Attention is currently required from: pespin, fixeria. neels has posted comments on this change. ( https://gerrit.osmocom.org/c/libosmo-gprs/+/31099 )

Change subject: rlcmac: Initial implementation of UL TBF assignment and scheduler ......................................................................

Patch Set 16:

(8 comments)

Commit Message:

https://gerrit.osmocom.org/c/libosmo-gprs/+/31099/comment/0b02cb09_fc0fce52 PS16, Line 9: er t ("step")

Patchset:

PS16: in general i wish you wouldn't have a habit of putting comments in the same line as code...

this is a lot and i can't really comment on functional accuracy, but looks sane in general. +1 when the T_def has a desc and a proper X-nr assigned

File src/rlcmac/rlcmac.c:

https://gerrit.osmocom.org/c/libosmo-gprs/+/31099/comment/5f53ac77_55fa7d08 PS16, Line 40: { .T=1, .default_val=1, .unit=OSMO_TDEF_S, .desc="foobar", .val=0 }, is this a regression test? doesn't look like one. so probably: add a desc, give it an unused T number like -32 currently and add X32 to https://osmocom.org/projects/cellular-infrastructure/wiki/List_of_Timer_numb...

drop .val = 0, it is redundant. (also because tdefs will be initialized to default_val so = 0 may be confusing)

File src/rlcmac/tbf_ul_ass_fsm.c:

https://gerrit.osmocom.org/c/libosmo-gprs/+/31099/comment/9a7e3b97_946dd565 PS16, Line 53: [GPRS_RLCMAC_TBF_UL_ASS_ST_COMPL] = { }, (could just omit unset ones)

File src/rlcmac/tbf_ul_fsm.c:

https://gerrit.osmocom.org/c/libosmo-gprs/+/31099/comment/d87efcda_1bebc6cb PS16, Line 42: [GPRS_RLCMAC_TBF_UL_ST_FLOW] = { }, (same)

https://gerrit.osmocom.org/c/libosmo-gprs/+/31099/comment/519c7a11_48025cc4 PS16, Line 93: } drop this func?

https://gerrit.osmocom.org/c/libosmo-gprs/+/31099/comment/f6c20afd_adaeff1c PS16, Line 147: (extra blank line)

https://gerrit.osmocom.org/c/libosmo-gprs/+/31099/comment/e2fd11a8_57d45d51 PS16, Line 207: (trailing blank line)