Update to C11, fix signal undefined behavior
Ian Rogers
reviewercover Update to C11, fix signal undefined behavior
The use of C11 is mainstream in the kernel [1]. There was some confusion on volatile and signal handlers in [2]. Switch to using stdatomic.h (requires C11) and sig_atomic_t as per [3]. Thanks to Leo Yan <[email protected]> for the suggestions.
[1] https://lore.kernel.org/lkml/CAHk-=whWbENRz-vLY6vpESDLj6kGUTKO3khGtVfipHqwewh2HQ@mail.gmail.com/
[2] https://lore.kernel.org/lkml/[email protected]/
[3] https://wiki.sei.cmu.edu/confluence/display/c/SIG31-C.+Do+not+access+shared+objects+in+signal+handlers
Ian Rogers (8):
perf build: Update to C standard to gnu11 perf record: Use sig_atomic_t for signal handlers perf daemon: Use sig_atomic_t to avoid UB perf ftrace: Use sig_atomic_t to avoid UB perf session: Change type to avoid UB perf stat: Use sig_atomic_t to avoid UB perf top: Use sig_atomic_t to avoid UB perf trace: Use sig_atomic_t to avoid UB
tools/perf/Makefile.config | 2 +- tools/perf/builtin-daemon.c | 3 ++- tools/perf/builtin-ftrace.c | 4 ++-- tools/perf/builtin-record.c | 9 +++++---- tools/perf/builtin-stat.c | 9 +++++---- tools/perf/builtin-top.c | 4 ++-- tools/perf/builtin-trace.c | 4 ++-- tools/perf/util/session.c | 3 ++- 8 files changed, 21 insertions(+), 17 deletions(-)
Signature · 1 line
2.38.0.135.g90850a2211-goog
Ian Rogers
reviewerPATCH 1/8 perf build: Update to C standard to gnu11
C11 has become the standard for mainstream kernel development [1], allowing it in the perf build enables libraries like stdatomic.h to be assumed to be present. This came up in the context of [2].
[1] https://lore.kernel.org/lkml/CAHk-=whWbENRz-vLY6vpESDLj6kGUTKO3khGtVfipHqwewh2HQ@mail.gmail.com/
[2] https://lore.kernel.org/lkml/[email protected]/
--- tools/perf/Makefile.config | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/Makefile.config +1 -1
@@ -307,7 +307,7 @@ CORE_CFLAGS += -ggdb3 CORE_CFLAGS += -funwind-tables CORE_CFLAGS += -Wall CORE_CFLAGS += -Wextra -CORE_CFLAGS += -std=gnu99 +CORE_CFLAGS += -std=gnu11 CXXFLAGS += -std=gnu++14 -fno-exceptions -fno-rtti CXXFLAGS += -Wall
Signature · 1 line
2.38.0.135.g90850a2211-goog
Ian Rogers
reviewerPATCH 2/8 perf record: Use sig_atomic_t for signal handlers
In reply to Ian RogersThis removes undefined behavior as described in:
https://wiki.sei.cmu.edu/confluence/display/c/SIG31-C.+Do+not+access+shared+objects+in+signal+handlers
--- tools/perf/builtin-record.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)
tools/perf/builtin-record.c +5 -4
@@ -61,6 +61,7 @@#include <locale.h> #include <poll.h> #include <pthread.h> +#include <stdatomic.h> #include <unistd.h> #ifndef HAVE_GETTID #include <syscall.h>
@@ -646,10 +647,10 @@ static int record__pushfn(struct mmap *map, void *to, void *bf, size_t size)return record__write (rec, map, bf, size); } -static volatile int signr =-1 ; -static volatile int child_finished; +static volatile sig_atomic_t signr =-1 ; +static volatile sig_atomic_t child_finished;#ifdef HAVE_EVENTFD_SUPPORT -static volatile int done_fd =-1 ; +static volatile sig_atomic_t done_fd =-1 ;#endif static void sig_handler (int sig)
@@ -1926,7 +1927,7 @@ static void record__read_lost_samples(struct record *rec) } -static volatile int workload_exec_errno; +static volatile sig_atomic_t workload_exec_errno;/* * evlist__prepare_workload will send a SIGUSR1
Signature · 1 line
2.38.0.135.g90850a2211-goog
Ian Rogers
reviewerPATCH 3/8 perf daemon: Use sig_atomic_t to avoid UB
