github.com/OpenHD/wifibroadcast commits | Ecosyste.ms: OpenCollective

add executable for testing injection max capabilities

ae4d8445aed8970cfb6d99f6de449d390a9b944a authored over 1 year ago

add executable for testing injection max capabilities

3cc970a29c3de58dd68e0168138d8a829341121c authored over 1 year ago

add executable for testing injection max capabilities

d06d3724b45a732b36dc4f032c5abf310739e837 authored over 1 year ago

add executable for testing injection max capabilities

6c493fddb54582e4ce6583cd60194a56bd598cad authored over 1 year ago

add executable for testing injection max capabilities

7e62d441368885440a68c6be2e53319f1f28136c authored over 1 year ago

add executable for testing injection max capabilities

3af2cb68d4a1ec9c80aa318453f350066b0afdce authored over 1 year ago

add executable for testing injection max capabilities

16f71f9f51285974a11c89a9a2cb46749ea9de83 authored over 1 year ago

add executable for testing injection max capabilities

2ffe9460cdc8e88e4a79fadbbcbc2a605193b8a9 authored over 1 year ago

allow mcs up to 31

00ad4fd1605a05afef2599bd4b28cd029d35427c authored over 1 year ago

fix comp after merge

0e55299afbb5083bd85acaa5cb6d069fb6ded450 authored over 1 year ago

document

49381046cf31578326fff23969ec03b8683264fa authored over 1 year ago

fix merge

d191811939faeed403998d2ed5eeb0c631ec873f authored over 1 year ago

Refactor - Unify TX RX and more (#8)

* exp TxRxInstance refactoring

* exp TxRxInstance refactoring - works !

* exp TxRxInstance...

4577993942a400da30bad0c927040acee19b22c6 authored over 1 year ago

Merge pull request #7 from rsaxvc/master

Fix narrowing conversion warning

88473da154aefdc6e586c625e836a78db7d985e1 authored over 1 year ago

Fix narrowing conversion warning

Fix the following warning by casting to the appropriate timeval member types:

In file included ...

a9d1942420955ea1902c13fd23711c4115386b21 authored over 1 year ago

fix comp ENABLE_ADVANCED_DEBUGGING

3c223a04a791178577130bf5bbaa580195ef69c5 authored almost 2 years ago

perform bytes per second and packet loss calculation(s) after decryption (which rejects possibly incorrect received packets)

f81799cdce9b83676d6f1bae8a34d0ca52eac275 authored almost 2 years ago

properly use and only ask for SSSE3 on X86. This way we should be backwards compatible "enough"

95d9b27635b3a422bdb63649b564947cad10902e authored almost 2 years ago

update documentation

11bd355a48dcdabec0bfb1369cc9febe24329d62 authored almost 2 years ago

let's just expose thread safe update methods for all these params

f868abe4e1f885892b0b47e9b6a365de437891c2 authored almost 2 years ago

add option to update STBC index at run time

8e9ea0ae2e19ae7a655e0c4f823e414ed1616cfa authored almost 2 years ago

do not use spd:: shortcut for spdlog

bf26f6900dbf7291bc0757ba1673a78b2dab91a7 authored almost 2 years ago

dummy commit

a1c1fd1040c6cea25fd7932c56e1192181386354 authored almost 2 years ago

only use dbm, mcs index and channel width from "good" packets, aka packets that originated from another wb link

832a05fbb9b15aee0bb68a662846fc03fbe08151 authored almost 2 years ago

only use dbm, mcs index and channel width from "good" packets, aka packets that originated from another wb link

9579521821ff6ec4968d9b8625368c96015bb4bc authored almost 2 years ago

only use dbm, mcs index and channel width from "good" packets, aka packets that originated from another wb link

775dd6c2da28894a8b38baec0d4a88a71d7ff01f authored almost 2 years ago

separate decrypt session key from decrypt success / error packets

4fa2401641127834b75b35fe67650beea0ec4375 authored almost 2 years ago

remove unused

90ee22df164eb8704392bfe34efbbe0bd68e3a00 authored almost 2 years ago

add channel width received on last packet - we need that in openhd for the scan functionality

54dadcb01732b84dfe1f4310870c3e95a3e7c2e9 authored almost 2 years ago

only search for spdlog package if not already added as a target (e.g. that is used in OpenHD)

