* [PATCH] Check if /proc is usable on gdbserver start
[not found] <CGME20180704073447eucas1p1541c2021074eb8faa58e49f1abe8d733@eucas1p1.samsung.com>
@ 2018-07-04 7:34 ` Vyacheslav Barinov
2018-07-04 9:57 ` Pedro Alves
0 siblings, 1 reply; 4+ messages in thread
From: Vyacheslav Barinov @ 2018-07-04 7:34 UTC (permalink / raw)
To: gdb-patches
[-- Attachment #1: Type: text/plain, Size: 542 bytes --]
Hello,
Just lost a couple of hours trying to find while gdb can't connect to gdbserver
in my debug session.
There's a buildroot where I want to debug a binary, and I tried to connect to
it from outside, but got very weird errors like architecture mismatch or
protocol errors. At last, after switching on '--debug' for gdbserver I found a
message 'Can't open /proc/pid/' message and suddenly found that I forgot to
mount procfs in my buildroot.
I think it's better to check this before running gdbserver.
Best Regards,
Slava Barinov.
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-Check-if-proc-is-usable-on-gdbserver-start.patch --]
[-- Type: text/x-patch, Size: 982 bytes --]
From 0d9bc48e145195b9ce04dedb3ca75ee93d90a32c Mon Sep 17 00:00:00 2001
From: Slava Barinov <v.barinov@samsung.com>
Date: Wed, 4 Jul 2018 10:20:42 +0300
Subject: [PATCH] Check if /proc is usable on gdbserver start
Add check if procfs is mount and can be accessed by gdbserver.
gdbserver/
* linux-low.c (initialize_low): Add /proc check.
Signed-off-by: Slava Barinov <v.barinov@samsung.com>
---
gdb/gdbserver/linux-low.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/gdb/gdbserver/linux-low.c b/gdb/gdbserver/linux-low.c
index 6e026f1aab..28273da1b9 100644
--- a/gdb/gdbserver/linux-low.c
+++ b/gdb/gdbserver/linux-low.c
@@ -7553,6 +7553,13 @@ void
initialize_low (void)
{
struct sigaction sigchld_action;
+ struct stat st;
+
+ if (stat ("/proc/self", &st) != 0)
+ {
+ fprintf (stderr, "/proc is not accessible.\n");
+ exit (1);
+ }
memset (&sigchld_action, 0, sizeof (sigchld_action));
set_target_ops (&linux_target_ops);
--
2.18.0
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] Check if /proc is usable on gdbserver start
2018-07-04 7:34 ` [PATCH] Check if /proc is usable on gdbserver start Vyacheslav Barinov
@ 2018-07-04 9:57 ` Pedro Alves
2018-07-04 11:19 ` Vyacheslav Barinov
0 siblings, 1 reply; 4+ messages in thread
From: Pedro Alves @ 2018-07-04 9:57 UTC (permalink / raw)
To: Vyacheslav Barinov, gdb-patches
On 07/04/2018 08:34 AM, Vyacheslav Barinov wrote:
> Hello,
>
> Just lost a couple of hours trying to find while gdb can't connect to gdbserver
> in my debug session.
>
> There's a buildroot where I want to debug a binary, and I tried to connect to
> it from outside, but got very weird errors like architecture mismatch or
> protocol errors. At last, after switching on '--debug' for gdbserver I found a
> message 'Can't open /proc/pid/' message and suddenly found that I forgot to
> mount procfs in my buildroot.
>
> I think it's better to check this before running gdbserver.
I'm not certain that it's a good idea to absolute require a /proc mount.
Even though we gradually moved into relying on /proc more (or better said, on
libthread_db less), over the years we've also tried to be tolerant to
missing /proc. I believe people had use cases for that (very constrained
containers or embedded systems? I don't recall exactly though...)
I assume that you're on either Aarch64 or x86_64 and that the architecture
mismatch is because gdbserver couldn't determine whether the process
was a 64-bit process, because that is done by reading the /proc/PID/exe
file. Maybe other 32-bit- or 64-bit-only archs still minimally work
without /proc.
I'm thinking that it may be prudent to downgrade this to a warning.
Would that work for you?
