1TM3Og==
=HsMj
-----END PGP SIGNATURE-----

--DA7REkwdj/3NVgqE--