Enable building more on Windows
Subsystems (2)
cover Enable building more on Windows
While diagnosing some Windows cross build errors; noticed that lots of important DPDK libraries are not being built on Windows.
This patchset enables build them.
Tested with MingW cross build.
Stephen Hemminger (8):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on Windows
rib: enable on Windows
fib: enable on Windows
pcapng: windows compatability
app/test/meson.build | 2 +- app/test/test_fib.c | 22 +------------- app/test/test_fib6.c | 24 ++------------- app/test/test_fib6_perf.c | 15 +-------- app/test/test_fib_perf.c | 19 ++---------- app/test/test_ipfrag.c | 12 -------- app/test/test_lpm.c | 15 +-------- app/test/test_lpm6.c | 12 -------- app/test/test_lpm6_data.h | 3 +- app/test/test_lpm6_perf.c | 14 +-------- app/test/test_lpm_perf.c | 39 ++++++++---------------- app/test/test_reorder.c | 11 ------- app/test/test_rib.c | 19 ------------ app/test/test_rib6.c | 24 ++------------- drivers/net/null/meson.build | 6 ---- drivers/net/ring/meson.build | 6 ---- lib/fib/meson.build | 6 ---- lib/fib/rte_fib.c | 1 + lib/fib/rte_fib6.c | 1 + lib/ip_frag/ip_frag_common.h | 2 ++ lib/ip_frag/meson.build | 6 ---- lib/lpm/meson.build | 6 ---- lib/pcapng/meson.build | 6 ---- lib/pcapng/rte_pcapng.c | 59 ++++++++++++++++++++++++++++++++++-- lib/reorder/meson.build | 6 ---- lib/reorder/rte_reorder.c | 1 + lib/rib/meson.build | 6 ---- lib/rib/rte_rib.c | 1 + lib/rib/rte_rib6.c | 1 + 29 files changed, 90 insertions(+), 255 deletions(-)
Signature · 1 line
2.35.1
PATCH 1/8 net/null: build null PMD on Windows
Builds fine with current code, no changes needed.
Signed-off-by Stephen Hemminger <[email protected]>
--- drivers/net/null/meson.build | 6 ------ 1 file changed, 6 deletions(-)
drivers/net/null/meson.build +0 -6
@@ -1,11 +1,5 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_eth_null.c') pmd_supports_disable_iova_as_pa = true
Signature · 1 line
2.35.1
Re: [PATCH 1/8] net/null: build null PMD on Windows
Quoted 4 lines · Stephen Hemminger
Builds fine with current code, no changes needed.
Signed-off-by Stephen Hemminger <[email protected]>---
Acked-by Tyler Retzlaff <[email protected]>
PATCH 2/8 net/ring: build on Windows
In reply to Stephen HemmingerThis driver builds ok on windows, but need to modify the test meson build to skip the eventdev test on Windows.
Signed-off-by Stephen Hemminger <[email protected]>
--- app/test/meson.build | 2 +- drivers/net/ring/meson.build | 6 ------ 2 files changed, 1 insertion(+), 7 deletions(-)
app/test/meson.build +1 -1
@@ -390,7 +390,7 @@ if dpdk_conf.has('RTE_NET_BOND') driver_test_names += 'link_bonding_mode4_autotest' endif endif -if dpdk_conf.has('RTE_NET_RING') +if dpdk_conf.has('RTE_LIB_EVENTDEV') and dpdk_conf.has('RTE_NET_RING') test_deps += 'net_ring' test_sources += 'test_pmd_ring_perf.c' test_sources += 'test_pmd_ring.c'
drivers/net/ring/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_eth_ring.c') headers = files('rte_eth_ring.h') pmd_supports_disable_iova_as_pa = true
Signature · 1 line
2.35.1
Re: [PATCH 2/8] net/ring: build on Windows
Quoted 5 lines · Stephen Hemminger
This driver builds ok on windows, but need to modify the test meson build to skip the eventdev test on Windows.
Signed-off-by Stephen Hemminger <[email protected]>---
Acked-by Tyler Retzlaff <[email protected]>
Re: [PATCH 2/8] net/ring: build on Windows
In reply to Stephen HemmingerHi Stephen,
Thanks for enabling this support.
Quoted 5 lines · Stephen Hemminger
This driver builds ok on windows, but need to modify the test meson build to skip the eventdev test on Windows.
Signed-off-by Stephen Hemminger<[email protected]>---
There is still a build error with clang compiler on Windows as mentioned by Dmitry:
../drivers/net/ring/rte_eth_ring.c:542:12: error: use of undeclared identifier 'PATH_MAX'
char name[PATH_MAX];
^#include <rte_os_shim.h> is missing
Quoted 1 line
Re: [PATCH 2/8] net/ring: build on Windows
Quoted 11 lines · Pallavi
Hi Stephen,
Thanks for enabling this support.
Quoted 5 lines · Stephen Hemminger
This driver builds ok on windows, but need to modify the test meson build to skip the eventdev test on Windows.
Signed-off-by Stephen Hemminger<[email protected]>---
Quoted 9 lines
There is still a build error with clang compiler on Windows as mentioned by Dmitry:
../drivers/net/ring/rte_eth_ring.c:542:12: error: use of undeclared identifier 'PATH_MAX'char name[PATH_MAX]; ^#include <
rte_os_shim.h> is missingQuoted 1 line
It is in the mingw version but not other compilers.
Re: [PATCH 2/8] net/ring: build on Windows
Quoted 17 lines · Stephen Hemminger
Quoted 14 lines · Pallavi
Hi Stephen,
Thanks for enabling this support.
Quoted 5 lines · Stephen Hemminger
This driver builds ok on windows, but need to modify the test meson build to skip the eventdev test on Windows.
Signed-off-by Stephen Hemminger<[email protected]>---
There is still a build error with clang compiler on Windows as mentioned by Dmitry:
Quoted 8 lines
Quoted 6 lines
../drivers/net/ring/rte_eth_ring.c:542:12: error: use of undeclared identifier 'PATH_MAX'char name[PATH_MAX]; ^#include <
rte_os_shim.h> is missingQuoted 1 line
It is in the mingw version but not other compilers.
not sure if it helps but in the sludge of my memory i think you can use MAX_PATH.
so maybe a little
#ifndef PATH_MAX
#define PATH_MAX MAX_PATH
#endif
in the os shim?
Re: [PATCH 2/8] net/ring: build on Windows
2022-12-15 09:48 (UTC-0800), Tyler Retzlaff:
Quoted 3 lines · Stephen Hemminger
Quoted 2 lines
On Wed, 14 Dec 2022 15:18:08 -0800
"Kadam, Pallavi" <[email protected]> wrote:
[...]
Quoted 15 lines
Quoted 11 lines
Quoted 9 lines
There is still a build error with clang compiler on Windows as mentioned by Dmitry:
../drivers/net/ring/rte_eth_ring.c:542:12: error: use of undeclared identifier 'PATH_MAX'char name[PATH_MAX]; ^#include <
rte_os_shim.h> is missingQuoted 1 line
It is in the mingw version but not other compilers.
not sure if it helps but in the sludge of my memory i think you can use
MAX_PATH.
Quoted 6 lines
so maybe a little
#ifndefPATH_MAX
#definePATH_MAXMAX_PATH
#endifin the os shim?
It is in rte_os_shim.h already (for MinGW), just needs to be included.
PATCH 3/8 lpm: enable on Windows
In reply to Stephen HemmingerThis builds on Windows without changes.
Signed-off-by Stephen Hemminger <[email protected]>
--- app/test/test_lpm.c | 15 +-------------- app/test/test_lpm6.c | 12 ------------ app/test/test_lpm6_data.h | 3 ++- app/test/test_lpm6_perf.c | 14 +------------- app/test/test_lpm_perf.c | 39 ++++++++++++--------------------------- lib/lpm/meson.build | 6 ------ 6 files changed, 16 insertions(+), 73 deletions(-)
app/test/test_lpm.c +1 -14
@@ -2,18 +2,6 @@* Copyright(c) 2010-2014 Intel Corporation */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm (void ) -{ -printf ("lpm not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <stdio.h> #include <stdint.h> #include <stdlib.h>
@@ -22,6 +10,7 @@ test_lpm(void)#include <rte_lpm.h> #include <rte_malloc.h> +#include "test.h" #include "test_xmmt_ops.h" #define TEST_LPM_ASSERT (cond) do { \
@@ -1595,6 +1584,4 @@ test_lpm(void)return global_status; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm_autotest, test_lpm);
app/test/test_lpm6.c +0 -12
@@ -10,16 +10,6 @@#include <string.h> #include <rte_memory.h> -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm6 (void ) -{ -printf ("lpm6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_lpm6.h> #include "test_lpm6_data.h"
@@ -1803,6 +1793,4 @@ test_lpm6(void)return global_status; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm6_autotest, test_lpm6);
app/test/test_lpm6_data.h +2 -1
@@ -6,6 +6,7 @@#include <stdint.h> #include <stdlib.h> +#include <rte_random.h> struct rules_tbl_entry {uint8_t ip[16 ];
@@ -1129,7 +1130,7 @@ static void generate_large_ips_table(int gen_expected_next_hop)for (i =0 ; i <NUM_IPS_ENTRIES ; i++) {for (j =0 ; j <16 ; j++) - large_ips_table[i].ip[j] =lrand48 (); + large_ips_table[i].ip[j] =rte_rand (); }for (k = j =0 , i =0 ; i <NUM_IPS_ENTRIES ; i++) {
app/test/test_lpm6_perf.c +1 -13
@@ -2,17 +2,6 @@* Copyright(c) 2010-2014 Intel Corporation */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm6_perf (void ) -{ -printf ("lpm6_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else #include <stdio.h> #include <stdint.h>
@@ -24,6 +13,7 @@ test_lpm6_perf(void)#include <rte_memory.h> #include <rte_lpm6.h> +#include "test.h" #include "test_lpm6_data.h" #define TEST_LPM_ASSERT (cond) do { \
@@ -171,6 +161,4 @@ test_lpm6_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm6_perf_autotest, test_lpm6_perf);
app/test/test_lpm_perf.c +12 -27
@@ -3,17 +3,6 @@* Copyright(c) 2020 Arm Limited */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm_perf (void ) -{ -printf ("lpm_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else #include <stdio.h> #include <stdint.h> #include <stdlib.h>
@@ -25,7 +14,9 @@ test_lpm_perf(void)#include <rte_malloc.h> #include <rte_ip.h> #include <rte_lpm.h> +#include <rte_spinlock.h> +#include "test.h" #include "test_xmmt_ops.h" struct rte_lpm *lpm;
@@ -34,8 +25,9 @@ static volatile uint8_t writer_done;static volatile uint32_t thr_id;static uint64_t gwrite_cycles;static uint32_t num_writers; -/* LPM APIs are not thread safe, use mutex to provide thread safety */ -static pthread_mutex_t lpm_mutex =PTHREAD_MUTEX_INITIALIZER ; + +/* LPM APIs are not thread safe, use spinlock */ +static rte_spinlock_t lpm_lock =RTE_SPINLOCK_INITIALIZER ;/* Report quiescent state interval every 1024 lookups. Larger critical * sections in reader will result in writer polling multiple times.
@@ -267,7 +259,7 @@ static void generate_random_rule_prefix(uint32_t ip_class, uint8_t depth)/* Only generate rest bits except the most significant * fixed bits for IP address class */ - start =lrand48 () & mask; + start =rte_rand () & mask; ptr_rule = &large_route_table[num_route_entries]; ptr_ldepth_rule = &large_ldepth_route_table[num_ldepth_route_entries];for (k =0 ; k < rule_num; k++) {
@@ -296,7 +288,7 @@ static void insert_rule_in_random_pos(uint32_t ip, uint8_t depth)struct route_rule tmp;do { - pos =lrand48 (); + pos =rte_rand (); try_count++; }while ((try_count <10 ) && (pos > num_route_entries));
@@ -452,8 +444,7 @@ test_lpm_rcu_qsbr_writer(void *arg)for (i =0 ; i <RCU_ITERATIONS ; i++) {/* Add all the entries */ for (j = si; j < ei; j++) { -if (num_writers >1 ) -pthread_mutex_lock (&lpm_mutex); +rte_spinlock_lock (&lpm_lock);if (rte_lpm_add (lpm, large_ldepth_route_table[j].ip, large_ldepth_route_table[j].depth, next_hop_add) !=0 ) {
@@ -461,22 +452,19 @@ test_lpm_rcu_qsbr_writer(void *arg) i, j); goto error; } -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock); }/* Delete all the entries */ for (j = si; j < ei; j++) { -if (num_writers >1 ) -pthread_mutex_lock (&lpm_mutex); +rte_spinlock_lock (&lpm_lock);if (rte_lpm_delete (lpm, large_ldepth_route_table[j].ip, large_ldepth_route_table[j].depth) !=0 ) {printf ("Failed to delete iteration %d, route# %d\n" , i, j); goto error; } -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock); } }
@@ -487,8 +475,7 @@ test_lpm_rcu_qsbr_writer(void *arg)return 0 ; error: -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock);return -1 ; }
@@ -773,6 +760,4 @@ test_lpm_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm_perf_autotest, test_lpm_perf);
lib/lpm/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_lpm.c', 'rte_lpm6.c') headers = files('rte_lpm.h', 'rte_lpm6.h') # since header files have different names, we can install all vector headers
Signature · 1 line
2.35.1
Re: [PATCH 3/8] lpm: enable on Windows
Quoted 4 lines · Stephen Hemminger
This builds on Windows without changes.
Signed-off-by Stephen Hemminger <[email protected]>---
Acked-by Tyler Retzlaff <[email protected]>
PATCH 4/8 reorder: build on Windows
In reply to Stephen HemmingerThis builds on Windows if sys/queue.h is included.
Signed-off-by Stephen Hemminger <[email protected]>
--- app/test/test_reorder.c | 11 ----------- lib/reorder/meson.build | 6 ------ lib/reorder/rte_reorder.c | 1 + 3 files changed, 1 insertion(+), 17 deletions(-)
app/test/test_reorder.c +0 -11
@@ -11,16 +11,6 @@#include <rte_cycles.h> #include <rte_errno.h> #include <rte_mbuf.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_reorder (void ) -{ -printf ("reorder not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_reorder.h> #include <rte_lcore.h> #include <rte_malloc.h>
@@ -395,6 +385,5 @@ test_reorder(void)return unit_test_suite_runner (&reorder_test_suite); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ REGISTER_TEST_COMMAND (reorder_autotest, test_reorder);
lib/reorder/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_reorder.c') headers = files('rte_reorder.h') deps += ['mbuf']
lib/reorder/rte_reorder.c +1 -0
@@ -3,6 +3,7 @@*/ #include <string.h> +#include <sys/queue.h> #include <rte_string_fns.h> #include <rte_log.h>
Signature · 1 line
2.35.1
Re: [PATCH 4/8] reorder: build on Windows
Quoted 4 lines · Stephen Hemminger
This builds on Windows if
sys/queue.his included.Signed-off-by Stephen Hemminger <[email protected]>---
Acked-by Tyler Retzlaff <[email protected]>
PATCH 5/8 ip_frag: enable build on Windows
In reply to Stephen HemmingerThis build works on Windows if sys/queue.h is included.
