d/numam-dpdk
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
numam-dpdk/examples/l3fwd/l3fwd_em_hlm_sse.h

48 lines
1.3 KiB
C
Raw Normal View History

examples: use SPDX tag for Intel copyright files Replace the BSD license header with the SPDX tag for files with only an Intel copyright on them. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
2017-12-19 15:49:02 +00:00
/* SPDX-License-Identifier: BSD-3-Clause
* Copyright(c) 2016 Intel Corporation
examples/l3fwd: rework exact-match Current implementation of Exact-Match uses different execution path than for LPM. Unifying them allows to reuse big part of LPM code and sightly increase performance of Exact-Match. Main changes: ------------- * Packet classification stage is separated from the rest of path for both LPM and EM. * Packet processing, modifying and transmit part is the same for LPM and EM and mostly based on the current LPM implementation. * Shared code is moved to the common file "l3fwd_sse.h". * While sequential packet classification in EM path, seems to be faster than using multi hash lookup, used before, it is used by default. Old implementation is moved to the file l3fwd_em_hlm_sse.h and can be enabled with HASH_LOOKUP_MULTI global define in compilation time. Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2016-02-29 11:33:07 +01:00
*/
#ifndef __L3FWD_EM_HLM_SSE_H__
#define __L3FWD_EM_HLM_SSE_H__
#include "l3fwd_sse.h"
eal: introduce macro for always inline Different drivers use internal macros like force_inline for compiler always inline feature. Standardizing it through __rte_always_inline macro. Verified the change by comparing the output binary file. No difference found in the output binary file with this change. Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2017-05-13 14:57:25 +05:30
static __rte_always_inline void
examples/l3fwd: reorganise multi hash lookup Extract common code from l3fwd_em_hlm_sse.h, and add to the new file l3fwd_em_hlm.h. Signed-off-by: Jianbo Liu <jianbo.liu@linaro.org>
2017-07-04 18:23:58 +08:00
get_ipv4_5tuple(struct rte_mbuf *m0, __m128i mask0,
union ipv4_5tuple_host *key)
examples/l3fwd: rework exact-match Current implementation of Exact-Match uses different execution path than for LPM. Unifying them allows to reuse big part of LPM code and sightly increase performance of Exact-Match. Main changes: ------------- * Packet classification stage is separated from the rest of path for both LPM and EM. * Packet processing, modifying and transmit part is the same for LPM and EM and mostly based on the current LPM implementation. * Shared code is moved to the common file "l3fwd_sse.h". * While sequential packet classification in EM path, seems to be faster than using multi hash lookup, used before, it is used by default. Old implementation is moved to the file l3fwd_em_hlm_sse.h and can be enabled with HASH_LOOKUP_MULTI global define in compilation time. Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2016-02-29 11:33:07 +01:00
{
examples/l3fwd: reorganise multi hash lookup Extract common code from l3fwd_em_hlm_sse.h, and add to the new file l3fwd_em_hlm.h. Signed-off-by: Jianbo Liu <jianbo.liu@linaro.org>
2017-07-04 18:23:58 +08:00
__m128i tmpdata0 = _mm_loadu_si128(
rte_pktmbuf_mtod_offset(m0, __m128i *,
net: add rte prefix to ether structures Add 'rte_' prefix to structures: - rename struct ether_addr as struct rte_ether_addr. - rename struct ether_hdr as struct rte_ether_hdr. - rename struct vlan_hdr as struct rte_vlan_hdr. - rename struct vxlan_hdr as struct rte_vxlan_hdr. - rename struct vxlan_gpe_hdr as struct rte_vxlan_gpe_hdr. Do not update the command line library to avoid adding a dependency to librte_net. Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Reviewed-by: Stephen Hemminger <stephen@networkplumber.org> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
2019-05-21 18:13:03 +02:00
sizeof(struct rte_ether_hdr) +
net: add rte prefix to IP structure Add 'rte_' prefix to structures: - rename struct ipv4_hdr as struct rte_ipv4_hdr. - rename struct ipv6_hdr as struct rte_ipv6_hdr. Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Reviewed-by: Stephen Hemminger <stephen@networkplumber.org> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
