From: Steven Rostedt <rostedt@goodmis.org>
To: linux-kernel@vger.kernel.org
Cc: Ingo Molnar <mingo@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Jiri Olsa <jolsa@redhat.com>
Subject: [for-next][PATCH 30/40] ftrace/direct: Do not disable when switching direct callers
Date: Fri, 22 Oct 2021 16:48:26 -0400 [thread overview]
Message-ID: <20211022204843.717752020@goodmis.org> (raw)
In-Reply-To: 20211022204756.099054287@goodmis.org
From: "Steven Rostedt (VMware)" <rostedt@goodmis.org>
Currently to switch a set of "multi" direct trampolines from one
trampoline to another, a full shutdown of the current set needs to be
done, followed by an update to what trampoline the direct callers would
call, and then re-enabling the callers. This leaves a time when the
functions will not be calling anything, and events may be missed.
Instead, use a trick to allow all the functions with direct trampolines
attached will always call either the new or old trampoline while the
switch is happening. To do this, first attach a "dummy" callback via
ftrace to all the functions that the current direct trampoline is attached
to. This will cause the functions to call the "list func" instead of the
direct trampoline. The list function will call the direct trampoline
"helper" that will set the function it should call as it returns back to
the ftrace trampoline.
At this moment, the direct caller descriptor can safely update the direct
call trampoline. The list function will pick either the new or old
function (depending on the memory coherency model of the architecture).
Now removing the dummy function from each of the locations of the direct
trampoline caller, will put back the direct call, but now to the new
trampoline.
A better visual is:
[ Changing direct call from my_direct_1 to my_direct_2 ]
<traced_func>:
call my_direct_1
||||||||||||||||||||
vvvvvvvvvvvvvvvvvvvv
<traced_func>:
call ftrace_caller
<ftrace_caller>:
[..]
call ftrace_ops_list_func
ftrace_ops_list_func()
{
ops->func() -> direct_helper -> set rax to my_direct_1 or my_direct_2
}
call rax (to either my_direct_1 or my_direct_2
||||||||||||||||||||
vvvvvvvvvvvvvvvvvvvv
<traced_func>:
call my_direct_2
Link: https://lore.kernel.org/all/20211014162819.5c85618b@gandalf.local.home/
Acked-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
---
kernel/trace/ftrace.c | 34 ++++++++++++++++++++--------------
1 file changed, 20 insertions(+), 14 deletions(-)
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 30120342176e..f90ed00a6d5b 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -5561,8 +5561,12 @@ EXPORT_SYMBOL_GPL(unregister_ftrace_direct_multi);
*/
int modify_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr)
{
- struct ftrace_hash *hash = ops->func_hash->filter_hash;
+ struct ftrace_hash *hash;
struct ftrace_func_entry *entry, *iter;
+ static struct ftrace_ops tmp_ops = {
+ .func = ftrace_stub,
+ .flags = FTRACE_OPS_FL_STUB,
+ };
int i, size;
int err;
@@ -5572,21 +5576,22 @@ int modify_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr)
return -EINVAL;
mutex_lock(&direct_mutex);
- mutex_lock(&ftrace_lock);
+
+ /* Enable the tmp_ops to have the same functions as the direct ops */
+ ftrace_ops_init(&tmp_ops);
+ tmp_ops.func_hash = ops->func_hash;
+
+ err = register_ftrace_function(&tmp_ops);
+ if (err)
+ goto out_direct;
/*
- * Shutdown the ops, change 'direct' pointer for each
- * ops entry in direct_functions hash and startup the
- * ops back again.
- *
- * Note there is no callback called for @ops object after
- * this ftrace_shutdown call until ftrace_startup is called
- * later on.
+ * Now the ftrace_ops_list_func() is called to do the direct callers.
+ * We can safely change the direct functions attached to each entry.
