CoreSight

coresight@lists.linaro.org

6 participants
2564 discussions

[PATCH 1/2] perf: cs-etm: No-op refactor of synth opt usage

by James Clark

sample_branches and sample_instructions are already saved in the synth_opts struct. Other usages like synth_opts.last_branch don't save a value, so make this more consistent by always going through synth_opts and not saving duplicate values. Signed-off-by: James Clark <james.clark(a)arm.com> --- tools/perf/util/cs-etm.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/tools/perf/util/cs-etm.c b/tools/perf/util/cs-etm.c index 4f672f7d008c..796a065a500e 100644 --- a/tools/perf/util/cs-etm.c +++ b/tools/perf/util/cs-etm.c @@ -50,8 +50,6 @@ struct cs_etm_auxtrace { u8 timeless_decoding; u8 snapshot_mode; u8 data_queued; - u8 sample_branches; - u8 sample_instructions; int num_cpu; u64 latest_kernel_timestamp; @@ -410,8 +408,8 @@ static void cs_etm__packet_swap(struct cs_etm_auxtrace *etm, { struct cs_etm_packet *tmp; - if (etm->sample_branches || etm->synth_opts.last_branch || - etm->sample_instructions) { + if (etm->synth_opts.branches || etm->synth_opts.last_branch || + etm->synth_opts.instructions) { /* * Swap PACKET with PREV_PACKET: PACKET becomes PREV_PACKET for * the next incoming packet. @@ -1365,7 +1363,6 @@ static int cs_etm__synth_events(struct cs_etm_auxtrace *etm, err = cs_etm__synth_event(session, &attr, id); if (err) return err; - etm->sample_branches = true; etm->branches_sample_type = attr.sample_type; etm->branches_id = id; id += 1; @@ -1389,7 +1386,6 @@ static int cs_etm__synth_events(struct cs_etm_auxtrace *etm, err = cs_etm__synth_event(session, &attr, id); if (err) return err; - etm->sample_instructions = true; etm->instructions_sample_type = attr.sample_type; etm->instructions_id = id; id += 1; @@ -1420,7 +1416,7 @@ static int cs_etm__sample(struct cs_etm_queue *etmq, tidq->prev_packet->last_instr_taken_branch) cs_etm__update_last_branch_rb(etmq, tidq); - if (etm->sample_instructions && + if (etm->synth_opts.instructions && tidq->period_instructions >= etm->instructions_sample_period) { /* * Emit instruction sample periodically @@ -1503,7 +1499,7 @@ static int cs_etm__sample(struct cs_etm_queue *etmq, } } - if (etm->sample_branches) { + if (etm->synth_opts.branches) { bool generate_sample = false; /* Generate sample for tracing on packet */ @@ -1582,7 +1578,7 @@ static int cs_etm__flush(struct cs_etm_queue *etmq, } - if (etm->sample_branches && + if (etm->synth_opts.branches && tidq->prev_packet->sample_type == CS_ETM_RANGE) { err = cs_etm__synth_branch_sample(etmq, tidq); if (err) -- 2.28.0