Signed-off-by Stephen Hemminger <[email protected]>
--- app/test/test_ipfrag.c | 12 ------------ lib/ip_frag/ip_frag_common.h | 2 ++ lib/ip_frag/meson.build | 6 ------ 3 files changed, 2 insertions(+), 18 deletions(-)
app/test/test_ipfrag.c +0 -12
@@ -10,17 +10,6 @@#include <rte_cycles.h> #include <rte_hexdump.h> #include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_ipfrag (void ) -{ -printf ("ipfrag not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_ip_frag.h> #include <rte_mbuf.h> #include <rte_random.h>
@@ -520,6 +509,5 @@ test_ipfrag(void)return unit_test_suite_runner (&ipfrag_testsuite); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ REGISTER_TEST_COMMAND (ipfrag_autotest, test_ipfrag);
lib/ip_frag/ip_frag_common.h +2 -0
@@ -8,6 +8,8 @@#include "rte_ip_frag.h" #include "ip_reassembly.h" +#include <sys/queue.h> +/* logging macros. */ #ifdef RTE_LIBRTE_IP_FRAG_DEBUG #define IP_FRAG_LOG(lvl, fmt, args...) RTE_LOG(lvl, USER1, fmt, ##args)
lib/ip_frag/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files( 'rte_ipv4_fragmentation.c', 'rte_ipv6_fragmentation.c',
Signature · 1 line
2.35.1
Re: [PATCH 5/8] ip_frag: enable build on Windows
Quoted 4 lines · Stephen Hemminger
This build works on Windows if
sys/queue.his included.Signed-off-by Stephen Hemminger <[email protected]>---
Acked-by Tyler Retzlaff <[email protected]>
PATCH 6/8 rib: enable on Windows
In reply to Stephen HemmingerThe RIB library builds on Windows as long as sys/queue.h is included
Signed-off-by Stephen Hemminger <[email protected]>
--- app/test/test_rib.c | 19 ------------------- app/test/test_rib6.c | 24 ++---------------------- lib/rib/meson.build | 6 ------ lib/rib/rte_rib.c | 1 + lib/rib/rte_rib6.c | 1 + 5 files changed, 4 insertions(+), 47 deletions(-)
app/test/test_rib.c +0 -19
@@ -10,23 +10,6 @@#include <stdlib.h> #include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_rib (void ) -{ -printf ("rib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_rib (void ) -{ -printf ("slow_rib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_rib.h> typedef int32_t (*rte_rib_test )(void );
@@ -380,7 +363,5 @@ test_slow_rib(void)return unit_test_suite_runner (&rib_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (rib_autotest, test_rib);REGISTER_TEST_COMMAND (rib_slow_autotest, test_slow_rib);
app/test/test_rib6.c +2 -22
@@ -3,32 +3,14 @@* Copyright(c) 2019 Intel Corporation */ -#include "test.h" -#include <stdio.h> #include <stdint.h> #include <stdlib.h> -#include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_rib6 (void ) -{ -printf ("rib6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_rib6 (void ) -{ -printf ("slow_rib6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_rib6.h> +#include "test.h" +typedef int32_t (*rte_rib6_test )(void );static int32_t test_create_invalid (void );
@@ -385,7 +367,5 @@ test_slow_rib6(void)return unit_test_suite_runner (&rib6_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (rib6_autotest, test_rib6);REGISTER_TEST_COMMAND (rib6_slow_autotest, test_slow_rib6);
lib/rib/meson.build +0 -6
@@ -2,12 +2,6 @@ # Copyright(c) 2018 Vladimir Medvedkin <[email protected]> # Copyright(c) 2019 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_rib.c', 'rte_rib6.c') headers = files('rte_rib.h', 'rte_rib6.h') deps += ['mempool']
lib/rib/rte_rib.c +1 -0
@@ -5,6 +5,7 @@#include <stdbool.h> #include <stdint.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
lib/rib/rte_rib6.c +1 -0
@@ -5,6 +5,7 @@#include <stdbool.h> #include <stdint.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
Signature · 1 line
2.35.1
Re: [PATCH 6/8] rib: enable on Windows
Quoted 4 lines · Stephen Hemminger
The RIB library builds on Windows as long as
sys/queue.his includedSigned-off-by Stephen Hemminger <[email protected]>---
Acked-by Tyler Retzlaff <[email protected]>
PATCH 7/8 fib: enable on Windows
In reply to Stephen HemmingerThe FIB library builds on Windows as long as sys/queue.h is defined.
Signed-off-by Stephen Hemminger <[email protected]>
--- app/test/test_fib.c | 22 +--------------------- app/test/test_fib6.c | 24 ++---------------------- app/test/test_fib6_perf.c | 15 +-------------- app/test/test_fib_perf.c | 19 +++---------------- lib/fib/meson.build | 6 ------ lib/fib/rte_fib.c | 1 + lib/fib/rte_fib6.c | 1 + 7 files changed, 9 insertions(+), 79 deletions(-)
app/test/test_fib.c +1 -21
@@ -9,28 +9,10 @@#include <rte_ip.h> #include <rte_log.h> +#include <rte_fib.h> #include "test.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib (void ) -{ -printf ("fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_fib (void ) -{ -printf ("slow_fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib.h> -typedef int32_t (*rte_fib_test )(void );static int32_t test_create_invalid (void );
@@ -433,7 +415,5 @@ test_slow_fib(void)return unit_test_suite_runner (&fib_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib_autotest, test_fib);REGISTER_TEST_COMMAND (fib_slow_autotest, test_slow_fib);
app/test/test_fib6.c +2 -22
@@ -9,29 +9,11 @@#include <rte_memory.h> #include <rte_log.h> - -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib6 (void ) -{ -printf ("fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_fib6 (void ) -{ -printf ("slow_fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_rib6.h> #include <rte_fib6.h> +#include "test.h" +typedef int32_t (*rte_fib6_test )(void );static int32_t test_create_invalid (void );
@@ -442,7 +424,5 @@ test_slow_fib6(void)return unit_test_suite_runner (&fib6_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib6_autotest, test_fib6);REGISTER_TEST_COMMAND (fib6_slow_autotest, test_slow_fib6);
app/test/test_fib6_perf.c +1 -14
@@ -10,21 +10,10 @@#include <rte_cycles.h> #include <rte_random.h> #include <rte_memory.h> +#include <rte_fib6.h> #include "test.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib6_perf (void ) -{ -printf ("fib6_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib6.h> -#include "test_lpm6_data.h" #define TEST_FIB_ASSERT (cond) do { \
@@ -167,6 +156,4 @@ test_fib6_perf(void)return 0 ; } -#endif /*ifdef RTE_EXEC_ENV_WINDOWS*/ -REGISTER_TEST_COMMAND (fib6_perf_autotest, test_fib6_perf);
app/test/test_fib_perf.c +3 -16
@@ -12,22 +12,11 @@#include <rte_random.h> #include <rte_branch_prediction.h> #include <rte_ip.h> +#include <rte_fib.h> #include "test.h" #include "test_xmmt_ops.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib_perf (void ) -{ -printf ("fib_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib.h> -#define TEST_FIB_ASSERT (cond) do { \ if (!(cond)) { \ printf("Error at line %d:\n", __LINE__); \
@@ -246,7 +235,7 @@ static void generate_random_rule_prefix(uint32_t ip_class, uint8_t depth)/* Only generate rest bits except the most significant * fixed bits for IP address class */ - start =lrand48 () & mask; + start =rte_rand () & mask; ptr_rule = &large_route_table[num_route_entries];for (k =0 ; k < rule_num; k++) { ptr_rule->ip = (start << (RTE_FIB_MAX_DEPTH - depth))
@@ -265,7 +254,7 @@ static void insert_rule_in_random_pos(uint32_t ip, uint8_t depth)struct route_rule tmp;do { - pos =lrand48 (); + pos =rte_rand (); try_count++; }while ((try_count <10 ) && (pos > num_route_entries));
@@ -420,6 +409,4 @@ test_fib_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib_perf_autotest, test_fib_perf);
lib/fib/meson.build +0 -6
@@ -2,12 +2,6 @@ # Copyright(c) 2018 Vladimir Medvedkin <[email protected]> # Copyright(c) 2019 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_fib.c', 'rte_fib6.c', 'dir24_8.c', 'trie.c') headers = files('rte_fib.h', 'rte_fib6.h') deps += ['rib']
lib/fib/rte_fib.c +1 -0
@@ -5,6 +5,7 @@#include <stdint.h> #include <string.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
lib/fib/rte_fib6.c +1 -0
@@ -5,6 +5,7 @@#include <stdint.h> #include <string.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_tailq.h>
Signature · 1 line
2.35.1
Re: [PATCH 7/8] fib: enable on Windows
Quoted 4 lines · Stephen Hemminger
The FIB library builds on Windows as long as
sys/queue.his defined.Signed-off-by Stephen Hemminger <[email protected]>---
Acked-by Tyler Retzlaff <[email protected]>
PATCH 8/8 pcapng: windows compatability
In reply to Stephen HemmingerAllow building on Windows, need to provide some compatability wrappers for writev() and if_nametoindex.
Signed-off-by Stephen Hemminger <[email protected]>
--- lib/pcapng/meson.build | 6 ----- lib/pcapng/rte_pcapng.c | 59 +++++++++++++++++++++++++++++++++++++++-- 2 files changed, 57 insertions(+), 8 deletions(-)
lib/pcapng/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2019 Microsoft Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_pcapng.c') headers = files('rte_pcapng.h')
lib/pcapng/rte_pcapng.c +57 -2
84 lines not on this page · raw message
Signature · 1 line
2.35.1
Re: [PATCH 8/8] pcapng: windows compatability
Quoted 5 lines · Stephen Hemminger
Allow building on Windows, need to provide some compatability wrappers for
writev()andif_nametoindex.Signed-off-by Stephen Hemminger <[email protected]>---
Acked-by Tyler Retzlaff <[email protected]>
v2 cover Enable building more libraries on Windows
In reply to Stephen HemmingerWhile diagnosing some Windows cross build errors; noticed that lots of important DPDK libraries are not being built on Windows.
Stephen Hemminger (8):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on Windows
rib: enable on Windows
fib: enable on Windows
pcapng: windows compatibility
v2 - fix unnecessary PATH_MAX in net/ring driver
app/test/meson.build | 2 +- app/test/test_fib.c | 22 +----------- app/test/test_fib6.c | 24 ++------------ app/test/test_fib6_perf.c | 15 +-------- app/test/test_fib_perf.c | 19 ++--------- app/test/test_ipfrag.c | 12 ------- app/test/test_lpm.c | 15 +-------- app/test/test_lpm6.c | 12 ------- app/test/test_lpm6_data.h | 3 +- app/test/test_lpm6_perf.c | 14 +------- app/test/test_lpm_perf.c | 39 +++++++--------------- app/test/test_reorder.c | 11 ------ app/test/test_rib.c | 19 ----------- app/test/test_rib6.c | 24 ++------------ drivers/net/null/meson.build | 6 ---- drivers/net/ring/meson.build | 6 ---- drivers/net/ring/rte_eth_ring.c | 3 +- lib/fib/meson.build | 6 ---- lib/fib/rte_fib.c | 1 + lib/fib/rte_fib6.c | 1 + lib/ip_frag/ip_frag_common.h | 2 ++ lib/ip_frag/meson.build | 6 ---- lib/lpm/meson.build | 6 ---- lib/pcapng/meson.build | 6 ---- lib/pcapng/rte_pcapng.c | 59 +++++++++++++++++++++++++++++++-- lib/reorder/meson.build | 6 ---- lib/reorder/rte_reorder.c | 1 + lib/rib/meson.build | 6 ---- lib/rib/rte_rib.c | 1 + lib/rib/rte_rib6.c | 1 + 30 files changed, 92 insertions(+), 256 deletions(-)
Signature · 1 line
2.39.1
PATCH v2 1/8 net/null: build null PMD on Windows
Builds fine with current code, no changes needed.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- drivers/net/null/meson.build | 6 ------ 1 file changed, 6 deletions(-)
drivers/net/null/meson.build +0 -6
@@ -1,11 +1,5 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_eth_null.c') pmd_supports_disable_iova_as_pa = true
Signature · 1 line
2.39.1
PATCH v2 2/8 net/ring: build on Windows
In reply to Stephen HemmingerThe command line arguments are stored in node_action_pair and the name[] was sized to PATH_MAX which does not exist on Windows. Since the name is either "CREATE" or "ATTACH" it is not related to PATH_MAX (4096).
