[dpdk-dev] [PATCH v6 3/3] ixgbe: Add LRO support
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Vlad Zolotarov
> Sent: Monday, March 16, 2015 6:27 PM
> To: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v6 3/3] ixgbe: Add LRO support
>
>
>
> On 03/09/15 21:07, Vlad Zolotarov wrote:
> > - Only x540 and 82599 devices support LRO.
> > - Add the appropriate HW configuration.
> > - Add RSC aware rx_pkt_burst() handlers:
> > - Implemented bulk allocation and non-bulk allocation versions.
> > - Add LRO-specific fields to rte_eth_rxmode, to rte_eth_dev_data
> > and to igb_rx_queue.
> > - Use the appropriate handler when LRO is requested.
> >
> > Signed-off-by: Vlad Zolotarov
> > ---
> > New in v5:
> > - Put the RTE_ETHDEV_HAS_LRO_SUPPORT definition at the beginning of
> > rte_ethdev.h.
> > - Removed the "TODO: Remove me" comment near RTE_ETHDEV_HAS_LRO_SUPPORT.
> >
> > New in v4:
> > - Define RTE_ETHDEV_HAS_LRO_SUPPORT in rte_ethdev.h instead of
> > RTE_ETHDEV_LRO_SUPPORT defined in config/common_linuxapp.
> >
> > New in v2:
> > - Removed rte_eth_dev_data.lro_bulk_alloc.
> > - Fixed a few styling and spelling issues.
> > ---
> > lib/librte_ether/rte_ethdev.h | 9 +-
> > lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 6 +
> > lib/librte_pmd_ixgbe/ixgbe_ethdev.h | 5 +
> > lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 562
> > +++-
> > lib/librte_pmd_ixgbe/ixgbe_rxtx.h | 6 +
> > 5 files changed, 581 insertions(+), 7 deletions(-)
> >
> > diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
> > index 8db3127..44f081f 100644
> > --- a/lib/librte_ether/rte_ethdev.h
> > +++ b/lib/librte_ether/rte_ethdev.h
> > @@ -172,6 +172,9 @@ extern "C" {
> >
> > #include
> >
> > +/* Use this macro to check if LRO API is supported */
> > +#define RTE_ETHDEV_HAS_LRO_SUPPORT
> > +
> > #include
> > #include
> > #include
> > @@ -320,14 +323,15 @@ struct rte_eth_rxmode {
> > enum rte_eth_rx_mq_mode mq_mode;
> > uint32_t max_rx_pkt_len; /**< Only used if jumbo_frame enabled. */
> > uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
> > - uint8_t header_split : 1, /**< Header Split enable. */
> > + uint16_t header_split : 1, /**< Header Split enable. */
> > hw_ip_checksum : 1, /**< IP/UDP/TCP checksum offload enable.
> > */
> > hw_vlan_filter : 1, /**< VLAN filter enable. */
> > hw_vlan_strip: 1, /**< VLAN strip enable. */
> > hw_vlan_extend : 1, /**< Extended VLAN enable. */
> > jumbo_frame : 1, /**< Jumbo Frame Receipt enable. */
> > hw_strip_crc : 1, /**< Enable CRC stripping by hardware. */
> > - enable_scatter : 1; /**< Enable scatter packets rx handler */
> > + enable_scatter : 1, /**< Enable scatter packets rx handler */
> > + enable_lro : 1; /**< Enable LRO */
> > };
> >
> > /**
> > @@ -1515,6 +1519,7 @@ struct rte_eth_dev_data {
> > uint8_t port_id; /**< Device [external] port identifier. */
> > uint8_t promiscuous : 1, /**< RX promiscuous mode ON(1) / OFF(0). */
> > scattered_rx : 1, /**< RX of scattered packets is ON(1) /
> > OFF(0) */
> > + lro : 1, /**< RX LRO is ON(1) / OFF(0) */
> > all_multicast : 1, /**< RX all multicast mode ON(1) / OFF(0). */
> > dev_started : 1; /**< Device state: STARTED(1) / STOPPED(0).
> > */
> > };
> > diff --git a/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
> > b/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
> > index 9d3de1a..765174d 100644
> > --- a/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
> > +++ b/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
> > @@ -1648,6 +1648,7 @@ ixgbe_dev_stop(struct rte_eth_dev *dev)
> >
> > /* Clear stored conf */
> > dev->data->scattered_rx = 0;
> > + dev->data->lro = 0;
> > hw->rx_bulk_alloc_allowed = false;
> > hw->rx_vec_allowed = false;
> >
> > @@ -2018,6 +2019,11 @@ ixgbe_dev_info_get(struct rte_eth_dev *dev, struct
> > rte_eth_dev_info *dev_info)
> > DEV_RX_OFFLOAD_IPV4_CKSUM |
> > DEV_RX_OFFLOAD_UDP_CKSUM |
> > DEV_RX_OFFLOAD_TCP_CKSUM;
> > +
> > + if (hw->mac.type == ixgbe_mac_82599EB ||
> > + hw->mac.type == ixgbe_mac_X540)
> > + dev_info->rx_offload_capa |= DEV_RX_OFFLOAD_TCP_LRO;
> > +
> > dev_info->tx_offload_capa =
> > DEV_TX_OFFLOAD_VLAN_INSERT |
> > DEV_TX_OFFLOAD_IPV4_CKSUM |
> > diff --git a/lib/librte_pmd_ixgbe/ixgbe_ethdev.h
> > b/lib/librte_pmd_ixgbe/ixgbe_ethdev.h
> > index a549f5c..e206584 100644
> > --- a/lib/librte_pmd_ixgbe/ixgbe_ethdev.h
> > +++ b/lib/librte_pmd_ixgbe/ixgbe_ethdev.h
> > @@ -349,6 +349,11 @@ uint16_t ixgbe_recv_pkts_bulk_alloc(void *rx_queue,
> > struct rte_mbuf **rx_pkts,
> > uint16_t ixgbe_recv_scattered_pkts(void *rx_queue,
> > struct rte_mbuf *
[dpdk-dev] [PATCH] common/rte_memcpy: Fix x86intrin.h missed
On 3/13/2015 5:45 PM, Ananyev, Konstantin wrote:
> Hi Michael,
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Michael Qiu
>> Sent: Friday, March 13, 2015 7:03 AM
>> To: dev at dpdk.org
>> Subject: [dpdk-dev] [PATCH] common/rte_memcpy: Fix x86intrin.h missed
>>
>> rte_memcpy.h(46): catastrophic error: cannot open source file "x86intrin.h"
>>
>> For icc and old gcc, this header is not included.
>>
>> Signed-off-by: Michael Qiu
>> ---
>> lib/librte_eal/common/include/arch/x86/rte_memcpy.h | 20
>>
>> 1 file changed, 20 insertions(+)
>>
>> diff --git a/lib/librte_eal/common/include/arch/x86/rte_memcpy.h
>> b/lib/librte_eal/common/include/arch/x86/rte_memcpy.h
>> index ac72069..bd10d36 100644
>> --- a/lib/librte_eal/common/include/arch/x86/rte_memcpy.h
>> +++ b/lib/librte_eal/common/include/arch/x86/rte_memcpy.h
>> @@ -43,7 +43,27 @@
>> #include
>> #include
>> #include
>> +#if (defined(__ICC) || (__GNUC__ == 4 && __GNUC_MINOR__ < 4))
>> +
>> +#ifdef __SSE__
>> +#include
>> +#endif
>> +
>> +#ifdef __SSE2__
>> +#include
>> +#endif
>> +
>> +#if defined(__SSE4_2__) || defined(__SSE4_1__)
>> +#include
>> +#endif
>> +
>> +#if defined(__AVX__)
>> +#include
>> +#endif
>> +
>> +#else
>> #include
>> +#endif
>>
>> #ifdef __cplusplus
>> extern "C" {
>> --
>> 1.9.3
> Wonder why to spread this thing over?
> Why not just #include ?
Yes, I will send other patch to fix this,
Thanks,
Michael
> Konstantin
>
>
>
>
[dpdk-dev] [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added + unittest
Hi, Konstantin,
>Shouldn't the line above be inside if (head != NULL) {...} block?
This is removed as Olivier commented before:
>> +{
> + if (likely(head != NULL)) {
>I think we should remove this test. The other mbuf functions do not
>check this.
Regards,
Vadim.
On Wed, Mar 18, 2015 at 1:46 AM, Ananyev, Konstantin <
konstantin.ananyev at intel.com> wrote:
> Hi Vadim,
>
> > -Original Message-
> > From: vadim.suraev at gmail.com [mailto:vadim.suraev at gmail.com]
> > Sent: Tuesday, March 17, 2015 9:36 PM
> > To: dev at dpdk.org
> > Cc: olivier.matz at 6wind.com; stephen at networkplumber.org; Ananyev,
> Konstantin; vadim.suraev at gmail.com
> > Subject: [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added +
> unittest
> >
> > From: "vadim.suraev at gmail.com"
> >
> > This patch adds mbuf bulk allocation/freeing functions and unittest
> >
> > Signed-off-by: Vadim Suraev
> >
> > ---
> > New in v2:
> > - function rte_pktmbuf_alloc_bulk added
> > - function rte_pktmbuf_bulk_free added
> > - function rte_pktmbuf_free_chain added
> > - applied reviewers' comments
> >
> > app/test/test_mbuf.c | 94
> +++-
> > lib/librte_mbuf/rte_mbuf.h | 89
> +
> > 2 files changed, 182 insertions(+), 1 deletion(-)
> >
> > diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c
> > index 1ff66cb..b20c6a4 100644
> > --- a/app/test/test_mbuf.c
> > +++ b/app/test/test_mbuf.c
> > @@ -77,6 +77,7 @@
> > #define REFCNT_RING_SIZE(REFCNT_MBUF_NUM * REFCNT_MAX_REF)
> >
> > #define MAKE_STRING(x) # x
> > +#define MBUF_POOL_LOCAL_CACHE_SIZE 32
> >
> > static struct rte_mempool *pktmbuf_pool = NULL;
> >
> > @@ -405,6 +406,84 @@ test_pktmbuf_pool(void)
> > return ret;
> > }
> >
> > +/* test pktmbuf bulk allocation and freeing
> > +*/
> > +static int
> > +test_pktmbuf_pool_bulk(void)
> > +{
> > + unsigned i;
> > + /* size of mempool - size of local cache, otherwise may fail */
> > + unsigned mbufs_to_allocate = NB_MBUF - MBUF_POOL_LOCAL_CACHE_SIZE;
> > + struct rte_mbuf *m[mbufs_to_allocate];
> > + int ret = 0;
> > + unsigned mbuf_count_before_allocation =
> rte_mempool_count(pktmbuf_pool);
> > +
> > + for (i = 0; i < mbufs_to_allocate; i++)
> > + m[i] = NULL;
> > + /* alloc NB_MBUF-MBUF_POOL_LOCAL_CACHE_SIZE mbufs */
> > + ret = rte_pktmbuf_alloc_bulk(pktmbuf_pool, m, mbufs_to_allocate);
> > + if (ret) {
> > + printf("cannot allocate %d mbufs bulk mempool_cnt=%d
> ret=%d\n",
> > + mbufs_to_allocate,
> > + rte_mempool_count(pktmbuf_pool),
> > + ret);
> > + return -1;
> > + }
> > + if ((rte_mempool_count(pktmbuf_pool) + mbufs_to_allocate) !=
> > + mbuf_count_before_allocation) {
> > + printf("mempool count %d + allocated %d != initial %d\n",
> > + rte_mempool_count(pktmbuf_pool),
> > + mbufs_to_allocate,
> > + mbuf_count_before_allocation);
> > + return -1;
> > + }
> > + /* free them */
> > + rte_pktmbuf_bulk_free(m, mbufs_to_allocate);
> > +
> > + if (rte_mempool_count(pktmbuf_pool) !=
> mbuf_count_before_allocation) {
> > + printf("mempool count %d != initial %d\n",
> > + rte_mempool_count(pktmbuf_pool),
> > + mbuf_count_before_allocation);
> > + return -1;
> > + }
> > + for (i = 0; i < mbufs_to_allocate; i++)
> > + m[i] = NULL;
> > +
> > + /* alloc NB_MBUF-MBUF_POOL_LOCAL_CACHE_SIZE mbufs */
> > + ret = rte_pktmbuf_alloc_bulk(pktmbuf_pool, m, mbufs_to_allocate);
> > + if (ret) {
> > + printf("cannot allocate %d mbufs bulk mempool_cnt=%d
> ret=%d\n",
> > + mbufs_to_allocate,
> > + rte_mempool_count(pktmbuf_pool),
> > + ret);
> > + return -1;
> > + }
> > + if ((rte_mempool_count(pktmbuf_pool) + mbufs_to_allocate) !=
> > + mbuf_count_before_allocation) {
> > + printf("mempool count %d + allocated %d != initial %d\n",
> > + rte_mempool_count(pktmbuf_pool),
> > + mbufs_to_allocate,
> > + mbuf_count_before_allocation);
> > + return -1;
> > + }
> > +
> > + /* chain it */
> > + for (i = 0; i < mbufs_to_allocate - 1; i++) {
> > + m[i]->next = m[i + 1];
> > + m[0]->nb_segs++;
> > + }
> > + /* free them */
> > + rte_pktmbuf_free_chain(m[0]);
> > +
> > + if (rte_mempool_count(pktmbuf_pool) !=
> mbuf_count_before_allocation) {
> > + printf("mempool count %d != initial %d\n",
> > + rte_mempool_count(pktmbuf_pool),
> > +
[dpdk-dev] [PATCH] fix build warning and failure in Suse11
Suse11 SP3 default gcc version is 4.3.4, some options not support on this
version.
error: implicit declaration of function ?_mm_alignr_epi8?
solution: include tmmintrin.h when enable SSE3
error: unrecognized command line option "-Wno-unused-but-set-variable"
solution: add version check in fm10k Makefile
error: enic_main.c:845: error: initialized field overwritten
solution: change struct initialization code
error: ?testfn_pci_cmd? defined but not used
solution: add __attribute__((unused)) before function definition
error: unrecognized command line option "-fno-var-tracking-assignments"
solution: add version check in app/test/Makefile
error: implicit declaration of function ?pread?
solution: add _GNU_SOURCE flag when compile eal_pci_uio and eal_interrupts
signed-off-by: Marvin Liu
diff --git a/app/test/Makefile b/app/test/Makefile
index 9f0262c..4aca77c 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -152,9 +152,11 @@ CFLAGS += -D_GNU_SOURCE
# Disable VTA for memcpy test
ifeq ($(CC), gcc)
+ifeq ($(shell test $(GCC_VERSION) -ge 44 && echo 1), 1)
CFLAGS_test_memcpy.o += -fno-var-tracking-assignments
CFLAGS_test_memcpy_perf.o += -fno-var-tracking-assignments
endif
+endif
# this application needs libraries first
DEPDIRS-y += lib
diff --git a/app/test/test.h b/app/test/test.h
index 5450986..13f6592 100644
--- a/app/test/test.h
+++ b/app/test/test.h
@@ -169,7 +169,7 @@ struct test_command {
void add_test_command(struct test_command *t);
#define REGISTER_TEST_COMMAND(t) \
-static void testfn_##t(void);\
+static void __attribute__((unused))testfn_##t(void);\
void __attribute__((constructor, used)) testfn_##t(void)\
{\
add_test_command(&t);\
diff --git a/lib/librte_eal/common/include/rte_common_vect.h
b/lib/librte_eal/common/include/rte_common_vect.h
index 54ec70f..df3dce4 100644
--- a/lib/librte_eal/common/include/rte_common_vect.h
+++ b/lib/librte_eal/common/include/rte_common_vect.h
@@ -50,6 +50,10 @@
#include
#endif
+#ifdef __SSE3__
+#include
+#endif
+
#if defined(__SSE4_2__) || defined(__SSE4_1__)
#include
#endif
diff --git a/lib/librte_eal/linuxapp/eal/Makefile
b/lib/librte_eal/linuxapp/eal/Makefile
index 23c2d48..21875b8 100644
--- a/lib/librte_eal/linuxapp/eal/Makefile
+++ b/lib/librte_eal/linuxapp/eal/Makefile
@@ -102,6 +102,8 @@ CFLAGS_eal_pci_vfio.o := -D_GNU_SOURCE
CFLAGS_eal_common_whitelist.o := -D_GNU_SOURCE
CFLAGS_eal_common_options.o := -D_GNU_SOURCE
CFLAGS_eal_common_thread.o := -D_GNU_SOURCE
+CFLAGS_eal_pci_uio.o := -D_GNU_SOURCE
+CFLAGS_eal_interrupts.o := -D_GNU_SOURCE
# workaround for a gcc bug with noreturn attribute
# http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12603
diff --git a/lib/librte_pmd_enic/enic_main.c b/lib/librte_pmd_enic/enic_main.c
index c66f139..0892b3e 100644
--- a/lib/librte_pmd_enic/enic_main.c
+++ b/lib/librte_pmd_enic/enic_main.c
@@ -840,10 +840,12 @@ static int enic_set_rsskey(struct enic *enic)
dma_addr_t rss_key_buf_pa;
union vnic_rss_key *rss_key_buf_va = NULL;
static union vnic_rss_key rss_key = {
- .key[0] = {.b = {85, 67, 83, 97, 119, 101, 115, 111, 109, 101}},
- .key[1] = {.b = {80, 65, 76, 79, 117, 110, 105, 113, 117, 101}},
- .key[2] = {.b = {76, 73, 78, 85, 88, 114, 111, 99, 107, 115}},
- .key[3] = {.b = {69, 78, 73, 67, 105, 115, 99, 111, 111, 108}},
+ .key = {
+ [0] = {.b = {85, 67, 83, 97, 119, 101, 115, 111, 109,
101}},
+ [1] = {.b = {80, 65, 76, 79, 117, 110, 105, 113, 117,
101}},
+ [2] = {.b = {76, 73, 78, 85, 88, 114, 111, 99, 107,
115}},
+ [3] = {.b = {69, 78, 73, 67, 105, 115, 99, 111, 111,
108}},
+ }
};
int err;
u8 name[NAME_MAX];
diff --git a/lib/librte_pmd_fm10k/Makefile b/lib/librte_pmd_fm10k/Makefile
index 998bf23..52fc315 100644
--- a/lib/librte_pmd_fm10k/Makefile
+++ b/lib/librte_pmd_fm10k/Makefile
@@ -62,13 +62,18 @@ else
#
# CFLAGS for gcc
#
-ifeq ($(shell test $(GCC_VERSION) -ge 44 && echo 1), 1)
-CFLAGS += -Wno-deprecated
-endif
CFLAGS_BASE_DRIVER = -Wno-unused-parameter -Wno-unused-value
CFLAGS_BASE_DRIVER += -Wno-strict-aliasing -Wno-format-extra-args
-CFLAGS_BASE_DRIVER += -Wno-unused-variable -Wno-unused-but-set-variable
+CFLAGS_BASE_DRIVER += -Wno-unused-variable
CFLAGS_BASE_DRIVER += -Wno-missing-field-initializers
+
+ifeq ($(shell test $(GCC_VERSION) -ge 44 && echo 1), 1)
+CFLAGS += -Wno-deprecated
+endif
+
+ifeq ($(shell test $(GCC_VERSION) -ge 46 && echo 1), 1)
+CFLAGS_BASE_DRIVER += -Wno-unused-but-set-variable
+endif
endif
#
--
1.9.3
[dpdk-dev] Undefined reference to FUSE
>> I have done what you said but it does not work. I did it in the >> server I usually use and in a VM, in both of them appears the error. >> To be clear, I have followed the next instructions: >> >> apt-get install fuse >> apt-get install libfuse-dev >> download dpdk-1.8.0 >> modify common_linuxapp: >> +CONFIG_RTE_BUILD_COMBINE_LIBS=y >> +CONFIG_RTE_LIBRTE_VHOST=y >> make config T=x86_64-native-linuxapp-gcc make install >> T=x86_64-native-linuxapp-gcc >> >> git clone ovs >> apply next patch: >> http://openvswitch.org/pipermail/dev/2015-March/052061.html >> ./boot.sh >> ./configure --with-dpdk=$DPDK_BUILD >> make >> >> Is anything wrong with it? > > I don't see anything obvious that's wrong. I've just checked it now on my own > system with the commit I rebased against and it's compiling ok. > I'm on F20 with 3.16. The part of the patch that adds the fuse library is > below, so you could check to make sure it has applied ok. > > diff --git a/lib/automake.mk b/lib/automake.mk index 2acfe18..594dec4 100644 > --- a/lib/automake.mk > +++ b/lib/automake.mk > @@ -346,6 +346,7 @@ lib_libopenvswitch_la_SOURCES += \ endif > > if DPDK_NETDEV > +lib_libopenvswitch_la_LDFLAGS += -lfuse > > > After that I'm not really sure where to go. I'll try it on the head of master > when I'm back in the office on Wednesday. I can confirm this problem, at least on Ubuntu (14.10). I believe it is due to the fact that dependencies on shared objects (-lfuse) are not captured in .la files. Hence when executables are linked with libopenvswitch.a, the -lfuse link flag doesn't get added automatically. I've seen this on another project but I'm not sure what is the proper solution. The difference with other shared objects (e.g. -ldl) is that those get added during configure step and land into $(LIBS) that then gets widely applied. The easy workaround though is to add -lfuse to LIBS at ovs make invocation time: make LIBS=-lfuse
[dpdk-dev] [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added + unittest
Hi Vadim,
> From: Vadim Suraev [mailto:vadim.suraev at gmail.com]
> Sent: Wednesday, March 18, 2015 5:19 AM
> To: Ananyev, Konstantin
> Cc: dev at dpdk.org; olivier.matz at 6wind.com; stephen at networkplumber.org
> Subject: Re: [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added +
> unittest
>
> Hi, Konstantin,
>
> >Shouldn't the line above be inside if (head != NULL) {...} block?
> This is removed as Olivier commented before:
>
> >> +{
> > +? ? ?if (likely(head != NULL)) {
>
> >I think we should remove this test. The other mbuf functions do not
> >check this.
> Regards,
> ?Vadim.
I meant that in my opinion it should be:
while (head) {
next = head->next;
- head->next = NULL;
head = __rte_pktmbuf_prefree_seg(head);
if (likely(head != NULL)) {
+ head->next = NULL;
RTE_MBUF_ASSERT(rte_mbuf_refcnt_read(head) == 0);
Same as rte_pktmbuf_free() doing.