3 years, 9 months

Re: [PATCH v3 10/10] ARM: dts: msm: Add coresight components for SM8250

by Mike Leach

Hi, It would appear here that normal coresight funnels appear between tpdm and tpda devices. This is contrary to my understanding of your hardware and the descriptions and conversations we had in patch 00/10 in this set. trace path appears to be tpdm_mm => funnel@6c0b000 => funnel@6c2d000 => tpda@6004000 => funnel@6005000 => rest of ATB The documentation states that the TPDA is responsible for for funneling and packetising - so why do we have extra funnels? I have no way of knowing if that topology is right or wrong. Are the funnels between tpdm and tpda actual coresight funnels, a component that is register compatible with CS funnels but designed for the TPDM / TPDA transport, or the funnels are being used, but without driving ATID from the sources. It is difficult to review this set without better information on the hardware represented. Regards Mike On Wed, 9 Feb 2022 at 10:58, Mao Jinlong <quic_jinlmao(a)quicinc.com> wrote: > > Add coresight device tree for sm8250. STM/ETM/TPDM are added. > > Signed-off-by: Tao Zhang <quic_taozha(a)quicinc.com> > Signed-off-by: Mao Jinlong <quic_jinlmao(a)quicinc.com> > --- > .../arm64/boot/dts/qcom/sm8250-coresight.dtsi | 690 ++++++++++++++++++ > arch/arm64/boot/dts/qcom/sm8250.dtsi | 2 + > 2 files changed, 692 insertions(+) > create mode 100644 arch/arm64/boot/dts/qcom/sm8250-coresight.dtsi > > diff --git a/arch/arm64/boot/dts/qcom/sm8250-coresight.dtsi b/arch/arm64/boot/dts/qcom/sm8250-coresight.dtsi > new file mode 100644 > index 000000000000..732dab7f8674 > --- /dev/null > +++ b/arch/arm64/boot/dts/qcom/sm8250-coresight.dtsi > @@ -0,0 +1,690 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved. > + */ > + > +&soc { > + > + stm@6002000 { > + compatible = "arm,coresight-stm", "arm,primecell"; > + reg = <0 0x06002000 0 0x1000>, > + <0 0x16280000 0 0x180000>; > + reg-names = "stm-base", "stm-stimulus-base"; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + stm_out: endpoint { > + remote-endpoint = > + <&funnel0_in7>; > + }; > + }; > + }; > + }; > + > + funnel@6041000 { > + compatible = "arm,coresight-dynamic-funnel", "arm,primecell"; > + reg = <0 0x06041000 0 0x1000>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + funnel0_out: endpoint { > + remote-endpoint = > + <&merge_funnel_in0>; > + }; > + }; > + }; > + > + in-ports { > + #address-cells = <1>; > + #size-cells = <0>; > + > + port@6 { > + reg = <6>; > + funnel_in0_in_funnel_qatb: endpoint { > + remote-endpoint = > + <&funnel_qatb_out_funnel_in0>; > + }; > + }; > + > + port@7 { > + reg = <7>; > + funnel0_in7: endpoint { > + remote-endpoint = <&stm_out>; > + }; > + }; > + }; > + }; > + > + funnel@6042000 { > + compatible = "arm,coresight-dynamic-funnel", "arm,primecell"; > + reg = <0 0x06042000 0 0x1000>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + funnel2_out: endpoint { > + remote-endpoint = > + <&merge_funnel_in2>; > + }; > + }; > + }; > + > + in-ports { > + #address-cells = <1>; > + #size-cells = <0>; > + > + port@2 { > + reg = <4>; > + funnel2_in5: endpoint { > + remote-endpoint = > + <&apss_merge_funnel_out>; > + }; > + }; > + }; > + }; > + > + funnel@6b04000 { > + compatible = "arm,coresight-dynamic-funnel", "arm,primecell"; > + arm,primecell-periphid = <0x000bb908>; > + > + reg = <0 0x6b04000 0 0x1000>; > + reg-names = "funnel-base"; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + merge_funnel_out: endpoint { > + remote-endpoint = > + <&etf_in>; > + }; > + }; > + }; > + > + in-ports { > + #address-cells = <1>; > + #size-cells = <0>; > + > + port@7 { > + reg = <7>; > + funnel_swao_in_funnel_merg: endpoint { > + remote-endpoint= > + <&funnel_merg_out_funnel_swao>; > + }; > + }; > + }; > + > + }; > + > + funnel@6045000 { > + compatible = "arm,coresight-dynamic-funnel", "arm,primecell"; > + reg = <0 0x06045000 0 0x1000>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + funnel_merg_out_funnel_swao: endpoint { > + remote-endpoint = <&funnel_swao_in_funnel_merg>; > + }; > + }; > + }; > + > + in-ports { > + #address-cells = <1>; > + #size-cells = <0>; > + > + port@1 { > + reg = <0>; > + merge_funnel_in0: endpoint { > + remote-endpoint = > + <&funnel0_out>; > + }; > + }; > + > + port@2 { > + reg = <1>; > + merge_funnel_in2: endpoint { > + remote-endpoint = > + <&funnel2_out>; > + }; > + }; > + }; > + }; > + > + replicator@6046000 { > + compatible = "arm,coresight-dynamic-replicator", "arm,primecell"; > + reg = <0 0x06046000 0 0x1000>; > + > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + replicator_out: endpoint { > + remote-endpoint = <&etr_in>; > + }; > + }; > + }; > + > + in-ports { > + port { > + replicator_cx_in_swao_out: endpoint { > + remote-endpoint = <&replicator_swao_out_cx_in>; > + }; > + }; > + }; > + }; > + > + replicator@6b06000 { > + compatible = "arm,coresight-dynamic-replicator", "arm,primecell"; > + reg = <0 0x06b06000 0 0x1000>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + replicator_swao_out_cx_in: endpoint { > + remote-endpoint = <&replicator_cx_in_swao_out>; > + }; > + }; > + }; > + > + in-ports { > + port { > + replicator_in: endpoint { > + remote-endpoint = <&etf_out>; > + }; > + }; > + }; > + }; > + > + etf@6b05000 { > + compatible = "arm,coresight-tmc", "arm,primecell"; > + reg = <0 0x6b05000 0 0x1000>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + etf_out: endpoint { > + remote-endpoint = > + <&replicator_in>; > + }; > + }; > + }; > + > + in-ports { > + #address-cells = <1>; > + #size-cells = <0>; > + > + port@1 { > + reg = <0>; > + etf_in: endpoint { > + remote-endpoint = > + <&merge_funnel_out>; > + }; > + }; > + }; > + }; > + > + etr@6048000 { > + compatible = "arm,coresight-tmc", "arm,primecell"; > + reg = <0 0x06048000 0 0x1000>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + arm,scatter-gather; > + > + in-ports { > + port { > + etr_in: endpoint { > + remote-endpoint = > + <&replicator_out>; > + }; > + }; > + }; > + }; > + > + etm@7040000 { > + compatible = "arm,coresight-etm4x", "arm,primecell"; > + reg = <0 0x07040000 0 0x1000>; > + > + cpu = <&CPU0>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + arm,coresight-loses-context-with-cpu; > + > + out-ports { > + port { > + etm0_out: endpoint { > + remote-endpoint = > + <&apss_funnel_in0>; > + }; > + }; > + }; > + }; > + > + etm@7140000 { > + compatible = "arm,coresight-etm4x", "arm,primecell"; > + reg = <0 0x07140000 0 0x1000>; > + > + cpu = <&CPU1>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + arm,coresight-loses-context-with-cpu; > + > + out-ports { > + port { > + etm1_out: endpoint { > + remote-endpoint = > + <&apss_funnel_in1>; > + }; > + }; > + }; > + }; > + > + etm@7240000 { > + compatible = "arm,coresight-etm4x", "arm,primecell"; > + reg = <0 0x07240000 0 0x1000>; > + > + cpu = <&CPU2>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + arm,coresight-loses-context-with-cpu; > + > + out-ports { > + port { > + etm2_out: endpoint { > + remote-endpoint = > + <&apss_funnel_in2>; > + }; > + }; > + }; > + }; > + > + etm@7340000 { > + compatible = "arm,coresight-etm4x", "arm,primecell"; > + reg = <0 0x07340000 0 0x1000>; > + > + cpu = <&CPU3>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + arm,coresight-loses-context-with-cpu; > + > + out-ports { > + port { > + etm3_out: endpoint { > + remote-endpoint = > + <&apss_funnel_in3>; > + }; > + }; > + }; > + }; > + > + etm@7440000 { > + compatible = "arm,coresight-etm4x", "arm,primecell"; > + reg = <0 0x07440000 0 0x1000>; > + > + cpu = <&CPU4>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + arm,coresight-loses-context-with-cpu; > + > + out-ports { > + port { > + etm4_out: endpoint { > + remote-endpoint = > + <&apss_funnel_in4>; > + }; > + }; > + }; > + }; > + > + etm@7540000 { > + compatible = "arm,coresight-etm4x", "arm,primecell"; > + reg = <0 0x07540000 0 0x1000>; > + > + cpu = <&CPU5>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + arm,coresight-loses-context-with-cpu; > + > + out-ports { > + port { > + etm5_out: endpoint { > + remote-endpoint = > + <&apss_funnel_in5>; > + }; > + }; > + }; > + }; > + > + etm@7640000 { > + compatible = "arm,coresight-etm4x", "arm,primecell"; > + reg = <0 0x07640000 0 0x1000>; > + > + cpu = <&CPU6>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + arm,coresight-loses-context-with-cpu; > + > + out-ports { > + port { > + etm6_out: endpoint { > + remote-endpoint = > + <&apss_funnel_in6>; > + }; > + }; > + }; > + }; > + > + etm@7740000 { > + compatible = "arm,coresight-etm4x", "arm,primecell"; > + reg = <0 0x07740000 0 0x1000>; > + > + cpu = <&CPU7>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + arm,coresight-loses-context-with-cpu; > + > + out-ports { > + port { > + etm7_out: endpoint { > + remote-endpoint = > + <&apss_funnel_in7>; > + }; > + }; > + }; > + }; > + > + funnel@7800000 { > + compatible = "arm,coresight-dynamic-funnel", "arm,primecell"; > + reg = <0 0x07800000 0 0x1000>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + apss_funnel_out: endpoint { > + remote-endpoint = > + <&apss_merge_funnel_in>; > + }; > + }; > + }; > + > + in-ports { > + #address-cells = <1>; > + #size-cells = <0>; > + > + port@0 { > + reg = <0>; > + apss_funnel_in0: endpoint { > + remote-endpoint = > + <&etm0_out>; > + }; > + }; > + > + port@1 { > + reg = <1>; > + apss_funnel_in1: endpoint { > + remote-endpoint = > + <&etm1_out>; > + }; > + }; > + > + port@2 { > + reg = <2>; > + apss_funnel_in2: endpoint { > + remote-endpoint = > + <&etm2_out>; > + }; > + }; > + > + port@3 { > + reg = <3>; > + apss_funnel_in3: endpoint { > + remote-endpoint = > + <&etm3_out>; > + }; > + }; > + > + port@4 { > + reg = <4>; > + apss_funnel_in4: endpoint { > + remote-endpoint = > + <&etm4_out>; > + }; > + }; > + > + port@5 { > + reg = <5>; > + apss_funnel_in5: endpoint { > + remote-endpoint = > + <&etm5_out>; > + }; > + }; > + > + port@6 { > + reg = <6>; > + apss_funnel_in6: endpoint { > + remote-endpoint = > + <&etm6_out>; > + }; > + }; > + > + port@7 { > + reg = <7>; > + apss_funnel_in7: endpoint { > + remote-endpoint = > + <&etm7_out>; > + }; > + }; > + }; > + }; > + > + funnel@7810000 { > + compatible = "arm,coresight-dynamic-funnel", "arm,primecell"; > + reg = <0 0x07810000 0 0x1000>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + apss_merge_funnel_out: endpoint { > + remote-endpoint = > + <&funnel2_in5>; > + }; > + }; > + }; > + > + in-ports { > + port@1 { > + reg = <0>; > + apss_merge_funnel_in: endpoint { > + remote-endpoint = > + <&apss_funnel_out>; > + }; > + }; > + }; > + }; > + > + tpdm_mm: mm.tpdm@6c08000 { > + compatible = "arm,primecell"; > + reg = <0 0x6c08000 0 0x1000>; > + reg-names = "tpdm-base"; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + qcom,msr-fix-req; > + > + out-ports { > + port { > + tpdm_mm_out_funnel_dl_mm: endpoint { > + remote-endpoint = > + <&funnel_dl_mm_in_tpdm_mm>; > + }; > + }; > + }; > + }; > + > + funnel@6c0b000 { > + compatible = "arm,coresight-dynamic-funnel", "arm,primecell"; > + > + reg = <0 0x6c0b000 0 0x1000>; > + reg-names = "funnel-base"; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + funnel_dl_mm_out_funnel_dl_center: endpoint { > + remote-endpoint = > + <&funnel_dl_center_in_funnel_dl_mm>; > + }; > + }; > + }; > + > + in-ports { > + #address-cells = <1>; > + #size-cells = <0>; > + > + port@3 { > + reg = <3>; > + funnel_dl_mm_in_tpdm_mm: endpoint { > + remote-endpoint = > + <&tpdm_mm_out_funnel_dl_mm>; > + }; > + }; > + }; > + }; > + > + funnel@6c2d000 { > + compatible = "arm,coresight-dynamic-funnel", "arm,primecell"; > + > + reg = <0 0x6c2d000 0 0x1000>; > + reg-names = "funnel-base"; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + #address-cells = <1>; > + #size-cells = <0>; > + port { > + tpdm_mm_out_tpda9: endpoint { > + remote-endpoint = > + <&tpda_9_in_tpdm_mm>; > + }; > + }; > + }; > + > + in-ports { > + #address-cells = <1>; > + #size-cells = <0>; > + > + port@2 { > + reg = <2>; > + funnel_dl_center_in_funnel_dl_mm: endpoint { > + remote-endpoint = > + <&funnel_dl_mm_out_funnel_dl_center>; > + }; > + }; > + }; > + }; > + > + tpda@6004000 { > + compatible = "arm,primecell"; > + reg = <0 0x6004000 0 0x1000>; > + reg-names = "tpda-base"; > + > + qcom,tpda-atid = <65>; > + > + qcom,dsb-elem-size = <9 32>, > + <10 32>, > + <14 32>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + reg = <0>; > + tpda_out_funnel_qatb: endpoint { > + remote-endpoint = > + <&funnel_qatb_in_tpda>; > + }; > + }; > + }; > + > + in-ports { > + #address-cells = <1>; > + #size-cells = <0>; > + > + port@9 { > + reg = <9>; > + tpda_9_in_tpdm_mm: endpoint { > + remote-endpoint = > + <&tpdm_mm_out_tpda9>; > + }; > + }; > + }; > + }; > + > + funnel@6005000 { > + compatible = "arm,primecell"; > + > + reg = <0 0x6005000 0 0x1000>; > + reg-names = "funnel-base"; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + funnel_qatb_out_funnel_in0: endpoint { > + remote-endpoint = > + <&funnel_in0_in_funnel_qatb>; > + }; > + }; > + }; > + > + in-ports { > + #address-cells = <1>; > + #size-cells = <0>; > + > + port@0 { > + reg = <0>; > + funnel_qatb_in_tpda: endpoint { > + remote-endpoint = > + <&tpda_out_funnel_qatb>; > + }; > + }; > + }; > + }; > +}; > diff --git a/arch/arm64/boot/dts/qcom/sm8250.dtsi b/arch/arm64/boot/dts/qcom/sm8250.dtsi > index 5617a46e5ccd..d1837e3df30a 100644 > --- a/arch/arm64/boot/dts/qcom/sm8250.dtsi > +++ b/arch/arm64/boot/dts/qcom/sm8250.dtsi > @@ -5323,3 +5323,5 @@ > }; > }; > }; > + > +#include "sm8250-coresight.dtsi" > -- > 2.17.1 > -- Mike Leach Principal Engineer, ARM Ltd. Manchester Design Centre. UK

