{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":18383364,"defaultBranch":"main","name":"ovs","ownerLogin":"openvswitch","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2014-04-02T22:15:28.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/7143863?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1712754171.0","currentOid":""},"activityList":{"items":[{"before":"755b66f68e8675b02e25ec7662b3800d3df1c6b1","after":"9d0de176c778a9f5566639585b57b1df4795a7f6","ref":"refs/heads/branch-2.17","pushedAt":"2024-05-03T14:08:15.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"ofproto-dpif-trace: Fix access to an out-of-scope stack memory.\n\nWhile tracing NAT actions, pointer to the action may be stored in the\nrecirculation node for future reference. However, while translating\nactions for the group bucket in xlate_group_bucket, the action list is\nallocated temporarily on stack. So, in case the group translation\nleads to NAT, the stack pointer can be stored in the recirculation node\nand accessed later by the tracing mechanism when this stack memory is\nlong gone:\n\n ==396230==ERROR: AddressSanitizer: stack-use-after-return on address\n 0x191844 at pc 0x64222a bp 0xa5da10 sp 0xa5da08\n READ of size 1 at 0x191844 thread T0\n 0 0x642229 in ofproto_trace_recirc_node ofproto/ofproto-dpif-trace.c:704:49\n 1 0x642229 in ofproto_trace ofproto/ofproto-dpif-trace.c:867:9\n 2 0x6434c1 in ofproto_unixctl_trace ofproto/ofproto-dpif-trace.c:489:9\n 3 0xc1e491 in process_command lib/unixctl.c:310:13\n 4 0xc1e491 in run_connection lib/unixctl.c:344:17\n 5 0xc1e491 in unixctl_server_run lib/unixctl.c:395:21\n 6 0x53eedf in main ovs/vswitchd/ovs-vswitchd.c:131:9\n 7 0x2be087 in __libc_start_call_main\n 8 0x2be14a in __libc_start_main@GLIBC_2.2.5\n 9 0x42dee4 in _start (vswitchd/ovs-vswitchd+0x42dee4)\n\n Address 0x191844 is located in stack of thread T0 at offset 68 in frame\n 0 0x6d391f in xlate_group_bucket ofproto/ofproto-dpif-xlate.c:4751\n\n This frame has 3 object(s):\n [32, 1056) 'action_list_stub' (line 4760) <== Memory access at\n offset 68 is inside\n this variable\n [1184, 1248) 'action_list' (line 4761)\n [1280, 1344) 'action_set' (line 4762)\n\n SUMMARY: AddressSanitizer: stack-use-after-return\n ofproto/ofproto-dpif-trace.c:704:49 in ofproto_trace_recirc_node\n\nFix that by copying the action.\n\nFixes: d072d2de011b (\"ofproto-dpif-trace: Improve NAT tracing.\")\nReported-by: Ales Musil \nReviewed-by: Adrian Moreno \nAcked-by: Eelco Chaudron \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ofproto-dpif-trace: Fix access to an out-of-scope stack memory."}},{"before":"c111cadf1c73aff58fb50799a759e12901fb168c","after":"3194d6d74e5f3afa5bbb5674e8ec2c417246fc48","ref":"refs/heads/branch-3.0","pushedAt":"2024-05-03T14:08:13.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"ofproto-dpif-trace: Fix access to an out-of-scope stack memory.\n\nWhile tracing NAT actions, pointer to the action may be stored in the\nrecirculation node for future reference. However, while translating\nactions for the group bucket in xlate_group_bucket, the action list is\nallocated temporarily on stack. So, in case the group translation\nleads to NAT, the stack pointer can be stored in the recirculation node\nand accessed later by the tracing mechanism when this stack memory is\nlong gone:\n\n ==396230==ERROR: AddressSanitizer: stack-use-after-return on address\n 0x191844 at pc 0x64222a bp 0xa5da10 sp 0xa5da08\n READ of size 1 at 0x191844 thread T0\n 0 0x642229 in ofproto_trace_recirc_node ofproto/ofproto-dpif-trace.c:704:49\n 1 0x642229 in ofproto_trace ofproto/ofproto-dpif-trace.c:867:9\n 2 0x6434c1 in ofproto_unixctl_trace ofproto/ofproto-dpif-trace.c:489:9\n 3 0xc1e491 in process_command lib/unixctl.c:310:13\n 4 0xc1e491 in run_connection lib/unixctl.c:344:17\n 5 0xc1e491 in unixctl_server_run lib/unixctl.c:395:21\n 6 0x53eedf in main ovs/vswitchd/ovs-vswitchd.c:131:9\n 7 0x2be087 in __libc_start_call_main\n 8 0x2be14a in __libc_start_main@GLIBC_2.2.5\n 9 0x42dee4 in _start (vswitchd/ovs-vswitchd+0x42dee4)\n\n Address 0x191844 is located in stack of thread T0 at offset 68 in frame\n 0 0x6d391f in xlate_group_bucket ofproto/ofproto-dpif-xlate.c:4751\n\n This frame has 3 object(s):\n [32, 1056) 'action_list_stub' (line 4760) <== Memory access at\n offset 68 is inside\n this variable\n [1184, 1248) 'action_list' (line 4761)\n [1280, 1344) 'action_set' (line 4762)\n\n SUMMARY: AddressSanitizer: stack-use-after-return\n ofproto/ofproto-dpif-trace.c:704:49 in ofproto_trace_recirc_node\n\nFix that by copying the action.\n\nFixes: d072d2de011b (\"ofproto-dpif-trace: Improve NAT tracing.\")\nReported-by: Ales Musil \nReviewed-by: Adrian Moreno \nAcked-by: Eelco Chaudron \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ofproto-dpif-trace: Fix access to an out-of-scope stack memory."}},{"before":"1937dbe20fc5200cb78c25810f8eae775a22a4cb","after":"59135a75ee4342ae2db08cc30cffcba0e40836a5","ref":"refs/heads/branch-3.1","pushedAt":"2024-05-03T14:08:10.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"ofproto-dpif-trace: Fix access to an out-of-scope stack memory.