net/mlx5: insertion rate optimization
Get series with b4
b4 am [email protected]Versions newest first · linked by author and title
- v6 Latest26 patches 1 replyIn review
- v526 patches 2 repliesSuperseded
- v426 patches 0 repliesSuperseded
Earlier versions · 2
| Version | Posted | Patches | Replies | Status |
|---|---|---|---|---|
| v3 | 22 patches | 0 replies | Superseded | |
| v2 | 22 patches | 0 replies | Superseded |
v4 from the cover letter
- split the list utility to commom patch
- fix one ipool crash issue in mlx5_ipool_get_cache() function.
- reorganize the common and mlx5 patches.
- split the doc update to an individual patch.
- update and rebase to the latest version.
Patches in v6 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/26 | net/mlx5: allow limiting the index pool maximum indexdrivers/ drivers/ | +13 | 0 replies | |
| 2/26 | net/mlx5: add indexed pool local cachedrivers/ drivers/ | +372 | 0 replies | |
| 3/26 | net/mlx5: add index pool foreach definedrivers/ drivers/ | +104 | 0 replies | |
| 4/26 | net/mlx5: support index pool non-lcore operationsdrivers/ drivers/ | +56 | 0 replies | |
| 5/26 | net/mlx5: replace flow list with index pooldrivers/ drivers/ drivers/ +5 more | +127 | 0 replies | |
| 6/26 | net/mlx5: optimize modify header action memorydrivers/ drivers/ drivers/ +1 more | +12 | 0 replies | |
| 7/26 | net/mlx5: remove cache term from the list utilitydrivers/ drivers/ drivers/ +7 more | +456 | 0 replies | |
| 8/26 | net/mlx5: add per lcore cache to the list utilitydrivers/ drivers/ drivers/ +6 more | +451 | 0 replies | |
| 9/26 | net/mlx5: minimize list critical sectionsdrivers/ drivers/ | +48 | 0 replies | |
| 10/26 | net/mlx5: manage list cache entries releasedrivers/ drivers/ | +54 | 0 replies | |
| 11/26 | net/mlx5: relax the list utility atomic operationsdrivers/ | +6 | 0 replies | |
| 12/26 | net/mlx5: allocate list memory by the create APIdrivers/ drivers/ drivers/ +6 more | +139 | 0 replies | |
| 13/26 | common/mlx5: move list utility to commondrivers/ drivers/ drivers/ +3 more | +464 | 0 replies | |
| 14/26 | common/mlx5: add list lcore sharedrivers/ drivers/ drivers/ +2 more | +56 | 0 replies | |
| 15/26 | common/mlx5: call list callbacks with contextdrivers/ drivers/ drivers/ +3 more | +132 | 0 replies | |
| 16/26 | common/mlx5: add per-lcore cache to hash list utilitydrivers/ drivers/ drivers/ +5 more | +482 | 0 replies | |
| 17/26 | common/mlx5: allocate cache list memory individuallydrivers/ drivers/ | +45 | 0 replies | |
| 18/26 | common/mlx5: optimize cache list object memorydrivers/ drivers/ | +201 | 0 replies | |
| 19/26 | common/mlx5: support list non-lcore operationsdrivers/ drivers/ | +71 | 0 replies | |
| 20/26 | net/mlx5: move modify header allocator to ipooldrivers/ drivers/ drivers/ +1 more | +86 | 0 replies | |
| 21/26 | net/mlx5: adjust the hash bucket sizedrivers/ drivers/ drivers/ +1 more | +7 | 0 replies | |
| 22/26 | net/mlx5: enable index pool per-core cachedrivers/ drivers/ drivers/ | +6 | 0 replies | |
| 23/26 | net/mlx5: optimize hash list table allocate on demanddrivers/ drivers/ drivers/ +1 more | +82 | 0 replies | |
| 24/26 | net/mlx5: change memory release configurationdrivers/ | +2 | 0 replies | |
| 25/26 | net/mlx5: optimize Rx queue matchdrivers/ | +7 | 0 replies | |
| 26/26 | doc: add mlx5 multiple-thread flow insertion optimizationdoc/ doc/ | +6 | 0 replies |
Reviews trailers per patch
All 26 patches have a Reviewed-by or Acked-by.
