From: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
To: peterz@infradead.org, mingo@redhat.com, acme@kernel.org,
alexander.shishkin@linux.intel.com, jolsa@redhat.com,
namhyung@kernel.org, linux-kernel@vger.kernel.org,
rostedt@goodmis.org, mhiramat@kernel.org,
ananth@linux.vnet.ibm.com, naveen.n.rao@linux.vnet.ibm.com,
srikar@linux.vnet.ibm.com, oleg@redhat.com
Cc: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Subject: [RFC 4/4] trace_uprobe: Fix multiple update of same semaphores
Date: Wed, 28 Feb 2018 13:23:45 +0530 [thread overview]
Message-ID: <20180228075345.674-5-ravi.bangoria@linux.vnet.ibm.com> (raw)
In-Reply-To: <20180228075345.674-1-ravi.bangoria@linux.vnet.ibm.com>
For tiny binaries/libraries, different mmap regions points to
the same file portion. In such cases, we may increment semaphore
multiple times. But while de-registration, semaphore will get
decremented only once, leaving semaphore > 0 even if no one is
tracing on that marker.
Ensure increment and decrement happens in sync by keeping list
of mms in trace_uprobe. Increment semaphore only if mm is not
present in the list and decrement only if mm is present in the
list.
Example
# echo "p:sdt_tick/loop2 /tmp/tick:0x6e4 *0x10036" > uprobe_events
Before patch:
# echo 1 > events/sdt_tick/loop2/enable
# ./Workspace/sdt_prog/tick &
# dd if=/proc//mem bs=1 count=1 skip=268566582 2>/dev/null | xxd
0000000: 02 .
# echo 0 > events/sdt_tick/loop2/enable
# dd if=/proc//mem bs=1 count=1 skip=268566582 2>/dev/null | xxd
0000000: 01 .
After patch:
# echo 1 > events/sdt_tick/loop2/enable
# ./Workspace/sdt_prog/tick &
# dd if=/proc//mem bs=1 count=1 skip=268566582 2>/dev/null | xxd
0000000: 01 .
# echo 0 > events/sdt_tick/loop2/enable
# dd if=/proc//mem bs=1 count=1 skip=268566582 2>/dev/null | xxd
0000000: 00 .
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
---
kernel/trace/trace_uprobe.c | 105 ++++++++++++++++++++++++++++++++++++++++++--
1 file changed, 102 insertions(+), 3 deletions(-)
diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c
index d14aafc..3f1e8bd 100644
--- a/kernel/trace/trace_uprobe.c
+++ b/kernel/trace/trace_uprobe.c
@@ -49,6 +49,11 @@ struct trace_uprobe_filter {
struct list_head perf_events;
};
+struct sdt_mm_list {
+ struct mm_struct *mm;
+ struct sdt_mm_list *next;
+};
+
/*
* uprobe event core functions
*/
@@ -60,6 +65,8 @@ struct trace_uprobe {
char *filename;
unsigned long offset;
unsigned long sdt_offset; /* sdt semaphore offset */
+ struct sdt_mm_list *sml;
+ struct rw_semaphore sml_rw_sem;
unsigned long nhit;
struct trace_probe tp;
};
@@ -273,6 +280,7 @@ static inline bool is_ret_probe(struct trace_uprobe *tu)
if (is_ret)
tu->consumer.ret_handler = uretprobe_dispatcher;
init_trace_uprobe_filter(&tu->filter);
+ init_rwsem(&tu->sml_rw_sem);
return tu;
error:
@@ -953,6 +961,75 @@ static bool sdt_valid_vma(struct trace_uprobe *tu, struct vm_area_struct *vma)
return 0;
}
+static bool sdt_check_mm_list(struct trace_uprobe *tu, struct mm_struct *mm)
+{
+ struct sdt_mm_list *tmp = tu->sml;
+
+ if (!tu->sml || !mm)
+ return false;
+
+ while (tmp) {
+ if (tmp->mm == mm)
+ return true;
+ tmp = tmp->next;
+ }
+
+ return false;
+}
+
+static void sdt_add_mm_list(struct trace_uprobe *tu, struct mm_struct *mm)
+{
+ struct sdt_mm_list *tmp;
+
+ tmp = kzalloc(sizeof(*tmp), GFP_KERNEL);
+ if (!tmp) {
+ pr_info("sdt_add_mm_list failed.\n");
+ return;
+ }
+ tmp->mm = mm;
+ tmp->next = tu->sml;
+ tu->sml = tmp;
+}
+
+static void sdt_del_mm_list(struct trace_uprobe *tu, struct mm_struct *mm)
+{
+ struct sdt_mm_list *prev, *curr;
+
+ if (!tu->sml)
+ return;
+
+ if (tu->sml->mm == mm) {
+ curr = tu->sml;
+ tu->sml = tu->sml->next;
+ kfree(curr);
+ return;
+ }
+
+ prev = tu->sml;
+ curr = tu->sml->next;
+ while (curr) {
+ if (curr->mm == mm) {
+ prev->next = curr->next;
+ kfree(curr);
+ return;
+ }
+ prev = curr;
+ curr = curr->next;
+ }
+}
+
+static void sdt_flush_mm_list(struct trace_uprobe *tu)
+{
+ struct sdt_mm_list *next, *curr = tu->sml;
+
+ while (curr) {
+ next = curr->next;
+ kfree(curr);
+ curr = next;
+ }
+ tu->sml = NULL;
+}
+
/*
* TODO: Adding this defination in include/linux/uprobes.h throws
* warnings about address_sapce. Adding it here for the time being.