Konstantin
>
> On Wed, Mar 18, 2015 at 1:46 AM, Ananyev, Konstantin intel.com> wrote:
> Hi Vadim,
>
> > -Original Message-
> > From: vadim.suraev at gmail.com [mailto:vadim.suraev at gmail.com]
> > Sent: Tuesday, March 17, 2015 9:36 PM
> > To: dev at dpdk.org
> > Cc: olivier.matz at 6wind.com; stephen at networkplumber.org; Ananyev,
> > Konstantin; vadim.suraev at gmail.com
> > Subject: [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added +
> > unittest
> >
> > From: "vadim.suraev at gmail.com"
> >
> > This patch adds mbuf bulk allocation/freeing functions and unittest
> >
> > Signed-off-by: Vadim Suraev
> >
> > ---
> > New in v2:
> >? ? ?- function rte_pktmbuf_alloc_bulk added
> >? ? ?- function rte_pktmbuf_bulk_free added
> >? ? ?- function rte_pktmbuf_free_chain added
> >? ? ?- applied reviewers' comments
> >
> >? app/test/test_mbuf.c? ? ? ?|? ?94
> >+++-
> >? lib/librte_mbuf/rte_mbuf.h |? ?89 +
> >? 2 files changed, 182 insertions(+), 1 deletion(-)
> >
> > diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c
> > index 1ff66cb..b20c6a4 100644
> > --- a/app/test/test_mbuf.c
> > +++ b/app/test/test_mbuf.c
> > @@ -77,6 +77,7 @@
> >? #define REFCNT_RING_SIZE? ? ? ? (REFCNT_MBUF_NUM * REFCNT_MAX_REF)
> >
> >? #define MAKE_STRING(x)? ? ? ? ? # x
> > +#define MBUF_POOL_LOCAL_CACHE_SIZE 32
> >
> >? static struct rte_mempool *pktmbuf_pool = NULL;
> >
> > @@ -405,6 +406,84 @@ test_pktmbuf_pool(void)
> >? ? ? ?return ret;
> >? }
> >
> > +/* test pktmbuf bulk allocation and freeing
> > +*/
> > +static int
> > +test_pktmbuf_pool_bulk(void)
> > +{
> > +? ? ?unsigned i;
> > +? ? ?/* size of mempool - size of local cache, otherwise may fail */
> > +? ? ?unsigned mbufs_to_allocate = NB_MBUF - MBUF_POOL_LOCAL_CACHE_SIZE;
> > +? ? ?struct rte_mbuf *m[mbufs_to_allocate];
> > +? ? ?int ret = 0;
> > +? ? ?unsigned mbuf_count_before_allocation =
> > rte_mempool_count(pktmbuf_pool);
> > +
> > +? ? ?for (i = 0; i < mbufs_to_allocate; i++)
> > +? ? ? ? ? ? ?m[i] = NULL;
> > +? ? ?/* alloc NB_MBUF-MBUF_POOL_LOCAL_CACHE_SIZE mbufs */
> > +? ? ?ret = rte_pktmbuf_alloc_bulk(pktmbuf_pool, m, mbufs_to_allocate);
> > +? ? ?if (ret) {
> > +? ? ? ? ? ? ?printf("cannot allocate %d mbufs bulk mempool_cnt=%d
> > ret=%d\n",
> > +? ? ? ? ? ? ? ? ? ? ?mbufs_to_allocate,
> > +? ? ? ? ? ? ? ? ? ? ?rte_mempool_count(pktmbuf_pool),
> > +? ? ? ? ? ? ? ? ? ? ?ret);
> > +? ? ? ? ? ? ?return -1;
> > +? ? ?}
> > +? ? ?if ((rte_mempool_count(pktmbuf_pool) + mbufs_to_allocate) !=
> > +? ? ? ? ?mbuf_count_before_allocation) {
> > +? ? ? ? ? ? ?printf("mempool count %d + allocated %d != initial %d\n",
> > +? ? ? ? ? ? ? ? ? ? ?rte_mempool_count(pktmbuf_pool),
> > +? ? ? ? ? ? ? ? ? ? ?mbufs_to_allocate,
> > +? ? ? ? ? ? ? ? ? ? ?mbuf_count_before_allocation);
> > +? ? ? ? ? ? ?return -1;
> > +? ? ?}
> > +? ? ?/* free them */
> > +? ? ?rte_pktmbuf_bulk_free(m, mbufs_to_allocate);
> > +
> > +? ? ?if (rte_mempool_count(pktmbuf_pool)? != mbuf_count_before_allocation)
> > {
> > +? ? ? ? ? ? ?printf("mempool count %d != initial %d\n",
> > +? ? ? ? ? ? ? ? ? ? ?rte_mempool_count(pktmbuf_pool),
> > +? ? ? ? ? ? ? ? ? ? ?mbuf_count_before_allocation);
> > +? ? ? ? ? ? ?return -1;
> > +? ? ?}
> > +? ? ?for (i = 0; i < mbufs_to_allocate; i++)
> > +? ? ? ? ? ? ?m[i] = NULL;
> > +
> > +? ? ?/* alloc NB_MBUF-MBUF_POOL_LOCAL_CACHE_SIZE mbufs */
> > +? ? ?ret = rte_pktmbuf_alloc_bulk(pktmbuf_pool, m, mbufs_to_allocate);
> > +? ? ?if (ret) {
> > +? ? ? ? ? ? ?printf("cannot allocate %d mbufs bulk mempool_cnt=%d
> > ret=%d\n",
> > +? ? ? ? ? ? ? ? ? ? ?mbufs_to_allocate,
> > +? ? ? ? ? ? ? ? ? ? ?rte_mempool_count(pktmbuf_pool),
> > +? ? ? ? ? ? ? ? ? ? ?ret);
> > +? ? ? ? ? ? ?return -1;
> > +? ? ?}
> > +? ? ?if ((rte_mempool_count(pktmbuf_pool) + mbufs_to_allocate) !=
> > +? ? ? ? ?mbuf_count_before_allocation) {
> > +? ? ? ? ? ? ?printf("mempool count %d + allocated %d != initial %d\n",
> > +? ? ? ? ? ? ?
[dpdk-dev] [PATCH] i40e: revert internal switch of PF
VEB switching is blocking VF.
If the source mac address of packet sent from VF is not listed in the
VEB?s mac table, the VEB will switch the packet back to the VF.
It's an hardware issue.
Reverts: 2ccabd8cd1f6 ("i40e: enable internal switch of PF").
Reported-by: Jingjing Wu
Signed-off-by: Thomas Monjalon
---
lib/librte_pmd_i40e/i40e_ethdev.c | 36
1 file changed, 36 deletions(-)
diff --git a/lib/librte_pmd_i40e/i40e_ethdev.c
b/lib/librte_pmd_i40e/i40e_ethdev.c
index 6888072..cf6685e 100644
--- a/lib/librte_pmd_i40e/i40e_ethdev.c
+++ b/lib/librte_pmd_i40e/i40e_ethdev.c
@@ -2868,40 +2868,6 @@ i40e_vsi_dump_bw_config(struct i40e_vsi *vsi)
return 0;
}
-/*
- * i40e_enable_pf_lb
- * @pf: pointer to the pf structure
- *
- * allow loopback on pf
- */
-static inline void
-i40e_enable_pf_lb(struct i40e_pf *pf)
-{
- struct i40e_hw *hw = I40E_PF_TO_HW(pf);
- struct i40e_vsi_context ctxt;
- int ret;
-
- memset(&ctxt, 0, sizeof(ctxt));
- ctxt.seid = pf->main_vsi_seid;
- ctxt.pf_num = hw->pf_id;
- ret = i40e_aq_get_vsi_params(hw, &ctxt, NULL);
- if (ret) {
- PMD_DRV_LOG(ERR, "couldn't get pf vsi config, err %d, aq_err
%d",
- ret, hw->aq.asq_last_status);
- return;
- }
- ctxt.flags = I40E_AQ_VSI_TYPE_PF;
- ctxt.info.valid_sections =
- rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);
- ctxt.info.switch_id |=
- rte_cpu_to_le_16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
-
- ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
- if (ret)
- PMD_DRV_LOG(ERR, "update vsi switch failed, aq_err=%d\n",
- hw->aq.asq_last_status);
-}
-
/* Setup a VSI */
struct i40e_vsi *
i40e_vsi_setup(struct i40e_pf *pf,
@@ -2937,8 +2903,6 @@ i40e_vsi_setup(struct i40e_pf *pf,
PMD_DRV_LOG(ERR, "VEB setup failed");
return NULL;
}
- /* set ALLOWLOOPBACk on pf, when veb is created */
- i40e_enable_pf_lb(pf);
}
vsi = rte_zmalloc("i40e_vsi", sizeof(struct i40e_vsi), 0);
--
2.2.2
[dpdk-dev] [PATCH v2] doc: add l2fwd-jobstats user guide
Signed-off-by: Pawel Wodkowski --- Changes v2 1. Fix trailing spaces and typos. 2. Add maintaners claim MAINTAINERS | 1 + doc/guides/sample_app_ug/index.rst| 1 + doc/guides/sample_app_ug/l2_forward_job_stats.rst | 637 ++ 3 files changed, 639 insertions(+) create mode 100644 doc/guides/sample_app_ug/l2_forward_job_stats.rst diff --git a/MAINTAINERS b/MAINTAINERS index 07fdf5e..a82e2f0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -391,6 +391,7 @@ Job statistics M: Pawel Wodkowski F: lib/librte_jobstats/ F: examples/l2fwd-jobstats/ +F: doc/guides/sample_app_ug/l2_forward_job_stats.rst Test Applications diff --git a/doc/guides/sample_app_ug/index.rst b/doc/guides/sample_app_ug/index.rst index 5720181..c89a2f0 100644 --- a/doc/guides/sample_app_ug/index.rst +++ b/doc/guides/sample_app_ug/index.rst @@ -47,6 +47,7 @@ Sample Applications User Guide ipv4_multicast ip_reassembly kernel_nic_interface +l2_forward_job_stats l2_forward_real_virtual l3_forward l3_forward_power_man diff --git a/doc/guides/sample_app_ug/l2_forward_job_stats.rst b/doc/guides/sample_app_ug/l2_forward_job_stats.rst new file mode 100644 index 000..e25d7b1 --- /dev/null +++ b/doc/guides/sample_app_ug/l2_forward_job_stats.rst @@ -0,0 +1,637 @@ +.. BSD LICENSE +Copyright(c) 2010-2015 Intel Corporation. All rights reserved. +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions +are met: + +* Redistributions of source code must retain the above copyright +notice, this list of conditions and the following disclaimer. +* Redistributions in binary form must reproduce the above copyright +notice, this list of conditions and the following disclaimer in +the documentation and/or other materials provided with the +distribution. +* Neither the name of Intel Corporation nor the names of its +contributors may be used to endorse or promote products derived +from this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +L2 Forwarding Sample Application (in Real and Virtualized Environments) with core load statistics. +== + +The L2 Forwarding sample application is a simple example of packet processing using +the Data Plane Development Kit (DPDK) which +also takes advantage of Single Root I/O Virtualization (SR-IOV) features in a virtualized environment. + +.. note:: + +This application is a variation of L2 Forwarding sample application. It demonstrate possible +scheme of job stats library usage therefore some parts of this document is identical with original +L2 forwarding application. + +Overview + + +The L2 Forwarding sample application, which can operate in real and virtualized environments, +performs L2 forwarding for each packet that is received. +The destination port is the adjacent port from the enabled portmask, that is, +if the first four ports are enabled (portmask 0xf), +ports 1 and 2 forward into each other, and ports 3 and 4 forward into each other. +Also, the MAC addresses are affected as follows: + +* The source MAC address is replaced by the TX port MAC address + +* The destination MAC address is replaced by 02:00:00:00:00:TX_PORT_ID + +This application can be used to benchmark performance using a traffic-generator, as shown in the Figure 3. + +The application can also be used in a virtualized environment as shown in Figure 4. + +The L2 Forwarding application can also be used as a starting point for developing a new application based on the DPDK. + +.. _figure_3: + +**Figure 3. Performance Benchmark Setup (Basic Environment)** + +.. image4_png has been replaced + +|l2_fwd_benchmark_setup| + +.. _figure_4: + +**Figure 4. Performance Benchmark Setup (Virtualized Environment)** + +.. image5_png has been renamed + +|l2_fwd_virtenv_benchmark_setup| + +Virtual Function Setup Instructions +~~~ + +
[dpdk-dev] [PATCH v6 3/3] ixgbe: Add LRO support
On 03/18/15 02:31, Ananyev, Konstantin wrote:
>
>> -Original Message-
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Vlad Zolotarov
>> Sent: Monday, March 16, 2015 6:27 PM
>> To: dev at dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH v6 3/3] ixgbe: Add LRO support
>>
>>
>>
>> On 03/09/15 21:07, Vlad Zolotarov wrote:
>>> - Only x540 and 82599 devices support LRO.
>>> - Add the appropriate HW configuration.
>>> - Add RSC aware rx_pkt_burst() handlers:
>>> - Implemented bulk allocation and non-bulk allocation versions.
>>> - Add LRO-specific fields to rte_eth_rxmode, to rte_eth_dev_data
>>>and to igb_rx_queue.
>>> - Use the appropriate handler when LRO is requested.
>>>
>>> Signed-off-by: Vlad Zolotarov
>>> ---
>>> New in v5:
>>> - Put the RTE_ETHDEV_HAS_LRO_SUPPORT definition at the beginning of
>>> rte_ethdev.h.
>>> - Removed the "TODO: Remove me" comment near
>>> RTE_ETHDEV_HAS_LRO_SUPPORT.
>>>
>>> New in v4:
>>> - Define RTE_ETHDEV_HAS_LRO_SUPPORT in rte_ethdev.h instead of
>>>RTE_ETHDEV_LRO_SUPPORT defined in config/common_linuxapp.
>>>
>>> New in v2:
>>> - Removed rte_eth_dev_data.lro_bulk_alloc.
>>> - Fixed a few styling and spelling issues.
>>> ---
>>>lib/librte_ether/rte_ethdev.h | 9 +-
>>>lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 6 +
>>>lib/librte_pmd_ixgbe/ixgbe_ethdev.h | 5 +
>>>lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 562
>>> +++-
>>>lib/librte_pmd_ixgbe/ixgbe_rxtx.h | 6 +
>>>5 files changed, 581 insertions(+), 7 deletions(-)
>>>
>>> diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
>>> index 8db3127..44f081f 100644
>>> --- a/lib/librte_ether/rte_ethdev.h
>>> +++ b/lib/librte_ether/rte_ethdev.h
>>> @@ -172,6 +172,9 @@ extern "C" {
>>>
>>>#include
>>>
>>> +/* Use this macro to check if LRO API is supported */
>>> +#define RTE_ETHDEV_HAS_LRO_SUPPORT
>>> +
>>>#include
>>>#include
>>>#include
>>> @@ -320,14 +323,15 @@ struct rte_eth_rxmode {
>>> enum rte_eth_rx_mq_mode mq_mode;
>>> uint32_t max_rx_pkt_len; /**< Only used if jumbo_frame enabled. */
>>> uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
>>> - uint8_t header_split : 1, /**< Header Split enable. */
>>> + uint16_t header_split : 1, /**< Header Split enable. */
>>> hw_ip_checksum : 1, /**< IP/UDP/TCP checksum offload enable.
>>> */
>>> hw_vlan_filter : 1, /**< VLAN filter enable. */
>>> hw_vlan_strip: 1, /**< VLAN strip enable. */
>>> hw_vlan_extend : 1, /**< Extended VLAN enable. */
>>> jumbo_frame : 1, /**< Jumbo Frame Receipt enable. */
>>> hw_strip_crc : 1, /**< Enable CRC stripping by hardware. */
>>> - enable_scatter : 1; /**< Enable scatter packets rx handler */
>>> + enable_scatter : 1, /**< Enable scatter packets rx handler */
>>> + enable_lro : 1; /**< Enable LRO */
>>>};
>>>
>>>/**
>>> @@ -1515,6 +1519,7 @@ struct rte_eth_dev_data {
>>> uint8_t port_id; /**< Device [external] port identifier. */
>>> uint8_t promiscuous : 1, /**< RX promiscuous mode ON(1) / OFF(0). */
>>> scattered_rx : 1, /**< RX of scattered packets is ON(1) /
>>> OFF(0) */
>>> + lro : 1, /**< RX LRO is ON(1) / OFF(0) */
>>> all_multicast : 1, /**< RX all multicast mode ON(1) / OFF(0). */
>>> dev_started : 1; /**< Device state: STARTED(1) / STOPPED(0).
>>> */
>>>};
>>> diff --git a/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
>>> b/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
>>> index 9d3de1a..765174d 100644
>>> --- a/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
>>> +++ b/lib/librte_pmd_ixgbe/ixgbe_ethdev.c
>>> @@ -1648,6 +1648,7 @@ ixgbe_dev_stop(struct rte_eth_dev *dev)
>>>
>>> /* Clear stored conf */
>>> dev->data->scattered_rx = 0;
>>> + dev->data->lro = 0;
>>> hw->rx_bulk_alloc_allowed = false;
>>> hw->rx_vec_allowed = false;
>>>
>>> @@ -2018,6 +2019,11 @@ ixgbe_dev_info_get(struct rte_eth_dev *dev, struct
>>> rte_eth_dev_info *dev_info)
>>> DEV_RX_OFFLOAD_IPV4_CKSUM |
>>> DEV_RX_OFFLOAD_UDP_CKSUM |
>>> DEV_RX_OFFLOAD_TCP_CKSUM;
>>> +
>>> + if (hw->mac.type == ixgbe_mac_82599EB ||
>>> + hw->mac.type == ixgbe_mac_X540)
>>> + dev_info->rx_offload_capa |= DEV_RX_OFFLOAD_TCP_LRO;
>>> +
>>> dev_info->tx_offload_capa =
>>> DEV_TX_OFFLOAD_VLAN_INSERT |
>>> DEV_TX_OFFLOAD_IPV4_CKSUM |
>>> diff --git a/lib/librte_pmd_ixgbe/ixgbe_ethdev.h
>>> b/lib/librte_pmd_ixgbe/ixgbe_ethdev.h
>>> index a549f5c..e206584 100644
>>> --- a/lib/librte_pmd_ixgbe/ixgbe_ethdev.h
>>> +++ b/lib/librte_pmd_ixgbe/ixgbe_ethdev.h
>>> @@ -349,6 +349,11 @@ uint16_t ixgbe_recv_pkts_bulk_alloc(void *rx_queue,
>>> struct rte_mbuf **rx_pkts,
>>>
[dpdk-dev] [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added + unittest
Hi, Konstantin,
Got it. To make the same, nulling the next should be inside of the block as
you said.
One question raises here: If a segment in the chain has refcnt > 1 (so its
next is not assigned NULL), and the next segment has refcnt == 1 (so it is
freed), do you think this scenario is real/should be considered? If so, the
former can be safely freed only by calling rte_pktmbuf_free_seg which does
not iterate. So why to keep next pointing to something?
Regards,
Vadim
On Wed, Mar 18, 2015 at 11:56 AM, Ananyev, Konstantin <
konstantin.ananyev at intel.com> wrote:
>
> Hi Vadim,
>
>
> > From: Vadim Suraev [mailto:vadim.suraev at gmail.com]
> > Sent: Wednesday, March 18, 2015 5:19 AM
> > To: Ananyev, Konstantin
> > Cc: dev at dpdk.org; olivier.matz at 6wind.com; stephen at
> > networkplumber.org
> > Subject: Re: [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added +
> unittest
> >
> > Hi, Konstantin,
> >
> > >Shouldn't the line above be inside if (head != NULL) {...} block?
> > This is removed as Olivier commented before:
> >
> > >> +{
> > > + if (likely(head != NULL)) {
> >
> > >I think we should remove this test. The other mbuf functions do not
> > >check this.
> > Regards,
> > Vadim.
>
> I meant that in my opinion it should be:
>
> while (head) {
> next = head->next;
> - head->next = NULL;
>
> head = __rte_pktmbuf_prefree_seg(head);
> if (likely(head != NULL)) {
> + head->next = NULL;
> RTE_MBUF_ASSERT(rte_mbuf_refcnt_read(head) == 0);
>
> Same as rte_pktmbuf_free() doing.
>
> Konstantin
>
> >
> > On Wed, Mar 18, 2015 at 1:46 AM, Ananyev, Konstantin <
> konstantin.ananyev at intel.com> wrote:
> > Hi Vadim,
> >
> > > -Original Message-
> > > From: vadim.suraev at gmail.com [mailto:vadim.suraev at gmail.com]
> > > Sent: Tuesday, March 17, 2015 9:36 PM
> > > To: dev at dpdk.org
> > > Cc: olivier.matz at 6wind.com; stephen at networkplumber.org; Ananyev,
> Konstantin; vadim.suraev at gmail.com
> > > Subject: [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added +
> unittest
> > >
> > > From: "vadim.suraev at gmail.com"
> > >
> > > This patch adds mbuf bulk allocation/freeing functions and unittest
> > >
> > > Signed-off-by: Vadim Suraev
> > >
> > > ---
> > > New in v2:
> > > - function rte_pktmbuf_alloc_bulk added
> > > - function rte_pktmbuf_bulk_free added
> > > - function rte_pktmbuf_free_chain added
> > > - applied reviewers' comments
> > >
> > > app/test/test_mbuf.c | 94
> +++-
> > > lib/librte_mbuf/rte_mbuf.h | 89
> +
> > > 2 files changed, 182 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c
> > > index 1ff66cb..b20c6a4 100644
> > > --- a/app/test/test_mbuf.c
> > > +++ b/app/test/test_mbuf.c
> > > @@ -77,6 +77,7 @@
> > > #define REFCNT_RING_SIZE(REFCNT_MBUF_NUM * REFCNT_MAX_REF)
> > >
> > > #define MAKE_STRING(x) # x
> > > +#define MBUF_POOL_LOCAL_CACHE_SIZE 32
> > >
> > > static struct rte_mempool *pktmbuf_pool = NULL;
> > >
> > > @@ -405,6 +406,84 @@ test_pktmbuf_pool(void)
> > > return ret;
> > > }
> > >
> > > +/* test pktmbuf bulk allocation and freeing
> > > +*/
> > > +static int
> > > +test_pktmbuf_pool_bulk(void)
> > > +{
> > > + unsigned i;
> > > + /* size of mempool - size of local cache, otherwise may fail */
> > > + unsigned mbufs_to_allocate = NB_MBUF -
> MBUF_POOL_LOCAL_CACHE_SIZE;
> > > + struct rte_mbuf *m[mbufs_to_allocate];
> > > + int ret = 0;
> > > + unsigned mbuf_count_before_allocation =
> rte_mempool_count(pktmbuf_pool);
> > > +
> > > + for (i = 0; i < mbufs_to_allocate; i++)
> > > + m[i] = NULL;
> > > + /* alloc NB_MBUF-MBUF_POOL_LOCAL_CACHE_SIZE mbufs */
> > > + ret = rte_pktmbuf_alloc_bulk(pktmbuf_pool, m, mbufs_to_allocate);
> > > + if (ret) {
> > > + printf("cannot allocate %d mbufs bulk mempool_cnt=%d
> ret=%d\n",
> > > + mbufs_to_allocate,
> > > + rte_mempool_count(pktmbuf_pool),
> > > + ret);
> > > + return -1;
> > > + }
> > > + if ((rte_mempool_count(pktmbuf_pool) + mbufs_to_allocate) !=
> > > + mbuf_count_before_allocation) {
> > > + printf("mempool count %d + allocated %d != initial %d\n",
> > > + rte_mempool_count(pktmbuf_pool),
> > > + mbufs_to_allocate,
> > > + mbuf_count_before_allocation);
> > > + return -1;
> > > + }
> > > + /* free them */
> > > + rte_pktmbuf_bulk_free(m, mbufs_to_allocate);
> > > +
> > > + if (rte_mempool_count(pktmbuf_pool) !=
> mbuf_count_before_allocation) {
> > > + printf("mempool count %d != initial %d\n",
> > > + rte_mempool_count(pktmbuf_pool),
> > > +
[dpdk-dev] [PATCH] fix build warning and failure in Suse11
Hi Yong, Thanks for working on these important fixes. 2015-03-18 15:10, Yong Liu: > Suse11 SP3 default gcc version is 4.3.4, some options not support on this > version. I guess some of these errors are not only specific to Suse-11? Maybe that 1 patch per issue would be easier to read and could provide a more accurate description. > error: implicit declaration of function ?_mm_alignr_epi8? > solution: include tmmintrin.h when enable SSE3 > > error: unrecognized command line option "-Wno-unused-but-set-variable" > solution: add version check in fm10k Makefile > > error: enic_main.c:845: error: initialized field overwritten > solution: change struct initialization code > > error: ?testfn_pci_cmd? defined but not used > solution: add __attribute__((unused)) before function definition Please could you explain more the problem? There are other constructors in DPDK which don't need the unused attribute. > > error: unrecognized command line option "-fno-var-tracking-assignments" > solution: add version check in app/test/Makefile > > error: implicit declaration of function ?pread? > solution: add _GNU_SOURCE flag when compile eal_pci_uio and eal_interrupts > > signed-off-by: Marvin Liu Please use -s git option to have an automatic well formatted Signed-off. Your previous contributions were signed "Yong Liu". Do you prefer Marvin Liu?
[dpdk-dev] [PATCH] EAL: move rte_common_vect.h into arch/x86
Signed-off-by: Konstantin Ananyev
---
lib/librte_eal/common/Makefile | 1 -
.../common/include/arch/x86/rte_common_vect.h | 128 +
lib/librte_eal/common/include/rte_common_vect.h| 128 -
3 files changed, 128 insertions(+), 129 deletions(-)
create mode 100644 lib/librte_eal/common/include/arch/x86/rte_common_vect.h
delete mode 100644 lib/librte_eal/common/include/rte_common_vect.h
diff --git a/lib/librte_eal/common/Makefile b/lib/librte_eal/common/Makefile
index cf961a7..3ea3bbf 100644
--- a/lib/librte_eal/common/Makefile
+++ b/lib/librte_eal/common/Makefile
@@ -39,7 +39,6 @@ INC += rte_rwlock.h rte_tailq.h rte_interrupts.h rte_alarm.h
INC += rte_string_fns.h rte_version.h
INC += rte_eal_memconfig.h rte_malloc_heap.h
INC += rte_hexdump.h rte_devargs.h rte_dev.h
-INC += rte_common_vect.h
INC += rte_pci_dev_feature_defs.h rte_pci_dev_features.h
ifeq ($(CONFIG_RTE_INSECURE_FUNCTION_WARNING),y)
diff --git a/lib/librte_eal/common/include/arch/x86/rte_common_vect.h
b/lib/librte_eal/common/include/arch/x86/rte_common_vect.h
new file mode 100644
index 000..54ec70f
--- /dev/null
+++ b/lib/librte_eal/common/include/arch/x86/rte_common_vect.h
@@ -0,0 +1,128 @@
+/*-
+ * BSD LICENSE
+ *
+ * Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * * Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef _RTE_COMMON_VECT_H_
+#define _RTE_COMMON_VECT_H_
+
+/**
+ * @file
+ *
+ * RTE SSE/AVX related header.
