linux-toradex.git/tools/perf/arch, branch v6.6

tools headers UAPI: Sync files changed by new fchmodat2 and map_shadow_stack syscalls with the kernel sources

2023-09-13T11:24:51+00:00

To pick the changes in these csets:

  c35559f94ebc3e3b ("x86/shstk: Introduce map_shadow_stack syscall")
  78252deb023cf087 ("arch: Register fchmodat2, usually as syscall 452")

That add support for this new syscall in tools such as 'perf trace'.

For instance, this is now possible:

  # perf trace -v -e fchmodat*,map_shadow_stack --max-events=4
  Using CPUID AuthenticAMD-25-21-0
  Reusing "openat" BPF sys_enter augmenter for "fchmodat"
  event qualifier tracepoint filter: (common_pid != 3499340 && common_pid != 11259) && (id == 268 || id == 452 || id == 453)
  ^C#

  And it'll work as with other syscalls, for instance openat:

  # perf trace -e openat* --max-events=4
     0.000 ( 0.015 ms): systemd-oomd/1150 openat(dfd: CWD, filename: "/proc/meminfo", flags: RDONLY|CLOEXEC)    = 11
     0.068 ( 0.019 ms): systemd-oomd/1150 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1001.slice/user@1001.service/memory.pressure", flags: RDONLY|CLOEXEC) = 11
     0.119 ( 0.008 ms): systemd-oomd/1150 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1001.slice/user@1001.service/memory.current", flags: RDONLY|CLOEXEC) = 11
     0.138 ( 0.006 ms): systemd-oomd/1150 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1001.slice/user@1001.service/memory.min", flags: RDONLY|CLOEXEC) = 11
  #

That is the filter expression attached to the raw_syscalls:sys_{enter,exit}
tracepoints.

  $ find tools/perf/arch/ -name "syscall*tbl" | xargs grep -E fchmodat\|sys_map_shadow_stack
  tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl:258	n64	fchmodat			sys_fchmodat
  tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl:452	n64	fchmodat2			sys_fchmodat2
  tools/perf/arch/powerpc/entry/syscalls/syscall.tbl:297	common	fchmodat			sys_fchmodat
  tools/perf/arch/powerpc/entry/syscalls/syscall.tbl:452	common	fchmodat2			sys_fchmodat2
  tools/perf/arch/s390/entry/syscalls/syscall.tbl:299  common	fchmodat		sys_fchmodat			sys_fchmodat
  tools/perf/arch/s390/entry/syscalls/syscall.tbl:452  common	fchmodat2		sys_fchmodat2			sys_fchmodat2
  tools/perf/arch/x86/entry/syscalls/syscall_64.tbl:268	common	fchmodat		sys_fchmodat
  tools/perf/arch/x86/entry/syscalls/syscall_64.tbl:452	common	fchmodat2		sys_fchmodat2
  tools/perf/arch/x86/entry/syscalls/syscall_64.tbl:453	64	map_shadow_stack	sys_map_shadow_stack
  $

  $ grep -Ew map_shadow_stack\|fchmodat2 /tmp/build/perf-tools/arch/x86/include/generated/asm/syscalls_64.c
	[452] = "fchmodat2",
	[453] = "map_shadow_stack",
  $

This addresses these perf build warnings:

  Warning: Kernel ABI header differences:
    diff -u tools/include/uapi/asm-generic/unistd.h include/uapi/asm-generic/unistd.h
    diff -u tools/perf/arch/x86/entry/syscalls/syscall_64.tbl arch/x86/entry/syscalls/syscall_64.tbl
    diff -u tools/perf/arch/powerpc/entry/syscalls/syscall.tbl arch/powerpc/kernel/syscalls/syscall.tbl
    diff -u tools/perf/arch/s390/entry/syscalls/syscall.tbl arch/s390/kernel/syscalls/syscall.tbl
    diff -u tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl arch/mips/kernel/syscalls/syscall_n64.tbl

