CoreSight October 2017

coresight@lists.linaro.org

11 participants
15 discussions

Re: [EXTERNAL] Re: [PATCH v2 01/23] perf tools: Add initial hooks for decoding coresight traces

by Mathieu Poirier

On 4 October 2017 at 15:30, Jeremiassen, Tor <tor(a)ti.com> wrote: > As far as I could tell, the C flags when compiling cs-etm.c comes from CFLAGS-$(CONFIG_AUXTRACE), while the C flags that are used when compiling auxtrace.c come from CFLAGS. > > That's why it has to be in both. My compilation failed otherwise. Even nowadays? Can you check with my branch? > > Tor > > -----Original Message----- > From: Mathieu Poirier [mailto:mathieu.poirier@linaro.org] > Sent: Wednesday, October 4, 2017 4:24 PM > To: Jeremiassen, Tor > Cc: coresight(a)lists.linaro.org > Subject: [EXTERNAL] Re: [PATCH v2 01/23] perf tools: Add initial hooks for decoding coresight traces > > On Fri, Sep 22, 2017 at 04:03:18AM -0500, Tor Jeremiassen wrote: >> Adds new file cs-etm.c that adds auxtrace info event handler for >> cs-etm (coresight ETM) traces, event handling data structures and the >> necessary callback functions required to be provided in the auxtrace struct. >> >> The coresight trace decoding is enabled by setting the CSTRACE_PATH >> environment variable to point to the decoder subdirectory in the >> coresight trace decoder library. >> >> Signed-off-by: Tor Jeremiassen <tor(a)ti.com> >> --- >> tools/perf/Makefile.config | 5 + >> tools/perf/util/Build | 5 + >> tools/perf/util/auxtrace.c | 2 + >> tools/perf/util/cs-etm.c | 230 +++++++++++++++++++++++++++++++++++++++++++++ >> tools/perf/util/cs-etm.h | 16 ++++ >> 5 files changed, 258 insertions(+) >> create mode 100644 tools/perf/util/cs-etm.c >> >> diff --git a/tools/perf/Makefile.config b/tools/perf/Makefile.config >> index 63f534a..d1a83aa 100644 >> --- a/tools/perf/Makefile.config >> +++ b/tools/perf/Makefile.config >> @@ -339,6 +339,11 @@ ifeq ($(feature-setns), 1) >> $(call detected,CONFIG_SETNS) >> endif >> >> +ifdef CSTRACE_PATH >> + CFLAGS += -DHAVE_CSTRACE_SUPPORT >> + CFLAGS-$(CONFIG_AUXTRACE) += -DHAVE_CSTRACE_SUPPORT > > I couldn't figure out what the above line was doing so I removed it. > >> +endif >> + >> ifndef NO_LIBELF >> CFLAGS += -DHAVE_LIBELF_SUPPORT >> EXTLIBS += -lelf >> diff --git a/tools/perf/util/Build b/tools/perf/util/Build index >> 94518c1..23439f4 100644 >> --- a/tools/perf/util/Build >> +++ b/tools/perf/util/Build >> @@ -83,6 +83,11 @@ libperf-$(CONFIG_AUXTRACE) += auxtrace.o >> libperf-$(CONFIG_AUXTRACE) += intel-pt-decoder/ >> libperf-$(CONFIG_AUXTRACE) += intel-pt.o >> libperf-$(CONFIG_AUXTRACE) += intel-bts.o >> + >> +ifdef CSTRACE_PATH >> +libperf-$(CONFIG_AUXTRACE) += cs-etm.o endif >> + >> libperf-y += parse-branch-options.o >> libperf-y += dump-insn.o >> libperf-y += parse-regs-options.o >> diff --git a/tools/perf/util/auxtrace.c b/tools/perf/util/auxtrace.c >> index 5547457..37f9d08 100644 >> --- a/tools/perf/util/auxtrace.c >> +++ b/tools/perf/util/auxtrace.c >> @@ -55,6 +55,7 @@ >> #include "debug.h" >> #include <subcmd/parse-options.h> >> >> +#include "cs-etm.h" >> #include "intel-pt.h" >> #include "intel-bts.h" >> >> @@ -914,6 +915,7 @@ int perf_event__process_auxtrace_info(struct perf_tool *tool __maybe_unused, >> case PERF_AUXTRACE_INTEL_BTS: >> return intel_bts_process_auxtrace_info(event, session); >> case PERF_AUXTRACE_CS_ETM: >> + return cs_etm__process_auxtrace_info(event, session); >> case PERF_AUXTRACE_UNKNOWN: >> default: >> return -EINVAL; >> diff --git a/tools/perf/util/cs-etm.c b/tools/perf/util/cs-etm.c new >> file mode 100644 index 0000000..e3f4e86 >> --- /dev/null >> +++ b/tools/perf/util/cs-etm.c >> @@ -0,0 +1,230 @@ >> +/* >> + * Copyright(C) 