In reply to Ian RogersUse sig_atomic_t for a variable written to in a signal handler and read elsewhere. This is undefined behavior as per:
https://wiki.sei.cmu.edu/confluence/display/c/SIG31-C.+Do+not+access+shared+objects+in+signal+handlers
--- tools/perf/builtin-daemon.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
tools/perf/builtin-daemon.c +2 -1
@@ -9,6 +9,7 @@#include <string.h> #include <sys/file.h> #include <signal.h> +#include <stdatomic.h> #include <stdlib.h> #include <time.h> #include <stdio.h>
@@ -105,7 +106,7 @@ static const char * const daemon_usage[] = { NULL }; -static bool done; +static volatile sig_atomic_t done;static void sig_handler (int sig__maybe_unused ) {
Signature · 1 line
2.38.0.135.g90850a2211-goog
Ian Rogers
reviewerPATCH 4/8 perf ftrace: Use sig_atomic_t to avoid UB
In reply to Ian RogersUse sig_atomic_t for a variable written to in a signal handler and read elsewhere. This is undefined behavior as per:
https://wiki.sei.cmu.edu/confluence/display/c/SIG31-C.+Do+not+access+shared+objects+in+signal+handlers
--- tools/perf/builtin-ftrace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/builtin-ftrace.c +2 -2
@@ -36,8 +36,8 @@#define DEFAULT_TRACER "function_graph" -static volatile int workload_exec_errno; -static bool done; +static volatile sig_atomic_t workload_exec_errno; +static volatile sig_atomic_t done;static void sig_handler (int sig__maybe_unused ) {
Signature · 1 line
2.38.0.135.g90850a2211-goog
Ian Rogers
reviewerPATCH 5/8 perf session: Change type to avoid UB
In reply to Ian Rogerssession_done is written to inside the signal handler of perf report and script. Switch its type to avoid undefined behavior.
--- tools/perf/util/session.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
tools/perf/util/session.c +2 -1
@@ -1,5 +1,6 @@// SPDX-License-Identifier: GPL-2.0 #include <errno.h> +#include <stdatomic.h> #include <inttypes.h> #include <linux/err.h> #include <linux/kernel.h>
@@ -2022,7 +2023,7 @@ static int perf_session__flush_thread_stacks(struct perf_session *session) NULL); } -volatile int session_done; +volatile sig_atomic_t session_done;static int __perf_session__process_decomp_events (struct perf_session *session);
Signature · 1 line
2.38.0.135.g90850a2211-goog
Ian Rogers
reviewerPATCH 6/8 perf stat: Use sig_atomic_t to avoid UB
In reply to Ian RogersUse sig_atomic_t for variables written/accessed in signal handlers. This is undefined behavior as per:
https://wiki.sei.cmu.edu/confluence/display/c/SIG31-C.+Do+not+access+shared+objects+in+signal+handlers
--- tools/perf/builtin-stat.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)
tools/perf/builtin-stat.c +5 -4
@@ -83,6 +83,7 @@#include <inttypes.h> #include <locale.h> #include <math.h> +#include <stdatomic.h> #include <sys/types.h> #include <sys/stat.h> #include <sys/wait.h>
@@ -173,7 +174,7 @@ static struct target target = {#define METRIC_ONLY_LEN 20 -static volatile pid_t child_pid =-1 ; +static volatile sig_atomic_t child_pid =-1 ;static int detailed_run =0 ;static bool transaction_run;static bool topdown_run = false;
@@ -208,7 +209,7 @@ struct perf_stat {static struct perf_stat perf_stat;#define STAT_RECORD perf_stat.record -static volatile int done =0 ; +static volatile sig_atomic_t done =0 ;static struct perf_stat_config stat_config = { .aggr_mode =AGGR_GLOBAL ,
@@ -569,7 +570,7 @@ static void disable_counters(void) } } -static volatile int workload_exec_errno; +static volatile sig_atomic_t workload_exec_errno;/* * evlist__prepare_workload will send a SIGUSR1