+ */
+
+#if (defined(__ICC) || (__GNUC__ == 4 && __GNUC_MINOR__ < 4))
+
+#ifdef __SSE__
+#include
+#endif
+
+#ifdef __SSE2__
+#include
+#endif
+
+#if defined(__SSE4_2__) || defined(__SSE4_1__)
+#include
+#endif
+
+#if defined(__AVX__)
+#include
+#endif
+
+#else
+
+#include
+
+#endif
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef __m128i xmm_t;
+
+#defineXMM_SIZE(sizeof(xmm_t))
+#defineXMM_MASK(XMM_SIZE - 1)
+
+typedef union rte_xmm {
+ xmm_tx;
+ uint8_t u8[XMM_SIZE / sizeof(uint8_t)];
+ uint16_t u16[XMM_SIZE / sizeof(uint16_t)];
+ uint32_t u32[XMM_SIZE / sizeof(uint32_t)];
+ uint64_t u64[XMM_SIZE / sizeof(uint64_t)];
+ double pd[XMM_SIZE / sizeof(double)];
+} rte_xmm_t;
+
+#ifdef __AVX__
+
+typedef __m256i ymm_t;
+
+#defineYMM_SIZE(sizeof(ymm_t))
+#defineYMM_MASK(YMM_SIZE - 1)
+
+typedef union rte_ymm {
+ ymm_ty;
+ xmm_tx[YMM_SIZE / sizeof(xmm_t)];
+ uint8_t u8[YMM_SIZE / sizeof(uint8_t)];
+ uint16_t u16[YMM_SIZE / sizeof(uint16_t)];
+ uint32_t u32[YMM_SIZE / sizeof(uint32_t)];
+ uint64_t u64[YMM_SIZE / sizeof(uint64_t)];
+ double pd[YMM_SIZE / sizeof(double)];
+} rte_ymm_t;
+
+#endif /* __AVX__ */
+
+#ifdef RTE_ARCH_I686
+#define _mm_cvtsi128_si64(a) ({ \
+ rte_xmm_t m;\
+ m.x = (a); \
+ (m.u64[0]); \
+})
+#endif
+
+/*
+ * Prior to version 12.1 icc doesn't support _mm_set_epi64x.
+ */
+#if (defined(__ICC) && __ICC < 1210)
+#define _mm_set_epi64x(a, b) ({ \
+ rte_xmm_t m; \
+ m.u64[0] = b;\
+ m.u64[1] = a;\
+ (m.x); \
+})
+#endif /* (defined(__ICC) && __ICC < 1210) */
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_COMMON__VECT_H_ */
diff --git a/lib/librte_eal/common/i
[dpdk-dev] [PATCH] igb: handle VF LPE mailbox message
This patch adds the handle function for the LPE mailbox message (VF to
PF) to set maximum packet size, which can be used to enable jumbo
frame support.
Signed-off-by: Sergio Gonzalez Monroy
---
lib/librte_pmd_e1000/igb_pf.c | 28
1 file changed, 28 insertions(+)
diff --git a/lib/librte_pmd_e1000/igb_pf.c b/lib/librte_pmd_e1000/igb_pf.c
index bc3816a..2d49379 100644
--- a/lib/librte_pmd_e1000/igb_pf.c
+++ b/lib/librte_pmd_e1000/igb_pf.c
@@ -395,6 +395,31 @@ igb_vf_set_vlan(struct rte_eth_dev *dev, uint32_t vf,
uint32_t *msgbuf)
}
static int
+igb_vf_set_rlpml(struct rte_eth_dev *dev, uint32_t vf, uint32_t *msgbuf)
+{
+ struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+ uint16_t rlpml = msgbuf[1] & E1000_VMOLR_RLPML_MASK;
+ uint32_t max_frame = rlpml + ETHER_HDR_LEN + ETHER_CRC_LEN;
+ uint32_t vmolr;
+
+ if ((max_frame < ETHER_MIN_LEN) || (max_frame >
ETHER_MAX_JUMBO_FRAME_LEN))
+ return -1;
+
+ vmolr = E1000_READ_REG(hw, E1000_VMOLR(vf));
+
+ vmolr &= ~E1000_VMOLR_RLPML_MASK;
+ vmolr |= rlpml;
+
+ /* Enable Long Packet support */
+ vmolr |= E1000_VMOLR_LPE;
+
+ E1000_WRITE_REG(hw, E1000_VMOLR(vf), vmolr);
+ E1000_WRITE_FLUSH(hw);
+
+ return 0;
+}
+
+static int
igb_rcv_msg_from_vf(struct rte_eth_dev *dev, uint16_t vf)
{
uint16_t mbx_size = E1000_VFMAILBOX_SIZE;
@@ -428,6 +453,9 @@ igb_rcv_msg_from_vf(struct rte_eth_dev *dev, uint16_t vf)
case E1000_VF_SET_MULTICAST:
retval = igb_vf_set_multicast(dev, vf, msgbuf);
break;
+ case E1000_VF_SET_LPE:
+ retval = igb_vf_set_rlpml(dev, vf, msgbuf);
+ break;
case E1000_VF_SET_VLAN:
retval = igb_vf_set_vlan(dev, vf, msgbuf);
break;
--
1.9.3
[dpdk-dev] [PATCH] igb: handle VF LPE mailbox message
On 18/03/2015 12:01, Sergio Gonzalez Monroy wrote:
> This patch adds the handle function for the LPE mailbox message (VF to
> PF) to set maximum packet size, which can be used to enable jumbo
> frame support.
>
> Signed-off-by: Sergio Gonzalez Monroy
> ---
> lib/librte_pmd_e1000/igb_pf.c | 28
> 1 file changed, 28 insertions(+)
>
> diff --git a/lib/librte_pmd_e1000/igb_pf.c b/lib/librte_pmd_e1000/igb_pf.c
> index bc3816a..2d49379 100644
> --- a/lib/librte_pmd_e1000/igb_pf.c
> +++ b/lib/librte_pmd_e1000/igb_pf.c
> @@ -395,6 +395,31 @@ igb_vf_set_vlan(struct rte_eth_dev *dev, uint32_t vf,
> uint32_t *msgbuf)
> }
>
> static int
> +igb_vf_set_rlpml(struct rte_eth_dev *dev, uint32_t vf, uint32_t *msgbuf)
> +{
> + struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
> + uint16_t rlpml = msgbuf[1] & E1000_VMOLR_RLPML_MASK;
> + uint32_t max_frame = rlpml + ETHER_HDR_LEN + ETHER_CRC_LEN;
> + uint32_t vmolr;
> +
> + if ((max_frame < ETHER_MIN_LEN) || (max_frame >
> ETHER_MAX_JUMBO_FRAME_LEN))
> + return -1;
> +
> + vmolr = E1000_READ_REG(hw, E1000_VMOLR(vf));
> +
> + vmolr &= ~E1000_VMOLR_RLPML_MASK;
> + vmolr |= rlpml;
> +
> + /* Enable Long Packet support */
> + vmolr |= E1000_VMOLR_LPE;
> +
> + E1000_WRITE_REG(hw, E1000_VMOLR(vf), vmolr);
> + E1000_WRITE_FLUSH(hw);
> +
> + return 0;
> +}
> +
> +static int
> igb_rcv_msg_from_vf(struct rte_eth_dev *dev, uint16_t vf)
> {
> uint16_t mbx_size = E1000_VFMAILBOX_SIZE;
> @@ -428,6 +453,9 @@ igb_rcv_msg_from_vf(struct rte_eth_dev *dev, uint16_t vf)
> case E1000_VF_SET_MULTICAST:
> retval = igb_vf_set_multicast(dev, vf, msgbuf);
> break;
> + case E1000_VF_SET_LPE:
> + retval = igb_vf_set_rlpml(dev, vf, msgbuf);
> + break;
> case E1000_VF_SET_VLAN:
> retval = igb_vf_set_vlan(dev, vf, msgbuf);
> break;
This patch is targeting 2.1 release.
Sergio
[dpdk-dev] [PATCH v2 1/4] mk: Remove combined library and related options
On 12/03/2015 16:27, Sergio Gonzalez Monroy wrote: > Remove CONFIG_RTE_BUILD_COMBINE_LIBS and CONFIG_RTE_LIBNAME. > > Signed-off-by: Sergio Gonzalez Monroy > --- > config/common_bsdapp| 6 -- > config/common_linuxapp | 6 -- > config/defconfig_ppc_64-power8-linuxapp-gcc | 2 - > lib/Makefile| 1 - > mk/rte.app.mk | 12 > mk/rte.lib.mk | 35 -- > mk/rte.sdkbuild.mk | 3 - > mk/rte.sharelib.mk | 101 > > mk/rte.vars.mk | 9 --- > 9 files changed, 175 deletions(-) > delete mode 100644 mk/rte.sharelib.mk > > diff --git a/config/common_bsdapp b/config/common_bsdapp > index 8ff4dc2..7ee5ecf 100644 > --- a/config/common_bsdapp > +++ b/config/common_bsdapp > @@ -79,12 +79,6 @@ CONFIG_RTE_FORCE_INTRINSICS=n > CONFIG_RTE_BUILD_SHARED_LIB=n > > # > -# Combine to one single library > -# > -CONFIG_RTE_BUILD_COMBINE_LIBS=n > -CONFIG_RTE_LIBNAME=intel_dpdk > - > -# > # Compile Environment Abstraction Layer > # > CONFIG_RTE_LIBRTE_EAL=y > diff --git a/config/common_linuxapp b/config/common_linuxapp > index 97f1c9e..ae13805 100644 > --- a/config/common_linuxapp > +++ b/config/common_linuxapp > @@ -79,12 +79,6 @@ CONFIG_RTE_FORCE_INTRINSICS=n > CONFIG_RTE_BUILD_SHARED_LIB=n > > # > -# Combine to one single library > -# > -CONFIG_RTE_BUILD_COMBINE_LIBS=n > -CONFIG_RTE_LIBNAME="intel_dpdk" > - > -# > # Compile Environment Abstraction Layer > # > CONFIG_RTE_LIBRTE_EAL=y > diff --git a/config/defconfig_ppc_64-power8-linuxapp-gcc > b/config/defconfig_ppc_64-power8-linuxapp-gcc > index d97a885..f1af518 100644 > --- a/config/defconfig_ppc_64-power8-linuxapp-gcc > +++ b/config/defconfig_ppc_64-power8-linuxapp-gcc > @@ -39,8 +39,6 @@ CONFIG_RTE_ARCH_64=y > CONFIG_RTE_TOOLCHAIN="gcc" > CONFIG_RTE_TOOLCHAIN_GCC=y > > -CONFIG_RTE_LIBNAME="powerpc_dpdk" > - > # Note: Power doesn't have this support > CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT=n > > diff --git a/lib/Makefile b/lib/Makefile > index d94355d..c34cf2f 100644 > --- a/lib/Makefile > +++ b/lib/Makefile > @@ -77,5 +77,4 @@ DIRS-$(CONFIG_RTE_LIBRTE_KNI) += librte_kni > DIRS-$(CONFIG_RTE_LIBRTE_IVSHMEM) += librte_ivshmem > endif > > -include $(RTE_SDK)/mk/rte.sharelib.mk > include $(RTE_SDK)/mk/rte.subdir.mk > diff --git a/mk/rte.app.mk b/mk/rte.app.mk > index 63a41e2..e2baa49 100644 > --- a/mk/rte.app.mk > +++ b/mk/rte.app.mk > @@ -61,12 +61,6 @@ ifeq ($(NO_AUTOLIBS),) > > LDLIBS += --whole-archive > > -ifeq ($(CONFIG_RTE_BUILD_COMBINE_LIBS),y) > -LDLIBS += -l$(RTE_LIBNAME) > -endif > - > -ifeq ($(CONFIG_RTE_BUILD_COMBINE_LIBS),n) > - > ifeq ($(CONFIG_RTE_LIBRTE_DISTRIBUTOR),y) > LDLIBS += -lrte_distributor > endif > @@ -137,8 +131,6 @@ ifeq ($(CONFIG_RTE_LIBRTE_VHOST), y) > LDLIBS += -lrte_vhost > endif > > -endif # ! CONFIG_RTE_BUILD_COMBINE_LIBS > - > ifeq ($(CONFIG_RTE_LIBRTE_PMD_PCAP),y) > LDLIBS += -lpcap > endif > @@ -153,8 +145,6 @@ endif > > LDLIBS += --start-group > > -ifeq ($(CONFIG_RTE_BUILD_COMBINE_LIBS),n) > - > ifeq ($(CONFIG_RTE_LIBRTE_KVARGS),y) > LDLIBS += -lrte_kvargs > endif > @@ -253,8 +243,6 @@ endif > > endif # plugins > > -endif # ! CONFIG_RTE_BUILD_COMBINE_LIBS > - > LDLIBS += $(EXECENV_LDLIBS) > > LDLIBS += --end-group > diff --git a/mk/rte.lib.mk b/mk/rte.lib.mk > index 0d7482d..d96101a 100644 > --- a/mk/rte.lib.mk > +++ b/mk/rte.lib.mk > @@ -87,24 +87,6 @@ O_TO_S_DO = @set -e; \ > $(O_TO_S) && \ > echo $(O_TO_S_CMD) > $(call exe2cmd,$(@)) > > -ifeq ($(RTE_BUILD_SHARED_LIB),n) > -O_TO_C = $(AR) crus $(LIB_ONE) $(OBJS-y) > -O_TO_C_STR = $(subst ','\'',$(O_TO_C)) #'# fix syntax highlight > -O_TO_C_DISP = $(if $(V),"$(O_TO_C_STR)"," AR_C $(@)") > -O_TO_C_DO = @set -e; \ > - $(lib_dir) \ > - $(copy_obj) > -else > -O_TO_C = $(LD) -shared $(OBJS-y) -o $(LIB_ONE) > -O_TO_C_STR = $(subst ','\'',$(O_TO_C)) #'# fix syntax highlight > -O_TO_C_DISP = $(if $(V),"$(O_TO_C_STR)"," LD_C $(@)") > -O_TO_C_DO = @set -e; \ > - $(lib_dir) \ > - $(copy_obj) > -endif > - > -copy_obj = cp -f $(OBJS-y) $(RTE_OUTPUT)/build/lib; > -lib_dir = [ -d $(RTE_OUTPUT)/lib ] || mkdir -p $(RTE_OUTPUT)/lib; > -include .$(LIB).cmd > > # > @@ -129,15 +111,6 @@ endif > $(depfile_missing),\ > $(depfile_newer)),\ > $(O_TO_S_DO)) > - > -ifeq ($(RTE_BUILD_COMBINE_LIBS),y) > - $(if $(or \ > -$(file_missing),\ > -$(call cmdline_changed,$(O_TO_C_STR)),\ > -$(depfile_missing),\ > -$(depfile_newer)),\ > -$(O_TO_C_DO)) > -endif > else > $(LIB): $(OBJS-y) $(DEP_$(LIB)) FORCE > @[ -d $(dir $@) ] || mkdir -p $(dir $@) > @@ -153,14 +126,6 @@ $(LIB): $(OBJS-y) $(DEP_$(LIB)) FORCE >
[dpdk-dev] [PATCH v2 1/4] mk: Remove combined library and related options
Hi Sergio, Thank you for explaining the situation. 2015-03-18 12:11, Gonzalez Monroy, Sergio: > Given that the patch to remove combined libraries is not welcome, I'll > try to explain the current situation so we can agree on the way forward. > > Currently we have build config option for shared libraries and combined > libraries. Thus, this results in four possible combinations when > building dpdk: > - not combined static > - not combined shared > - combined static > - combined shared > > The makefile rules/targets for combined are different than for not > combined. Thus, we currently have two different files for > archive/linking (rte.lib.mk and rte.sharelib.mk). > > Since having versioning, combined shared libraries build will be broken > the moment we add a versioned API, as we do not have a global version > map that we use when linking such library. > Also in my opinion, we would want to prevent users linking against a > combined libdpdk.so that may have different features built-in, with the > corresponding debugging difficulties when users > report different problems/errors. I think this would defeat many of the > advantages of using shared libraries. > > By removing the combined library build option, we would simplify the > build system with only two possible choices: > - static > - shared +1 I believe that simplification is the way go. > This would allow us to remove one file (rte.sharelib.mk) and have a > single file with archive/linking rules. > > For the convenience of linking against a single library instead of the > multiple dpdk libraries, there are a few ways to go around it: > - for combined static lib, we can either have a script to re-archive > all libraries into a single/combined library (ie. extract all archives > into one directory, the re-archive all objects into a combined library), >or use a linker script (ie. GROUP ( -lrte_eal -lrte_malloc ... ) ). > - for combined shared lib, we can use a linker script (ie. INPUT ( > -lrte_eal -lrte_malloc ... AS_NEEDED -lrte_hash ...) ) or we could use a > global version map (either somehow merging all independent version maps > or maintaining a global version map). > > My preference would be to remove the combined libs as a build config > option, then either add scripts to create those linker scripts or > document it so users know how to create their own linker scripts. > This would simplify the build process and still be able to provide the > convenience of the combined library by using a linker script. > > Comments? You're right about the word convenience. There are many ways to provide such convenience. The first one is to simply use the DPDK makefiles which abstract linking problems. If using DPDK framework is not an option, we can add new conveniences like scripts or pkgconfig support.
[dpdk-dev] [PATCH] EAL: move rte_common_vect.h into arch/x86
Hi Konstantin, 2015-03-18 10:58, Konstantin Ananyev: > lib/librte_eal/common/Makefile | 1 - > .../common/include/arch/x86/rte_common_vect.h | 128 > + > lib/librte_eal/common/include/rte_common_vect.h| 128 > - I think rte_vect.h is a better name as common is not anymore relevant. Should we add an empty file in ppc_64 directory?
[dpdk-dev] [PATCH v2] Fix `eventfd_link' module leakages and races
The `eventfd_link' module provides an API to "steal" fd from another
process had been written with a bug that leaks `struct file' because
of the extra reference counter increment and missing `fput' call.
The other bug is using another process' `task_struct' without incrementing
a reference counter.
Fix these bugs and refactor the module.
---
Changes since last submission:
* Rebased to the `master' version,
* Corrected error codes returned.
lib/librte_vhost/eventfd_link/eventfd_link.c | 212 ---
1 file changed, 125 insertions(+), 87 deletions(-)
diff --git a/lib/librte_vhost/eventfd_link/eventfd_link.c
b/lib/librte_vhost/eventfd_link/eventfd_link.c
index 7755dd6..57b0a8a 100644
--- a/lib/librte_vhost/eventfd_link/eventfd_link.c
+++ b/lib/librte_vhost/eventfd_link/eventfd_link.c
@@ -65,100 +65,138 @@ put_files_struct(struct files_struct *files)
BUG();
}
+static struct file *
+fget_from_files(struct files_struct *files, unsigned fd)
+{
+ struct file *file;
-static long
-eventfd_link_ioctl(struct file *f, unsigned int ioctl, unsigned long arg)
+ rcu_read_lock();
+ file = fcheck_files(files, fd);
+ if (file)
+ {
+ if (file->f_mode & FMODE_PATH
+ || !atomic_long_inc_not_zero(&file->f_count))
+ file = NULL;
+ }
+ rcu_read_unlock();
+
+ return file;
+}
+
+static int
+close_fd(unsigned fd)
{
- void __user *argp = (void __user *) arg;
- struct task_struct *task_target = NULL;
struct file *file;
- struct files_struct *files;
+ struct files_struct *files = current->files;
struct fdtable *fdt;
+
+ spin_lock(&files->file_lock);
+ fdt = files_fdtable(files);
+ if (fd >= fdt->max_fds)
+ goto out_unlock;
+ file = fdt->fd[fd];
+ if (!file)
+ goto out_unlock;
+ rcu_assign_pointer(fdt->fd[fd], NULL);
+ __clear_bit(fd, fdt->close_on_exec);
+ spin_unlock(&files->file_lock);
+ return filp_close(file, files);
+
+out_unlock:
+ spin_unlock(&files->file_lock);
+ return -EBADF;
+}
+
+
+static long
+eventfd_link_ioctl_copy(unsigned long arg)
+{
+ long ret = -EFAULT;
+ struct task_struct *task_target = NULL;
+ struct file *target_file = NULL;
+ struct files_struct *target_files = NULL;
struct eventfd_copy eventfd_copy;
+ struct pid *pid;
+
+ if (copy_from_user(&eventfd_copy, (void __user*)arg,
+ sizeof(struct eventfd_copy)))
+ goto out;
+
+ /*
+* Find the task struct for the target pid
+*/
+ ret = -ESRCH;
+
+ pid = find_vpid(eventfd_copy.target_pid);
+ if (pid == NULL) {
+ pr_info("Unable to find pid %d\n", eventfd_copy.target_pid);
+ goto out;
+ }
+
+ task_target = get_pid_task(pid, PIDTYPE_PID);
+ if (task_target == NULL) {
+ pr_info("Failed to get task for pid %d\n",
+ eventfd_copy.target_pid);
+ goto out;
+ }
+
+ ret = close_fd(eventfd_copy.source_fd);
+ if (ret)
+ goto out_task;
+ ret = -ESTALE;
+
+ /*
+* Find the file struct associated with the target fd.
+*/
+
+ target_files = get_files_struct(task_target);
+ if (target_files == NULL) {
+ pr_info("Failed to get target files struct\n");
+ goto out_task;
+ }
+
+ ret = -EBADF;
+ target_file = fget_from_files(target_files, eventfd_copy.target_fd);
+
+ if (target_file == NULL) {
+ pr_info("Failed to get file from target pid\n");
+ goto out_target_files;
+ }
- switch (ioctl) {
- case EVENTFD_COPY:
- if (copy_from_user(&eventfd_copy, argp,
- sizeof(struct eventfd_copy)))
- return -EFAULT;
-
- /*
-* Find the task struct for the target pid
-*/
- task_target =
- pid_task(find_vpid(eventfd_copy.target_pid),
PIDTYPE_PID);
- if (task_target == NULL) {
- pr_debug("Failed to get mem ctx for target pid\n");
- return -EFAULT;
- }
-
- files = get_files_struct(current);
- if (files == NULL) {
- pr_debug("Failed to get files struct\n");
- return -EFAULT;
- }
-
- rcu_read_lock();
- file = fcheck_files(files, eventfd_copy.source_fd);
- if (file) {
- if (file->f_mode & FMODE_PATH ||
- !atomic_long_inc_not_zero(&file->f_count))
- file = NULL;
- }
- rcu_read_unlock();
- put_files_s
[dpdk-dev] [PATCH 1/3 v3] librte_hash: Fix unsupported instruction `crc32' in i686 platform
Michael & Thomas,
Should we use software crc function replace of hardware crc function in
'crc32c_sse42_u64' when arch is i686?
Thus application still can use CRC32_SSE42_x64 algorithm for crc calculation
when build with i686 configuration.
This may helpful for simplify application code.
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Michael Qiu
> Sent: Monday, March 09, 2015 1:58 PM
> To: dev at dpdk.org
> Cc: yerden.zhumabekov at sts.kz
> Subject: [dpdk-dev] [PATCH 1/3 v3] librte_hash: Fix unsupported
> instruction `crc32' in i686 platform
>
> CC rte_hash.o
> Error: unsupported instruction `crc32'
>
> The root cause is that i686 platform does not support 'crc32q'
> Need make it only available in x86_64 platform
>
> Signed-off-by: Michael Qiu
> Acked-by: Yerden Zhumabekov
> ---
> v3 --> v2:
> Add sub function for #else which returns 0
> v2 --> v1:
> Make crc32 instruction only works in X86 platform
>
> lib/librte_hash/rte_hash_crc.h | 46 +
> -
> 1 file changed, 36 insertions(+), 10 deletions(-)
>
> diff --git a/lib/librte_hash/rte_hash_crc.h
> b/lib/librte_hash/rte_hash_crc.h
> index d28bb2a..f1dbded 100644
> --- a/lib/librte_hash/rte_hash_crc.h
> +++ b/lib/librte_hash/rte_hash_crc.h
> @@ -47,6 +47,7 @@ extern "C" {
> #include
> #include
> #include
> +#include
>
> /* Lookup tables for software implementation of CRC32C */
> static const uint32_t crc32c_tables[8][256] = {{
> @@ -364,6 +365,7 @@ crc32c_2words(uint64_t data, uint32_t init_val)
> return crc;
> }
>
> +#if defined(RTE_ARCH_I686) || defined(RTE_ARCH_X86_64)
> static inline uint32_t
> crc32c_sse42_u32(uint32_t data, uint32_t init_val)
> {
> @@ -375,16 +377,6 @@ crc32c_sse42_u32(uint32_t data, uint32_t init_val)
> }
>
> static inline uint32_t
> -crc32c_sse42_u64(uint64_t data, uint64_t init_val)
> -{
> - __asm__ volatile(
> - "crc32q %[data], %[init_val];"
> - : [init_val] "+r" (init_val)
> - : [data] "rm" (data));
> - return init_val;
> -}
> -
> -static inline uint32_t
> crc32c_sse42_u64_mimic(uint64_t data, uint64_t init_val)
> {
> union {
> @@ -397,6 +389,40 @@ crc32c_sse42_u64_mimic(uint64_t data, uint64_t
> init_val)
> init_val = crc32c_sse42_u32(d.u32[1], init_val);
> return init_val;
> }
> +#else
> +static inline uint32_t
> +crc32c_sse42_u32(__rte_unused uint32_t data,
> + __rte_unused uint32_t init_val)
> +{
> + return 0;
> +}
> +
> +static inline uint32_t
> +crc32c_sse42_u64_mimic(__rte_unused uint32_t data,
> +__rte_unused uint32_t init_val)
> +{
> + return 0;
> +}
> +#endif
> +
> +#ifdef RTE_ARCH_X86_64
> +static inline uint32_t
> +crc32c_sse42_u64(uint64_t data, uint64_t init_val)
> +{
> + __asm__ volatile(
> + "crc32q %[data], %[init_val];"
> + : [init_val] "+r" (init_val)
> + : [data] "rm" (data));
> + return init_val;
> +}
> +#else
> +static inline uint32_t
> +crc32c_sse42_u64(__rte_unused uint64_t data,
> + __rte_unused uint64_t init_val)
> +{
return crc32c_2words(data, init_val);
> + return 0;
> +}
> +#endif
>
> #define CRC32_SW(1U << 0)
> #define CRC32_SSE42 (1U << 1)
> --
> 1.9.3
[dpdk-dev] [PATCH 1/6] eal: Fix cording style of eal_pci.c and eal_pci_uio.c
On Tue, Mar 17, 2015 at 06:30:40PM +0900, Tetsuya Mukawa wrote:
> This patch fixes cording style of below files in linuxapp and bsdapp.