With this fix driver builds ok on windows, but need to modify the test meson build to skip the eventdev test on Windows.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/meson.build | 2 +- drivers/net/ring/meson.build | 6 ------ drivers/net/ring/rte_eth_ring.c | 3 ++- 3 files changed, 3 insertions(+), 8 deletions(-)
app/test/meson.build +1 -1
@@ -390,7 +390,7 @@ if dpdk_conf.has('RTE_NET_BOND') driver_test_names += 'link_bonding_mode4_autotest' endif endif -if dpdk_conf.has('RTE_NET_RING') +if dpdk_conf.has('RTE_LIB_EVENTDEV') and dpdk_conf.has('RTE_NET_RING') test_deps += 'net_ring' test_sources += 'test_pmd_ring_perf.c' test_sources += 'test_pmd_ring.c'
drivers/net/ring/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_eth_ring.c') headers = files('rte_eth_ring.h') pmd_supports_disable_iova_as_pa = true
drivers/net/ring/rte_eth_ring.c +2 -1
@@ -17,6 +17,7 @@#define ETH_RING_NUMA_NODE_ACTION_ARG "nodeaction"#define ETH_RING_ACTION_CREATE "CREATE"#define ETH_RING_ACTION_ATTACH "ATTACH" +#define ETH_RING_ACTION_MAX_LEN 8/* CREATE | ACTION */ #define ETH_RING_INTERNAL_ARG "internal"#define ETH_RING_INTERNAL_ARG_MAX_LEN 19/* "0x..16chars..\0" */
@@ -539,7 +540,7 @@ eth_dev_ring_create(const char *name, }struct node_action_pair { -char name[PATH_MAX ]; +char name[ETH_RING_ACTION_MAX_LEN ];unsigned int node;enum dev_action action; };
Signature · 1 line
2.39.1
PATCH v2 3/8 lpm: enable on Windows
In reply to Stephen HemmingerThis builds on Windows without changes.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_lpm.c | 15 +-------------- app/test/test_lpm6.c | 12 ------------ app/test/test_lpm6_data.h | 3 ++- app/test/test_lpm6_perf.c | 14 +------------- app/test/test_lpm_perf.c | 39 ++++++++++++--------------------------- lib/lpm/meson.build | 6 ------ 6 files changed, 16 insertions(+), 73 deletions(-)
app/test/test_lpm.c +1 -14
@@ -2,18 +2,6 @@* Copyright(c) 2010-2014 Intel Corporation */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm (void ) -{ -printf ("lpm not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <stdio.h> #include <stdint.h> #include <stdlib.h>
@@ -22,6 +10,7 @@ test_lpm(void)#include <rte_lpm.h> #include <rte_malloc.h> +#include "test.h" #include "test_xmmt_ops.h" #define TEST_LPM_ASSERT (cond) do { \
@@ -1595,6 +1584,4 @@ test_lpm(void)return global_status; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm_autotest, test_lpm);
app/test/test_lpm6.c +0 -12
@@ -10,16 +10,6 @@#include <string.h> #include <rte_memory.h> -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm6 (void ) -{ -printf ("lpm6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_lpm6.h> #include "test_lpm6_data.h"
@@ -1803,6 +1793,4 @@ test_lpm6(void)return global_status; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm6_autotest, test_lpm6);
app/test/test_lpm6_data.h +2 -1
@@ -6,6 +6,7 @@#include <stdint.h> #include <stdlib.h> +#include <rte_random.h> struct rules_tbl_entry {uint8_t ip[16 ];
@@ -1129,7 +1130,7 @@ static void generate_large_ips_table(int gen_expected_next_hop)for (i =0 ; i <NUM_IPS_ENTRIES ; i++) {for (j =0 ; j <16 ; j++) - large_ips_table[i].ip[j] =lrand48 (); + large_ips_table[i].ip[j] =rte_rand (); }for (k = j =0 , i =0 ; i <NUM_IPS_ENTRIES ; i++) {
app/test/test_lpm6_perf.c +1 -13
@@ -2,17 +2,6 @@* Copyright(c) 2010-2014 Intel Corporation */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm6_perf (void ) -{ -printf ("lpm6_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else #include <stdio.h> #include <stdint.h>
@@ -24,6 +13,7 @@ test_lpm6_perf(void)#include <rte_memory.h> #include <rte_lpm6.h> +#include "test.h" #include "test_lpm6_data.h" #define TEST_LPM_ASSERT (cond) do { \
@@ -171,6 +161,4 @@ test_lpm6_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm6_perf_autotest, test_lpm6_perf);
app/test/test_lpm_perf.c +12 -27
@@ -3,17 +3,6 @@* Copyright(c) 2020 Arm Limited */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm_perf (void ) -{ -printf ("lpm_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else #include <stdio.h> #include <stdint.h> #include <stdlib.h>
@@ -25,7 +14,9 @@ test_lpm_perf(void)#include <rte_malloc.h> #include <rte_ip.h> #include <rte_lpm.h> +#include <rte_spinlock.h> +#include "test.h" #include "test_xmmt_ops.h" struct rte_lpm *lpm;
@@ -34,8 +25,9 @@ static volatile uint8_t writer_done;static volatile uint32_t thr_id;static uint64_t gwrite_cycles;static uint32_t num_writers; -/* LPM APIs are not thread safe, use mutex to provide thread safety */ -static pthread_mutex_t lpm_mutex =PTHREAD_MUTEX_INITIALIZER ; + +/* LPM APIs are not thread safe, use spinlock */ +static rte_spinlock_t lpm_lock =RTE_SPINLOCK_INITIALIZER ;/* Report quiescent state interval every 1024 lookups. Larger critical * sections in reader will result in writer polling multiple times.
@@ -267,7 +259,7 @@ static void generate_random_rule_prefix(uint32_t ip_class, uint8_t depth)/* Only generate rest bits except the most significant * fixed bits for IP address class */ - start =lrand48 () & mask; + start =rte_rand () & mask; ptr_rule = &large_route_table[num_route_entries]; ptr_ldepth_rule = &large_ldepth_route_table[num_ldepth_route_entries];for (k =0 ; k < rule_num; k++) {
@@ -296,7 +288,7 @@ static void insert_rule_in_random_pos(uint32_t ip, uint8_t depth)struct route_rule tmp;do { - pos =lrand48 (); + pos =rte_rand (); try_count++; }while ((try_count <10 ) && (pos > num_route_entries));
@@ -452,8 +444,7 @@ test_lpm_rcu_qsbr_writer(void *arg)for (i =0 ; i <RCU_ITERATIONS ; i++) {/* Add all the entries */ for (j = si; j < ei; j++) { -if (num_writers >1 ) -pthread_mutex_lock (&lpm_mutex); +rte_spinlock_lock (&lpm_lock);if (rte_lpm_add (lpm, large_ldepth_route_table[j].ip, large_ldepth_route_table[j].depth, next_hop_add) !=0 ) {
@@ -461,22 +452,19 @@ test_lpm_rcu_qsbr_writer(void *arg) i, j); goto error; } -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock); }/* Delete all the entries */ for (j = si; j < ei; j++) { -if (num_writers >1 ) -pthread_mutex_lock (&lpm_mutex); +rte_spinlock_lock (&lpm_lock);if (rte_lpm_delete (lpm, large_ldepth_route_table[j].ip, large_ldepth_route_table[j].depth) !=0 ) {printf ("Failed to delete iteration %d, route# %d\n" , i, j); goto error; } -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock); } }
@@ -487,8 +475,7 @@ test_lpm_rcu_qsbr_writer(void *arg)return 0 ; error: -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock);return -1 ; }
@@ -773,6 +760,4 @@ test_lpm_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm_perf_autotest, test_lpm_perf);
lib/lpm/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_lpm.c', 'rte_lpm6.c') headers = files('rte_lpm.h', 'rte_lpm6.h') # since header files have different names, we can install all vector headers
Signature · 1 line
2.39.1
PATCH v2 4/8 reorder: build on Windows
In reply to Stephen HemmingerThis builds on Windows if sys/queue.h is included.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_reorder.c | 11 ----------- lib/reorder/meson.build | 6 ------ lib/reorder/rte_reorder.c | 1 + 3 files changed, 1 insertion(+), 17 deletions(-)
app/test/test_reorder.c +0 -11
@@ -11,16 +11,6 @@#include <rte_cycles.h> #include <rte_errno.h> #include <rte_mbuf.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_reorder (void ) -{ -printf ("reorder not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_reorder.h> #include <rte_lcore.h> #include <rte_malloc.h>
@@ -395,6 +385,5 @@ test_reorder(void)return unit_test_suite_runner (&reorder_test_suite); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ REGISTER_TEST_COMMAND (reorder_autotest, test_reorder);
lib/reorder/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_reorder.c') headers = files('rte_reorder.h') deps += ['mbuf']
lib/reorder/rte_reorder.c +1 -0
@@ -3,6 +3,7 @@*/ #include <string.h> +#include <sys/queue.h> #include <rte_string_fns.h> #include <rte_log.h>
Signature · 1 line
2.39.1
PATCH v2 5/8 ip_frag: enable build on Windows
In reply to Stephen HemmingerThis build works on Windows if sys/queue.h is included.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_ipfrag.c | 12 ------------ lib/ip_frag/ip_frag_common.h | 2 ++ lib/ip_frag/meson.build | 6 ------ 3 files changed, 2 insertions(+), 18 deletions(-)
app/test/test_ipfrag.c +0 -12
@@ -10,17 +10,6 @@#include <rte_cycles.h> #include <rte_hexdump.h> #include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_ipfrag (void ) -{ -printf ("ipfrag not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_ip_frag.h> #include <rte_mbuf.h> #include <rte_random.h>
@@ -520,6 +509,5 @@ test_ipfrag(void)return unit_test_suite_runner (&ipfrag_testsuite); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ REGISTER_TEST_COMMAND (ipfrag_autotest, test_ipfrag);
lib/ip_frag/ip_frag_common.h +2 -0
@@ -8,6 +8,8 @@#include "rte_ip_frag.h" #include "ip_reassembly.h" +#include <sys/queue.h> +/* logging macros. */ #ifdef RTE_LIBRTE_IP_FRAG_DEBUG #define IP_FRAG_LOG(lvl, fmt, args...) RTE_LOG(lvl, USER1, fmt, ##args)
lib/ip_frag/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files( 'rte_ipv4_fragmentation.c', 'rte_ipv6_fragmentation.c',
Signature · 1 line
2.39.1
PATCH v2 6/8 rib: enable on Windows
In reply to Stephen HemmingerThe RIB library builds on Windows as long as sys/queue.h is included
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_rib.c | 19 ------------------- app/test/test_rib6.c | 24 ++---------------------- lib/rib/meson.build | 6 ------ lib/rib/rte_rib.c | 1 + lib/rib/rte_rib6.c | 1 + 5 files changed, 4 insertions(+), 47 deletions(-)
app/test/test_rib.c +0 -19
@@ -10,23 +10,6 @@#include <stdlib.h> #include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_rib (void ) -{ -printf ("rib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_rib (void ) -{ -printf ("slow_rib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_rib.h> typedef int32_t (*rte_rib_test )(void );
@@ -380,7 +363,5 @@ test_slow_rib(void)return unit_test_suite_runner (&rib_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (rib_autotest, test_rib);REGISTER_TEST_COMMAND (rib_slow_autotest, test_slow_rib);
app/test/test_rib6.c +2 -22
@@ -3,32 +3,14 @@* Copyright(c) 2019 Intel Corporation */ -#include "test.h" -#include <stdio.h> #include <stdint.h> #include <stdlib.h> -#include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_rib6 (void ) -{ -printf ("rib6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_rib6 (void ) -{ -printf ("slow_rib6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_rib6.h> +#include "test.h" +typedef int32_t (*rte_rib6_test )(void );static int32_t test_create_invalid (void );
@@ -385,7 +367,5 @@ test_slow_rib6(void)return unit_test_suite_runner (&rib6_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (rib6_autotest, test_rib6);REGISTER_TEST_COMMAND (rib6_slow_autotest, test_slow_rib6);
lib/rib/meson.build +0 -6
@@ -2,12 +2,6 @@ # Copyright(c) 2018 Vladimir Medvedkin <[email protected]> # Copyright(c) 2019 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_rib.c', 'rte_rib6.c') headers = files('rte_rib.h', 'rte_rib6.h') deps += ['mempool']
lib/rib/rte_rib.c +1 -0
@@ -5,6 +5,7 @@#include <stdbool.h> #include <stdint.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
lib/rib/rte_rib6.c +1 -0
@@ -5,6 +5,7 @@#include <stdbool.h> #include <stdint.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
Signature · 1 line
2.39.1
PATCH v2 7/8 fib: enable on Windows
In reply to Stephen HemmingerThe FIB library builds on Windows as long as sys/queue.h is defined.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_fib.c | 22 +--------------------- app/test/test_fib6.c | 24 ++---------------------- app/test/test_fib6_perf.c | 15 +-------------- app/test/test_fib_perf.c | 19 +++---------------- lib/fib/meson.build | 6 ------ lib/fib/rte_fib.c | 1 + lib/fib/rte_fib6.c | 1 + 7 files changed, 9 insertions(+), 79 deletions(-)
app/test/test_fib.c +1 -21
@@ -9,28 +9,10 @@#include <rte_ip.h> #include <rte_log.h> +#include <rte_fib.h> #include "test.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib (void ) -{ -printf ("fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_fib (void ) -{ -printf ("slow_fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib.h> -typedef int32_t (*rte_fib_test )(void );static int32_t test_create_invalid (void );
@@ -433,7 +415,5 @@ test_slow_fib(void)return unit_test_suite_runner (&fib_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib_autotest, test_fib);REGISTER_TEST_COMMAND (fib_slow_autotest, test_slow_fib);
app/test/test_fib6.c +2 -22
@@ -9,29 +9,11 @@#include <rte_memory.h> #include <rte_log.h> - -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib6 (void ) -{ -printf ("fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_fib6 (void ) -{ -printf ("slow_fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_rib6.h> #include <rte_fib6.h> +#include "test.h" +typedef int32_t (*rte_fib6_test )(void );static int32_t test_create_invalid (void );
@@ -442,7 +424,5 @@ test_slow_fib6(void)return unit_test_suite_runner (&fib6_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib6_autotest, test_fib6);REGISTER_TEST_COMMAND (fib6_slow_autotest, test_slow_fib6);
app/test/test_fib6_perf.c +1 -14
@@ -10,21 +10,10 @@#include <rte_cycles.h> #include <rte_random.h> #include <rte_memory.h> +#include <rte_fib6.h> #include "test.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib6_perf (void ) -{ -printf ("fib6_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib6.h> -#include "test_lpm6_data.h" #define TEST_FIB_ASSERT (cond) do { \
@@ -167,6 +156,4 @@ test_fib6_perf(void)return 0 ; } -#endif /*ifdef RTE_EXEC_ENV_WINDOWS*/ -REGISTER_TEST_COMMAND (fib6_perf_autotest, test_fib6_perf);
app/test/test_fib_perf.c +3 -16
@@ -12,22 +12,11 @@#include <rte_random.h> #include <rte_branch_prediction.h> #include <rte_ip.h> +#include <rte_fib.h> #include "test.h" #include "test_xmmt_ops.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib_perf (void ) -{ -printf ("fib_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib.h> -#define TEST_FIB_ASSERT (cond) do { \ if (!(cond)) { \ printf("Error at line %d:\n", __LINE__); \
@@ -246,7 +235,7 @@ static void generate_random_rule_prefix(uint32_t ip_class, uint8_t depth)/* Only generate rest bits except the most significant * fixed bits for IP address class */ - start =lrand48 () & mask; + start =rte_rand () & mask; ptr_rule = &large_route_table[num_route_entries];for (k =0 ; k < rule_num; k++) { ptr_rule->ip = (start << (RTE_FIB_MAX_DEPTH - depth))
@@ -265,7 +254,7 @@ static void insert_rule_in_random_pos(uint32_t ip, uint8_t depth)struct route_rule tmp;do { - pos =lrand48 (); + pos =rte_rand (); try_count++; }while ((try_count <10 ) && (pos > num_route_entries));
@@ -420,6 +409,4 @@ test_fib_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib_perf_autotest, test_fib_perf);
lib/fib/meson.build +0 -6
@@ -2,12 +2,6 @@ # Copyright(c) 2018 Vladimir Medvedkin <[email protected]> # Copyright(c) 2019 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_fib.c', 'rte_fib6.c', 'dir24_8.c', 'trie.c') headers = files('rte_fib.h', 'rte_fib6.h') deps += ['rib']
lib/fib/rte_fib.c +1 -0
@@ -5,6 +5,7 @@#include <stdint.h> #include <string.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
lib/fib/rte_fib6.c +1 -0
@@ -5,6 +5,7 @@#include <stdint.h> #include <string.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_tailq.h>
Signature · 1 line
2.39.1
PATCH v2 8/8 pcapng: windows compatibility
In reply to Stephen HemmingerAllow building on Windows, need to provide some comparability wrappers for writev() and if_nametoindex.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- lib/pcapng/meson.build | 6 ----- lib/pcapng/rte_pcapng.c | 59 +++++++++++++++++++++++++++++++++++++++-- 2 files changed, 57 insertions(+), 8 deletions(-)
lib/pcapng/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2019 Microsoft Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_pcapng.c') headers = files('rte_pcapng.h')
lib/pcapng/rte_pcapng.c +57 -2
84 lines not on this page · raw message
Signature · 1 line
2.39.1
Re: [PATCH v2 0/8] Enable building more libraries on Windows
In reply to Stephen Hemminger2023-02-06 16:19 (UTC-0800), Stephen Hemminger:
Quoted 14 lines
While diagnosing some Windows cross build errors; noticed that lots of important DPDK libraries are not being built on Windows.
