From: "Michal Koutný" <mkoutny@suse.com>
To: linux-kernel@vger.kernel.org, linux-trace-kernel@vger.kernel.org
Cc: "Steven Rostedt" <rostedt@goodmis.org>,
"Masami Hiramatsu" <mhiramat@kernel.org>,
"Mathieu Desnoyers" <mathieu.desnoyers@efficios.com>,
"Christian Brauner" <brauner@kernel.org>,
"Oleg Nesterov" <oleg@redhat.com>,
"Kent Overstreet" <kent.overstreet@linux.dev>,
"Kees Cook" <keescook@chromium.org>,
"Michal Koutný" <mkoutny@suse.com>,
"Andrew Morton" <akpm@linux-foundation.org>,
"Tycho Andersen" <tandersen@netflix.com>,
"Jens Axboe" <axboe@kernel.dk>,
"Aleksa Sarai" <cyphar@cyphar.com>
Subject: [PATCH 1/3] tracing: Remove dependency of saved_cmdlines_buffer on PID_MAX_DEFAULT
Date: Mon, 8 Apr 2024 16:58:17 +0200 [thread overview]
Message-ID: <20240408145819.8787-2-mkoutny@suse.com> (raw)
In-Reply-To: <20240408145819.8787-1-mkoutny@suse.com>
Calculations into map_pid_to_cmdline use PID_MAX_DEFAULT but they
actually depend on the size of map_pid_to_cmdline. The size of the map
may be arbitrary. First, refer to the map size where necessary, second,
pick a good value for the size of the map.
Since the buffer is allocated at boot (i.e. user cannot affect its size
later), accounting for full PID_MAX_LIMIT would inflate map's size
unnecessarily (4*4M) for all users. Stick to the original value of
4*32k, the commit 785e3c0a3a87 ("tracing: Map all PIDs to command
lines") explains why it still works for higher pids.
The point of this exercise is to remove dependency on PID_MAX_DEFAULT.
Signed-off-by: Michal Koutný <mkoutny@suse.com>
---
kernel/trace/trace_sched_switch.c | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c
index 8a407adb0e1c..aca2dafdd97a 100644
--- a/kernel/trace/trace_sched_switch.c
+++ b/kernel/trace/trace_sched_switch.c
@@ -161,6 +161,7 @@ static size_t tgid_map_max;
#define SAVED_CMDLINES_DEFAULT 128
#define NO_CMDLINE_MAP UINT_MAX
+#define PID_MAP_SIZE (CONFIG_BASE_SMALL ? 0x1000 : 0x8000)
/*
* Preemption must be disabled before acquiring trace_cmdline_lock.
* The various trace_arrays' max_lock must be acquired in a context
@@ -168,7 +169,7 @@ static size_t tgid_map_max;
*/
static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED;
struct saved_cmdlines_buffer {
- unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1];
+ unsigned map_pid_to_cmdline[PID_MAP_SIZE];
unsigned *map_cmdline_to_pid;
unsigned cmdline_num;
int cmdline_idx;
@@ -248,7 +249,7 @@ int trace_save_cmdline(struct task_struct *tsk)
if (!tsk->pid)
return 1;
- tpid = tsk->pid & (PID_MAX_DEFAULT - 1);
+ tpid = tsk->pid % PID_MAP_SIZE;
/*
* It's not the end of the world if we don't get
@@ -294,7 +295,7 @@ static void __trace_find_cmdline(int pid, char comm[])
return;
}
- tpid = pid & (PID_MAX_DEFAULT - 1);
+ tpid = pid % PID_MAP_SIZE;
map = savedcmd->map_pid_to_cmdline[tpid];
if (map != NO_CMDLINE_MAP) {
tpid = savedcmd->map_cmdline_to_pid[map];
@@ -645,8 +646,8 @@ tracing_saved_cmdlines_size_write(struct file *filp, const char __user *ubuf,
if (ret)
return ret;
- /* must have at least 1 entry or less than PID_MAX_DEFAULT */
- if (!val || val > PID_MAX_DEFAULT)
+ /* must have at least 1 entry or fit into map_pid_to_cmdline */
+ if (!val || val >= PID_MAP_SIZE)
return -EINVAL;
ret = tracing_resize_saved_cmdlines((unsigned int)val);
--
2.44.0
next prev parent reply other threads:[~2024-04-08 14:58 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-04-08 14:58 [PATCH 0/3] kernel/pid: Remove default pid_max value Michal Koutný
2024-04-08 14:58 ` Michal Koutný [this message]
2024-04-09 15:01 ` [PATCH 1/3] tracing: Remove dependency of saved_cmdlines_buffer on PID_MAX_DEFAULT Steven Rostedt
2024-05-13 17:30 ` Michal Koutný
2024-04-08 14:58 ` [PATCH 2/3] kernel/pid: Remove default pid_max value Michal Koutný
2024-04-08 20:29 ` Andrew Morton
2024-04-11 15:40 ` Michal Koutný
2024-04-11 22:03 ` Andrew Morton
2024-04-12 14:32 ` Michal Koutný
2024-04-09 0:45 ` kernel test robot
2024-04-09 1:38 ` kernel test robot
2024-05-13 17:26 ` Michal Koutný
2024-04-08 14:58 ` [PATCH 3/3] tracing: Compare pid_max against pid_list capacity Michal Koutný
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240408145819.8787-2-mkoutny@suse.com \
--to=mkoutny@suse.com \
--cc=akpm@linux-foundation.org \
--cc=axboe@kernel.dk \
--cc=brauner@kernel.org \
--cc=cyphar@cyphar.com \
--cc=keescook@chromium.org \
--cc=kent.overstreet@linux.dev \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-trace-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mhiramat@kernel.org \
--cc=oleg@redhat.com \
--cc=rostedt@goodmis.org \
--cc=tandersen@netflix.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox