On 30.05.2023 12:20, Konrad Dybcio wrote: > Ever since the introduction of SMD RPM ICC, we've been dividing the > clock rate by the wrong bus width. This has resulted in: > > - setting wrong (mostly too low) rates, affecting performance > - most often /2 or /4 > - things like DDR never hit their full potential > - the rates were only correct if src bus width == dst bus width > for all src, dst pairs on a given bus > > - Qualcomm using the same wrong logic in their BSP driver in msm-5.x > that ships in production devices today > > - me losing my sanity trying to find this > > Resolve it by using dst_qn, if it exists. > > Fixes: 5e4e6c4d3ae0 ("interconnect: qcom: Add QCS404 interconnect provider driver") > Signed-off-by: Konrad Dybcio <konrad.dybcio@xxxxxxxxxx> > --- The problem is deeper. Chatting with Stephan (+CC), we tackled a few issues (that I will send fixes for in v2): 1. qcom_icc_rpm_set() should take per-node (src_qn->sum_avg, dst_qn->sum_avg) and NOT aggregated bw (unless you want ALL of your nodes on a given provider to "go very fast") 2. the aggregate bw/clk rate calculation should use the node-specific bus widths and not only the bus width of the src/dst node, otherwise the average bw values will be utterly meaningless 3. thanks to (1) and (2) qcom_icc_bus_aggregate() can be remodeled to instead calculate the clock rates for the two rpm contexts, which we can then max() and pass on to the ratesetting call ----8<---- Cutting off Stephan's seal of approval, this is my thinking ---- 4. I *think* Qualcomm really made a mistake in their msm-5.4 driver where they took most of the logic from the current -next state and should have been setting the rate based on the *DST* provider, or at least that's my understanding trying to read the "known good" msm-4.19 driver (which remembers msm-3.0 lol).. Or maybe we should keep src but ensure there's also a final (dst, dst) vote cast: provider->inter_set = false // current state upstream setting apps_proc<->slv_bimc_snoc setting mas_bimc_snoc<->slv_snoc_cnoc setting mas_snoc_cnoc<->qhs_sdc2 provider->inter_set = true // I don't think there's effectively a difference? setting apps_proc<->slv_bimc_snoc setting slv_bimc_snoc<->mas_bimc_snoc setting mas_bimc_snoc<->slv_snoc_cnoc setting slv_snoc_cnoc<->mas_snoc_cnoc setting mas_snoc_cnoc<->qhs_sdc2 all the (mas|slv)_bus1_bus2 are very wide whereas the target nodes are usually 4-, 8- or 16-wide, which without this patch or something equivalent decimates (or actually 2^n-ates) the calculated rates.. Konrad > drivers/interconnect/qcom/icc-rpm.c | 8 +++++--- > 1 file changed, 5 insertions(+), 3 deletions(-) > > diff --git a/drivers/interconnect/qcom/icc-rpm.c b/drivers/interconnect/qcom/icc-rpm.c > index 59be704364bb..58e2a8b1b7c3 100644 > --- a/drivers/interconnect/qcom/icc-rpm.c > +++ b/drivers/interconnect/qcom/icc-rpm.c > @@ -340,7 +340,7 @@ static void qcom_icc_bus_aggregate(struct icc_provider *provider, > static int qcom_icc_set(struct icc_node *src, struct icc_node *dst) > { > struct qcom_icc_provider *qp; > - struct qcom_icc_node *src_qn = NULL, *dst_qn = NULL; > + struct qcom_icc_node *src_qn = NULL, *dst_qn = NULL, *qn = NULL; > struct icc_provider *provider; > u64 active_rate, sleep_rate; > u64 agg_avg[QCOM_SMD_RPM_STATE_NUM], agg_peak[QCOM_SMD_RPM_STATE_NUM]; > @@ -353,6 +353,8 @@ static int qcom_icc_set(struct icc_node *src, struct icc_node *dst) > provider = src->provider; > qp = to_qcom_provider(provider); > > + qn = dst_qn ? dst_qn : src_qn; > + > qcom_icc_bus_aggregate(provider, agg_avg, agg_peak, &max_agg_avg); > > ret = qcom_icc_rpm_set(src_qn, agg_avg); > @@ -372,11 +374,11 @@ static int qcom_icc_set(struct icc_node *src, struct icc_node *dst) > /* Intentionally keep the rates in kHz as that's what RPM accepts */ > active_rate = max(agg_avg[QCOM_SMD_RPM_ACTIVE_STATE], > agg_peak[QCOM_SMD_RPM_ACTIVE_STATE]); > - do_div(active_rate, src_qn->buswidth); > + do_div(active_rate, qn->buswidth); > > sleep_rate = max(agg_avg[QCOM_SMD_RPM_SLEEP_STATE], > agg_peak[QCOM_SMD_RPM_SLEEP_STATE]); > - do_div(sleep_rate, src_qn->buswidth); > + do_div(sleep_rate, qn->buswidth); > > /* > * Downstream checks whether the requested rate is zero, but it makes little sense >