| Message ID | 20221025113242.58271-4-yangyicong@huawei.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Add TLP filter support and some fixes for HiSilicon PCIe PMU | expand |
On Tue, 25 Oct 2022 19:32:42 +0800 Yicong Yang <yangyicong@huawei.com> wrote: > From: Yicong Yang <yangyicong@hisilicon.com> > > The PMU support to filter the TLP when counting the bandwidth with below > options: > > - only count the TLP headers > - only count the TLP payloads > - count both TLP headers and payloads > > In the current driver it's default to count the TLP payloads only, which > will have an implicity side effects that on the traffic only have header > only TLPs, we'll get no data. > > Make this user configuration through "len_mode" parameter and make it > default to count both TLP headers and payloads when user not specified. > Also update the documentation for it. I suppose this filter option is useful. Default of "both" makes sense. A few minor suggestions inline. Jonathan > > Signed-off-by: Yicong Yang <yangyicong@hisilicon.com> > --- > .../admin-guide/perf/hisi-pcie-pmu.rst | 20 +++++++++++++++++++ > drivers/perf/hisilicon/hisi_pcie_pmu.c | 14 ++++++++++++- > 2 files changed, 33 insertions(+), 1 deletion(-) > > diff --git a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst > index bbe66480ff85..83a2ef11b1a0 100644 > --- a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst > +++ b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst > @@ -104,3 +104,23 @@ when TLP length < threshold. > Example usage of perf:: > > $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,thr_len=0x4,thr_mode=1/ sleep 5 > + > +4. TLP Length filter > +When counting bandwidth, the data can be composed of certain parts of TLP > +packets. You can specify it through "len_mode": > + > +- 2'b00: Reserved (Do not use this since the behaviour is undefined) > +- 2'b01: Bandwidth of TLP payloads > +- 2'b10: Bandwidth of TLP headers > +- 2'b11: Bandwidth of both TLP payloads and headers We could describe this as a bitmap, but i think you are right in thinking it is clearer to present it as 3 values as not always obvious that the bitmap is controlling sum of two different things. > + > +For example, "len_mode=2" means only counting the bandwidth of TLP headers > +and "len_mode=3" means the final bandwidth data is composed of both TLP > +headers and payloads. You need to carefully using this to avoid losing > +data. I'm not sure this warning makes sense. If user has set this filter then they don't want to measure the types they haven't picked. I would drop the warning and example. It is fairly obvious that no_counts == nothing to count. > For example you're likely to get no counts by "len_mode=1" if the TLPs > +on the traffic has no payload. Drop this example. > This config is optional, by default it'll > +be 2'b11. Anything with a default is inherently optional, so could we just say Default value if not specified is 2'b11. > + > +Example usage of perf:: > + > + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,len_mode=0x1/ sleep 5 > diff --git a/drivers/perf/hisilicon/hisi_pcie_pmu.c b/drivers/perf/hisilicon/hisi_pcie_pmu.c > index 071e63d9a9ac..6fee0b6e163b 100644 > --- a/drivers/perf/hisilicon/hisi_pcie_pmu.c > +++ b/drivers/perf/hisilicon/hisi_pcie_pmu.c > @@ -47,10 +47,14 @@ > #define HISI_PCIE_EVENT_M GENMASK_ULL(15, 0) > #define HISI_PCIE_THR_MODE_M GENMASK_ULL(27, 27) > #define HISI_PCIE_THR_M GENMASK_ULL(31, 28) > +#define HISI_PCIE_LEN_M GENMASK_ULL(35, 34) > #define HISI_PCIE_TARGET_M GENMASK_ULL(52, 36) > #define HISI_PCIE_TRIG_MODE_M GENMASK_ULL(53, 53) > #define HISI_PCIE_TRIG_M GENMASK_ULL(59, 56) > > +/* Default config of TLP length mode, will count both TLP headers and payloads */ > +#define HISI_PCIE_LEN_M_DEFAULT 3ULL > + > #define HISI_PCIE_MAX_COUNTERS 8 > #define HISI_PCIE_REG_STEP 8 > #define HISI_PCIE_THR_MAX_VAL 10 > @@ -91,6 +95,7 @@ HISI_PCIE_PMU_FILTER_ATTR(thr_len, config1, 3, 0); > HISI_PCIE_PMU_FILTER_ATTR(thr_mode, config1, 4, 4); > HISI_PCIE_PMU_FILTER_ATTR(trig_len, config1, 8, 5); > HISI_PCIE_PMU_FILTER_ATTR(trig_mode, config1, 