3 years, 9 months

Re: [PATCH v3 05/10] coresight-tpdm: Add DSB dataset support

by Mike Leach

Hi This patch does not apply cleanly:- ======================================= git apply -v ../patch.in/qcom/qcom-05.patch Checking patch drivers/hwtracing/coresight/coresight-tpdm.c... error: while searching for: DEFINE_CORESIGHT_DEVLIST(tpdm_devs, "tpdm"); /** TPDM enable operations **/ static int tpdm_enable(struct coresight_device *csdev, struct perf_event *event, u32 mode) { error: patch failed: drivers/hwtracing/coresight/coresight-tpdm.c:20 error: drivers/hwtracing/coresight/coresight-tpdm.c: patch does not apply Checking patch drivers/hwtracing/coresight/coresight-tpdm.h... ======================================= In patch 3 - the first time coresight-tpdm.c. is added the comment line is /* TPDM enable operations */ Note the single * in the comment. This would seem to indicate missing intermediate patches, or patches from different trees / times. A coherent patch set is needed for review Regards Mike On Wed, 9 Feb 2022 at 10:57, Mao Jinlong <quic_jinlmao(a)quicinc.com> wrote: > > TPDM serves as data collection component for various dataset types. > DSB(Discrete Single Bit) is one of the dataset types. DSB subunit > can be enabled for data collection by writing 1 to the first bit of > DSB_CR register. This change is to add enable/disable function for > DSB dataset by writing DSB_CR register. > > Signed-off-by: Tao Zhang <quic_taozha(a)quicinc.com> > Signed-off-by: Mao Jinlong <quic_jinlmao(a)quicinc.com> > --- > drivers/hwtracing/coresight/coresight-tpdm.c | 57 ++++++++++++++++++++ > drivers/hwtracing/coresight/coresight-tpdm.h | 21 ++++++++ > 2 files changed, 78 insertions(+) > > diff --git a/drivers/hwtracing/coresight/coresight-tpdm.c b/drivers/hwtracing/coresight/coresight-tpdm.c > index 51b8b17e6a80..c6480b7389b0 100644 > --- a/drivers/hwtracing/coresight/coresight-tpdm.c > +++ b/drivers/hwtracing/coresight/coresight-tpdm.c > @@ -20,7 +20,28 @@ > > DEFINE_CORESIGHT_DEVLIST(tpdm_devs, "tpdm"); > > +static void tpdm_enable_dsb(struct tpdm_drvdata *drvdata) > +{ > + u32 val; > + > + /* Set the enable bit of DSB control register to 1 */ > + val = readl_relaxed(drvdata->base + TPDM_DSB_CR); > + val = val | BIT(0); > + writel_relaxed(val, drvdata->base + TPDM_DSB_CR); > +} > + > /** TPDM enable operations **/ > +static void _tpdm_enable(struct tpdm_drvdata *drvdata) > +{ > + CS_UNLOCK(drvdata->base); > + > + /* Check if DSB datasets is present for TPDM. */ > + if (test_bit(TPDM_DS_DSB, drvdata->datasets)) > + tpdm_enable_dsb(drvdata); > + > + CS_LOCK(drvdata->base); > +} > + > static int tpdm_enable(struct coresight_device *csdev, > struct perf_event *event, u32 mode) > { > @@ -32,6 +53,7 @@ static int tpdm_enable(struct coresight_device *csdev, > return -EBUSY; > } > > + _tpdm_enable(drvdata); > drvdata->enable = true; > mutex_unlock(&drvdata->lock); > > @@ -39,7 +61,29 @@ static int tpdm_enable(struct coresight_device *csdev, > return 0; > } > > +static void tpdm_disable_dsb(struct tpdm_drvdata *drvdata) > +{ > + u32 val; > + > + /* Set the enable bit of DSB control register to 0 */ > + val = readl_relaxed(drvdata->base + TPDM_DSB_CR); > + val = val & ~BIT(0); > + writel_relaxed(val, drvdata->base + TPDM_DSB_CR); > +} > + > /** TPDM disable operations **/ > +static void _tpdm_disable(struct tpdm_drvdata *drvdata) > +{ > + CS_UNLOCK(drvdata->base); > + > + /* Check if DSB datasets is present for TPDM. */ > + if (test_bit(TPDM_DS_DSB, drvdata->datasets)) > + tpdm_disable_dsb(drvdata); > + > + CS_LOCK(drvdata->base); > + > +} > + > static void tpdm_disable(struct coresight_device *csdev, > struct perf_event *event) > { > @@ -51,6 +95,7 @@ static void tpdm_disable(struct coresight_device *csdev, > return; > } > > + _tpdm_disable(drvdata); > drvdata->enable = false; > mutex_unlock(&drvdata->lock); > > @@ -76,7 +121,19 @@ static const struct coresight_ops tpdm_cs_ops = { > > static void tpdm_init_default_data(struct tpdm_drvdata *drvdata) > { > + int i; > + u32 pidr; > + > drvdata->traceid = coresight_get_system_trace_id(); > + > + CS_UNLOCK(drvdata->base); > + /* Get the datasets present on the TPDM. */ > + pidr = readl_relaxed(drvdata->base + CORESIGHT_PERIPHIDR0); > + for (i = 0; i < TPDM_DATASETS; i++) { > + if (pidr & BIT(i)) > + __set_bit(i, drvdata->datasets); > + } > + CS_LOCK(drvdata->base); > } > > static int tpdm_probe(struct amba_device *adev, const struct amba_id *id) > diff --git a/drivers/hwtracing/coresight/coresight-tpdm.h b/drivers/hwtracing/coresight/coresight-tpdm.h > index 2effbabf349b..cb3ddc6c89ae 100644 > --- a/drivers/hwtracing/coresight/coresight-tpdm.h > +++ b/drivers/hwtracing/coresight/coresight-tpdm.h > @@ -6,6 +6,25 @@ > #ifndef _CORESIGHT_CORESIGHT_TPDM_H > #define _CORESIGHT_CORESIGHT_TPDM_H > > +/* The max number of the datasets that TPDM supports */ > +#define TPDM_DATASETS 7 > + > +/* DSB Subunit Registers */ > +#define TPDM_DSB_CR (0x780) > + > +/** > + * This enum is for PERIPHIDR0 register of TPDM. > + * The fields [6:0] of PERIPHIDR0 are used to determine what > + * interfaces and subunits are present on a given TPDM. > + * > + * PERIPHIDR0[0] : Fix to 1 if ImplDef subunit present, else 0 > + * PERIPHIDR0[1] : Fix to 1 if DSB subunit present, else 0 > + */ > +enum tpdm_dataset { > + TPDM_DS_IMPLDEF, > + TPDM_DS_DSB, > +}; > + > /** > * struct tpdm_drvdata - specifics associated to an TPDM component > * @base: memory mapped base address for this component. > @@ -13,6 +32,7 @@ > * @csdev: component vitals needed by the framework. > * @lock: lock for the enable value. > * @enable: enable status of the component. > + * @datasets: The datasets types present of the TPDM. > * @traceid: value of the current ID for this component. > */ > > @@ -22,6 +42,7 @@ struct tpdm_drvdata { > struct coresight_device *csdev; > struct mutex lock; > bool enable; > + DECLARE_BITMAP(datasets, TPDM_DATASETS); > int traceid; > }; > > -- > 2.17.1 > -- Mike Leach Principal Engineer, ARM Ltd. Manchester Design Centre. UK