2015-2017 Linaro Limited. All rights reserved. >> + * Author: Tor Jeremiassen <tor(a)ti.com> >> + * >> + * This program is free software; you can redistribute it and/or >> +modify it >> + * under the terms of the GNU General Public License version 2 as >> +published by >> + * the Free Software Foundation. >> + * >> + * This program is distributed in the hope that it will be useful, >> +but WITHOUT >> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY >> +or >> + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public >> +License for >> + * more details. >> + * >> + * You should have received a copy of the GNU General Public License >> +along with >> + * this program. If not, see <http://www.gnu.org/licenses/>. >> + */ >> + >> +#include <linux/bitops.h> >> +#include <linux/err.h> >> +#include <linux/kernel.h> >> +#include <linux/log2.h> >> +#include <linux/types.h> >> + >> +#include <stdlib.h> >> + >> +#include "auxtrace.h" >> +#include "color.h" >> +#include "cs-etm.h" >> +#include "debug.h" >> +#include "evlist.h" >> +#include "intlist.h" >> +#include "machine.h" >> +#include "perf.h" >> +#include "thread.h" >> +#include "thread_map.h" >> +#include "thread-stack.h" >> +#include "util.h" >> + >> +struct cs_etm_auxtrace { >> + struct auxtrace auxtrace; >> + struct auxtrace_queues queues; >> + struct auxtrace_heap heap; >> + u64 **metadata; >> + u32 auxtrace_type; >> + struct perf_session *session; >> + struct machine *machine; >> + struct perf_evsel *switch_evsel; >> + struct thread *unknown_thread; >> + uint32_t num_cpu; >> + bool timeless_decoding; >> + bool sampling_mode; >> + bool snapshot_mode; >> + bool data_queued; >> + bool synth_needs_swap; >> + bool sample_instructions; >> + u64 instructions_sample_type; >> + u64 instructions_sample_period; >> + u64 instructions_id; >> + struct itrace_synth_opts synth_opts; >> + unsigned int pmu_type; >> + u64 kernel_start; >> +}; >> + >> +struct cs_etm_queue { >> + struct cs_etm_auxtrace *etm; >> + unsigned int queue_nr; >> + struct auxtrace_buffer *buffer; >> + const struct cs_etm_state *state; >> + union perf_event *event_buf; >> + bool on_heap; >> + bool step_through_buffers; >> + bool use_buffer_pid_tid; >> + pid_t pid, tid; >> + int cpu; >> + struct thread *thread; >> + u64 time; >> + u64 timestamp; >> + bool stop; >> + struct cs_etm_decoder *decoder; >> + u64 offset; >> + bool eot; >> +}; >> + >> +static int cs_etm__flush_events(struct perf_session *session, >> + struct perf_tool *tool) >> +{ >> + (void) session; >> + (void) tool; >> + return 0; >> +} >> + >> +static void cs_etm__free_queue(void *priv) { >> + struct cs_etm_queue *etmq = priv; >> + >> + if (!etmq) >> + return; >> + >> + thread__zput(etmq->thread); >> + free(etmq); >> +} >> + >> +static void cs_etm__free_events(struct perf_session *session) { >> + struct cs_etm_auxtrace *aux = container_of(session->auxtrace, >> + struct cs_etm_auxtrace, >> + auxtrace); >> + struct auxtrace_queues *queues = &aux->queues; >> + unsigned int i; >> + >> + for (i = 0; i < queues->nr_queues; i++) { >> + cs_etm__free_queue(queues->queue_array[i].priv); >> + queues->queue_array[i].priv = NULL; >> + } >> + >> + auxtrace_queues__free(queues); >> +} >> + >> +static void cs_etm__free(struct perf_session *session) { >> + struct cs_etm_auxtrace *aux = container_of(session->auxtrace, >> + struct cs_etm_auxtrace, >> + auxtrace); >> + auxtrace_heap__free(&aux->heap); >> + cs_etm__free_events(session); >> + session->auxtrace = NULL; >> + >> + zfree(&aux); >> +} >> + >> +static int cs_etm__process_event(struct perf_session *session, >> + union perf_event *event, >> + struct perf_sample *sample, >> + struct perf_tool *tool) >> +{ >> + (void) session; >> + (void) event; >> + (void) sample; >> + (void) tool; >> + return 0; >> +} >> + >> +static int cs_etm__process_auxtrace_event(struct perf_session *session, >> + union perf_event *event, >> + struct perf_tool *tool) >> +{ >> + (void) session; >> + (void) event; >> + (void) tool; >> + return 0; >> +} >> + >> +int cs_etm__process_auxtrace_info(union perf_event *event, >> + struct perf_session *session) >> +{ >> + struct auxtrace_info_event *auxtrace_info = &event->auxtrace_info; >> + size_t event_header_size = sizeof(struct perf_event_header); >> + size_t info_header_size; >> + size_t total_size = auxtrace_info->header.size; >> + struct cs_etm_auxtrace *etm = NULL; >> + int err = 0; >> + >> + /* >> + * sizeof(auxtrace_info_event::type) + >> + * sizeof(auxtrace_info_event::reserved) == 8 >> + */ >> + info_header_size = 8; >> + >> + if (total_size < (event_header_size + info_header_size)) >> + return -EINVAL; >> + >> + etm = zalloc(sizeof(*etm)); >> + >> + if (!etm) >> + return -ENOMEM; >> + >> + err = auxtrace_queues__init(&etm->queues); >> + if (err) >> + goto err_free_etm; >> + >> + etm->unknown_thread = thread__new(999999999, 999999999); >> + if (!etm->unknown_thread) { >> + err = -ENOMEM; >> + goto err_free_queues; >> + } >> + >> + err = thread__set_comm(etm->unknown_thread, "unknown", 0); >> + if (err) >> + goto err_delete_thread; >> + >> + etm->session = session; >> + etm->machine = &session->machines.host; >> + etm->kernel_start = machine__kernel_start(etm->machine); >> + >> + if (thread__init_map_groups(etm->unknown_thread, >> + etm->machine)) { >> + err = -ENOMEM; >> + goto err_delete_thread; >> + } >> + >> + etm->auxtrace_type = auxtrace_info->type; >> + >> + etm->auxtrace.process_event = cs_etm__process_event; >> + etm->auxtrace.process_auxtrace_event = cs_etm__process_auxtrace_event; >> + etm->auxtrace.flush_events = cs_etm__flush_events; >> + etm->auxtrace.free_events = cs_etm__free_events; >> + etm->auxtrace.free = cs_etm__free; >> + session->auxtrace = &etm->auxtrace; >> + >> + if (dump_trace) >> + return 0; >> + >> + err = auxtrace_queues__process_index(&etm->queues, session); >> + if (err) >> + goto err_delete_thread; >> + >> + etm->data_queued = etm->queues.populated; >> + >> + return 0; >> + >> +err_delete_thread: >> + thread__delete(etm->unknown_thread); >> +err_free_queues: >> + auxtrace_queues__free(&etm->queues); >> + session->auxtrace = NULL; >> +err_free_etm: >> + zfree(&etm); >> + >> + return err; >> +} >> diff --git a/tools/perf/util/cs-etm.h b/tools/perf/util/cs-etm.h index >> 3cc6bc3..9e43ede 100644 >> --- a/tools/perf/util/cs-etm.h >> +++ b/tools/perf/util/cs-etm.h >> @@ -18,6 +18,9 @@ >> #ifndef INCLUDE__UTIL_PERF_CS_ETM_H__ #define >> INCLUDE__UTIL_PERF_CS_ETM_H__ >> >> +#include "util/event.h" >> +#include "util/session.h" >> + >> /* Versionning header in case things need tro change in the future. That way >> * decoding of old snapshot is still possible. >> */ >> @@ -71,4 +74,17 @@ static const u64 __perf_cs_etmv4_magic = 0x4040404040404040ULL; >> #define CS_ETMV3_PRIV_SIZE (CS_ETM_PRIV_MAX * sizeof(u64)) #define >> CS_ETMV4_PRIV_SIZE (CS_ETMV4_PRIV_MAX * sizeof(u64)) >> >> +#ifdef HAVE_CSTRACE_SUPPORT >> +int cs_etm__process_auxtrace_info(union perf_event *event, >> + struct perf_session *session); >> +#else >> +static inline int cs_etm__process_auxtrace_info(union perf_event *event, >> + struct perf_session *session) >> +{ >> + (void) event; >> + (void) session; >> + return -1; >> +} >> +#endif >> + >> #endif >> -- >> 2.7.4 >> >> _______________________________________________ >> CoreSight mailing list >> CoreSight(a)lists.linaro.org >> https://lists.linaro.org/mailman/listinfo/coresight