Stephen Hemminger (8):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on Windows
rib: enable on Windows
fib: enable on Windows
pcapng: windows compatibility
Quoted 1 line
v2 - fix unnecessary
PATH_MAXin net/ring driver
Looks like #include <rte_os_shim.h> is still needed in rte_eth_ring.h to avoid the fatal warning about strdup() with clang (MS CRT, actually). For the series,
Acked-by Dmitry Kozlyuk <[email protected]>
Re: [PATCH v2 0/8] Enable building more libraries on Windows
Hello Stephen,
Quoted 16 lines · Dmitry Kozlyuk
2023-02-06 16:19 (UTC-0800), Stephen Hemminger:
Quoted 14 lines
While diagnosing some Windows cross build errors; noticed that lots of important DPDK libraries are not being built on Windows.
Stephen Hemminger (8):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on Windows
rib: enable on Windows
fib: enable on Windows
pcapng: windows compatibility
Quoted 4 lines
Quoted 1 line
v2 - fix unnecessary
PATH_MAXin net/ring driverLooks like #include <
rte_os_shim.h> is still needed inrte_eth_ring.hto avoid the fatal warning aboutstrdup()with clang (MS CRT, actually).
We need a new revision.
Thanks.
Signature · 1 line
David Marchand
v3 cover Enable building more libraries on Windows
In reply to Stephen HemmingerWhile diagnosing some Windows cross build errors; noticed that lots of important DPDK libraries are not being built on Windows.
Stephen Hemminger (8):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on Windows
rib: enable on Windows
fib: enable on Windows
pcapng: windows compatibility
v3 - add rte_os_shim.h to rte_ring to fix use of
strdup() breaking with stricter MS compiler
v2 - fix unnecessary PATH_MAX in net/ring driver
Stephen Hemminger (8):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on Windows
rib: enable on Windows
fib: enable on Windows
pcapng: windows compatibility
app/test/meson.build | 2 +- app/test/test_fib.c | 22 +----------- app/test/test_fib6.c | 24 ++------------ app/test/test_fib6_perf.c | 15 +-------- app/test/test_fib_perf.c | 19 ++--------- app/test/test_ipfrag.c | 12 ------- app/test/test_lpm.c | 15 +-------- app/test/test_lpm6.c | 12 ------- app/test/test_lpm6_data.h | 3 +- app/test/test_lpm6_perf.c | 14 +------- app/test/test_lpm_perf.c | 39 +++++++--------------- app/test/test_reorder.c | 11 ------ app/test/test_rib.c | 19 ----------- app/test/test_rib6.c | 24 ++------------ drivers/net/null/meson.build | 6 ---- drivers/net/ring/meson.build | 6 ---- drivers/net/ring/rte_eth_ring.c | 4 ++- lib/fib/meson.build | 6 ---- lib/fib/rte_fib.c | 1 + lib/fib/rte_fib6.c | 1 + lib/ip_frag/ip_frag_common.h | 2 ++ lib/ip_frag/meson.build | 6 ---- lib/lpm/meson.build | 6 ---- lib/pcapng/meson.build | 6 ---- lib/pcapng/rte_pcapng.c | 59 +++++++++++++++++++++++++++++++-- lib/reorder/meson.build | 6 ---- lib/reorder/rte_reorder.c | 1 + lib/rib/meson.build | 6 ---- lib/rib/rte_rib.c | 1 + lib/rib/rte_rib6.c | 1 + 30 files changed, 93 insertions(+), 256 deletions(-)
Signature · 1 line
2.39.1
PATCH v3 1/8 net/null: build null PMD on Windows
Builds fine with current code, no changes needed.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- drivers/net/null/meson.build | 6 ------ 1 file changed, 6 deletions(-)
drivers/net/null/meson.build +0 -6
@@ -1,11 +1,5 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_eth_null.c') pmd_supports_disable_iova_as_pa = true
Signature · 1 line
2.39.1
PATCH v3 2/8 net/ring: build on Windows
In reply to Stephen HemmingerThe command line arguments are stored in node_action_pair and the name[] was sized to PATH_MAX which does not exist on Windows. Since the name is either "CREATE" or "ATTACH" it is not related to PATH_MAX (4096).
With this fix driver builds ok on windows, but need to modify the test meson build to skip the eventdev test on Windows.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/meson.build | 2 +- drivers/net/ring/meson.build | 6 ------ drivers/net/ring/rte_eth_ring.c | 4 +++- 3 files changed, 4 insertions(+), 8 deletions(-)
app/test/meson.build +1 -1
@@ -390,7 +390,7 @@ if dpdk_conf.has('RTE_NET_BOND') driver_test_names += 'link_bonding_mode4_autotest' endif endif -if dpdk_conf.has('RTE_NET_RING') +if dpdk_conf.has('RTE_LIB_EVENTDEV') and dpdk_conf.has('RTE_NET_RING') test_deps += 'net_ring' test_sources += 'test_pmd_ring_perf.c' test_sources += 'test_pmd_ring.c'
drivers/net/ring/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_eth_ring.c') headers = files('rte_eth_ring.h') pmd_supports_disable_iova_as_pa = true
drivers/net/ring/rte_eth_ring.c +3 -1
@@ -9,6 +9,7 @@#include <ethdev_driver.h> #include <rte_malloc.h> #include <rte_memcpy.h> +#include <rte_os_shim.h> #include <rte_string_fns.h> #include <bus_vdev_driver.h> #include <rte_kvargs.h>
@@ -17,6 +18,7 @@#define ETH_RING_NUMA_NODE_ACTION_ARG "nodeaction"#define ETH_RING_ACTION_CREATE "CREATE"#define ETH_RING_ACTION_ATTACH "ATTACH" +#define ETH_RING_ACTION_MAX_LEN 8/* CREATE | ACTION */ #define ETH_RING_INTERNAL_ARG "internal"#define ETH_RING_INTERNAL_ARG_MAX_LEN 19/* "0x..16chars..\0" */
@@ -539,7 +541,7 @@ eth_dev_ring_create(const char *name, }struct node_action_pair { -char name[PATH_MAX ]; +char name[ETH_RING_ACTION_MAX_LEN ];unsigned int node;enum dev_action action; };
Signature · 1 line
2.39.1
PATCH v3 3/8 lpm: enable on Windows
In reply to Stephen HemmingerThis builds on Windows without changes.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_lpm.c | 15 +-------------- app/test/test_lpm6.c | 12 ------------ app/test/test_lpm6_data.h | 3 ++- app/test/test_lpm6_perf.c | 14 +------------- app/test/test_lpm_perf.c | 39 ++++++++++++--------------------------- lib/lpm/meson.build | 6 ------ 6 files changed, 16 insertions(+), 73 deletions(-)
app/test/test_lpm.c +1 -14
@@ -2,18 +2,6 @@* Copyright(c) 2010-2014 Intel Corporation */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm (void ) -{ -printf ("lpm not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <stdio.h> #include <stdint.h> #include <stdlib.h>
@@ -22,6 +10,7 @@ test_lpm(void)#include <rte_lpm.h> #include <rte_malloc.h> +#include "test.h" #include "test_xmmt_ops.h" #define TEST_LPM_ASSERT (cond) do { \
@@ -1595,6 +1584,4 @@ test_lpm(void)return global_status; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm_autotest, test_lpm);
app/test/test_lpm6.c +0 -12
@@ -10,16 +10,6 @@#include <string.h> #include <rte_memory.h> -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm6 (void ) -{ -printf ("lpm6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_lpm6.h> #include "test_lpm6_data.h"
@@ -1803,6 +1793,4 @@ test_lpm6(void)return global_status; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm6_autotest, test_lpm6);
app/test/test_lpm6_data.h +2 -1
@@ -6,6 +6,7 @@#include <stdint.h> #include <stdlib.h> +#include <rte_random.h> struct rules_tbl_entry {uint8_t ip[16 ];
@@ -1129,7 +1130,7 @@ static void generate_large_ips_table(int gen_expected_next_hop)for (i =0 ; i <NUM_IPS_ENTRIES ; i++) {for (j =0 ; j <16 ; j++) - large_ips_table[i].ip[j] =lrand48 (); + large_ips_table[i].ip[j] =rte_rand (); }for (k = j =0 , i =0 ; i <NUM_IPS_ENTRIES ; i++) {
app/test/test_lpm6_perf.c +1 -13
@@ -2,17 +2,6 @@* Copyright(c) 2010-2014 Intel Corporation */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm6_perf (void ) -{ -printf ("lpm6_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else #include <stdio.h> #include <stdint.h>
@@ -24,6 +13,7 @@ test_lpm6_perf(void)#include <rte_memory.h> #include <rte_lpm6.h> +#include "test.h" #include "test_lpm6_data.h" #define TEST_LPM_ASSERT (cond) do { \
@@ -171,6 +161,4 @@ test_lpm6_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm6_perf_autotest, test_lpm6_perf);
app/test/test_lpm_perf.c +12 -27
@@ -3,17 +3,6 @@* Copyright(c) 2020 Arm Limited */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm_perf (void ) -{ -printf ("lpm_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else #include <stdio.h> #include <stdint.h> #include <stdlib.h>
@@ -25,7 +14,9 @@ test_lpm_perf(void)#include <rte_malloc.h> #include <rte_ip.h> #include <rte_lpm.h> +#include <rte_spinlock.h> +#include "test.h" #include "test_xmmt_ops.h" struct rte_lpm *lpm;
@@ -34,8 +25,9 @@ static volatile uint8_t writer_done;static volatile uint32_t thr_id;static uint64_t gwrite_cycles;static uint32_t num_writers; -/* LPM APIs are not thread safe, use mutex to provide thread safety */ -static pthread_mutex_t lpm_mutex =PTHREAD_MUTEX_INITIALIZER ; + +/* LPM APIs are not thread safe, use spinlock */ +static rte_spinlock_t lpm_lock =RTE_SPINLOCK_INITIALIZER ;/* Report quiescent state interval every 1024 lookups. Larger critical * sections in reader will result in writer polling multiple times.
