pespin has submitted this change. ( https://gerrit.osmocom.org/c/libosmocore/+/39348?usp=email )
Change subject: osmo_io: Improve osmo_iofd_notify_connected() documentation
......................................................................
osmo_io: Improve osmo_iofd_notify_connected() documentation
The mentioned scenario happens when using osmo_stream_cli inside
libosmo-abis BTS code connecting to an osmo-bsc in the loopback
interface.
The client sends SYN and the server sends SYN+ACK and then quickly sends
an IPA ID GET.
Change-Id: Ica20a050b98d117995a5b625b23ab9faa61aabee
(cherry picked from commit 59753c57a5d084549b43e63e3f0a0f6b4d21d766)
---
M src/core/osmo_io.c
1 file changed, 12 insertions(+), 2 deletions(-)
Approvals:
neels: Looks good to me, but someone else must approve
Jenkins Builder: Verified
osmith: Looks good to me, approved
diff --git a/src/core/osmo_io.c b/src/core/osmo_io.c
index 98d8ef3..9c93e25 100644
--- a/src/core/osmo_io.c
+++ b/src/core/osmo_io.c
@@ -994,12 +994,22 @@
/*! Request notification of the user if/when a client socket is connected.
* Calling this function will request osmo_io to notify the user (via
- * write call-back) once a non-blocking outbound connect() of the
- * socket completes.
+ * write call-back with res=0 and msgb=NULL) once a non-blocking outbound
+ * connect() of the socket completes.
*
* This only works for connection oriented sockets in either
* OSMO_IO_FD_MODE_READ_WRITE or OSMO_IO_FD_MODE_RECVMSG_SENDMSG mode.
*
+ * The fact that the write call-back is called with msgb=NULL can be used to
+ * distinguish before this "connected" notification and a socket write failure.
+ *
+ * If the server transmits data quick enough after accepting the connection,
+ * it may happen that a read call-back is triggered towards the user before this
+ * special write-callback, since both events may come together from the kernel.
+ * Hence under those scenarios where server starts the communication, it is
+ * important not to assume or require that the write-callback(res=0, msgb=NULL)
+ * will be the first one triggered.
+ *
* \param[in] iofd the file descriptor */
void osmo_iofd_notify_connected(struct osmo_io_fd *iofd)
{
--
To view, visit https://gerrit.osmocom.org/c/libosmocore/+/39348?usp=email
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email
Gerrit-MessageType: merged
Gerrit-Project: libosmocore
Gerrit-Branch: pespin/rel-1.10.0
Gerrit-Change-Id: Ica20a050b98d117995a5b625b23ab9faa61aabee
Gerrit-Change-Number: 39348
Gerrit-PatchSet: 1
Gerrit-Owner: pespin <pespin(a)sysmocom.de>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: neels <nhofmeyr(a)sysmocom.de>
Gerrit-Reviewer: osmith <osmith(a)sysmocom.de>
Gerrit-Reviewer: pespin <pespin(a)sysmocom.de>
pespin has submitted this change. ( https://gerrit.osmocom.org/c/libosmocore/+/39347?usp=email )
Change subject: gsmtap: Increase txqueue max_length from 64 to 1024
......................................................................
gsmtap: Increase txqueue max_length from 64 to 1024
The default value may be too low for main loop iterations writing a lot
of output. Increase it to a saner value.
Change-Id: Ie737e1d060ea3ee48631604bbae1b8b84cf787b9
(cherry picked from commit 5edb7d54a8265ad6f44afdd0d07eda041f54ed98)
---
M src/core/gsmtap_util.c
1 file changed, 2 insertions(+), 2 deletions(-)
Approvals:
neels: Looks good to me, but someone else must approve
osmith: Looks good to me, approved
Jenkins Builder: Verified
diff --git a/src/core/gsmtap_util.c b/src/core/gsmtap_util.c
index b64c7b0..62d3ffa 100644
--- a/src/core/gsmtap_util.c
+++ b/src/core/gsmtap_util.c
@@ -496,8 +496,8 @@
if (osmo_iofd_register(gti->out, gti->wq.bfd.fd) < 0)
goto err_cleanup;
- /* osmo write queue previously used was set up with value of 64 */
- osmo_iofd_set_txqueue_max_length(gti->out, 64);
+ /* Use a big enough tx queue to avoid gsmtap messages being dropped: */
+ osmo_iofd_set_txqueue_max_length(gti->out, 1024);
}
return gti;
--
To view, visit https://gerrit.osmocom.org/c/libosmocore/+/39347?usp=email
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email
Gerrit-MessageType: merged
Gerrit-Project: libosmocore
Gerrit-Branch: pespin/rel-1.10.0
Gerrit-Change-Id: Ie737e1d060ea3ee48631604bbae1b8b84cf787b9
Gerrit-Change-Number: 39347
Gerrit-PatchSet: 1
Gerrit-Owner: pespin <pespin(a)sysmocom.de>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: neels <nhofmeyr(a)sysmocom.de>
Gerrit-Reviewer: osmith <osmith(a)sysmocom.de>
Gerrit-Reviewer: pespin <pespin(a)sysmocom.de>
pespin has submitted this change. ( https://gerrit.osmocom.org/c/libosmocore/+/39346?usp=email )
Change subject: osmo_io: Increase default queue max_length from 32 to 1024
......................................................................
osmo_io: Increase default queue max_length from 32 to 1024
The default value is way too low, and can be hit easily by apps
submitting several messages in one main loop run.