> - eal_pci.c
> - eal_pci_uio.c
>
> Signed-off-by: Tetsuya Mukawa
Acked-by: Bruce Richardson
> ---
> lib/librte_eal/bsdapp/eal/eal_pci.c | 24 +---
> lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 12
> 2 files changed, 21 insertions(+), 15 deletions(-)
>
> diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
> b/lib/librte_eal/bsdapp/eal/eal_pci.c
> index fe3ef86..3a22b49 100644
> --- a/lib/librte_eal/bsdapp/eal/eal_pci.c
> +++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
> @@ -161,9 +161,10 @@ fail:
> static int
> pci_uio_map_secondary(struct rte_pci_device *dev)
> {
> -size_t i;
> -struct uio_resource *uio_res;
> - struct uio_res_list *uio_res_list = RTE_TAILQ_CAST(rte_uio_tailq.head,
> uio_res_list);
> + size_t i;
> + struct uio_resource *uio_res;
> + struct uio_res_list *uio_res_list =
> + RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
>
> TAILQ_FOREACH(uio_res, uio_res_list, next) {
>
> @@ -179,10 +180,10 @@ pci_uio_map_secondary(struct rte_pci_device *dev)
> != uio_res->maps[i].addr) {
> RTE_LOG(ERR, EAL,
> "Cannot mmap device resource\n");
> - return (-1);
> + return -1;
> }
> }
> - return (0);
> + return 0;
> }
>
> RTE_LOG(ERR, EAL, "Cannot find resource for device\n");
> @@ -201,7 +202,8 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> uint64_t pagesz;
> struct rte_pci_addr *loc = &dev->addr;
> struct uio_resource *uio_res;
> - struct uio_res_list *uio_res_list = RTE_TAILQ_CAST(rte_uio_tailq.head,
> uio_res_list);
> + struct uio_res_list *uio_res_list =
> + RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
> struct uio_map *maps;
>
> dev->intr_handle.fd = -1;
> @@ -209,7 +211,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
>
> /* secondary processes - use already recorded details */
> if (rte_eal_process_type() != RTE_PROC_PRIMARY)
> - return (pci_uio_map_secondary(dev));
> + return pci_uio_map_secondary(dev);
>
> snprintf(devname, sizeof(devname), "/dev/uio at pci:%u:%u:%u",
> dev->addr.bus, dev->addr.devid, dev->addr.function);
> @@ -233,7 +235,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> if ((uio_res = rte_zmalloc("UIO_RES", sizeof (*uio_res), 0)) == NULL) {
> RTE_LOG(ERR, EAL,
> "%s(): cannot store uio mmap details\n", __func__);
> - return (-1);
> + return -1;
> }
>
> snprintf(uio_res->path, sizeof(uio_res->path), "%s", devname);
> @@ -261,7 +263,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> (size_t)maps[j].size)
> ) == NULL) {
> rte_free(uio_res);
> - return (-1);
> + return -1;
> }
>
> maps[j].addr = mapaddr;
> @@ -271,7 +273,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
>
> TAILQ_INSERT_TAIL(uio_res_list, uio_res, next);
>
> - return (0);
> + return 0;
> }
>
> /* Scan one pci sysfs entry, and fill the devices list from it. */
> @@ -311,7 +313,7 @@ pci_scan_one(int dev_pci_fd, struct pci_conf *conf)
> /* FreeBSD has no NUMA support (yet) */
> dev->numa_node = 0;
>
> -/* parse resources */
> + /* parse resources */
> switch (conf->pc_hdr & PCIM_HDRTYPE) {
> case PCIM_HDRTYPE_NORMAL:
> max = PCIR_MAX_BAR_0;
> diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> index 2d1c69b..9cdf24f 100644
> --- a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> +++ b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> @@ -92,7 +92,8 @@ pci_uio_map_secondary(struct rte_pci_device *dev)
> {
> int fd, i;
> struct mapped_pci_resource *uio_res;
> - struct mapped_pci_res_list *uio_res_list =
> RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
> + struct mapped_pci_res_list *uio_res_list =
> + RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
>
> TAILQ_FOREACH(uio_res, uio_res_list, next) {
>
> @@ -272,7 +273,8 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> uint64_t phaddr;
> struct rte_pci_addr *loc = &dev->addr;
> struct mapped_pci_resource *uio_res;
> - struct mapped_pci_res_list *uio_res_list =
> RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
> + struct mapped_pci_res_list *uio_res_list =
> + RTE_TAILQ_CAST(rte_uio_tailq.head,
[dpdk-dev] [PATCH 1/3 v3] librte_hash: Fix unsupported instruction `crc32' in i686 platform
Hi, Yong
If the platform is i686, dpdk will use software crc function.
Thanks,
Michael
-Original Message-
From: Liu, Yong
Sent: Wednesday, March 18, 2015 9:21 PM
To: Qiu, Michael; thomas.monjalon at 6wind.com; dev at dpdk.org
Cc: yerden.zhumabekov at sts.kz
Subject: RE: [dpdk-dev] [PATCH 1/3 v3] librte_hash: Fix unsupported instruction
`crc32' in i686 platform
Michael & Thomas,
Should we use software crc function replace of hardware crc function in
'crc32c_sse42_u64' when arch is i686?
Thus application still can use CRC32_SSE42_x64 algorithm for crc calculation
when build with i686 configuration.
This may helpful for simplify application code.
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Michael Qiu
> Sent: Monday, March 09, 2015 1:58 PM
> To: dev at dpdk.org
> Cc: yerden.zhumabekov at sts.kz
> Subject: [dpdk-dev] [PATCH 1/3 v3] librte_hash: Fix unsupported
> instruction `crc32' in i686 platform
>
> CC rte_hash.o
> Error: unsupported instruction `crc32'
>
> The root cause is that i686 platform does not support 'crc32q'
> Need make it only available in x86_64 platform
>
> Signed-off-by: Michael Qiu
> Acked-by: Yerden Zhumabekov
> ---
> v3 --> v2:
> Add sub function for #else which returns 0
> v2 --> v1:
> Make crc32 instruction only works in X86 platform
>
> lib/librte_hash/rte_hash_crc.h | 46
> +
> -
> 1 file changed, 36 insertions(+), 10 deletions(-)
>
> diff --git a/lib/librte_hash/rte_hash_crc.h
> b/lib/librte_hash/rte_hash_crc.h index d28bb2a..f1dbded 100644
> --- a/lib/librte_hash/rte_hash_crc.h
> +++ b/lib/librte_hash/rte_hash_crc.h
> @@ -47,6 +47,7 @@ extern "C" {
> #include
> #include
> #include
> +#include
>
> /* Lookup tables for software implementation of CRC32C */ static
> const uint32_t crc32c_tables[8][256] = {{ @@ -364,6 +365,7 @@
> crc32c_2words(uint64_t data, uint32_t init_val)
> return crc;
> }
>
> +#if defined(RTE_ARCH_I686) || defined(RTE_ARCH_X86_64)
> static inline uint32_t
> crc32c_sse42_u32(uint32_t data, uint32_t init_val) { @@ -375,16
> +377,6 @@ crc32c_sse42_u32(uint32_t data, uint32_t init_val) }
>
> static inline uint32_t
> -crc32c_sse42_u64(uint64_t data, uint64_t init_val) -{
> - __asm__ volatile(
> - "crc32q %[data], %[init_val];"
> - : [init_val] "+r" (init_val)
> - : [data] "rm" (data));
> - return init_val;
> -}
> -
> -static inline uint32_t
> crc32c_sse42_u64_mimic(uint64_t data, uint64_t init_val) {
> union {
> @@ -397,6 +389,40 @@ crc32c_sse42_u64_mimic(uint64_t data, uint64_t
> init_val)
> init_val = crc32c_sse42_u32(d.u32[1], init_val);
> return init_val;
> }
> +#else
> +static inline uint32_t
> +crc32c_sse42_u32(__rte_unused uint32_t data,
> + __rte_unused uint32_t init_val)
> +{
> + return 0;
> +}
> +
> +static inline uint32_t
> +crc32c_sse42_u64_mimic(__rte_unused uint32_t data,
> +__rte_unused uint32_t init_val) {
> + return 0;
> +}
> +#endif
> +
> +#ifdef RTE_ARCH_X86_64
> +static inline uint32_t
> +crc32c_sse42_u64(uint64_t data, uint64_t init_val) {
> + __asm__ volatile(
> + "crc32q %[data], %[init_val];"
> + : [init_val] "+r" (init_val)
> + : [data] "rm" (data));
> + return init_val;
> +}
> +#else
> +static inline uint32_t
> +crc32c_sse42_u64(__rte_unused uint64_t data,
> + __rte_unused uint64_t init_val)
> +{
return crc32c_2words(data, init_val);
> + return 0;
> +}
> +#endif
>
> #define CRC32_SW(1U << 0)
> #define CRC32_SSE42 (1U << 1)
> --
> 1.9.3
[dpdk-dev] [PATCH] tools: Fix some strings and functions regarding VFIO support
This patch fixes several minor issues in setup.sh:
- show_nics() would not display the current Ethernet settings if
the user only loads the vfio-pci module, b/c it only checks for
presence of igb_uio. Fix this by adding a check for vfio-pci.
- unbind_nics(): Fix option naming and string inside function.
- Exchange a forgotten "igb_uio" with "vfio-pci" in a comment.
---
tools/setup.sh | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/tools/setup.sh b/tools/setup.sh
index ac438c2..5a8b2f3 100755
--- a/tools/setup.sh
+++ b/tools/setup.sh
@@ -427,16 +427,16 @@ grep_meminfo()
#
show_nics()
{
- if /sbin/lsmod | grep -q igb_uio ; then
+ if /sbin/lsmod | grep -q -e igb_uio -e vfio_pci; then
${RTE_SDK}/tools/dpdk_nic_bind.py --status
else
- echo "# Please load the 'igb_uio' kernel module before querying
or "
- echo "# adjusting NIC device bindings"
+ echo "# Please load the 'igb_uio' or 'vfio-pci' kernel module
before "
+ echo "# querying or adjusting NIC device bindings"
fi
}
#
-# Uses dpdk_nic_bind.py to move devices to work with igb_uio
+# Uses dpdk_nic_bind.py to move devices to work with vfio-pci
#
bind_nics_to_vfio()
{
@@ -477,7 +477,7 @@ unbind_nics()
{
${RTE_SDK}/tools/dpdk_nic_bind.py --status
echo ""
- echo -n "Enter PCI address of device to bind to IGB UIO driver: "
+ echo -n "Enter PCI address of device to unbind: "
read PCI_PATH
echo ""
echo -n "Enter name of kernel driver to bind the device to: "
@@ -574,7 +574,7 @@ step5_func()
TEXT[1]="Uninstall all targets"
FUNC[1]="uninstall_targets"
- TEXT[2]="Unbind NICs from IGB UIO driver"
+ TEXT[2]="Unbind NICs from IGB UIO or VFIO driver"
FUNC[2]="unbind_nics"
TEXT[3]="Remove IGB UIO module"
--
1.9.1
[dpdk-dev] [PATCH 1/6] eal: Fix cording style of eal_pci.c and eal_pci_uio.c
Hello Tetsuya,
On Tue, Mar 17, 2015 at 10:30 AM, Tetsuya Mukawa wrote:
> This patch fixes cording style of below files in linuxapp and bsdapp.
> - eal_pci.c
> - eal_pci_uio.c
>
> Signed-off-by: Tetsuya Mukawa
>
I suppose typo for "coding style" in both title and commit log.
The rest looks fine to me.
Acked-by: David Marchand
> ---
> lib/librte_eal/bsdapp/eal/eal_pci.c | 24 +---
> lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 12
> 2 files changed, 21 insertions(+), 15 deletions(-)
>
> diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c
> b/lib/librte_eal/bsdapp/eal/eal_pci.c
> index fe3ef86..3a22b49 100644
> --- a/lib/librte_eal/bsdapp/eal/eal_pci.c
> +++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
> @@ -161,9 +161,10 @@ fail:
> static int
> pci_uio_map_secondary(struct rte_pci_device *dev)
> {
> -size_t i;
> -struct uio_resource *uio_res;
> - struct uio_res_list *uio_res_list =
> RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
> + size_t i;
> + struct uio_resource *uio_res;
> + struct uio_res_list *uio_res_list =
> + RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
>
> TAILQ_FOREACH(uio_res, uio_res_list, next) {
>
> @@ -179,10 +180,10 @@ pci_uio_map_secondary(struct rte_pci_device *dev)
> != uio_res->maps[i].addr) {
> RTE_LOG(ERR, EAL,
> "Cannot mmap device resource\n");
> - return (-1);
> + return -1;
> }
> }
> - return (0);
> + return 0;
> }
>
> RTE_LOG(ERR, EAL, "Cannot find resource for device\n");
> @@ -201,7 +202,8 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> uint64_t pagesz;
> struct rte_pci_addr *loc = &dev->addr;
> struct uio_resource *uio_res;
> - struct uio_res_list *uio_res_list =
> RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
> + struct uio_res_list *uio_res_list =
> + RTE_TAILQ_CAST(rte_uio_tailq.head, uio_res_list);
> struct uio_map *maps;
>
> dev->intr_handle.fd = -1;
> @@ -209,7 +211,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
>
> /* secondary processes - use already recorded details */
> if (rte_eal_process_type() != RTE_PROC_PRIMARY)
> - return (pci_uio_map_secondary(dev));
> + return pci_uio_map_secondary(dev);
>
> snprintf(devname, sizeof(devname), "/dev/uio at pci:%u:%u:%u",
> dev->addr.bus, dev->addr.devid,
> dev->addr.function);
> @@ -233,7 +235,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> if ((uio_res = rte_zmalloc("UIO_RES", sizeof (*uio_res), 0)) ==
> NULL) {
> RTE_LOG(ERR, EAL,
> "%s(): cannot store uio mmap details\n", __func__);
> - return (-1);
> + return -1;
> }
>
> snprintf(uio_res->path, sizeof(uio_res->path), "%s", devname);
> @@ -261,7 +263,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> (size_t)maps[j].size)
> ) == NULL) {
> rte_free(uio_res);
> - return (-1);
> + return -1;
> }
>
> maps[j].addr = mapaddr;
> @@ -271,7 +273,7 @@ pci_uio_map_resource(struct rte_pci_device *dev)
>
> TAILQ_INSERT_TAIL(uio_res_list, uio_res, next);
>
> - return (0);
> + return 0;
> }
>
> /* Scan one pci sysfs entry, and fill the devices list from it. */
> @@ -311,7 +313,7 @@ pci_scan_one(int dev_pci_fd, struct pci_conf *conf)
> /* FreeBSD has no NUMA support (yet) */
> dev->numa_node = 0;
>
> -/* parse resources */
> + /* parse resources */
> switch (conf->pc_hdr & PCIM_HDRTYPE) {
> case PCIM_HDRTYPE_NORMAL:
> max = PCIR_MAX_BAR_0;
> diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> index 2d1c69b..9cdf24f 100644
> --- a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> +++ b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> @@ -92,7 +92,8 @@ pci_uio_map_secondary(struct rte_pci_device *dev)
> {
> int fd, i;
> struct mapped_pci_resource *uio_res;
> - struct mapped_pci_res_list *uio_res_list =
> RTE_TAILQ_CAST(rte_uio_tailq.head, mapped_pci_res_list);
> + struct mapped_pci_res_list *uio_res_list =
> + RTE_TAILQ_CAST(rte_uio_tailq.head,
> mapped_pci_res_list);
>
> TAILQ_FOREACH(uio_res, uio_res_list, next) {
>
> @@ -272,7 +273,8 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> uint64_t phaddr;
> struct rte_pci_addr *loc = &dev->addr;
> struct mapped_pci_resource *uio_res;
> - struct map
[dpdk-dev] [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added + unittest
> From: Vadim Suraev [mailto:vadim.suraev at gmail.com]
> Sent: Wednesday, March 18, 2015 10:41 AM
> To: Ananyev, Konstantin
> Cc: dev at dpdk.org
> Subject: Re: [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added +
> unittest
>
> Hi, Konstantin,
>
> Got it. To make the same, nulling the next should be inside of the block as
> you said.
> One question raises here: If a segment in the chain has refcnt > 1 (so its
> next is not assigned NULL), and the next segment has refcnt
> == 1 (so it is freed), do you think this scenario is real/should be
> considered? If so, the former can be safely freed only by calling
> rte_pktmbuf_free_seg which does not iterate. So why to keep next pointing to
> something?
I think we need it, not just to keep things the same with rte_pktmbuf_free(),
but because it is a right thing to do.
Let say you have a packet in 2 mbufs chained together, both mbufs have
refcnt==2.
Then:
rte_pktmbuf_free(firs_mbuf);
rte_pktmbuf_free(firs_mbuf);
Would work correctly and free both mbufs back to the mempool.
While after:
rte_pktmbuf_free_chain(first_mbuf);
rte_pktmbuf_free_chain(first_mbuf);
We would have first_mbuf freed back into the mempool, while second would get
lost(memory leaking).
Basically free() shouldn't modify any filed inside mbuf, except refcnt if
rte_mbuf_refcnt_update(m, -1) > 0
About your case, when: first_mbuf->refcnt==2 and second_mbuf->refcnt==1.
Right now, rte_pktmbuf_free() can't handle such cases properly,
and, as I know, such situation is not considered as valid one.
Konstantin
> Regards,
> ?Vadim
>
> On Wed, Mar 18, 2015 at 11:56 AM, Ananyev, Konstantin intel.com> wrote:
>
> Hi Vadim,
>
>
> > From: Vadim Suraev [mailto:vadim.suraev at gmail.com]
> > Sent: Wednesday, March 18, 2015 5:19 AM
> > To: Ananyev, Konstantin
> > Cc: dev at dpdk.org; olivier.matz at 6wind.com; stephen at
> > networkplumber.org
> > Subject: Re: [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added +
> > unittest
> >
> > Hi, Konstantin,
> >
> > >Shouldn't the line above be inside if (head != NULL) {...} block?
> > This is removed as Olivier commented before:
> >
> > >> +{
> > > +? ? ?if (likely(head != NULL)) {
> >
> > >I think we should remove this test. The other mbuf functions do not
> > >check this.
> > Regards,
> > ?Vadim.
>
> I meant that in my opinion it should be:
>
> while (head) {
> ? ? ? ? ? ? ?next = head->next;
> -? ? ? ? ? ? ?head->next = NULL;
>
> ? ? ? ? ? ? ?head = __rte_pktmbuf_prefree_seg(head);
> ? ? ? ? ? ? ?if (likely(head != NULL)) {
> +? ? ? ? ? ? ? ? ? head->next = NULL;
> ? ? ? ? ? ? ? ? ? ? ?RTE_MBUF_ASSERT(rte_mbuf_refcnt_read(head) == 0);
>
> Same as rte_pktmbuf_free() doing.
>
> Konstantin
>
> >
> > On Wed, Mar 18, 2015 at 1:46 AM, Ananyev, Konstantin > intel.com> wrote:
> > Hi Vadim,
> >
> > > -Original Message-
> > > From: vadim.suraev at gmail.com [mailto:vadim.suraev at gmail.com]
> > > Sent: Tuesday, March 17, 2015 9:36 PM
> > > To: dev at dpdk.org
> > > Cc: olivier.matz at 6wind.com; stephen at networkplumber.org; Ananyev,
> > > Konstantin; vadim.suraev at gmail.com
> > > Subject: [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added +
> > > unittest
> > >
> > > From: "vadim.suraev at gmail.com"
> > >
> > > This patch adds mbuf bulk allocation/freeing functions and unittest
> > >
> > > Signed-off-by: Vadim Suraev
> > >
> > > ---
> > > New in v2:
> > >? ? ?- function rte_pktmbuf_alloc_bulk added
> > >? ? ?- function rte_pktmbuf_bulk_free added
> > >? ? ?- function rte_pktmbuf_free_chain added
> > >? ? ?- applied reviewers' comments
> > >
> > >? app/test/test_mbuf.c? ? ? ?|? ?94
> > >+++-
> > >? lib/librte_mbuf/rte_mbuf.h |? ?89
> > >+
> > >? 2 files changed, 182 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c
> > > index 1ff66cb..b20c6a4 100644
> > > --- a/app/test/test_mbuf.c
> > > +++ b/app/test/test_mbuf.c
> > > @@ -77,6 +77,7 @@
> > >? #define REFCNT_RING_SIZE? ? ? ? (REFCNT_MBUF_NUM * REFCNT_MAX_REF)
> > >
> > >? #define MAKE_STRING(x)? ? ? ? ? # x
> > > +#define MBUF_POOL_LOCAL_CACHE_SIZE 32
> > >
> > >? static struct rte_mempool *pktmbuf_pool = NULL;
> > >
> > > @@ -405,6 +406,84 @@ test_pktmbuf_pool(void)
> > >? ? ? ?return ret;
> > >? }
> > >
> > > +/* test pktmbuf bulk allocation and freeing
> > > +*/
> > > +static int
> > > +test_pktmbuf_pool_bulk(void)
> > > +{
> > > +? ? ?unsigned i;
> > > +? ? ?/* size of mempool - size of local cache, otherwise may fail */
> > > +? ? ?unsigned mbufs_to_allocate = NB_MBUF - MBUF_POOL_LOCAL_CACHE_SIZE;
> > > +? ? ?struct rte_mbuf *m[mbufs_to_allocate];
> > > +? ? ?int ret = 0;
> > > +? ? ?unsigned mbuf_count_before_allocation =
> > > rte_mempool_count(pktmbuf_pool);
> > > +
> > > +? ? ?for (i = 0; i < mbufs_to_allocate; i++)
> > > +? ? ? ? ? ? ?m[i] = NULL;
> > > +? ? ?/* alloc NB_MBUF-MBUF_POOL_LOCAL_CACHE_SIZE mbufs */
> > > +? ?
[dpdk-dev] [PATCH] fix build warning and failure in Suse11
Hi Thomas, I'll separated this patch set into several patches and send them out later. I think this patch can also fix some issue on gcc 44. > -Original Message- > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com] > Sent: Wednesday, March 18, 2015 6:51 PM > To: Liu, Yong > Cc: dev at dpdk.org > Subject: Re: [dpdk-dev] [PATCH] fix build warning and failure in Suse11 > > Hi Yong, > > Thanks for working on these important fixes. > > 2015-03-18 15:10, Yong Liu: > > Suse11 SP3 default gcc version is 4.3.4, some options not support on > this version. > > I guess some of these errors are not only specific to Suse-11? > Maybe that 1 patch per issue would be easier to read and could provide a > more > accurate description. > > > error: implicit declaration of function ?_mm_alignr_epi8? > > solution: include tmmintrin.h when enable SSE3 > > > > error: unrecognized command line option "-Wno-unused-but-set-variable" > > solution: add version check in fm10k Makefile > > > > error: enic_main.c:845: error: initialized field overwritten > > solution: change struct initialization code > > > > error: ?testfn_pci_cmd? defined but not used > > solution: add __attribute__((unused)) before function definition > > Please could you explain more the problem? > There are other constructors in DPDK which don't need the unused attribute. > > > > > error: unrecognized command line option "-fno-var-tracking-assignments" > > solution: add version check in app/test/Makefile > > > > error: implicit declaration of function ?pread? > > solution: add _GNU_SOURCE flag when compile eal_pci_uio and > eal_interrupts > > > > signed-off-by: Marvin Liu > > Please use -s git option to have an automatic well formatted Signed-off. > Your previous contributions were signed "Yong Liu". Do you prefer Marvin > Liu? Thanks Thomas, "Marvin" the name I used in DTS branch. For my major task is in DTS project, I preferred to use "Marvin Liu":)
[dpdk-dev] [PATCH 2/6] eal: Close file descriptor of uio configuration
On Tue, Mar 17, 2015 at 10:30 AM, Tetsuya Mukawa wrote:
> When pci_uio_unmap_resource() is called, a file descriptor that is used
> for uio configuration should be closed.
>
> Signed-off-by: Tetsuya Mukawa
> ---
> lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 8 +++-
> 1 file changed, 7 insertions(+), 1 deletion(-)
>
> diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> index 9cdf24f..b971ec9 100644
> --- a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> +++ b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> @@ -459,8 +459,14 @@ pci_uio_unmap_resource(struct rte_pci_device *dev)
>
> /* close fd if in primary process */
> close(dev->intr_handle.fd);
> -
> dev->intr_handle.fd = -1;
> +
> + /* close cfg_fd if in primary process */
+ if (dev->intr_handle.uio_cfg_fd >= 0) {
> + close(dev->intr_handle.uio_cfg_fd);
> + dev->intr_handle.uio_cfg_fd = -1;
> + }
> +
> dev->intr_handle.type = RTE_INTR_HANDLE_UNKNOWN;
> }
> #endif /* RTE_LIBRTE_EAL_HOTPLUG */
>
Hum, why check for < 0 value ?