9, 9); > +HISI_PCIE_PMU_FILTER_ATTR(len_mode, config1, 11, 10); > HISI_PCIE_PMU_FILTER_ATTR(port, config2, 15, 0); > HISI_PCIE_PMU_FILTER_ATTR(bdf, config2, 31, 16); > > @@ -215,8 +220,8 @@ static void hisi_pcie_pmu_config_filter(struct perf_event *event) > { > struct hisi_pcie_pmu *pcie_pmu = to_pcie_pmu(event->pmu); > struct hw_perf_event *hwc = &event->hw; > + u64 port, trig_len, thr_len, len_mode; > u64 reg = HISI_PCIE_INIT_SET; > - u64 port, trig_len, thr_len; > > /* Config HISI_PCIE_EVENT_CTRL according to event. */ > reg |= FIELD_PREP(HISI_PCIE_EVENT_M, hisi_pcie_get_real_event(event)); > @@ -245,6 +250,12 @@ static void hisi_pcie_pmu_config_filter(struct perf_event *event) > reg |= HISI_PCIE_THR_EN; > } > > + len_mode = hisi_pcie_get_len_mode(event); > + if (len_mode) > + reg |= FIELD_PREP(HISI_PCIE_LEN_M, len_mode); > + else > + reg |= FIELD_PREP(HISI_PCIE_LEN_M, HISI_PCIE_LEN_M_DEFAULT); > + > hisi_pcie_pmu_writeq(pcie_pmu, HISI_PCIE_EVENT_CTRL, hwc->idx, reg); > } > > @@ -711,6 +722,7 @@ static struct attribute *hisi_pcie_pmu_format_attr[] = { > HISI_PCIE_PMU_FORMAT_ATTR(thr_mode, "config1:4"), > HISI_PCIE_PMU_FORMAT_ATTR(trig_len, "config1:5-8"), > HISI_PCIE_PMU_FORMAT_ATTR(trig_mode, "config1:9"), > + HISI_PCIE_PMU_FORMAT_ATTR(len_mode, "config1:10-11"), > HISI_PCIE_PMU_FORMAT_ATTR(port, "config2:0-15"), > HISI_PCIE_PMU_FORMAT_ATTR(bdf, "config2:16-31"), > NULL
On 2022/11/10 1:09, Jonathan Cameron wrote: > On Tue, 25 Oct 2022 19:32:42 +0800 > Yicong Yang <yangyicong@huawei.com> wrote: > >> From: Yicong Yang <yangyicong@hisilicon.com> >> >> The PMU support to filter the TLP when counting the bandwidth with below >> options: >> >> - only count the TLP headers >> - only count the TLP payloads >> - count both TLP headers and payloads >> >> In the current driver it's default to count the TLP payloads only, which >> will have an implicity side effects that on the traffic only have header >> only TLPs, we'll get no data. >> >> Make this user configuration through "len_mode" parameter and make it >> default to count both TLP headers and payloads when user not specified. >> Also update the documentation for it. > > I suppose this filter option is useful. Default of "both" makes sense. > > A few minor suggestions inline. > > Jonathan > >> >> Signed-off-by: Yicong Yang <yangyicong@hisilicon.com> >> --- >> .../admin-guide/perf/hisi-pcie-pmu.rst | 20 +++++++++++++++++++ >> drivers/perf/hisilicon/hisi_pcie_pmu.c | 14 ++++++++++++- >> 2 files changed, 33 insertions(+), 1 deletion(-) >> >> diff --git a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst >> index bbe66480ff85..83a2ef11b1a0 100644 >> --- a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst >> +++ b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst >> @@ -104,3 +104,23 @@ when TLP length < threshold. >> Example usage of perf:: >> >> $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,thr_len=0x4,thr_mode=1/ sleep 5 >> + >> +4. TLP Length filter >> +When counting bandwidth, the data can be composed of certain parts of TLP >> +packets. You can specify it through "len_mode": >> + >> +- 2'b00: Reserved (Do not use this since the behaviour is undefined) >> +- 2'b01: Bandwidth of TLP payloads >> +- 2'b10: Bandwidth of TLP headers >> +- 2'b11: Bandwidth of both TLP payloads and headers > > We could describe this as a bitmap, but i think you are right in thinking > it is clearer to present it as 3 values as not always obvious that the bitmap > is controlling sum of two different things. > I just to keep the style consistence with the existing doc and yes I also think it's clearer. >> + >> +For example, "len_mode=2" means only counting the bandwidth of TLP headers >> +and "len_mode=3" means the final bandwidth data is composed of both TLP >> +headers and payloads. You need to carefully using this to avoid losing >> +data. > > I'm not sure this warning makes sense. If user has set this filter then > they don't want to measure the types they haven't picked. I would drop > the warning and example. It is fairly obvious that no_counts == nothing > to count. > Just hope