Thanks,
Pedro Alves
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] Check if /proc is usable on gdbserver start
2018-07-04 9:57 ` Pedro Alves
@ 2018-07-04 11:19 ` Vyacheslav Barinov
2018-07-04 15:23 ` [pushed] Warn if /proc is not accessible (was: Re: [PATCH] Check if /proc is usable on gdbserver start) Pedro Alves
0 siblings, 1 reply; 4+ messages in thread
From: Vyacheslav Barinov @ 2018-07-04 11:19 UTC (permalink / raw)
To: Pedro Alves; +Cc: gdb-patches
Pedro Alves <palves@redhat.com> writes:
> On 07/04/2018 08:34 AM, Vyacheslav Barinov wrote:
>> Hello,
>>
>> Just lost a couple of hours trying to find while gdb can't connect to gdbserver
>> in my debug session.
>>
>> There's a buildroot where I want to debug a binary, and I tried to connect to
>> it from outside, but got very weird errors like architecture mismatch or
>> protocol errors. At last, after switching on '--debug' for gdbserver I found a
>> message 'Can't open /proc/pid/' message and suddenly found that I forgot to
>> mount procfs in my buildroot.
>>
>> I think it's better to check this before running gdbserver.
> I'm not certain that it's a good idea to absolute require a /proc mount.
> Even though we gradually moved into relying on /proc more (or better said, on
> libthread_db less), over the years we've also tried to be tolerant to
> missing /proc. I believe people had use cases for that (very constrained
> containers or embedded systems? I don't recall exactly though...)
>
> I assume that you're on either Aarch64 or x86_64 and that the architecture
> mismatch is because gdbserver couldn't determine whether the process
> was a 64-bit process, because that is done by reading the /proc/PID/exe
> file. Maybe other 32-bit- or 64-bit-only archs still minimally work
> without /proc.
>
> I'm thinking that it may be prudent to downgrade this to a warning.
> Would that work for you?
>
> Thanks,
> Pedro Alves
You are right, I use x86_64 architecture.
Okay, if the gdbserver can work without /proc, then just warning for cases
like this is enough.
Best Regards,
Vyacheslav Barinov
^ permalink raw reply [flat|nested] 4+ messages in thread
* [pushed] Warn if /proc is not accessible (was: Re: [PATCH] Check if /proc is usable on gdbserver start)
2018-07-04 11:19 ` Vyacheslav Barinov
@ 2018-07-04 15:23 ` Pedro Alves
0 siblings, 0 replies; 4+ messages in thread
From: Pedro Alves @ 2018-07-04 15:23 UTC (permalink / raw)
To: Vyacheslav Barinov; +Cc: gdb-patches
On 07/04/2018 12:19 PM, Vyacheslav Barinov wrote:
>
> Okay, if the gdbserver can work without /proc, then just warning for cases
> like this is enough.
Alright, here's what I've merged.
I added a new linux_ptrace_init_warnings routine to nat/linux-procfs.c
very similar in spirit to the existing linux_ptrace_init_warnings
and called it from the same places. This makes native debugging
with GDB warn too.
From 7c65a87e84f79155a9e9f59890340f85387df8a1 Mon Sep 17 00:00:00 2001
From: Vyacheslav Barinov <v.barinov@samsung.com>
Date: Wed, 4 Jul 2018 16:13:29 +0100
Subject: [PATCH] Warn if /proc is not accessible
There's a buildroot where I want to debug a binary, and I tried to
connect to it from outside, but got very weird errors like
architecture mismatch or protocol errors. At last, after switching on
'--debug' for gdbserver I found a message 'Can't open /proc/pid/'
message and suddenly found that I forgot to mount procfs in my
buildroot.
Make discovering the problem easier by making GDB / GDBserver warn
(even without --debug) if /proc can not be accessed.
Native debugging:
(gdb) start
Temporary breakpoint 1 at 0x400835: file test.c, line 10.
Starting program: /tmp/test
warning: /proc is not accessible.
GDBserver/remote debugging:
$ ./gdbserver :9999 ./gdbserver
gdbserver: /proc is not accessible.
gdb/ChangeLog:
2018-07-04 Vyacheslav Barinov <v.barinov@samsung.com>
Pedro Alves <palves@redhat.com>
* linux-nat.c (linux_init_ptrace): Rename to ...
(linux_init_ptrace_procfs): ... this. Call
linux_proc_init_warnings.
(linux_nat_target::post_attach)
(linux_nat_target::post_startup_inferior): Adjust.
* nat/linux-procfs.c (linux_proc_init_warnings): Define function.