@@ -267,7 +259,7 @@ static void generate_random_rule_prefix(uint32_t ip_class, uint8_t depth)/* Only generate rest bits except the most significant * fixed bits for IP address class */ - start =lrand48 () & mask; + start =rte_rand () & mask; ptr_rule = &large_route_table[num_route_entries]; ptr_ldepth_rule = &large_ldepth_route_table[num_ldepth_route_entries];for (k =0 ; k < rule_num; k++) {
@@ -296,7 +288,7 @@ static void insert_rule_in_random_pos(uint32_t ip, uint8_t depth)struct route_rule tmp;do { - pos =lrand48 (); + pos =rte_rand (); try_count++; }while ((try_count <10 ) && (pos > num_route_entries));
@@ -452,8 +444,7 @@ test_lpm_rcu_qsbr_writer(void *arg)for (i =0 ; i <RCU_ITERATIONS ; i++) {/* Add all the entries */ for (j = si; j < ei; j++) { -if (num_writers >1 ) -pthread_mutex_lock (&lpm_mutex); +rte_spinlock_lock (&lpm_lock);if (rte_lpm_add (lpm, large_ldepth_route_table[j].ip, large_ldepth_route_table[j].depth, next_hop_add) !=0 ) {
@@ -461,22 +452,19 @@ test_lpm_rcu_qsbr_writer(void *arg) i, j); goto error; } -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock); }/* Delete all the entries */ for (j = si; j < ei; j++) { -if (num_writers >1 ) -pthread_mutex_lock (&lpm_mutex); +rte_spinlock_lock (&lpm_lock);if (rte_lpm_delete (lpm, large_ldepth_route_table[j].ip, large_ldepth_route_table[j].depth) !=0 ) {printf ("Failed to delete iteration %d, route# %d\n" , i, j); goto error; } -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock); } }
@@ -487,8 +475,7 @@ test_lpm_rcu_qsbr_writer(void *arg)return 0 ; error: -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock);return -1 ; }
@@ -773,6 +760,4 @@ test_lpm_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm_perf_autotest, test_lpm_perf);
lib/lpm/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_lpm.c', 'rte_lpm6.c') headers = files('rte_lpm.h', 'rte_lpm6.h') # since header files have different names, we can install all vector headers
Signature · 1 line
2.39.1
PATCH v3 4/8 reorder: build on Windows
In reply to Stephen HemmingerThis builds on Windows if sys/queue.h is included.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_reorder.c | 11 ----------- lib/reorder/meson.build | 6 ------ lib/reorder/rte_reorder.c | 1 + 3 files changed, 1 insertion(+), 17 deletions(-)
app/test/test_reorder.c +0 -11
@@ -11,16 +11,6 @@#include <rte_cycles.h> #include <rte_errno.h> #include <rte_mbuf.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_reorder (void ) -{ -printf ("reorder not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_reorder.h> #include <rte_lcore.h> #include <rte_malloc.h>
@@ -395,6 +385,5 @@ test_reorder(void)return unit_test_suite_runner (&reorder_test_suite); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ REGISTER_TEST_COMMAND (reorder_autotest, test_reorder);
lib/reorder/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_reorder.c') headers = files('rte_reorder.h') deps += ['mbuf']
lib/reorder/rte_reorder.c +1 -0
@@ -3,6 +3,7 @@*/ #include <string.h> +#include <sys/queue.h> #include <rte_string_fns.h> #include <rte_log.h>
Signature · 1 line
2.39.1
PATCH v3 5/8 ip_frag: enable build on Windows
In reply to Stephen HemmingerThis build works on Windows if sys/queue.h is included.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_ipfrag.c | 12 ------------ lib/ip_frag/ip_frag_common.h | 2 ++ lib/ip_frag/meson.build | 6 ------ 3 files changed, 2 insertions(+), 18 deletions(-)
app/test/test_ipfrag.c +0 -12
@@ -10,17 +10,6 @@#include <rte_cycles.h> #include <rte_hexdump.h> #include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_ipfrag (void ) -{ -printf ("ipfrag not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_ip_frag.h> #include <rte_mbuf.h> #include <rte_random.h>
@@ -520,6 +509,5 @@ test_ipfrag(void)return unit_test_suite_runner (&ipfrag_testsuite); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ REGISTER_TEST_COMMAND (ipfrag_autotest, test_ipfrag);
lib/ip_frag/ip_frag_common.h +2 -0
@@ -8,6 +8,8 @@#include "rte_ip_frag.h" #include "ip_reassembly.h" +#include <sys/queue.h> +/* logging macros. */ #ifdef RTE_LIBRTE_IP_FRAG_DEBUG #define IP_FRAG_LOG(lvl, fmt, args...) RTE_LOG(lvl, USER1, fmt, ##args)
lib/ip_frag/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files( 'rte_ipv4_fragmentation.c', 'rte_ipv6_fragmentation.c',
Signature · 1 line
2.39.1
PATCH v3 6/8 rib: enable on Windows
In reply to Stephen HemmingerThe RIB library builds on Windows as long as sys/queue.h is included
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_rib.c | 19 ------------------- app/test/test_rib6.c | 24 ++---------------------- lib/rib/meson.build | 6 ------ lib/rib/rte_rib.c | 1 + lib/rib/rte_rib6.c | 1 + 5 files changed, 4 insertions(+), 47 deletions(-)
app/test/test_rib.c +0 -19
@@ -10,23 +10,6 @@#include <stdlib.h> #include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_rib (void ) -{ -printf ("rib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_rib (void ) -{ -printf ("slow_rib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_rib.h> typedef int32_t (*rte_rib_test )(void );
@@ -380,7 +363,5 @@ test_slow_rib(void)return unit_test_suite_runner (&rib_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (rib_autotest, test_rib);REGISTER_TEST_COMMAND (rib_slow_autotest, test_slow_rib);
app/test/test_rib6.c +2 -22
@@ -3,32 +3,14 @@* Copyright(c) 2019 Intel Corporation */ -#include "test.h" -#include <stdio.h> #include <stdint.h> #include <stdlib.h> -#include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_rib6 (void ) -{ -printf ("rib6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_rib6 (void ) -{ -printf ("slow_rib6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_rib6.h> +#include "test.h" +typedef int32_t (*rte_rib6_test )(void );static int32_t test_create_invalid (void );
@@ -385,7 +367,5 @@ test_slow_rib6(void)return unit_test_suite_runner (&rib6_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (rib6_autotest, test_rib6);REGISTER_TEST_COMMAND (rib6_slow_autotest, test_slow_rib6);
lib/rib/meson.build +0 -6
@@ -2,12 +2,6 @@ # Copyright(c) 2018 Vladimir Medvedkin <[email protected]> # Copyright(c) 2019 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_rib.c', 'rte_rib6.c') headers = files('rte_rib.h', 'rte_rib6.h') deps += ['mempool']
lib/rib/rte_rib.c +1 -0
@@ -5,6 +5,7 @@#include <stdbool.h> #include <stdint.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
lib/rib/rte_rib6.c +1 -0
@@ -5,6 +5,7 @@#include <stdbool.h> #include <stdint.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
Signature · 1 line
2.39.1
PATCH v3 7/8 fib: enable on Windows
In reply to Stephen HemmingerThe FIB library builds on Windows as long as sys/queue.h is defined.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_fib.c | 22 +--------------------- app/test/test_fib6.c | 24 ++---------------------- app/test/test_fib6_perf.c | 15 +-------------- app/test/test_fib_perf.c | 19 +++---------------- lib/fib/meson.build | 6 ------ lib/fib/rte_fib.c | 1 + lib/fib/rte_fib6.c | 1 + 7 files changed, 9 insertions(+), 79 deletions(-)
app/test/test_fib.c +1 -21
@@ -9,28 +9,10 @@#include <rte_ip.h> #include <rte_log.h> +#include <rte_fib.h> #include "test.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib (void ) -{ -printf ("fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_fib (void ) -{ -printf ("slow_fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib.h> -typedef int32_t (*rte_fib_test )(void );static int32_t test_create_invalid (void );
@@ -433,7 +415,5 @@ test_slow_fib(void)return unit_test_suite_runner (&fib_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib_autotest, test_fib);REGISTER_TEST_COMMAND (fib_slow_autotest, test_slow_fib);
app/test/test_fib6.c +2 -22
@@ -9,29 +9,11 @@#include <rte_memory.h> #include <rte_log.h> - -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib6 (void ) -{ -printf ("fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_fib6 (void ) -{ -printf ("slow_fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_rib6.h> #include <rte_fib6.h> +#include "test.h" +typedef int32_t (*rte_fib6_test )(void );static int32_t test_create_invalid (void );
@@ -442,7 +424,5 @@ test_slow_fib6(void)return unit_test_suite_runner (&fib6_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib6_autotest, test_fib6);REGISTER_TEST_COMMAND (fib6_slow_autotest, test_slow_fib6);
app/test/test_fib6_perf.c +1 -14
@@ -10,21 +10,10 @@#include <rte_cycles.h> #include <rte_random.h> #include <rte_memory.h> +#include <rte_fib6.h> #include "test.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib6_perf (void ) -{ -printf ("fib6_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib6.h> -#include "test_lpm6_data.h" #define TEST_FIB_ASSERT (cond) do { \
@@ -167,6 +156,4 @@ test_fib6_perf(void)return 0 ; } -#endif /*ifdef RTE_EXEC_ENV_WINDOWS*/ -REGISTER_TEST_COMMAND (fib6_perf_autotest, test_fib6_perf);
app/test/test_fib_perf.c +3 -16
@@ -12,22 +12,11 @@#include <rte_random.h> #include <rte_branch_prediction.h> #include <rte_ip.h> +#include <rte_fib.h> #include "test.h" #include "test_xmmt_ops.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib_perf (void ) -{ -printf ("fib_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib.h> -#define TEST_FIB_ASSERT (cond) do { \ if (!(cond)) { \ printf("Error at line %d:\n", __LINE__); \
@@ -246,7 +235,7 @@ static void generate_random_rule_prefix(uint32_t ip_class, uint8_t depth)/* Only generate rest bits except the most significant * fixed bits for IP address class */ - start =lrand48 () & mask; + start =rte_rand () & mask; ptr_rule = &large_route_table[num_route_entries];for (k =0 ; k < rule_num; k++) { ptr_rule->ip = (start << (RTE_FIB_MAX_DEPTH - depth))
@@ -265,7 +254,7 @@ static void insert_rule_in_random_pos(uint32_t ip, uint8_t depth)struct route_rule tmp;do { - pos =lrand48 (); + pos =rte_rand (); try_count++; }while ((try_count <10 ) && (pos > num_route_entries));
@@ -420,6 +409,4 @@ test_fib_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib_perf_autotest, test_fib_perf);
lib/fib/meson.build +0 -6
@@ -2,12 +2,6 @@ # Copyright(c) 2018 Vladimir Medvedkin <[email protected]> # Copyright(c) 2019 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_fib.c', 'rte_fib6.c', 'dir24_8.c', 'trie.c') headers = files('rte_fib.h', 'rte_fib6.h') deps += ['rib']
lib/fib/rte_fib.c +1 -0
@@ -5,6 +5,7 @@#include <stdint.h> #include <string.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
lib/fib/rte_fib6.c +1 -0
@@ -5,6 +5,7 @@#include <stdint.h> #include <string.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_tailq.h>
Signature · 1 line
2.39.1
PATCH v3 8/8 pcapng: windows compatibility
In reply to Stephen HemmingerAllow building on Windows, need to provide some comparability wrappers for writev() and if_nametoindex.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- lib/pcapng/meson.build | 6 ----- lib/pcapng/rte_pcapng.c | 59 +++++++++++++++++++++++++++++++++++++++-- 2 files changed, 57 insertions(+), 8 deletions(-)
lib/pcapng/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2019 Microsoft Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_pcapng.c') headers = files('rte_pcapng.h')
lib/pcapng/rte_pcapng.c +57 -2
84 lines not on this page · raw message
Signature · 1 line
2.39.1
v4 cover Enable building more libraries on Windows
In reply to Stephen HemmingerWhile diagnosing some Windows cross build errors; noticed that lots of important DPDK libraries are not being built on Windows.
Stephen Hemminger (8):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on Windows
rib: enable on Windows
fib: enable on Windows
pcapng: windows compatibility
v4 - fix spelling in commnet
enable build of lib/table as well
v3 - add rte_os_shim.h to rte_ring to fix use of
strdup() breaking with stricter MS compiler
v2 - fix unnecessary PATH_MAX in net/ring driver
Stephen Hemminger (9):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on Windows
rib: enable on Windows
fib: enable on Windows
pcapng: windows compatibility
table: enable build on Windows
app/test/meson.build | 2 +- app/test/test_fib.c | 22 +----------- app/test/test_fib6.c | 24 ++------------ app/test/test_fib6_perf.c | 15 +-------- app/test/test_fib_perf.c | 19 ++--------- app/test/test_ipfrag.c | 12 ------- app/test/test_lpm.c | 15 +-------- app/test/test_lpm6.c | 12 ------- app/test/test_lpm6_data.h | 3 +- app/test/test_lpm6_perf.c | 14 +------- app/test/test_lpm_perf.c | 39 +++++++--------------- app/test/test_reorder.c | 11 ------ app/test/test_rib.c | 19 ----------- app/test/test_rib6.c | 24 ++------------ drivers/net/null/meson.build | 6 ---- drivers/net/ring/meson.build | 6 ---- drivers/net/ring/rte_eth_ring.c | 4 ++- lib/fib/meson.build | 6 ---- lib/fib/rte_fib.c | 1 + lib/fib/rte_fib6.c | 1 + lib/ip_frag/ip_frag_common.h | 2 ++ lib/ip_frag/meson.build | 6 ---- lib/lpm/meson.build | 6 ---- lib/pcapng/meson.build | 6 ---- lib/pcapng/rte_pcapng.c | 59 +++++++++++++++++++++++++++++++-- lib/reorder/meson.build | 6 ---- lib/reorder/rte_reorder.c | 1 + lib/rib/meson.build | 6 ---- lib/rib/rte_rib.c | 1 + lib/rib/rte_rib6.c | 1 + lib/table/meson.build | 6 ---- 31 files changed, 93 insertions(+), 262 deletions(-)
Signature · 1 line
2.39.1
PATCH v4 1/9 net/null: build null PMD on Windows
Builds fine with current code, no changes needed.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- drivers/net/null/meson.build | 6 ------ 1 file changed, 6 deletions(-)
drivers/net/null/meson.build +0 -6
@@ -1,11 +1,5 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_eth_null.c') pmd_supports_disable_iova_as_pa = true
Signature · 1 line
2.39.1
PATCH v4 2/9 net/ring: build on Windows
In reply to Stephen HemmingerThe command line arguments are stored in node_action_pair and the name[] was sized to PATH_MAX which does not exist on Windows. Since the name is either "CREATE" or "ATTACH" it is not related to PATH_MAX (4096).
With this fix driver builds ok on windows, but need to modify the test meson build to skip the eventdev test on Windows.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/meson.build | 2 +- drivers/net/ring/meson.build | 6 ------ drivers/net/ring/rte_eth_ring.c | 4 +++- 3 files changed, 4 insertions(+), 8 deletions(-)
app/test/meson.build +1 -1
@@ -390,7 +390,7 @@ if dpdk_conf.has('RTE_NET_BOND') driver_test_names += 'link_bonding_mode4_autotest' endif endif -if dpdk_conf.has('RTE_NET_RING') +if dpdk_conf.has('RTE_LIB_EVENTDEV') and dpdk_conf.has('RTE_NET_RING') test_deps += 'net_ring' test_sources += 'test_pmd_ring_perf.c' test_sources += 'test_pmd_ring.c'
drivers/net/ring/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_eth_ring.c') headers = files('rte_eth_ring.h') pmd_supports_disable_iova_as_pa = true
drivers/net/ring/rte_eth_ring.c +3 -1
@@ -9,6 +9,7 @@#include <ethdev_driver.h> #include <rte_malloc.h> #include <rte_memcpy.h> +#include <rte_os_shim.h> #include <rte_string_fns.h> #include <bus_vdev_driver.h> #include <rte_kvargs.h>
@@ -17,6 +18,7 @@#define ETH_RING_NUMA_NODE_ACTION_ARG "nodeaction"#define ETH_RING_ACTION_CREATE "CREATE"#define ETH_RING_ACTION_ATTACH "ATTACH" +#define ETH_RING_ACTION_MAX_LEN 8/* CREATE | ACTION */ #define ETH_RING_INTERNAL_ARG "internal"#define ETH_RING_INTERNAL_ARG_MAX_LEN 19/* "0x..16chars..\0" */
@@ -539,7 +541,7 @@ eth_dev_ring_create(const char *name, }struct node_action_pair { -char name[PATH_MAX ]; +char name[ETH_RING_ACTION_MAX_LEN ];unsigned int node;enum dev_action action; };
Signature · 1 line
2.39.1
Re: [PATCH v4 2/9] net/ring: build on Windows
Quoted 52 lines · Stephen Hemminger
The command line arguments are stored in
node_action_pairand the name[] was sized toPATH_MAXwhich does not exist on Windows. Since the name is either "CREATE" or "ATTACH" it is not related toPATH_MAX(4096).With this fix driver builds ok on windows, but need to modify the test meson build to skip the eventdev test on Windows.
Signed-off-by Stephen Hemminger <[email protected]>Acked-by Tyler Retzlaff <[email protected]>--- app/test/meson.build | 2 +- drivers/net/ring/meson.build | 6 ------ drivers/net/ring/rte_eth_ring.c | 4 +++- 3 files changed, 4 insertions(+), 8 deletions(-)app/test/meson.build +1
-1@@ -390,7 +390,7 @@ if dpdk_conf.has('RTE_NET_BOND') driver_test_names += 'link_bonding_mode4_autotest' endif endif -if dpdk_conf.has('RTE_NET_RING') +if dpdk_conf.has('RTE_LIB_EVENTDEV') and dpdk_conf.has('RTE_NET_RING') test_deps += 'net_ring' test_sources += 'test_pmd_ring_perf.c' test_sources += 'test_pmd_ring.c'drivers/net/ring/meson.build +0
-6@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_eth_ring.c') headers = files('rte_eth_ring.h') pmd_supports_disable_iova_as_pa = truedrivers/net/ring/rte_eth_ring.c +0
-0
drivers/net/ring/rte_eth_ring.c line 12
Quoted 4 lines
drivers/net/ring/rte_eth_ring.c +1
-0@@ -9,6 +9,7 @@#include <ethdev_driver.h> #include <rte_malloc.h> #include <rte_memcpy.h> +#include <rte_os_shim.h>
Is it still needed, since this patch drops use of PATH_MAX?