Cc: Adrian Hunter 
Cc: Christian Brauner 
Cc: Dave Hansen 
Cc: Ian Rogers 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Palmer Dabbelt 
Cc: Rick Edgecombe 
Link: https://lore.kernel.org/lkml/ZP8bE7aXDBu%2Fdrak@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf pmu: Remove logic for PMU name being NULL

2023-08-25T13:22:16+00:00

The PMU name could be NULL in the case of the fake_pmu. Initialize the
name for the fake_pmu to "fake" so that all other logic can assume it
is initialized. Add a const to the type of name so that a literal can
be used to avoid additional initialization code. Propagate the cost
through related routines and remove now unnecessary "(char *)"
casts. Doing this located a bug in builtin-list for the pmu_glob that
was missing a strdup.

Signed-off-by: Ian Rogers 
Link: https://lore.kernel.org/r/20230825024002.801955-3-irogers@google.com
Cc: K Prateek Nayak 
Cc: Ravi Bangoria 
Cc: James Clark 
Cc: Mark Rutland 
Cc: Suzuki Poulouse 
Cc: Sean Christopherson 
Cc: Wei Li 
Cc: Peter Zijlstra 
Cc: Adrian Hunter 
Cc: Arnaldo Carvalho de Melo 
Cc: Huacai Chen 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Will Deacon 
Cc: Leo Yan 
Cc: Mike Leach 
Cc: Kajol Jain 
Cc: Alexander Shishkin 
Cc: Kan Liang 
Cc: Athira Rajeev 
Cc: linux-arm-kernel@lists.infradead.org
Cc: Ming Wang 
Cc: John Garry 
Cc: Ingo Molnar 
Cc: linux-kernel@vger.kernel.org
Cc: linux-perf-users@vger.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf pmu: Parse sysfs events directly from a file

2023-08-24T14:02:59+00:00

Rather than read a sysfs events file into a 256 byte char buffer, pass
the FILE* directly to the lex/yacc parser.

This avoids there being a maximum events file size.

While changing the API, constify some arguments to remove unnecessary
casts.

Allocating the read buffer decreases the performance of pmu-scan by
around 3%.

Signed-off-by: Ian Rogers 
Cc: Adrian Hunter 
Cc: Alexander Shishkin 
Cc: Gaosheng Cui 
Cc: Ingo Molnar 
Cc: James Clark 
Cc: Jing Zhang 
Cc: Jiri Olsa 
Cc: John Garry 
Cc: Kajol Jain 
Cc: Kan Liang 
Cc: Mark Rutland 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Ravi Bangoria 
Cc: Rob Herring 
Link: https://lore.kernel.org/r/20230824041330.266337-10-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo

perf pmu: Avoid passing format list to perf_pmu__format_bits()

2023-08-23T17:27:34+00:00

Pass the PMU so the format list can be better abstracted and later
lazily loaded.

Signed-off-by: Ian Rogers 
Cc: Adrian Hunter 
Cc: Alexander Shishkin 
Cc: Gaosheng Cui 
Cc: Ingo Molnar 
Cc: James Clark 
Cc: Jing Zhang 
Cc: Jiri Olsa 
Cc: John Garry 
Cc: Kajol Jain 
Cc: Kan Liang 
Cc: Mark Rutland 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Ravi Bangoria 
Cc: Rob Herring 
Link: https://lore.kernel.org/r/20230823080828.1460376-8-irogers@google.com
[ Did missing conversions in tools/perf/arch/arm*/util/cs-etm.c ]
Signed-off-by: Arnaldo Carvalho de Melo

perf pmu: Avoid passing format list to perf_pmu__config_terms()

2023-08-23T11:49:35+00:00

Abstract the format list better, hiding it in the PMU, by changing
perf_pmu__config_terms() the PMU rather than the format list in the PMU.

Change the PMU test to pass a dummy PMU for this purpose. Changing the
test allows perf_pmu__del_formats() to become static.

