Reland "Remove chromeos::LibraryLoader"

Reland "Remove chromeos::LibraryLoader"

CrosLibrary::libcros_loaded()/load_error_string() will be removed later when there is no user for them
--stub-cros command line option is removed since our code uses stub CrosLibrary whenever libcros loading fails

Originally reviewed at: https://chromiumcodereview.appspot.com/10574040/
Reverted in: http://crrev.com/143365

BUG=127051
TEST=git try -b linux_chromeos,cros_x86

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=143526

[hashimoto, 2012-06-21 15:37:20]

https://chromiumcodereview.appspot.com/10626002/diff/2001/chrome/browser/chro...
File chrome/browser/chromeos/cros/cert_library.cc (left):

https://chromiumcodereview.appspot.com/10626002/diff/2001/chrome/browser/chro...
chrome/browser/chromeos/cros/cert_library.cc:157:
CommandLine::ForCurrentProcess()->HasSwitch(switches::kStubCros)) {
Steven, loading Opencryptoki on Linux resulted in build bot failure.
(http://build.chromium.org/p/chromium.chromiumos/builders/Linux%20ChromiumOS%2...)
Can we remove this stub hack at all?
It looks like no one uses this code path on Linux recently.

[hashimoto, 2012-06-21 15:38:17]

Steven, loading Opencryptoki on Linux resulted in build bot failure.
(http://build.chromium.org/p/chromium.chromiumos/builders/Linux%20ChromiumOS%2...)
Can we remove the stub hack at all?
It looks like no one uses this code path on Linux recently.

[satorux1, 2012-06-21 15:43:00]

LGTM.

I think it's safe to remove the code path. IIRC, opencryptoki was replaced with
Chaps. adding gspencer@ and dkrahn@ who'd know the status.

[stevenjb, 2012-06-21 16:29:54]

lgtm

https://chromiumcodereview.appspot.com/10626002/diff/2001/chrome/browser/chro...
File chrome/browser/chromeos/cros/cert_library.cc (left):

https://chromiumcodereview.appspot.com/10626002/diff/2001/chrome/browser/chro...
chrome/browser/chromeos/cros/cert_library.cc:157:
CommandLine::ForCurrentProcess()->HasSwitch(switches::kStubCros)) {
On 2012/06/21 15:37:20, hashimoto wrote:
> Steven, loading Opencryptoki on Linux resulted in build bot failure.
>
(http://build.chromium.org/p/chromium.chromiumos/builders/Linux%2520ChromiumOS...)
> Can we remove this stub hack at all?
> It looks like no one uses this code path on Linux recently.

I'm not sure why we were loading this for --stub-cros, removing it seems fine.
Otherwise we would need to protect it with IsRunningOnChromeOS().

[commit-bot: I haz the power, 2012-06-22 02:30:12]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/hashimoto@chromium.org/10626002/2001

[commit-bot: I haz the power, 2012-06-22 03:22:08]

Try job failure for 10626002-2001 (retry) on mac_rel for step "browser_tests".
It's a second try, previously, step "browser_tests" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=mac_rel&nu...

[Darren Krahn, 2012-06-22 03:58:40]

On 2012/06/21 15:43:00, satorux1 wrote:
> LGTM.
> 
> I think it's safe to remove the code path. IIRC, opencryptoki was replaced
with
> Chaps. adding gspencer@ and dkrahn@ who'd know the status.

FWIW, opencryptoki has been removed as of R21.

[hashimoto, 2012-06-22 06:03:37]

On 2012/06/22 03:58:40, dkrahn wrote:
> On 2012/06/21 15:43:00, satorux1 wrote:
> > LGTM.
> > 
> > I think it's safe to remove the code path. IIRC, opencryptoki was replaced
> with
> > Chaps. adding gspencer@ and dkrahn@ who'd know the status.
> 
> FWIW, opencryptoki has been removed as of R21.

I see "--load-opencryptoki" option in /sbin/session_manager_setup.sh.
AFAIK CertLibrary::user_logged_in_ never gets true without "--load-opencryptoki"
and EncryptToken/DecryptToken does not work. Was opencryptoki really removed?