There is no such check for intr_handle.fd and I can see no reason why we
should behave differently.
--
David Marchand
[dpdk-dev] [PATCH] EAL: move rte_common_vect.h into arch/x86
Hi Thomas, > -Original Message- > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com] > Sent: Wednesday, March 18, 2015 1:14 PM > To: Ananyev, Konstantin > Cc: dev at dpdk.org > Subject: Re: [dpdk-dev] [PATCH] EAL: move rte_common_vect.h into arch/x86 > > Hi Konstantin, > > 2015-03-18 10:58, Konstantin Ananyev: > > lib/librte_eal/common/Makefile | 1 - > > .../common/include/arch/x86/rte_common_vect.h | 128 > > + > > lib/librte_eal/common/include/rte_common_vect.h| 128 > > - > > I think rte_vect.h is a better name as common is not anymore relevant. I don't mind, but it means more changes - all files which include it, would need to be changed. > Should we add an empty file in ppc_64 directory? I thought about that too, but it seems not necessary. It is included by: lib/librte_lpm/rte_lpm.h lib/librte_eal/common/include/arch/x86/rte_memcpy.h lib/librte_acl/rte_acl_osdep.h As I understand, neither LPM, neither ACL are supported on PPC right now. Again, if we'll provide an empty one for PPC, it wouldn't help to compile LPM/ACL on PPC anyway, as both use SSE instrincts inside their code. Konstantin
[dpdk-dev] [PATCH v2 1/4] mk: Remove combined library and related options
Hi, On Wed, Mar 18, 2015 at 2:59 PM, Thomas Monjalon wrote: > Hi Sergio, > > Thank you for explaining the situation. > > 2015-03-18 12:11, Gonzalez Monroy, Sergio: >> Given that the patch to remove combined libraries is not welcome, I'll >> try to explain the current situation so we can agree on the way forward. >> >> Currently we have build config option for shared libraries and combined >> libraries. Thus, this results in four possible combinations when >> building dpdk: >> - not combined static >> - not combined shared >> - combined static >> - combined shared >> >> The makefile rules/targets for combined are different than for not >> combined. Thus, we currently have two different files for >> archive/linking (rte.lib.mk and rte.sharelib.mk). >> >> Since having versioning, combined shared libraries build will be broken >> the moment we add a versioned API, as we do not have a global version >> map that we use when linking such library. >> Also in my opinion, we would want to prevent users linking against a >> combined libdpdk.so that may have different features built-in, with the >> corresponding debugging difficulties when users >> report different problems/errors. I think this would defeat many of the >> advantages of using shared libraries. >> >> By removing the combined library build option, we would simplify the >> build system with only two possible choices: >> - static >> - shared > > +1 > I believe that simplification is the way go. > >> This would allow us to remove one file (rte.sharelib.mk) and have a >> single file with archive/linking rules. >> >> For the convenience of linking against a single library instead of the >> multiple dpdk libraries, there are a few ways to go around it: >> - for combined static lib, we can either have a script to re-archive >> all libraries into a single/combined library (ie. extract all archives >> into one directory, the re-archive all objects into a combined library), >>or use a linker script (ie. GROUP ( -lrte_eal -lrte_malloc ... ) ). Would the linker script be provided in the repository or would it be the responsibility of people building against the DPDK? If I'd need to make a linker script with the list of libraries to link against, might as well put that list in my SConstruct / Makefile and be done with it. So the "write your own linker script" and "just deal with separate libraries" options don't seem that different to me. Let me ask you something - I understand your concerns about simplifying Makefiles and the concerns about versioning. How significant is the "separate libs" use case? And especially the "separate libs" in the current division of the code / libraries? I counted about ~30 libs in 1.8.0 under build/lib. Are there people using librte_eal without rte_malloc? Or rte_malloc without rte_mempool? I noticed that some examples I built ended up using --whole-archive -lrte_eal -lrte_etc To me, --whole-archive is one way of saying "we have lots of libraries with obscure dependencies", maybe reducing the number of libs might also be a way to make the combined lib unnecessary? I wouldn't bother with the combined lib if I had 3-4 libs to link against instead of the number of libs needed now. Just asking - obviously you guys are maintaining the code and know best, but I want to better understand the context from your side, as opposed to my (selfish) user perspective :). Thanks, Stefan. >> - for combined shared lib, we can use a linker script (ie. INPUT ( >> -lrte_eal -lrte_malloc ... AS_NEEDED -lrte_hash ...) ) or we could use a >> global version map (either somehow merging all independent version maps >> or maintaining a global version map). >> >> My preference would be to remove the combined libs as a build config >> option, then either add scripts to create those linker scripts or >> document it so users know how to create their own linker scripts. >> This would simplify the build process and still be able to provide the >> convenience of the combined library by using a linker script. >> >> Comments? > > You're right about the word convenience. > There are many ways to provide such convenience. > The first one is to simply use the DPDK makefiles which abstract linking > problems. > If using DPDK framework is not an option, we can add new conveniences like > scripts or pkgconfig support. >
[dpdk-dev] [PATCH 3/6] eal: Fix memory leaks and needless incrementation of pci uio implementation
On Tue, Mar 17, 2015 at 10:30 AM, Tetsuya Mukawa wrote:
> When pci_map_resource() is failed but path is allocated correctly,
> path won't be freed. Also, when open() is failed, uio_res won't be freed.
> This patch fixes these memory leaks.
> When pci_map_resource() is failed, mapaddr will be MAP_FAILED.
> In this case, pci_map_addr should not be incremented.
>
> Also, the patch fixes belows.
> - To shrink code, move close().
> - Remove fail variable.
>
> Signed-off-by: Tetsuya Mukawa
> ---
> lib/librte_eal/linuxapp/eal/eal_pci_uio.c | 28
> ++--
> 1 file changed, 14 insertions(+), 14 deletions(-)
>
> diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> index b971ec9..5044884 100644
> --- a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> +++ b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
> @@ -333,7 +333,6 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> maps = uio_res->maps;
> for (i = 0, map_idx = 0; i != PCI_MAX_RESOURCE; i++) {
> int fd;
> - int fail = 0;
>
> /* skip empty BAR */
> phaddr = dev->mem_resource[i].phys_addr;
> @@ -347,6 +346,11 @@ pci_uio_map_resource(struct rte_pci_device *dev)
> loc->domain, loc->bus, loc->devid,
> loc->function,
> i);
>
> + /* allocate memory to keep path */
> + maps[map_idx].path = rte_malloc(NULL, strlen(devname) + 1,
> 0);
> + if (maps[map_idx].path == NULL)
> + goto fail0;
> +
>
[snip]
Neither fail0 nor fail1 labels seem to free previous allocations.
Did I miss something ?
[snip]
+
> +fail1:
> + rte_free(maps[map_idx].path);
> +fail0:
> + rte_free(uio_res);
> + return -1;
> }
>
--
David Marchand
[dpdk-dev] [PATCH v2 1/4] mk: Remove combined library and related options
On 18/03/2015 15:30, Stefan Puiu wrote: > Hi, > > On Wed, Mar 18, 2015 at 2:59 PM, Thomas Monjalon > wrote: >> Hi Sergio, >> >> Thank you for explaining the situation. >> >> 2015-03-18 12:11, Gonzalez Monroy, Sergio: >>> Given that the patch to remove combined libraries is not welcome, I'll >>> try to explain the current situation so we can agree on the way forward. >>> >>> Currently we have build config option for shared libraries and combined >>> libraries. Thus, this results in four possible combinations when >>> building dpdk: >>> - not combined static >>> - not combined shared >>> - combined static >>> - combined shared >>> >>> The makefile rules/targets for combined are different than for not >>> combined. Thus, we currently have two different files for >>> archive/linking (rte.lib.mk and rte.sharelib.mk). >>> >>> Since having versioning, combined shared libraries build will be broken >>> the moment we add a versioned API, as we do not have a global version >>> map that we use when linking such library. >>> Also in my opinion, we would want to prevent users linking against a >>> combined libdpdk.so that may have different features built-in, with the >>> corresponding debugging difficulties when users >>> report different problems/errors. I think this would defeat many of the >>> advantages of using shared libraries. >>> >>> By removing the combined library build option, we would simplify the >>> build system with only two possible choices: >>> - static >>> - shared >> +1 >> I believe that simplification is the way go. >> >>> This would allow us to remove one file (rte.sharelib.mk) and have a >>> single file with archive/linking rules. >>> >>> For the convenience of linking against a single library instead of the >>> multiple dpdk libraries, there are a few ways to go around it: >>>- for combined static lib, we can either have a script to re-archive >>> all libraries into a single/combined library (ie. extract all archives >>> into one directory, the re-archive all objects into a combined library), >>> or use a linker script (ie. GROUP ( -lrte_eal -lrte_malloc ... ) ). > Would the linker script be provided in the repository or would it be > the responsibility of people building against the DPDK? If I'd need to > make a linker script with the list of libraries to link against, might > as well put that list in my SConstruct / Makefile and be done with it. > So the "write your own linker script" and "just deal with separate > libraries" options don't seem that different to me. > > Let me ask you something - I understand your concerns about > simplifying Makefiles and the concerns about versioning. How > significant is the "separate libs" use case? And especially the > "separate libs" in the current division of the code / libraries? I > counted about ~30 libs in 1.8.0 under build/lib. Are there people > using librte_eal without rte_malloc? Or rte_malloc without > rte_mempool? > > I noticed that some examples I built ended up using --whole-archive > -lrte_eal -lrte_etc To me, --whole-archive is one way of saying > "we have lots of libraries with obscure dependencies", maybe reducing > the number of libs might also be a way to make the combined lib > unnecessary? I wouldn't bother with the combined lib if I had 3-4 libs > to link against instead of the number of libs needed now. > > Just asking - obviously you guys are maintaining the code and know > best, but I want to better understand the context from your side, as > opposed to my (selfish) user perspective :). > > Thanks, > Stefan. > Some of this questions have been discussed previously: http://dpdk.org/ml/archives/dev/2014-October/007389.html http://dpdk.org/ml/archives/dev/2015-January/010917.html http://dpdk.org/ml/archives/dev/2015-January/011912.html I think those threads will provide enough context but as a very general summary, only eal, malloc, mempool and ring would have circular dependencies and you could consider them as 'core' libraries in the sense that almost (if not all) dpdk apps are going to be linked against them. Most of dpdk libraries are optional features that your application may or may not be using. Sergio >>> - for combined shared lib, we can use a linker script (ie. INPUT ( >>> -lrte_eal -lrte_malloc ... AS_NEEDED -lrte_hash ...) ) or we could use a >>> global version map (either somehow merging all independent version maps >>> or maintaining a global version map). >>> >>> My preference would be to remove the combined libs as a build config >>> option, then either add scripts to create those linker scripts or >>> document it so users know how to create their own linker scripts. >>> This would simplify the build process and still be able to provide the >>> convenience of the combined library by using a linker script. >>> >>> Comments? >> You're right about the word convenience. >> There are many ways to provide such convenience. >> The first one is to simply use the DPDK makefiles which abstract linking >> p
[dpdk-dev] [PATCH] igb: handle VF LPE mailbox message
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Sergio Gonzalez Monroy
> Sent: Wednesday, March 18, 2015 12:02 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] igb: handle VF LPE mailbox message
>
> This patch adds the handle function for the LPE mailbox message (VF to
> PF) to set maximum packet size, which can be used to enable jumbo
> frame support.
>
> Signed-off-by: Sergio Gonzalez Monroy
Acked-by: Konstantin Ananyev
> ---
> lib/librte_pmd_e1000/igb_pf.c | 28
> 1 file changed, 28 insertions(+)
>
> diff --git a/lib/librte_pmd_e1000/igb_pf.c b/lib/librte_pmd_e1000/igb_pf.c
> index bc3816a..2d49379 100644
> --- a/lib/librte_pmd_e1000/igb_pf.c
> +++ b/lib/librte_pmd_e1000/igb_pf.c
> @@ -395,6 +395,31 @@ igb_vf_set_vlan(struct rte_eth_dev *dev, uint32_t vf,
> uint32_t *msgbuf)
> }
>
> static int
> +igb_vf_set_rlpml(struct rte_eth_dev *dev, uint32_t vf, uint32_t *msgbuf)
> +{
> + struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
> + uint16_t rlpml = msgbuf[1] & E1000_VMOLR_RLPML_MASK;
> + uint32_t max_frame = rlpml + ETHER_HDR_LEN + ETHER_CRC_LEN;
> + uint32_t vmolr;
> +
> + if ((max_frame < ETHER_MIN_LEN) || (max_frame >
> ETHER_MAX_JUMBO_FRAME_LEN))
> + return -1;
> +
> + vmolr = E1000_READ_REG(hw, E1000_VMOLR(vf));
> +
> + vmolr &= ~E1000_VMOLR_RLPML_MASK;
> + vmolr |= rlpml;
> +
> + /* Enable Long Packet support */
> + vmolr |= E1000_VMOLR_LPE;
> +
> + E1000_WRITE_REG(hw, E1000_VMOLR(vf), vmolr);
> + E1000_WRITE_FLUSH(hw);
> +
> + return 0;
> +}
> +
> +static int
> igb_rcv_msg_from_vf(struct rte_eth_dev *dev, uint16_t vf)
> {
> uint16_t mbx_size = E1000_VFMAILBOX_SIZE;
> @@ -428,6 +453,9 @@ igb_rcv_msg_from_vf(struct rte_eth_dev *dev, uint16_t vf)
> case E1000_VF_SET_MULTICAST:
> retval = igb_vf_set_multicast(dev, vf, msgbuf);
> break;
> + case E1000_VF_SET_LPE:
> + retval = igb_vf_set_rlpml(dev, vf, msgbuf);
> + break;
> case E1000_VF_SET_VLAN:
> retval = igb_vf_set_vlan(dev, vf, msgbuf);
> break;
> --
> 1.9.3
[dpdk-dev] [PATCH] EAL: move rte_common_vect.h into arch/x86
2015-03-18 15:24, Ananyev, Konstantin: > > Hi Thomas, > > > -Original Message- > > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com] > > Sent: Wednesday, March 18, 2015 1:14 PM > > To: Ananyev, Konstantin > > Cc: dev at dpdk.org > > Subject: Re: [dpdk-dev] [PATCH] EAL: move rte_common_vect.h into arch/x86 > > > > Hi Konstantin, > > > > 2015-03-18 10:58, Konstantin Ananyev: > > > lib/librte_eal/common/Makefile | 1 - > > > .../common/include/arch/x86/rte_common_vect.h | 128 > > > + > > > lib/librte_eal/common/include/rte_common_vect.h| 128 > > > - > > > > I think rte_vect.h is a better name as common is not anymore relevant. > > I don't mind, but it means more changes - all files which include it, would > need to be changed. I think file naming deserves it. > > Should we add an empty file in ppc_64 directory? > > I thought about that too, but it seems not necessary. > It is included by: > lib/librte_lpm/rte_lpm.h > lib/librte_eal/common/include/arch/x86/rte_memcpy.h > lib/librte_acl/rte_acl_osdep.h > > As I understand, neither LPM, neither ACL are supported on PPC right now. > Again, if we'll provide an empty one for PPC, it wouldn't help to compile > LPM/ACL on PPC anyway, > as both use SSE instrincts inside their code. Yes, it was an open question. It's probably better to create the PPC file when really needed.
[dpdk-dev] [PATCH] i40e: revert internal switch of PF
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, March 18, 2015 3:07 AM
> To: Zhang, Helin
> Cc: dev at dpdk.org; Wu, Jingjing
> Subject: [PATCH] i40e: revert internal switch of PF
>
> VEB switching is blocking VF.
> If the source mac address of packet sent from VF is not listed in the
> VEB?s mac table, the VEB will switch the packet back to the VF.
> It's an hardware issue.
>
> Reverts: 2ccabd8cd1f6 ("i40e: enable internal switch of PF").
>
> Reported-by: Jingjing Wu
> Signed-off-by: Thomas Monjalon
> ---
> lib/librte_pmd_i40e/i40e_ethdev.c | 36
> 1 file changed, 36 deletions(-)
>
> diff --git a/lib/librte_pmd_i40e/i40e_ethdev.c
> b/lib/librte_pmd_i40e/i40e_ethdev.c
> index 6888072..cf6685e 100644
> --- a/lib/librte_pmd_i40e/i40e_ethdev.c
> +++ b/lib/librte_pmd_i40e/i40e_ethdev.c
> @@ -2868,40 +2868,6 @@ i40e_vsi_dump_bw_config(struct i40e_vsi *vsi)
> return 0;
> }
>
> -/*
> - * i40e_enable_pf_lb
> - * @pf: pointer to the pf structure
> - *
> - * allow loopback on pf
> - */
> -static inline void
> -i40e_enable_pf_lb(struct i40e_pf *pf)
> -{
> - struct i40e_hw *hw = I40E_PF_TO_HW(pf);
> - struct i40e_vsi_context ctxt;
> - int ret;
> -
> - memset(&ctxt, 0, sizeof(ctxt));
> - ctxt.seid = pf->main_vsi_seid;
> - ctxt.pf_num = hw->pf_id;
> - ret = i40e_aq_get_vsi_params(hw, &ctxt, NULL);
> - if (ret) {
> - PMD_DRV_LOG(ERR, "couldn't get pf vsi config, err %d, aq_err
> %d",
> - ret, hw->aq.asq_last_status);
> - return;
> - }
> - ctxt.flags = I40E_AQ_VSI_TYPE_PF;
> - ctxt.info.valid_sections =
> - rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);
> - ctxt.info.switch_id |=
> - rte_cpu_to_le_16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
> -
> - ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
> - if (ret)
> - PMD_DRV_LOG(ERR, "update vsi switch failed, aq_err=%d\n",
> - hw->aq.asq_last_status);
> -}
> -
> /* Setup a VSI */
> struct i40e_vsi *
> i40e_vsi_setup(struct i40e_pf *pf,
> @@ -2937,8 +2903,6 @@ i40e_vsi_setup(struct i40e_pf *pf,
> PMD_DRV_LOG(ERR, "VEB setup failed");
> return NULL;
> }
> - /* set ALLOWLOOPBACk on pf, when veb is created */
> - i40e_enable_pf_lb(pf);
> }
>
> vsi = rte_zmalloc("i40e_vsi", sizeof(struct i40e_vsi), 0);
> --
> 2.2.2
Acked-by: Jingjing Wu
[dpdk-dev] i40e_aq_get_firmware_version failure
Hi, I am trying to start DPDK with 40G Intel NIC and get a failure at initialization stage in i40e_aq_get_firmware_version(). For some reason this function reaches TIMEOUT for more than maximum allowed times (10 times). In the note below I understand that several failures may be considerable but not as many as I have. Should I enlarge the retries number ? Is it a HW issue ? Anyone who faced it or may assist please comment. Thanks, Yan
[dpdk-dev] Virtual NIC interface fails to receive packets
Hi, I'm trying to test the l2fwd sample application by passing packets from traffic generator. I'm doing this testing in a VM with a configuration hav?ing virtual NIC port. When I bind the virtual NIC port with the igb driver, the interface doesnt get listed in the "ifconfig" output. Also the virtual NIC port fails to receive packets from the traffic generator after the interface is bound to igb driver. Please let me know whether this is the expected behavior and also let me know how I can make the traffic pass through the port bound to the igb driver. While starting the VM the interfaces currently active: controller at controller-VirtualBox:~$ ifconfig -a eth0 Link encap:Ethernet HWaddr 08:00:27:21:a9:02 inet addr:192.168.56.101 Bcast:192.168.56.255 Mask:255.255.255.0 inet6 addr: fe80::a00:27ff:fe21:a902/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2 errors:0 dropped:0 overruns:0 frame:0 TX packets:70 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1180 (1.1 KB) TX bytes:14504 (14.5 KB) eth1 Link encap:Ethernet HWaddr 08:00:27:ef:8b:a1 inet addr:192.168.56.101 Bcast:192.168.56.255 Mask:255.255.255.0 inet6 addr: fe80::a00:27ff:feef:8ba1/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2 errors:0 dropped:0 overruns:0 frame:0 TX packets:80 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1180 (1.1 KB) TX bytes:15132 (15.1 KB) eth2 Link encap:Ethernet HWaddr 08:00:27:bc:04:b6 inet addr:10.0.2.15 Bcast:10.0.2.255 Mask:255.255.255.0 inet6 addr: fe80::a00:27ff:febc:4b6/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:150 errors:0 dropped:0 overruns:0 frame:0 TX packets:215 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:12304 (12.3 KB) TX bytes:25233 (25.2 KB) eth4 Link encap:Ethernet HWaddr 08:00:27:b7:3a:25 inet6 addr: fe80::a00:27ff:feb7:3a25/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:196 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:0 (0.0 B) TX bytes:44609 (44.6 KB) loLink encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:65536 Metric:1 RX packets:300 errors:0 dropped:0 overruns:0 frame:0 TX packets:300 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:18580 (18.5 KB) TX bytes:18580 (18.5 KB) After executing this command controller at controller-VirtualBox:~/dpdk-1.7.1$ sudo insmod $RTE_SDK/$RTE_TARGET/kmod/igb_uio.ko controller at controller-VirtualBox:~/dpdk-1.7.1$ sudo $RTE_SDK/tools/dpdk_nic_bind.py --bind=igb_uio 00:08.0 00:09.0 00:0a.0 Routing table indicates that interface :00:08.0 is active. Not modifying Routing table indicates that interface :00:09.0 is active. Not modifying controller at controller-VirtualBox:~/dpdk-1.7.1$ sudo $RTE_SDK/tools/dpdk_nic_bind.py --status Network devices using DPDK-compatible driver :00:0a.0 '82545EM Gigabit Ethernet Controller (Copper)' drv=igb_uio unused=e1000 Network devices using kernel driver === :00:08.0 '82545EM Gigabit Ethernet Controller (Copper)' if=eth0 drv=e1000 unused=igb_uio *Active* :00:09.0 '82545EM Gigabit Ethernet Controller (Copper)' if=eth1 drv=e1000 unused=igb_uio *Active* :00:11.0 '82545EM Gigabit Ethernet Controller (Copper)' if=eth2 drv=e1000 unused=igb_uio *Active* Other network devices = got the below output: controller at controller-VirtualBox:~$ ifconfig eth4 >>> Corresponds to the interface bound to igb driver eth4: error fetching interface information: Device not found controller at controller-VirtualBox:~$ ifconfig eth2 eth2 Link encap:Ethernet HWaddr 08:00:27:bc:04:b6 inet addr:10.0.2.15 Bcast:10.0.2.255 Mask:255.255.255.0 inet6 addr: fe80::a00:27ff:febc:4b6/64 Scope:Link UP BROADCAST RUNNING PROMISC MULTICAST MTU:1500 Metric:1 RX packets:197 errors:0 dropped:0 overruns:0 frame:0 TX packets:271 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:15864 (15.8 KB) TX bytes:30592 (30.5 KB) controller at controller-VirtualBox:~$ ifconfig eth1 eth1 Link encap:Ethernet HWaddr 08:00:27:ef:8b:a1 inet addr:192.168.56.101 Bcast:192.168.56.255 Mask:255.255.255.0 inet6 addr: fe80::a00:27ff:feef:8ba1/64 Scope:Link
