[dpdk-dev] Fwd: config: disable KNI for 32-bit because cannot work
On Thu, Jan 2, 2014 at 3:49 PM, Thomas Monjalon wrote: > Hello, > > Thank you for your suggestion. > I think you should send a patch on the mailing list to re-enable it. > > Thanks > -- > Thomas > > > 02/01/2014 11:10, neeraj kumar : > > Hi Thomas, > > > > According to commit no. (*d92c66f5ac53a7b803efcb4782e601ea99c38a7e*) we > > can't create KNI on 32 bit system b'coz > > > > *lib/librte_eal/linuxapp/kni/kni_misc.c:304:2: error: format '%llx' > expects > > argument of type 'long long unsigned int', but argument 2 has type > > 'phys_addr_t' [-Werror=format*] > > > > It is because by default in 32 bit kernel *High Memory Support* option of > > make menuconfig is selected with size 4 GB, therefore it gives > compilation > > error. > > > > Thus to use KNI on 32 bit system we need kernel with High memory support > of > > 64 GB, i.e *PAE* (*Physical Address Extension*) *Support*, then we can > be > > able to use it in 32 bit environment. > > > > hope you will update this on dpdk.org > > > > > > Regards, > > > > Neeraj >
[dpdk-dev] [PATCH] app/testpmd: fix RSS rx by setting mq_mode
Hello,
11/01/2014 03:46, Daniel Kan:
> --- a/app/test-pmd/testpmd.c
> +++ b/app/test-pmd/testpmd.c
> @@ -1546,6 +1546,9 @@ init_port_config(void)
> if (nb_rxq > 0) {
> port->dev_conf.rx_adv_conf.rss_conf.rss_key = NULL;
> port->dev_conf.rx_adv_conf.rss_conf.rss_hf = rss_hf;
> + if (nb_rxq > 1 && rss_hf != 0) {
> + port->dev_conf.rxmode.mq_mode = ETH_MQ_RX_RSS;
> + }
The fix looks good. Thanks.
Please, could you add a detailed explanation and a signed-off as described in
the guidelines:
http://dpdk.org/dev#send
I think it's a good idea to point to the commit which has introduced the bug.
Note also that "RSS rx" may be simply "RSS".
Thank you
--
Thomas
[dpdk-dev] [PATCH] app/testpmd: fix RSS rx by setting mq_mode
Hi Daniel,
Good catch :)
We plan to deliver a fix for that bug in coming DPDK release.
It is very similar to yours one, though we need extra check that there are no
VFs enabled for that port.
Thanks
Konstantin
-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Thomas Monjalon
Sent: Monday, January 13, 2014 10:31 AM
To: Daniel Kan
Cc: dev at dpdk.org
Subject: Re: [dpdk-dev] [PATCH] app/testpmd: fix RSS rx by setting mq_mode
Hello,
11/01/2014 03:46, Daniel Kan:
> --- a/app/test-pmd/testpmd.c
> +++ b/app/test-pmd/testpmd.c
> @@ -1546,6 +1546,9 @@ init_port_config(void)
> if (nb_rxq > 0) {
> port->dev_conf.rx_adv_conf.rss_conf.rss_key = NULL;
> port->dev_conf.rx_adv_conf.rss_conf.rss_hf = rss_hf;
> + if (nb_rxq > 1 && rss_hf != 0) {
> + port->dev_conf.rxmode.mq_mode = ETH_MQ_RX_RSS;
> + }
The fix looks good. Thanks.
Please, could you add a detailed explanation and a signed-off as described in
the guidelines:
http://dpdk.org/dev#send
I think it's a good idea to point to the commit which has introduced the bug.
Note also that "RSS rx" may be simply "RSS".
Thank you
--
Thomas
--
Intel Shannon Limited
Registered in Ireland
Registered Office: Collinstown Industrial Park, Leixlip, County Kildare
Registered Number: 308263
Business address: Dromore House, East Park, Shannon, Co. Clare
This e-mail and any attachments may contain confidential material for the sole
use of the intended recipient(s). Any review or distribution by others is
strictly prohibited. If you are not the intended recipient, please contact the
sender and delete all copies.
[dpdk-dev] [PATCH] app/testpmd: fix RSS rx by setting mq_mode
13/01/2014 12:18, Ananyev, Konstantin: > We plan to deliver a fix for that bug in coming DPDK release. > It is very similar to yours one, though we need extra check that there are > no VFs enabled for that port. Thanks for your comments about the VF. I think it will be clearer in a patch. Please, send a v2 patch or a new one on top of this one, following the guidelines: http://dpdk.org/dev#send so it can be tracked with git. Note: please remove your "confidential emails" footer for mailing list. -- Thomas
[dpdk-dev] [PATCH v2] app/testpmd: fix RSS by setting mq_mode
The mq_mode was not set when rxq is > 1; it's defaulted to ETH_MQ_RX_NONE.
As a result, RSS remains inactive. The fix is to set mq_mode to ETH_MQ_RX_RSS
when rxq is > 1 and hf is non-zero.