3 years, 9 months

Re: [PATCH v3 03/10] Coresight: Add coresight TPDM source driver

by Mike Leach

HI, On Wed, 9 Feb 2022 at 10:57, Mao Jinlong <quic_jinlmao(a)quicinc.com> wrote: > > Add driver to support Coresight device TPDM (Trace, Profiling and > Diagnostics Monitor). TPDM is a monitor to collect data from > different datasets. This change is to add probe/enable/disable > functions for tpdm source. > > Signed-off-by: Tao Zhang <quic_taozha(a)quicinc.com> > Signed-off-by: Mao Jinlong <quic_jinlmao(a)quicinc.com> > --- > drivers/hwtracing/coresight/Kconfig | 13 ++ > drivers/hwtracing/coresight/Makefile | 1 + > drivers/hwtracing/coresight/coresight-core.c | 3 +- > drivers/hwtracing/coresight/coresight-tpdm.c | 159 +++++++++++++++++++ > drivers/hwtracing/coresight/coresight-tpdm.h | 28 ++++ > include/linux/coresight.h | 1 + > 6 files changed, 204 insertions(+), 1 deletion(-) > create mode 100644 drivers/hwtracing/coresight/coresight-tpdm.c > create mode 100644 drivers/hwtracing/coresight/coresight-tpdm.h > > diff --git a/drivers/hwtracing/coresight/Kconfig b/drivers/hwtracing/coresight/Kconfig > index 514a9b8086e3..5c506a1cd08f 100644 > --- a/drivers/hwtracing/coresight/Kconfig > +++ b/drivers/hwtracing/coresight/Kconfig > @@ -201,4 +201,17 @@ config CORESIGHT_TRBE > > To compile this driver as a module, choose M here: the module will be > called coresight-trbe. > + > +config CORESIGHT_TPDM > + tristate "CoreSight Trace, Profiling & Diagnostics Monitor driver" > + select CORESIGHT_LINKS_AND_SINKS > + help > + This driver provides support for configuring monitor. Monitors are > + primarily responsible for data set collection and support the > + ability to collect any permutation of data set types. Monitors are > + also responsible for interaction with system cross triggering. > + > + To compile this driver as a module, choose M here: the module will be > + called coresight-tpdm. > + > endif > diff --git a/drivers/hwtracing/coresight/Makefile b/drivers/hwtracing/coresight/Makefile > index b6c4a48140ec..e7392a0dddeb 100644 > --- a/drivers/hwtracing/coresight/Makefile > +++ b/drivers/hwtracing/coresight/Makefile > @@ -25,5 +25,6 @@ obj-$(CONFIG_CORESIGHT_CPU_DEBUG) += coresight-cpu-debug.o > obj-$(CONFIG_CORESIGHT_CATU) += coresight-catu.o > obj-$(CONFIG_CORESIGHT_CTI) += coresight-cti.o > obj-$(CONFIG_CORESIGHT_TRBE) += coresight-trbe.o > +obj-$(CONFIG_CORESIGHT_TPDM) += coresight-tpdm.o > coresight-cti-y := coresight-cti-core.o coresight-cti-platform.o \ > coresight-cti-sysfs.o > diff --git a/drivers/hwtracing/coresight/coresight-core.c b/drivers/hwtracing/coresight/coresight-core.c > index 6cb55c3f41d5..b56826d9a2b3 100644 > --- a/drivers/hwtracing/coresight/coresight-core.c > +++ b/drivers/hwtracing/coresight/coresight-core.c > @@ -1117,7 +1117,8 @@ static int coresight_validate_source(struct coresight_device *csdev, > } > > if (subtype != CORESIGHT_DEV_SUBTYPE_SOURCE_PROC && > - subtype != CORESIGHT_DEV_SUBTYPE_SOURCE_SOFTWARE) { > + subtype != CORESIGHT_DEV_SUBTYPE_SOURCE_SOFTWARE && > + subtype != CORESIGHT_DEV_SUBTYPE_SOURCE_SYS) { > dev_err(&csdev->dev, "wrong device subtype in %s\n", function); > return -EINVAL; > } > diff --git a/drivers/hwtracing/coresight/coresight-tpdm.c b/drivers/hwtracing/coresight/coresight-tpdm.c > new file mode 100644 > index 000000000000..51b8b17e6a80 > --- /dev/null > +++ b/drivers/hwtracing/coresight/coresight-tpdm.c > @@ -0,0 +1,159 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved. > + */ > + > +#include <linux/amba/bus.h> > +#include <linux/bitmap.h> > +#include <linux/coresight.h> > +#include <linux/coresight-pmu.h> > +#include <linux/device.h> > +#include <linux/err.h> > +#include <linux/fs.h> > +#include <linux/io.h> > +#include <linux/kernel.h> > +#include <linux/module.h> > +#include <linux/of.h> > + > +#include "coresight-priv.h" > +#include "coresight-tpdm.h" > + > +DEFINE_CORESIGHT_DEVLIST(tpdm_devs, "tpdm"); > + > +/* TPDM enable operations */ > +static int tpdm_enable(struct coresight_device *csdev, > + struct perf_event *event, u32 mode) > +{ > + struct tpdm_drvdata *drvdata = dev_get_drvdata(csdev->dev.parent); > + > + mutex_lock(&drvdata->lock); > + if (drvdata->enable) { > + mutex_unlock(&drvdata->lock); > + return -EBUSY; > + } > + > + drvdata->enable = true; > + mutex_unlock(&drvdata->lock); > + > + dev_info(drvdata->dev, "TPDM tracing enabled\n"); > + return 0; > +} > + > +/* TPDM disable operations */ > +static void tpdm_disable(struct coresight_device *csdev, > + struct perf_event *event) > +{ > + struct tpdm_drvdata *drvdata = dev_get_drvdata(csdev->dev.parent); > + > + mutex_lock(&drvdata->lock); > + if (!drvdata->enable) { > + mutex_unlock(&drvdata->lock); > + return; > + } > + > + drvdata->enable = false; > + mutex_unlock(&drvdata->lock); > + > + dev_info(drvdata->dev, "TPDM tracing disabled\n"); > +} > + > +static int tpdm_trace_id(struct coresight_device *csdev) > +{ > + struct tpdm_drvdata *drvdata = dev_get_drvdata(csdev->dev.parent); > + > + return drvdata->traceid; > +} > + > +static const struct coresight_ops_source tpdm_source_ops = { > + .trace_id = tpdm_trace_id, > + .enable = tpdm_enable, > + .disable = tpdm_disable, > +}; > + > +static const struct coresight_ops tpdm_cs_ops = { > + .source_ops = &tpdm_source_ops, > +}; > + > +static void tpdm_init_default_data(struct tpdm_drvdata *drvdata) > +{ > + drvdata->traceid = coresight_get_system_trace_id(); > +} > + > +static int tpdm_probe(struct amba_device *adev, const struct amba_id *id) > +{ > + struct device *dev = &adev->dev; > + struct coresight_platform_data *pdata; > + struct tpdm_drvdata *drvdata; > + struct coresight_desc desc = { 0 }; > + > + pdata = coresight_get_platform_data(dev); > + if (IS_ERR(pdata)) > + return PTR_ERR(pdata); > + adev->dev.platform_data = pdata; > + > + /* driver data*/ > + drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL); > + if (!drvdata) > + return -ENOMEM; > + drvdata->dev = &adev->dev; > + dev_set_drvdata(dev, drvdata); > + > + drvdata->base = devm_ioremap_resource(dev, &adev->res); > + if (!drvdata->base) > + return -ENOMEM; > + > + mutex_init(&drvdata->lock); > + tpdm_init_default_data(drvdata); > + > + /* Set up coresight component description */ > + desc.name = coresight_alloc_device_name(&tpdm_devs, dev); > + if (!desc.name) > + return -ENOMEM; > + desc.type = CORESIGHT_DEV_TYPE_SOURCE; > + desc.subtype.source_subtype = CORESIGHT_DEV_SUBTYPE_SOURCE_SYS; > + desc.ops = &tpdm_cs_ops; > + desc.pdata = adev->dev.platform_data; > + desc.dev = &adev->dev; > + drvdata->csdev = coresight_register(&desc); > + if (IS_ERR(drvdata->csdev)) > + return PTR_ERR(drvdata->csdev); > + > + /* Decrease pm refcount when probe is done.*/ > + pm_runtime_put(&adev->dev); > + > + return 0; > +} > + > +static void __exit tpdm_remove(struct amba_device *adev) > +{ > + struct tpdm_drvdata *drvdata = dev_get_drvdata(&adev->dev); > + > + coresight_unregister(drvdata->csdev); > +} > + > +/* > + * Different TPDM has different periph id. > + * The difference is 0-7 bits' value. So ignore 0-7 bits. > + */ > +static struct amba_id tpdm_ids[] = { > + { > + .id = 0x000f0e00, > + .mask = 0x000fff00, > + }, > + { 0, 0}, > +}; > + > +static struct amba_driver tpdm_driver = { > + .drv = { > + .name = "coresight-tpdm", > + .owner = THIS_MODULE, > + .suppress_bind_attrs = true, > + }, > + .probe = tpdm_probe, > + .id_table = tpdm_ids, > +}; > + > +module_amba_driver(tpdm_driver); > + > +MODULE_LICENSE("GPL v2"); > +MODULE_DESCRIPTION("Trace, Profiling & Diagnostic Monitor driver"); > diff --git a/drivers/hwtracing/coresight/coresight-tpdm.h b/drivers/hwtracing/coresight/coresight-tpdm.h > new file mode 100644 > index 000000000000..2effbabf349b > --- /dev/null > +++ b/drivers/hwtracing/coresight/coresight-tpdm.h > @@ -0,0 +1,28 @@ > +/* SPDX-License-Identifier: GPL-2.0 */ > +/* > + * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved. > + */ > + > +#ifndef _CORESIGHT_CORESIGHT_TPDM_H > +#define _CORESIGHT_CORESIGHT_TPDM_H > + > +/** > + * struct tpdm_drvdata - specifics associated to an TPDM component > + * @base: memory mapped base address for this component. > + * @dev: The device entity associated to this component. > + * @csdev: component vitals needed by the framework. > + * @lock: lock for the enable value. > + * @enable: enable status of the component. > + * @traceid: value of the current ID for this component. > + */ > + > +struct tpdm_drvdata { > + void __iomem *base; > + struct device *dev; > + struct coresight_device *csdev; > + struct mutex lock; > + bool enable; > + int traceid; TraceID appears to have no purpose in the hardware and does not apparently drive onto the ATB - ATID signals from this device. This should be dropped and the core code fixed to allow for sources that have no trace id. Regards Mike > +}; > + > +#endif /* _CORESIGHT_CORESIGHT_TPDM_H */ > diff --git a/include/linux/coresight.h b/include/linux/coresight.h > index 93a2922b7653..e48d463be63b 100644 > --- a/include/linux/coresight.h > +++ b/include/linux/coresight.h > @@ -65,6 +65,7 @@ enum coresight_dev_subtype_source { > CORESIGHT_DEV_SUBTYPE_SOURCE_PROC, > CORESIGHT_DEV_SUBTYPE_SOURCE_BUS, > CORESIGHT_DEV_SUBTYPE_SOURCE_SOFTWARE, > + CORESIGHT_DEV_SUBTYPE_SOURCE_SYS, > }; > > enum coresight_dev_subtype_helper { > -- > 2.17.1 > -- Mike Leach Principal Engineer, ARM Ltd. Manchester Design Centre. UK