2019-05-21 18:13:10 +02:00
offsetof(struct rte_ipv4_hdr, time_to_live)));
examples/l3fwd: rework exact-match Current implementation of Exact-Match uses different execution path than for LPM. Unifying them allows to reuse big part of LPM code and sightly increase performance of Exact-Match. Main changes: ------------- * Packet classification stage is separated from the rest of path for both LPM and EM. * Packet processing, modifying and transmit part is the same for LPM and EM and mostly based on the current LPM implementation. * Shared code is moved to the common file "l3fwd_sse.h". * While sequential packet classification in EM path, seems to be faster than using multi hash lookup, used before, it is used by default. Old implementation is moved to the file l3fwd_em_hlm_sse.h and can be enabled with HASH_LOOKUP_MULTI global define in compilation time. Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2016-02-29 11:33:07 +01:00
examples/l3fwd: reorganise multi hash lookup Extract common code from l3fwd_em_hlm_sse.h, and add to the new file l3fwd_em_hlm.h. Signed-off-by: Jianbo Liu <jianbo.liu@linaro.org>
2017-07-04 18:23:58 +08:00
key->xmm = _mm_and_si128(tmpdata0, mask0);
examples/l3fwd: rework exact-match Current implementation of Exact-Match uses different execution path than for LPM. Unifying them allows to reuse big part of LPM code and sightly increase performance of Exact-Match. Main changes: ------------- * Packet classification stage is separated from the rest of path for both LPM and EM. * Packet processing, modifying and transmit part is the same for LPM and EM and mostly based on the current LPM implementation. * Shared code is moved to the common file "l3fwd_sse.h". * While sequential packet classification in EM path, seems to be faster than using multi hash lookup, used before, it is used by default. Old implementation is moved to the file l3fwd_em_hlm_sse.h and can be enabled with HASH_LOOKUP_MULTI global define in compilation time. Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2016-02-29 11:33:07 +01:00
}
static inline void
get_ipv6_5tuple(struct rte_mbuf *m0, __m128i mask0,
__m128i mask1, union ipv6_5tuple_host *key)
{
__m128i tmpdata0 = _mm_loadu_si128(
rte_pktmbuf_mtod_offset(m0, __m128i *,
net: add rte prefix to ether structures Add 'rte_' prefix to structures: - rename struct ether_addr as struct rte_ether_addr. - rename struct ether_hdr as struct rte_ether_hdr. - rename struct vlan_hdr as struct rte_vlan_hdr. - rename struct vxlan_hdr as struct rte_vxlan_hdr. - rename struct vxlan_gpe_hdr as struct rte_vxlan_gpe_hdr. Do not update the command line library to avoid adding a dependency to librte_net. Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Reviewed-by: Stephen Hemminger <stephen@networkplumber.org> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
2019-05-21 18:13:03 +02:00
sizeof(struct rte_ether_hdr) +
net: add rte prefix to IP structure Add 'rte_' prefix to structures: - rename struct ipv4_hdr as struct rte_ipv4_hdr. - rename struct ipv6_hdr as struct rte_ipv6_hdr. Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Reviewed-by: Stephen Hemminger <stephen@networkplumber.org> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
2019-05-21 18:13:10 +02:00
offsetof(struct rte_ipv6_hdr, payload_len)));
examples/l3fwd: rework exact-match Current implementation of Exact-Match uses different execution path than for LPM. Unifying them allows to reuse big part of LPM code and sightly increase performance of Exact-Match. Main changes: ------------- * Packet classification stage is separated from the rest of path for both LPM and EM. * Packet processing, modifying and transmit part is the same for LPM and EM and mostly based on the current LPM implementation. * Shared code is moved to the common file "l3fwd_sse.h". * While sequential packet classification in EM path, seems to be faster than using multi hash lookup, used before, it is used by default. Old implementation is moved to the file l3fwd_em_hlm_sse.h and can be enabled with HASH_LOOKUP_MULTI global define in compilation time. Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2016-02-29 11:33:07 +01:00
