Hi
when I run the l3fwd application with the below command line, I am observing
the error "Cannot configure device: err=-22, port=0". The complete log is
posted below.
./build/l3fwd -c 0x03 -n 2 -- -p 0x03 --config="(0,0,0),(1,1,1)"
Setup:
a) Virtual machine booted with Fedora20 with 2 virtual
2014-04-18 09:20, Neil Horman:
> On Fri, Apr 18, 2014 at 02:08:56PM +0200, Thomas Monjalon wrote:
> > 2014-04-18 08:04, Neil Horman:
> > > On Fri, Apr 18, 2014 at 04:42:01AM -0700, Thomas Monjalon wrote:
> > > > 2014-04-15 14:05, Neil Horman:
> > > > > Rather than have each driver have to remember
2014-04-18 09:18, Neil Horman:
> On Fri, Apr 18, 2014 at 01:23:19PM +0200, Thomas Monjalon wrote:
> > I think that CPU_LDFLAGS should be prefixed with -Wl, in case of CC
> > linking. So blindly assigning CC to LD variable seems a bad idea.
> > Other makefiles have different O_TO_S commands dependin
There is no need for a 'magic' field in struct rte_config, as this part of the
structure is local to each process. All threads of a process are synchronised
because of the run_once atomic.
So remove this field, as it is only adding confusion when reading code that
references 'magic' field from stru
There should be no real need for this initialised field as the whole structure
is set to 0 in rte_config_init() by primary process, and secondary processes
wait for this to happen before anything else (looking at mem_config magic).
Signed-off-by: David Marchand
---
lib/librte_eal/common/include/
We don't really need this field as it is only used when creating the memzone
object associated to this heap.
Signed-off-by: David Marchand
---
lib/librte_eal/common/include/rte_malloc_heap.h |1 -
lib/librte_malloc/malloc_heap.c | 13 +
2 files changed, 5 insert
Following Neil's suggestion, here is a patchset that removes the need for a
synchronisation mechanism when initialising heap objects.
As a consequence, this patchset replaces the two patches Didier proposed
earlier.
--
David Marchand
David Marchand (2):
malloc: get rid of numa_socket field
m
Hi Gopi
I recently run into the same problem when using 82576 with igb_uio on VM.
The problem is not e1000 or igb. The problem is with the other network
device managed by virtio-pci driver
Notice that when application polls for ETH devices it prints out
EAL: probe driver: 1af4:1000 rte_virtio_pm
Hi,
2014-04-18 17:39, B Gopikrishna:
> when I run the l3fwd application with the below command line, I am observing
> the error "Cannot configure device: err=-22, port=0". The complete log is
> posted below.
Could you try to reproduce it with the git HEAD version?
It may be fixed by this commit:
2014-04-18 08:04, Neil Horman:
> On Fri, Apr 18, 2014 at 04:42:01AM -0700, Thomas Monjalon wrote:
> > 2014-04-15 14:05, Neil Horman:
> > > Rather than have each driver have to remember to add a constructor to it
> > > to
> > > make sure its gets registered properly, wrap that process up in a macro
I have some comments inline.
2014-04-15 14:06, Neil Horman:
> Currently, physical device pmds use a separate initalization path
> (rte_pmd_init_all) while virtual devices use a constructor registration and
> rte_eal_dev_init. Theres no reason to have them be separate. This patch
> removes the vd
On Fri, Apr 18, 2014 at 03:32:15PM +0200, Thomas Monjalon wrote:
> 2014-04-18 09:20, Neil Horman:
> > On Fri, Apr 18, 2014 at 02:08:56PM +0200, Thomas Monjalon wrote:
> > > 2014-04-18 08:04, Neil Horman:
> > > > On Fri, Apr 18, 2014 at 04:42:01AM -0700, Thomas Monjalon wrote:
> > > > > 2014-04-15 1
On Fri, Apr 18, 2014 at 03:29:01PM +0200, Thomas Monjalon wrote:
> 2014-04-18 09:18, Neil Horman:
> > On Fri, Apr 18, 2014 at 01:23:19PM +0200, Thomas Monjalon wrote:
> > > I think that CPU_LDFLAGS should be prefixed with -Wl, in case of CC
> > > linking. So blindly assigning CC to LD variable seem
Hi Neil,
2014-04-16 09:51, Neil Horman:
> The shared libraries built with the current makefile set produce static
> libraries rather than actual shared objects. This is due to several missing
> options that are required to correctly build shared objects using ld, as
> well as a mis-specified -sha
On Fri, Apr 18, 2014 at 02:58:08PM +0200, David Marchand wrote:
> There is no need for a 'magic' field in struct rte_config, as this part of the
> structure is local to each process. All threads of a process are synchronised
> because of the run_once atomic.