3 years, 9 months

Re: [PATCH v3 04/10] dt-bindings: arm: Adds CoreSight TPDM hardware definitions

by Mike Leach

Hi, On Wed, 9 Feb 2022 at 10:57, Mao Jinlong <quic_jinlmao(a)quicinc.com> wrote: > > Adds new coresight-tpdm.yaml file describing the bindings required > to define tpdm in the device trees. > > Signed-off-by: Tao Zhang <quic_taozha(a)quicinc.com> > Signed-off-by: Mao Jinlong <quic_jinlmao(a)quicinc.com> > --- > .../bindings/arm/coresight-tpdm.yaml | 81 +++++++++++++++++++ > .../devicetree/bindings/arm/coresight.txt | 7 ++ > MAINTAINERS | 1 + > 3 files changed, 89 insertions(+) > create mode 100644 Documentation/devicetree/bindings/arm/coresight-tpdm.yaml > > diff --git a/Documentation/devicetree/bindings/arm/coresight-tpdm.yaml b/Documentation/devicetree/bindings/arm/coresight-tpdm.yaml > new file mode 100644 > index 000000000000..7ebc1bf7abc2 > --- /dev/null > +++ b/Documentation/devicetree/bindings/arm/coresight-tpdm.yaml > @@ -0,0 +1,81 @@ > +# SPDX-License-Identifier: GPL-2.0-only or BSD-2-Clause > +# Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved. > +%YAML 1.2 > +--- > +$id: http://devicetree.org/schemas/arm/coresight-tpdm.yaml# > +$schema: http://devicetree.org/meta-schemas/core.yaml# > + > +title: Trace, Profiling and Diagnostics Monitor - TPDM > + > +description: | > + The TPDM or Monitor serves as data collection component for various dataset > + types specified in the QPMDA spec. It covers Implementation defined ((ImplDef), > + Basic Counts (BC), Tenure Counts (TC), Continuous Multi-Bit (CMB), and Discrete > + Single Bit (DSB). It performs data collection in the data producing clock > + domain and transfers it to the data collection time domain, generally ATB > + clock domain. > + > + The primary use case of the TPDM is to collect data from different data > + sources and send it to a TPDA for packetization, timestamping, and funneling. > + > +maintainers: > + - Suzuki K Poulose <suzuki.poulose(a)arm.com> > + - Mathieu Poirier <mathieu.poirier(a)linaro.org> > + > +properties: > + $nodename: > + pattern: "^tpdm(@[0-9a-f]+)$" > + compatible: > + items: > + - const: qcom,coresight-tpdm > + - const: arm,primecell > + > + reg: > + maxItems: 1 > + > + clocks: > + maxItems: 1 > + > + clock-names: > + items: > + - const: apb_pclk > + > + out-ports: > + description: | > + Output connections from the TPDM to legacy CoreSight trace bus. > + $ref: /schemas/graph.yaml#/properties/ports > + properties: > + port: > + description: Output connection from the TPDM to legacy CoreSight > + Trace bus. > + $ref: /schemas/graph.yaml#/properties/port > + As we discussed in patch 0/10 of this set - the TDPM actually sends data to TDPA - not to coresight. These descriptions should match that. > +required: > + - compatible > + - reg > + - clocks > + - clock-names > + > +additionalProperties: false > + > +examples: > + # minimum TPDM definition. > + - | > + tpdm@6980000 { > + compatible = "qcom,coresight-tpdm", "arm,primecell"; > + reg = <0x6980000 0x1000>; > + > + clocks = <&aoss_qmp>; > + clock-names = "apb_pclk"; > + > + out-ports { > + port { > + tpdm_turing_out_funnel_turing: endpoint { > + remote-endpoint = > + <&funnel_turing_in_tpdm_turing>; > + }; > + }; > + }; > + }; > + example looks OK, but I would like to see something of a _tdpa in the nameing as that is the funnel / link the tdpm sends data to. Regards Mike > +... > diff --git a/Documentation/devicetree/bindings/arm/coresight.txt b/Documentation/devicetree/bindings/arm/coresight.txt > index c68d93a35b6c..f7ce8af48574 100644 > --- a/Documentation/devicetree/bindings/arm/coresight.txt > +++ b/Documentation/devicetree/bindings/arm/coresight.txt > @@ -52,6 +52,10 @@ its hardware characteristcs. > "arm,coresight-cti", "arm,primecell"; > See coresight-cti.yaml for full CTI definitions. > > + - Trace, Profiling and Diagnostics Monitor (TPDM): > + "qcom,coresight-tpdm", "arm,primecell"; > + See coresight-tpdm.yaml for full TPDM definitions. > + > * reg: physical base address and length of the register > set(s) of the component. > > @@ -82,6 +86,9 @@ its hardware characteristcs. > * Required properties for Coresight Cross Trigger Interface (CTI) > See coresight-cti.yaml for full CTI definitions. > > +* Required properties for Trace, Profiling and Diagnostics Monitor (TPDM) > + See coresight-tpdm.yaml for full TPDM definitions. > + > * Required properties for devices that don't show up on the AMBA bus, such as > non-configurable replicators and non-configurable funnels: > > diff --git a/MAINTAINERS b/MAINTAINERS > index ea3e6c914384..434bbe98d22b 100644 > --- a/MAINTAINERS > +++ b/MAINTAINERS > @@ -1906,6 +1906,7 @@ T: git git://git.kernel.org/pub/scm/linux/kernel/git/coresight/linux.git > F: Documentation/ABI/testing/sysfs-bus-coresight-devices-* > F: Documentation/devicetree/bindings/arm/coresight-cpu-debug.txt > F: Documentation/devicetree/bindings/arm/coresight-cti.yaml > +F: Documentation/devicetree/bindings/arm/coresight-tpdm.yaml > F: Documentation/devicetree/bindings/arm/coresight.txt > F: Documentation/devicetree/bindings/arm/ete.yaml > F: Documentation/devicetree/bindings/arm/trbe.yaml > -- > 2.17.1 > -- Mike Leach Principal Engineer, ARM Ltd. Manchester Design Centre. UK

3 years, 9 months

Re: [PATCH v3 02/10] coresight: Use bitmap to assign trace id to the sources

by Mike Leach

Hi, On Wed, 9 Feb 2022 at 10:57, Mao Jinlong <quic_jinlmao(a)quicinc.com> wrote: > > Except from STM and ETM/ETE, there could be other sources. Each > source needs a unique trace id. Define a bitmap for the trace ids. > The position of each bit represents trace id of the source. > > Signed-off-by: Mao Jinlong <quic_jinlmao(a)quicinc.com> > --- > drivers/hwtracing/coresight/coresight-core.c | 48 ++++++++++++++++++++ > include/linux/coresight-pmu.h | 11 +++++ > 2 files changed, 59 insertions(+) > > diff --git a/drivers/hwtracing/coresight/coresight-core.c b/drivers/hwtracing/coresight/coresight-core.c > index a90097f88425..6cb55c3f41d5 100644 > --- a/drivers/hwtracing/coresight/coresight-core.c > +++ b/drivers/hwtracing/coresight/coresight-core.c > @@ -16,6 +16,7 @@ > #include <linux/mutex.h> > #include <linux/clk.h> > #include <linux/coresight.h> > +#include <linux/coresight-pmu.h> see my comment below about using coresigh-priv.h > #include <linux/of_platform.h> > #include <linux/delay.h> > #include <linux/pm_runtime.h> > @@ -25,8 +26,11 @@ > #include "coresight-syscfg.h" > > static DEFINE_MUTEX(coresight_mutex); > +static DEFINE_MUTEX(coresight_id_mutex); > static DEFINE_PER_CPU(struct coresight_device *, csdev_sink); > > +static DECLARE_BITMAP(coresight_trace_id, CORESIGHT_TRACE_ID_NUM); > + > /* > * Use IDR to map the hash length of the source's device name > * to the pointer of path for the source > @@ -51,6 +55,48 @@ struct coresight_node { > const u32 coresight_barrier_pkt[4] = {0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff}; > EXPORT_SYMBOL_GPL(coresight_barrier_pkt); > > +/* Init the coresight_trace_id bit map. */ > +static void coresight_init_trace_id(void) > +{ > + int i; > + > + /* Trace id 0 is invalid. */ > + set_bit(CORESIGHT_TRACE_ID_0, coresight_trace_id); > + /* Trace id 1 is fixed for STM. */ > + set_bit(CORESIGHT_TRACE_ID_1, coresight_trace_id); > + /* Trace id from 112 to 127 are reserved. */ > + for (i = CORESIGHT_TRACE_ID_112; i <= CORESIGHT_TRACE_ID_127; i++) > + set_bit(i, coresight_trace_id); > + /* Skip the trace ids of ETM/ETE. */ > + for (i = 0; i <= cpumask_last(cpu_possible_mask); i++) > + set_bit(coresight_get_trace_id(i), coresight_trace_id); > + > +} > + > +/* > + * Return the first zero bit position of bitmap coresight_trace_id > + * as source's trace id. > + * > + */ > +int coresight_get_system_trace_id(void) > +{ > + int id; > + > + mutex_lock(&coresight_id_mutex); > + id = find_first_zero_bit(coresight_trace_id, CORESIGHT_TRACE_ID_NUM); > + /* If no zero bit is found, return error value. */ > + if (id == CORESIGHT_TRACE_ID_NUM) { > + mutex_unlock(&coresight_id_mutex); > + return -EINVAL; > + } > + > + set_bit(id, coresight_trace_id); > + mutex_unlock(&coresight_id_mutex); > + > + return id; > +} > +EXPORT_SYMBOL(coresight_get_system_trace_id); > + > static const struct cti_assoc_op *cti_assoc_ops; > > void coresight_set_cti_ops(const struct cti_assoc_op *cti_op) > @@ -1750,6 +1796,8 @@ static int __init coresight_init(void) > return 0; > > etm_perf_exit(); > + > + coresight_init_trace_id(); > exit_bus_unregister: > bus_unregister(&coresight_bustype); > return ret; > diff --git a/include/linux/coresight-pmu.h b/include/linux/coresight-pmu.h > index 4ac5c081af93..1e2c5ca4c6e6 100644 > --- a/include/linux/coresight-pmu.h > +++ b/include/linux/coresight-pmu.h > @@ -32,6 +32,14 @@ > #define ETM4_CFG_BIT_RETSTK 12 > #define ETM4_CFG_BIT_VMID_OPT 15 > The following additional defines and function should appear in coresight-priv.h The coresight-pmu.h file contains data for the interface between the drivers and perf. > +/* Coresight component supports 7 bits trace id. */ additional comment here to explain that 0, 0x70- 0x7F IDs are reserved by the architecture, 1 is default for STM > +#define CORESIGHT_TRACE_ID_NUM 128 > + > +#define CORESIGHT_TRACE_ID_0 0 > +#define CORESIGHT_TRACE_ID_1 1 > +#define CORESIGHT_TRACE_ID_112 112 > +#define CORESIGHT_TRACE_ID_127 127 > + can we have these names a little more descriptive - e.g. CORESIGHT_TRACE_ID_0_RES 0 CORESIGHT_TRACE_ID_STM 1 CORESIGHT_TRACE_ID_RANGE_LO_RES 0x70 CORESIGHT_TRACE_ID_RANGE_HI_RES 0x7F Additionally - now we are declaring a #define for the STM ID - it would be better to use that in coresight-stm.c stm_init_default_data() to set the trace id for the STM. Regards Mike > static inline int coresight_get_trace_id(int cpu) > { > /* > @@ -43,4 +51,7 @@ static inline int coresight_get_trace_id(int cpu) > return (CORESIGHT_ETM_PMU_SEED + (cpu * 2)); > } > > +/* Get the trace id for the sources except from STM, ETM/ETE. */ > +extern int coresight_get_system_trace_id(void); > + > #endif > -- > 2.17.1 > -- Mike Leach Principal Engineer, ARM Ltd. Manchester Design Centre. UK

