<p>pespin <strong>submitted</strong> this change.</p><p><a href="https://gerrit.osmocom.org/c/osmo-pcu/+/22436">View Change</a></p><div style="white-space:pre-wrap">Approvals:
pespin: Looks good to me, approved
Jenkins Builder: Verified
</div><pre style="font-family: monospace,monospace; white-space: pre-wrap;">sched: Avoid picking TBF with nacked dl blocks when GMSK is required<br><br>Sine we don't yet implement properly all resegmentation of blocks from<br>same MCS family type, when requiring a GMSK DL block (due to GPRS+EGPRS<br>multiplexing limitations) we need to skip retransmitions, otherwise we'd<br>be incorrectly picking a DL block which was already built with a<br>potentially higher MCS value.<br><br>The "DL_PRIO_NEW_DATA" prio serves two purposes:<br>* There's new data to send<br>* There's some nacked data to be retransmitted<br><br>The 2nd purpose has, later on, more priority over the 1st one when the tbf<br>is selected (see gprs_rlcmac_dl_tbf::take_next_bsn()).<br><br>Until now we were handling correctly the case where the tbf was skipped<br>in case the prio was to resend unacked data (DL_PRIO_SENT_DATA), but<br>was incorrectly selected when it'd send nacked data. Let's fix it by<br>specifically checking w->resend_needed() < 0.<br><br>Change-Id: I253de8e1a190a9adb56160f38892c9e43e2c0272<br>---<br>M src/gprs_rlcmac_sched.cpp<br>1 file changed, 28 insertions(+), 3 deletions(-)<br><br></pre><pre style="font-family: monospace,monospace; white-space: pre-wrap;"><span>diff --git a/src/gprs_rlcmac_sched.cpp b/src/gprs_rlcmac_sched.cpp</span><br><span>index 49a8ff2..6dea949 100644</span><br><span>--- a/src/gprs_rlcmac_sched.cpp</span><br><span>+++ b/src/gprs_rlcmac_sched.cpp</span><br><span>@@ -275,6 +275,31 @@</span><br><span> return DL_PRIO_NONE;</span><br><span> }</span><br><span> </span><br><span style="color: hsl(120, 100%, 40%);">+/* Check if next data block of a TBF can be encoded in GMSK [(M)CS1-4]. */</span><br><span style="color: hsl(120, 100%, 40%);">+static bool can_produce_gmsk_data_block_next(struct gprs_rlcmac_dl_tbf *tbf, enum tbf_dl_prio prio)</span><br><span style="color: hsl(120, 100%, 40%);">+{</span><br><span style="color: hsl(120, 100%, 40%);">+ const gprs_rlc_dl_window *w;</span><br><span style="color: hsl(120, 100%, 40%);">+</span><br><span style="color: hsl(120, 100%, 40%);">+ /* GPRS TBFs can always send GMSK */</span><br><span style="color: hsl(120, 100%, 40%);">+ if (!tbf->is_egprs_enabled())</span><br><span style="color: hsl(120, 100%, 40%);">+ return true;</span><br><span style="color: hsl(120, 100%, 40%);">+</span><br><span style="color: hsl(120, 100%, 40%);">+ switch (prio) {</span><br><span style="color: hsl(120, 100%, 40%);">+ case DL_PRIO_CONTROL:</span><br><span style="color: hsl(120, 100%, 40%);">+ /* Control blocks are always CS-1 */</span><br><span style="color: hsl(120, 100%, 40%);">+ return true;</span><br><span style="color: hsl(120, 100%, 40%);">+ case DL_PRIO_NEW_DATA:</span><br><span style="color: hsl(120, 100%, 40%);">+ /* We can send any new data (no block generated yet) using any</span><br><span style="color: hsl(120, 100%, 40%);">+ * MCS. However, we don't (yet) support resegmenting already</span><br><span style="color: hsl(120, 100%, 40%);">+ * sent blocks (NACKed blocks in this case) into lower MCS of</span><br><span style="color: hsl(120, 100%, 40%);">+ * the same family. See OS#4966 */</span><br><span style="color: hsl(120, 100%, 40%);">+ w = static_cast<gprs_rlc_dl_window *>(tbf->window());</span><br><span style="color: hsl(120, 100%, 40%);">+ return w->resend_needed() < 0;</span><br><span style="color: hsl(120, 100%, 40%);">+ default:</span><br><span style="color: hsl(120, 100%, 40%);">+ return false;</span><br><span style="color: hsl(120, 100%, 40%);">+ }</span><br><span style="color: hsl(120, 100%, 40%);">+}</span><br><span style="color: hsl(120, 100%, 40%);">+</span><br><span> static struct msgb *sched_select_downlink(struct gprs_rlcmac_bts *bts,</span><br><span> uint8_t trx, uint8_t ts, uint32_t fn,</span><br><span> uint8_t block_nr, struct gprs_rlcmac_pdch *pdch, enum mcs_kind req_mcs_kind, bool *is_egprs)</span><br><span>@@ -319,9 +344,9 @@</span><br><span> /* If a GPRS (CS1-4/MCS1-4) Dl block is required, downgrade MCS</span><br><span> * below instead of skipping. However, downgrade can only be</span><br><span> * done on new data BSNs (not yet sent) and control blocks. */</span><br><span style="color: hsl(0, 100%, 40%);">- if (req_mcs_kind == EGPRS_GMSK && tbf->is_egprs_enabled() &&</span><br><span style="color: hsl(0, 100%, 40%);">- (prio !=DL_PRIO_CONTROL && prio != DL_PRIO_NEW_DATA)) {</span><br><span style="color: hsl(0, 100%, 40%);">- LOGP(DRLCMACSCHED, LOGL_DEBUG, "Cannot downgrade EGPRS TBF with prio %d\n", prio);</span><br><span style="color: hsl(120, 100%, 40%);">+ if (req_mcs_kind == EGPRS_GMSK && !can_produce_gmsk_data_block_next(tbf, prio)) {</span><br><span style="color: hsl(120, 100%, 40%);">+ LOGP(DRLCMACSCHED, LOGL_DEBUG, "%s Cannot downgrade EGPRS TBF with prio %d\n",</span><br><span style="color: hsl(120, 100%, 40%);">+ tbf_name(tbf), prio);</span><br><span> continue;</span><br><span> }</span><br><span> </span><br><span></span><br></pre><p>To view, visit <a href="https://gerrit.osmocom.org/c/osmo-pcu/+/22436">change 22436</a>. To unsubscribe, or for help writing mail filters, visit <a href="https://gerrit.osmocom.org/settings">settings</a>.</p><div itemscope itemtype="http://schema.org/EmailMessage"><div itemscope itemprop="action" itemtype="http://schema.org/ViewAction"><link itemprop="url" href="https://gerrit.osmocom.org/c/osmo-pcu/+/22436"/><meta itemprop="name" content="View Change"/></div></div>
<div style="display:none"> Gerrit-Project: osmo-pcu </div>
<div style="display:none"> Gerrit-Branch: master </div>
<div style="display:none"> Gerrit-Change-Id: I253de8e1a190a9adb56160f38892c9e43e2c0272 </div>
<div style="display:none"> Gerrit-Change-Number: 22436 </div>
<div style="display:none"> Gerrit-PatchSet: 3 </div>
<div style="display:none"> Gerrit-Owner: pespin <pespin@sysmocom.de> </div>
<div style="display:none"> Gerrit-Reviewer: Jenkins Builder </div>
<div style="display:none"> Gerrit-Reviewer: laforge <laforge@osmocom.org> </div>
<div style="display:none"> Gerrit-Reviewer: osmith <osmith@sysmocom.de> </div>
<div style="display:none"> Gerrit-Reviewer: pespin <pespin@sysmocom.de> </div>
<div style="display:none"> Gerrit-MessageType: merged </div>