DPDK patches and discussions
 help / color / mirror / Atom feed
From: Hamid Ramazani <mrramazani@gmail.com>
To: Dan Kan <dan@nyansa.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] Useful: Simple C++ program & Makefile
Date: Tue, 29 Apr 2014 20:16:02 +0330	[thread overview]
Message-ID: <CACu58pCX0dcxeBkTkdtuTZnzbW1Bf6VcJvJkzoGO05GJiEjpzA@mail.gmail.com> (raw)
In-Reply-To: <CA+RRbcp9VqCkfDKYwYLHsOWJsCKMWGwQ1ViXB=iujPa1rRY4jA@mail.gmail.com>

Hi,

this post might be a little old, but there is point which might be
really important for C++ developers using DPDK:

C++, especially C++11, doesn't support "designated initializes" of C,
instead of that you can use "memset" and set the instance of a
struct/class to 0/NULL/nullptr and then you can assign proper values
to the members of that struct/class by . or -> (if it is pointer).

Remember that even empty-initialization list in C++ (just a simple {}
) doesn't set things to 0/NULL/nullptr/flase (default values), and you
may have garbage values which cause problems in your application.

Hope this saves you from lots of headache.

All the Best,
Hamid

On 1/11/14, Dan Kan <dan@nyansa.com> wrote:
> I forgot to add that you will also need to add -D__STDC_LIMIT_MACROS
> because C99 standard specifies that limits such as INT8_MAX should only be
> defined if explicitly requested. Also with g++, you can no longer use
> non-trivial designated initializers which are used extensively throughout
> dpdk sample apps. For example,
>
> static const struct rte_eth_rxconf rx_conf = {
>     .rx_thresh = {
>         .pthresh = RX_PTHRESH,
>         .hthresh = RX_HTHRESH,
>         .wthresh = RX_WTHRESH,
>     },
>     .rx_free_thresh = 32,
> };
>
>
> Basically, use gcc to build the dpdk libraries. Then, use g++ to build your
> own code and link against the built libraries.
>
> Dan
>
>
> On Fri, Jan 10, 2014 at 12:19 PM, Dan Kan <dan@nyansa.com> wrote:
>
>> Hamid,
>> I'm in the same situation as you in which I would like to write most of
>> the application logic in C++. I was able to use CC=g++ by slightly
>> modifying the makefiles in mk. I replaced all occurrences of "%.c" with
>> "%.cc" using the following command.
>>
>> find . -name "*.mk" -exec sed -i 's/%\.c\([^[:alpha:]]\)/%\.cc\1/g' {} \;
>>
>> I also remove some warning flags as errors.
>>
>> Here are the diffs:
>>
>> diff -r mk/internal/rte.compile-pre.mk ../temp/dpdk-1.5.1r2/mk/internal/
>> rte.compile-pre.mk
>> 39c39
>> < src2obj = $(strip $(patsubst %.cc,%.o,\
>> ---
>> > src2obj = $(strip $(patsubst %.c,%.o,\
>> 48c48
>> < src2dep = $(strip $(call dotfile,$(patsubst %.cc,%.o.d, \
>> ---
>> > src2dep = $(strip $(call dotfile,$(patsubst %.c,%.o.d, \
>> 53c53
>> < src2cmd = $(strip $(call dotfile,$(patsubst %.cc,%.o.cmd, \
>> ---
>> > src2cmd = $(strip $(call dotfile,$(patsubst %.c,%.o.cmd, \
>> 125c125
>> < %.o: %.cc $$(wildcard $$(dep_$$@)) $$(DEP_$$(@)) FORCE
>> ---
>> > %.o: %.c $$(wildcard $$(dep_$$@)) $$(DEP_$$(@)) FORCE
>> diff -r mk/rte.module.mk ../temp/dpdk-1.5.1r2/mk/rte.module.mk
>> 36,37c36,37
>> < ifneq ($(MODULE),$(notdir $(SRCS-y:%.cc=%)))
>> < $(MODULE)-objs += $(notdir $(SRCS-y:%.cc=%.o))
>> ---
>> > ifneq ($(MODULE),$(notdir $(SRCS-y:%.c=%)))
>> > $(MODULE)-objs += $(notdir $(SRCS-y:%.c=%.o))
>> diff -r
>> mk/toolchain/gcc/rte.vars.mk../temp/dpdk-1.5.1r2/mk/toolchain/gcc/
