Age | Commit message (Collapse) | Author | Files | Lines |
|
Currently, buffers-per-numa is used for all NUMA domains, and there's
no way to allocate different amounts for each domain.
Therefore, this adds a per-NUMA-domain buffer config section, which
currently has a single option to set the number of buffers.
If it's not specified or set to zero, the code falls back to
buffers-per-numa as before.
Type: improvement
Change-Id: If35d7a9eff6f8d1d78063ea7873dbf50780d0ec3
Signed-off-by: Lukas Stockner <lstockner@genesiscloud.com>
|
|
Type: improvement
Change-Id: I88c4c45bed0bdd8686e17e4f77a7d32a08c995aa
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Type: feature
Change-Id: Icd9de05f2cbac0e5a6dfb1f1414f21dc4b893104
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Type: improvement
Change-Id: Ie86a5edf2ada21355543e9a0382052b16ff86927
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
- pass buffer pool name trough va
- make buffers naturaly aligned
- fix calculation of total number of buffers
Type: improvement
Change-Id: I6aebf249ebd67823b4632ac08905bfa3aa7d1ee5
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
This plugin allow to keep track of buffer usage in VPP graph nodes. The
main use is to detect buffer leakages.
Type: feature
Change-Id: Iadcf4ab98207fab6e2fa375060879bc2a25b711e
Signed-off-by: Benoît Ganne <bganne@cisco.com>
|
|
trajectory trace has been broken for a while because we used to save the
buffer trajectory in a vector pointed to in opaque2. This does not work
well when opaque2 is copied (eg. because of a clone) as 2 buffers end up
sharing the same vector.
This dedicates a full cacheline in the buffer metadata instead when
trajectory is compiled in. No dynamic allocation, no sharing, no tears.
Type: refactor
Change-Id: I6a028ca1b48d38f393a36979e5e452c2dd48ad3f
Signed-off-by: Benoît Ganne <bganne@cisco.com>
|
|
Type: feature
Add a `buffers {page-size}` parameter to specify page size
for buffers. This also fixes an issue with the parsing in
unformat_log2_page_size.
Change-Id: I7d7b1fa0bb7febaa7509cf2c625882f07eeafaad
Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
|
|
Fixes issues on systems with 128-byte cacheline.
Type: fix
Change-Id: I33689ffa5ef0a879b4bf93c25d48618dd43daf58
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
GCC-10 increase overflows-related warnings but gets confused by 0-length
arrays. Use C99 flexible length array instead.
Type: fix
Change-Id: Ie62cfa8faaa408479a598785fd3f06ffd0233c7a
Signed-off-by: Benoît Ganne <bganne@cisco.com>
|
|
Type: feature
For some deployments(mobile traffic) we need few extra bytes of space
Signed-off-by: Srikanth Akula <srakula@cisco.com>
Change-Id: I3367561dc23dbb8b266dea583e23b9430fd7dcab
|
|
Several nodes include buffers in their traces, but only the 1st. When
formatting the trace we must not try to iterate through all chained
buffers.
Default to display only the 1st buffer.
Type: fix
Change-Id: Ib3c668bbf4ab70ae68eba2ac402c7b7329825b70
Signed-off-by: Benoît Ganne <bganne@cisco.com>
|
|
Type: refactor
Change-Id: I72221b97d7e0bf5c93e20bbda4473ca67bfcdeb4
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Type: feature
Change-Id: Ia3d9a47679202c2a47cd3746b50e86c6b8627ef6
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
A punt/exception path that provides:
1) clients that use the infra
2) clients can create punt reasons
3) clients can register to recieve packets that are punted
for a given reason to be sent to the desired node.
4) nodes which punt packets fill in the {reason,protocol} of the
buffere (in the meta-data) and send to the new node "punt-dispatch"
5) punt-dispatch sends packets to the registered nodes or drops
Change-Id: Ia4f144337f1387cbe585b4f375d0842aefffcde5
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: I550ef893e41d86310d4e5af16d6100b8e557b68a
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Example:
buffers {
default data-size 1536
}
Change-Id: I5b4436850ca18025c9fdcfc7ed648c2c2732d660
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
This patch introduces following changes:
- deprecated free lists which are not used and not compatible
with external buffer managers (i.e. DPDK)
- introduces native support for per-numa buffer pools
- significantly improves performance of buffer alloc and free
Change-Id: I4a8e723ae47056717afd6cac0efe87cb731b5be7
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I9df96264e30806ac3daf7121f314f34f06232413
Signed-off-by: Damjan Marion <damjan.marion@gmail.com>
|
|
Change-Id: I1c12e2941cae198ededbb65eb5be51a4eabe2c1b
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I0ba5175be077c40556f2a3ce629c5bbcd71e0a81
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I5fe01e918b534d1ac1d47f0d7610536f45c3158c
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I3bb1d9f83dd08f4b93acd4a281bfec0674e39c2e
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I6048c6a51efa826ac333f7d15919cb87dd766d74
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Storing buffer in local template seems to be better option....