@@ -1029,7 +1030,7 @@ static void print_counters(struct timespec *ts, int argc, const char **argv)evlist__print_counters (evsel_list, &stat_config, &target, ts, argc, argv); } -static volatile int signr =-1 ; +static volatile sig_atomic_t signr =-1 ;static void skip_signal (int signo) {
Signature · 1 line
2.38.0.135.g90850a2211-goog
Ian Rogers
reviewerPATCH 7/8 perf top: Use sig_atomic_t to avoid UB
In reply to Ian RogersUse sig_atomic_t for variables written/accessed in signal handlers. This is undefined behavior as per:
https://wiki.sei.cmu.edu/confluence/display/c/SIG31-C.+Do+not+access+shared+objects+in+signal+handlers
--- tools/perf/builtin-top.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/builtin-top.c +2 -2
@@ -87,8 +87,8 @@#include <linux/ctype.h> #include <perf/mmap.h> -static volatile int done; -static volatile int resize; +static volatile sig_atomic_t done; +static volatile sig_atomic_t resize;#define HEADER_LINE_NR 5
Signature · 1 line
2.38.0.135.g90850a2211-goog
Ian Rogers
reviewerPATCH 8/8 perf trace: Use sig_atomic_t to avoid UB
In reply to Ian RogersUse sig_atomic_t for variables written/accessed in signal handlers. This is undefined behavior as per:
https://wiki.sei.cmu.edu/confluence/display/c/SIG31-C.+Do+not+access+shared+objects+in+signal+handlers
--- tools/perf/builtin-trace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/builtin-trace.c +2 -2
@@ -1535,8 +1535,8 @@ static size_t trace__fprintf_tstamp(struct trace *trace, u64 tstamp, FILE *fp) }static pid_t workload_pid =-1 ; -static bool done = false; -static bool interrupted = false; +static volatile sig_atomic_t done = false; +static volatile sig_atomic_t interrupted = false;static void sighandler_interrupt (int sig__maybe_unused ) {
Signature · 1 line
2.38.0.135.g90850a2211-goog
Arnaldo Carvalho de Melo
maintainerEm Mon, Oct 24, 2022 at 10:35:15AM -0700, Ian Rogers escreveu:
Quoted 8 lines
The use of C11 is mainstream in the kernel [1]. There was some confusion on volatile and signal handlers in [2]. Switch to using
stdatomic.h(requires C11) andsig_atomic_tas per [3]. Thanks to Leo Yan <[email protected]> for the suggestions.[1] https://lore.kernel.org/lkml/CAHk-=whWbENRz-vLY6vpESDLj6kGUTKO3khGtVfipHqwewh2HQ@mail.gmail.com/
[2] https://lore.kernel.org/lkml/[email protected]/
[3] https://wiki.sei.cmu.edu/confluence/display/c/SIG31-C.+Do+not+access+shared+objects+in+signal+handlers
I think I'll apply this to perf/core, i.e. for 6.3, ok?
- Arnaldo
Quoted 23 lines
Ian Rogers (8):
perf build: Update to C standard to gnu11 perf record: Usesig_atomic_tfor signal handlers perf daemon: Usesig_atomic_tto avoid UB perf ftrace: Usesig_atomic_tto avoid UB perf session: Change type to avoid UB perf stat: Usesig_atomic_tto avoid UB perf top: Usesig_atomic_tto avoid UB perf trace: Usesig_atomic_tto avoid UBtools/perf/Makefile.config | 2 +- tools/perf/builtin-daemon.c | 3 ++- tools/perf/builtin-ftrace.c | 4 ++-- tools/perf/builtin-record.c | 9 +++++---- tools/perf/builtin-stat.c | 9 +++++---- tools/perf/builtin-top.c | 4 ++-- tools/perf/builtin-trace.c | 4 ++-- tools/perf/util/session.c | 3 ++- 8 files changed, 21 insertions(+), 17 deletions(-)Signature · 1 line
2.38.0.135.g90850a2211-goog
Signature · 1 line
- Arnaldo
Ian Rogers
reviewerQuoted 7 lines · Arnaldo Carvalho de Melo
Em Mon, Oct 24, 2022 at 10:35:15AM -0700, Ian Rogers escreveu:
Quoted 5 lines
The use of C11 is mainstream in the kernel [1]. There was some confusion on volatile and signal handlers in [2]. Switch to using
stdatomic.h(requires C11) andsig_atomic_tas per [3]. Thanks to Leo Yan <[email protected]> for the suggestions.
