debugging blink process with --renderer-startup-dialog broken?

[Stefan Zager]

Howdy people,

My regular debugging flow is to launch content_shell with --renderer-startup-dialog, like this:

$ content_shell --no-sandbox --renderer-startup-dialog test.html

...

[16218:16218:0525/112126:1188505707021:ERROR:child_process.cc(136)] Renderer (16218) paused waiting for debugger to attach. Send SIGUSR1 to unpause.

Then, in another terminal:

$ gdb

(gdb) attach 16218

<set breakpoints>

(gdb) c

And back in the first terminal:

$ kill -s SIGUSR1 16218

Today, this isn't working.  After running the `kill` command, content_shell hangs for a few seconds, and then I get this from the content_shell terminal

[15153:15153:0525/112035:1188454431795:INFO:child_thread_impl.cc(723)] ChildThreadImpl::EnsureConnected()

... and the renderer process in gdb receives SIGTERM.

Does anybody know why this might be broken now?

[Jeremy Roman]

https://bugs.chromium.org/p/chromium/issues/detail?id=612844

[Jeremy Roman]

Whoops, sent too early.

On Wed, May 25, 2016 at 4:30 PM, Jeremy Roman <jbr...@chromium.org> wrote:

https://bugs.chromium.org/p/chromium/issues/detail?id=612844

tl;dr: probably IPC::ChannelMojo

 

On Wed, May 25, 2016 at 2:27 PM, Stefan Zager <sza...@chromium.org> wrote:

Howdy people,

My regular debugging flow is to launch content_shell with --renderer-startup-dialog, like this:

$ content_shell --no-sandbox --renderer-startup-dialog test.html

...

[16218:16218:0525/112126:1188505707021:ERROR:child_process.cc(136)] Renderer (16218) paused waiting for debugger to attach. Send SIGUSR1 to unpause.

Then, in another terminal:

$ gdb

(gdb) attach 16218

<set breakpoints>

(gdb) c

And back in the first terminal:

$ kill -s SIGUSR1 16218

By the way, you can do this from within gdb by running "signal SIGUSR1" rather than "c". (This is what the debug-renderer script, which uses --renderer-startup-dialog, advises.)

[Robert Flack]

I've also had good luck with using the --single-process flag to run the renderer in the same process which lets you run content_shell from within gdb. This has worked so far for all of the things I've needed to debug, but I recognize this exercises different code paths so it may not run the same way as without --single-process.

[Stefan Zager]

I've been burned too many times by --single-process flag; enough so that I long ago added a warning against --single-process to the blink debugging guide:

https://www.chromium.org/blink/getting-started-with-blink-debugging#TOC-Starting-the-Debugger

[neil scanlan]

I know 

Whoops, sent too early.

On Wed, May 25, 2016 at 4:30 PM, Jeremy Roman <jbr...@chromium.org> wrote:

https://bugs.chromium.org/p/chromium/issues/detail?id=612844

tl;dr: probably IPC::ChannelMojo

 

On Wed, May 25, 2016 at 2:27 PM, Stefan Zager <sza...@chromium.org> wrote:

Howdy people,

My regular debugging flow is to launch content_shell with --renderer-startup-dialog, like this:

$ content_shell --no-sandbox --renderer-startup-dialog test.html

...

[16218:16218:0525/112126:1188505707021:ERROR:child_process.cc(136)] Renderer (16218) paused waiting for debugger to attach. Send SIGUSR1 to unpause.

Then, in another terminal:

$ gdb

(gdb) attach 16218

<set breakpoints>

(gdb) c

And back in the first terminal:

$ kill -s SIGUSR1 16218

By the way, you can do this from within gdb by running "signal SIGUSR1" rather than "c". (This is what the debug-renderer script, which uses --renderer-startup-dialog, advises.)

 

Today, this isn't working.  After running the `kill` command, content_shell hangs for a few seconds, and then I get this from the content_shell terminal

[15153:15153:0525/112035:1188454431795:INFO:child_thread_impl.cc(723)] ChildThreadImpl::EnsureConnected()

... and the renderer process in gdb receives SIGTERM.

Does anybody know why this might be broken now?

--
--
Chromium Developers mailing list: chromi...@chromium.org
View archives, change email options, or unsubscribe:
http://groups.google.com/a/chromium.org/group/chromium-dev
---
You received this message because you are subscribed to the Google Groups "Chromium-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chromium-dev...@chromium.org.