Quoted 23 lines
drivers/net/ring/rte_eth_ring.c +2
-2#include <rte_string_fns.h> #include <bus_vdev_driver.h> #include <rte_kvargs.h> @@ -17,6 +18,7 @@#define ETH_RING_NUMA_NODE_ACTION_ARG "nodeaction"#define ETH_RING_ACTION_CREATE "CREATE"#define ETH_RING_ACTION_ATTACH "ATTACH" +#define ETH_RING_ACTION_MAX_LEN 8/* CREATE | ACTION */ #define ETH_RING_INTERNAL_ARG "internal"#define ETH_RING_INTERNAL_ARG_MAX_LEN 19/* "0x..16chars..\0" */ @@ -539,7 +541,7 @@ eth_dev_ring_create(const char *name, }struct node_action_pair { -char name[PATH_MAX ]; +char name[ETH_RING_ACTION_MAX_LEN ];unsigned int node;enum dev_action action; }; --2.39.1
Signature · 1 line
David Marchand
Re: [PATCH v4 2/9] net/ring: build on Windows
Quoted 11 lines · On Thu, 9 Mar 2023 22:10:03 +0100 David Marchand <[email protected]>…
drivers/net/ring/rte_eth_ring.c line 12
Quoted 9 lines
drivers/net/ring/rte_eth_ring.c +1
-0@@ -9,6 +9,7 @@#include <ethdev_driver.h> #include <rte_malloc.h> #include <rte_memcpy.h> +#include <rte_os_shim.h> Is it still needed, since this patch drops use of
PATH_MAX?
Let me submit a version without, I don't run windows just do cross builds.
Re: [PATCH v4 2/9] net/ring: build on Windows
Quoted 16 lines · Stephen Hemminger
Quoted 11 lines · On Thu, 9 Mar 2023 22:10:03 +0100 David Marchand <[email protected]>…
drivers/net/ring/rte_eth_ring.c line 12
Quoted 9 lines
drivers/net/ring/rte_eth_ring.c +1
-0@@ -9,6 +9,7 @@#include <ethdev_driver.h> #include <rte_malloc.h> #include <rte_memcpy.h> +#include <rte_os_shim.h> Is it still needed, since this patch drops use of
PATH_MAX?
Quoted 1 line
Let me submit a version without, I don't run windows just do cross builds.
Good thing you did.
So we still need this include for hack on strdup.
So the v4 series is enough.
If I manage to merge it today..., I'll simply update the commitlog.
libtmp_rte_net_ring.a.p" "-Idrivers" "-I..\drivers" "-Idrivers\net\ring" "-I..\drivers\net\ring" "-Ilib\ethdev" "-I..\lib\ethdev" "-I." "-I.." "-Iconfig" "-I..\config" "-Ilib\eal\include" "-I..\lib\eal\include" "-Ilib\eal\windows\include" "-I..\lib\eal\windows\include" "-Ilib\eal\x86\include" "-I..\lib\eal\x86\include" "-Ilib\eal\common" "-I..\lib\eal\common" "-Ilib\eal" "-I..\lib\eal" "-Ilib\kvargs" "-I..\lib\kvargs" "-Ilib\net" "-I..\lib\net" "-Ilib\mbuf" "-I..\lib\mbuf" "-Ilib\mempool" "-I..\lib\mempool" "-Ilib\ring" "-I..\lib\ring" "-Ilib\metrics" "-I..\lib\metrics" "-Ilib\telemetry" "-I..\lib\telemetry" "-Ilib\meter" "-I..\lib\meter" "-Idrivers\bus\pci" "-I..\drivers\bus\pci" "-I..\drivers\bus\pci\windows" "-Ilib\pci" "-I..\lib\pci" "-Idrivers\bus\vdev" "-I..\drivers\bus\vdev" "-IC:\Program Files\Mellanox\MLNX_WinOF2_DevX_SDK\inc" "-Xclang" "-fcolor-diagnostics" "-pipe" "-D_FILE_OFFSET_BITS=64" "-Wall" "-Winvalid-pch" "-Wextra" "-Werror" "-O3" "-include" "rte_config.h" "-Wcast-qual" "-Wdeprecated" "-Wformat" "-Wformat-nonliteral" "-Wformat-security" "-Wmissing-declarations" "-Wmissing-prototypes" "-Wnested-externs" "-Wold-style-definition" "-Wpointer-arith" "-Wsign-compare" "-Wstrict-prototypes" "-Wundef" "-Wwrite-strings" "-Wno-address-of-packed-member" "-Wno-missing-field-initializers" "-D_GNU_SOURCE" "-D_WIN32_WINNT=0x0A00" "-D_CRT_SECURE_NO_WARNINGS" "-march=native" "-DALLOW_EXPERIMENTAL_API" "-DALLOW_INTERNAL_API" "-DRTE_LOG_DEFAULT_LOGTYPE=pmd.net.ring" -MD -MQ drivers/libtmp_rte_net_ring.a.p/net_ring_rte_eth_ring.c.obj -MF "drivers\libtmp_rte_net_ring.a.p\net_ring_rte_eth_ring.c.obj.d" -o drivers/libtmp_rte_net_ring.a.p/net_ring_rte_eth_ring.c.obj "-c" ../drivers/net/ring/rte_eth_ring.c ../drivers/net/ring/rte_eth_ring.c:564:9: error: 'strdup' is
deprecated: The POSIX name for this item is deprecated. Instead, use the ISO C and C++ conformant name: _strdup. See online help for details. [-Werror,-Wdeprecated-declarations]
name = strdup(value);
^
C:\Program Files (x86)\Windows
Kits\10\Include\10.0.18362.0\ucrt\string.h:535:20: note: 'strdup' has
been explicitly marked deprecated here
_Check_return_ _CRT_NONSTDC_DEPRECATE(_strdup)
^
C:\Program Files (x86)\Windows
Kits\10\Include\10.0.18362.0\ucrt\corecrt.h:335:50: note: expanded
from macro '_CRT_NONSTDC_DEPRECATE'
#define _CRT_NONSTDC_DEPRECATE(_NewName) _CRT_DEPRECATE_TEXT(
\
^
C:\Program Files (x86)\Microsoft Visual
Studio\2019\BuildTools\VC\Tools\MSVC\14.29.30133\include\vcruntime.h:310:47:
note: expanded from macro '_CRT_DEPRECATE_TEXT'
#define _CRT_DEPRECATE_TEXT(_Text) __declspec(deprecated(_Text))
^1 error generated
Signature · 1 line
David Marchand
PATCH v4 3/9 lpm: enable on Windows
In reply to Stephen HemmingerThis builds on Windows without changes.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_lpm.c | 15 +-------------- app/test/test_lpm6.c | 12 ------------ app/test/test_lpm6_data.h | 3 ++- app/test/test_lpm6_perf.c | 14 +------------- app/test/test_lpm_perf.c | 39 ++++++++++++--------------------------- lib/lpm/meson.build | 6 ------ 6 files changed, 16 insertions(+), 73 deletions(-)
app/test/test_lpm.c +1 -14
@@ -2,18 +2,6 @@* Copyright(c) 2010-2014 Intel Corporation */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm (void ) -{ -printf ("lpm not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <stdio.h> #include <stdint.h> #include <stdlib.h>
@@ -22,6 +10,7 @@ test_lpm(void)#include <rte_lpm.h> #include <rte_malloc.h> +#include "test.h" #include "test_xmmt_ops.h" #define TEST_LPM_ASSERT (cond) do { \
@@ -1595,6 +1584,4 @@ test_lpm(void)return global_status; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm_autotest, test_lpm);
app/test/test_lpm6.c +0 -12
@@ -10,16 +10,6 @@#include <string.h> #include <rte_memory.h> -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm6 (void ) -{ -printf ("lpm6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_lpm6.h> #include "test_lpm6_data.h"
@@ -1803,6 +1793,4 @@ test_lpm6(void)return global_status; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm6_autotest, test_lpm6);
app/test/test_lpm6_data.h +2 -1
@@ -6,6 +6,7 @@#include <stdint.h> #include <stdlib.h> +#include <rte_random.h> struct rules_tbl_entry {uint8_t ip[16 ];
@@ -1129,7 +1130,7 @@ static void generate_large_ips_table(int gen_expected_next_hop)for (i =0 ; i <NUM_IPS_ENTRIES ; i++) {for (j =0 ; j <16 ; j++) - large_ips_table[i].ip[j] =lrand48 (); + large_ips_table[i].ip[j] =rte_rand (); }for (k = j =0 , i =0 ; i <NUM_IPS_ENTRIES ; i++) {
app/test/test_lpm6_perf.c +1 -13
@@ -2,17 +2,6 @@* Copyright(c) 2010-2014 Intel Corporation */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm6_perf (void ) -{ -printf ("lpm6_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else #include <stdio.h> #include <stdint.h>
@@ -24,6 +13,7 @@ test_lpm6_perf(void)#include <rte_memory.h> #include <rte_lpm6.h> +#include "test.h" #include "test_lpm6_data.h" #define TEST_LPM_ASSERT (cond) do { \
@@ -171,6 +161,4 @@ test_lpm6_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm6_perf_autotest, test_lpm6_perf);
app/test/test_lpm_perf.c +12 -27
@@ -3,17 +3,6 @@* Copyright(c) 2020 Arm Limited */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm_perf (void ) -{ -printf ("lpm_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else #include <stdio.h> #include <stdint.h> #include <stdlib.h>
@@ -25,7 +14,9 @@ test_lpm_perf(void)#include <rte_malloc.h> #include <rte_ip.h> #include <rte_lpm.h> +#include <rte_spinlock.h> +#include "test.h" #include "test_xmmt_ops.h" struct rte_lpm *lpm;
@@ -34,8 +25,9 @@ static volatile uint8_t writer_done;static volatile uint32_t thr_id;static uint64_t gwrite_cycles;static uint32_t num_writers; -/* LPM APIs are not thread safe, use mutex to provide thread safety */ -static pthread_mutex_t lpm_mutex =PTHREAD_MUTEX_INITIALIZER ; + +/* LPM APIs are not thread safe, use spinlock */ +static rte_spinlock_t lpm_lock =RTE_SPINLOCK_INITIALIZER ;/* Report quiescent state interval every 1024 lookups. Larger critical * sections in reader will result in writer polling multiple times.
@@ -267,7 +259,7 @@ static void generate_random_rule_prefix(uint32_t ip_class, uint8_t depth)/* Only generate rest bits except the most significant * fixed bits for IP address class */ - start =lrand48 () & mask; + start =rte_rand () & mask; ptr_rule = &large_route_table[num_route_entries]; ptr_ldepth_rule = &large_ldepth_route_table[num_ldepth_route_entries];for (k =0 ; k < rule_num; k++) {
@@ -296,7 +288,7 @@ static void insert_rule_in_random_pos(uint32_t ip, uint8_t depth)struct route_rule tmp;do { - pos =lrand48 (); + pos =rte_rand (); try_count++; }while ((try_count <10 ) && (pos > num_route_entries));
@@ -452,8 +444,7 @@ test_lpm_rcu_qsbr_writer(void *arg)for (i =0 ; i <RCU_ITERATIONS ; i++) {/* Add all the entries */ for (j = si; j < ei; j++) { -if (num_writers >1 ) -pthread_mutex_lock (&lpm_mutex); +rte_spinlock_lock (&lpm_lock);if (rte_lpm_add (lpm, large_ldepth_route_table[j].ip, large_ldepth_route_table[j].depth, next_hop_add) !=0 ) {
@@ -461,22 +452,19 @@ test_lpm_rcu_qsbr_writer(void *arg) i, j); goto error; } -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock); }/* Delete all the entries */ for (j = si; j < ei; j++) { -if (num_writers >1 ) -pthread_mutex_lock (&lpm_mutex); +rte_spinlock_lock (&lpm_lock);if (rte_lpm_delete (lpm, large_ldepth_route_table[j].ip, large_ldepth_route_table[j].depth) !=0 ) {printf ("Failed to delete iteration %d, route# %d\n" , i, j); goto error; } -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock); } }
@@ -487,8 +475,7 @@ test_lpm_rcu_qsbr_writer(void *arg)return 0 ; error: -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock);return -1 ; }
@@ -773,6 +760,4 @@ test_lpm_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm_perf_autotest, test_lpm_perf);
lib/lpm/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_lpm.c', 'rte_lpm6.c') headers = files('rte_lpm.h', 'rte_lpm6.h') # since header files have different names, we can install all vector headers
Signature · 1 line
2.39.1
PATCH v4 4/9 reorder: build on Windows
In reply to Stephen HemmingerThis builds on Windows if sys/queue.h is included.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_reorder.c | 11 ----------- lib/reorder/meson.build | 6 ------ lib/reorder/rte_reorder.c | 1 + 3 files changed, 1 insertion(+), 17 deletions(-)
app/test/test_reorder.c +0 -11
@@ -11,16 +11,6 @@#include <rte_cycles.h> #include <rte_errno.h> #include <rte_mbuf.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_reorder (void ) -{ -printf ("reorder not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_reorder.h> #include <rte_lcore.h> #include <rte_malloc.h>
@@ -395,6 +385,5 @@ test_reorder(void)return unit_test_suite_runner (&reorder_test_suite); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ REGISTER_TEST_COMMAND (reorder_autotest, test_reorder);
lib/reorder/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_reorder.c') headers = files('rte_reorder.h') deps += ['mbuf']
lib/reorder/rte_reorder.c +1 -0
@@ -3,6 +3,7 @@*/ #include <string.h> +#include <sys/queue.h> #include <rte_string_fns.h> #include <rte_log.h>
Signature · 1 line
2.39.1
PATCH v4 5/9 ip_frag: enable build on Windows
In reply to Stephen HemmingerThis build works on Windows if sys/queue.h is included.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_ipfrag.c | 12 ------------ lib/ip_frag/ip_frag_common.h | 2 ++ lib/ip_frag/meson.build | 6 ------ 3 files changed, 2 insertions(+), 18 deletions(-)
app/test/test_ipfrag.c +0 -12
@@ -10,17 +10,6 @@#include <rte_cycles.h> #include <rte_hexdump.h> #include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_ipfrag (void ) -{ -printf ("ipfrag not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_ip_frag.h> #include <rte_mbuf.h> #include <rte_random.h>
@@ -520,6 +509,5 @@ test_ipfrag(void)return unit_test_suite_runner (&ipfrag_testsuite); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ REGISTER_TEST_COMMAND (ipfrag_autotest, test_ipfrag);
lib/ip_frag/ip_frag_common.h +2 -0
@@ -8,6 +8,8 @@#include "rte_ip_frag.h" #include "ip_reassembly.h" +#include <sys/queue.h> +/* logging macros. */ #ifdef RTE_LIBRTE_IP_FRAG_DEBUG #define IP_FRAG_LOG(lvl, fmt, args...) RTE_LOG(lvl, USER1, fmt, ##args)
lib/ip_frag/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files( 'rte_ipv4_fragmentation.c', 'rte_ipv6_fragmentation.c',
Signature · 1 line
2.39.1
PATCH v4 6/9 rib: enable on Windows
In reply to Stephen HemmingerThe RIB library builds on Windows as long as sys/queue.h is included
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_rib.c | 19 ------------------- app/test/test_rib6.c | 24 ++---------------------- lib/rib/meson.build | 6 ------ lib/rib/rte_rib.c | 1 + lib/rib/rte_rib6.c | 1 + 5 files changed, 4 insertions(+), 47 deletions(-)
app/test/test_rib.c +0 -19
@@ -10,23 +10,6 @@#include <stdlib.h> #include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_rib (void ) -{ -printf ("rib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_rib (void ) -{ -printf ("slow_rib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_rib.h> typedef int32_t (*rte_rib_test )(void );
@@ -380,7 +363,5 @@ test_slow_rib(void)return unit_test_suite_runner (&rib_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (rib_autotest, test_rib);REGISTER_TEST_COMMAND (rib_slow_autotest, test_slow_rib);
app/test/test_rib6.c +2 -22
@@ -3,32 +3,14 @@* Copyright(c) 2019 Intel Corporation */ -#include "test.h" -#include <stdio.h> #include <stdint.h> #include <stdlib.h> -#include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_rib6 (void ) -{ -printf ("rib6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_rib6 (void ) -{ -printf ("slow_rib6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_rib6.h> +#include "test.h" +typedef int32_t (*rte_rib6_test )(void );static int32_t test_create_invalid (void );
@@ -385,7 +367,5 @@ test_slow_rib6(void)return unit_test_suite_runner (&rib6_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (rib6_autotest, test_rib6);REGISTER_TEST_COMMAND (rib6_slow_autotest, test_slow_rib6);
lib/rib/meson.build +0 -6