users will not make it "len_mode=0" as it's reserved. Will drop it. >> For example you're likely to get no counts by "len_mode=1" if the TLPs >> +on the traffic has no payload. > Drop this example. > ok. >> This config is optional, by default it'll >> +be 2'b11. > > Anything with a default is inherently optional, so could we just say > Default value if not specified is 2'b11. > ok, it'll be better. Will update the doc. Thanks for the comments! Regards. >> + >> +Example usage of perf:: >> + >> + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,len_mode=0x1/ sleep 5 >> diff --git a/drivers/perf/hisilicon/hisi_pcie_pmu.c b/drivers/perf/hisilicon/hisi_pcie_pmu.c >> index 071e63d9a9ac..6fee0b6e163b 100644 >> --- a/drivers/perf/hisilicon/hisi_pcie_pmu.c >> +++ b/drivers/perf/hisilicon/hisi_pcie_pmu.c >> @@ -47,10 +47,14 @@ >> #define HISI_PCIE_EVENT_M GENMASK_ULL(15, 0) >> #define HISI_PCIE_THR_MODE_M GENMASK_ULL(27, 27) >> #define HISI_PCIE_THR_M GENMASK_ULL(31, 28) >> +#define HISI_PCIE_LEN_M GENMASK_ULL(35, 34) >> #define HISI_PCIE_TARGET_M GENMASK_ULL(52, 36) >> #define HISI_PCIE_TRIG_MODE_M GENMASK_ULL(53, 53) >> #define HISI_PCIE_TRIG_M GENMASK_ULL(59, 56) >> >> +/* Default config of TLP length mode, will count both TLP headers and payloads */ >> +#define HISI_PCIE_LEN_M_DEFAULT 3ULL >> + >> #define HISI_PCIE_MAX_COUNTERS 8 >> #define HISI_PCIE_REG_STEP 8 >> #define HISI_PCIE_THR_MAX_VAL 10 >> @@ -91,6 +95,7 @@ HISI_PCIE_PMU_FILTER_ATTR(thr_len, config1, 3, 0); >> HISI_PCIE_PMU_FILTER_ATTR(thr_mode, config1, 4, 4); >> HISI_PCIE_PMU_FILTER_ATTR(trig_len, config1, 8, 5); >> HISI_PCIE_PMU_FILTER_ATTR(trig_mode, config1, 9, 9); >> +HISI_PCIE_PMU_FILTER_ATTR(len_mode, config1, 11, 10); >> HISI_PCIE_PMU_FILTER_ATTR(port, config2, 15, 0); >> HISI_PCIE_PMU_FILTER_ATTR(bdf, config2, 31, 16); >> >> @@ -215,8 +220,8 @@ static void hisi_pcie_pmu_config_filter(struct perf_event *event) >> { >> struct hisi_pcie_pmu *pcie_pmu = to_pcie_pmu(event->pmu); >> struct hw_perf_event *hwc = &event->hw; >> + u64 port, trig_len, thr_len, len_mode; >> u64 reg = HISI_PCIE_INIT_SET; >> - u64 port, trig_len, thr_len; >> >> /* Config HISI_PCIE_EVENT_CTRL according to event. */ >> reg |= FIELD_PREP(HISI_PCIE_EVENT_M, hisi_pcie_get_real_event(event)); >> @@ -245,6 +250,12 @@ static void hisi_pcie_pmu_config_filter(struct perf_event *event) >> reg |= HISI_PCIE_THR_EN; >> } >> >> + len_mode = hisi_pcie_get_len_mode(event); >> + if (len_mode) >> + reg |= FIELD_PREP(HISI_PCIE_LEN_M, len_mode); >> + else >> + reg |= FIELD_PREP(HISI_PCIE_LEN_M, HISI_PCIE_LEN_M_DEFAULT); >> + >> hisi_pcie_pmu_writeq(pcie_pmu, HISI_PCIE_EVENT_CTRL, hwc->idx, reg); >> } >> >> @@ -711,6 +722,7 @@ static struct attribute *hisi_pcie_pmu_format_attr[] = { >> HISI_PCIE_PMU_FORMAT_ATTR(thr_mode, "config1:4"), >> HISI_PCIE_PMU_FORMAT_ATTR(trig_len, "config1:5-8"), >> HISI_PCIE_PMU_FORMAT_ATTR(trig_mode, "config1:9"), >> + HISI_PCIE_PMU_FORMAT_ATTR(len_mode, "config1:10-11"), >> HISI_PCIE_PMU_FORMAT_ATTR(port, "config2:0-15"), >> HISI_PCIE_PMU_FORMAT_ATTR(bdf, "config2:16-31"), >> NULL > > . >
On Tue, Oct 25, 2022 at 07:32:42PM +0800, Yicong Yang wrote: > +4. TLP Length filter > +When counting bandwidth, the data can be composed of certain parts of TLP > +packets. You can specify it through "len_mode": > + > +- 2'b00: Reserved (Do not use this since the behaviour is undefined) > +- 2'b01: Bandwidth of TLP payloads > +- 2'b10: Bandwidth of TLP headers > +- 2'b11: Bandwidth of both TLP payloads and headers > + > +For example, "len_mode=2" means only counting the bandwidth of TLP headers > +and "len_mode=3" means the final bandwidth data is composed of both TLP > +headers and payloads. You need to carefully using this to avoid losing > +data. For example you're likely to get no counts by "len_mode=1" if the TLPs > +on the traffic has no payload. This config is optional, by default it'll > +be 2'b11. > + > +Example usage of perf:: > + > + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,len_mode=0x1/ sleep 5 The indentation is rather ugly, so I have applied the fixup: ---- >8 ---- diff --git