Hi Johan it looks like the tg3 driver is not in sync with latest pf_ring aware api, we need to fix it.
Best Regards Alfredo On Jul 31, 2013, at 6:01 PM, Johan Karlsson <[email protected]> wrote: > Hi! > > I’m currently running PF_RING 5.2.0 with the bundled tg3-3.102 driver on a > server running Debian 6 (x64). For a few reasons I wan’t to upgrade to a > newer PF_RING release (or even the SVN). However, in all versions above > 5.2.0, the bundled tg3-3.102 driver fails to compile. > > First, just a minor issue: it seems to now be found under the path > “drivers/PF_RING_aware/2.6.x/Broadcom/…”. This should be > “drivers/PF_RING_aware/Broadcom”. Easily solved though, and isn’t the root > cause of the issues I’m experiencing, just a side note. > > Here’s the output from the make: > > --- > sh makeflags.sh /lib/modules/2.6.32-5-amd64/source > tg3_flags.h > make -C /lib/modules/2.6.32-5-amd64/build > SUBDIRS=/usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102 > modules > make[1]: Entering directory `/usr/src/linux-headers-2.6.32-5-amd64' > CC [M] > /usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/tg3.o > In file included from > /usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/tg3.c:67: > /usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/tg3.h:2814:1: > warning: "PHY_ID_BCM5411" redefined > In file included from > /usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/tg3.c:36: > /usr/src/linux-headers-2.6.32-5-common/include/linux/brcmphy.h:7:1: warning: > this is the location of the previous definition > In file included from > /usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/tg3.c:73: > /usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/../../../../../kernel/linux/pf_ring.h:522:81: > error: missing binary operator before token "(" > /usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/tg3.c: > In function ‘tg3_rx’: > /usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/tg3.c:4646: > warning: passing argument 4 of ‘hook->ring_handler’ makes pointer from > integer without a cast > /usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/tg3.c:4646: > error: too few arguments to function ‘hook->ring_handler’ > make[4]: *** > [/usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/tg3.o] > Error 1 > make[3]: *** > [_module_/usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102] > Error 2 > make[2]: *** [sub-make] Error 2 > make[1]: *** [all] Error 2 > make[1]: Leaving directory `/usr/src/linux-headers-2.6.32-5-amd64' > make: *** [default] Error 2 > --- > > The following line can sort of be ignored: > > --- > /usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/../../../../../kernel/linux/pf_ring.h:522:81: > error: missing binary operator before token "(" > --- > > Checking in pf_ring.h, this line is the offender: > > --- > #if defined(I82599_HW_FILTERING_SUPPORT) && (LINUX_VERSION_CODE < > KERNEL_VERSION(2,6,40)) > --- > > Since I’m running 2.6.32, I omitted the kernel version check: > > --- > #if defined(I82599_HW_FILTERING_SUPPORT) > --- > > This got rid of the line in the make output above, but the make still fails > with what can be described as the “real” error here: > > --- > /usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/tg3.c:4646: > warning: passing argument 4 of ‘hook->ring_handler’ makes pointer from > integer without a cast > /usr/local/src/PF_RING-6641/drivers/PF_RING_aware/broadcom/tg3/tg3-3.102/tg3.c:4646: > error: too few arguments to function ‘hook->ring_handler’ > --- > > Comparing the code from the tg3 driver bundled with PF_RING 5.2.0 it hasn’t > changed at all, so I don’t know why it’s complaining regarding that function > since it’s identical as the one in 5.2.0. Before I noticed they were > identical I tried to simply copy the code (the whole tg3 directory) from my > 5.2.0 directory to the one where I grabbed the SVN (6641) to. But with the > same error, naturally. I’m no programmer, but I’m guessing something in the > PF_RING header files or such are triggering this condition. > > Any ideas? > > Regards, > > Johan > _______________________________________________ > Ntop-misc mailing list > [email protected] > http://listgateway.unipi.it/mailman/listinfo/ntop-misc
_______________________________________________ Ntop-misc mailing list [email protected] http://listgateway.unipi.it/mailman/listinfo/ntop-misc
