Attention is currently required from: osmith.
laforge has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-ggsn/+/36068?usp=email )
Change subject: gtp/gtp_internal.h: new file
......................................................................
Patch Set 1:
(1 comment)
File gtp/gtp_internal.h:
https://gerrit.osmocom.org/c/osmo-ggsn/+/36068/comment/e1c5d1ef_64a4f4b3
PS1, Line 3: int gtp_echo_resp(struct gsn_t *gsn, int version,
: struct sockaddr_in *peer, int fd,
: void *pack, unsigned len);
: int gtp_echo_ind(struct gsn_t *gsn, int version,
: struct sockaddr_in *peer, int fd,
: void *pack, unsigned len);
: int gtp_echo_conf(struct gsn_t *gsn, int version,
: struct sockaddr_in *peer, void *pack, unsigned len);
:
: int gtp_unsup_req(struct gsn_t *gsn, int version,
: struct sockaddr_in *peer,
: int fd, void *pack, unsigned len);
: int gtp_unsup_ind(struct gsn_t *gsn, struct sockaddr_in *peer,
: void *pack, unsigned len);
:
: int gtp_create_pdp_resp(struct gsn_t *gsn, int version,
: struct pdp_t *pdp, uint8_t cause);
:
: int gtp_create_pdp_ind(struct gsn_t *gsn, int version,
: struct sockaddr_in *peer, int fd,
: void *pack, unsigned len);
:
: int gtp_create_pdp_conf(struct gsn_t *gsn, int version,
: struct sockaddr_in *peer,
: void *pack, unsigned len);
:
: int gtp_update_pdp_req(struct gsn_t *gsn, int version, void *cbp,
: struct in_addr *inetaddr, struct pdp_t *pdp);
:
: int gtp_delete_pdp_req(struct gsn_t *gsn, int version, void *cbp,
: struct pdp_t *pdp);
:
: int gtp_delete_pdp_resp(struct gsn_t *gsn, int version,
: struct sockaddr_in *peer, int fd,
: void *pack, unsigned len,
: struct pdp_t *pdp, struct pdp_t *linked_pdp,
: uint8_t cause, int teardown);
:
: int gtp_delete_pdp_ind(struct gsn_t *gsn, int version,
: struct sockaddr_in *peer, int fd,
: void *pack, unsigned len);
:
: int gtp_delete_pdp_conf(struct gsn_t *gsn, int version,
: struct sockaddr_in *peer,
: void *pack, unsigned len);
why do those gtp_* symbols need to be non-static and declared in any header file at all? It looks to me like they are defined in gtp.c and the only user is in gtp.c itself?a
--
To view, visit https://gerrit.osmocom.org/c/osmo-ggsn/+/36068?usp=email
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings
Gerrit-Project: osmo-ggsn
Gerrit-Branch: master
Gerrit-Change-Id: I72c40cbdec33449ca8104fb3cad8df1a9e07dfd7
Gerrit-Change-Number: 36068
Gerrit-PatchSet: 1
Gerrit-Owner: osmith <osmith(a)sysmocom.de>
Gerrit-Reviewer: Jenkins Builder
Gerrit-CC: laforge <laforge(a)osmocom.org>
Gerrit-Attention: osmith <osmith(a)sysmocom.de>
Gerrit-Comment-Date: Wed, 20 Mar 2024 12:32:31 +0000
Gerrit-HasComments: Yes
Gerrit-Has-Labels: No
Gerrit-MessageType: comment
Attention is currently required from: dexter, laforge.
pespin has posted comments on this change. ( https://gerrit.osmocom.org/c/libosmocore/+/36324?usp=email )
Change subject: msgb: Also copy msgb {dst, lchan, cb} during msgb_copy_*
......................................................................
Patch Set 2: Code-Review+1
(1 comment)
File src/core/msgb.c:
https://gerrit.osmocom.org/c/libosmocore/+/36324/comment/7bd7ce73_efa9a6dd
PS1, Line 320: * and adjusts the pointers (incl l1h-l4h) accordingly. The cb part is not copied.
> I did a rather exhaustive study and could not find any osmocoom code that would rely on the existing […]
I think the cb was not copied as a performance reason, since lots of code doesn't use cb at all.