__m128i tmpdata1 = _mm_loadu_si128(
rte_pktmbuf_mtod_offset(m0, __m128i *,
net: add rte prefix to ether structures Add 'rte_' prefix to structures: - rename struct ether_addr as struct rte_ether_addr. - rename struct ether_hdr as struct rte_ether_hdr. - rename struct vlan_hdr as struct rte_vlan_hdr. - rename struct vxlan_hdr as struct rte_vxlan_hdr. - rename struct vxlan_gpe_hdr as struct rte_vxlan_gpe_hdr. Do not update the command line library to avoid adding a dependency to librte_net. Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Reviewed-by: Stephen Hemminger <stephen@networkplumber.org> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
2019-05-21 18:13:03 +02:00
sizeof(struct rte_ether_hdr) +
net: add rte prefix to IP structure Add 'rte_' prefix to structures: - rename struct ipv4_hdr as struct rte_ipv4_hdr. - rename struct ipv6_hdr as struct rte_ipv6_hdr. Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Reviewed-by: Stephen Hemminger <stephen@networkplumber.org> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
2019-05-21 18:13:10 +02:00
offsetof(struct rte_ipv6_hdr, payload_len) +
examples/l3fwd: rework exact-match Current implementation of Exact-Match uses different execution path than for LPM. Unifying them allows to reuse big part of LPM code and sightly increase performance of Exact-Match. Main changes: ------------- * Packet classification stage is separated from the rest of path for both LPM and EM. * Packet processing, modifying and transmit part is the same for LPM and EM and mostly based on the current LPM implementation. * Shared code is moved to the common file "l3fwd_sse.h". * While sequential packet classification in EM path, seems to be faster than using multi hash lookup, used before, it is used by default. Old implementation is moved to the file l3fwd_em_hlm_sse.h and can be enabled with HASH_LOOKUP_MULTI global define in compilation time. Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2016-02-29 11:33:07 +01:00
sizeof(__m128i)));
__m128i tmpdata2 = _mm_loadu_si128(
rte_pktmbuf_mtod_offset(m0, __m128i *,
net: add rte prefix to ether structures Add 'rte_' prefix to structures: - rename struct ether_addr as struct rte_ether_addr. - rename struct ether_hdr as struct rte_ether_hdr. - rename struct vlan_hdr as struct rte_vlan_hdr. - rename struct vxlan_hdr as struct rte_vxlan_hdr. - rename struct vxlan_gpe_hdr as struct rte_vxlan_gpe_hdr. Do not update the command line library to avoid adding a dependency to librte_net. Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Reviewed-by: Stephen Hemminger <stephen@networkplumber.org> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
2019-05-21 18:13:03 +02:00
sizeof(struct rte_ether_hdr) +
net: add rte prefix to IP structure Add 'rte_' prefix to structures: - rename struct ipv4_hdr as struct rte_ipv4_hdr. - rename struct ipv6_hdr as struct rte_ipv6_hdr. Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Reviewed-by: Stephen Hemminger <stephen@networkplumber.org> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
2019-05-21 18:13:10 +02:00
offsetof(struct rte_ipv6_hdr, payload_len) +
examples/l3fwd: rework exact-match Current implementation of Exact-Match uses different execution path than for LPM. Unifying them allows to reuse big part of LPM code and sightly increase performance of Exact-Match. Main changes: ------------- * Packet classification stage is separated from the rest of path for both LPM and EM. * Packet processing, modifying and transmit part is the same for LPM and EM and mostly based on the current LPM implementation. * Shared code is moved to the common file "l3fwd_sse.h". * While sequential packet classification in EM path, seems to be faster than using multi hash lookup, used before, it is used by default. Old implementation is moved to the file l3fwd_em_hlm_sse.h and can be enabled with HASH_LOOKUP_MULTI global define in compilation time. Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2016-02-29 11:33:07 +01:00
sizeof(__m128i) + sizeof(__m128i)));
key->xmm[0] = _mm_and_si128(tmpdata0, mask0);
key->xmm[1] = tmpdata1;
key->xmm[2] = _mm_and_si128(tmpdata2, mask1);
}
#endif /* __L3FWD_EM_SSE_HLM_H__ */
Reference in New Issue Copy Permalink