65c75bb72829b914e400065e16ec9a57bc3e8f32 authored almost 2 years ago

improve logging

b941e4bf4c56c7c6db4fa3114888aaa9386cefc4 authored about 2 years ago

improve logging

e7a1d962f7ae2bd8252b499a7b5c8f71e0c434a1 authored about 2 years ago

add option to update the channel width at run time (I am still not 100% sure which is actually applied, the one you set during iw dev blabla or from the radiotap header)

e0b8fe43228d27a1371940657beb5dd41f737e39 authored about 2 years ago

document

259097ef203d7c71e90dcb0ecbd1d74ef672b8d1 authored about 2 years ago

dirty fix some stats with multiple rx-es

795f948bc56c504a965f3a6e86c55df6bea054a1 authored about 2 years ago

dirty fix some stats with multiple rx-es

54e2a1cfd64ba5cdb5dda3a897beac613dfd08a6 authored about 2 years ago

dirty fix some stats with multiple rx-es

0dbe97384047dd8057324cfbef2c22845316e65d authored about 2 years ago

dirty fix some stats with multiple rx-es

bcc6360ab3bfd63c0057fc37313ce11929a74173 authored about 2 years ago

dirty fix some stats with multiple rx-es

e0455623b6c7562d64fcb1abc11ec02124b5f110 authored about 2 years ago

dirty fix some stats with multiple rx-es

7c7e71453c3940cb8ad173d89770ea5d6c7ad23e authored about 2 years ago

dirty fix some stats with multiple rx-es

af82806a8e29e63a38ac5514ff45d2603a5e0231 authored about 2 years ago

be more close to google c++ style code

828b68325d8fd28973c05a2475c04b47c7bce71d authored about 2 years ago

be more close to google c++ style code

4d668f3e7a9c1ed6702e9791a845f05bdff8684f authored about 2 years ago

dirty fix required for multiple rx-es

b25cbfc962b705d159a1ffbc57b754e30954e1dc authored about 2 years ago

dirty fix required for multiple rx-es

9eff77d41d23bdbb222ff52f8d76cc63d285401b authored about 2 years ago

argh, there seems to be some issue with BlockingConcurrentQueue when used in OpenHD. For now, put the thread safety responsibility one layer up and see if that fixes things.

2996ab3e78b65a388f81641a9ed9c9512770a724 authored about 2 years ago

with the current layout with data queues (that work) this idea doesn't make much sense anymore. delete the unused code

e3222d9ac861a26fe27d42b9110c121e89bb7c03 authored about 2 years ago

move into subdirectory

1b33daa2227e8d07d22fff95f874ca2ddc290b3b authored about 2 years ago

test if we do not add any latency by changing the queue type

896b99e2830c41fa642a512d5aa41ad4c2d91f28 authored about 2 years ago

test if we do not add any latency by changing the queue type

c1d13783b3bcf39300efd709f5b7592f08ac6453 authored about 2 years ago

use BlockingConcurrentQueue - openhd telemetry requires thread-safe enqueueing

73bab2215a338a0abdf5354d63141654893d3770 authored about 2 years ago

store mcs index of last received packet in the stats

9381cf6f9be4d702eaf24070e15ed3ef8be66c50 authored about 2 years ago

use 0 directly for memset

0e2d04f0d4341054693024465f80df687872f1c7 authored about 2 years ago

debug latency in queue

7fdcd060acebf76899e0ca07d4f858ea2da9f7c4 authored about 2 years ago

debug latency in queue

7c4a10857d5857af5ad8fc284d90c8a26892d61f authored about 2 years ago

also set sched max realtime for rx thread

2de12db408d4fd607cb8bbfa48f762c5acfc61c9 authored about 2 years ago

Merge pull request #6 from OpenHD/test

annoying, but if we wanna keep the option to not use aligned fec bloc…

6cfc5b8da5658bca1970fe4fe0ab6b826e1bc5f7 authored about 2 years ago

annoying, but if we wanna keep the option to not use aligned fec blocks (for whatever, discouraged reason) we need this differentiation

13326559d3cd0d863eb1d608568d82731f412390 authored about 2 years ago

Merge pull request #5 from OpenHD/test

feature - add 2 modes of operation, packet queue or block (frames in …

f33c862b6df4b5b6177861b2f28a0374d032d2f5 authored about 2 years ago

feature - add 2 modes of operation, packet queue or block (frames in OpenHD) queue. Using a queue of frames instead of (agnostic) rtp fragments improves the behaviour when the tx needs to drop data (tx errors), since the tx now drops whole frame(s) instead of parts of a frame (which results in non-decodable data). Latency is untouched by that (as long as we are not dropping frame(s) - but when frames need to be dropped, we are already in a wrongly configured state anyways). Also fixes the issue of dropped data on frame(s) consisting of more than packet queue size fragments.

afd6ecd93244d9ee53d944aeaaa924f6beebed8d authored about 2 years ago

Merge pull request #4 from OpenHD/test

eature - add 2 modes of operation, packet queue or block

750afd5d63283e01f16e3439547535783ed774c1 authored about 2 years ago

feature - add 2 modes of operation, packet queue or block (frames in OpenHD) queue. Using a queue of frames instead of (agnostic) rtp fragments improves the behaviour when the tx needs to drop data (tx errors), since the tx now drops whole frame(s) instead of parts of a frame (which results in non-decodable data). Latency is untouched by that (as long as we are not dropping frame(s) - but when frames need to be dropped, we are already in a wrongly configured state anyways). Also fixes the issue of dropped data on frame(s) consisting of more than packet queue size fragments.