> So remove this field, as it is only ad
On Fri, Apr 18, 2014 at 02:08:56PM +0200, Thomas Monjalon wrote:
> 2014-04-18 08:04, Neil Horman:
> > On Fri, Apr 18, 2014 at 04:42:01AM -0700, Thomas Monjalon wrote:
> > > 2014-04-15 14:05, Neil Horman:
> > > > Rather than have each driver have to remember to add a constructor to it
> > > > to
> >
On Fri, Apr 18, 2014 at 01:23:19PM +0200, Thomas Monjalon wrote:
> Hi Neil,
>
> 2014-04-16 09:51, Neil Horman:
> > The shared libraries built with the current makefile set produce static
> > libraries rather than actual shared objects. This is due to several missing
> > options that are required
On Fri, Apr 18, 2014 at 02:56:18PM +0200, David Marchand wrote:
> There should be no real need for this initialised field as the whole structure
> is set to 0 in rte_config_init() by primary process, and secondary processes
> wait for this to happen before anything else (looking at mem_config magic
On Fri, Apr 18, 2014 at 02:56:17PM +0200, David Marchand wrote:
> We don't really need this field as it is only used when creating the memzone
> object associated to this heap.
>
> Signed-off-by: David Marchand
Acked-by: Neil Horman
> ---
> lib/librte_eal/common/include/rte_malloc_heap.h |
Thanks, Thomas.
2014-04-18 5:58 GMT+08:00 Thomas Monjalon :
> Hi,
>
> 2014-04-15 11:03, Wang Sheng-Hui:
> > Parse args first, to resolve any invalid args and give out the usage
> string.
> > E.g './helloworld --invalid', the '--invalid' will be checked before any
> > init. After the options are
On Fri, Apr 18, 2014 at 04:42:01AM -0700, Thomas Monjalon wrote:
> 2014-04-15 14:05, Neil Horman:
> > Rather than have each driver have to remember to add a constructor to it to
> > make sure its gets registered properly, wrap that process up in a macro to
> > make registration a one line affair.
2014-04-15 14:05, Neil Horman:
> Rather than have each driver have to remember to add a constructor to it to
> make sure its gets registered properly, wrap that process up in a macro to
> make registration a one line affair. This also sets the stage for us to
> make registration of vdev pmds and p
Hi.
I'm facing a problem where I stop receiving packets after I have
received as many packets as the number of RX descriptors. More details
follow:
I have a small program where a server machine and a client machine
execute the following loop:
int nb_rx_new = rte_eth_rx_burst(0, 0, rx_pkts_burst,
> > When loading a library "libfoo.so" (depending on "libbar.so", located in
> > an
> > entirely different folder), with a LD_LIBRARY_PATH=/path/to/libfoo.so", it
> >
> > returns an error:
> > EAL: ./libfoo.so: cannot open shared object file: No such file or
> > directory
> >
> > If the first d
> > lcores that are set in coremask should be checked against lcores detected
> > on system. This way, we won't need to check them later.
> >
> > Besides, if specifying an unavailable lcore, we currently panic in
> > eal_thread_loop() because pthread_setaffinity_np fails.
> > So this check will re
2014-04-16 06:50, Neil Horman:
> The return code for rte_cpu_get_flag_enabled is only checked on the
> termination of the for loop that it is called inside, but should be checked
> for every iteration it makes through the for loop. This is caused by some
> silly missing brackets. Simply add them
2014-04-15 15:51, David Marchand:
> From: Jean-Mickael Guerin
>
> For RH 6.5:
> - always include mdio.h to get the definitions of MDIO_EEE, ETHTOOL_GEEE
> - is_link_local_ether_addr(), pcie_capability_clear_and_set_word(), and
> ether_addr_equal() have been backported
>
> For RH 6.4:
> - same
> From: Jean-Mickael Guerin
>
> On RH 6.5:
> igb_main.c:2298: error: unknown field ?ndo_fdb_add? specified in
> initializer
>
> FDB ops are present in RH 6.5 via the extension of netdev, so add the
> ifdef inside the netdev ops definition of igb.
>
> However, FDB functions are not set for RHEL
28 matches
Mail list logo