3 years, 9 months

Re: [PATCH v3 00/10] Coresight: Add support for TPDM and TPDA

by Mike Leach

Hi, On Fri, 11 Feb 2022 at 04:17, Jinlong Mao <quic_jinlmao(a)quicinc.com> wrote: > > > On 2/10/2022 6:30 PM, Mike Leach wrote: > > Hello Mao, > > > > I have looked through this set and have a few general questions. > > > > My understanding based on the information in the code is that the TPDM > > devices will always send data to the TPDA device, the TPDM is not > > capable of directly driving the ATB itself? > > The TPDA device will then packetize the inputs and output these to > > the ATB over the normal CoreSight architecture to a standard ETR / ETF > > for collection. > > > > Looking at the TPDM driver - it is assigned a trace ID but never > > actually uses it in the hardware. My assumption here is that this is > > used purely to satisfy the requirement that the CoreSight core has > > that all sources have a unique trace id? > > > > For the TPDA driver you assign an ATID as an attribute in device tree, > > and then program this into the devices control register. > > > > The trace IDs in ETM / ETE / STM, are programmed into the hardware and > > these values drive the ATID value on the trace bus. So assigning an > > ATID value to the TPDA driver through device tree will lead to clashes > > with the assignment of trace IDs in the other driver software. > > > > The topology here appears to me that you have multiple "data source" > > devices TPDM, supplying a TPDA - which is the real CoreSight "trace > > source" from the viewpoint of the trace bus and CoreSight > > infrastructure. > > To get this to work in the current CoreSight driver stack, you have > > assigned the TPDM as a source type, and the TPDA as a link to ensure > > that when a TPDM is started, all the components on the path to the > > sink are activated. > > This is fine. > > > > If my assumptions above are all accurate I suggest the following improvements > > > > For TPDA drop the device tree assignment of ATID and instead use the > > coresight_get_system_trace_id() function you introduce in the 2nd > > patch in this set. > > > > For TPDM you have assigned a unique source sub-type > > CORESIGHT_DEV_SUBTYPE_SOURCE_SYS.- this could become > > CORESIGHT_DEV_SUBTYPE_SOURCE_DATA_ONLY. If the trace ID assigned to > > this device is only to satisfy the unique ID requirement and is not > > used elsewhere, then the sub type could become > > CORESIGHT_DEV_SUBTYPE_SOURCE_DATA_ONLY. We can agree that this sub > > type does not need a unique ID and acts as none ATB a source for > > another component, The core code can be altered to drop the > > requirement for this sub-type and trace ID can be dropped for this > > component. > > > > You should be aware that we are in the process of re-designed how > > trace IDs are allocated. The current mechanism does not scale for > > large multi-core systems (currently broken for any system > 46 cores), > > and as you have discovered there is a need for additional allocation > > of IDs. Also the ETE / TRBE combination does not need a trace ID. A > > dynamic allocation system is being proposed. > > > > Regards > > > > Mike > > > Hi Mike, > > Your assumptions above are all correct. > TPDMs connect to the same TPDA will share the atid of the TPDA. > We have a PC tool to parse the TPDM trace data. It needs the fixed atid > for each TPDA to identify the data. > So we configure the atid for TPDA in device tree with fixed ids. > I will discuss with internal tool team to see if TPDA's id can become > dynamic when parse the data. > I understand that it is essential to know the ID when extracting the trace from a sink with a coresight frame fomatter. However, tools such as perf, achieve this by saving the metadata for the session. If you are programming the trace ID, then it should be possible to read back the same register to get the required information for your trace session. If this is genuinely not possible for your system, then treat as a fixed ID as I describe below. > Apart from the TPDA's atid, we also have some other sources with fixed > id in HW on our internal device. > Do you have any suggestion to how to allocate the IDs for the source > with fixed id in HW ? > > Fixed IDs in hardware - assuming that these devices do write directly onto the trace bus using the ATID, should be accommodated by reserving the fixed ID in any scheme used to allocate IDs, In your case this would be to set a bit at the correct area in the bitfield. As mentioned we are working on an updated system to dynamically allocate IDs for CoreSight sources - a reservation scheme for fixed ID devices will be provided . Regards Mike >< Thanks > Jinlong Mao > > > > > > > > On Wed, 9 Feb 2022 at 10:57, Mao Jinlong <quic_jinlmao(a)quicinc.com> wrote: > >> This series adds support for the trace performance monitoring and > >> diagnostics hardware (TPDM and TPDA). It is composed of two major > >> elements. > >> a) Changes for original coresight framework to support for TPDM and TPDA. > >> b) Add driver code for TPDM and TPDA. > >> > >> Introduction of changes for original coresight framework > >> Support TPDM as new coresight source. > >> Since only STM and ETM are supported as coresight source originally. > >> TPDM is a newly added coresight source. We need to change > >> the original way of saving coresight path to support more types source > >> for coresight driver. > >> The following patch is to add support more coresight sources. > >> Use IDR to maintain all the enabled sources' paths. > >> coresight: Use bitmap to assign trace id to the sources > >> > >> Introduction of TPDM and TPDA > >> TPDM - The trace performance monitoring and diagnostics monitor or TPDM in > >> short serves as data collection component for various dataset types > >> specified in the QPMDA(Qualcomm performance monitoring and diagnostics > >> architecture) spec. The primary use case of the TPDM is to collect data > >> from different data sources and send it to a TPDA for packetization, > >> timestamping and funneling. > >> Coresight: Add coresight TPDM source driver > >> dt-bindings: arm: Adds CoreSight TPDM hardware definitions > >> coresight-tpdm: Add DSB dataset support > >> coresight-tpdm: Add integration test support > >> docs: sysfs: coresight: Add sysfs ABI documentation for TPDM > >> > >> TPDA - The trace performance monitoring and diagnostics aggregator or > >> TPDA in short serves as an arbitration and packetization engine for the > >> performance monitoring and diagnostics network as specified in the QPMDA > >> (Qualcomm performance monitoring and diagnostics architecture) > >> specification. The primary use case of the TPDA is to provide > >> packetization, funneling and timestamping of Monitor data as specified > >> in the QPMDA specification. > >> The following patch is to add driver for TPDA. > >> Coresight: Add TPDA link driver > >> dt-bindings: arm: Adds CoreSight TPDA hardware definitions > >> > >> The last patch of this series is a device tree modification, which add > >> the TPDM and TPDA configuration to device tree for validating. > >> ARM: dts: msm: Add coresight components for SM8250 > >> > >> Once this series patches are applied properly, the tpdm and tpda nodes > >> should be observed at the coresight path /sys/bus/coresight/devices > >> e.g. > >> /sys/bus/coresight/devices # ls -l | grep tpd > >> tpda0 -> ../../../devices/platform/soc(a)0/6004000.tpda/tpda0 > >> tpdm0 -> ../../../devices/platform/soc(a)0/6c08000.mm.tpdm/tpdm0 > >> > >> We can use the commands are similar to the below to validate TPDMs. > >> Enable coresight sink first. > >> > >> echo 1 > /sys/bus/coresight/devices/tmc_etf0/enable_sink > >> echo 1 > /sys/bus/coresight/devices/tpdm0/enable_source > >> echo 1 > /sys/bus/coresight/devices/tpdm0/integration_test > >> echo 2 > /sys/bus/coresight/devices/tpdm0/integration_test > >> The test data will be collected in the coresight sink which is enabled. > >> If rwp register of the sink is keeping updating when do > >> integration_test (by cat tmc_etf0/mgmt/rwp), it means there is data > >> generated from TPDM to sink. > >> > >> Changes from V2: > >> 1. Use bitmap to assign the trace id. (Mathieu Poirier) > >> > >> Mao Jinlong (10): > >> Use IDR to maintain all the enabled sources' paths. > >> coresight: Use bitmap to assign trace id to the sources > >> Coresight: Add coresight TPDM source driver > >> dt-bindings: arm: Adds CoreSight TPDM hardware definitions > >> coresight-tpdm: Add DSB dataset support > >> coresight-tpdm: Add integration test support > >> docs: sysfs: coresight: Add sysfs ABI documentation for TPDM > >> Coresight: Add TPDA link driver > >> dt-bindings: arm: Adds CoreSight TPDA hardware definitions > >> ARM: dts: msm: Add coresight components for SM8250 > >> > >> .../testing/sysfs-bus-coresight-devices-tpdm | 6 + > >> .../bindings/arm/coresight-tpda.yaml | 129 ++++ > >> .../bindings/arm/coresight-tpdm.yaml | 81 ++ > >> .../devicetree/bindings/arm/coresight.txt | 7 + > >> MAINTAINERS | 1 + > >> .../arm64/boot/dts/qcom/sm8250-coresight.dtsi | 690 ++++++++++++++++++ > >> arch/arm64/boot/dts/qcom/sm8250.dtsi | 2 + > >> drivers/hwtracing/coresight/Kconfig | 33 + > >> drivers/hwtracing/coresight/Makefile | 2 + > >> drivers/hwtracing/coresight/coresight-core.c | 127 ++-- > >> drivers/hwtracing/coresight/coresight-tpda.c | 193 +++++ > >> drivers/hwtracing/coresight/coresight-tpda.h | 32 + > >> drivers/hwtracing/coresight/coresight-tpdm.c | 270 +++++++ > >> drivers/hwtracing/coresight/coresight-tpdm.h | 57 ++ > >> include/linux/coresight-pmu.h | 11 + > >> include/linux/coresight.h | 1 + > >> 16 files changed, 1592 insertions(+), 50 deletions(-) > >> create mode 100644 Documentation/ABI/testing/sysfs-bus-coresight-devices-tpdm > >> create mode 100644 Documentation/devicetree/bindings/arm/coresight-tpda.yaml > >> create mode 100644 Documentation/devicetree/bindings/arm/coresight-tpdm.yaml > >> create mode 100644 arch/arm64/boot/dts/qcom/sm8250-coresight.dtsi > >> create mode 100644 drivers/hwtracing/coresight/coresight-tpda.c > >> create mode 100644 drivers/hwtracing/coresight/coresight-tpda.h > >> create mode 100644 drivers/hwtracing/coresight/coresight-tpdm.c > >> create mode 100644 drivers/hwtracing/coresight/coresight-tpdm.h > >> > >> -- > >> 2.17.1 > >> > > > > -- > > Mike Leach > > Principal Engineer, ARM Ltd. > > Manchester Design Centre. UK -- Mike Leach Principal Engineer, ARM Ltd. Manchester Design Centre. UK