\n\nWhile tracing NAT actions, pointer to the action may be stored in the\nrecirculation node for future reference. However, while translating\nactions for the group bucket in xlate_group_bucket, the action list is\nallocated temporarily on stack. So, in case the group translation\nleads to NAT, the stack pointer can be stored in the recirculation node\nand accessed later by the tracing mechanism when this stack memory is\nlong gone:\n\n ==396230==ERROR: AddressSanitizer: stack-use-after-return on address\n 0x191844 at pc 0x64222a bp 0xa5da10 sp 0xa5da08\n READ of size 1 at 0x191844 thread T0\n 0 0x642229 in ofproto_trace_recirc_node ofproto/ofproto-dpif-trace.c:704:49\n 1 0x642229 in ofproto_trace ofproto/ofproto-dpif-trace.c:867:9\n 2 0x6434c1 in ofproto_unixctl_trace ofproto/ofproto-dpif-trace.c:489:9\n 3 0xc1e491 in process_command lib/unixctl.c:310:13\n 4 0xc1e491 in run_connection lib/unixctl.c:344:17\n 5 0xc1e491 in unixctl_server_run lib/unixctl.c:395:21\n 6 0x53eedf in main ovs/vswitchd/ovs-vswitchd.c:131:9\n 7 0x2be087 in __libc_start_call_main\n 8 0x2be14a in __libc_start_main@GLIBC_2.2.5\n 9 0x42dee4 in _start (vswitchd/ovs-vswitchd+0x42dee4)\n\n Address 0x191844 is located in stack of thread T0 at offset 68 in frame\n 0 0x6d391f in xlate_group_bucket ofproto/ofproto-dpif-xlate.c:4751\n\n This frame has 3 object(s):\n [32, 1056) 'action_list_stub' (line 4760) <== Memory access at\n offset 68 is inside\n this variable\n [1184, 1248) 'action_list' (line 4761)\n [1280, 1344) 'action_set' (line 4762)\n\n SUMMARY: AddressSanitizer: stack-use-after-return\n ofproto/ofproto-dpif-trace.c:704:49 in ofproto_trace_recirc_node\n\nFix that by copying the action.\n\nFixes: d072d2de011b (\"ofproto-dpif-trace: Improve NAT tracing.\")\nReported-by: Ales Musil \nReviewed-by: Adrian Moreno \nAcked-by: Eelco Chaudron \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ofproto-dpif-trace: Fix access to an out-of-scope stack memory."}},{"before":"03c165e43633aafd67284d89bb647054bb902cf5","after":"a38ff0a233c4fa5187c8d3363e97ee52aec1e71b","ref":"refs/heads/branch-3.2","pushedAt":"2024-05-03T14:08:08.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"ofproto-dpif-trace: Fix access to an out-of-scope stack memory.\n\nWhile tracing NAT actions, pointer to the action may be stored in the\nrecirculation node for future reference. However, while translating\nactions for the group bucket in xlate_group_bucket, the action list is\nallocated temporarily on stack. So, in case the group translation\nleads to NAT, the stack pointer can be stored in the recirculation node\nand accessed later by the tracing mechanism when this stack memory is\nlong gone:\n\n ==396230==ERROR: AddressSanitizer: stack-use-after-return on address\n 0x191844 at pc 0x64222a bp 0xa5da10 sp 0xa5da08\n READ of size 1 at 0x191844 thread T0\n 0 0x642229 in ofproto_trace_recirc_node ofproto/ofproto-dpif-trace.c:704:49\n 1 0x642229 in ofproto_trace ofproto/ofproto-dpif-trace.c:867:9\n 2 0x6434c1 in ofproto_unixctl_trace ofproto/ofproto-dpif-trace.c:489:9\n 3 0xc1e491 in process_command lib/unixctl.c:310:13\n 4 0xc1e491 in run_connection lib/unixctl.c:344:17\n 5 0xc1e491 in unixctl_server_run lib/unixctl.c:395:21\n 6 0x53eedf in main ovs/vswitchd/ovs-vswitchd.c:131:9\n 7 0x2be087 in __libc_start_call_main\n 8 0x2be14a in __libc_start_main@GLIBC_2.2.5\n 9 0x42dee4 in _start (vswitchd/ovs-vswitchd+0x42dee4)\n\n Address 0x191844 is located in stack of thread T0 at offset 68 in frame\n 0 0x6d391f in xlate_group_bucket ofproto/ofproto-dpif-xlate.c:4751\n\n This frame has 3 object(s):\n [32, 1056) 'action_list_stub' (line 4760) <== Memory access at\n offset 68 is inside\n this variable\n [1184, 1248) 'action_list' (line 4761)\n [1280, 1344) 'action_set' (line 4762)\n\n SUMMARY: AddressSanitizer: stack-use-after-return\n ofproto/ofproto-dpif-trace.c:704:49 in ofproto_trace_recirc_node\n\nFix that by copying the action.\n\nFixes: d072d2de011b (\"ofproto-dpif-trace: Improve NAT tracing.\")\nReported-by: Ales Musil \nReviewed-by: Adrian Moreno \nAcked-by: Eelco Chaudron \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ofproto-dpif-trace: Fix access to an out-of-scope stack memory."}},{"before":"9a5c24d70fb6b004034c748146de81bd8edc47f9","after":"4756bf4baf1e3e05b2526064be5d8e6965f97b76","ref":"refs/heads/branch-3.3","pushedAt":"2024-05-03T14:08:06.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"ofproto-dpif-trace: Fix access to an out-of-scope stack memory.