*/
- err = ftrace_shutdown(ops, 0);
- if (err)
- goto out_unlock;
+ mutex_lock(&ftrace_lock);
+ hash = ops->func_hash->filter_hash;
size = 1 << hash->size_bits;
for (i = 0; i < size; i++) {
hlist_for_each_entry(iter, &hash->buckets[i], hlist) {
@@ -5597,10 +5602,11 @@ int modify_ftrace_direct_multi(struct ftrace_ops *ops, unsigned long addr)
}
}
- err = ftrace_startup(ops, 0);
+ /* Removing the tmp_ops will add the updated direct callers to the functions */
+ unregister_ftrace_function(&tmp_ops);
- out_unlock:
mutex_unlock(&ftrace_lock);
+ out_direct:
mutex_unlock(&direct_mutex);
return err;
}
--
2.33.0
next prev parent reply other threads:[~2021-10-22 20:50 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-22 20:47 [for-next][PATCH 00/40] tracing: Updates for 5.16 Steven Rostedt
2021-10-22 20:47 ` [for-next][PATCH 01/40] tracing: Initialize upper and lower vars in pid_list_refill_irq() Steven Rostedt
2021-10-22 20:47 ` [for-next][PATCH 02/40] tracefs: Have tracefs directories not set OTH permission bits by default Steven Rostedt
2021-10-22 20:47 ` [for-next][PATCH 03/40] tracing: Disable "other" permission bits in the tracefs files Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 04/40] ftrace: Cleanup ftrace_dyn_arch_init() Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 05/40] bootconfig: Allocate xbc_data inside xbc_init() Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 06/40] bootconfig: Add xbc_get_info() for the node information Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 07/40] tools/bootconfig: Run test script when build all Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 08/40] bootconfig: Rename xbc_destroy_all() to xbc_exit() Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 09/40] bootconfig: Split parse-tree part from xbc_init Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 10/40] bootconfig: Remove unused debug function Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 11/40] tools/bootconfig: Print all error message in stderr Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 12/40] bootconfig: Replace u16 and u32 with uint16_t and uint32_t Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 13/40] bootconfig: Cleanup dummy headers in tools/bootconfig Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 14/40] ftrace: Add unit test for removing trace function Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 15/40] tracing: in_irq() cleanup Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 16/40] tracing: Use linker magic instead of recasting ftrace_ops_list_func() Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 17/40] tracing/cfi: Fix cmp_entries_* functions signature mismatch Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 18/40] tracing: Reuse logic from perfs get_recursion_context() Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 19/40] tracing/perf: Add interrupt_context_level() helper Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 20/40] x86/ftrace: Remove extra orig rax move Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 21/40] x86/ftrace: Remove fault protection code in prepare_ftrace_return Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 22/40] ftrace/x86_64: Have function graph tracer depend on DYNAMIC_FTRACE Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 23/40] x86/ftrace: Make function graph use ftrace directly Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 24/40] tracing: Add trampoline/graph selftest Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 25/40] tracing: Fix selftest config check for function graph start up test Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 26/40] ftrace: Add ftrace_add_rec_direct function Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 27/40] ftrace: Add multi direct register/unregister interface Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 28/40] ftrace: Add multi direct modify interface Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 29/40] ftrace/samples: Add multi direct interface test module Steven Rostedt
2021-10-22 20:48 ` Steven Rostedt [this message]
2021-10-22 20:48 ` [for-next][PATCH 31/40] tracing: Explain the trace recursion transition bit better Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 32/40] tracing: use %ps format string to print symbols Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 33/40] kprobes: convert tests to kunit Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 34/40] x86/unwind: Compile kretprobe fixup code only if CONFIG_KRETPROBES=y Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 35/40] arm64: kprobes: Record frame pointer with kretprobe instance Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 36/40] arm64: kprobes: Make a frame pointer on __kretprobe_trampoline Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 37/40] arm64: Recover kretprobe modified return address in stacktrace Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 38/40] ARM: clang: Do not rely on lr register for stacktrace Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 39/40] ARM: kprobes: Make a frame pointer on __kretprobe_trampoline Steven Rostedt
2021-10-22 20:48 ` [for-next][PATCH 40/40] ARM: Recover kretprobe modified return address in stacktrace Steven Rostedt
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=20211022204843.717752020@goodmis.org \
--to=rostedt@goodmis.org \
--cc=akpm@linux-foundation.org \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.