fe89993b2daf2da8acc52b2c7ff4044026668610 authored about 2 years ago

feature - add 2 modes of operation, packet queue or block (frames in OpenHD) queue. Using a queue of frames instead of (agnostic) rtp fragments improves the behaviour when the tx needs to drop data (tx errors), since the tx now drops whole frame(s) instead of parts of a frame (which results in non-decodable data). Latency is untouched by that (as long as we are not dropping frame(s) - but when frames need to be dropped, we are already in a wrongly configured state anyways). Also fixes the issue of dropped data on frame(s) consisting of more than packet queue size fragments.

ff96415e6611329b790a9bb96c4db89c1bab6075 authored about 2 years ago

feature - add 2 modes of operation, packet queue or block (frames in OpenHD) queue. Using a queue of frames instead of (agnostic) rtp fragments improves the behaviour when the tx needs to drop data (tx errors), since the tx now drops whole frame(s) instead of parts of a frame (which results in non-decodable data). Latency is untouched by that (as long as we are not dropping frame(s) - but when frames need to be dropped, we are already in a wrongly configured state anyways). Also fixes the issue of dropped data on frame(s) consisting of more than packet queue size fragments.

4709d9487ca83fb8e5fa90484030b9744a2a4111 authored about 2 years ago

feature - add 2 modes of operation, packet queue or block (frames in OpenHD) queue. Using a queue of frames instead of (agnostic) rtp fragments improves the behaviour when the tx needs to drop data (tx errors), since the tx now drops whole frame(s) instead of parts of a frame (which results in non-decodable data). Latency is untouched by that (as long as we are not dropping frame(s) - but when frames need to be dropped, we are already in a wrongly configured state anyways). Also fixes the issue of dropped data on frame(s) consisting of more than packet queue size fragments.

b2b084a6adbe359c189b91626ef505730ce4e65e authored about 2 years ago

feature - add 2 modes of operation, packet queue or block (frames in OpenHD) queue. Using a queue of frames instead of (agnostic) rtp fragments improves the behaviour when the tx needs to drop data (tx errors), since the tx now drops whole frame(s) instead of parts of a frame (which results in non-decodable data). Latency is untouched by that (as long as we are not dropping frame(s) - but when frames need to be dropped, we are already in a wrongly configured state anyways). Also fixes the issue of dropped data on frame(s) consisting of more than packet queue size fragments.

a7c0108c9b091ab81b597cdcc36125439eccf479 authored about 2 years ago

feature - add 2 modes of operation, packet queue or block (frames in OpenHD) queue. Using a queue of frames instead of (agnostic) rtp fragments improves the behaviour when the tx needs to drop data (tx errors), since the tx now drops whole frame(s) instead of parts of a frame (which results in non-decodable data). Latency is untouched by that (as long as we are not dropping frame(s) - but when frames need to be dropped, we are already in a wrongly configured state anyways). Also fixes the issue of dropped data on frame(s) consisting of more than packet queue size fragments.

b3ae1c650ad7be64c10137f89c55a16db4f24024 authored about 2 years ago

feature - add 2 modes of operation, packet queue or block (frames in OpenHD) queue. Using a queue of frames instead of (agnostic) rtp fragments improves the behaviour when the tx needs to drop data (tx errors), since the tx now drops whole frame(s) instead of parts of a frame (which results in non-decodable data). Latency is untouched by that (as long as we are not dropping frame(s) - but when frames need to be dropped, we are already in a wrongly configured state anyways). Also fixes the issue of dropped data on frame(s) consisting of more than packet queue size fragments.

43702242e55d28986608e50698efee7c69bd8958 authored about 2 years ago

refactor

0a3f9916752052f68dd590c3d9344e3d3530a93a authored about 2 years ago

refactor

cae7331f0743ccecd88a8db3c345eebb43497881 authored about 2 years ago

Merge pull request #3 from OpenHD/test

refactor and document

6ff209989f026b741c65d7669a80b6f7fc830a8d authored about 2 years ago

refactor

229f59525ff64c48c9772564d9a0e635346e1b15 authored about 2 years ago

document

3955bec6ee7a9b4b019a95a05771e5afb063380d authored about 2 years ago

document

58d8193a67ce5ea2520f23421659ea4ef09c304b authored about 2 years ago

Merge pull request #2 from OpenHD/test

move packet sequence number (loss perc calc) into its own class,

d3cc991fa4b50f85c73439a823f467f1a42ed7f5 authored about 2 years ago