\n\nWhile tracing NAT actions, pointer to the action may be stored in the\nrecirculation node for future reference. However, while translating\nactions for the group bucket in xlate_group_bucket, the action list is\nallocated temporarily on stack. So, in case the group translation\nleads to NAT, the stack pointer can be stored in the recirculation node\nand accessed later by the tracing mechanism when this stack memory is\nlong gone:\n\n ==396230==ERROR: AddressSanitizer: stack-use-after-return on address\n 0x191844 at pc 0x64222a bp 0xa5da10 sp 0xa5da08\n READ of size 1 at 0x191844 thread T0\n 0 0x642229 in ofproto_trace_recirc_node ofproto/ofproto-dpif-trace.c:704:49\n 1 0x642229 in ofproto_trace ofproto/ofproto-dpif-trace.c:867:9\n 2 0x6434c1 in ofproto_unixctl_trace ofproto/ofproto-dpif-trace.c:489:9\n 3 0xc1e491 in process_command lib/unixctl.c:310:13\n 4 0xc1e491 in run_connection lib/unixctl.c:344:17\n 5 0xc1e491 in unixctl_server_run lib/unixctl.c:395:21\n 6 0x53eedf in main ovs/vswitchd/ovs-vswitchd.c:131:9\n 7 0x2be087 in __libc_start_call_main\n 8 0x2be14a in __libc_start_main@GLIBC_2.2.5\n 9 0x42dee4 in _start (vswitchd/ovs-vswitchd+0x42dee4)\n\n Address 0x191844 is located in stack of thread T0 at offset 68 in frame\n 0 0x6d391f in xlate_group_bucket ofproto/ofproto-dpif-xlate.c:4751\n\n This frame has 3 object(s):\n [32, 1056) 'action_list_stub' (line 4760) <== Memory access at\n offset 68 is inside\n this variable\n [1184, 1248) 'action_list' (line 4761)\n [1280, 1344) 'action_set' (line 4762)\n\n SUMMARY: AddressSanitizer: stack-use-after-return\n ofproto/ofproto-dpif-trace.c:704:49 in ofproto_trace_recirc_node\n\nFix that by copying the action.\n\nFixes: d072d2de011b (\"ofproto-dpif-trace: Improve NAT tracing.\")\nReported-by: Ales Musil \nReviewed-by: Adrian Moreno \nAcked-by: Eelco Chaudron \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ofproto-dpif-trace: Fix access to an out-of-scope stack memory."}},{"before":"139b564dbd0a1934a4a135cc0bfb700249afd572","after":"b91f6788c4be0dd35b9f5edae14f372d68fced08","ref":"refs/heads/main","pushedAt":"2024-05-03T14:07:58.000Z","pushType":"push","commitsCount":4,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"ofproto-dpif-trace: Fix access to an out-of-scope stack memory.\n\nWhile tracing NAT actions, pointer to the action may be stored in the\nrecirculation node for future reference. However, while translating\nactions for the group bucket in xlate_group_bucket, the action list is\nallocated temporarily on stack. So, in case the group translation\nleads to NAT, the stack pointer can be stored in the recirculation node\nand accessed later by the tracing mechanism when this stack memory is\nlong gone:\n\n ==396230==ERROR: AddressSanitizer: stack-use-after-return on address\n 0x191844 at pc 0x64222a bp 0xa5da10 sp 0xa5da08\n READ of size 1 at 0x191844 thread T0\n 0 0x642229 in ofproto_trace_recirc_node ofproto/ofproto-dpif-trace.c:704:49\n 1 0x642229 in ofproto_trace ofproto/ofproto-dpif-trace.c:867:9\n 2 0x6434c1 in ofproto_unixctl_trace ofproto/ofproto-dpif-trace.c:489:9\n 3 0xc1e491 in process_command lib/unixctl.c:310:13\n 4 0xc1e491 in run_connection lib/unixctl.c:344:17\n 5 0xc1e491 in unixctl_server_run lib/unixctl.c:395:21\n 6 0x53eedf in main ovs/vswitchd/ovs-vswitchd.c:131:9\n 7 0x2be087 in __libc_start_call_main\n 8 0x2be14a in __libc_start_main@GLIBC_2.2.5\n 9 0x42dee4 in _start (vswitchd/ovs-vswitchd+0x42dee4)\n\n Address 0x191844 is located in stack of thread T0 at offset 68 in frame\n 0 0x6d391f in xlate_group_bucket ofproto/ofproto-dpif-xlate.c:4751\n\n This frame has 3 object(s):\n [32, 1056) 'action_list_stub' (line 4760) <== Memory access at\n offset 68 is inside\n this variable\n [1184, 1248) 'action_list' (line 4761)\n [1280, 1344) 'action_set' (line 4762)\n\n SUMMARY: AddressSanitizer: stack-use-after-return\n ofproto/ofproto-dpif-trace.c:704:49 in ofproto_trace_recirc_node\n\nFix that by copying the action.\n\nFixes: d072d2de011b (\"ofproto-dpif-trace: Improve NAT tracing.\")\nReported-by: Ales Musil \nReviewed-by: Adrian Moreno \nAcked-by: Eelco Chaudron \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ofproto-dpif-trace: Fix access to an out-of-scope stack memory."}},{"before":"bf82d2cce2d94443fbd40a32023eeb1d29c48e1b","after":"139b564dbd0a1934a4a135cc0bfb700249afd572","ref":"refs/heads/main","pushedAt":"2024-05-03T12:05:01.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"horms","name":"Simon Horman","path":"/horms","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/65151?s=80&v=4"},"commit":{"message":"conntrack: Key connections by zone.\n\nCurrently conntrack uses a single large cmap for all connections stored.\nThis cmap contains all connections for all conntrack zones which are\ncompletely separate from each other. By separating each zone to its own\ncmap we can significantly optimize the performance when using multiple\nzones.\n\nThe change fixes a similar issue as [1] where slow conntrack zone flush\noperations significantly slow down OVN router failover. The difference is\njust that this fix is used whith dpdk, while [1] was when using the ovs\nkernel module.\n\nAs we now need to store more cmap's the memory usage of struct conntrack\nincreases by 524280 bytes. Additionally we need 65535 cmaps with 128\nbytes each. This leads to a total memory increase of around 10MB.\n\nRunning \"./ovstest test-conntrack benchmark 4 33554432 32 1\" shows no\nreal difference in the multithreading behaviour against a single zone.\n\nRunning the new \"./ovstest test-conntrack benchmark-zones\" show\nsignificant speedups as shown below. The values for \"ct execute\" are for\nacting on the complete zone with all its entries in total (so in the\nfirst case adding 10,000 new conntrack entries). All tests are run 1000\ntimes.