a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst index 83a2ef11b1a08d..bae690dddbebfd 100644 --- a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst +++ b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst @@ -106,21 +106,22 @@ Example usage of perf:: $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,thr_len=0x4,thr_mode=1/ sleep 5 4. TLP Length filter -When counting bandwidth, the data can be composed of certain parts of TLP -packets. You can specify it through "len_mode": -- 2'b00: Reserved (Do not use this since the behaviour is undefined) -- 2'b01: Bandwidth of TLP payloads -- 2'b10: Bandwidth of TLP headers -- 2'b11: Bandwidth of both TLP payloads and headers + When counting bandwidth, the data can be composed of certain parts of TLP + packets. You can specify it through "len_mode": -For example, "len_mode=2" means only counting the bandwidth of TLP headers -and "len_mode=3" means the final bandwidth data is composed of both TLP -headers and payloads. You need to carefully using this to avoid losing -data. For example you're likely to get no counts by "len_mode=1" if the TLPs -on the traffic has no payload. This config is optional, by default it'll -be 2'b11. + - 2'b00: Reserved (Do not use this since the behaviour is undefined) + - 2'b01: Bandwidth of TLP payloads + - 2'b10: Bandwidth of TLP headers + - 2'b11: Bandwidth of both TLP payloads and headers -Example usage of perf:: + For example, "len_mode=2" means only counting the bandwidth of TLP headers + and "len_mode=3" means the final bandwidth data is composed of both TLP + headers and payloads. You need to carefully using this to avoid losing + data. For example you're likely to get no counts by "len_mode=1" if the TLPs + on the traffic has no payload. This config is optional, by default it'll + be 2'b11. - $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,len_mode=0x1/ sleep 5 + Example usage of perf:: + + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,len_mode=0x1/ sleep 5 Also, for formatting consistency, you need to pick up the fix for rest of documentation (as separate patch in your series): ---- >8 ---- From 5f286aceb959a7c70a56cb5c453d3ac0b844fb49 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya <bagasdotme@gmail.com> Date: Thu, 10 Nov 2022 11:03:08 +0700 Subject: [PATCH] Documentation: perf: Indent filter options list The "Filter options" list have a rather ugly indentation. Also, the first paragraph after list name is rendered without separator (as continuation from the name). Align the list by indenting the list items and add a blank line separator for each list name. Fixes: c8602008e247f5 ("docs: perf: Add description for HiSilicon PCIe PMU driver") Signed-off-by: Bagas Sanjaya <bagasdotme@gmail.com> --- .../admin-guide/perf/hisi-pcie-pmu.rst | 80 ++++++++++--------- 1 file changed, 43 insertions(+), 37 deletions(-) diff --git a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst index bae690dddbebfd..54fab870a2aefb 100644 --- a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst +++ b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst @@ -48,62 +48,68 @@ Filter options -------------- 1. Target filter -PMU could only monitor the performance of traffic downstream target Root Ports -or downstream target Endpoint. PCIe PMU driver support "port" and "bdf" -interfaces for users, and these two interfaces aren't supported at the same -time. --port -"port" filter can be used in all PCIe PMU events, target Root Port can be -selected by configuring the 16-bits-bitmap "port". Multi ports can be selected -for AP-layer-events, and only one port can be selected for TL/DL-layer-events. + PMU could only monitor the performance of traffic downstream target Root + Ports or downstream target Endpoint. PCIe PMU driver support "port" and + "bdf" interfaces for users, and these two interfaces aren't supported at the + same time. -For example, if target Root Port is 0000:00:00.0 (x8 lanes), bit0 of bitmap -should be set, port=0x1; if target Root Port is 0000:00:04.0 (x4 lanes), -bit8 is set, port=0x100; if these two Root Ports are both