Quoted 5 lines
Quoted 3 lines
[1] https://lore.kernel.org/lkml/CAHk-=whWbENRz-vLY6vpESDLj6kGUTKO3khGtVfipHqwewh2HQ@mail.gmail.com/
[2] https://lore.kernel.org/lkml/[email protected]/
[3] https://wiki.sei.cmu.edu/confluence/display/c/SIG31-C.+Do+not+access+shared+objects+in+signal+handlersI think I'll apply this to perf/core, i.e. for 6.3, ok?
Sounds good to me. 6.3 or 6.2? I suspect there is more cleanup like this and to the iterators (from C11) that can be done.
Thanks,
Ian
Quoted 29 lines
- Arnaldo
Quoted 21 lines
Ian Rogers (8):
perf build: Update to C standard to gnu11 perf record: Usesig_atomic_tfor signal handlers perf daemon: Usesig_atomic_tto avoid UB perf ftrace: Usesig_atomic_tto avoid UB perf session: Change type to avoid UB perf stat: Usesig_atomic_tto avoid UB perf top: Usesig_atomic_tto avoid UB perf trace: Usesig_atomic_tto avoid UBtools/perf/Makefile.config | 2 +- tools/perf/builtin-daemon.c | 3 ++- tools/perf/builtin-ftrace.c | 4 ++-- tools/perf/builtin-record.c | 9 +++++---- tools/perf/builtin-stat.c | 9 +++++---- tools/perf/builtin-top.c | 4 ++-- tools/perf/builtin-trace.c | 4 ++-- tools/perf/util/session.c | 3 ++- 8 files changed, 21 insertions(+), 17 deletions(-)Quoted 2 lines
--
2.38.0.135.g90850a2211-goog--
- Arnaldo
Ian Rogers
reviewerQuoted 14 lines · Ian Rogers
Quoted 11 lines · Arnaldo Carvalho de Melo
Em Mon, Oct 24, 2022 at 10:35:15AM -0700, Ian Rogers escreveu:
Quoted 8 lines
The use of C11 is mainstream in the kernel [1]. There was some confusion on volatile and signal handlers in [2]. Switch to using
stdatomic.h(requires C11) andsig_atomic_tas per [3]. Thanks to Leo Yan <[email protected]> for the suggestions.[1] https://lore.kernel.org/lkml/CAHk-=whWbENRz-vLY6vpESDLj6kGUTKO3khGtVfipHqwewh2HQ@mail.gmail.com/
[2] https://lore.kernel.org/lkml/[email protected]/
[3] https://wiki.sei.cmu.edu/confluence/display/c/SIG31-C.+Do+not+access+shared+objects+in+signal+handlers
Quoted 7 lines
Quoted 1 line
I think I'll apply this to perf/core, i.e. for 6.3, ok?
Sounds good to me. 6.3 or 6.2? I suspect there is more cleanup like this and to the iterators (from C11) that can be done.
Thanks,
Ian
So I noticed a few changes missing #include-ing stdatomic.h and sig_atomic_t is actually in signal.h. I'm not sure we need the C11 change then, but it seems like the right thing to do anyway. I'll do a v2 to drop the unneeded (currently) include of stdatomic.h.
Thanks,
Ian
Quoted 29 lines
Quoted 29 lines
- Arnaldo
Quoted 21 lines
Ian Rogers (8):
perf build: Update to C standard to gnu11 perf record: Usesig_atomic_tfor signal handlers perf daemon: Usesig_atomic_tto avoid UB perf ftrace: Usesig_atomic_tto avoid UB perf session: Change type to avoid UB perf stat: Usesig_atomic_tto avoid UB perf top: Usesig_atomic_tto avoid UB perf trace: Usesig_atomic_tto avoid UBtools/perf/Makefile.config | 2 +- tools/perf/builtin-daemon.c | 3 ++- tools/perf/builtin-ftrace.c | 4 ++-- tools/perf/builtin-record.c | 9 +++++---- tools/perf/builtin-stat.c | 9 +++++---- tools/perf/builtin-top.c | 4 ++-- tools/perf/builtin-trace.c | 4 ++-- tools/perf/util/session.c | 3 ++- 8 files changed, 21 insertions(+), 17 deletions(-)Quoted 2 lines
--
2.38.0.135.g90850a2211-goog--
- Arnaldo
From: Ian Rogers
Quoted 24 lines
Sent: 24 October 2022 19:12
Quoted 11 lines · Ian Rogers
Quoted 8 lines · Arnaldo Carvalho de Melo
Em Mon, Oct 24, 2022 at 10:35:15AM -0700, Ian Rogers escreveu:
Quoted 6 lines
The use of C11 is mainstream in the kernel [1]. There was some confusion on volatile and signal handlers in [2]. Switch to using
stdatomic.h(requires C11) andsig_atomic_tas per [3]. Thanks to Leo Yan <[email protected]> for the suggestions.Quoted 2 lines
Quoted 2 lines
C.+Do+not+access+shared+objects+in+signal+handlers
Quoted 6 lines
Quoted 2 lines
I think I'll apply this to perf/core, i.e. for 6.3, ok?