\n\nWhen running with 1,000 zones with 10,000 entries each we see the\nfollowing results (all in microseconds):\n\"./ovstest test-conntrack benchmark-zones 10000 1000 1000\"\n\n +------+--------+---------+---------+\n | Min | Max | 95%ile | Avg |\n+------------------------+------+--------+---------+---------+\n| ct execute (commit) | | | | |\n| with commit | 2266 | 3505 | 2707.06 | 2592.06 |\n| without commit | 2411 | 12730 | 4432.50 | 2736.78 |\n+------------------------+------+--------+---------+---------+\n| ct execute (no commit) | | | | |\n| with commit | 699 | 1238 | 886.15 | 722.67 |\n| without commit | 700 | 3377 | 1934.42 | 803.53 |\n+------------------------+------+--------+---------+---------+\n| flush full zone | | | | |\n| with commit | 619 | 1122 | 901.36 | 679.15 |\n| without commit | 618 | 105078 | 64591 | 2886.46 |\n+------------------------+------+--------+---------+---------+\n| flush empty zone | | | | |\n| with commit | 0 | 5 | 1.00 | 0.64 |\n| without commit | 54 | 87469 | 64520 | 2172.25 |\n+------------------------+------+--------+---------+---------+\n\nWhen running with 10,000 zones with 1,000 entries each we see the\nfollowing results (all in microseconds):\n\"./ovstest test-conntrack benchmark-zones 1000 10000 1000\"\n\n +------+--------+---------+---------+\n | Min | Max | 95%ile | Avg |\n+------------------------+------+--------+---------+---------+\n| ct execute (commit) | | | | |\n| with commit | 215 | 287 | 231.88 | 222.30 |\n| without commit | 214 | 1692 | 569.18 | 285.83 |\n+------------------------+------+--------+---------+---------+\n| ct execute (no commit) | | | | |\n| with commit | 68 | 97 | 74.69 | 70.09 |\n| without commit | 68 | 300 | 158.40 | 82.06 |\n+------------------------+------+--------+---------+---------+\n| flush full zone | | | | |\n| with commit | 47 | 211 | 56.34 | 50.34 |\n| without commit | 48 | 96330 | 63392 | 63923 |\n+------------------------+------+--------+---------+---------+\n| flush empty zone | | | | |\n| with commit | 0 | 1 | 1.00 | 0.44 |\n| without commit | 3 | 109728 | 63923 | 3629.44 |\n+------------------------+------+--------+---------+---------+\n\nComparing the averages we see:\n* a moderate performance improvement for conntrack_execute with or\n without commiting of around 6% to 23%\n* a significant performance improvement for flushing a full zone of\n around 75% to 99%\n* an even more significant improvement for flushing empty zones since we\n no longer need to check any unrelated connections\n\n[1] 9ec849e8aa869b646c372fac552ae2609a4b5f66\n\nSigned-off-by: Felix Huettner \nSigned-off-by: Simon Horman ","shortMessageHtmlLink":"conntrack: Key connections by zone."}},{"before":"bd8e9f48f180800292c10e12f26824833f18506a","after":"bf82d2cce2d94443fbd40a32023eeb1d29c48e1b","ref":"refs/heads/main","pushedAt":"2024-05-03T11:36:12.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"horms","name":"Simon Horman","path":"/horms","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/65151?s=80&v=4"},"commit":{"message":"Documentation: Update Pacemaker main page link.\n\nUpdate link to pacemaker main page as the existing link is broken.\nAlso, use HTTPS.\n\nBroken link flagged by make check-docs\n\nAcked-by: Eelco Chaudron \nSigned-off-by: Simon Horman ","shortMessageHtmlLink":"Documentation: Update Pacemaker main page link."}},{"before":"b2e19110394e65125079b2c614454d4b4bd1d8e1","after":"755b66f68e8675b02e25ec7662b3800d3df1c6b1","ref":"refs/heads/branch-2.17","pushedAt":"2024-04-30T19:01:02.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"sparse: Add immintrin.h header.\n\nSparse doesn't understand _Float16 and some other types used by\nimmintrin.h from GCC 13. This breaks sparse builds with DPDK on\nFedora 38+ and Ubuntu 24.04.\n\nAdd another sparse-specific header to workaround the problem. We do\nneed some of the functions and types defined in these headers, so we\ncan't really stab out the whole header. Carving out the main offenders\ninstead by defining the inclusion guards.\n\nThis is fragile and depends on internals of immintrin and underlying\nheaders, but I'm not sure what the better way to solve the issue\nwould be. This approach should be more or less portable between\ncompilers, because it only defines a few specific variables. We may\nhave to add more as GCC headers change over time.\n\nThis fixes the build with a following config on F38 and Ubuntu 24.04:\n\n ./configure --enable-sparse --with-dpdk=yes --enable-Werror\n\nAcked-by: Ales Musil \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"sparse: Add immintrin.h header."}},{"before":"e5f237e9a95d6dc97eca8b819f17b8b3b909a9f7","after":"c111cadf1c73aff58fb50799a759e12901fb168c","ref":"refs/heads/branch-3.0","pushedAt":"2024-04-30T19:00:59.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"sparse: Add immintrin.h header.\n\nSparse doesn't understand _Float16 and some other types used by\nimmintrin.h from GCC 13. This breaks sparse builds with DPDK on\nFedora 38+ and Ubuntu 24.04.