Change-Id: I1a2fdd68cb956f99a5b36d2cd810fc623e089bcf
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: If2bbfbc52994f5de0879763e0b7a7864498debb6
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Add a check to make sure that the vlib and vnet buffer flag bit
definitions do not overlap.
The VNET_BUFFER_F_AVAIL1...8 definitions allow out-of-tree codes to:
#define VNET_BUFFER_F_MY_USECASE VNET_BUFFER_F_AVAIL1
and so on. This avoids introducing irrelevant and/or proprietary bit
definitions into vnet/buffer.h, and hopefully minimizes merge pain for
everyone involved.
Change-Id: I5be4f61dceb81b5bfca005f6d609ade074af205b
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
dhcp packets might (when flooded) arrive in chains of cloned buffers
Change-Id: Ifddecd656b6a5d6ba8cd94184f5c021684e35548
Signed-off-by: Eyal Bari <ebari@cisco.com>
|
|
Change-Id: Iba750a41262cc028ad0363fff78cc219e4a33538
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I9b5f7b264f9978e3dd97b2d1eb103b7d10ac3170
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Ic4c46bc733afae8bf0d8146623ed15633928de30
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I0b42ac6b05bc9910904a97924ea4bebc84507d4d
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I1f54b994425c58776e1445c8d9fe142e7a644d3d
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I2794384557c6272fe217269b14a9db09eda19220
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I32f68e2ee8f5d32962acdefb0193583f71d342b3
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
- buffer_main is no longer part of vlib_main_t
- pool of free lists is still part of vlib_main_t
- mheap is not used anymore for buffer allocation
- simple bitmap bassed buffer alloc scheme is introduced
Change-Id: I3e1e6d00e2c8122293ed0a741245eb841315a1ff
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I99cf3e7cc991aa7d32385a155c707a6516516117
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I853386aebfe488ebb10328435b81b6e3403c5dd0
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Currently, every alloc involves callback function call. With this change
callback function is called only if there is no empty buffers on the
free list.
Change-Id: I2238c19ece7ce182c49ba0f2485add52f335f3b6
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: If581feca0d51d0420c971801aecdf9250c671b36
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Icaf7d7ad47284aea7a56e8006b69f45874d64202
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Use a proper u16 * vector to capture node indices, since vpp w/
plugins now exceeds 255 graph nodes
Change-Id: Ic48cad676fa3a6116413ddf08c083dd9660783f1
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
This patch adds supprot support for multiple numa-aware physmem regions.
Change-Id: I5c69a6f4da33c8ee21bdb8604d52fd2886f2327e
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
First attempt to make internal buffer manager thread safe was not
succesfull, so trying again. This time with more testing.
Change-Id: I01b8385a9c26d233934a3339255ea4bd31c865ac
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I45845b952aa42a854e1c2c396b85f905de987020
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
- api-segment prefix not used when unlinking shm files
- unlink root region on exit if no clients referenced
- stale reference to freed segment name
- don't add fake client to /db unless CLIB_DEBUG > 2
- turn off the gmond plugin
- clean up unused vars in vpp/api
Change-Id: I66451fcfd6ee64a12466c2d6c209050e3cdb74b7
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
To save space in the first cacheline following is changed:
- total_length_not_including_first_buffer moved to the 2nd cacheline.
This field is used only when VLIB_BUFFER_TOTAL_LENGTH_VALID and
VLIB_BUFFER_NEXT_PRESENT are both set.
- free_list_index is now stored in 4bits inside flags, which
allows up to 16 free lists. In case we need more we can store index
in the 2nd cachelin
Change-Id: Ic8521350819391af470d31d3fa1013e67ecb7681
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Currently, buffer index is calculated as a offset to the physmem
region shifted by log2_cacheline size.
When DPDK is used we "hack" physmem data with information taken from
dpdk mempool. This makes physmem code not usable with DPDK.
This change makes buffer memory start and size independent of physmem
basically allowing physmem to be used when DPDK plugin is loaded.
Change-Id: Ieb399d398f147583b9baab467152a352d58c9c31
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Ibe55e4399c6b78d83268d7c49ed498cab7bfdb43
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
|
|
Change-Id: I49e5ce0aae6e4ff634024387ceaf7dbc432a0351
Signed-off-by: Dave Barach <dave@barachs.net>
Signed-off-by: Florin Coras <fcoras@cisco.com>
|