monitored, port=0x101. + - port -Example usage of perf:: + "port" filter can be used in all PCIe PMU events, target Root Port can be + selected by configuring the 16-bits-bitmap "port". Multi ports can be + selected for AP-layer-events, and only one port can be selected for + TL/DL-layer-events. - $# perf stat -e hisi_pcie0_core0/rx_mwr_latency,port=0x1/ sleep 5 + For example, if target Root Port is 0000:00:00.0 (x8 lanes), bit0 of + bitmap should be set, port=0x1; if target Root Port is 0000:00:04.0 (x4 + lanes), bit8 is set, port=0x100; if these two Root Ports are both + monitored, port=0x101. --bdf + Example usage of perf:: -"bdf" filter can only be used in bandwidth events, target Endpoint is selected -by configuring BDF to "bdf". Counter only counts the bandwidth of message -requested by target Endpoint. + $# perf stat -e hisi_pcie0_core0/rx_mwr_latency,port=0x1/ sleep 5 -For example, "bdf=0x3900" means BDF of target Endpoint is 0000:39:00.0. + - bdf -Example usage of perf:: + "bdf" filter can only be used in bandwidth events, target Endpoint is + selected by configuring BDF to "bdf". Counter only counts the bandwidth of + message requested by target Endpoint. - $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,bdf=0x3900/ sleep 5 + For example, "bdf=0x3900" means BDF of target Endpoint is 0000:39:00.0. + + Example usage of perf:: + + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,bdf=0x3900/ sleep 5 2. Trigger filter -Event statistics start when the first time TLP length is greater/smaller -than trigger condition. You can set the trigger condition by writing "trig_len", -and set the trigger mode by writing "trig_mode". This filter can only be used -in bandwidth events. -For example, "trig_len=4" means trigger condition is 2^4 DW, "trig_mode=0" -means statistics start when TLP length > trigger condition, "trig_mode=1" -means start when TLP length < condition. + Event statistics start when the first time TLP length is greater/smaller + than trigger condition. You can set the trigger condition by writing + "trig_len", and set the trigger mode by writing "trig_mode". This filter can + only be used in bandwidth events. -Example usage of perf:: + For example, "trig_len=4" means trigger condition is 2^4 DW, "trig_mode=0" + means statistics start when TLP length > trigger condition, "trig_mode=1" + means start when TLP length < condition. - $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,trig_len=0x4,trig_mode=1/ sleep 5 + Example usage of perf:: + + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,trig_len=0x4,trig_mode=1/ sleep 5 3. Threshold filter -Counter counts when TLP length within the specified range. You can set the -threshold by writing "thr_len", and set the threshold mode by writing -"thr_mode". This filter can only be used in bandwidth events. -For example, "thr_len=4" means threshold is 2^4 DW, "thr_mode=0" means -counter counts when TLP length >= threshold, and "thr_mode=1" means counts -when TLP length < threshold. + Counter counts when TLP length within the specified range. You can set the + threshold by writing "thr_len", and set the threshold mode by writing + "thr_mode". This filter can only be used in bandwidth events. -Example usage of perf:: + For example, "thr_len=4" means threshold is 2^4 DW, "thr_mode=0" means + counter counts when TLP length >= threshold, and "thr_mode=1" means counts + when TLP length < threshold. - $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,thr_len=0x4,thr_mode=1/ sleep 5 + Example usage of perf:: + + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,thr_len=0x4,thr_mode=1/ sleep 5 4. TLP Length filter