\n\nAdd another sparse-specific header to workaround the problem. We do\nneed some of the functions and types defined in these headers, so we\ncan't really stab out the whole header. Carving out the main offenders\ninstead by defining the inclusion guards.\n\nThis is fragile and depends on internals of immintrin and underlying\nheaders, but I'm not sure what the better way to solve the issue\nwould be. This approach should be more or less portable between\ncompilers, because it only defines a few specific variables. We may\nhave to add more as GCC headers change over time.\n\nThis fixes the build with a following config on F38 and Ubuntu 24.04:\n\n ./configure --enable-sparse --with-dpdk=yes --enable-Werror\n\nAcked-by: Ales Musil \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"sparse: Add immintrin.h header."}},{"before":"6ce0130f12399b14fd27260cc186bc0bf0187928","after":"1937dbe20fc5200cb78c25810f8eae775a22a4cb","ref":"refs/heads/branch-3.1","pushedAt":"2024-04-30T19:00:57.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"sparse: Add immintrin.h header.\n\nSparse doesn't understand _Float16 and some other types used by\nimmintrin.h from GCC 13. This breaks sparse builds with DPDK on\nFedora 38+ and Ubuntu 24.04.\n\nAdd another sparse-specific header to workaround the problem. We do\nneed some of the functions and types defined in these headers, so we\ncan't really stab out the whole header. Carving out the main offenders\ninstead by defining the inclusion guards.\n\nThis is fragile and depends on internals of immintrin and underlying\nheaders, but I'm not sure what the better way to solve the issue\nwould be. This approach should be more or less portable between\ncompilers, because it only defines a few specific variables. We may\nhave to add more as GCC headers change over time.\n\nThis fixes the build with a following config on F38 and Ubuntu 24.04:\n\n ./configure --enable-sparse --with-dpdk=yes --enable-Werror\n\nAcked-by: Ales Musil \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"sparse: Add immintrin.h header."}},{"before":"4e4463ca55392ca841f95bf51c06b1e03d6c5492","after":"03c165e43633aafd67284d89bb647054bb902cf5","ref":"refs/heads/branch-3.2","pushedAt":"2024-04-30T19:00:54.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"sparse: Add immintrin.h header.\n\nSparse doesn't understand _Float16 and some other types used by\nimmintrin.h from GCC 13. This breaks sparse builds with DPDK on\nFedora 38+ and Ubuntu 24.04.\n\nAdd another sparse-specific header to workaround the problem. We do\nneed some of the functions and types defined in these headers, so we\ncan't really stab out the whole header. Carving out the main offenders\ninstead by defining the inclusion guards.\n\nThis is fragile and depends on internals of immintrin and underlying\nheaders, but I'm not sure what the better way to solve the issue\nwould be. This approach should be more or less portable between\ncompilers, because it only defines a few specific variables. We may\nhave to add more as GCC headers change over time.\n\nThis fixes the build with a following config on F38 and Ubuntu 24.04:\n\n ./configure --enable-sparse --with-dpdk=yes --enable-Werror\n\nAcked-by: Ales Musil \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"sparse: Add immintrin.h header."}},{"before":"a6c3b5202c2fc9c075a929b052e433fef06b42c0","after":"9a5c24d70fb6b004034c748146de81bd8edc47f9","ref":"refs/heads/branch-3.3","pushedAt":"2024-04-30T19:00:49.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"sparse: Add immintrin.h header.\n\nSparse doesn't understand _Float16 and some other types used by\nimmintrin.h from GCC 13. This breaks sparse builds with DPDK on\nFedora 38+ and Ubuntu 24.04.\n\nAdd another sparse-specific header to workaround the problem. We do\nneed some of the functions and types defined in these headers, so we\ncan't really stab out the whole header. Carving out the main offenders\ninstead by defining the inclusion guards.\n\nThis is fragile and depends on internals of immintrin and underlying\nheaders, but I'm not sure what the better way to solve the issue\nwould be. This approach should be more or less portable between\ncompilers, because it only defines a few specific variables. We may\nhave to add more as GCC headers change over time.\n\nThis fixes the build with a following config on F38 and Ubuntu 24.04:\n\n ./configure --enable-sparse --with-dpdk=yes --enable-Werror\n\nAcked-by: Ales Musil \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"sparse: Add immintrin.h header."}},{"before":"8ce5c95f089c2006709d8f6039d597d75aa73a2f","after":"bd8e9f48f180800292c10e12f26824833f18506a","ref":"refs/heads/main","pushedAt":"2024-04-30T19:00:38.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"sparse: Add immintrin.h header.\n\nSparse doesn't understand _Float16 and some other types used by\nimmintrin.h from GCC 13. This breaks sparse builds with DPDK on\nFedora 38+ and Ubuntu 24.04.\n\nAdd another sparse-specific header to workaround the problem. We do\nneed some of the functions and types defined in these headers, so we\ncan't really stab out the whole header. Carving out the main offenders\ninstead by defining the inclusion guards.