This is the case for instance of the Abis OML protocol, where a cascade
of "State Changed Event Reports" are sent for the whole object tree.
Set the default to a saner bigger value. In the end, increasing the
value doesn't hurt normal use of apps, and 1024 is still a fairly small
number to avoid the app eating the whole system memory if something
unexpected happens.
Change-Id: I92d5369af523f6297d45d4b534bb0f349ae74485
(cherry picked from commit 9cbc92e80e8a70131851b363c4438c216e559335)
---
M src/core/osmo_io.c
1 file changed, 2 insertions(+), 2 deletions(-)
Approvals:
osmith: Looks good to me, approved
Jenkins Builder: Verified
neels: Looks good to me, but someone else must approve
diff --git a/src/core/osmo_io.c b/src/core/osmo_io.c
index af096e6..98d8ef3 100644
--- a/src/core/osmo_io.c
+++ b/src/core/osmo_io.c
@@ -637,7 +637,7 @@
* The created osmo_io_fd is initialized with some default settings:
* * msgb allocations size: OSMO_IO_DEFAULT_MSGB_SIZE (1024)
* * msgb headroom: OSMO_IO_DEFAULT_MSGB_HEADROOM (128)
- * * tx_queue depth: 32
+ * * tx_queue depth: 1024
*
* Those values may be adjusted from their defaults by using osmo_iofd_set_alloc_info() and
* osmo_iofd_set_txqueue_max_length() on the osmo_io_fd.
@@ -696,7 +696,7 @@
iofd->msgb_alloc.size = OSMO_IO_DEFAULT_MSGB_SIZE;
iofd->msgb_alloc.headroom = OSMO_IO_DEFAULT_MSGB_HEADROOM;
- iofd->tx_queue.max_length = 32;
+ iofd->tx_queue.max_length = 1024;
INIT_LLIST_HEAD(&iofd->tx_queue.msg_queue);
return iofd;
--
To view, visit https://gerrit.osmocom.org/c/libosmocore/+/39346?usp=email
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email
Gerrit-MessageType: merged
Gerrit-Project: libosmocore
Gerrit-Branch: pespin/rel-1.10.0
Gerrit-Change-Id: I92d5369af523f6297d45d4b534bb0f349ae74485
Gerrit-Change-Number: 39346
Gerrit-PatchSet: 1
Gerrit-Owner: pespin <pespin(a)sysmocom.de>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: neels <nhofmeyr(a)sysmocom.de>
Gerrit-Reviewer: osmith <osmith(a)sysmocom.de>
Gerrit-Reviewer: pespin <pespin(a)sysmocom.de>
Attention is currently required from: dexter.
laforge has posted comments on this change by dexter. ( https://gerrit.osmocom.org/c/pysim/+/39199?usp=email )
Change subject: global_platform: add new command "install_cap"
......................................................................
Patch Set 6:
(1 comment)
File docs/shell.rst:
https://gerrit.osmocom.org/c/pysim/+/39199/comment/44d70710_93c9fc65?usp=em… :
PS6, Line 1058: ~~~~
does this work if the line is not as long as the word above it? I never tried...
--
To view, visit https://gerrit.osmocom.org/c/pysim/+/39199?usp=email
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email
Gerrit-MessageType: comment
Gerrit-Project: pysim
Gerrit-Branch: master
Gerrit-Change-Id: I6cbd37f0fad5579b20e83c27349bd5acc129e6d0
Gerrit-Change-Number: 39199
Gerrit-PatchSet: 6
Gerrit-Owner: dexter <pmaier(a)sysmocom.de>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: laforge <laforge(a)osmocom.org>
Gerrit-Attention: dexter <pmaier(a)sysmocom.de>
Gerrit-Comment-Date: Thu, 16 Jan 2025 13:28:16 +0000
Gerrit-HasComments: Yes
Gerrit-Has-Labels: No
Attention is currently required from: dexter.
laforge has posted comments on this change by laforge. ( https://gerrit.osmocom.org/c/pysim/+/37454?usp=email )
Change subject: global_platform: LOAD and INSTALL [for load] support
......................................................................
Patch Set 7: Code-Review+1
--
To view, visit https://gerrit.osmocom.org/c/pysim/+/37454?usp=email
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email
Gerrit-MessageType: comment
Gerrit-Project: pysim
Gerrit-Branch: master
Gerrit-Change-Id: I924aaeecbb3a72bdb65eefbff6135e4e9570579e
Gerrit-Change-Number: 37454
Gerrit-PatchSet: 7
Gerrit-Owner: laforge <laforge(a)osmocom.org>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: laforge <laforge(a)osmocom.org>
Gerrit-Attention: dexter <pmaier(a)sysmocom.de>
Gerrit-Comment-Date: Thu, 16 Jan 2025 13:27:49 +0000
Gerrit-HasComments: No
Gerrit-Has-Labels: Yes