@@ -970,20 +1047,26 @@ static void sdt_increment_sem(struct trace_uprobe *tu)
if (IS_ERR(info))
goto out;
+ down_write(&tu->sml_rw_sem);
while (info) {
down_write(&info->mm->mmap_sem);
vma = sdt_find_vma(info->mm, tu);
if (!vma)
goto cont;
+ if (sdt_check_mm_list(tu, info->mm))
+ goto cont;
+
vaddr = offset_to_vaddr(vma, tu->sdt_offset);
- sdt_update_sem(info->mm, vaddr, 1);
+ if (!sdt_update_sem(info->mm, vaddr, 1))
+ sdt_add_mm_list(tu, info->mm);
cont:
up_write(&info->mm->mmap_sem);
mmput(info->mm);
info = free_uprobe_map_info(info);
}
+ up_write(&tu->sml_rw_sem);
out:
uprobe_end_dup_mmap();
@@ -1001,8 +1084,16 @@ void trace_uprobe_mmap_callback(struct vm_area_struct *vma)
!trace_probe_is_enabled(&tu->tp))
continue;
+ down_write(&tu->sml_rw_sem);
+ if (sdt_check_mm_list(tu, vma->vm_mm))
+ goto cont;
+
vaddr = offset_to_vaddr(vma, tu->sdt_offset);
- sdt_update_sem(vma->vm_mm, vaddr, 1);
+ if (!sdt_update_sem(vma->vm_mm, vaddr, 1))
+ sdt_add_mm_list(tu, vma->vm_mm);
+
+cont:
+ up_write(&tu->sml_rw_sem);
}
mutex_unlock(&uprobe_lock);
}
@@ -1017,7 +1108,11 @@ static void sdt_decrement_sem(struct trace_uprobe *tu)
if (IS_ERR(info))
return;
+ down_write(&tu->sml_rw_sem);
while (info) {
+ if (!sdt_check_mm_list(tu, info->mm))
+ goto cont;
+
down_write(&info->mm->mmap_sem);
vma = sdt_find_vma(info->mm, tu);
if (vma) {
@@ -1025,10 +1120,14 @@ static void sdt_decrement_sem(struct trace_uprobe *tu)
sdt_update_sem(info->mm, vaddr, -1);
}
up_write(&info->mm->mmap_sem);
-
+ sdt_del_mm_list(tu, info->mm);
+
+cont:
mmput(info->mm);
info = free_uprobe_map_info(info);
}
+ sdt_flush_mm_list(tu);
+ up_write(&tu->sml_rw_sem);
}
typedef bool (*filter_func_t)(struct uprobe_consumer *self,
--
1.8.3.1
next prev parent reply other threads:[~2018-02-28 7:52 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-02-28 7:53 [RFC 0/4] trace_uprobe: Support SDT markers having semaphore Ravi Bangoria
2018-02-28 7:53 ` [RFC 1/4] Uprobe: Rename map_info to uprobe_map_info Ravi Bangoria
2018-02-28 12:09 ` Srikar Dronamraju
2018-03-01 5:11 ` Ravi Bangoria
2018-02-28 7:53 ` [RFC 2/4] Uprobe: Export few functions / data structures Ravi Bangoria
2018-02-28 12:24 ` Srikar Dronamraju
2018-03-01 5:25 ` Ravi Bangoria
2018-02-28 7:53 ` [RFC 3/4] trace_uprobe: Support SDT markers having semaphore Ravi Bangoria
2018-03-01 14:07 ` Masami Hiramatsu
2018-03-02 3:54 ` Ravi Bangoria
2018-03-06 11:59 ` Peter Zijlstra
2018-03-07 8:46 ` Ravi Bangoria
2018-03-07 8:57 ` Peter Zijlstra
2018-02-28 7:53 ` Ravi Bangoria [this message]
2018-02-28 12:06 ` [RFC 0/4] " Srikar Dronamraju
2018-03-01 5:10 ` Ravi Bangoria
2018-02-28 14:25 ` Masami Hiramatsu
2018-03-01 5:32 ` Ravi Bangoria
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=20180228075345.674-5-ravi.bangoria@linux.vnet.ibm.com \
--to=ravi.bangoria@linux.vnet.ibm.com \
--cc=acme@kernel.org \
--cc=alexander.shishkin@linux.intel.com \
--cc=ananth@linux.vnet.ibm.com \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mhiramat@kernel.org \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=naveen.n.rao@linux.vnet.ibm.com \
--cc=oleg@redhat.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=srikar@linux.vnet.ibm.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
all inboxes | Powered by JetHome®