* nat/linux-procfs.h (linux_proc_init_warnings): Declare function.
gdb/gdbserver/ChangeLog:
2018-07-04 Vyacheslav Barinov <v.barinov@samsung.com>
Pedro Alves <palves@redhat.com>
* linux-low.c (initialize_low): Call linux_proc_init_warnings.
---
gdb/gdbserver/linux-low.c | 1 +
gdb/linux-nat.c | 11 ++++++-----
gdb/nat/linux-procfs.c | 17 +++++++++++++++++
gdb/nat/linux-procfs.h | 5 +++++
4 files changed, 29 insertions(+), 5 deletions(-)
diff --git a/gdb/gdbserver/linux-low.c b/gdb/gdbserver/linux-low.c
index 6e026f1aab8..7c396ae3662 100644
--- a/gdb/gdbserver/linux-low.c
+++ b/gdb/gdbserver/linux-low.c
@@ -7558,6 +7558,7 @@ initialize_low (void)
set_target_ops (&linux_target_ops);
linux_ptrace_init_warnings ();
+ linux_proc_init_warnings ();
sigchld_action.sa_handler = sigchld_handler;
sigemptyset (&sigchld_action.sa_mask);
diff --git a/gdb/linux-nat.c b/gdb/linux-nat.c
index 567cb064d3a..18d9a3872a5 100644
--- a/gdb/linux-nat.c
+++ b/gdb/linux-nat.c
@@ -384,18 +384,19 @@ linux_nat_ptrace_options (int attached)
return options;
}
-/* Initialize ptrace warnings and check for supported ptrace
- features given PID.
+/* Initialize ptrace and procfs warnings and check for supported
+ ptrace features given PID.
ATTACHED should be nonzero iff we attached to the inferior. */
static void
-linux_init_ptrace (pid_t pid, int attached)
+linux_init_ptrace_procfs (pid_t pid, int attached)
{
int options = linux_nat_ptrace_options (attached);
linux_enable_event_reporting (pid, options);
linux_ptrace_init_warnings ();
+ linux_proc_init_warnings ();
}
linux_nat_target::~linux_nat_target ()
@@ -404,13 +405,13 @@ linux_nat_target::~linux_nat_target ()
void
linux_nat_target::post_attach (int pid)
{
- linux_init_ptrace (pid, 1);
+ linux_init_ptrace_procfs (pid, 1);
}
void
linux_nat_target::post_startup_inferior (ptid_t ptid)
{
- linux_init_ptrace (ptid.pid (), 0);
+ linux_init_ptrace_procfs (ptid.pid (), 0);
}
/* Return the number of known LWPs in the tgid given by PID. */
diff --git a/gdb/nat/linux-procfs.c b/gdb/nat/linux-procfs.c
index 1c236c543d9..6c15ea5cce3 100644
--- a/gdb/nat/linux-procfs.c
+++ b/gdb/nat/linux-procfs.c
@@ -357,3 +357,20 @@ linux_proc_pid_to_exec_file (int pid)
return buf;
}
+
+/* See linux-procfs.h. */
+
+void
+linux_proc_init_warnings ()
+{
+ static bool warned = false;
+
+ if (warned)
+ return;
+ warned = true;
+
+ struct stat st;
+
+ if (stat ("/proc/self", &st) != 0)
+ warning (_("/proc is not accessible."));
+}
diff --git a/gdb/nat/linux-procfs.h b/gdb/nat/linux-procfs.h
index efb2911cb8f..2f8932e89c7 100644
--- a/gdb/nat/linux-procfs.h
+++ b/gdb/nat/linux-procfs.h
@@ -80,4 +80,9 @@ extern int linux_proc_task_list_dir_exists (pid_t pid);
extern char *linux_proc_pid_to_exec_file (int pid);
+/* Display possible problems on this system. Display them only once
+ per GDB execution. */
+
+extern void linux_proc_init_warnings ();
+
#endif /* COMMON_LINUX_PROCFS_H */
--
2.14.4
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2018-07-04 15:23 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <CGME20180704073447eucas1p1541c2021074eb8faa58e49f1abe8d733@eucas1p1.samsung.com>
2018-07-04 7:34 ` [PATCH] Check if /proc is usable on gdbserver start Vyacheslav Barinov
2018-07-04 9:57 ` Pedro Alves
2018-07-04 11:19 ` Vyacheslav Barinov
2018-07-04 15:23 ` [pushed] Warn if /proc is not accessible (was: Re: [PATCH] Check if /proc is usable on gdbserver start) Pedro Alves
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).