@@ -2,12 +2,6 @@ # Copyright(c) 2018 Vladimir Medvedkin <[email protected]> # Copyright(c) 2019 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_rib.c', 'rte_rib6.c') headers = files('rte_rib.h', 'rte_rib6.h') deps += ['mempool']
lib/rib/rte_rib.c +1 -0
@@ -5,6 +5,7 @@#include <stdbool.h> #include <stdint.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
lib/rib/rte_rib6.c +1 -0
@@ -5,6 +5,7 @@#include <stdbool.h> #include <stdint.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
Signature · 1 line
2.39.1
PATCH v4 7/9 fib: enable on Windows
In reply to Stephen HemmingerThe FIB library builds on Windows as long as sys/queue.h is defined.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_fib.c | 22 +--------------------- app/test/test_fib6.c | 24 ++---------------------- app/test/test_fib6_perf.c | 15 +-------------- app/test/test_fib_perf.c | 19 +++---------------- lib/fib/meson.build | 6 ------ lib/fib/rte_fib.c | 1 + lib/fib/rte_fib6.c | 1 + 7 files changed, 9 insertions(+), 79 deletions(-)
app/test/test_fib.c +1 -21
@@ -9,28 +9,10 @@#include <rte_ip.h> #include <rte_log.h> +#include <rte_fib.h> #include "test.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib (void ) -{ -printf ("fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_fib (void ) -{ -printf ("slow_fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib.h> -typedef int32_t (*rte_fib_test )(void );static int32_t test_create_invalid (void );
@@ -433,7 +415,5 @@ test_slow_fib(void)return unit_test_suite_runner (&fib_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib_autotest, test_fib);REGISTER_TEST_COMMAND (fib_slow_autotest, test_slow_fib);
app/test/test_fib6.c +2 -22
@@ -9,29 +9,11 @@#include <rte_memory.h> #include <rte_log.h> - -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib6 (void ) -{ -printf ("fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_fib6 (void ) -{ -printf ("slow_fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_rib6.h> #include <rte_fib6.h> +#include "test.h" +typedef int32_t (*rte_fib6_test )(void );static int32_t test_create_invalid (void );
@@ -442,7 +424,5 @@ test_slow_fib6(void)return unit_test_suite_runner (&fib6_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib6_autotest, test_fib6);REGISTER_TEST_COMMAND (fib6_slow_autotest, test_slow_fib6);
app/test/test_fib6_perf.c +1 -14
@@ -10,21 +10,10 @@#include <rte_cycles.h> #include <rte_random.h> #include <rte_memory.h> +#include <rte_fib6.h> #include "test.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib6_perf (void ) -{ -printf ("fib6_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib6.h> -#include "test_lpm6_data.h" #define TEST_FIB_ASSERT (cond) do { \
@@ -167,6 +156,4 @@ test_fib6_perf(void)return 0 ; } -#endif /*ifdef RTE_EXEC_ENV_WINDOWS*/ -REGISTER_TEST_COMMAND (fib6_perf_autotest, test_fib6_perf);
app/test/test_fib_perf.c +3 -16
@@ -12,22 +12,11 @@#include <rte_random.h> #include <rte_branch_prediction.h> #include <rte_ip.h> +#include <rte_fib.h> #include "test.h" #include "test_xmmt_ops.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib_perf (void ) -{ -printf ("fib_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib.h> -#define TEST_FIB_ASSERT (cond) do { \ if (!(cond)) { \ printf("Error at line %d:\n", __LINE__); \
@@ -246,7 +235,7 @@ static void generate_random_rule_prefix(uint32_t ip_class, uint8_t depth)/* Only generate rest bits except the most significant * fixed bits for IP address class */ - start =lrand48 () & mask; + start =rte_rand () & mask; ptr_rule = &large_route_table[num_route_entries];for (k =0 ; k < rule_num; k++) { ptr_rule->ip = (start << (RTE_FIB_MAX_DEPTH - depth))
@@ -265,7 +254,7 @@ static void insert_rule_in_random_pos(uint32_t ip, uint8_t depth)struct route_rule tmp;do { - pos =lrand48 (); + pos =rte_rand (); try_count++; }while ((try_count <10 ) && (pos > num_route_entries));
@@ -420,6 +409,4 @@ test_fib_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib_perf_autotest, test_fib_perf);
lib/fib/meson.build +0 -6
@@ -2,12 +2,6 @@ # Copyright(c) 2018 Vladimir Medvedkin <[email protected]> # Copyright(c) 2019 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_fib.c', 'rte_fib6.c', 'dir24_8.c', 'trie.c') headers = files('rte_fib.h', 'rte_fib6.h') deps += ['rib']
lib/fib/rte_fib.c +1 -0
@@ -5,6 +5,7 @@#include <stdint.h> #include <string.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
lib/fib/rte_fib6.c +1 -0
@@ -5,6 +5,7 @@#include <stdint.h> #include <string.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_tailq.h>
Signature · 1 line
2.39.1
PATCH v4 8/9 pcapng: windows compatibility
In reply to Stephen HemmingerAllow building on Windows, need to provide some comparability wrappers for writev() and if_nametoindex.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- lib/pcapng/meson.build | 6 ----- lib/pcapng/rte_pcapng.c | 59 +++++++++++++++++++++++++++++++++++++++-- 2 files changed, 57 insertions(+), 8 deletions(-)
lib/pcapng/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2019 Microsoft Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_pcapng.c') headers = files('rte_pcapng.h')
lib/pcapng/rte_pcapng.c +57 -2
84 lines not on this page · raw message
Signature · 1 line
2.39.1
PATCH v4 9/9 table: enable build on Windows
In reply to Stephen HemmingerThere doesn't seem to be anything keeping this from building on Windows.
Signed-off-by Stephen Hemminger <[email protected]>
--- lib/table/meson.build | 6 ------ 1 file changed, 6 deletions(-)
lib/table/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files( 'rte_swx_keycmp.c', 'rte_swx_table_em.c',
Signature · 1 line
2.39.1
Re: [PATCH v4 9/9] table: enable build on Windows
Quoted 4 lines · Stephen Hemminger
There doesn't seem to be anything keeping this from building on Windows.
Signed-off-by Stephen Hemminger <[email protected]>---
Acked-by Tyler Retzlaff <[email protected]>
this series looks good to me, i see no barrier to having it merged?
Re: [PATCH v4 0/9] Enable building more libraries on Windows
In reply to Stephen HemmingerQuoted 5 lines · Stephen Hemminger
While diagnosing some Windows cross build errors; noticed that lots of important DPDK libraries are not being built on Windows.
Quoted 9 lines
Stephen Hemminger (8):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on Windows
rib: enable on Windows
fib: enable on Windows
pcapng: windows compatibility
Tweaked commitlogs and applied.
Thanks.
Signature · 1 line
David Marchand
v5 cover Enable building more libraries on Windows
In reply to Stephen HemmingerWhile diagnosing some Windows cross build errors; noticed that lots of important DPDK libraries are not being built on Windows.
Stephen Hemminger (8):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on Windows
rib: enable on Windows
fib: enable on Windows
pcapng: windows compatibility
v5 - remove no longer needed rte_os_shim.h in ring
v4 - fix spelling in commnet
enable build of lib/table as well
v3 - add rte_os_shim.h to rte_ring to fix use of
strdup() breaking with stricter MS compiler
v2 - fix unnecessary PATH_MAX in net/ring driver
Stephen Hemminger (9):
net/null: build null PMD on Windows
net/ring: build on Windows
lpm: enable on Windows
reorder: build on Windows
ip_frag: enable build on Windows
rib: enable on Windows
fib: enable on Windows
pcapng: windows compatibility
table: enable build on Windows
app/test/meson.build | 2 +- app/test/test_fib.c | 22 +----------- app/test/test_fib6.c | 24 ++------------ app/test/test_fib6_perf.c | 15 +-------- app/test/test_fib_perf.c | 19 ++--------- app/test/test_ipfrag.c | 12 ------- app/test/test_lpm.c | 15 +-------- app/test/test_lpm6.c | 12 ------- app/test/test_lpm6_data.h | 3 +- app/test/test_lpm6_perf.c | 14 +------- app/test/test_lpm_perf.c | 39 +++++++--------------- app/test/test_reorder.c | 11 ------ app/test/test_rib.c | 19 ----------- app/test/test_rib6.c | 24 ++------------ drivers/net/null/meson.build | 6 ---- drivers/net/ring/meson.build | 6 ---- drivers/net/ring/rte_eth_ring.c | 3 +- lib/fib/meson.build | 6 ---- lib/fib/rte_fib.c | 1 + lib/fib/rte_fib6.c | 1 + lib/ip_frag/ip_frag_common.h | 2 ++ lib/ip_frag/meson.build | 6 ---- lib/lpm/meson.build | 6 ---- lib/pcapng/meson.build | 6 ---- lib/pcapng/rte_pcapng.c | 59 +++++++++++++++++++++++++++++++-- lib/reorder/meson.build | 6 ---- lib/reorder/rte_reorder.c | 1 + lib/rib/meson.build | 6 ---- lib/rib/rte_rib.c | 1 + lib/rib/rte_rib6.c | 1 + lib/table/meson.build | 6 ---- 31 files changed, 92 insertions(+), 262 deletions(-)
Signature · 1 line
2.39.2
PATCH v5 1/9 net/null: build null PMD on Windows
Builds fine with current code, no changes needed.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- drivers/net/null/meson.build | 6 ------ 1 file changed, 6 deletions(-)
drivers/net/null/meson.build +0 -6
@@ -1,11 +1,5 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_eth_null.c') pmd_supports_disable_iova_as_pa = true
Signature · 1 line
2.39.2
PATCH v5 2/9 net/ring: build on Windows
In reply to Stephen HemmingerThe command line arguments are stored in node_action_pair and the name[] was sized to PATH_MAX which does not exist on Windows. Since the name is either "CREATE" or "ATTACH" it is not related to PATH_MAX (4096).
With this fix driver builds ok on windows, but need to modify the test meson build to skip the eventdev test on Windows.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/meson.build | 2 +- drivers/net/ring/meson.build | 6 ------ drivers/net/ring/rte_eth_ring.c | 3 ++- 3 files changed, 3 insertions(+), 8 deletions(-)
app/test/meson.build +1 -1
@@ -391,7 +391,7 @@ if dpdk_conf.has('RTE_NET_BOND') driver_test_names += 'link_bonding_mode4_autotest' endif endif -if dpdk_conf.has('RTE_NET_RING') +if dpdk_conf.has('RTE_LIB_EVENTDEV') and dpdk_conf.has('RTE_NET_RING') test_deps += 'net_ring' test_sources += 'test_pmd_ring_perf.c' test_sources += 'test_pmd_ring.c'
drivers/net/ring/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_eth_ring.c') headers = files('rte_eth_ring.h') pmd_supports_disable_iova_as_pa = true
drivers/net/ring/rte_eth_ring.c +2 -1
@@ -17,6 +17,7 @@#define ETH_RING_NUMA_NODE_ACTION_ARG "nodeaction"#define ETH_RING_ACTION_CREATE "CREATE"#define ETH_RING_ACTION_ATTACH "ATTACH" +#define ETH_RING_ACTION_MAX_LEN 8/* CREATE | ACTION */ #define ETH_RING_INTERNAL_ARG "internal"#define ETH_RING_INTERNAL_ARG_MAX_LEN 19/* "0x..16chars..\0" */
@@ -539,7 +540,7 @@ eth_dev_ring_create(const char *name, }struct node_action_pair { -char name[PATH_MAX ]; +char name[ETH_RING_ACTION_MAX_LEN ];unsigned int node;enum dev_action action; };
Signature · 1 line
2.39.2
PATCH v5 3/9 lpm: enable on Windows
In reply to Stephen HemmingerThis builds on Windows without changes.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_lpm.c | 15 +-------------- app/test/test_lpm6.c | 12 ------------ app/test/test_lpm6_data.h | 3 ++- app/test/test_lpm6_perf.c | 14 +------------- app/test/test_lpm_perf.c | 39 ++++++++++++--------------------------- lib/lpm/meson.build | 6 ------ 6 files changed, 16 insertions(+), 73 deletions(-)
app/test/test_lpm.c +1 -14
@@ -2,18 +2,6 @@* Copyright(c) 2010-2014 Intel Corporation */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm (void ) -{ -printf ("lpm not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <stdio.h> #include <stdint.h> #include <stdlib.h>
@@ -22,6 +10,7 @@ test_lpm(void)#include <rte_lpm.h> #include <rte_malloc.h> +#include "test.h" #include "test_xmmt_ops.h" #define TEST_LPM_ASSERT (cond) do { \
@@ -1595,6 +1584,4 @@ test_lpm(void)return global_status; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm_autotest, test_lpm);
app/test/test_lpm6.c +0 -12
@@ -10,16 +10,6 @@#include <string.h> #include <rte_memory.h> -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm6 (void ) -{ -printf ("lpm6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_lpm6.h> #include "test_lpm6_data.h"
@@ -1803,6 +1793,4 @@ test_lpm6(void)return global_status; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm6_autotest, test_lpm6);
app/test/test_lpm6_data.h +2 -1
@@ -6,6 +6,7 @@#include <stdint.h> #include <stdlib.h> +#include <rte_random.h> struct rules_tbl_entry {uint8_t ip[16 ];
@@ -1129,7 +1130,7 @@ static void generate_large_ips_table(int gen_expected_next_hop)for (i =0 ; i <NUM_IPS_ENTRIES ; i++) {for (j =0 ; j <16 ; j++) - large_ips_table[i].ip[j] =lrand48 (); + large_ips_table[i].ip[j] =rte_rand (); }for (k = j =0 , i =0 ; i <NUM_IPS_ENTRIES ; i++) {
app/test/test_lpm6_perf.c +1 -13
@@ -2,17 +2,6 @@* Copyright(c) 2010-2014 Intel Corporation */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm6_perf (void ) -{ -printf ("lpm6_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else #include <stdio.h> #include <stdint.h>
@@ -24,6 +13,7 @@ test_lpm6_perf(void)#include <rte_memory.h> #include <rte_lpm6.h> +#include "test.h" #include "test_lpm6_data.h" #define TEST_LPM_ASSERT (cond) do { \
@@ -171,6 +161,4 @@ test_lpm6_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm6_perf_autotest, test_lpm6_perf);
app/test/test_lpm_perf.c +12 -27
@@ -3,17 +3,6 @@* Copyright(c) 2020 Arm Limited */ -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_lpm_perf (void ) -{ -printf ("lpm_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else #include <stdio.h> #include <stdint.h> #include <stdlib.h>
@@ -25,7 +14,9 @@ test_lpm_perf(void)#include <rte_malloc.h> #include <rte_ip.h> #include <rte_lpm.h> +#include <rte_spinlock.h> +#include "test.h" #include "test_xmmt_ops.h" struct rte_lpm *lpm;
@@ -34,8 +25,9 @@ static volatile uint8_t writer_done;static volatile uint32_t thr_id;static uint64_t gwrite_cycles;static uint32_t num_writers; -/* LPM APIs are not thread safe, use mutex to provide thread safety */ -static pthread_mutex_t lpm_mutex =PTHREAD_MUTEX_INITIALIZER ; + +/* LPM APIs are not thread safe, use spinlock */ +static rte_spinlock_t lpm_lock =RTE_SPINLOCK_INITIALIZER ;/* Report quiescent state interval every 1024 lookups. Larger critical * sections in reader will result in writer polling multiple times.