\n\nThis is fragile and depends on internals of immintrin and underlying\nheaders, but I'm not sure what the better way to solve the issue\nwould be. This approach should be more or less portable between\ncompilers, because it only defines a few specific variables. We may\nhave to add more as GCC headers change over time.\n\nThis fixes the build with a following config on F38 and Ubuntu 24.04:\n\n ./configure --enable-sparse --with-dpdk=yes --enable-Werror\n\nAcked-by: Ales Musil \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"sparse: Add immintrin.h header."}},{"before":"fb46f5d29ab86f4e95997ca47098a455ce91ed67","after":"8ce5c95f089c2006709d8f6039d597d75aa73a2f","ref":"refs/heads/main","pushedAt":"2024-04-26T09:11:51.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"horms","name":"Simon Horman","path":"/horms","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/65151?s=80&v=4"},"commit":{"message":"dpctl: Fix segfault on ct-{set,del}-limits.\n\nWhen no parameters other than the datapath are specified a segfault\noccurs.\n\nFix it by checking the argument access is inside the bounds.\n\nSigned-off-by: Paolo Valerio \nAcked-by: Kevin Traynor \nSigned-off-by: Simon Horman ","shortMessageHtmlLink":"dpctl: Fix segfault on ct-{set,del}-limits."}},{"before":"bcacd805fefa776d4943190d5b1c19238b3dcf75","after":"fb46f5d29ab86f4e95997ca47098a455ce91ed67","ref":"refs/heads/main","pushedAt":"2024-04-26T09:09:04.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"horms","name":"Simon Horman","path":"/horms","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/65151?s=80&v=4"},"commit":{"message":"netdev-dpdk: Improve error print to the user for flow control error.\n\nWhen failing to get flow control parameters use VLOG_WARN_BUF()\nto expose the error string in ovs-vsctl show.\n\nSigned-off-by: Roi Dayan \nSuggested-by: Simon Horman \nAcked-by: Eli Britstein \nSigned-off-by: Simon Horman ","shortMessageHtmlLink":"netdev-dpdk: Improve error print to the user for flow control error."}},{"before":"07fc1d1ae4f8e4143cc10fd045df3277c83e7c2c","after":"92626bbf315015770660972e521debd740387885","ref":"refs/heads/dpdk-latest","pushedAt":"2024-04-25T11:57:43.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"kevintraynor","name":null,"path":"/kevintraynor","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6536388?s=80&v=4"},"commit":{"message":"ci: Check compilation with DPDK experimental API.\n\nAdd jobs to check compilation with DPDK experimental API enabled.\nThis will help us catch issues for the day we need one of them.\n\nNote: this should not be merged to master, intended for dpdk-latest\nbranch only.\n\nSigned-off-by: David Marchand \nSigned-off-by: Ian Stokes \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ci: Check compilation with DPDK experimental API."}},{"before":"091961542a88c606c21d3226e3ca91dc48bef6ca","after":"b2e19110394e65125079b2c614454d4b4bd1d8e1","ref":"refs/heads/branch-2.17","pushedAt":"2024-04-24T13:33:25.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"horms","name":"Simon Horman","path":"/horms","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/65151?s=80&v=4"},"commit":{"message":"netdev-dpdk: Fix possible memory leak configuring VF MAC address.\n\nVLOG_WARN_BUF() is allocating memory for the error string and should\ne used if the configuration cannot continue and error is being returned\nso the caller has indication of releasing the pointer.\nChange to VLOG_WARN() to keep the logic that error is not being\nreturned.\n\nFixes: f4336f504b17 (\"netdev-dpdk: Add option to configure VF MAC address.\")\nSigned-off-by: Roi Dayan \nAcked-by: Gaetan Rivet \nAcked-by: Eli Britstein \nSigned-off-by: Simon Horman ","shortMessageHtmlLink":"netdev-dpdk: Fix possible memory leak configuring VF MAC address."}},{"before":"f42b58a37d72ed8371a602ee16669b237da09a08","after":"e5f237e9a95d6dc97eca8b819f17b8b3b909a9f7","ref":"refs/heads/branch-3.0","pushedAt":"2024-04-24T13:32:51.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"horms","name":"Simon Horman","path":"/horms","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/65151?s=80&v=4"},"commit":{"message":"netdev-dpdk: Fix possible memory leak configuring VF MAC address.\n\nVLOG_WARN_BUF() is allocating memory for the error string and should\ne used if the configuration cannot continue and error is being returned\nso the caller has indication of releasing the pointer.\nChange to VLOG_WARN() to keep the logic that error is not being\nreturned.\n\nFixes: f4336f504b17 (\"netdev-dpdk: Add option to configure VF MAC address.\")\nSigned-off-by: Roi Dayan \nAcked-by: Gaetan Rivet \nAcked-by: Eli Britstein \nSigned-off-by: Simon Horman ","shortMessageHtmlLink":"netdev-dpdk: Fix possible memory leak configuring VF MAC address."}},{"before":"c71c87d5c928fa538439519cca8a609a7e52ad3d","after":"6ce0130f12399b14fd27260cc186bc0bf0187928","ref":"refs/heads/branch-3.1","pushedAt":"2024-04-24T13:32:18.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"horms","name":"Simon Horman","path":"/horms","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/65151?s=80&v=4"},"commit":{"message":"netdev-dpdk: Fix possible memory leak configuring VF MAC address.