I personally think it's fine keeping the old behavior. You may want to add a copy_full() API which also copies the cb and other stuff.
--
To view, visit https://gerrit.osmocom.org/c/libosmocore/+/36324?usp=email
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings
Gerrit-Project: libosmocore
Gerrit-Branch: master
Gerrit-Change-Id: I68328adb952ca8833ba047cb3b49ccc6f8a1f1b5
Gerrit-Change-Number: 36324
Gerrit-PatchSet: 2
Gerrit-Owner: laforge <laforge(a)osmocom.org>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: dexter <pmaier(a)sysmocom.de>
Gerrit-Reviewer: pespin <pespin(a)sysmocom.de>
Gerrit-Attention: laforge <laforge(a)osmocom.org>
Gerrit-Attention: dexter <pmaier(a)sysmocom.de>
Gerrit-Comment-Date: Wed, 20 Mar 2024 12:31:22 +0000
Gerrit-HasComments: Yes
Gerrit-Has-Labels: Yes
Comment-In-Reply-To: laforge <laforge(a)osmocom.org>
Comment-In-Reply-To: dexter <pmaier(a)sysmocom.de>
Gerrit-MessageType: comment
Attention is currently required from: dexter, laforge, neels, pespin.
Hello Jenkins Builder, dexter, neels, pespin,
I'd like you to reexamine a change. Please visit
https://gerrit.osmocom.org/c/osmo-mgw/+/36363?usp=email
to look at the new patch set (#3).
The following approvals got outdated and were removed:
Verified-1 by Jenkins Builder
Change subject: Convert RTP/RTCP/OSMUX I/O from osmo_fd to osmo_io
......................................................................
Convert RTP/RTCP/OSMUX I/O from osmo_fd to osmo_io
Converting from osmo_fd to osmo_io allows us to switch to the new
io_uring backend and benefit from related performance benefits.
In a benchmark running 200 concurrent bi-directional voice calls with
GSM-EFR codec, I am observing:
* the code before this patch uses 40..42% of a single core on a
Ryzen 5950X at 200 calls (=> 200 endpoints with each two connections)
* no increase in CPU utilization before/after this patch, i.e. the
osmo_io overhead for the osmo_fd backend is insignificant compared
to the direct osmo_fd mode before
* an almost exactly 50% reduction of CPU utilization when running the
same osmo-mgw build with LIBOSMO_IO_BACKEND=IO_URING - top shows
19..21% for the same workload instead of 40..42% with the OSMO_FD
default backend.
* An increase of about 4 Megabytes in both RSS and VIRT size when
enabling the OSMO_IO backend. This is likely the memory-mapped rings.
No memory leakage is observed when using either of the backends.
Change-Id: I8471960d5d8088a70cf105f2f40dfa5d5458169a
---
M include/osmocom/mgcp/mgcp.h
M include/osmocom/mgcp/mgcp_network.h
M src/libosmo-mgcp/mgcp_conn.c
M src/libosmo-mgcp/mgcp_iuup.c
M src/libosmo-mgcp/mgcp_network.c
M src/libosmo-mgcp/mgcp_osmux.c
6 files changed, 225 insertions(+), 165 deletions(-)
git pull ssh://gerrit.osmocom.org:29418/osmo-mgw refs/changes/63/36363/3
--
To view, visit https://gerrit.osmocom.org/c/osmo-mgw/+/36363?usp=email
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings
Gerrit-Project: osmo-mgw
Gerrit-Branch: master
Gerrit-Change-Id: I8471960d5d8088a70cf105f2f40dfa5d5458169a
Gerrit-Change-Number: 36363
Gerrit-PatchSet: 3
Gerrit-Owner: laforge <laforge(a)osmocom.org>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: dexter <pmaier(a)sysmocom.de>
Gerrit-Reviewer: neels <nhofmeyr(a)sysmocom.de>
Gerrit-Reviewer: pespin <pespin(a)sysmocom.de>
Gerrit-Attention: neels <nhofmeyr(a)sysmocom.de>
Gerrit-Attention: laforge <laforge(a)osmocom.org>
Gerrit-Attention: pespin <pespin(a)sysmocom.de>
Gerrit-Attention: dexter <pmaier(a)sysmocom.de>
Gerrit-MessageType: newpatchset