>> rte.vars.mk
>> 71,73c71,73
>> < WERROR_FLAGS := -W -Wall -Werror
>> < WERROR_FLAGS += -Wmissing-declarations -Wpointer-arith
>> < WERROR_FLAGS += -Wcast-align -Wcast-qual
>> ---
>> > WERROR_FLAGS := -W -Wall -Werror -Wstrict-prototypes
>> > -Wmissing-prototypes
>> > WERROR_FLAGS += -Wmissing-declarations -Wold-style-definition
>> -Wpointer-arith
>> > WERROR_FLAGS += -Wcast-align -Wnested-externs -Wcast-qual
>>
>>
>> Dan
>>
>>
>> On Thu, Jan 9, 2014 at 10:30 PM, Hamid Ramazani
>> <mrramazani@gmail.com>wrote:
>>
>>> > I don't exactly know what is needed for C++. Please keep us informed.
>>>
>>> Hey Thomas,
>>>
>>> I've attached a simple program (main.cpp main.h and Makefile) that has
>>> a C++ class and just prints some messages in the output.
>>> Despite the fact that it's working fine, I'm sure the Makefile could
>>> be written much better; Maybe I made it completer in future.
>>>
>>> All the Best,
>>> --Hamid
>>>
>>> On 1/3/14, Thomas Monjalon <thomas.monjalon@6wind.com> wrote:
>>> > Hello,
>>> >
>>> > 03/01/2014 11:48, Hamid Ramazani :
>>> >> eal_timer.c:(.text+0x42c): undefined reference to `clock_gettime'
>>> >
>>> > From "man clock_gettime":
>>> > Link with -lrt (only for glibc versions before 2.17).
>>> >
>>> >>      g++ -m64 -pthread  -march=native -DRTE_MACHINE_CPUFLAG_SSE
>>> >> -DRTE_MACHINE_CPUFLAG_SSE2 -DRTE_MACHINE_CPUFLAG_SSSE3
>>> >>
>>> -DRTE_COMPILE_TIME_CPUFLAGS=RTE_CPUFLAG_SSE,RTE_CPUFLAG_SSE2,RTE_CPUFLAG_SS
>>> >> SE3 -I/home/hamid/dpdk/dpdk-1.5.1r1/examples/sample/build/include
>>> >> -I/home/hamid/dpdk/dpdk-1.5.1r1/x86_64-default-linuxapp-gcc/include
>>> >> -include
>>> >>
>>> /home/hamid/dpdk/dpdk-1.5.1r1/x86_64-default-linuxapp-gcc/include/rte_conf
>>> >> ig.h -O3 -W -Wall -Werror -Wmissing-declarations -Wpointer-arith
>>> >> -Wcast-align -Wcast-qual -Wformat-nonliteral -Wformat-security
>>> >> -Wundef
>>> >> -Wwrite-strings -Wl,-melf_x86_64 -Wl,-export-dynamic sample.cpp -o
>>> >> sample -Wl,-L/home/hamid/dpdk/dpdk-1.5.1r1/examples/sample/build/lib
>>> >> -Wl,-L/home/hamid/dpdk/dpdk-1.5.1r1/x86_64-default-linuxapp-gcc/lib
>>> >> -Wl,-L/home/hamid/dpdk/dpdk-1.5.1r1/x86_64-default-linuxapp-gcc/lib
>>> >> -Wl,-lrte_kni -Wl,-lrte_pmd_e1000 -Wl,-lrte_pmd_ixgbe -Wl,-lrte_mbuf
>>> >> -Wl,-lrte_cmdline -Wl,-lrte_timer -Wl,-lrte_hash -Wl,-lrte_lpm
>>> >> -Wl,--start-group -Wl,-lethdev -Wl,-lrte_malloc -Wl,-lrte_mempool
>>> >> -Wl,-lrte_ring -Wl,-lrte_eal -Wl,-ldl -Wl,--end-group
>>> >
>>> > Try CONFIG_RTE_BUILD_COMBINE_LIBS=y and -lintel_dpdk instead of all
>>> these
>>> > libraries. You can also remove the warning options if you want.
>>> >
>>> > You can also try to build your Makefile by including files like
>>> > mk/rte.extapp.mk and defining CC=g++.
>>> > I don't exactly know what is needed for C++. Please keep us informed.
>>> >
>>> > --
>>> > Thomas
>>> >
>>>
>>
>>
>

      reply	other threads:[~2014-04-29 16:45 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-01-10  6:30 Hamid Ramazani
2014-01-10 20:19 ` Dan Kan
2014-01-10 23:42   ` Dan Kan
2014-04-29 16:46     ` Hamid Ramazani [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CACu58pCX0dcxeBkTkdtuTZnzbW1Bf6VcJvJkzoGO05GJiEjpzA@mail.gmail.com \
    --to=mrramazani@gmail.com \
    --cc=dan@nyansa.com \
    --cc=dev@dpdk.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).