@@ -267,7 +259,7 @@ static void generate_random_rule_prefix(uint32_t ip_class, uint8_t depth)/* Only generate rest bits except the most significant * fixed bits for IP address class */ - start =lrand48 () & mask; + start =rte_rand () & mask; ptr_rule = &large_route_table[num_route_entries]; ptr_ldepth_rule = &large_ldepth_route_table[num_ldepth_route_entries];for (k =0 ; k < rule_num; k++) {
@@ -296,7 +288,7 @@ static void insert_rule_in_random_pos(uint32_t ip, uint8_t depth)struct route_rule tmp;do { - pos =lrand48 (); + pos =rte_rand (); try_count++; }while ((try_count <10 ) && (pos > num_route_entries));
@@ -452,8 +444,7 @@ test_lpm_rcu_qsbr_writer(void *arg)for (i =0 ; i <RCU_ITERATIONS ; i++) {/* Add all the entries */ for (j = si; j < ei; j++) { -if (num_writers >1 ) -pthread_mutex_lock (&lpm_mutex); +rte_spinlock_lock (&lpm_lock);if (rte_lpm_add (lpm, large_ldepth_route_table[j].ip, large_ldepth_route_table[j].depth, next_hop_add) !=0 ) {
@@ -461,22 +452,19 @@ test_lpm_rcu_qsbr_writer(void *arg) i, j); goto error; } -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock); }/* Delete all the entries */ for (j = si; j < ei; j++) { -if (num_writers >1 ) -pthread_mutex_lock (&lpm_mutex); +rte_spinlock_lock (&lpm_lock);if (rte_lpm_delete (lpm, large_ldepth_route_table[j].ip, large_ldepth_route_table[j].depth) !=0 ) {printf ("Failed to delete iteration %d, route# %d\n" , i, j); goto error; } -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock); } }
@@ -487,8 +475,7 @@ test_lpm_rcu_qsbr_writer(void *arg)return 0 ; error: -if (num_writers >1 ) -pthread_mutex_unlock (&lpm_mutex); +rte_spinlock_unlock (&lpm_lock);return -1 ; }
@@ -773,6 +760,4 @@ test_lpm_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (lpm_perf_autotest, test_lpm_perf);
lib/lpm/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_lpm.c', 'rte_lpm6.c') headers = files('rte_lpm.h', 'rte_lpm6.h') # since header files have different names, we can install all vector headers
Signature · 1 line
2.39.2
PATCH v5 4/9 reorder: build on Windows
In reply to Stephen HemmingerThis builds on Windows if sys/queue.h is included.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_reorder.c | 11 ----------- lib/reorder/meson.build | 6 ------ lib/reorder/rte_reorder.c | 1 + 3 files changed, 1 insertion(+), 17 deletions(-)
app/test/test_reorder.c +0 -11
@@ -11,16 +11,6 @@#include <rte_cycles.h> #include <rte_errno.h> #include <rte_mbuf.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_reorder (void ) -{ -printf ("reorder not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_reorder.h> #include <rte_lcore.h> #include <rte_malloc.h>
@@ -557,6 +547,5 @@ test_reorder(void)return unit_test_suite_runner (&reorder_test_suite); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ REGISTER_TEST_COMMAND (reorder_autotest, test_reorder);
lib/reorder/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_reorder.c') headers = files('rte_reorder.h') deps += ['mbuf']
lib/reorder/rte_reorder.c +1 -0
@@ -3,6 +3,7 @@*/ #include <string.h> +#include <sys/queue.h> #include <rte_string_fns.h> #include <rte_log.h>
Signature · 1 line
2.39.2
PATCH v5 5/9 ip_frag: enable build on Windows
In reply to Stephen HemmingerThis build works on Windows if sys/queue.h is included.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_ipfrag.c | 12 ------------ lib/ip_frag/ip_frag_common.h | 2 ++ lib/ip_frag/meson.build | 6 ------ 3 files changed, 2 insertions(+), 18 deletions(-)
app/test/test_ipfrag.c +0 -12
@@ -10,17 +10,6 @@#include <rte_cycles.h> #include <rte_hexdump.h> #include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_ipfrag (void ) -{ -printf ("ipfrag not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_ip_frag.h> #include <rte_mbuf.h> #include <rte_random.h>
@@ -520,6 +509,5 @@ test_ipfrag(void)return unit_test_suite_runner (&ipfrag_testsuite); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ REGISTER_TEST_COMMAND (ipfrag_autotest, test_ipfrag);
lib/ip_frag/ip_frag_common.h +2 -0
@@ -8,6 +8,8 @@#include "rte_ip_frag.h" #include "ip_reassembly.h" +#include <sys/queue.h> +/* logging macros. */ #ifdef RTE_LIBRTE_IP_FRAG_DEBUG #define IP_FRAG_LOG(lvl, fmt, args...) RTE_LOG(lvl, USER1, fmt, ##args)
lib/ip_frag/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files( 'rte_ipv4_fragmentation.c', 'rte_ipv6_fragmentation.c',
Signature · 1 line
2.39.2
PATCH v5 6/9 rib: enable on Windows
In reply to Stephen HemmingerThe RIB library builds on Windows as long as sys/queue.h is included
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_rib.c | 19 ------------------- app/test/test_rib6.c | 24 ++---------------------- lib/rib/meson.build | 6 ------ lib/rib/rte_rib.c | 1 + lib/rib/rte_rib6.c | 1 + 5 files changed, 4 insertions(+), 47 deletions(-)
app/test/test_rib.c +0 -19
@@ -10,23 +10,6 @@#include <stdlib.h> #include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_rib (void ) -{ -printf ("rib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_rib (void ) -{ -printf ("slow_rib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_rib.h> typedef int32_t (*rte_rib_test )(void );
@@ -380,7 +363,5 @@ test_slow_rib(void)return unit_test_suite_runner (&rib_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (rib_autotest, test_rib);REGISTER_TEST_COMMAND (rib_slow_autotest, test_slow_rib);
app/test/test_rib6.c +2 -22
@@ -3,32 +3,14 @@* Copyright(c) 2019 Intel Corporation */ -#include "test.h" -#include <stdio.h> #include <stdint.h> #include <stdlib.h> -#include <rte_ip.h> - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_rib6 (void ) -{ -printf ("rib6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_rib6 (void ) -{ -printf ("slow_rib6 not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} -#else -#include <rte_rib6.h> +#include "test.h" +typedef int32_t (*rte_rib6_test )(void );static int32_t test_create_invalid (void );
@@ -385,7 +367,5 @@ test_slow_rib6(void)return unit_test_suite_runner (&rib6_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (rib6_autotest, test_rib6);REGISTER_TEST_COMMAND (rib6_slow_autotest, test_slow_rib6);
lib/rib/meson.build +0 -6
@@ -2,12 +2,6 @@ # Copyright(c) 2018 Vladimir Medvedkin <[email protected]> # Copyright(c) 2019 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_rib.c', 'rte_rib6.c') headers = files('rte_rib.h', 'rte_rib6.h') deps += ['mempool']
lib/rib/rte_rib.c +1 -0
@@ -5,6 +5,7 @@#include <stdbool.h> #include <stdint.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
lib/rib/rte_rib6.c +1 -0
@@ -5,6 +5,7 @@#include <stdbool.h> #include <stdint.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
Signature · 1 line
2.39.2
PATCH v5 7/9 fib: enable on Windows
In reply to Stephen HemmingerThe FIB library builds on Windows as long as sys/queue.h is defined.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- app/test/test_fib.c | 22 +--------------------- app/test/test_fib6.c | 24 ++---------------------- app/test/test_fib6_perf.c | 15 +-------------- app/test/test_fib_perf.c | 19 +++---------------- lib/fib/meson.build | 6 ------ lib/fib/rte_fib.c | 1 + lib/fib/rte_fib6.c | 1 + 7 files changed, 9 insertions(+), 79 deletions(-)
app/test/test_fib.c +1 -21
@@ -9,28 +9,10 @@#include <rte_ip.h> #include <rte_log.h> +#include <rte_fib.h> #include "test.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib (void ) -{ -printf ("fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_fib (void ) -{ -printf ("slow_fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib.h> -typedef int32_t (*rte_fib_test )(void );static int32_t test_create_invalid (void );
@@ -433,7 +415,5 @@ test_slow_fib(void)return unit_test_suite_runner (&fib_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib_autotest, test_fib);REGISTER_TEST_COMMAND (fib_slow_autotest, test_slow_fib);
app/test/test_fib6.c +2 -22
@@ -9,29 +9,11 @@#include <rte_memory.h> #include <rte_log.h> - -#include "test.h" - -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib6 (void ) -{ -printf ("fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -static int -test_slow_fib6 (void ) -{ -printf ("slow_fib not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else -#include <rte_rib6.h> #include <rte_fib6.h> +#include "test.h" +typedef int32_t (*rte_fib6_test )(void );static int32_t test_create_invalid (void );
@@ -442,7 +424,5 @@ test_slow_fib6(void)return unit_test_suite_runner (&fib6_slow_tests); } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib6_autotest, test_fib6);REGISTER_TEST_COMMAND (fib6_slow_autotest, test_slow_fib6);
app/test/test_fib6_perf.c +1 -14
@@ -10,21 +10,10 @@#include <rte_cycles.h> #include <rte_random.h> #include <rte_memory.h> +#include <rte_fib6.h> #include "test.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib6_perf (void ) -{ -printf ("fib6_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib6.h> -#include "test_lpm6_data.h" #define TEST_FIB_ASSERT (cond) do { \
@@ -167,6 +156,4 @@ test_fib6_perf(void)return 0 ; } -#endif /*ifdef RTE_EXEC_ENV_WINDOWS*/ -REGISTER_TEST_COMMAND (fib6_perf_autotest, test_fib6_perf);
app/test/test_fib_perf.c +3 -16
@@ -12,22 +12,11 @@#include <rte_random.h> #include <rte_branch_prediction.h> #include <rte_ip.h> +#include <rte_fib.h> #include "test.h" #include "test_xmmt_ops.h" -#ifdef RTE_EXEC_ENV_WINDOWS -static int -test_fib_perf (void ) -{ -printf ("fib_perf not supported on Windows, skipping test\n" ); -return TEST_SKIPPED ; -} - -#else - -#include <rte_fib.h> -#define TEST_FIB_ASSERT (cond) do { \ if (!(cond)) { \ printf("Error at line %d:\n", __LINE__); \
@@ -246,7 +235,7 @@ static void generate_random_rule_prefix(uint32_t ip_class, uint8_t depth)/* Only generate rest bits except the most significant * fixed bits for IP address class */ - start =lrand48 () & mask; + start =rte_rand () & mask; ptr_rule = &large_route_table[num_route_entries];for (k =0 ; k < rule_num; k++) { ptr_rule->ip = (start << (RTE_FIB_MAX_DEPTH - depth))
@@ -265,7 +254,7 @@ static void insert_rule_in_random_pos(uint32_t ip, uint8_t depth)struct route_rule tmp;do { - pos =lrand48 (); + pos =rte_rand (); try_count++; }while ((try_count <10 ) && (pos > num_route_entries));
@@ -420,6 +409,4 @@ test_fib_perf(void)return 0 ; } -#endif /* !RTE_EXEC_ENV_WINDOWS */ -REGISTER_TEST_COMMAND (fib_perf_autotest, test_fib_perf);
lib/fib/meson.build +0 -6
@@ -2,12 +2,6 @@ # Copyright(c) 2018 Vladimir Medvedkin <[email protected]> # Copyright(c) 2019 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_fib.c', 'rte_fib6.c', 'dir24_8.c', 'trie.c') headers = files('rte_fib.h', 'rte_fib6.h') deps += ['rib']
lib/fib/rte_fib.c +1 -0
@@ -5,6 +5,7 @@#include <stdint.h> #include <string.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_errno.h>
lib/fib/rte_fib6.c +1 -0
@@ -5,6 +5,7 @@#include <stdint.h> #include <string.h> +#include <sys/queue.h> #include <rte_eal_memconfig.h> #include <rte_tailq.h>
Signature · 1 line
2.39.2
PATCH v5 8/9 pcapng: windows compatibility
In reply to Stephen HemmingerAllow building on Windows, need to provide some comparability wrappers for writev() and if_nametoindex.
Signed-off-by Stephen Hemminger <[email protected]>
Acked-by Tyler Retzlaff <[email protected]>
--- lib/pcapng/meson.build | 6 ----- lib/pcapng/rte_pcapng.c | 59 +++++++++++++++++++++++++++++++++++++++-- 2 files changed, 57 insertions(+), 8 deletions(-)
lib/pcapng/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2019 Microsoft Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files('rte_pcapng.c') headers = files('rte_pcapng.h')
lib/pcapng/rte_pcapng.c +57 -2
84 lines not on this page · raw message
Signature · 1 line
2.39.2
PATCH v5 9/9 table: enable build on Windows
In reply to Stephen HemmingerThere doesn't seem to be anything keeping this from building on Windows.
Signed-off-by Stephen Hemminger <[email protected]>
--- lib/table/meson.build | 6 ------ 1 file changed, 6 deletions(-)
lib/table/meson.build +0 -6
@@ -1,12 +1,6 @@ # SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2017 Intel Corporation -if is_windows - build = false - reason = 'not supported on Windows' - subdir_done() -endif - sources = files( 'rte_swx_keycmp.c', 'rte_swx_table_em.c',
Signature · 1 line
2.39.2