Sounds good to me. 6.3 or 6.2? I suspect there is more cleanup like this and to the iterators (from C11) that can be done.
Quoted 7 lines
Quoted 2 lines
Thanks,
IanSo I noticed a few changes missing #include-ing
stdatomic.handsig_atomic_tis actually insignal.h. I'm not sure we need the C11 change then, but it seems like the right thing to do anyway. I'll do a v2 to drop the unneeded (currently) include ofstdatomic.h.
While the C standard requires you use sig_atomic_t (to avoid wider RMW being done for writes) the kernel very much requires that volatiles accesses are atomic.
So the compilers used will do that even when the C standard would allow them to do otherwise.
Pretty much the last mainstream cpu that couldn't do atomic byte writes was an old alpha.
So for anything that Linux is going to run on these changes aren't really needed.
David
-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK Registration No: 1397386 (Wales)
On Tue, Oct 25, 2022 at 10:36:16AM +0000, David Laight wrote:
[...]
Quoted 5 lines
Quoted 4 lines
So I noticed a few changes missing #include-ing
stdatomic.handsig_atomic_tis actually insignal.h. I'm not sure we need the C11 change then, but it seems like the right thing to do anyway. I'll do a v2 to drop the unneeded (currently) include ofstdatomic.h.
Quoted 8 lines
While the C standard requires you use
sig_atomic_t(to avoid wider RMW being done for writes) the kernel very much requires that volatiles accesses are atomic.
So the compilers used will do that even when the C standard would allow them to do otherwise.Pretty much the last mainstream cpu that couldn't do atomic byte writes was an old alpha.
One case that it might break atomicity for the data with int type, I.e. on Arm CPU when CPU access data without alignment, then it cannot promise atomicity. I am not sure if the data with int type will be always compiled with 4-byte alignment, if it's packed without alignment then sig_atomic_t is still required.
Clarify that I don't have deep understanding for compiler, so sorry if I introduce any noise.
Thanks,
Leo
Quoted 9 lines
So for anything that Linux is going to run on these changes aren't really needed.
David
-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK Registration No: 1397386 (Wales)
Arnaldo Carvalho de Melo
maintainerEm Mon, Oct 24, 2022 at 10:59:03AM -0700, Ian Rogers escreveu:
Quoted 9 lines
Quoted 7 lines · Arnaldo Carvalho de Melo
Em Mon, Oct 24, 2022 at 10:35:15AM -0700, Ian Rogers escreveu:
Quoted 5 lines
The use of C11 is mainstream in the kernel [1]. There was some confusion on volatile and signal handlers in [2]. Switch to using
stdatomic.h(requires C11) andsig_atomic_tas per [3]. Thanks to Leo Yan <[email protected]> for the suggestions.
Quoted 8 lines
Quoted 5 lines
Quoted 3 lines
[1] https://lore.kernel.org/lkml/CAHk-=whWbENRz-vLY6vpESDLj6kGUTKO3khGtVfipHqwewh2HQ@mail.gmail.com/
[2] https://lore.kernel.org/lkml/[email protected]/
[3] https://wiki.sei.cmu.edu/confluence/display/c/SIG31-C.+Do+not+access+shared+objects+in+signal+handlersI think I'll apply this to perf/core, i.e. for 6.3, ok?
Sounds good to me. 6.3 or 6.2? I suspect there is more cleanup like this and to the iterators (from C11) that can be done.
oops, 6.2, sure, 6.1 is the current one, merge window closed. :-)
- Arnaldo