On 3/6/2018 8:45 AM, Varghese, Vipin wrote: > Hi Ferruh, > >> -----Original Message----- >> From: Yigit, Ferruh >> Sent: Monday, March 5, 2018 2:33 PM >> To: Varghese, Vipin <vipin.vargh...@intel.com>; firstname.lastname@example.org; Pattan, >> Reshma <reshma.pat...@intel.com> >> Cc: Mcnamara, John <john.mcnam...@intel.com> >> Subject: Re: [dpdk-dev] [PATCH v1] app/pdump: add check for PCAP PMD >> >> On 3/5/2018 7:57 AM, Vipin Varghese wrote: >>> dpdk-pdump makes use of LIBRTE_PMD_PCAP for interfacing the ring to >>> the device-queue pair. Updating Makefile to check for the same. >>> >>> Signed-off-by: Vipin Varghese <vipin.vargh...@intel.com> >>> --- >>> app/pdump/Makefile | 4 ++++ >>> 1 file changed, 4 insertions(+) >>> >>> diff --git a/app/pdump/Makefile b/app/pdump/Makefile index >>> bd3c208..038a34f 100644 >>> --- a/app/pdump/Makefile >>> +++ b/app/pdump/Makefile >>> @@ -3,6 +3,10 @@ >>> >>> include $(RTE_SDK)/mk/rte.vars.mk >>> >>> +ifeq ($(CONFIG_RTE_LIBRTE_PMD_PCAP),n) $(error "Please enable >>> +CONFIG_RTE_LIBRTE_PMD_PCAP") endif >> >> pdump is enabled default, so won't this break the default build? > > Yes, you are right it will fail. Which then forces the user to enable PCAP.
We shouldn't break the default build because of missing dependencies. > >> >> What about moving this to lib/librte_pdump, convert $(error ..) to $(warning >> ..) >> and disable CONFIG_RTE_LIBRTE_PDUMP there? > > If we set to warning and there are no PCAP headers in build system. The > application gets built, but will fail internally becz the pcap API will fails > during execution. if CONFIG_RTE_LIBRTE_PDUMP disabled application won't be compiled > >> >>> + >>> ifeq ($(CONFIG_RTE_LIBRTE_PDUMP),y) >>> >>> APP = dpdk-pdump >>> >