\n\nVLOG_WARN_BUF() is allocating memory for the error string and should\ne used if the configuration cannot continue and error is being returned\nso the caller has indication of releasing the pointer.\nChange to VLOG_WARN() to keep the logic that error is not being\nreturned.\n\nFixes: f4336f504b17 (\"netdev-dpdk: Add option to configure VF MAC address.\")\nSigned-off-by: Roi Dayan \nAcked-by: Gaetan Rivet \nAcked-by: Eli Britstein \nSigned-off-by: Simon Horman ","shortMessageHtmlLink":"netdev-dpdk: Fix possible memory leak configuring VF MAC address."}},{"before":"f0ccdfc0b3996c1144ac5d98b8580d8f38dc4293","after":"4e4463ca55392ca841f95bf51c06b1e03d6c5492","ref":"refs/heads/branch-3.2","pushedAt":"2024-04-24T13:31:53.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"horms","name":"Simon Horman","path":"/horms","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/65151?s=80&v=4"},"commit":{"message":"netdev-dpdk: Fix possible memory leak configuring VF MAC address.\n\nVLOG_WARN_BUF() is allocating memory for the error string and should\ne used if the configuration cannot continue and error is being returned\nso the caller has indication of releasing the pointer.\nChange to VLOG_WARN() to keep the logic that error is not being\nreturned.\n\nFixes: f4336f504b17 (\"netdev-dpdk: Add option to configure VF MAC address.\")\nSigned-off-by: Roi Dayan \nAcked-by: Gaetan Rivet \nAcked-by: Eli Britstein \nSigned-off-by: Simon Horman ","shortMessageHtmlLink":"netdev-dpdk: Fix possible memory leak configuring VF MAC address."}},{"before":"42e685916e5963774c3c815c5b9c6680c0a8568e","after":"a6c3b5202c2fc9c075a929b052e433fef06b42c0","ref":"refs/heads/branch-3.3","pushedAt":"2024-04-24T13:31:16.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"horms","name":"Simon Horman","path":"/horms","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/65151?s=80&v=4"},"commit":{"message":"netdev-dpdk: Fix possible memory leak configuring VF MAC address.\n\nVLOG_WARN_BUF() is allocating memory for the error string and should\ne used if the configuration cannot continue and error is being returned\nso the caller has indication of releasing the pointer.\nChange to VLOG_WARN() to keep the logic that error is not being\nreturned.\n\nFixes: f4336f504b17 (\"netdev-dpdk: Add option to configure VF MAC address.\")\nSigned-off-by: Roi Dayan \nAcked-by: Gaetan Rivet \nAcked-by: Eli Britstein \nSigned-off-by: Simon Horman ","shortMessageHtmlLink":"netdev-dpdk: Fix possible memory leak configuring VF MAC address."}},{"before":"1fe98e41f0b62b71f4c190daf2d56d864f85ab47","after":"091961542a88c606c21d3226e3ca91dc48bef6ca","ref":"refs/heads/branch-2.17","pushedAt":"2024-04-23T21:34:40.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"ovsdb: raft: Fix probe intervals after install snapshot request.\n\nIf the new snapshot received with INSTALL_SNAPSHOT request contains\na different election timer value, the timer is updated, but the\nprobe intervals for RAFT connections are not.\n\nFix that by updating probe intervals whenever we get election timer\nfrom the log.\n\nFixes: 14b2b0aad7ae (\"raft: Reintroduce jsonrpc inactivity probes.\")\nAcked-by: Mike Pattrick \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ovsdb: raft: Fix probe intervals after install snapshot request."}},{"before":"d4d309bbd59e6af466c6164188fea18183e4ddc5","after":"f42b58a37d72ed8371a602ee16669b237da09a08","ref":"refs/heads/branch-3.0","pushedAt":"2024-04-23T21:34:33.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"ovsdb: raft: Fix probe intervals after install snapshot request.\n\nIf the new snapshot received with INSTALL_SNAPSHOT request contains\na different election timer value, the timer is updated, but the\nprobe intervals for RAFT connections are not.\n\nFix that by updating probe intervals whenever we get election timer\nfrom the log.\n\nFixes: 14b2b0aad7ae (\"raft: Reintroduce jsonrpc inactivity probes.\")\nAcked-by: Mike Pattrick \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ovsdb: raft: Fix probe intervals after install snapshot request."}},{"before":"6b32502aa694af877979432ab5deba81fbb5acbc","after":"c71c87d5c928fa538439519cca8a609a7e52ad3d","ref":"refs/heads/branch-3.1","pushedAt":"2024-04-23T21:34:30.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"ovsdb: raft: Fix probe intervals after install snapshot request.\n\nIf the new snapshot received with INSTALL_SNAPSHOT request contains\na different election timer value, the timer is updated, but the\nprobe intervals for RAFT connections are not.\n\nFix that by updating probe intervals whenever we get election timer\nfrom the log.\n\nFixes: 14b2b0aad7ae (\"raft: Reintroduce jsonrpc inactivity probes.\")\nAcked-by: Mike Pattrick \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ovsdb: raft: Fix probe intervals after install snapshot request."}},{"before":"c88a35fc29f0c0eb6189853bfc738c2100d4860f","after":"f0ccdfc0b3996c1144ac5d98b8580d8f38dc4293","ref":"refs/heads/branch-3.2","pushedAt":"2024-04-23T21:34:25.