[dpdk-dev] [PATCH v2 1/4] mk: Remove combined library and related options
On Wed, Mar 18, 2015 at 12:11:50PM +, Gonzalez Monroy, Sergio wrote: > On 12/03/2015 16:27, Sergio Gonzalez Monroy wrote: > >Remove CONFIG_RTE_BUILD_COMBINE_LIBS and CONFIG_RTE_LIBNAME. > > > >Signed-off-by: Sergio Gonzalez Monroy > >--- > > config/common_bsdapp| 6 -- > > config/common_linuxapp | 6 -- > > config/defconfig_ppc_64-power8-linuxapp-gcc | 2 - > > lib/Makefile| 1 - > > mk/rte.app.mk | 12 > > mk/rte.lib.mk | 35 -- > > mk/rte.sdkbuild.mk | 3 - > > mk/rte.sharelib.mk | 101 > > > > mk/rte.vars.mk | 9 --- > > 9 files changed, 175 deletions(-) > > delete mode 100644 mk/rte.sharelib.mk > > > >diff --git a/config/common_bsdapp b/config/common_bsdapp > >index 8ff4dc2..7ee5ecf 100644 > >--- a/config/common_bsdapp > >+++ b/config/common_bsdapp > >@@ -79,12 +79,6 @@ CONFIG_RTE_FORCE_INTRINSICS=n > > CONFIG_RTE_BUILD_SHARED_LIB=n > > # > >-# Combine to one single library > >-# > >-CONFIG_RTE_BUILD_COMBINE_LIBS=n > >-CONFIG_RTE_LIBNAME=intel_dpdk > >- > >-# > > # Compile Environment Abstraction Layer > > # > > CONFIG_RTE_LIBRTE_EAL=y > >diff --git a/config/common_linuxapp b/config/common_linuxapp > >index 97f1c9e..ae13805 100644 > >--- a/config/common_linuxapp > >+++ b/config/common_linuxapp > >@@ -79,12 +79,6 @@ CONFIG_RTE_FORCE_INTRINSICS=n > > CONFIG_RTE_BUILD_SHARED_LIB=n > > # > >-# Combine to one single library > >-# > >-CONFIG_RTE_BUILD_COMBINE_LIBS=n > >-CONFIG_RTE_LIBNAME="intel_dpdk" > >- > >-# > > # Compile Environment Abstraction Layer > > # > > CONFIG_RTE_LIBRTE_EAL=y > >diff --git a/config/defconfig_ppc_64-power8-linuxapp-gcc > >b/config/defconfig_ppc_64-power8-linuxapp-gcc > >index d97a885..f1af518 100644 > >--- a/config/defconfig_ppc_64-power8-linuxapp-gcc > >+++ b/config/defconfig_ppc_64-power8-linuxapp-gcc > >@@ -39,8 +39,6 @@ CONFIG_RTE_ARCH_64=y > > CONFIG_RTE_TOOLCHAIN="gcc" > > CONFIG_RTE_TOOLCHAIN_GCC=y > >-CONFIG_RTE_LIBNAME="powerpc_dpdk" > >- > > # Note: Power doesn't have this support > > CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT=n > >diff --git a/lib/Makefile b/lib/Makefile > >index d94355d..c34cf2f 100644 > >--- a/lib/Makefile > >+++ b/lib/Makefile > >@@ -77,5 +77,4 @@ DIRS-$(CONFIG_RTE_LIBRTE_KNI) += librte_kni > > DIRS-$(CONFIG_RTE_LIBRTE_IVSHMEM) += librte_ivshmem > > endif > >-include $(RTE_SDK)/mk/rte.sharelib.mk > > include $(RTE_SDK)/mk/rte.subdir.mk > >diff --git a/mk/rte.app.mk b/mk/rte.app.mk > >index 63a41e2..e2baa49 100644 > >--- a/mk/rte.app.mk > >+++ b/mk/rte.app.mk > >@@ -61,12 +61,6 @@ ifeq ($(NO_AUTOLIBS),) > > LDLIBS += --whole-archive > >-ifeq ($(CONFIG_RTE_BUILD_COMBINE_LIBS),y) > >-LDLIBS += -l$(RTE_LIBNAME) > >-endif > >- > >-ifeq ($(CONFIG_RTE_BUILD_COMBINE_LIBS),n) > >- > > ifeq ($(CONFIG_RTE_LIBRTE_DISTRIBUTOR),y) > > LDLIBS += -lrte_distributor > > endif > >@@ -137,8 +131,6 @@ ifeq ($(CONFIG_RTE_LIBRTE_VHOST), y) > > LDLIBS += -lrte_vhost > > endif > >-endif # ! CONFIG_RTE_BUILD_COMBINE_LIBS > >- > > ifeq ($(CONFIG_RTE_LIBRTE_PMD_PCAP),y) > > LDLIBS += -lpcap > > endif > >@@ -153,8 +145,6 @@ endif > > LDLIBS += --start-group > >-ifeq ($(CONFIG_RTE_BUILD_COMBINE_LIBS),n) > >- > > ifeq ($(CONFIG_RTE_LIBRTE_KVARGS),y) > > LDLIBS += -lrte_kvargs > > endif > >@@ -253,8 +243,6 @@ endif > > endif # plugins > >-endif # ! CONFIG_RTE_BUILD_COMBINE_LIBS > >- > > LDLIBS += $(EXECENV_LDLIBS) > > LDLIBS += --end-group > >diff --git a/mk/rte.lib.mk b/mk/rte.lib.mk > >index 0d7482d..d96101a 100644 > >--- a/mk/rte.lib.mk > >+++ b/mk/rte.lib.mk > >@@ -87,24 +87,6 @@ O_TO_S_DO = @set -e; \ > > $(O_TO_S) && \ > > echo $(O_TO_S_CMD) > $(call exe2cmd,$(@)) > >-ifeq ($(RTE_BUILD_SHARED_LIB),n) > >-O_TO_C = $(AR) crus $(LIB_ONE) $(OBJS-y) > >-O_TO_C_STR = $(subst ','\'',$(O_TO_C)) #'# fix syntax highlight > >-O_TO_C_DISP = $(if $(V),"$(O_TO_C_STR)"," AR_C $(@)") > >-O_TO_C_DO = @set -e; \ > >-$(lib_dir) \ > >-$(copy_obj) > >-else > >-O_TO_C = $(LD) -shared $(OBJS-y) -o $(LIB_ONE) > >-O_TO_C_STR = $(subst ','\'',$(O_TO_C)) #'# fix syntax highlight > >-O_TO_C_DISP = $(if $(V),"$(O_TO_C_STR)"," LD_C $(@)") > >-O_TO_C_DO = @set -e; \ > >-$(lib_dir) \ > >-$(copy_obj) > >-endif > >- > >-copy_obj = cp -f $(OBJS-y) $(RTE_OUTPUT)/build/lib; > >-lib_dir = [ -d $(RTE_OUTPUT)/lib ] || mkdir -p $(RTE_OUTPUT)/lib; > > -include .$(LIB).cmd > > # > >@@ -129,15 +111,6 @@ endif > > $(depfile_missing),\ > > $(depfile_newer)),\ > > $(O_TO_S_DO)) > >- > >-ifeq ($(RTE_BUILD_COMBINE_LIBS),y) > >-$(if $(or \ > >-$(file_missing),\ > >-$(call cmdline_changed,$(O_TO_C_STR)),\ > >-$(depfile_missing),\ > >-$(depfile_newer)),\ > >-$(O_TO_C_DO)) > >-endif > > else > > $(LIB): $(OBJ
[dpdk-dev] [PATCH v2 1/4] mk: Remove combined library and related options
On Wed, Mar 18, 2015 at 05:30:12PM +0200, Stefan Puiu wrote: > Hi, > > On Wed, Mar 18, 2015 at 2:59 PM, Thomas Monjalon > wrote: > > Hi Sergio, > > > > Thank you for explaining the situation. > > > > 2015-03-18 12:11, Gonzalez Monroy, Sergio: > >> Given that the patch to remove combined libraries is not welcome, I'll > >> try to explain the current situation so we can agree on the way forward. > >> > >> Currently we have build config option for shared libraries and combined > >> libraries. Thus, this results in four possible combinations when > >> building dpdk: > >> - not combined static > >> - not combined shared > >> - combined static > >> - combined shared > >> > >> The makefile rules/targets for combined are different than for not > >> combined. Thus, we currently have two different files for > >> archive/linking (rte.lib.mk and rte.sharelib.mk). > >> > >> Since having versioning, combined shared libraries build will be broken > >> the moment we add a versioned API, as we do not have a global version > >> map that we use when linking such library. > >> Also in my opinion, we would want to prevent users linking against a > >> combined libdpdk.so that may have different features built-in, with the > >> corresponding debugging difficulties when users > >> report different problems/errors. I think this would defeat many of the > >> advantages of using shared libraries. > >> > >> By removing the combined library build option, we would simplify the > >> build system with only two possible choices: > >> - static > >> - shared > > > > +1 > > I believe that simplification is the way go. > > > >> This would allow us to remove one file (rte.sharelib.mk) and have a > >> single file with archive/linking rules. > >> > >> For the convenience of linking against a single library instead of the > >> multiple dpdk libraries, there are a few ways to go around it: > >> - for combined static lib, we can either have a script to re-archive > >> all libraries into a single/combined library (ie. extract all archives > >> into one directory, the re-archive all objects into a combined library), > >>or use a linker script (ie. GROUP ( -lrte_eal -lrte_malloc ... ) ). > > Would the linker script be provided in the repository or would it be > the responsibility of people building against the DPDK? If I'd need to > make a linker script with the list of libraries to link against, might > as well put that list in my SConstruct / Makefile and be done with it. > So the "write your own linker script" and "just deal with separate > libraries" options don't seem that different to me. > just to level set, I think you're thinking of a linker script in too grand a scale. Technically what we're proposing is a linker script, but its literally a single line. If you want an example take a look at /usr/lib64/libc.so. that said, I think it makes more sense for the linker script in question to be part of the dpdk distribution so that the combined library name picks up new libraries as they are created. > Let me ask you something - I understand your concerns about > simplifying Makefiles and the concerns about versioning. How > significant is the "separate libs" use case? And especially the > "separate libs" in the current division of the code / libraries? I > counted about ~30 libs in 1.8.0 under build/lib. Are there people > using librte_eal without rte_malloc? Or rte_malloc without > rte_mempool? > Highly doubtful/impossible since they are explicitly dependent on one another. > I noticed that some examples I built ended up using --whole-archive > -lrte_eal -lrte_etc To me, --whole-archive is one way of saying > "we have lots of libraries with obscure dependencies", maybe reducing > the number of libs might also be a way to make the combined lib > unnecessary? I wouldn't bother with the combined lib if I had 3-4 libs > to link against instead of the number of libs needed now. > This isn't a bad suggestion. combining the low level malloc/mempool/eal libraries into a libdpdk_core probably makes sense. Not sure if doing it right now makes sense (this close to the release). But as a next release goal that seems reasonable. Neil
[dpdk-dev] [PATCH] hash: fix breaking strict-aliasing rules
Fix rte_hash_crc() function. Casting uint64_t pointer to uin32_t
may trigger a compiler warning about breaking strict-aliasing rules.
To avoid that, introduce a lookup table which is used to mask out
a remainder of data.
See issue #1, http://dpdk.org/ml/archives/dev/2015-March/015174.html
Signed-off-by: Yerden Zhumabekov
---
lib/librte_hash/rte_hash_crc.h | 31 +++
1 file changed, 15 insertions(+), 16 deletions(-)
diff --git a/lib/librte_hash/rte_hash_crc.h b/lib/librte_hash/rte_hash_crc.h
index 3dcd362..e81920f 100644
--- a/lib/librte_hash/rte_hash_crc.h
+++ b/lib/librte_hash/rte_hash_crc.h
@@ -323,6 +323,16 @@ static const uint32_t crc32c_tables[8][256] = {{
0xE54C35A1, 0xAC704886, 0x7734CFEF, 0x3E08B2C8, 0xC451B7CC, 0x8D6DCAEB,
0x56294D82, 0x1F1530A5
}};
+static const uint64_t odd_8byte_mask[] = {
+ 0x00FF,
+ 0x,
+ 0x00FF,
+ 0x,
+ 0x00FF,
+ 0x,
+ 0x00FF,
+};
+
#define CRC32_UPD(crc, n) \
(crc32c_tables[(n)][(crc) & 0xFF] ^ \
crc32c_tables[(n)-1][((crc) >> 8) & 0xFF])
@@ -535,38 +545,27 @@ static inline uint32_t
rte_hash_crc(const void *data, uint32_t data_len, uint32_t init_val)
{
unsigned i;
- uint64_t temp = 0;
+ uint64_t temp;
const uint64_t *p64 = (const uint64_t *)data;
for (i = 0; i < data_len / 8; i++) {
init_val = rte_hash_crc_8byte(*p64++, init_val);
}
- switch (7 - (data_len & 0x07)) {
+ i = 7 - (data_len & 0x07);
+ switch (i) {
case 0:
- temp |= (uint64_t) *((const uint8_t *)p64 + 6) << 48;
- /* Fallthrough */
case 1:
- temp |= (uint64_t) *((const uint8_t *)p64 + 5) << 40;
- /* Fallthrough */
case 2:
- temp |= (uint64_t) *((const uint8_t *)p64 + 4) << 32;
- temp |= *((const uint32_t *)p64);
+ temp = odd_8byte_mask[i] & *p64;
init_val = rte_hash_crc_8byte(temp, init_val);
break;
case 3:
- init_val = rte_hash_crc_4byte(*(const uint32_t *)p64, init_val);
- break;
case 4:
- temp |= *((const uint8_t *)p64 + 2) << 16;
- /* Fallthrough */
case 5:
- temp |= *((const uint8_t *)p64 + 1) << 8;
- /* Fallthrough */
case 6:
- temp |= *((const uint8_t *)p64);
+ temp = odd_8byte_mask[i] & *p64;
init_val = rte_hash_crc_4byte(temp, init_val);
- /* Fallthrough */
default:
break;
}
--
1.7.9.5
[dpdk-dev] [PATCHv2] EAL: rename rte_common_vect.h into arch/x86/rte_vect.h
Signed-off-by: Konstantin Ananyev
---
examples/l3fwd/main.c | 2 +-
lib/librte_acl/rte_acl_osdep.h | 2 +-
lib/librte_eal/common/Makefile | 1 -
.../common/include/arch/x86/rte_memcpy.h | 2 +-
lib/librte_eal/common/include/arch/x86/rte_vect.h | 128 +
lib/librte_eal/common/include/rte_common_vect.h| 128 -
lib/librte_lpm/rte_lpm.h | 2 +-
7 files changed, 132 insertions(+), 133 deletions(-)
create mode 100644 lib/librte_eal/common/include/arch/x86/rte_vect.h
delete mode 100644 lib/librte_eal/common/include/rte_common_vect.h
diff --git a/examples/l3fwd/main.c b/examples/l3fwd/main.c
index 3edb250..90e177f 100644
--- a/examples/l3fwd/main.c
+++ b/examples/l3fwd/main.c
@@ -43,7 +43,7 @@
#include
#include
-#include
+#include
#include
#include
#include
diff --git a/lib/librte_acl/rte_acl_osdep.h b/lib/librte_acl/rte_acl_osdep.h
index 6287c84..81fdefb 100644
--- a/lib/librte_acl/rte_acl_osdep.h
+++ b/lib/librte_acl/rte_acl_osdep.h
@@ -61,7 +61,7 @@
#define DIM(x) RTE_DIM(x)
#include
-#include
+#include
#include
#include
#include
diff --git a/lib/librte_eal/common/Makefile b/lib/librte_eal/common/Makefile
index cf961a7..3ea3bbf 100644
--- a/lib/librte_eal/common/Makefile
+++ b/lib/librte_eal/common/Makefile
@@ -39,7 +39,6 @@ INC += rte_rwlock.h rte_tailq.h rte_interrupts.h rte_alarm.h
INC += rte_string_fns.h rte_version.h
INC += rte_eal_memconfig.h rte_malloc_heap.h
INC += rte_hexdump.h rte_devargs.h rte_dev.h
-INC += rte_common_vect.h
INC += rte_pci_dev_feature_defs.h rte_pci_dev_features.h
ifeq ($(CONFIG_RTE_INSECURE_FUNCTION_WARNING),y)
diff --git a/lib/librte_eal/common/include/arch/x86/rte_memcpy.h
b/lib/librte_eal/common/include/arch/x86/rte_memcpy.h
index 82ea7c3..6a57426 100644
--- a/lib/librte_eal/common/include/arch/x86/rte_memcpy.h
+++ b/lib/librte_eal/common/include/arch/x86/rte_memcpy.h
@@ -43,7 +43,7 @@
#include
#include
#include
-#include
+#include
#ifdef __cplusplus
extern "C" {
diff --git a/lib/librte_eal/common/include/arch/x86/rte_vect.h
b/lib/librte_eal/common/include/arch/x86/rte_vect.h
new file mode 100644
index 000..d5bcdb9
--- /dev/null
+++ b/lib/librte_eal/common/include/arch/x86/rte_vect.h
@@ -0,0 +1,128 @@
+/*-
+ * BSD LICENSE
+ *
+ * Copyright(c) 2010-2015 Intel Corporation. All rights reserved.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * * Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef _RTE_VECT_H_
+#define _RTE_VECT_H_
+
+/**
+ * @file
+ *
+ * RTE SSE/AVX related header.
+ */
+
+#if (defined(__ICC) || (__GNUC__ == 4 && __GNUC_MINOR__ < 4))
+
+#ifdef __SSE__
+#include
+#endif
+
+#ifdef __SSE2__
+#include
+#endif
+
+#if defined(__SSE4_2__) || defined(__SSE4_1__)
+#include
+#endif
+
+#if defined(__AVX__)
+#include
+#endif
+
+#else
+
+#include
+
+#endif
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef __m128i xmm_t;
+
+#defineXMM_SIZE(sizeof(xmm_t))
+#defineXMM_MASK(XMM_SIZE - 1)
+
+typedef union rte_xmm {
+ xmm_tx;
+ uint8_t u8[XMM_SIZE / sizeof(uint8_t)];
+ uint16_t u16[XMM_SIZE / sizeof(uint16_t)];
+ uint32_t u32[XMM_SIZE / sizeof(uint32_t)];
+ uint64_t u64[XMM_SIZE / sizeof(uint64_t)];
+ double pd[XMM_SIZE / sizeof(double)];
+} rte_xmm_t;
+
+#ifdef __AVX__
+
+typedef
[dpdk-dev] Interactive/dynamic QoS scheduler
Hi, I'm currently working with the QoS api. I would like to know if it is possible to have interaction at runtime with the parameter applied to pipes, subport and queue. Looking at the Scheduler sample, I understand that all of these parameters are predetermined using profiles in the config file. What I would like to do is interact with the scheduler to change subport, pipe and tc rate or size with arbitrary value at runtime to have control over traffic. For example: The scheduler sample is running and at any given time I want a number of pipe to drop from 10Gbit to 4.5 Gbit and then put it back to 8Gbit. Profiles are not useful in this case because I would have to set a different profile for each value from rate 1bit to 10Gbit. Is it actually possible to do that and how? Thanks, Alexandre F.
[dpdk-dev] [dpdk=dev] [PATCH v8 0/3]: Add LRO support to ixgbe PMD
This series adds the missing flow for enabling the LRO in the ethdev and
adds a support for this feature in the ixgbe PMD. There is a big hope that this
initiative is going to be picked up by some Intel developer that would add the
LRO support
to other Intel PMDs.
The series starts with some cleanup work in the code the final patch (the
actual adding of
the LRO support) is going to touch/use/change. There are still quite a few
issues in the ixgbe
PMD code left but they have to be a matter of a different series and I've left
a few "TODO"
remarks in the code.
The LRO ("RSC" in Intel's context) PMD completion handling code follows the
same design as the
corresponding Linux and FreeBSD implementation: pass the aggregation's cluster
HEAD buffer to
the NEXTP entry of the software ring till EOP is met.
HW configuration follows the corresponding specs: this feature is supported
only by x540 and
82599 PF devices.
The feature has been tested with seastar TCP stack with the following
configuration on Tx side:
- MTU: 400B
- 100 concurrent TCP connections.
The results were:
- Without LRO: total throughput: 0.12Gbps, coefficient of variance: 1.41%
- With LRO:total throughput: 8.21Gbps, coefficient of variance: 0.59%
This is an almost factor 80 improvement.
New in v8:
- Fixed the structs naming: igb_xxx -> ixgbe_xxx (some leftovers in PATCH2).
- Took the RSC configuration code from ixgbe_dev_rx_init() into a separate
function - ixgbe_set_rsc().
- Added some missing macros for HW configuration.
- Styling adjustments:
- Functions names.
- Functions descriptions.
- Reworked the ixgbe_free_rsc_cluster() code to make it more readable.
- Kill the HEADER_SPLIT flow in ixgbe_set_rsc() since it's not supported by
ixgbe PMD.
New in v7:
- Free not-yet-completed RSC aggregations in rte_eth_dev_stop() flow.
- Fixed rx_bulk_alloc_allowed and rx_vec_allowed initialization:
- Don't set them to FALSE in rte_eth_dev_stop() flow - the following
rte_eth_dev_start() will need them.
- Reset them to TRUE in rte_eth_dev_configure() and not in a probe() flow.
This will ensure the proper behaviour if port is re-configured.
- Reset the sw_ring[].mbuf entry in a bulk allocation case.
This is needed for ixgbe_rx_queue_release_mbufs().
- _recv_pkts_lro(): added the missing memory barrier before RDT update in a
non-bulk allocation case.
- Don't allow RSC when device is configured in an SR-IOV mode.
New in v6:
- Fix of the typo in the "bug fixes" series that broke the compilation
caused a
minor change in this follow-up series.
New in v5:
- Split the series into "bug fixes" and "all the rest" so that the former
could be
integrated into a 2.0 release.
- Put the RTE_ETHDEV_HAS_LRO_SUPPORT definition at the beginning of
rte_ethdev.h.
- Removed the "TODO: Remove me" comment near RTE_ETHDEV_HAS_LRO_SUPPORT.
New in v4:
- Remove CONFIG_RTE_ETHDEV_LRO_SUPPORT from config/common_linuxapp.
- Define RTE_ETHDEV_HAS_LRO_SUPPORT in rte_ethdev.h.
- As a result of "ixgbe: check rxd number to avoid mbuf leak" (352078e8e)
Vector Rx
had to get the same treatment as Rx Bulk Alloc (see PATCH4 for more
details).
New in v3:
- ixgbe_rx_alloc_bufs(): Always reset refcnt of the buffers to 1. Otherwise
rte_pktmbuf_free()
won't free them.
New in v2:
- Removed rte_eth_dev_data.lro_bulk_alloc and added
ixgbe_hw.rx_bulk_alloc_allowed
instead.
- Unified the rx_pkt_bulk callback setting (a separate new patch).
- Fixed a few styling and spelling issues.
Vlad Zolotarov (3):
ixgbe: Cleanups
ixgbe: Code refactoring
ixgbe: Add LRO support
lib/librte_ether/rte_ethdev.h | 9 +-
lib/librte_net/rte_ip.h | 3 +
lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h | 6 +
lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 11 +
lib/librte_pmd_ixgbe/ixgbe_ethdev.h | 5 +
lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 766 +---
lib/librte_pmd_ixgbe/ixgbe_rxtx.h | 6 +
7 files changed, 737 insertions(+), 69 deletions(-)
--
2.1.0
[dpdk-dev] [dpdk=dev] [PATCH v8 1/3] ixgbe: Cleanups
- Removed the not needed casting.
- ixgbe_dev_rx_init(): shorten the lines by defining a local alias variable
to access
&dev->data->dev_conf.rxmode.
Signed-off-by: Vlad Zolotarov
---
New in v6:
- Fixed a compilation error caused by a patches recomposition during series
separation.
---
lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 29 +
1 file changed, 13 insertions(+), 16 deletions(-)
diff --git a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
index 42f0aa5..f17e8e1 100644
--- a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
+++ b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
@@ -1031,8 +1031,7 @@ ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq)
int diag, i;
/* allocate buffers in bulk directly into the S/W ring */
- alloc_idx = (uint16_t)(rxq->rx_free_trigger -
- (rxq->rx_free_thresh - 1));
+ alloc_idx = rxq->rx_free_trigger - (rxq->rx_free_thresh - 1);
rxep = &rxq->sw_ring[alloc_idx];
diag = rte_mempool_get_bulk(rxq->mb_pool, (void *)rxep,
rxq->rx_free_thresh);
@@ -1060,10 +1059,9 @@ ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq)
IXGBE_PCI_REG_WRITE(rxq->rdt_reg_addr, rxq->rx_free_trigger);
/* update state of internal queue structure */
- rxq->rx_free_trigger = (uint16_t)(rxq->rx_free_trigger +
- rxq->rx_free_thresh);
+ rxq->rx_free_trigger = rxq->rx_free_trigger + rxq->rx_free_thresh;
if (rxq->rx_free_trigger >= rxq->nb_rx_desc)
- rxq->rx_free_trigger = (uint16_t)(rxq->rx_free_thresh - 1);
+ rxq->rx_free_trigger = rxq->rx_free_thresh - 1;
/* no errors */
return 0;
@@ -3579,6 +3577,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
uint32_t rxcsum;
uint16_t buf_size;
uint16_t i;
+ struct rte_eth_rxmode *rx_conf = &dev->data->dev_conf.rxmode;
PMD_INIT_FUNC_TRACE();
hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
@@ -3601,7 +3600,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
* Configure CRC stripping, if any.
*/
hlreg0 = IXGBE_READ_REG(hw, IXGBE_HLREG0);
- if (dev->data->dev_conf.rxmode.hw_strip_crc)
+ if (rx_conf->hw_strip_crc)
hlreg0 |= IXGBE_HLREG0_RXCRCSTRP;
else
hlreg0 &= ~IXGBE_HLREG0_RXCRCSTRP;
@@ -3609,11 +3608,11 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
/*
* Configure jumbo frame support, if any.
*/
- if (dev->data->dev_conf.rxmode.jumbo_frame == 1) {
+ if (rx_conf->jumbo_frame == 1) {
hlreg0 |= IXGBE_HLREG0_JUMBOEN;
maxfrs = IXGBE_READ_REG(hw, IXGBE_MAXFRS);
maxfrs &= 0x;
- maxfrs |= (dev->data->dev_conf.rxmode.max_rx_pkt_len << 16);
+ maxfrs |= (rx_conf->max_rx_pkt_len << 16);
IXGBE_WRITE_REG(hw, IXGBE_MAXFRS, maxfrs);
} else
hlreg0 &= ~IXGBE_HLREG0_JUMBOEN;
@@ -3637,9 +3636,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
* Reset crc_len in case it was changed after queue setup by a
* call to configure.
