Date: Mon, 24 Feb 2014 15:26:14 +0200 From: John Hay <jhay@meraka.org.za> To: Ian Lepore <ian@FreeBSD.org> Cc: freebsd-arm@FreeBSD.org Subject: Re: status of AVILA and CAMBRIA code Message-ID: <20140224132614.GA31984@zibbi.meraka.csir.co.za> In-Reply-To: <1393001249.1145.98.camel@revolution.hippie.lan> References: <20140219105934.GA74731@zibbi.meraka.csir.co.za> <20140219172938.GH34851@funkthat.com> <20140221130530.GA202@zibbi.meraka.csir.co.za> <1393001249.1145.98.camel@revolution.hippie.lan>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi Ian,
On Fri, Feb 21, 2014 at 09:47:29AM -0700, Ian Lepore wrote:
> On Fri, 2014-02-21 at 15:05 +0200, John Hay wrote:
> > On Wed, Feb 19, 2014 at 09:29:38AM -0800, John-Mark Gurney wrote:
> > > John Hay wrote this message on Wed, Feb 19, 2014 at 12:59 +0200:
> > > > What is the status of AVILA and CAMBRIA builds in our tree? Have anybody
> > > > had success recently? From the lists I can see that other people have
> > > > also asked in September 2013, but I cannot figure out if there was any
> > > > successes at that stage. :-)
> > ...
> > > >
> > > > Somewhere along the line the ethernet npe0 device also broke, writing
> > > > "npe0: npestart_locked: too many fragments 0". But I did not test the
> > > > npe0 device with every build and only realised it this morning, so I
> > > > do not know where it broke. :-)
> > >
> > > Not sure about these issues...
> >
> > Ok, I found the place. It is svn 246713 by kib:
> >
> > ###########
> > Reform the busdma API so that new types may be added without modifying
> > every architecture's busdma_machdep.c. It is done by unifying the
> > bus_dmamap_load_buffer() routines so that they may be called from MI
> > code. The MD busdma is then given a chance to do any final processing
> > in the complete() callback.
> >
> > The cam changes unify the bus_dmamap_load* handling in cam drivers.
> >
> > The arm and mips implementations are updated to track virtual
> > addresses for sync(). Previously this was done in a type specific
> > way. Now it is done in a generic way by recording the list of
> > virtuals in the map.
> >
> > Submitted by: jeff (sponsored by EMC/Isilon)
> > Reviewed by: kan (previous version), scottl,
> > mjacob (isp(4), no objections for target mode changes)
> > Discussed with: ian (arm changes)
> > Tested by: marius (sparc64), mips (jmallet), isci(4) on x86 (jharris),
> > amd64 (Fabian Keil <freebsd-listen@fabiankeil.de>
> > ###########
> >
> > After that tx packets will cause this message:
> > npe0: npestart_locked: too many fragments.
> >
> > Then updating to 246881 by ian:
> >
> > #############
> > In _bus_dmamap_addseg(), the return value must be zero for error, or the size
> > actually added to the segment (possibly smaller than the requested size if
> > boundary crossings had to be avoided).
> > #############
> >
> > This makes it a bit better in that some packets seem to go through. It
> > looks like 3 out of 4 will go out and the forth will cause the same
> > message as above.
> >
> > I have added a printf just above bus_dmamap_load_mbuf_sg() in
> > npestart_locked() to show some of the mbuf values:
> >
> > ############
> > npe0: npestart_locked: m_len 42, data 0xc0d3dcd6, next 0
> > [...]
> > ############
> >
> > Any ideas?
> >
> > John
>
> I can't see the path through the busdma code that leads to that result.
> It looks like there are two ways the mapping can fail, maybe it would
> help to know which path it's taking. In arm/busdma_machdep.c the two
> error paths out of the mapping loop are at lines 1066 and 1077, could
> you try putting printfs at those locations so we know which case is
> happening? Maybe printing some of the values involved with taking those
> exits would be helpful too.
>
> Maybe also check for map->sync_count being non-zero on entry to
> _bus_dmamap_load_buffer() and print something if it is. The printfs you
> did earlier make it look like there's never actually a second mbuf
> chained off the first (which makes sense for such small packets), I
> think that number should always be zero on entry because the outer loop
> in kern/subr_bus_dma.c should only ever run once.