7 years

Re: Re: Re: Access to QDSS

by marekzmyslowski

RTFM to me :) Solution was easy: echo 1 > /sys/kernel/debug/coresight_cpu_debug/enable Regards Marek W dniu 2017-10-02 19:16:12 użytkownik Mathieu Poirier <mathieu.poirier(a)linaro.org> napisał: > On 2 October 2017 at 09:41, marekzmyslowski > <marekzmyslowski(a)poczta.onet.pl> wrote: > > Hello Mathieu, > > > > I'm trying to use CSAL for DragonBoard. The fisrt thing there is rom_init. According to the docs, the rom that needs to be mmaped is 0x800000. Is there any way to check if the power domain or clock turned om? > > There is no way to check from user space. When using the CSAL people > will typically connect to the board using a Dstream (or another Jtag > solution). That typically switch on the power domain and the required > clock, given the initialisation script for the board has done that > correctly. > > The dragonboard is popular and ubiquitous, it shouldn't be hard to > find a Jtag init script on the internet somewhere. If you don't have > a Jtag you can read the Jtag init script for that board and enact the > same operations by mmap'ing memory areas and flipping bits. > > > > > Regards > > Marek > > > > W dniu 2017-10-02 16:50:08 użytkownik Mathieu Poirier <mathieu.poirier(a)linaro.org> napisał: > >> On 2 October 2017 at 02:15, marekzmyslowski > >> <marekzmyslowski(a)poczta.onet.pl> wrote: > >> > Hello, > >> > > >> > I'm trying to get access to Qualcomm Debug Subsystem on my DragonBoard 410c by using the mmap. Just want to read data available in DQSS_DAPROM. However when I try to do that, the system is rebooting. Any idea why I can't do that? > >> > > >> > int size = 4096; > >> > off_t addr_to_map = 0x00800000; //according to docs this is an address of QDSS_DAPROM > >> > > >> > int mem_fd = open("/dev/mem", O_RDWR); > >> > localv = mmap(0, size, PROT_READ, MAP_SHARED, mem_fd, addr_to_map); > >> > printf ("mmap completed %p.\n", localv); > >> > tab = (unsigned int *)localv; > >> > printf ("%u\n", tab[0]); > >> > > >> > Result of my application execution: > >> > mmap completed 0x7fa29f0000.[ 484.585864] Synchronous External Abort: synchronous external abort (0x92000010) at 0x0000007fa29f0000 > >> > >> > >> This is usually an indication the memory area you are trying to reach > >> is not accessible - either the power domain hasn't been switched on or > >> the clock(s) aren't enabled. Is there a device driver specific to the > >> DQSS_DAPROM? > >> > >> > > >> > Regards > >> > Marek > >> > > >> > Linux linaro-developer 4.14.0-rc1+ #2 SMP PREEMPT Thu Sep 21 10:39:33 CEST 2017 aarch64 GNU/Linux > >> > > >> > _______________________________________________ > >> > CoreSight mailing list > >> > CoreSight(a)lists.linaro.org > >> > https://lists.linaro.org/mailman/listinfo/coresight > >> > > > > > > >