| Patch | Matan Azrad | Suanming Mou |
|---|---|---|
| 1/26 | A-b | |
| 2/26 | A-b | |
| 3/26 | A-b | |
| 4/26 | A-b | |
| 5/26 | A-b | |
| 6/26 | A-b | |
| 7/26 | A-b | |
| 8/26 | A-b | |
| 9/26 | A-b | |
| 10/26 | A-b | |
| 11/26 | A-b | |
| 12/26 | A-b | |
| 13/26 | A-b | |
| 14/26 | A-b | |
| 15/26 | A-b | |
| 16/26 | A-b | |
| 17/26 | A-b | |
| 18/26 | A-b | |
| 19/26 | A-b | |
| 20/26 | A-b | |
| 21/26 | A-b | |
| 22/26 | A-b | |
| 23/26 | A-b | |
| 24/26 | A-b | |
| 25/26 | A-b | |
| 26/26 | A-b |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [dpdk-dev] [PATCH v6 00/26] net/mlx5: insertion rate optimization
This patch series optimize the flow insertion rate with adding local cache to index pool and list.
For object which wants efficient index allocate and free, local cache will be very helpful.
For index pool, two level cache is added, one as local and another as global. The global cache is able to save all the allocated index. That means all the allocated index will not be freed. Once the local cache is full, the extra index will be flushed to the global cache. Once local cache is empty, first try to fetch more index from global, if global is still empty, allocate new trunk and more index.
For list, sub local core list is introduced. The allocated objects will be added and released only from local list without any locks. Only the objects need to be shared will be synced from global list.
---
v6: fix compilation issue in "common/mlx5: add list lcore share" patch.
v5: code rebase to the latest version.
v4:
- split the list utility to commom patch
- fix one ipool crash issue in mlx5_ipool_get_cache() function.
- reorganize the common and mlx5 patches.
- split the doc update to an individual patch.
- update and rebase to the latest version.
v3: fix Windows compilation issue
v2: add the list per-lcore cache optimization
---
Matan Azrad (9):
net/mlx5: optimize modify header action memory net/mlx5: remove cache term from the list utility net/mlx5: add per lcore cache to the list utility net/mlx5: minimize list critical sections net/mlx5: manage list cache entries release net/mlx5: relax the list utility atomic operations net/mlx5: allocate list memory by the create API common/mlx5: add per-lcore cache to hash list utility net/mlx5: move modify header allocator to ipool
Suanming Mou (17):
net/mlx5: allow limiting the index pool maximum index net/mlx5: add indexed pool local cache
net/mlx5: add index pool foreach define net/mlx5: support index pool non-lcore operations net/mlx5: replace flow list with index pool common/mlx5: move list utility to common common/mlx5: add list lcore share
common/mlx5: call list callbacks with context common/mlx5: allocate cache list memory individually common/mlx5: optimize cache list object memory common/mlx5: support list non-lcore operations net/mlx5: adjust the hash bucket size
net/mlx5: enable index pool per-core cache net/mlx5: optimize hash list table allocate on demand net/mlx5: change memory release configuration net/mlx5: optimize Rx queue match
doc: add mlx5 multiple-thread flow insertion optimization
doc/guides/nics/mlx5.rst | 5 + doc/guides/rel_notes/release_21_08.rst | 1 + drivers/common/mlx5/linux/mlx5_glue.h | 1 + drivers/common/mlx5/mlx5_common.h | 2 + drivers/common/mlx5/mlx5_common_utils.c | 569 ++++++++--- drivers/common/mlx5/mlx5_common_utils.h | 294 ++++-- drivers/common/mlx5/version.map | 7 + drivers/net/mlx5/linux/mlx5_flow_os.h | 3 +- drivers/net/mlx5/linux/mlx5_os.c | 209 ++-- drivers/net/mlx5/mlx5.c | 34 +- drivers/net/mlx5/mlx5.h | 46 +- drivers/net/mlx5/mlx5_defs.h | 12 +- drivers/net/mlx5/mlx5_flow.c | 308 +++--- drivers/net/mlx5/mlx5_flow.h | 209 ++-- drivers/net/mlx5/mlx5_flow_dv.c | 1206 +++++++++++++++-------- drivers/net/mlx5/mlx5_rx.h | 14 +- drivers/net/mlx5/mlx5_rxq.c | 136 ++- drivers/net/mlx5/mlx5_trigger.c | 8 +- drivers/net/mlx5/mlx5_utils.c | 627 ++++++++---- drivers/net/mlx5/mlx5_utils.h | 255 ++--- drivers/net/mlx5/windows/mlx5_os.c | 11 +- 21 files changed, 2543 insertions(+), 1414 deletions(-)
Signature · 1 line
2.25.1