Signed-off-by: Ian Rogers 
Cc: Adrian Hunter 
Cc: Alexander Shishkin 
Cc: Gaosheng Cui 
Cc: Ingo Molnar 
Cc: James Clark 
Cc: Jing Zhang 
Cc: Jiri Olsa 
Cc: John Garry 
Cc: Kajol Jain 
Cc: Kan Liang 
Cc: Mark Rutland 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Ravi Bangoria 
Cc: Rob Herring 
Link: https://lore.kernel.org/r/20230823080828.1460376-6-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo

perf jevents: Add a new expression builtin strcmp_cpuid_str()

2023-08-17T17:12:14+00:00

This will allow writing formulas that are conditional on a specific
CPU type or CPU version. It calls through to the existing
strcmp_cpuid_str() function in Perf which has a default weak version,
and an arch specific version for x86 and arm64.

The function takes an 'ID' type value, which is a string. But in this
case Arm CPU IDs are hex numbers prefixed with '0x'. metric.py
assumes strings are only used by event names, and that they can't start
with a number ('0'), so an additional change has to be made to the
regex to convert hex numbers back to 'ID' types.

Signed-off-by: James Clark 
Reviewed-by: John Garry 
Cc: Adrian Hunter 
Cc: Alexander Shishkin 
Cc: Eduard Zingerman 
Cc: Haixin Yu 
Cc: Ian Rogers 
Cc: Ingo Molnar 
Cc: Jing Zhang 
Cc: Jiri Olsa 
Cc: Kajol Jain 
Cc: Kan Liang 
Cc: Leo Yan 
Cc: Mark Rutland 
Cc: Mike Leach 
Cc: Namhyung Kim 
Cc: Nick Forrington 
Cc: Peter Zijlstra 
Cc: Rob Herring 
Cc: Sohom Datta 
Cc: Will Deacon 
Cc: linux-arm-kernel@lists.infradead.org
Link: https://lore.kernel.org/r/20230816114841.1679234-5-james.clark@arm.com
Signed-off-by: Arnaldo Carvalho de Melo

perf test: Add a test for the new Arm CPU ID comparison behavior

2023-08-17T17:11:54+00:00

Now that variant and revision fields are taken into account the behavior
is slightly more complicated so add a test to ensure that this behaves
as expected.

Reviewed-by: John Garry 
Signed-off-by: James Clark 
Cc: Adrian Hunter 
Cc: Alexander Shishkin 
Cc: Eduard Zingerman 
Cc: Haixin Yu 
Cc: Ian Rogers 
Cc: Ingo Molnar 
Cc: Jing Zhang 
Cc: Jiri Olsa 
Cc: Kajol Jain 
Cc: Kan Liang 
Cc: Leo Yan 
Cc: Mark Rutland 
Cc: Mike Leach 
Cc: Namhyung Kim 
Cc: Nick Forrington 
Cc: Peter Zijlstra 
Cc: Rob Herring 
Cc: Sohom Datta 
Cc: Will Deacon 
Cc: linux-arm-kernel@lists.infradead.org
Link: https://lore.kernel.org/r/20230816114841.1679234-3-james.clark@arm.com
Signed-off-by: Arnaldo Carvalho de Melo

perf arm64: Allow version comparisons of CPU IDs

2023-08-17T17:11:37+00:00

Currently variant and revision fields are masked out of the MIDR so
it's not possible to compare different versions of the same CPU.
In a later commit a workaround will be removed just for N2 r0p3, so
enable comparisons on version.

This has the side effect of changing the MIDR stored in the header of
the perf.data file to no longer have masked version fields. It also
affects the lookups in mapfile.csv, but as that currently only has
zeroed version fields, it has no actual effect. The mapfile.csv
documentation also states to zero the version fields, so unless this
isn't done it will continue to have no effect.

There is an existing weak default strcmp_cpuid_str() function, and an
x86 version. This adds another version for arm64.

