Date: Mon, 10 Feb 2014 10:36:10 +0200 From: Viktor Penkoff <vpenkoff@gmail.com> To: Viktor <vpenkoff@gmail.com>, freebsd-drivers@freebsd.org Subject: Re: Receiving jumbo frames Message-ID: <CAE-a5gZRGyBQPbNEASvEKc-CzH4oVshkdj4b_Xd5ikdZauJ=gA@mail.gmail.com> In-Reply-To: <20140207164122.GE89104@funkthat.com> References: <52F49D36.8000103@gmail.com> <20140207164122.GE89104@funkthat.com>
next in thread | previous in thread | raw e-mail | index | archive | help
The line that the panic happens is in a standard invoking receive
interrupt function. Packet data is stored in memory buffers, with any
single packet spanning multiple buffers if necessary. The buffers are
allocated by the CPU and are managed through chained descriptor lists.
When I'm sending jumbo packets, e.g. 8008 bytes in size each, due to live
debugging with jtag and kgdb, I've got the following information for the
mbuf:
% ping -I eth0 -c 1 -s 8000 -M dont A.B.C.D -v -p ff
{m_hdr =3D {mh_next =3D 0x0, mh_nextpkt =3D 0x0, mh_data =3D 0xc3d41800 "",
mh_len =3D 2048, mh_flags =3D 3, mh_type =3D 1, pad =3D "\000"}, M_dat =
=3D {
MH =3D {MH_pkthdr =3D {rcvif =3D 0x0, header =3D 0x0, len =3D 2048, flo=
wid =3D 0,
csum_flags =3D 0, csum_data =3D 0, tso_segsz =3D 0, PH_vt =3D {
vt_vtag =3D 0, vt_nrecs =3D 0}, tags =3D {slh_first =3D 0x0},
dsa_tag =3D {0, 0}}, MH_dat =3D {MH_ext =3D {ext_buf =3D 0xc3d41800=
"",
ext_free =3D 0, ext_arg1 =3D 0x0, ext_arg2 =3D 0x0, ext_size =3D =
2048,
ref_cnt =3D 0xc3d3f244, ext_type =3D 6},
In this case, the given mbuf must occupy ~4 descriptors, IMO. Following the
next pointer in the descriptor list, the next one shows that the CPU has
an ownership. Nevertheless, =F2he byte count of the received frame shows th=
at
8056 bytes are received.
More interesting is the fact that an echo reply is given back. But after a
constant count of received packets, the kernel hangs.
2014-02-07 18:41 GMT+02:00 John-Mark Gurney <jmg@funkthat.com>:
> Viktor wrote this message on Fri, Feb 07, 2014 at 10:45 +0200:
> > Hi, folks! I'm writing an extension functionality to not-yet published
> > network driver.
> > I'm receiving the typical ethernet frames without problems. Considering
> > the datasheet of the device,
> > I'm capable of receiving jumbo frames. When I try to do that, e.g. to
> > send jumbo frame of 8000 bytes, I'm receiving only a limited count of
> > them - 105, then the kernel crashes with the following message:
> > "panic: vm_fault: fault on nofault entry, addr: cfcec000".
>
> This happens when you access an address in kmem (kernel space) that
> no longer has a page mapped... This is probably due to a pointer
> that was previously valid, but then you aren't suppose to have and the
> kernel unmapped the page because the subsystem free'd it...
>
> > I have inspected a kernel dump with kgdb and the problem occurs at the
> > function bus_dmamap_sync.
>
> What part of this? It could be that you are trying to _sync a buffer
> that doesn't belong to you anymore...
>
> Are you making sure that you allocate new frames to replace them in the
> receive buffer?
>
> > Some background information:
> > To enable the jumbo frame, one must set the appropriate register.
> > At the software level, a ring buffer with the descriptors is implemente=
d.
> > The device, under which the driver runs, is arm-based.
> > The freebsd version is 8.0. The device uses SerialDMA queues for
> > transmitting and receiving.
> > To receive packets, the CPU must perform the following:
> > 1. Prepare a linked list of descriptors
> > 2. Configure a given queue with the address of the first descriptor
> > in the list,
> > 3. enable SerialDMA;
> >
> > With the transmission - I don't have any problems. The logic is the sam=
e
> > as by the reception of packets - ring buffer with descriptors.
> >
> > Any ideas what can cause this type of crashes?
>
> With out better idea of what the driver is doing, or a line on which
> the panic happens, the above is about all I can think of...
>
> --
> John-Mark Gurney Voice: +1 415 225 5579
>
> "All that I will do, has been done, All that I have, has not."
>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAE-a5gZRGyBQPbNEASvEKc-CzH4oVshkdj4b_Xd5ikdZauJ=gA>