*/
- rxq->crc_len = (uint8_t)
- ((dev->data->dev_conf.rxmode.hw_strip_crc) ? 0 :
- ETHER_CRC_LEN);
+ rxq->crc_len = rx_conf->hw_strip_crc ? 0 : ETHER_CRC_LEN;
/* Setup the Base and Length of the Rx Descriptor Rings */
bus_addr = rxq->rx_ring_phys_addr;
@@ -3657,7 +3654,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
/*
* Configure Header Split
*/
- if (dev->data->dev_conf.rxmode.header_split) {
+ if (rx_conf->header_split) {
if (hw->mac.type == ixgbe_mac_82599EB) {
/* Must setup the PSRTYPE register */
uint32_t psrtype;
@@ -3667,7 +3664,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
IXGBE_PSRTYPE_IPV6HDR;
IXGBE_WRITE_REG(hw,
IXGBE_PSRTYPE(rxq->reg_idx), psrtype);
}
- srrctl = ((dev->data->dev_conf.rxmode.split_hdr_size <<
+ srrctl = ((rx_conf->split_hdr_size <<
IXGBE_SRRCTL_BSIZEHDRSIZE_SHIFT) &
IXGBE_SRRCTL_BSIZEHDR_MASK);
srrctl |= IXGBE_SRRCTL_DESCTYPE_HDR_SPLIT_ALWAYS;
@@ -3701,7 +3698,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
dev->data->scattered_rx = 1;
}
- if (dev->data->dev_conf.rxmode.enable_scatter)
+ if (rx_conf->enable_scatter)
dev->data->scattered_rx = 1;
ixgbe_set_rx_f
[dpdk-dev] [dpdk=dev] [PATCH v8 2/3] ixgbe: Code refactoring
- ixgbe_rx_alloc_bufs():
- Reset the rte_mbuf fields only when requested.
- Take the RDT update out of the function.
- Add the stub when RTE_LIBRTE_IXGBE_RX_ALLOW_BULK_ALLOC is not defined.
- ixgbe_recv_scattered_pkts():
- Take the code that updates the fields of the cluster's HEAD buffer into
the inline function.
Signed-off-by: Vlad Zolotarov
---
New in v8:
- Fixed the structs naming: igb_xxx -> ixgbe_xxx
- Adjust a code style with the ixgbe PMD styling.
New in v3:
- ixgbe_rx_alloc_bufs(): Always reset refcnt of the buffers to 1.
---
lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 127 --
1 file changed, 82 insertions(+), 45 deletions(-)
diff --git a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
index f17e8e1..a08ae6a 100644
--- a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
+++ b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
@@ -1021,7 +1021,7 @@ ixgbe_rx_scan_hw_ring(struct ixgbe_rx_queue *rxq)
}
static inline int
-ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq)
+ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq, bool reset_mbuf)
{
volatile union ixgbe_adv_rx_desc *rxdp;
struct ixgbe_rx_entry *rxep;
@@ -1042,11 +1042,14 @@ ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq)
for (i = 0; i < rxq->rx_free_thresh; ++i) {
/* populate the static rte mbuf fields */
mb = rxep[i].mbuf;
+ if (reset_mbuf) {
+ mb->next = NULL;
+ mb->nb_segs = 1;
+ mb->port = rxq->port_id;
+ }
+
rte_mbuf_refcnt_set(mb, 1);
- mb->next = NULL;
mb->data_off = RTE_PKTMBUF_HEADROOM;
- mb->nb_segs = 1;
- mb->port = rxq->port_id;
/* populate the descriptors */
dma_addr = rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mb));
@@ -1054,10 +1057,6 @@ ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq)
rxdp[i].read.pkt_addr = dma_addr;
}
- /* update tail pointer */
- rte_wmb();
- IXGBE_PCI_REG_WRITE(rxq->rdt_reg_addr, rxq->rx_free_trigger);
-
/* update state of internal queue structure */
rxq->rx_free_trigger = rxq->rx_free_trigger + rxq->rx_free_thresh;
if (rxq->rx_free_trigger >= rxq->nb_rx_desc)
@@ -1109,7 +1108,9 @@ rx_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
/* if required, allocate new buffers to replenish descriptors */
if (rxq->rx_tail > rxq->rx_free_trigger) {
- if (ixgbe_rx_alloc_bufs(rxq) != 0) {
+ uint16_t cur_free_trigger = rxq->rx_free_trigger;
+
+ if (ixgbe_rx_alloc_bufs(rxq, true) != 0) {
int i, j;
PMD_RX_LOG(DEBUG, "RX mbuf alloc failed port_id=%u "
"queue_id=%u", (unsigned) rxq->port_id,
@@ -1129,6 +1130,10 @@ rx_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
return 0;
}
+
+ /* update tail pointer */
+ rte_wmb();
+ IXGBE_PCI_REG_WRITE(rxq->rdt_reg_addr, cur_free_trigger);
}
if (rxq->rx_tail >= rxq->nb_rx_desc)
@@ -1168,6 +1173,13 @@ ixgbe_recv_pkts_bulk_alloc(void *rx_queue, struct
rte_mbuf **rx_pkts,
return nb_rx;
}
+#else
+static inline int
+ixgbe_rx_alloc_bufs(__rte_unused struct ixgbe_rx_queue *rxq,
+ __rte_unused bool reset_mbuf)
+{
+ return -ENOMEM;
+}
#endif /* RTE_LIBRTE_IXGBE_RX_ALLOW_BULK_ALLOC */
uint16_t
@@ -1352,6 +1364,64 @@ ixgbe_recv_pkts(void *rx_queue, struct rte_mbuf
**rx_pkts,
return (nb_rx);
}
+/**
+ * Detect an RSC descriptor.
+ */
+static inline uint32_t
+ixgbe_rsc_count(union ixgbe_adv_rx_desc *rx)
+{
+ return (rte_le_to_cpu_32(rx->wb.lower.lo_dword.data) &
+ IXGBE_RXDADV_RSCCNT_MASK) >> IXGBE_RXDADV_RSCCNT_SHIFT;
+}
+
+/**
+ * ixgbe_fill_cluster_head_buf - fill the first mbuf of the returned packet
+ *
+ * Fill the following info in the HEAD buffer of the Rx cluster:
+ *- RX port identifier
+ *- hardware offload data, if any:
+ * - RSS flag & hash
+ * - IP checksum flag
+ * - VLAN TCI, if any
+ * - error flags
+ * @head HEAD of the packet cluster
+ * @desc HW descriptor to get data from
+ * @port_id Port ID of the Rx queue
+ */
+static inline void
+ixgbe_fill_cluster_head_buf(
+ struct rte_mbuf *head,
+ union ixgbe_adv_rx_desc *desc,
+ uint8_t port_id,
+ uint32_t staterr)
+{
+ uint32_t hlen_type_rss;
+ uint64_t pkt_flags;
+
+ head->port = port_id;
+
+ /*
+* The vlan_tci field is only valid when PKT_RX_VLAN_PKT is
+* set in the pkt_flags field.
+*/
+ head->vlan_tci = rte_le_to_cpu_16(desc->wb.upper.vlan);
+ hlen_type_rss = rte_le_to_cpu_32(desc->wb.lower.lo_dword.data);
+ pkt_f
[dpdk-dev] [dpdk=dev] [PATCH v8 3/3] ixgbe: Add LRO support
- Only x540 and 82599 devices support LRO.
- Add the appropriate HW configuration.
- Add RSC aware rx_pkt_burst() handlers:
- Implemented bulk allocation and non-bulk allocation versions.
- Add LRO-specific fields to rte_eth_rxmode, to rte_eth_dev_data
and to ixgbe_rx_queue.
- Use the appropriate handler when LRO is requested.
Signed-off-by: Vlad Zolotarov
---
New in v8:
- Took the RSC configuration code from ixgbe_dev_rx_init() into a separate
function - ixgbe_set_rsc().
- Added some missing macros for HW configuration.
- Styling adjustments:
- Functions names.
- Functions descriptions.
- Reworked the ixgbe_free_rsc_cluster() code to make it more readable.
- Kill the HEADER_SPLIT flow in ixgbe_set_rsc() since it's not supported by
ixgbe PMD.
New in v7:
- Free not-yet-completed RSC aggregations in rte_eth_dev_stop() flow.
- Reset the sw_ring[].mbuf entry in a bulk allocation case.
This is needed for ixgbe_rx_queue_release_mbufs().
- _recv_pkts_lro(): added the missing memory barrier before RDT update in a
non-bulk allocation case.
- Don't allow RSC when device is configured in an SR-IOV mode.
New in v5:
- Put the RTE_ETHDEV_HAS_LRO_SUPPORT definition at the beginning of
rte_ethdev.h.
- Removed the "TODO: Remove me" comment near RTE_ETHDEV_HAS_LRO_SUPPORT.
New in v4:
- Define RTE_ETHDEV_HAS_LRO_SUPPORT in rte_ethdev.h instead of
RTE_ETHDEV_LRO_SUPPORT defined in config/common_linuxapp.
New in v2:
- Removed rte_eth_dev_data.lro_bulk_alloc.
- Fixed a few styling and spelling issues.
---
lib/librte_ether/rte_ethdev.h | 9 +-
lib/librte_net/rte_ip.h | 3 +
lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h | 6 +
lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 11 +
lib/librte_pmd_ixgbe/ixgbe_ethdev.h | 5 +
lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 610 +++-
lib/librte_pmd_ixgbe/ixgbe_rxtx.h | 6 +
7 files changed, 642 insertions(+), 8 deletions(-)
diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
index 21aa359..61dc49a 100644
--- a/lib/librte_ether/rte_ethdev.h
+++ b/lib/librte_ether/rte_ethdev.h
@@ -172,6 +172,9 @@ extern "C" {
#include
+/* Use this macro to check if LRO API is supported */
+#define RTE_ETHDEV_HAS_LRO_SUPPORT
+
#include
#include
#include
@@ -320,14 +323,15 @@ struct rte_eth_rxmode {
enum rte_eth_rx_mq_mode mq_mode;
uint32_t max_rx_pkt_len; /**< Only used if jumbo_frame enabled. */
uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
- uint8_t header_split : 1, /**< Header Split enable. */
+ uint16_t header_split : 1, /**< Header Split enable. */
hw_ip_checksum : 1, /**< IP/UDP/TCP checksum offload enable.
*/
hw_vlan_filter : 1, /**< VLAN filter enable. */
hw_vlan_strip: 1, /**< VLAN strip enable. */
hw_vlan_extend : 1, /**< Extended VLAN enable. */
jumbo_frame : 1, /**< Jumbo Frame Receipt enable. */
hw_strip_crc : 1, /**< Enable CRC stripping by hardware. */
- enable_scatter : 1; /**< Enable scatter packets rx handler */
+ enable_scatter : 1, /**< Enable scatter packets rx handler */
+ enable_lro : 1; /**< Enable LRO */
};
/**
@@ -1515,6 +1519,7 @@ struct rte_eth_dev_data {
uint8_t port_id; /**< Device [external] port identifier. */
uint8_t promiscuous : 1, /**< RX promiscuous mode ON(1) / OFF(0). */
scattered_rx : 1, /**< RX of scattered packets is ON(1) /
OFF(0) */
+ lro : 1, /**< RX LRO is ON(1) / OFF(0) */
all_multicast : 1, /**< RX all multicast mode ON(1) / OFF(0). */
dev_started : 1; /**< Device state: STARTED(1) / STOPPED(0).
*/
};
diff --git a/lib/librte_net/rte_ip.h b/lib/librte_net/rte_ip.h
index 64935d9..74c9ced 100644
--- a/lib/librte_net/rte_ip.h
+++ b/lib/librte_net/rte_ip.h
@@ -110,6 +110,9 @@ struct ipv4_hdr {
(((c) & 0xff) << 8) | \
((d) & 0xff))
+/** Maximal IPv4 packet length (including a header) */
+#define IPV4_MAX_PKT_LEN65535
+
/** Internet header length mask for version_ihl field */
#define IPV4_HDR_IHL_MASK (0x0f)
/**
diff --git a/lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h
b/lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h
index 9a66370..4998627 100644
--- a/lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h
+++ b/lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h
@@ -234,8 +234,14 @@ struct ixgbe_thermal_sensor_data {
#define IXGBE_EITR(_i) (((_i) <= 23) ? (0x00820 + ((_i) * 4)) : \
(0x012300 + (((_i) - 24) * 4)))
#define IXGBE_EITR_ITR_INT_MASK0x0FF8
+#define IXGBE_EITR_ITR_INT_SHIFT 3
#define I
[dpdk-dev] [PATCH v8 0/3]: Add LRO support to ixgbe PMD
This series adds the missing flow for enabling the LRO in the ethdev and
adds a support for this feature in the ixgbe PMD. There is a big hope that this
initiative is going to be picked up by some Intel developer that would add the
LRO support
to other Intel PMDs.
The series starts with some cleanup work in the code the final patch (the
actual adding of
the LRO support) is going to touch/use/change. There are still quite a few
issues in the ixgbe
PMD code left but they have to be a matter of a different series and I've left
a few "TODO"
remarks in the code.
The LRO ("RSC" in Intel's context) PMD completion handling code follows the
same design as the
corresponding Linux and FreeBSD implementation: pass the aggregation's cluster
HEAD buffer to
the NEXTP entry of the software ring till EOP is met.
HW configuration follows the corresponding specs: this feature is supported
only by x540 and
82599 PF devices.
The feature has been tested with seastar TCP stack with the following
configuration on Tx side:
- MTU: 400B
- 100 concurrent TCP connections.
The results were:
- Without LRO: total throughput: 0.12Gbps, coefficient of variance: 1.41%
- With LRO:total throughput: 8.21Gbps, coefficient of variance: 0.59%
This is an almost factor 80 improvement.
New in v8:
- Fixed the structs naming: igb_xxx -> ixgbe_xxx (some leftovers in PATCH2).
- Took the RSC configuration code from ixgbe_dev_rx_init() into a separate
function - ixgbe_set_rsc().
- Added some missing macros for HW configuration.
- Styling adjustments:
- Functions names.
- Functions descriptions.
- Reworked the ixgbe_free_rsc_cluster() code to make it more readable.
- Kill the HEADER_SPLIT flow in ixgbe_set_rsc() since it's not supported by
ixgbe PMD.
New in v7:
- Free not-yet-completed RSC aggregations in rte_eth_dev_stop() flow.
- Fixed rx_bulk_alloc_allowed and rx_vec_allowed initialization:
- Don't set them to FALSE in rte_eth_dev_stop() flow - the following
rte_eth_dev_start() will need them.
- Reset them to TRUE in rte_eth_dev_configure() and not in a probe() flow.
This will ensure the proper behaviour if port is re-configured.
- Reset the sw_ring[].mbuf entry in a bulk allocation case.
This is needed for ixgbe_rx_queue_release_mbufs().
- _recv_pkts_lro(): added the missing memory barrier before RDT update in a
non-bulk allocation case.
- Don't allow RSC when device is configured in an SR-IOV mode.
New in v6:
- Fix of the typo in the "bug fixes" series that broke the compilation
caused a
minor change in this follow-up series.
New in v5:
- Split the series into "bug fixes" and "all the rest" so that the former
could be
integrated into a 2.0 release.
- Put the RTE_ETHDEV_HAS_LRO_SUPPORT definition at the beginning of
rte_ethdev.h.
- Removed the "TODO: Remove me" comment near RTE_ETHDEV_HAS_LRO_SUPPORT.
New in v4:
- Remove CONFIG_RTE_ETHDEV_LRO_SUPPORT from config/common_linuxapp.
- Define RTE_ETHDEV_HAS_LRO_SUPPORT in rte_ethdev.h.
- As a result of "ixgbe: check rxd number to avoid mbuf leak" (352078e8e)
Vector Rx
had to get the same treatment as Rx Bulk Alloc (see PATCH4 for more
details).
New in v3:
- ixgbe_rx_alloc_bufs(): Always reset refcnt of the buffers to 1. Otherwise
rte_pktmbuf_free()
won't free them.
New in v2:
- Removed rte_eth_dev_data.lro_bulk_alloc and added
ixgbe_hw.rx_bulk_alloc_allowed
instead.
- Unified the rx_pkt_bulk callback setting (a separate new patch).
- Fixed a few styling and spelling issues.
Vlad Zolotarov (3):
ixgbe: Cleanups
ixgbe: Code refactoring
ixgbe: Add LRO support
lib/librte_ether/rte_ethdev.h | 9 +-
lib/librte_net/rte_ip.h | 3 +
lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h | 6 +
lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 11 +
lib/librte_pmd_ixgbe/ixgbe_ethdev.h | 5 +
lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 766 +---
lib/librte_pmd_ixgbe/ixgbe_rxtx.h | 6 +
7 files changed, 737 insertions(+), 69 deletions(-)
--
2.1.0
[dpdk-dev] [PATCH v8 1/3] ixgbe: Cleanups
- Removed the not needed casting.
- ixgbe_dev_rx_init(): shorten the lines by defining a local alias variable
to access
&dev->data->dev_conf.rxmode.
Signed-off-by: Vlad Zolotarov
---
New in v6:
- Fixed a compilation error caused by a patches recomposition during series
separation.
---
lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 29 +
1 file changed, 13 insertions(+), 16 deletions(-)
diff --git a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
index 42f0aa5..f17e8e1 100644
--- a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
+++ b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
@@ -1031,8 +1031,7 @@ ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq)
int diag, i;
/* allocate buffers in bulk directly into the S/W ring */
- alloc_idx = (uint16_t)(rxq->rx_free_trigger -
- (rxq->rx_free_thresh - 1));
+ alloc_idx = rxq->rx_free_trigger - (rxq->rx_free_thresh - 1);
rxep = &rxq->sw_ring[alloc_idx];
diag = rte_mempool_get_bulk(rxq->mb_pool, (void *)rxep,
rxq->rx_free_thresh);
@@ -1060,10 +1059,9 @@ ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq)
IXGBE_PCI_REG_WRITE(rxq->rdt_reg_addr, rxq->rx_free_trigger);
/* update state of internal queue structure */
- rxq->rx_free_trigger = (uint16_t)(rxq->rx_free_trigger +
- rxq->rx_free_thresh);
+ rxq->rx_free_trigger = rxq->rx_free_trigger + rxq->rx_free_thresh;
if (rxq->rx_free_trigger >= rxq->nb_rx_desc)
- rxq->rx_free_trigger = (uint16_t)(rxq->rx_free_thresh - 1);
+ rxq->rx_free_trigger = rxq->rx_free_thresh - 1;
/* no errors */
return 0;
@@ -3579,6 +3577,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
uint32_t rxcsum;
uint16_t buf_size;
uint16_t i;
+ struct rte_eth_rxmode *rx_conf = &dev->data->dev_conf.rxmode;
PMD_INIT_FUNC_TRACE();
hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
@@ -3601,7 +3600,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
* Configure CRC stripping, if any.
*/
hlreg0 = IXGBE_READ_REG(hw, IXGBE_HLREG0);
- if (dev->data->dev_conf.rxmode.hw_strip_crc)
+ if (rx_conf->hw_strip_crc)
hlreg0 |= IXGBE_HLREG0_RXCRCSTRP;
else
hlreg0 &= ~IXGBE_HLREG0_RXCRCSTRP;
@@ -3609,11 +3608,11 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
/*
* Configure jumbo frame support, if any.
*/
- if (dev->data->dev_conf.rxmode.jumbo_frame == 1) {
+ if (rx_conf->jumbo_frame == 1) {
hlreg0 |= IXGBE_HLREG0_JUMBOEN;
maxfrs = IXGBE_READ_REG(hw, IXGBE_MAXFRS);
maxfrs &= 0x;
- maxfrs |= (dev->data->dev_conf.rxmode.max_rx_pkt_len << 16);
+ maxfrs |= (rx_conf->max_rx_pkt_len << 16);
IXGBE_WRITE_REG(hw, IXGBE_MAXFRS, maxfrs);
} else
hlreg0 &= ~IXGBE_HLREG0_JUMBOEN;
@@ -3637,9 +3636,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
* Reset crc_len in case it was changed after queue setup by a
* call to configure.
*/
- rxq->crc_len = (uint8_t)
- ((dev->data->dev_conf.rxmode.hw_strip_crc) ? 0 :
- ETHER_CRC_LEN);
+ rxq->crc_len = rx_conf->hw_strip_crc ? 0 : ETHER_CRC_LEN;
/* Setup the Base and Length of the Rx Descriptor Rings */
bus_addr = rxq->rx_ring_phys_addr;
@@ -3657,7 +3654,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
/*
* Configure Header Split
*/
- if (dev->data->dev_conf.rxmode.header_split) {
+ if (rx_conf->header_split) {
if (hw->mac.type == ixgbe_mac_82599EB) {
/* Must setup the PSRTYPE register */
uint32_t psrtype;
@@ -3667,7 +3664,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
IXGBE_PSRTYPE_IPV6HDR;
IXGBE_WRITE_REG(hw,
IXGBE_PSRTYPE(rxq->reg_idx), psrtype);
}
- srrctl = ((dev->data->dev_conf.rxmode.split_hdr_size <<
+ srrctl = ((rx_conf->split_hdr_size <<
IXGBE_SRRCTL_BSIZEHDRSIZE_SHIFT) &
IXGBE_SRRCTL_BSIZEHDR_MASK);
srrctl |= IXGBE_SRRCTL_DESCTYPE_HDR_SPLIT_ALWAYS;
@@ -3701,7 +3698,7 @@ ixgbe_dev_rx_init(struct rte_eth_dev *dev)
dev->data->scattered_rx = 1;
}
- if (dev->data->dev_conf.rxmode.enable_scatter)
+ if (rx_conf->enable_scatter)
dev->data->scattered_rx = 1;
ixgbe_set_rx_f
[dpdk-dev] [PATCH v8 2/3] ixgbe: Code refactoring
- ixgbe_rx_alloc_bufs():
- Reset the rte_mbuf fields only when requested.
- Take the RDT update out of the function.
- Add the stub when RTE_LIBRTE_IXGBE_RX_ALLOW_BULK_ALLOC is not defined.
- ixgbe_recv_scattered_pkts():
- Take the code that updates the fields of the cluster's HEAD buffer into
the inline function.
Signed-off-by: Vlad Zolotarov
---
New in v8:
- Fixed the structs naming: igb_xxx -> ixgbe_xxx
- Adjust a code style with the ixgbe PMD styling.
New in v3:
- ixgbe_rx_alloc_bufs(): Always reset refcnt of the buffers to 1.
---
lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 127 --
1 file changed, 82 insertions(+), 45 deletions(-)
diff --git a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
index f17e8e1..a08ae6a 100644
--- a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
+++ b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
@@ -1021,7 +1021,7 @@ ixgbe_rx_scan_hw_ring(struct ixgbe_rx_queue *rxq)
}
static inline int
-ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq)
+ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq, bool reset_mbuf)
{
volatile union ixgbe_adv_rx_desc *rxdp;
struct ixgbe_rx_entry *rxep;
@@ -1042,11 +1042,14 @@ ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq)
for (i = 0; i < rxq->rx_free_thresh; ++i) {
/* populate the static rte mbuf fields */
mb = rxep[i].mbuf;
+ if (reset_mbuf) {
+ mb->next = NULL;
+ mb->nb_segs = 1;
+ mb->port = rxq->port_id;
+ }
+
rte_mbuf_refcnt_set(mb, 1);
- mb->next = NULL;
mb->data_off = RTE_PKTMBUF_HEADROOM;
- mb->nb_segs = 1;
- mb->port = rxq->port_id;
/* populate the descriptors */
dma_addr = rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mb));
@@ -1054,10 +1057,6 @@ ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq)
rxdp[i].read.pkt_addr = dma_addr;
}
- /* update tail pointer */
- rte_wmb();
- IXGBE_PCI_REG_WRITE(rxq->rdt_reg_addr, rxq->rx_free_trigger);
-
/* update state of internal queue structure */
rxq->rx_free_trigger = rxq->rx_free_trigger + rxq->rx_free_thresh;
if (rxq->rx_free_trigger >= rxq->nb_rx_desc)
@@ -1109,7 +1108,9 @@ rx_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
/* if required, allocate new buffers to replenish descriptors */
if (rxq->rx_tail > rxq->rx_free_trigger) {
- if (ixgbe_rx_alloc_bufs(rxq) != 0) {
+ uint16_t cur_free_trigger = rxq->rx_free_trigger;
+
+ if (ixgbe_rx_alloc_bufs(rxq, true) != 0) {
int i, j;
PMD_RX_LOG(DEBUG, "RX mbuf alloc failed port_id=%u "
"queue_id=%u", (unsigned) rxq->port_id,
@@ -1129,6 +1130,10 @@ rx_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
return 0;
}
+
+ /* update tail pointer */
+ rte_wmb();
+ IXGBE_PCI_REG_WRITE(rxq->rdt_reg_addr, cur_free_trigger);
}
if (rxq->rx_tail >= rxq->nb_rx_desc)
@@ -1168,6 +1173,13 @@ ixgbe_recv_pkts_bulk_alloc(void *rx_queue, struct
rte_mbuf **rx_pkts,
return nb_rx;
}
+#else
+static inline int
+ixgbe_rx_alloc_bufs(__rte_unused struct ixgbe_rx_queue *rxq,
+ __rte_unused bool reset_mbuf)
+{
+ return -ENOMEM;
+}
#endif /* RTE_LIBRTE_IXGBE_RX_ALLOW_BULK_ALLOC */
uint16_t
@@ -1352,6 +1364,64 @@ ixgbe_recv_pkts(void *rx_queue, struct rte_mbuf
**rx_pkts,
return (nb_rx);
}
+/**
+ * Detect an RSC descriptor.