Signed-off-by: James Clark 
Reviewed-by: John Garry 
Cc: Adrian Hunter 
Cc: Alexander Shishkin 
Cc: Andrii Nakryiko 
Cc: Eduard Zingerman 
Cc: Haixin Yu 
Cc: Ian Rogers 
Cc: Ingo Molnar 
Cc: Jing Zhang 
Cc: Jiri Olsa 
Cc: Kajol Jain 
Cc: Kan Liang 
Cc: Leo Yan 
Cc: Mark Rutland 
Cc: Mike Leach 
Cc: Namhyung Kim 
Cc: Nick Forrington 
Cc: Peter Zijlstra 
Cc: Rob Herring 
Cc: Sohom Datta 
Cc: Will Deacon 
Cc: linux-arm-kernel@lists.infradead.org
Link: https://lore.kernel.org/r/20230816114841.1679234-2-james.clark@arm.com
Signed-off-by: Arnaldo Carvalho de Melo

perf parse-regs: Move out arch specific header from util/perf_regs.h

2023-08-16T11:52:45+00:00

util/perf_regs.h includes another perf_regs.h:

  #include 

Here it includes architecture specific header, for example, if we build
arm64 target, the header tools/perf/arch/arm64/include/perf_regs.h is
included.

We use this implicit way to include architecture specific header, which
is not directive; furthermore, util/perf_regs.c is coupled with the
architecture specific definitions.

This patch moves out arch specific header from util/perf_regs.h for
generalizing the 'util' folder, as a result, the source files in 'arch'
folder explicitly include architecture's perf_regs.h.

Signed-off-by: Leo Yan 
Acked-by: Ian Rogers 
Cc: Adrian Hunter 
Cc: Albert Ou 
Cc: Alexander Shishkin 
Cc: Eric Lin 
Cc: Fangrui Song 
Cc: Guo Ren 
Cc: Huacai Chen 
Cc: Ingo Molnar 
Cc: Ivan Babrou 
Cc: James Clark 
Cc: Jiri Olsa 
Cc: John Garry 
Cc: Kan Liang 
Cc: Mark Rutland 
Cc: Mike Leach 
Cc: Ming Wang 
Cc: Namhyung Kim 
Cc: Palmer Dabbelt 
Cc: Paul Walmsley 
Cc: Peter Zijlstra 
Cc: Sandipan Das 
Cc: Will Deacon 
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-csky@vger.kernel.org
Cc: linux-riscv@lists.infradead.org
Link: https://lore.kernel.org/r/20230606014559.21783-7-leo.yan@linaro.org
Signed-off-by: Arnaldo Carvalho de Melo

perf parse-regs: Remove PERF_REGS_{MAX|MASK} from common code

2023-08-16T11:49:10+00:00

The macros PERF_REGS_MAX and PERF_REGS_MASK are architecture specific,
let's remove them from the common file util/perf_regs.c.

As a side effect, the weak functions arch__intr_reg_mask() and
arch__user_reg_mask() just return zeros, every arch defines its own
functions in the 'arch' folder for returning right values.

Note, we don't need to return intr/user register masks dynamically, this
is because these two functions are invoked during recording phase but
not decoding phase, they are always invoked on the native environment,
thus we don't need to parse them dynamically.

Signed-off-by: Leo Yan 
Acked-by: Ian Rogers 
Cc: Adrian Hunter 
Cc: Albert Ou 
Cc: Alexander Shishkin 
Cc: Eric Lin 
Cc: Fangrui Song 
Cc: Guo Ren 
Cc: Huacai Chen 
Cc: Ingo Molnar 
Cc: Ivan Babrou 
Cc: James Clark 
Cc: Jiri Olsa 
Cc: John Garry 
Cc: Kan Liang 
Cc: Mark Rutland 
Cc: Mike Leach 
Cc: Ming Wang 
Cc: Namhyung Kim 
Cc: Palmer Dabbelt 
Cc: Paul Walmsley 
Cc: Peter Zijlstra 
Cc: Sandipan Das 
Cc: Will Deacon 
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-csky@vger.kernel.org
Cc: linux-riscv@lists.infradead.org
Link: https://lore.kernel.org/r/20230606014559.21783-6-leo.yan@linaro.org
Signed-off-by: Arnaldo Carvalho de Melo