diff mbox

[v6,12/20] trace: don't abort qemu if ftrace can't be initialized

Message ID 1475588159-30598-13-git-send-email-berrange@redhat.com
State New
Headers show

Commit Message

Daniel P. Berrangé Oct. 4, 2016, 1:35 p.m. UTC
If the ftrace backend is compiled into QEMU, any attempt
to start QEMU while non-root will fail due to the
inability to open /sys/kernel/debug/tracing/tracing_on.

Add a fallback into the code so that it connects up the
trace_marker_fd variable to /dev/null when setting
EACCESS on the 'trace_on' file. This allows QEMU to
run, with ftrace turned into a no-op.

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
---
 trace/ftrace.c | 6 ++++++
 1 file changed, 6 insertions(+)

Comments

Eric Blake Oct. 4, 2016, 8:12 p.m. UTC | #1
On 10/04/2016 08:35 AM, Daniel P. Berrange wrote:
> If the ftrace backend is compiled into QEMU, any attempt
> to start QEMU while non-root will fail due to the
> inability to open /sys/kernel/debug/tracing/tracing_on.
> 
> Add a fallback into the code so that it connects up the
> trace_marker_fd variable to /dev/null when setting

s/setting/getting/ ?

> EACCESS on the 'trace_on' file. This allows QEMU to

s/EACCESS/EACCES/

> run, with ftrace turned into a no-op.
> 
> Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
> ---
>  trace/ftrace.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/trace/ftrace.c b/trace/ftrace.c
> index e953922..3588bb0 100644
> --- a/trace/ftrace.c
> +++ b/trace/ftrace.c
> @@ -51,6 +51,12 @@ bool ftrace_init(void)
>          snprintf(path, PATH_MAX, "%s/tracing/tracing_on", debugfs);

Unrelated, but it would be nice to get rid of the PATH_MAX stack
allocation, and instead use g_strdup_printf() or similar in a followup
patch.

>          trace_fd = open(path, O_WRONLY);
>          if (trace_fd < 0) {
> +            if (errno == EACCES) {
> +                trace_marker_fd = open("/dev/null", O_WRONLY);
> +                if (trace_marker_fd != -1) {
> +                    return true;
> +                }
> +            }
>              perror("Could not open ftrace 'tracing_on' file");
>              return false;
>          } else {
>
Stefan Hajnoczi Oct. 5, 2016, 3:39 p.m. UTC | #2
On Tue, Oct 04, 2016 at 03:12:31PM -0500, Eric Blake wrote:
> On 10/04/2016 08:35 AM, Daniel P. Berrange wrote:
> > If the ftrace backend is compiled into QEMU, any attempt
> > to start QEMU while non-root will fail due to the
> > inability to open /sys/kernel/debug/tracing/tracing_on.
> > 
> > Add a fallback into the code so that it connects up the
> > trace_marker_fd variable to /dev/null when setting
> 
> s/setting/getting/ ?
> 
> > EACCESS on the 'trace_on' file. This allows QEMU to
> 
> s/EACCESS/EACCES/

Thanks, I've applied these changes while merging.

Stefan
diff mbox

Patch

diff --git a/trace/ftrace.c b/trace/ftrace.c
index e953922..3588bb0 100644
--- a/trace/ftrace.c
+++ b/trace/ftrace.c
@@ -51,6 +51,12 @@  bool ftrace_init(void)
         snprintf(path, PATH_MAX, "%s/tracing/tracing_on", debugfs);
         trace_fd = open(path, O_WRONLY);
         if (trace_fd < 0) {
+            if (errno == EACCES) {
+                trace_marker_fd = open("/dev/null", O_WRONLY);
+                if (trace_marker_fd != -1) {
+                    return true;
+                }
+            }
             perror("Could not open ftrace 'tracing_on' file");
             return false;
         } else {