3 years, 9 months

Re: [PATCH v3 00/10] Coresight: Add support for TPDM and TPDA

by Mike Leach

Hello Mao, I have looked through this set and have a few general questions. My understanding based on the information in the code is that the TPDM devices will always send data to the TPDA device, the TPDM is not capable of directly driving the ATB itself? The TPDA device will then packetize the inputs and output these to the ATB over the normal CoreSight architecture to a standard ETR / ETF for collection. Looking at the TPDM driver - it is assigned a trace ID but never actually uses it in the hardware. My assumption here is that this is used purely to satisfy the requirement that the CoreSight core has that all sources have a unique trace id? For the TPDA driver you assign an ATID as an attribute in device tree, and then program this into the devices control register. The trace IDs in ETM / ETE / STM, are programmed into the hardware and these values drive the ATID value on the trace bus. So assigning an ATID value to the TPDA driver through device tree will lead to clashes with the assignment of trace IDs in the other driver software. The topology here appears to me that you have multiple "data source" devices TPDM, supplying a TPDA - which is the real CoreSight "trace source" from the viewpoint of the trace bus and CoreSight infrastructure. To get this to work in the current CoreSight driver stack, you have assigned the TPDM as a source type, and the TPDA as a link to ensure that when a TPDM is started, all the components on the path to the sink are activated. This is fine. If my assumptions above are all accurate I suggest the following improvements For TPDA drop the device tree assignment of ATID and instead use the coresight_get_system_trace_id() function you introduce in the 2nd patch in this set. For TPDM you have assigned a unique source sub-type CORESIGHT_DEV_SUBTYPE_SOURCE_SYS.- this could become CORESIGHT_DEV_SUBTYPE_SOURCE_DATA_ONLY. If the trace ID assigned to this device is only to satisfy the unique ID requirement and is not used elsewhere, then the sub type could become CORESIGHT_DEV_SUBTYPE_SOURCE_DATA_ONLY. We can agree that this sub type does not need a unique ID and acts as none ATB a source for another component, The core code can be altered to drop the requirement for this sub-type and trace ID can be dropped for this component. You should be aware that we are in the process of re-designed how trace IDs are allocated. The current mechanism does not scale for large multi-core systems (currently broken for any system > 46 cores), and as you have discovered there is a need for additional allocation of IDs. Also the ETE / TRBE combination does not need a trace ID. A dynamic allocation system is being proposed. Regards Mike On Wed, 9 Feb 2022 at 10:57, Mao Jinlong <quic_jinlmao(a)quicinc.com> wrote: > > This series adds support for the trace performance monitoring and > diagnostics hardware (TPDM and TPDA). It is composed of two major > elements. > a) Changes for original coresight framework to support for TPDM and TPDA. > b) Add driver code for TPDM and TPDA. > > Introduction of changes for original coresight framework > Support TPDM as new coresight source. > Since only STM and ETM are supported as coresight source originally. > TPDM is a newly added coresight source. We need to change > the original way of saving coresight path to support more types source > for coresight driver. > The following patch is to add support more coresight sources. > Use IDR to maintain all the enabled sources' paths. > coresight: Use bitmap to assign trace id to the sources > > Introduction of TPDM and TPDA > TPDM - The trace performance monitoring and diagnostics monitor or TPDM in > short serves as data collection component for various dataset types > specified in the QPMDA(Qualcomm performance monitoring and diagnostics > architecture) spec. The primary use case of the TPDM is to collect data > from different data sources and send it to a TPDA for packetization, > timestamping and funneling. > Coresight: Add coresight TPDM source driver > dt-bindings: arm: Adds CoreSight TPDM hardware definitions > coresight-tpdm: Add DSB dataset support > coresight-tpdm: Add integration test support > docs: sysfs: coresight: Add sysfs ABI documentation for TPDM > > TPDA - The trace performance monitoring and diagnostics aggregator or > TPDA in short serves as an arbitration and packetization engine for the > performance monitoring and diagnostics network as specified in the QPMDA > (Qualcomm performance monitoring and diagnostics architecture) > specification. The primary use case of the TPDA is to provide > packetization, funneling and timestamping of Monitor data as specified > in the QPMDA specification. > The following patch is to add driver for TPDA. > Coresight: Add TPDA link driver > dt-bindings: arm: Adds CoreSight TPDA hardware definitions > > The last patch of this series is a device tree modification, which add > the TPDM and TPDA configuration to device tree for validating. > ARM: dts: msm: Add coresight components for SM8250 > > Once this series patches are applied properly, the tpdm and tpda nodes > should be observed at the coresight path /sys/bus/coresight/devices > e.g. > /sys/bus/coresight/devices # ls -l | grep tpd > tpda0 -> ../../../devices/platform/soc(a)0/6004000.tpda/tpda0 > tpdm0 -> ../../../devices/platform/soc(a)0/6c08000.mm.tpdm/tpdm0 > > We can use the commands are similar to the below to validate TPDMs. > Enable coresight sink first. > > echo 1 > /sys/bus/coresight/devices/tmc_etf0/enable_sink > echo 1 > /sys/bus/coresight/devices/tpdm0/enable_source > echo 1 > /sys/bus/coresight/devices/tpdm0/integration_test > echo 2 > /sys/bus/coresight/devices/tpdm0/integration_test > The test data will be collected in the coresight sink which is enabled. > If rwp register of the sink is keeping updating when do > integration_test (by cat tmc_etf0/mgmt/rwp), it means there is data > generated from TPDM to sink. > > Changes from V2: > 1. Use bitmap to assign the trace id. (Mathieu Poirier) > > Mao Jinlong (10): > Use IDR to maintain all the enabled sources' paths. > coresight: Use bitmap to assign trace id to the sources > Coresight: Add coresight TPDM source driver > dt-bindings: arm: Adds CoreSight TPDM hardware definitions > coresight-tpdm: Add DSB dataset support > coresight-tpdm: Add integration test support > docs: sysfs: coresight: Add sysfs ABI documentation for TPDM > Coresight: Add TPDA link driver > dt-bindings: arm: Adds CoreSight TPDA hardware definitions > ARM: dts: msm: Add coresight components for SM8250 > > .../testing/sysfs-bus-coresight-devices-tpdm | 6 + > .../bindings/arm/coresight-tpda.yaml | 129 ++++ > .../bindings/arm/coresight-tpdm.yaml | 81 ++ > .../devicetree/bindings/arm/coresight.txt | 7 + > MAINTAINERS | 1 + > .../arm64/boot/dts/qcom/sm8250-coresight.dtsi | 690 ++++++++++++++++++ > arch/arm64/boot/dts/qcom/sm8250.dtsi | 2 + > drivers/hwtracing/coresight/Kconfig | 33 + > drivers/hwtracing/coresight/Makefile | 2 + > drivers/hwtracing/coresight/coresight-core.c | 127 ++-- > drivers/hwtracing/coresight/coresight-tpda.c | 193 +++++ > drivers/hwtracing/coresight/coresight-tpda.h | 32 + > drivers/hwtracing/coresight/coresight-tpdm.c | 270 +++++++ > drivers/hwtracing/coresight/coresight-tpdm.h | 57 ++ > include/linux/coresight-pmu.h | 11 + > include/linux/coresight.h | 1 + > 16 files changed, 1592 insertions(+), 50 deletions(-) > create mode 100644 Documentation/ABI/testing/sysfs-bus-coresight-devices-tpdm > create mode 100644 Documentation/devicetree/bindings/arm/coresight-tpda.yaml > create mode 100644 Documentation/devicetree/bindings/arm/coresight-tpdm.yaml > create mode 100644 arch/arm64/boot/dts/qcom/sm8250-coresight.dtsi > create mode 100644 drivers/hwtracing/coresight/coresight-tpda.c > create mode 100644 drivers/hwtracing/coresight/coresight-tpda.h > create mode 100644 drivers/hwtracing/coresight/coresight-tpdm.c > create mode 100644 drivers/hwtracing/coresight/coresight-tpdm.h > > -- > 2.17.1 > -- Mike Leach Principal Engineer, ARM Ltd. Manchester Design Centre. UK

3 years, 9 months

Re: [PATCH v3 01/10] Use IDR to maintain all the enabled sources' paths.