On 2022/11/10 12:16, Bagas Sanjaya wrote: > On Tue, Oct 25, 2022 at 07:32:42PM +0800, Yicong Yang wrote: >> +4. TLP Length filter >> +When counting bandwidth, the data can be composed of certain parts of TLP >> +packets. You can specify it through "len_mode": >> + >> +- 2'b00: Reserved (Do not use this since the behaviour is undefined) >> +- 2'b01: Bandwidth of TLP payloads >> +- 2'b10: Bandwidth of TLP headers >> +- 2'b11: Bandwidth of both TLP payloads and headers >> + >> +For example, "len_mode=2" means only counting the bandwidth of TLP headers >> +and "len_mode=3" means the final bandwidth data is composed of both TLP >> +headers and payloads. You need to carefully using this to avoid losing >> +data. For example you're likely to get no counts by "len_mode=1" if the TLPs >> +on the traffic has no payload. This config is optional, by default it'll >> +be 2'b11. >> + >> +Example usage of perf:: >> + >> + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,len_mode=0x1/ sleep 5 > > The indentation is rather ugly, so I have applied the fixup: > will apply the fix and include your patch in v2. Thanks. > ---- >8 ---- > > diff --git a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst > index 83a2ef11b1a08d..bae690dddbebfd 100644 > --- a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst > +++ b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst > @@ -106,21 +106,22 @@ Example usage of perf:: > $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,thr_len=0x4,thr_mode=1/ sleep 5 > > 4. TLP Length filter > -When counting bandwidth, the data can be composed of certain parts of TLP > -packets. You can specify it through "len_mode": > > -- 2'b00: Reserved (Do not use this since the behaviour is undefined) > -- 2'b01: Bandwidth of TLP payloads > -- 2'b10: Bandwidth of TLP headers > -- 2'b11: Bandwidth of both TLP payloads and headers > + When counting bandwidth, the data can be composed of certain parts of TLP > + packets. You can specify it through "len_mode": > > -For example, "len_mode=2" means only counting the bandwidth of TLP headers > -and "len_mode=3" means the final bandwidth data is composed of both TLP > -headers and payloads. You need to carefully using this to avoid losing > -data. For example you're likely to get no counts by "len_mode=1" if the TLPs > -on the traffic has no payload. This config is optional, by default it'll > -be 2'b11. > + - 2'b00: Reserved (Do not use this since the behaviour is undefined) > + - 2'b01: Bandwidth of TLP payloads > + - 2'b10: Bandwidth of TLP headers > + - 2'b11: Bandwidth of both TLP payloads and headers > > -Example usage of perf:: > + For example, "len_mode=2" means only counting the bandwidth of TLP headers > + and "len_mode=3" means the final bandwidth data is composed of both TLP > + headers and payloads. You need to carefully using this to avoid losing > + data. For example you're likely to get no counts by "len_mode=1" if the TLPs > + on the traffic has no payload. This config is optional, by default it'll > + be 2'b11. > > - $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,len_mode=0x1/ sleep 5 > + Example usage of perf:: > + > + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,len_mode=0x1/ sleep 5 > > Also, for formatting consistency, you need to pick up the fix for rest of > documentation (as separate patch in your series): > > ---- >8 ---- > > From 5f286aceb959a7c70a56cb5c453d3ac0b844fb49 Mon Sep 17 00:00:00 2001 > From: Bagas Sanjaya <bagasdotme@gmail.com> > Date: Thu, 10 Nov 2022 11:03:08 +0700 > Subject: [PATCH] Documentation: perf: Indent filter options list > > The "Filter options" list have a rather ugly indentation. Also, the first > paragraph after list name is rendered without separator (as continuation > from the name). > > Align the list by indenting the list items and add a blank line > separator for each list name. > > Fixes: c8602008e247f5 ("docs: perf: Add description for HiSilicon PCIe PMU driver") > Signed-off-by: Bagas Sanjaya <bagasdotme@gmail.com> > --- > .../admin-guide/perf/hisi-pcie-pmu.rst | 80 ++++++++++--------- > 1 file changed, 43 insertions(+), 37 deletions(-) > > diff --git a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst > index bae690dddbebfd..54fab870a2aefb 100644 > --- a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst > +++ b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst > @@ -48,62 +48,68 @@ Filter options > -------------- > > 1. Target filter > -PMU could only monitor the performance of traffic downstream target Root Ports > -or downstream target