This bug was introduced by commit 243db2ddee3094a2cb39fdd4b17e26df4e7735e1
igb/ixgbe: ETH_MQ_RX_NONE should disable RSS
Signed-off-by: Daniel Kan
---
Updated commit log in accordance with dpdk guidelines
app/test-pmd/testpmd.c |3 +++
1 file changed, 3 insertions(+)
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index b11eb2e..355db0f 100644
--- a/app/test-pmd/testpmd.c
+++ b/app/test-pmd/testpmd.c
@@ -1546,6 +1546,9 @@ init_port_config(void)
if (nb_rxq > 0) {
port->dev_conf.rx_adv_conf.rss_conf.rss_key = NULL;
port->dev_conf.rx_adv_conf.rss_conf.rss_hf = rss_hf;
+ if (nb_rxq > 1 && rss_hf != 0) {
+ port->dev_conf.rxmode.mq_mode = ETH_MQ_RX_RSS;
+ }
} else {
port->dev_conf.rx_adv_conf.rss_conf.rss_key = NULL;
port->dev_conf.rx_adv_conf.rss_conf.rss_hf = 0;
--
1.7.9.5
[dpdk-dev] Changes to makefiles to allow building apps using g++
I made changes to makefiles to at least allow app development to be in done C++. I'm wondering if the community is interested in getting these changes as a patch. The dpdk library must still be compiled using gcc. However, the applications can be compiled using g++ by setting CC=g++, e.g. "make CC=g++". With g++, non-trivial designated initializers will no longer be allowed among others. As a result, most sample apps cannot be compiled using g++ without modifications; hello world app works with g++. You will also need to specify your c++ file extension if it's not named cpp, e.g. make CC=g++ CXX-suffix=cc. The behavior for gcc or icc compilation remains unmodified. Anyway, let me know if it's worthwhile to submit a patch. Thanks. Here is the preview of the patch. +++ mk/internal/rte.compile-pre.mk2014-01-13 14:14:34.278816220 -0800 @@ -33,11 +33,14 @@ # Common to rte.lib.mk, rte.app.mk, rte.obj.mk # +CXX-suffix = cpp + SRCS-all := $(SRCS-y) $(SRCS-n) $(SRCS-) # convert source to obj file src2obj = $(strip $(patsubst %.c,%.o,\ -$(patsubst %.S,%_s.o,$(1 +$(patsubst %.$(CXX-suffix),%.o,\ +$(patsubst %.S,%_s.o,$(1) # add a dot in front of the file name dotfile = $(strip $(foreach f,$(1),\ @@ -46,12 +49,14 @@ # convert source/obj files into dot-dep filename (does not # include .S files) src2dep = $(strip $(call dotfile,$(patsubst %.c,%.o.d, \ -$(patsubst %.S,,$(1) +$(patsubst %.$(CXX-suffix),%.o.d, \ +$(patsubst %.S,,$(1)) obj2dep = $(strip $(call dotfile,$(patsubst %.o,%.o.d,$(1 # convert source/obj files into dot-cmd filename src2cmd = $(strip $(call dotfile,$(patsubst %.c,%.o.cmd, \ -$(patsubst %.S,%_s.o.cmd,$(1) +$(patsubst %.$(CXX-suffix),%.o.cmd, \ +$(patsubst %.S,%_s.o.cmd,$(1)) obj2cmd = $(strip $(call dotfile,$(patsubst %.o,%.o.cmd,$(1 OBJS-y := $(call src2obj,$(SRCS-y)) @@ -137,6 +142,22 @@ $(depfile_newer)),\ $(C_TO_O_DO)) +%.o: %.$(CXX-suffix) $$(wildcard $$(dep_$$@)) $$(DEP_$$(@)) FORCE +@[ -d $(dir $@) ] || mkdir -p $(dir $@) +$(if $(D),\ +@echo -n "$< -> $@ " ; \ +echo -n "file_missing=$(call boolean,$(file_missing)) " ; \ +echo -n "cmdline_changed=$(call boolean,$(call cmdline_changed,$(C_TO_O))) " ; \ +echo -n "depfile_missing=$(call boolean,$(depfile_missing)) " ; \ +echo "depfile_newer=$(call boolean,$(depfile_newer))") +$(if $(or \ +$(file_missing),\ +$(call cmdline_changed,$(C_TO_O)),\ +$(depfile_missing),\ +$(depfile_newer)),\ +$(C_TO_O_DO)) + + # command to assemble a .S file to generate an object ifeq ($(USE_HOST),1) S_TO_O = $(CPP) $(HOST_CPPFLAGS) $($(@)_CPPFLAGS) $(HOST_EXTRA_CPPFLAGS) $< $(@).tmp && \ diff -ur ../temp/dpdk-1.5.1r2/mk/toolchain/gcc/rte.vars.mk mk/toolchain/gcc/ rte.vars.mk --- ../temp/dpdk-1.5.1r2/mk/toolchain/gcc/rte.vars.mk2014-01-02 07:03:19.0 -0800 +++ mk/toolchain/gcc/rte.vars.mk2014-01-13 14:07:33.148292590 -0800 @@ -68,9 +68,16 @@ endif endif +ifeq ($(CC), $(CROSS)g++) +TOOLCHAIN_CFLAGS += -D__STDC_LIMIT_MACROS +WERROR_FLAGS := -W -Wall -Werror +WERROR_FLAGS += -Wmissing-declarations -Wpointer-arith +WERROR_FLAGS += -Wcast-align -Wcast-qual +else WERROR_FLAGS := -W -Wall -Werror -Wstrict-prototypes -Wmissing-prototypes WERROR_FLAGS += -Wmissing-declarations -Wold-style-definition -Wpointer-arith WERROR_FLAGS += -Wcast-align -Wnested-externs -Wcast-qual +endif WERROR_FLAGS += -Wformat-nonliteral -Wformat-security ifeq ($(CONFIG_RTE_EXEC_ENV),"linuxapp")
[dpdk-dev] Any benefit of using DPDK's makefiles instead of using your own and linking against DPDK library
I already have existing makefiles for my current application. I would like to integrate dpdk into the application. ?m wondering if there is any benefit to use dpdk?s makefiles instead of using your own makefile and linking against the library (e.g. libintel_dpdk.a). Thanks. Dan