+ */
+static inline uint32_t
+ixgbe_rsc_count(union ixgbe_adv_rx_desc *rx)
+{
+ return (rte_le_to_cpu_32(rx->wb.lower.lo_dword.data) &
+ IXGBE_RXDADV_RSCCNT_MASK) >> IXGBE_RXDADV_RSCCNT_SHIFT;
+}
+
+/**
+ * ixgbe_fill_cluster_head_buf - fill the first mbuf of the returned packet
+ *
+ * Fill the following info in the HEAD buffer of the Rx cluster:
+ *- RX port identifier
+ *- hardware offload data, if any:
+ * - RSS flag & hash
+ * - IP checksum flag
+ * - VLAN TCI, if any
+ * - error flags
+ * @head HEAD of the packet cluster
+ * @desc HW descriptor to get data from
+ * @port_id Port ID of the Rx queue
+ */
+static inline void
+ixgbe_fill_cluster_head_buf(
+ struct rte_mbuf *head,
+ union ixgbe_adv_rx_desc *desc,
+ uint8_t port_id,
+ uint32_t staterr)
+{
+ uint32_t hlen_type_rss;
+ uint64_t pkt_flags;
+
+ head->port = port_id;
+
+ /*
+* The vlan_tci field is only valid when PKT_RX_VLAN_PKT is
+* set in the pkt_flags field.
+*/
+ head->vlan_tci = rte_le_to_cpu_16(desc->wb.upper.vlan);
+ hlen_type_rss = rte_le_to_cpu_32(desc->wb.lower.lo_dword.data);
+ pkt_f
[dpdk-dev] [PATCH v8 3/3] ixgbe: Add LRO support
- Only x540 and 82599 devices support LRO.
- Add the appropriate HW configuration.
- Add RSC aware rx_pkt_burst() handlers:
- Implemented bulk allocation and non-bulk allocation versions.
- Add LRO-specific fields to rte_eth_rxmode, to rte_eth_dev_data
and to ixgbe_rx_queue.
- Use the appropriate handler when LRO is requested.
Signed-off-by: Vlad Zolotarov
---
New in v8:
- Took the RSC configuration code from ixgbe_dev_rx_init() into a separate
function - ixgbe_set_rsc().
- Added some missing macros for HW configuration.
- Styling adjustments:
- Functions names.
- Functions descriptions.
- Reworked the ixgbe_free_rsc_cluster() code to make it more readable.
- Kill the HEADER_SPLIT flow in ixgbe_set_rsc() since it's not supported by
ixgbe PMD.
New in v7:
- Free not-yet-completed RSC aggregations in rte_eth_dev_stop() flow.
- Reset the sw_ring[].mbuf entry in a bulk allocation case.
This is needed for ixgbe_rx_queue_release_mbufs().
- _recv_pkts_lro(): added the missing memory barrier before RDT update in a
non-bulk allocation case.
- Don't allow RSC when device is configured in an SR-IOV mode.
New in v5:
- Put the RTE_ETHDEV_HAS_LRO_SUPPORT definition at the beginning of
rte_ethdev.h.
- Removed the "TODO: Remove me" comment near RTE_ETHDEV_HAS_LRO_SUPPORT.
New in v4:
- Define RTE_ETHDEV_HAS_LRO_SUPPORT in rte_ethdev.h instead of
RTE_ETHDEV_LRO_SUPPORT defined in config/common_linuxapp.
New in v2:
- Removed rte_eth_dev_data.lro_bulk_alloc.
- Fixed a few styling and spelling issues.
---
lib/librte_ether/rte_ethdev.h | 9 +-
lib/librte_net/rte_ip.h | 3 +
lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h | 6 +
lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 11 +
lib/librte_pmd_ixgbe/ixgbe_ethdev.h | 5 +
lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 610 +++-
lib/librte_pmd_ixgbe/ixgbe_rxtx.h | 6 +
7 files changed, 642 insertions(+), 8 deletions(-)
diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
index 21aa359..61dc49a 100644
--- a/lib/librte_ether/rte_ethdev.h
+++ b/lib/librte_ether/rte_ethdev.h
@@ -172,6 +172,9 @@ extern "C" {
#include
+/* Use this macro to check if LRO API is supported */
+#define RTE_ETHDEV_HAS_LRO_SUPPORT
+
#include
#include
#include
@@ -320,14 +323,15 @@ struct rte_eth_rxmode {
enum rte_eth_rx_mq_mode mq_mode;
uint32_t max_rx_pkt_len; /**< Only used if jumbo_frame enabled. */
uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
- uint8_t header_split : 1, /**< Header Split enable. */
+ uint16_t header_split : 1, /**< Header Split enable. */
hw_ip_checksum : 1, /**< IP/UDP/TCP checksum offload enable.
*/
hw_vlan_filter : 1, /**< VLAN filter enable. */
hw_vlan_strip: 1, /**< VLAN strip enable. */
hw_vlan_extend : 1, /**< Extended VLAN enable. */
jumbo_frame : 1, /**< Jumbo Frame Receipt enable. */
hw_strip_crc : 1, /**< Enable CRC stripping by hardware. */
- enable_scatter : 1; /**< Enable scatter packets rx handler */
+ enable_scatter : 1, /**< Enable scatter packets rx handler */
+ enable_lro : 1; /**< Enable LRO */
};
/**
@@ -1515,6 +1519,7 @@ struct rte_eth_dev_data {
uint8_t port_id; /**< Device [external] port identifier. */
uint8_t promiscuous : 1, /**< RX promiscuous mode ON(1) / OFF(0). */
scattered_rx : 1, /**< RX of scattered packets is ON(1) /
OFF(0) */
+ lro : 1, /**< RX LRO is ON(1) / OFF(0) */
all_multicast : 1, /**< RX all multicast mode ON(1) / OFF(0). */
dev_started : 1; /**< Device state: STARTED(1) / STOPPED(0).
*/
};
diff --git a/lib/librte_net/rte_ip.h b/lib/librte_net/rte_ip.h
index 64935d9..74c9ced 100644
--- a/lib/librte_net/rte_ip.h
+++ b/lib/librte_net/rte_ip.h
@@ -110,6 +110,9 @@ struct ipv4_hdr {
(((c) & 0xff) << 8) | \
((d) & 0xff))
+/** Maximal IPv4 packet length (including a header) */
+#define IPV4_MAX_PKT_LEN65535
+
/** Internet header length mask for version_ihl field */
#define IPV4_HDR_IHL_MASK (0x0f)
/**
diff --git a/lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h
b/lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h
index 9a66370..4998627 100644
--- a/lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h
+++ b/lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h
@@ -234,8 +234,14 @@ struct ixgbe_thermal_sensor_data {
#define IXGBE_EITR(_i) (((_i) <= 23) ? (0x00820 + ((_i) * 4)) : \
(0x012300 + (((_i) - 24) * 4)))
#define IXGBE_EITR_ITR_INT_MASK0x0FF8
+#define IXGBE_EITR_ITR_INT_SHIFT 3
#define I
[dpdk-dev] [dpdk=dev] [PATCH v8 0/3]: Add LRO support to ixgbe PMD
There was a typo in a format-patch command - pls., ignore the whole
series. I'm respinning it with the proper subject.
On 03/18/15 19:48, Vlad Zolotarov wrote:
> This series adds the missing flow for enabling the LRO in the ethdev and
> adds a support for this feature in the ixgbe PMD. There is a big hope that
> this
> initiative is going to be picked up by some Intel developer that would add
> the LRO support
> to other Intel PMDs.
>
> The series starts with some cleanup work in the code the final patch (the
> actual adding of
> the LRO support) is going to touch/use/change. There are still quite a few
> issues in the ixgbe
> PMD code left but they have to be a matter of a different series and I've
> left a few "TODO"
> remarks in the code.
>
> The LRO ("RSC" in Intel's context) PMD completion handling code follows the
> same design as the
> corresponding Linux and FreeBSD implementation: pass the aggregation's
> cluster HEAD buffer to
> the NEXTP entry of the software ring till EOP is met.
>
> HW configuration follows the corresponding specs: this feature is supported
> only by x540 and
> 82599 PF devices.
>
> The feature has been tested with seastar TCP stack with the following
> configuration on Tx side:
> - MTU: 400B
> - 100 concurrent TCP connections.
>
> The results were:
> - Without LRO: total throughput: 0.12Gbps, coefficient of variance: 1.41%
> - With LRO:total throughput: 8.21Gbps, coefficient of variance: 0.59%
>
> This is an almost factor 80 improvement.
>
> New in v8:
> - Fixed the structs naming: igb_xxx -> ixgbe_xxx (some leftovers in
> PATCH2).
> - Took the RSC configuration code from ixgbe_dev_rx_init() into a separate
> function - ixgbe_set_rsc().
> - Added some missing macros for HW configuration.
> - Styling adjustments:
>- Functions names.
>- Functions descriptions.
> - Reworked the ixgbe_free_rsc_cluster() code to make it more readable.
> - Kill the HEADER_SPLIT flow in ixgbe_set_rsc() since it's not supported
> by
> ixgbe PMD.
>
> New in v7:
> - Free not-yet-completed RSC aggregations in rte_eth_dev_stop() flow.
> - Fixed rx_bulk_alloc_allowed and rx_vec_allowed initialization:
>- Don't set them to FALSE in rte_eth_dev_stop() flow - the following
> rte_eth_dev_start() will need them.
>- Reset them to TRUE in rte_eth_dev_configure() and not in a probe()
> flow.
> This will ensure the proper behaviour if port is re-configured.
> - Reset the sw_ring[].mbuf entry in a bulk allocation case.
> This is needed for ixgbe_rx_queue_release_mbufs().
> - _recv_pkts_lro(): added the missing memory barrier before RDT update in
> a
> non-bulk allocation case.
> - Don't allow RSC when device is configured in an SR-IOV mode.
>
> New in v6:
> - Fix of the typo in the "bug fixes" series that broke the compilation
> caused a
> minor change in this follow-up series.
>
> New in v5:
> - Split the series into "bug fixes" and "all the rest" so that the former
> could be
> integrated into a 2.0 release.
> - Put the RTE_ETHDEV_HAS_LRO_SUPPORT definition at the beginning of
> rte_ethdev.h.
> - Removed the "TODO: Remove me" comment near RTE_ETHDEV_HAS_LRO_SUPPORT.
>
> New in v4:
> - Remove CONFIG_RTE_ETHDEV_LRO_SUPPORT from config/common_linuxapp.
> - Define RTE_ETHDEV_HAS_LRO_SUPPORT in rte_ethdev.h.
> - As a result of "ixgbe: check rxd number to avoid mbuf leak" (352078e8e)
> Vector Rx
> had to get the same treatment as Rx Bulk Alloc (see PATCH4 for more
> details).
>
> New in v3:
> - ixgbe_rx_alloc_bufs(): Always reset refcnt of the buffers to 1.
> Otherwise rte_pktmbuf_free()
> won't free them.
>
> New in v2:
> - Removed rte_eth_dev_data.lro_bulk_alloc and added
> ixgbe_hw.rx_bulk_alloc_allowed
> instead.
> - Unified the rx_pkt_bulk callback setting (a separate new patch).
> - Fixed a few styling and spelling issues.
>
>
> Vlad Zolotarov (3):
>ixgbe: Cleanups
>ixgbe: Code refactoring
>ixgbe: Add LRO support
>
> lib/librte_ether/rte_ethdev.h | 9 +-
> lib/librte_net/rte_ip.h | 3 +
> lib/librte_pmd_ixgbe/ixgbe/ixgbe_type.h | 6 +
> lib/librte_pmd_ixgbe/ixgbe_ethdev.c | 11 +
> lib/librte_pmd_ixgbe/ixgbe_ethdev.h | 5 +
> lib/librte_pmd_ixgbe/ixgbe_rxtx.c | 766
> +---
> lib/librte_pmd_ixgbe/ixgbe_rxtx.h | 6 +
> 7 files changed, 737 insertions(+), 69 deletions(-)
>
[dpdk-dev] "FATAL: Module uio not found" for dpdk usage on Amazon AWS
When I tried to install dpdk on Amazon AWS ubuntu based linux, I see uio module error. Can anyone help me to how to load uio module on ubuntu linux so that IGB UIO module. ubuntu at ip-172-31-29-51:~$ uname --allLinux ip-172-31-29-51 3.13.0-44-generic #73-Ubuntu SMP Tue Dec 16 00:22:43 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux ubuntu at ip-172-31-29-51:~$ uname -r3.13.0-44-genericubuntu at ip-172-31-29-51:~$? root at ip-172-31-29-51:/home/pfe/riot/dpdk# ./tools/setup.sh?--?RTE_SDK exported as /home/pfe/riot/dpdk?Step 1: Select the DPDK environment to build--[1] i686-native-linuxapp-gcc[2] i686-native-linuxapp-icc[3] x86_64-default-linuxapp-gcc[4] x86_64-ivshmem-linuxapp-gcc[5] x86_64-ivshmem-linuxapp-icc[6] x86_64-native-bsdapp-gcc[7] x86_64-native-linuxapp-gcc[8] x86_64-native-linuxapp-icc --?Step 2: Setup linuxapp environment--[9] Insert IGB UIO module[10] Insert VFIO module[11] Insert KNI module[12] Setup hugepage mappings for non-NUMA systems[13] Setup hugepage mappings for NUMA systems[14] Display current Ethernet device settings[15] Bind Ethernet device to IGB UIO module[16] Bind Ethernet device to VFIO module[17] Setup VFIO permissions --?Step 3: Run test application for linuxapp environment--[18] Run test application ($RTE_TARGET/app/test)[19] Run testpmd application in interactive mode ($RTE_TARGET/app/testpmd) --?Step 4: Other tools--[20] List hugepage info from /proc/meminfo --?Step 5: Uninstall and system cleanup--[21] Uninstall all targets[22] Unbind NICs from IGB UIO driver[23] Remove IGB UIO module[24] Remove VFIO module[25] Remove KNI module[26] Remove hugepage mappings [27] Exit Script Option: 9 Unloading any existing DPDK UIO moduleLoading uio modulemodprobe: FATAL: Module uio not found.Loading DPDK UIO moduleinsmod: ERROR: could not insert module /home/pfe/riot/dpdk/x86_64-default-linuxapp-gcc/kmod/igb_uio.ko: Unknown symbol in module## ERROR: Could not load kmod/igb_uio.ko. Press enter to continue ...root at ip-172-31-29-51:/home/pfe/riot/dpdk# modinfo uiomodinfo: ERROR: Module uio not found.root at ip-172-31-29-51:/home/pfe/riot/dpdk# insmod uioinsmod: ERROR: could not load module uio: No such file or directoryroot at ip-172-31-29-51:/home/pfe/riot/dpdk# modprobe uiomodprobe: FATAL: Module uio not found.root at ip-172-31-29-51:/home/pfe/riot/dpdk#?
[dpdk-dev] [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added + unittest
From: "vadim.suraev at gmail.com"
This patch adds mbuf bulk allocation/freeing functions and unittest
Signed-off-by: Vadim Suraev
---
New in v2:
- function rte_pktmbuf_alloc_bulk added
- function rte_pktmbuf_bulk_free added
- function rte_pktmbuf_free_chain added
- applied reviewers' comments
app/test/test_mbuf.c | 94 +++-
lib/librte_mbuf/rte_mbuf.h | 91 ++
2 files changed, 184 insertions(+), 1 deletion(-)
diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c
index 1ff66cb..b20c6a4 100644
--- a/app/test/test_mbuf.c
+++ b/app/test/test_mbuf.c
@@ -77,6 +77,7 @@
#define REFCNT_RING_SIZE(REFCNT_MBUF_NUM * REFCNT_MAX_REF)
#define MAKE_STRING(x) # x
+#define MBUF_POOL_LOCAL_CACHE_SIZE 32
static struct rte_mempool *pktmbuf_pool = NULL;
@@ -405,6 +406,84 @@ test_pktmbuf_pool(void)
return ret;
}
+/* test pktmbuf bulk allocation and freeing
+*/
+static int
+test_pktmbuf_pool_bulk(void)
+{
+ unsigned i;
+ /* size of mempool - size of local cache, otherwise may fail */
+ unsigned mbufs_to_allocate = NB_MBUF - MBUF_POOL_LOCAL_CACHE_SIZE;
+ struct rte_mbuf *m[mbufs_to_allocate];
+ int ret = 0;
+ unsigned mbuf_count_before_allocation = rte_mempool_count(pktmbuf_pool);
+
+ for (i = 0; i < mbufs_to_allocate; i++)
+ m[i] = NULL;
+ /* alloc NB_MBUF-MBUF_POOL_LOCAL_CACHE_SIZE mbufs */
+ ret = rte_pktmbuf_alloc_bulk(pktmbuf_pool, m, mbufs_to_allocate);
+ if (ret) {
+ printf("cannot allocate %d mbufs bulk mempool_cnt=%d ret=%d\n",
+ mbufs_to_allocate,
+ rte_mempool_count(pktmbuf_pool),
+ ret);
+ return -1;
+ }
+ if ((rte_mempool_count(pktmbuf_pool) + mbufs_to_allocate) !=
+ mbuf_count_before_allocation) {
+ printf("mempool count %d + allocated %d != initial %d\n",
+ rte_mempool_count(pktmbuf_pool),
+ mbufs_to_allocate,
+ mbuf_count_before_allocation);
+ return -1;
+ }
+ /* free them */
+ rte_pktmbuf_bulk_free(m, mbufs_to_allocate);
+
+ if (rte_mempool_count(pktmbuf_pool) != mbuf_count_before_allocation) {
+ printf("mempool count %d != initial %d\n",
+ rte_mempool_count(pktmbuf_pool),
+ mbuf_count_before_allocation);
+ return -1;
+ }
+ for (i = 0; i < mbufs_to_allocate; i++)
+ m[i] = NULL;
+
+ /* alloc NB_MBUF-MBUF_POOL_LOCAL_CACHE_SIZE mbufs */
+ ret = rte_pktmbuf_alloc_bulk(pktmbuf_pool, m, mbufs_to_allocate);
+ if (ret) {
+ printf("cannot allocate %d mbufs bulk mempool_cnt=%d ret=%d\n",
+ mbufs_to_allocate,
+ rte_mempool_count(pktmbuf_pool),
+ ret);
+ return -1;
+ }
+ if ((rte_mempool_count(pktmbuf_pool) + mbufs_to_allocate) !=
+ mbuf_count_before_allocation) {
+ printf("mempool count %d + allocated %d != initial %d\n",
+ rte_mempool_count(pktmbuf_pool),
+ mbufs_to_allocate,
+ mbuf_count_before_allocation);
+ return -1;
+ }
+
+ /* chain it */
+ for (i = 0; i < mbufs_to_allocate - 1; i++) {
+ m[i]->next = m[i + 1];
+ m[0]->nb_segs++;
+ }
+ /* free them */
+ rte_pktmbuf_free_chain(m[0]);
+
+ if (rte_mempool_count(pktmbuf_pool) != mbuf_count_before_allocation) {
+ printf("mempool count %d != initial %d\n",
+ rte_mempool_count(pktmbuf_pool),
+ mbuf_count_before_allocation);
+ return -1;
+ }
+ return ret;
+}
+
/*
* test that the pointer to the data on a packet mbuf is set properly
*/
@@ -766,7 +845,8 @@ test_mbuf(void)
if (pktmbuf_pool == NULL) {
pktmbuf_pool =
rte_mempool_create("test_pktmbuf_pool", NB_MBUF,
- MBUF_SIZE, 32,
+ MBUF_SIZE,
+ MBUF_POOL_LOCAL_CACHE_SIZE,
sizeof(struct
rte_pktmbuf_pool_private),
rte_pktmbuf_pool_init, NULL,
rte_pktmbuf_init, NULL,
@@ -790,6 +870,18 @@ test_mbuf(void)
return -1;
}
+ /* test bulk allocation and freeing */
+ if (test_pktmbuf_pool_bulk() < 0) {
+ printf("test_pktmbuf_pool_bulk() failed\n");
+ return -1;
+ }
+
+ /* once again to ensure al
[dpdk-dev] [PATCH v2] rte_mbuf: mbuf bulk alloc/free functions added + unittest
On Wed, Mar 18, 2015 at 10:21:18PM +0200, vadim.suraev at gmail.com wrote:
> From: "vadim.suraev at gmail.com"
>
> This patch adds mbuf bulk allocation/freeing functions and unittest
>
> Signed-off-by: Vadim Suraev
>
> ---
> New in v2:
> - function rte_pktmbuf_alloc_bulk added
> - function rte_pktmbuf_bulk_free added
> - function rte_pktmbuf_free_chain added
> - applied reviewers' comments
>
> app/test/test_mbuf.c | 94
> +++-
> lib/librte_mbuf/rte_mbuf.h | 91 ++
> 2 files changed, 184 insertions(+), 1 deletion(-)
>
> diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c
> index 1ff66cb..b20c6a4 100644
> --- a/app/test/test_mbuf.c
> +++ b/app/test/test_mbuf.c
> @@ -77,6 +77,7 @@
> #define REFCNT_RING_SIZE(REFCNT_MBUF_NUM * REFCNT_MAX_REF)
>
> #define MAKE_STRING(x) # x
> +#define MBUF_POOL_LOCAL_CACHE_SIZE 32
>
> static struct rte_mempool *pktmbuf_pool = NULL;
>
> @@ -405,6 +406,84 @@ test_pktmbuf_pool(void)
> return ret;
> }
>
>
> diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
> index 17ba791..fabeae2 100644
> --- a/lib/librte_mbuf/rte_mbuf.h
> +++ b/lib/librte_mbuf/rte_mbuf.h
> @@ -825,6 +825,97 @@ static inline void rte_pktmbuf_free(struct rte_mbuf *m)
> }
>
> /**
> + * Allocate a bulk of mbufs, initiate refcnt and resets
> + *
> + * @param pool
> + *memory pool to allocate from
> + * @param mbufs
> + *Array of pointers to mbuf
> + * @param count
> + *Array size
> + */
> +static inline int rte_pktmbuf_alloc_bulk(struct rte_mempool *pool,
> + struct rte_mbuf **mbufs,
> + unsigned count)
> +{
> + unsigned idx;
> + int rc = 0;
> +
> + rc = rte_mempool_get_bulk(pool, (void **)mbufs, count);
> + if (unlikely(rc))
> + return rc;
> +
> + for (idx = 0; idx < count; idx++) {
> + RTE_MBUF_ASSERT(rte_mbuf_refcnt_read(mbufs[idx]) == 0);
> + rte_mbuf_refcnt_set(mbufs[idx], 1);
> + rte_pktmbuf_reset(mbufs[idx]);
> + }
> + return rc;
> +}
> +
> +/**
> + * Free a bulk of mbufs into its original mempool.
> + * This function assumes:
> + * - refcnt equals 1
> + * - mbufs are direct
> + * - all mbufs must belong to the same mempool
> + *
> + * @param mbufs
> + *Array of pointers to mbuf
> + * @param count
> + *Array size
> + */
> +static inline void rte_pktmbuf_bulk_free(struct rte_mbuf **mbufs,
> + unsigned count)
> +{
> + unsigned idx;
> +
> + RTE_MBUF_ASSERT(count > 0);
> +
> + for (idx = 0; idx < count; idx++) {
> + RTE_MBUF_ASSERT(mbufs[idx]->pool == mbufs[0]->pool);
> + RTE_MBUF_ASSERT(rte_mbuf_refcnt_read(mbufs[idx]) == 1);
> + rte_mbuf_refcnt_set(mbufs[idx], 0);
This is really a misuse of the API. The entire point of reference counting is
to know when an mbuf has no more references and can be freed. By forcing all
the reference counts to zero here, you allow the refcnt infrastructure to be
circumvented, causing memory leaks.
I think what you need to do here is enhance the underlying pktmbuf interface
such that an rte_mbuf structure has a destructor method association with it
which is called when its refcnt reaches zero. That way the
rte_pktmbuf_bulk_free function can just decrement the refcnt on each
mbuf_structure, and the pool as a whole can be returned when the destructor
function discovers that all mbufs in that bulk pool are freed.
Neil
[dpdk-dev] [PATCH] maintainers: claim pcap pmd library
Signed-off-by: Nicol?s Pernas Maradei --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index bef7f59..4c780db 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -247,6 +247,7 @@ F: lib/librte_pmd_vmxnet3/ F: doc/guides/prog_guide/poll_mode_drv_paravirtual_vmxnets_nic.rst PCAP PMD +M: Nicol?s Pernas Maradei F: lib/librte_pmd_pcap/ F: doc/guides/prog_guide/libpcap_ring_based_poll_mode_drv.rst -- 1.9.1
[dpdk-dev] "FATAL: Module uio not found" for dpdk usage on Amazon AWS
On Wed, Mar 18, 2015 at 06:41:05PM +, Naveen Gamini wrote: > When I tried to install dpdk on Amazon AWS ubuntu based linux, I see uio > module error. Can anyone help me to how to load uio module on ubuntu linux > so that IGB UIO module. Amazon Cloud Images usually include minimized kernel packages with Cloud drivers only. You usually have to fix it by reinstalling the server kernel which has the full driver set. On Ubuntu these two packages will usually fix it for me on minimal Ubuntu Cloud images / vagrant images / etc. linux-generic linux-headers-generic Matthew.