Ok, here is the patch I used to add printfs to arm/busdma_machdep.c
###################
Index: arm/busdma_machdep.c
===================================================================
--- arm/busdma_machdep.c (revision 246713)
+++ arm/busdma_machdep.c (working copy)
@@ -1008,6 +1008,8 @@
vm_offset_t vaddr = (vm_offset_t)buf;
int error = 0;
+ if (map->sync_count != 0)
+ printf("_bus_dmamap_load_buffer: map->sync_count %d\n", map->sync_count);
if (segs == NULL)
segs = dmat->segments;
if ((flags & BUS_DMA_LOAD_MBUF) != 0)
@@ -1052,8 +1054,10 @@
sl = &map->slist[map->sync_count - 1];
if (map->sync_count == 0 ||
vaddr != sl->vaddr + sl->datacount) {
- if (++map->sync_count > dmat->nsegments)
+ if (++map->sync_count > dmat->nsegments) {
+ printf("_bus_dmamap_load_buffer: map->sync_count %d, dmat->nsegments %d\n", map->sync_count, dmat->nsegments);
goto cleanup;
+ }
sl++;
sl->vaddr = vaddr;
sl->datacount = sgsize;
@@ -1064,6 +1068,8 @@
sgsize = _bus_dmamap_addseg(dmat, map, curaddr, sgsize, segs,
segp);
if (sgsize == 0)
+ printf("_bus_dmamap_load_buffer: sgsize == 0, dmat flags %x\n", dmat->flags);
+ if (sgsize == 0)
break;
vaddr += sgsize;
buflen -= sgsize;
@@ -1075,6 +1081,7 @@
*/
if (buflen != 0) {
_bus_dmamap_unload(dmat, map);
+ printf("_bus_dmamap_load_buffer: buflen %jd\n", (intmax_t)buflen);
return (EFBIG); /* XXX better return value here? */
}
return (0);
###################
The result looks like this. With printfs in _bus_dmamap_load_buffer(), we
also see messages for the receive side.
###################
ping 146.64.5.1
PING 146.64.5.1 (146.64.5.1): 56 data bytes
npe0: npestart_locked: m_len 98, data 0xc0d3ed66, next 0
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=0 ttl=64 time=9.089 ms
npe0: npestart_locked: m_len 98, data 0xc0d3ea66, next 0
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=1 ttl=64 time=9.262 ms
npe0: npestart_locked: m_len 98, data 0xc0d3e766, next 0
_bus_dmamap_load_buffer: map->sync_count 2
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=2 ttl=64 time=9.291 ms
npe0: npestart_locked: m_len 98, data 0xc0d3e466, next 0
_bus_dmamap_load_buffer: map->sync_count 3
_bus_dmamap_load_buffer: map->sync_count 4, dmat->nsegments 3
_bus_dmamap_load_buffer: buflen 98
npe0: npestart_locked: too many fragments 0
npe0: npestart_locked: m_len 98, data 0xc0d3e366, next 0
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=4 ttl=64 time=5.494 ms
npe0: npestart_locked: m_len 98, data 0xc0d3e066, next 0
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=5 ttl=64 time=9.294 ms
npe0: npestart_locked: m_len 98, data 0xc0d3dc66, next 0
_bus_dmamap_load_buffer: map->sync_count 2
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=6 ttl=64 time=9.323 ms
npe0: npestart_locked: m_len 98, data 0xc0d3d966, next 0
_bus_dmamap_load_buffer: map->sync_count 3
_bus_dmamap_load_buffer: map->sync_count 4, dmat->nsegments 3
_bus_dmamap_load_buffer: buflen 98
npe0: npestart_locked: too many fragments 0
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
npe0: npestart_locked: m_len 98, data 0xc0d3d866, next 0
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=8 ttl=64 time=5.509 ms
npe0: npestart_locked: m_len 98, data 0xc0d3d566, next 0
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=9 ttl=64 time=9.316 ms
npe0: npestart_locked: m_len 98, data 0xc0d3d266, next 0
_bus_dmamap_load_buffer: map->sync_count 2
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=10 ttl=64 time=9.266 ms
npe0: npestart_locked: m_len 98, data 0xc0d3d166, next 0
_bus_dmamap_load_buffer: map->sync_count 3
_bus_dmamap_load_buffer: map->sync_count 4, dmat->nsegments 3
_bus_dmamap_load_buffer: buflen 98
npe0: npestart_locked: too many fragments 0
npe0: npestart_locked: m_len 98, data 0xc0d3d066, next 0
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=12 ttl=64 time=5.469 ms
npe0: npestart_locked: m_len 98, data 0xc0d3d366, next 0
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=13 ttl=64 time=9.276 ms
npe0: npestart_locked: m_len 98, data 0xc0d3d666, next 0
_bus_dmamap_load_buffer: map->sync_count 2
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=14 ttl=64 time=9.302 ms
npe0: npestart_locked: m_len 98, data 0xc0d3db66, next 0
_bus_dmamap_load_buffer: map->sync_count 3
_bus_dmamap_load_buffer: map->sync_count 4, dmat->nsegments 3
_bus_dmamap_load_buffer: buflen 98
npe0: npestart_locked: too many fragments 0
npe0: npestart_locked: m_len 98, data 0xc0d3da66, next 0
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=16 ttl=64 time=5.484 ms
npe0: npestart_locked: m_len 98, data 0xc0d3dd66, next 0
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 1
_bus_dmamap_load_buffer: map->sync_count 2, dmat->nsegments 1
_bus_dmamap_load_buffer: buflen 1536
64 bytes from 146.64.5.1: icmp_seq=17 ttl=64 time=9.306 ms
###################
Regards
John
--
John Hay -- jhay@meraka.csir.co.za / jhay@meraka.org.za
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20140224132614.GA31984>