7 years

Re: Re: Access to QDSS

by marekzmyslowski

Hello Mathieu, I'm trying to use CSAL for DragonBoard. The fisrt thing there is rom_init. According to the docs, the rom that needs to be mmaped is 0x800000. Is there any way to check if the power domain or clock turned om? Regards Marek W dniu 2017-10-02 16:50:08 użytkownik Mathieu Poirier <mathieu.poirier(a)linaro.org> napisał: > On 2 October 2017 at 02:15, marekzmyslowski > <marekzmyslowski(a)poczta.onet.pl> wrote: > > Hello, > > > > I'm trying to get access to Qualcomm Debug Subsystem on my DragonBoard 410c by using the mmap. Just want to read data available in DQSS_DAPROM. However when I try to do that, the system is rebooting. Any idea why I can't do that? > > > > int size = 4096; > > off_t addr_to_map = 0x00800000; //according to docs this is an address of QDSS_DAPROM > > > > int mem_fd = open("/dev/mem", O_RDWR); > > localv = mmap(0, size, PROT_READ, MAP_SHARED, mem_fd, addr_to_map); > > printf ("mmap completed %p.\n", localv); > > tab = (unsigned int *)localv; > > printf ("%u\n", tab[0]); > > > > Result of my application execution: > > mmap completed 0x7fa29f0000.[ 484.585864] Synchronous External Abort: synchronous external abort (0x92000010) at 0x0000007fa29f0000 > > > This is usually an indication the memory area you are trying to reach > is not accessible - either the power domain hasn't been switched on or > the clock(s) aren't enabled. Is there a device driver specific to the > DQSS_DAPROM? > > > > > Regards > > Marek > > > > Linux linaro-developer 4.14.0-rc1+ #2 SMP PREEMPT Thu Sep 21 10:39:33 CEST 2017 aarch64 GNU/Linux > > > > _______________________________________________ > > CoreSight mailing list > > CoreSight(a)lists.linaro.org > > https://lists.linaro.org/mailman/listinfo/coresight >

7 years

Access to QDSS

by marekzmyslowski

Hello, I'm trying to get access to Qualcomm Debug Subsystem on my DragonBoard 410c by using the mmap. Just want to read data available in DQSS_DAPROM. However when I try to do that, the system is rebooting. Any idea why I can't do that? int size = 4096; off_t addr_to_map = 0x00800000; //according to docs this is an address of QDSS_DAPROM int mem_fd = open("/dev/mem", O_RDWR); localv = mmap(0, size, PROT_READ, MAP_SHARED, mem_fd, addr_to_map); printf ("mmap completed %p.\n", localv); tab = (unsigned int *)localv; printf ("%u\n", tab[0]); Result of my application execution: mmap completed 0x7fa29f0000.[ 484.585864] Synchronous External Abort: synchronous external abort (0x92000010) at 0x0000007fa29f0000 Regards Marek Linux linaro-developer 4.14.0-rc1+ #2 SMP PREEMPT Thu Sep 21 10:39:33 CEST 2017 aarch64 GNU/Linux

7 years

Re: Re: Access to QDSS

by marekzmyslowski

Hi Leo, Busybox gives the same results: root@linaro-developer:~# busybox devmem 0x00800000 [ 1713.501755] Synchronous External Abort: synchronous external abort (0x92000010) at 0x0000007fbb6eb000 How this clock can be enabled? I though it is already enabled. I used default configuration for db410c and it includes the msm8916.dtsi file. Regards Marek W dniu 2017-10-02 12:02:55 użytkownik Leo Yan <leo.yan(a)linaro.org> napisał: > On Mon, Oct 02, 2017 at 11:15:27AM +0200, marekzmyslowski wrote: > > Hello, > > > > I'm trying to get access to Qualcomm Debug Subsystem on my DragonBoard 410c by using the mmap. Just want to read data available in DQSS_DAPROM. However when I try to do that, the system is rebooting. Any idea why I can't do that? > > > > int size = 4096; > > off_t addr_to_map = 0x00800000; //according to docs this is an address of QDSS_DAPROM > > > > int mem_fd = open("/dev/mem", O_RDWR); > > localv = mmap(0, size, PROT_READ, MAP_SHARED, mem_fd, addr_to_map); > > printf ("mmap completed %p.\n", localv); > > tab = (unsigned int *)localv; > > printf ("%u\n", tab[0]); > > > > Result of my application execution: > > mmap completed 0x7fa29f0000.[ 484.585864] Synchronous External Abort: synchronous external abort (0x92000010) at 0x0000007fa29f0000 > > Seems this external exception relate with clock, so need enable > RPM_QDSS_CLK clock firstly? > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arc… > > Thanks, > Leo Yan >

7 years

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

CoreSight October 2017