Solaris patch

Patch for Solaris support, mostly ifdefs and header files, plus adds libevent configuration.

BUG=30101
TEST=compiles

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=90494

[ruben, 2011-06-24 07:26:32]

Evan: main reviewer
Tony: another set of eyes
James: Solaris-related feedback

I pulled in an unmodified MP-licensed file from the system NSPR, just as has
been done on other platforms.  I had to add the system NSS directory to the
runtime search path, as Solaris doesn't put the NSS libraries in a common
location like other OS's.  The long OS ifdefs in Skia don't look good, but I
don't know of any other accepted solution.  I followed James in using __sun to
distinguish the platform.

[Evan Martin, 2011-06-24 17:51:18]

Can you do a pass over the files you modified and ensure the platform-specific
includes are in a block below the normal platform includes?

http://codereview.chromium.org/7238021/diff/1/base/base_paths_linux.cc
File base/base_paths_linux.cc (left):

http://codereview.chromium.org/7238021/diff/1/base/base_paths_linux.cc#oldcode29
base/base_paths_linux.cc:29: const char kSelfExe[] = getexecname();
Thanks for fixing this!

http://codereview.chromium.org/7238021/diff/1/base/debug/debugger_posix.cc
File base/debug/debugger_posix.cc (right):

http://codereview.chromium.org/7238021/diff/1/base/debug/debugger_posix.cc#ne...
base/debug/debugger_posix.cc:15: #if defined(OS_OPENBSD) || defined(OS_FREEBSD)
Can you pull these platform-specific includes below this list?  See where e.g.
MACOSX is tested.

http://codereview.chromium.org/7238021/diff/1/base/sync_socket_posix.cc
File base/sync_socket_posix.cc (right):

http://codereview.chromium.org/7238021/diff/1/base/sync_socket_posix.cc#newco...
base/sync_socket_posix.cc:13: #if defined(OS_SOLARIS)
Can you put a newline before this?

http://codereview.chromium.org/7238021/diff/1/content/common/set_process_titl...
File content/common/set_process_title.cc (right):

http://codereview.chromium.org/7238021/diff/1/content/common/set_process_titl...
content/common/set_process_title.cc:25: #include
"content/common/set_process_title_linux.h"
Does this code work on Solaris?  I'd be surprised if it did.

http://codereview.chromium.org/7238021/diff/1/ipc/ipc_channel_posix.cc
File ipc/ipc_channel_posix.cc (right):

http://codereview.chromium.org/7238021/diff/1/ipc/ipc_channel_posix.cc#newcod...
ipc/ipc_channel_posix.cc:942: return false;
two space indnet

http://codereview.chromium.org/7238021/diff/1/net/net.gyp
File net/net.gyp (right):

http://codereview.chromium.org/7238021/diff/1/net/net.gyp#newcode760
net/net.gyp:760: 'libraries': [
ldflags, I think?

(it ends up in the same command line as libraries, but in a different place
within the command line)

[ruben, 2011-06-24 20:22:23]

On 2011/06/24 17:51:18, Evan Martin wrote:
> Can you do a pass over the files you modified and ensure the platform-specific
> includes are in a block below the normal platform includes?
Sorry about the style errors, fixed.

>
http://codereview.chromium.org/7238021/diff/1/base/debug/debugger_posix.cc#ne...
> base/debug/debugger_posix.cc:15: #if defined(OS_OPENBSD) ||
defined(OS_FREEBSD)
> Can you pull these platform-specific includes below this list?  See where e.g.
> MACOSX is tested.
Done.

>
http://codereview.chromium.org/7238021/diff/1/base/sync_socket_posix.cc#newco...
> base/sync_socket_posix.cc:13: #if defined(OS_SOLARIS)
> Can you put a newline before this?
Fixed.

>
http://codereview.chromium.org/7238021/diff/1/content/common/set_process_titl...
> content/common/set_process_title.cc:25: #include
> "content/common/set_process_title_linux.h"
> Does this code work on Solaris?  I'd be surprised if it did.
You're right, it didn't make a difference after taking it out, removed.
 
>
http://codereview.chromium.org/7238021/diff/1/ipc/ipc_channel_posix.cc#newcod...
> ipc/ipc_channel_posix.cc:942: return false;
> two space indnet
Fixed.
  
> http://codereview.chromium.org/7238021/diff/1/net/net.gyp#newcode760
> net/net.gyp:760: 'libraries': [
> ldflags, I think?
> 
> (it ends up in the same command line as libraries, but in a different place
> within the command line)
Yep, good catch, fixed.

[Evan Martin, 2011-06-24 21:08:17]

LGTM

[tony, 2011-06-24 21:11:51]

Adding jam for content rubber stamp and willchan and eroman for net rubber
stamp.

[eroman, 2011-06-24 21:13:44]

LGTM on net/ changes

[jchoi42, 2011-06-24 21:39:25]

On 2011/06/24 21:36:19, jchoi42 wrote:

LGTM

[jam, 2011-06-25 01:47:14]

content lgtm

[ruben, 2011-06-25 01:59:44]

On 2011/06/24 21:11:51, tony wrote:
> Adding jam for content rubber stamp and willchan and eroman for net rubber
> stamp.
Does Will need to sign off too or should I go ahead and commit?

[eroman, 2011-06-25 02:25:11]

You don't need to wait for Will, since I already approved for net.

If he does have comments, they can be addressed as a follow-up change.

[ruben, 2011-06-25 02:26:30]

On 2011/06/25 02:25:11, eroman wrote:
> You don't need to wait for Will, since I already approved for net.
> 
> If he does have comments, they can be addressed as a follow-up change.
OK, I flipped the commit bit. :)

[ruben, 2011-06-25 03:26:54]

Try failure on Mac, looks like I missed some sysctl stuff.  Fixed them and
retrying.

[ruben, 2011-06-25 04:33:04]

On 2011/06/25 03:26:54, ruben wrote:
> Try failure on Mac, looks like I missed some sysctl stuff.  Fixed them and
> retrying.
More failures, but these look like flakes.  Can somebody with a commit bit force
the commit?

[commit-bot: I haz the power, 2011-06-25 04:54:42]

Change committed as 90494