by Mike Leach

Hi, On Wed, 9 Feb 2022 at 10:57, Mao Jinlong <quic_jinlmao(a)quicinc.com> wrote: > > Use hash length of the source's device name to map to the pointer > of the enabled path. Using IDR will be more efficient than using > the list. And there could be other sources except STM and CPU etms > in the new HWs. It is better to maintain all the paths together. > > Signed-off-by: Mao Jinlong <quic_jinlmao(a)quicinc.com> > --- > drivers/hwtracing/coresight/coresight-core.c | 76 +++++++------------- > 1 file changed, 27 insertions(+), 49 deletions(-) > > diff --git a/drivers/hwtracing/coresight/coresight-core.c b/drivers/hwtracing/coresight/coresight-core.c > index 88653d1c06a4..a90097f88425 100644 > --- a/drivers/hwtracing/coresight/coresight-core.c > +++ b/drivers/hwtracing/coresight/coresight-core.c > @@ -7,6 +7,7 @@ > #include <linux/init.h> > #include <linux/types.h> > #include <linux/device.h> > +#include <linux/idr.h> > #include <linux/io.h> > #include <linux/err.h> > #include <linux/export.h> > @@ -26,6 +27,12 @@ > static DEFINE_MUTEX(coresight_mutex); > static DEFINE_PER_CPU(struct coresight_device *, csdev_sink); > > +/* > + * Use IDR to map the hash length of the source's device name > + * to the pointer of path for the source > + */ > +static DEFINE_IDR(path_idr); > + > /** > * struct coresight_node - elements of a path, from source to sink > * @csdev: Address of an element. > @@ -36,20 +43,6 @@ struct coresight_node { > struct list_head link; > }; > > -/* > - * When operating Coresight drivers from the sysFS interface, only a single > - * path can exist from a tracer (associated to a CPU) to a sink. > - */ > -static DEFINE_PER_CPU(struct list_head *, tracer_path); > - > -/* > - * As of this writing only a single STM can be found in CS topologies. Since > - * there is no way to know if we'll ever see more and what kind of > - * configuration they will enact, for the time being only define a single path > - * for STM. > - */ > -static struct list_head *stm_path; > - > /* > * When losing synchronisation a new barrier packet needs to be inserted at the > * beginning of the data collected in a buffer. That way the decoder knows that > @@ -1088,10 +1081,11 @@ static int coresight_validate_source(struct coresight_device *csdev, > > int coresight_enable(struct coresight_device *csdev) > { > - int cpu, ret = 0; > + int ret = 0; > struct coresight_device *sink; > struct list_head *path; > enum coresight_dev_subtype_source subtype; > + u32 hash; > > subtype = csdev->subtype.source_subtype; > > @@ -1133,26 +1127,14 @@ int coresight_enable(struct coresight_device *csdev) > if (ret) > goto err_source; > > - switch (subtype) { > - case CORESIGHT_DEV_SUBTYPE_SOURCE_PROC: > - /* > - * When working from sysFS it is important to keep track > - * of the paths that were created so that they can be > - * undone in 'coresight_disable()'. Since there can only > - * be a single session per tracer (when working from sysFS) > - * a per-cpu variable will do just fine. > - */ > - cpu = source_ops(csdev)->cpu_id(csdev); > - per_cpu(tracer_path, cpu) = path; > - break; > - case CORESIGHT_DEV_SUBTYPE_SOURCE_SOFTWARE: > - stm_path = path; > - break; > - default: > - /* We can't be here */ > - break; > - } > - > + /* > + * Use the hash length of source's device name as ID Slightly confusing comment: hashlen_string creates a hash and a length for the string. We are using the hash here so comment should be: "Use the hash of source's device name as ID" > + * and map the ID to the pointer of the path. > + */ > + hash = hashlen_hash(hashlen_string(NULL, dev_name(&csdev->dev))); > + ret = idr_alloc_u32(&path_idr, path, &hash, hash, GFP_KERNEL); > + if (ret) > + goto err_source; > out: > mutex_unlock(&coresight_mutex); > return ret; > @@ -1168,8 +1150,9 @@ EXPORT_SYMBOL_GPL(coresight_enable); > > void coresight_disable(struct coresight_device *csdev) > { > - int cpu, ret; > + int ret; > struct list_head *path = NULL; > + u32 hash; > > mutex_lock(&coresight_mutex); > > @@ -1180,21 +1163,16 @@ void coresight_disable(struct coresight_device *csdev) > if (!csdev->enable || !coresight_disable_source(csdev)) > goto out; > > - switch (csdev->subtype.source_subtype) { > - case CORESIGHT_DEV_SUBTYPE_SOURCE_PROC: > - cpu = source_ops(csdev)->cpu_id(csdev); > - path = per_cpu(tracer_path, cpu); > - per_cpu(tracer_path, cpu) = NULL; > - break; > - case CORESIGHT_DEV_SUBTYPE_SOURCE_SOFTWARE: > - path = stm_path; > - stm_path = NULL; > - break; > - default: > - /* We can't be here */ > - break; > + hash = hashlen_hash(hashlen_string(NULL, dev_name(&csdev->dev))); > + /* Find the path by the hash length. */ same comment as above re hash / hash length > + path = idr_find(&path_idr, hash); > + if (path == NULL) { > + dev_info(&csdev->dev, "Path is not found for %s\n", > + dev_name(&csdev->dev)); pr_err() is used in other parts of this file to print error messages. > + return; this skips the mutex_unlock(). use goto out; > } > > + idr_remove(&path_idr, hash); > coresight_disable_path(path); > coresight_release_path(path); > > -- > 2.17.1 > Regards Mike -- Mike Leach Principal Engineer, ARM Ltd. Manchester Design Centre. UK

3 years, 9 months

[PATCH v2 00/15] Make ETM register accesses consistent with sysreg.h

by James Clark

Changes since v1: * Change base to "[PATCH 1/1] coresight: no-op refactor to make INSTP0 check more idiomatic" * Change 'name' to 'field' in REG_VAL() macro * Add comment for REG_VAL() macro While working on the branch broadcast change I found it difficult to search for usages of registers and fields because of the magic numbers. I also found it difficult to decide which style to make new code in because of the varying ones used. There was also a code review comment from Suzuki about replacing a magic number so I'm proposing to refactor as many as possible into the style used in sysreg.h which seems to be the new and most consistently used method. For example it was used in the SPE and TRBE drivers. This isn't an exhaustive refactor, but it does get all the basic accesses. There are a couple of odd other cases remaining, mainly in the ETM3x code. These can be found by searching for BMVAL. There is one compromise to ensure this is a complete no-op and has binary equivalence with the old version. I needed to keep two register accesses here, where something like etmidr0 & TRCIDR0_INSTP0_LOAD_STORE would be better: - if (BMVAL(etmidr0, 1, 1) && BMVAL(etmidr0, 2, 2)) - drvdata->instrp0 = true; - else - drvdata->instrp0 = false; - + drvdata->instrp0 = !!((REG_VAL(etmidr0, TRCIDR0_INSTP0) & 0b01) && + (REG_VAL(etmidr0, TRCIDR0_INSTP0) & 0b10)); I think this change fixes quite a few issues like: * Some registers aren't referred to by name but a different variable name. For example eventctrl1 in mode_store() where TRCEVENTCTL1R isn't mentioned in that function. * Some bits aren't referred to by the name in the manual, even in the comments. For example TRCCONFIGR.RS only occurs as /* bit[12], Return stack enable bit */. * Some bits in the same register are referred to either as magic numbers or the publicly exported config macros, neiher of which are consistent with any other register accesses. For example config->cfg |= BIT(11); config->cfg |= BIT(ETM4_CFG_BIT_CTXTID); * Some fields are already partially referred to in the sysfs.h style way: TRCVICTLR_EXLEVEL_... etc. But other fields in the same register are not * Some fields are magic numbers that are repeated many times in different functions. For example vinst_ctrl |= BIT(9) * Some fields were referred to by magic numbers, even when there were already existing #defines. For example ETMTECR1_INC_EXC * Another benefit is that the #defines could be automatically checked against the reference manual because the style is uniform. Testing ======= To test this I used gcc-11 which doesn't have a quirk about changing register widths in some cases (as in w -> x). I also used elf_diff which showed me exactly where I'd made a mistake when I did (https://github.com/noseglasses/elf_diff). But now that there is no difference, objdump and diff also work for validation. make CC=gcc-11 modules diff <(objdump -d drivers/hwtracing/coresight/coresight-etm4x.ko) <(objdump -d ../linux2/drivers/hwtracing/coresight/coresight-etm4x.ko) diff <(objdump -d drivers/hwtracing/coresight/coresight.ko) <(objdump -d ../linux2/drivers/hwtracing/coresight/coresight.ko) And for ETM3x (doesn't need gcc 11): make ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- modules diff <(objdump -d drivers/hwtracing/coresight/coresight-etm3x.ko) <(objdump -d ../linux2/drivers/hwtracing/coresight/coresight-etm3x.ko) When there are no differences, the diff output looks like this with only the filename listed: < drivers/hwtracing/coresight/coresight-etm4x.ko: file format elf64-littleaarch64 --- > ../linux2/drivers/hwtracing/coresight/coresight-etm4x.ko: file format elf64-littleaarch64 Applies to "[PATCH 1/1] coresight: no-op refactor to make INSTP0 check more idiomatic" James Clark (15): coresight: Make ETM4x TRCIDR0 register accesses consistent with sysreg.h coresight: Make ETM4x TRCIDR2 register accesses consistent with sysreg.h coresight: Make ETM4x TRCIDR3 register accesses consistent with sysreg.h coresight: Make ETM4x TRCIDR4 register accesses consistent with sysreg.h coresight: Make ETM4x TRCIDR5 register accesses consistent with sysreg.h coresight: Make ETM4x TRCCONFIGR register accesses consistent with sysreg.h coresight: Make ETM4x TRCEVENTCTL1R register accesses consistent with sysreg.h coresight: Make ETM4x TRCSTALLCTLR register accesses consistent with sysreg.h coresight: Make ETM4x TRCVICTLR register accesses consistent with sysreg.h coresight: Make ETM3x ETMTECR1 register accesses consistent with sysreg.h coresight: Make ETM4x TRCACATRn register accesses consistent with sysreg.h coresight: Make ETM4x TRCSSCCRn and TRCSSCSRn register accesses consistent with sysreg.h coresight: Make ETM4x TRCSSPCICRn register accesses consistent with sysreg.h coresight: Make ETM4x TRCBBCTLR register accesses consistent with sysreg.h coresight: Make ETM4x TRCRSCTLRn register accesses consistent with sysreg.h .../coresight/coresight-etm3x-core.c | 2 +- .../coresight/coresight-etm3x-sysfs.c | 2 +- .../coresight/coresight-etm4x-core.c | 134 +++++-------- .../coresight/coresight-etm4x-sysfs.c | 178 +++++++++--------- drivers/hwtracing/coresight/coresight-etm4x.h | 159 ++++++++++++++-- drivers/hwtracing/coresight/coresight-priv.h | 5 + 6 files changed, 286 insertions(+), 194 deletions(-) -- 2.28.0

3 years, 9 months

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

CoreSight