Endpoint. PCIe PMU driver support "port" and "bdf" > -interfaces for users, and these two interfaces aren't supported at the same > -time. > > --port > -"port" filter can be used in all PCIe PMU events, target Root Port can be > -selected by configuring the 16-bits-bitmap "port". Multi ports can be selected > -for AP-layer-events, and only one port can be selected for TL/DL-layer-events. > + PMU could only monitor the performance of traffic downstream target Root > + Ports or downstream target Endpoint. PCIe PMU driver support "port" and > + "bdf" interfaces for users, and these two interfaces aren't supported at the > + same time. > > -For example, if target Root Port is 0000:00:00.0 (x8 lanes), bit0 of bitmap > -should be set, port=0x1; if target Root Port is 0000:00:04.0 (x4 lanes), > -bit8 is set, port=0x100; if these two Root Ports are both monitored, port=0x101. > + - port > > -Example usage of perf:: > + "port" filter can be used in all PCIe PMU events, target Root Port can be > + selected by configuring the 16-bits-bitmap "port". Multi ports can be > + selected for AP-layer-events, and only one port can be selected for > + TL/DL-layer-events. > > - $# perf stat -e hisi_pcie0_core0/rx_mwr_latency,port=0x1/ sleep 5 > + For example, if target Root Port is 0000:00:00.0 (x8 lanes), bit0 of > + bitmap should be set, port=0x1; if target Root Port is 0000:00:04.0 (x4 > + lanes), bit8 is set, port=0x100; if these two Root Ports are both > + monitored, port=0x101. > > --bdf > + Example usage of perf:: > > -"bdf" filter can only be used in bandwidth events, target Endpoint is selected > -by configuring BDF to "bdf". Counter only counts the bandwidth of message > -requested by target Endpoint. > + $# perf stat -e hisi_pcie0_core0/rx_mwr_latency,port=0x1/ sleep 5 > > -For example, "bdf=0x3900" means BDF of target Endpoint is 0000:39:00.0. > + - bdf > > -Example usage of perf:: > + "bdf" filter can only be used in bandwidth events, target Endpoint is > + selected by configuring BDF to "bdf". Counter only counts the bandwidth of > + message requested by target Endpoint. > > - $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,bdf=0x3900/ sleep 5 > + For example, "bdf=0x3900" means BDF of target Endpoint is 0000:39:00.0. > + > + Example usage of perf:: > + > + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,bdf=0x3900/ sleep 5 > > 2. Trigger filter > -Event statistics start when the first time TLP length is greater/smaller > -than trigger condition. You can set the trigger condition by writing "trig_len", > -and set the trigger mode by writing "trig_mode". This filter can only be used > -in bandwidth events. > > -For example, "trig_len=4" means trigger condition is 2^4 DW, "trig_mode=0" > -means statistics start when TLP length > trigger condition, "trig_mode=1" > -means start when TLP length < condition. > + Event statistics start when the first time TLP length is greater/smaller > + than trigger condition. You can set the trigger condition by writing > + "trig_len", and set the trigger mode by writing "trig_mode". This filter can > + only be used in bandwidth events. > > -Example usage of perf:: > + For example, "trig_len=4" means trigger condition is 2^4 DW, "trig_mode=0" > + means statistics start when TLP length > trigger condition, "trig_mode=1" > + means start when TLP length < condition. > > - $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,trig_len=0x4,trig_mode=1/ sleep 5 > + Example usage of perf:: > + > + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,trig_len=0x4,trig_mode=1/ sleep 5 > > 3. Threshold filter > -Counter counts when TLP length within the specified range. You can set the > -threshold by writing "thr_len", and set the threshold mode by writing > -"thr_mode". This filter can only be used in bandwidth events. > > -For example, "thr_len=4" means threshold is 2^4 DW, "thr_mode=0" means > -counter counts when TLP length >= threshold, and "thr_mode=1" means counts > -when TLP length < threshold. > + Counter counts when TLP length within the specified range. You can set the > + threshold by writing "thr_len", and set the threshold mode by writing > + "thr_mode". This filter can only be used in bandwidth events. > > -Example usage of perf:: > + For example, "thr_len=4" means threshold is 2^4 DW, "thr_mode=0" means > + counter counts when TLP length >= threshold, and "thr_mode=1" means counts > + when TLP length < threshold. > > - $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,thr_len=0x4,thr_mode=1/ sleep 5 > + Example usage of perf:: > + > + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,thr_len=0x4,thr_mode=1/ sleep 5 > > 4. TLP Length filter > >