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"ovsdb: raft: Fix probe intervals after install snapshot request.\n\nIf the new snapshot received with INSTALL_SNAPSHOT request contains\na different election timer value, the timer is updated, but the\nprobe intervals for RAFT connections are not.\n\nFix that by updating probe intervals whenever we get election timer\nfrom the log.\n\nFixes: 14b2b0aad7ae (\"raft: Reintroduce jsonrpc inactivity probes.\")\nAcked-by: Mike Pattrick \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ovsdb: raft: Fix probe intervals after install snapshot request."}},{"before":"f19448b8618967a108ec6f34713dd811ce1d1334","after":"42e685916e5963774c3c815c5b9c6680c0a8568e","ref":"refs/heads/branch-3.3","pushedAt":"2024-04-23T21:34:17.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"ovsdb: raft: Fix probe intervals after install snapshot request.\n\nIf the new snapshot received with INSTALL_SNAPSHOT request contains\na different election timer value, the timer is updated, but the\nprobe intervals for RAFT connections are not.\n\nFix that by updating probe intervals whenever we get election timer\nfrom the log.\n\nFixes: 14b2b0aad7ae (\"raft: Reintroduce jsonrpc inactivity probes.\")\nAcked-by: Mike Pattrick \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ovsdb: raft: Fix probe intervals after install snapshot request."}},{"before":"fbade819d2de1685c49e1deaf62d45049a5c2a27","after":"bcacd805fefa776d4943190d5b1c19238b3dcf75","ref":"refs/heads/main","pushedAt":"2024-04-23T21:33:57.000Z","pushType":"push","commitsCount":4,"pusher":{"login":"igsilya","name":"Ilya Maximets","path":"/igsilya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/11501231?s=80&v=4"},"commit":{"message":"ovsdb: raft: Fix probe intervals after install snapshot request.\n\nIf the new snapshot received with INSTALL_SNAPSHOT request contains\na different election timer value, the timer is updated, but the\nprobe intervals for RAFT connections are not.\n\nFix that by updating probe intervals whenever we get election timer\nfrom the log.\n\nFixes: 14b2b0aad7ae (\"raft: Reintroduce jsonrpc inactivity probes.\")\nAcked-by: Mike Pattrick \nSigned-off-by: Ilya Maximets ","shortMessageHtmlLink":"ovsdb: raft: Fix probe intervals after install snapshot request."}},{"before":"2b7efee031c3a2205ad2ee999275893edd083c1c","after":"fbade819d2de1685c49e1deaf62d45049a5c2a27","ref":"refs/heads/main","pushedAt":"2024-04-23T10:57:50.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"horms","name":"Simon Horman","path":"/horms","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/65151?s=80&v=4"},"commit":{"message":"ovsdb-idl: Add python keyword to persistent UUID test.\n\nThe Python persistent UUID tests should have the keyword \"python\"\nadded so that TESTSUITEFLAGS=\"-k python\" will not miss testing\nthem.\n\nFixes: 55b9507e6824 (\"ovsdb-idl: Add the support to specify the uuid for row insert.\")\nSigned-off-by: Terry Wilson \nTested-by: Simon Horman \nSigned-off-by: Simon Horman ","shortMessageHtmlLink":"ovsdb-idl: Add python keyword to persistent UUID test."}},{"before":"4f29804f249bf5994603b73c3250198ce0adb8d7","after":"2b7efee031c3a2205ad2ee999275893edd083c1c","ref":"refs/heads/main","pushedAt":"2024-04-23T10:53:32.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"horms","name":"Simon Horman","path":"/horms","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/65151?s=80&v=4"},"commit":{"message":"socket: Increase listen backlog to 64 everywhere.\n\nBefore the patch, the size of the backlog depended on the type of socket\n(UNIX vs INET) as well as on the language (C vs Python), specifically:\n\n- python used backlog size = 10 for all sockets;\n- C used 64 for UNIX sockets but 10 for INET sockets.\n\nThis consolidates the values across the board. It effectively bumps the\nnumber of simultaneous connections to python unixctl servers to 64. Also\nfor INET C servers too.\n\nThe rationale to do it, on top of consistency, is as follows:\n\n- fmt_pkt in ovn testsuite is limited by python server listen backlog,\n and as was found out when adopting the tool, it is sometimes useful to\n run lots of parallel calls to fmt_pkt unixctl server in some tests.\n (See [1] for example.)\n\n- there is a recent report [2] on discuss@ ML where the reporter noticed\n significant listen queue overflows in some scenarios (large openstack\n deployments; happens during leader transition when hundreds of neutron\n nodes - with dozens of neutron api workers each - simultaneously\n reconnect to the same northbound leader.) Note: While there is no\n clear indication that this backlog size bump would resolve the\n reported issues, it would probably help somewhat.\n\n[1] https://github.com/ovn-org/ovn/commit/0baca3e519756cbe98a32526ccc637bb73468743\n[2] https://mail.openvswitch.org/pipermail/ovs-discuss/2024-April/053049.html\n\nSigned-off-by: Ihar Hrachyshka \nAcked-by: Eelco Chaudron \nSigned-off-by: Simon Horman ","shortMessageHtmlLink":"socket: Increase listen backlog to 64 everywhere."}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEQNIf7AA","startCursor":null,"endCursor":null}},"title":"Activity ยท openvswitch/ovs"}