diff --git a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst index bbe66480ff85..83a2ef11b1a0 100644 --- a/Documentation/admin-guide/perf/hisi-pcie-pmu.rst +++ b/Documentation/admin-guide/perf/hisi-pcie-pmu.rst @@ -104,3 +104,23 @@ when TLP length < threshold. Example usage of perf:: $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,thr_len=0x4,thr_mode=1/ sleep 5 + +4. TLP Length filter +When counting bandwidth, the data can be composed of certain parts of TLP +packets. You can specify it through "len_mode": + +- 2'b00: Reserved (Do not use this since the behaviour is undefined) +- 2'b01: Bandwidth of TLP payloads +- 2'b10: Bandwidth of TLP headers +- 2'b11: Bandwidth of both TLP payloads and headers + +For example, "len_mode=2" means only counting the bandwidth of TLP headers +and "len_mode=3" means the final bandwidth data is composed of both TLP +headers and payloads. You need to carefully using this to avoid losing +data. For example you're likely to get no counts by "len_mode=1" if the TLPs +on the traffic has no payload. This config is optional, by default it'll +be 2'b11. + +Example usage of perf:: + + $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,len_mode=0x1/ sleep 5 diff --git a/drivers/perf/hisilicon/hisi_pcie_pmu.c b/drivers/perf/hisilicon/hisi_pcie_pmu.c index 071e63d9a9ac..6fee0b6e163b 100644 --- a/drivers/perf/hisilicon/hisi_pcie_pmu.c +++ b/drivers/perf/hisilicon/hisi_pcie_pmu.c @@ -47,10 +47,14 @@ #define HISI_PCIE_EVENT_M GENMASK_ULL(15, 0) #define HISI_PCIE_THR_MODE_M GENMASK_ULL(27, 27) #define HISI_PCIE_THR_M GENMASK_ULL(31, 28) +#define HISI_PCIE_LEN_M GENMASK_ULL(35, 34) #define HISI_PCIE_TARGET_M GENMASK_ULL(52, 36) #define HISI_PCIE_TRIG_MODE_M GENMASK_ULL(53, 53) #define HISI_PCIE_TRIG_M GENMASK_ULL(59, 56) +/* Default config of TLP length mode, will count both TLP headers and payloads */ +#define HISI_PCIE_LEN_M_DEFAULT 3ULL + #define HISI_PCIE_MAX_COUNTERS 8 #define HISI_PCIE_REG_STEP 8 #define HISI_PCIE_THR_MAX_VAL 10 @@ -91,6 +95,7 @@ HISI_PCIE_PMU_FILTER_ATTR(thr_len, config1, 3, 0); HISI_PCIE_PMU_FILTER_ATTR(thr_mode, config1, 4, 4); HISI_PCIE_PMU_FILTER_ATTR(trig_len, config1, 8, 5); HISI_PCIE_PMU_FILTER_ATTR(trig_mode, config1, 9, 9); +HISI_PCIE_PMU_FILTER_ATTR(len_mode, config1, 11, 10); HISI_PCIE_PMU_FILTER_ATTR(port, config2, 15, 0); HISI_PCIE_PMU_FILTER_ATTR(bdf, config2, 31, 16); @@ -215,8 +220,8 @@ static void hisi_pcie_pmu_config_filter(struct perf_event *event) { struct hisi_pcie_pmu *pcie_pmu = to_pcie_pmu(event->pmu); struct hw_perf_event *hwc = &event->hw; + u64 port, trig_len, thr_len, len_mode; u64 reg = HISI_PCIE_INIT_SET; - u64 port, trig_len, thr_len; /* Config HISI_PCIE_EVENT_CTRL according to event. */ reg |= FIELD_PREP(HISI_PCIE_EVENT_M, hisi_pcie_get_real_event(event)); @@ -245,6 +250,12 @@ static void hisi_pcie_pmu_config_filter(struct perf_event *event) reg |= HISI_PCIE_THR_EN; } + len_mode = hisi_pcie_get_len_mode(event); + if (len_mode) + reg |= FIELD_PREP(HISI_PCIE_LEN_M, len_mode); + else + reg |= FIELD_PREP(HISI_PCIE_LEN_M, HISI_PCIE_LEN_M_DEFAULT); + hisi_pcie_pmu_writeq(pcie_pmu, HISI_PCIE_EVENT_CTRL, hwc->idx, reg); } @@ -711,6 +722,7 @@ static struct attribute *hisi_pcie_pmu_format_attr[] = { HISI_PCIE_PMU_FORMAT_ATTR(thr_mode, "config1:4"), HISI_PCIE_PMU_FORMAT_ATTR(trig_len, "config1:5-8"), HISI_PCIE_PMU_FORMAT_ATTR(trig_mode, "config1:9"), + HISI_PCIE_PMU_FORMAT_ATTR(len_mode, "config1:10-11"), HISI_PCIE_PMU_FORMAT_ATTR(port, "config2:0-15"), HISI_PCIE_PMU_FORMAT_ATTR(bdf, "config2:16-31"), NULL