Stop doing unnecessary initialization in non-SFI

Stop doing unnecessary initialization in non-SFI

We do not need to call NaClChromeMainInit for non-SFI NaCl.
It seems the log module is initialized in other places, and
SRPC module is initialized in nonsfi_main.

Also set uses_irt=false for non-SFI NaCl. We do not need to
send the unnecessary FD of IRT.

With this patch, we will not call modify_ldt in non-SFI mode.

BUG=https://code.google.com/p/nativeclient/issues/detail?id=3734
BUG=350487
TEST=out/Release/browser_tests --gtest_filter='NaCl*'
TEST=Our app works without modify_ldt

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=261691

[hamaji, 2014-04-01 09:20:50]

I'm not sure if this is the right approach. If we want to keep non-SFI code path
to be similar to SFI code path, this patch would be wrong. If this is the case,
I guess we might want to add nonsfi::NaClChromeMainArgsCreate and use it from
nacl_listener.

[Mark Seaborn, 2014-04-01 14:35:36]

LGTM

https://codereview.chromium.org/218633011/diff/1/components/nacl/loader/nacl_...
File components/nacl/loader/nacl_listener.cc (right):

https://codereview.chromium.org/218633011/diff/1/components/nacl/loader/nacl_...
components/nacl/loader/nacl_listener.cc:270: if (!params.uses_nonsfi_mode) {
This conflicts with your other change,
https://codereview.chromium.org/216603002/, where uses_nonsfi_mode is
conditional on OS_LINUX.  You might want to make uses_nonsfi_mode defined
unconditionally.

https://codereview.chromium.org/218633011/diff/1/components/nacl/loader/nacl_...
components/nacl/loader/nacl_listener.cc:363: args = NaClChromeMainArgsCreate();
Combine with previous line?

https://codereview.chromium.org/218633011/diff/1/ppapi/native_client/src/trus...
File ppapi/native_client/src/trusted/plugin/plugin.cc (right):

https://codereview.chromium.org/218633011/diff/1/ppapi/native_client/src/trus...
ppapi/native_client/src/trusted/plugin/plugin.cc:349: const bool uses_irt =
!uses_nonsfi_mode;
It might be slightly preferable to do this in nacl_process_host.cc instead, i.e.

  params.uses_irt = uses_irt_ && !uses_nonsfi_mode_;

since this stops the renderer requesting a non-working combination.

[hamaji, 2014-04-03 07:05:54]

https://codereview.chromium.org/218633011/diff/1/components/nacl/loader/nacl_...
File components/nacl/loader/nacl_listener.cc (right):

https://codereview.chromium.org/218633011/diff/1/components/nacl/loader/nacl_...
components/nacl/loader/nacl_listener.cc:270: if (!params.uses_nonsfi_mode) {
On 2014/04/01 14:35:36, Mark Seaborn wrote:
> This conflicts with your other change,
> https://codereview.chromium.org/216603002/, where uses_nonsfi_mode is
> conditional on OS_LINUX.  You might want to make uses_nonsfi_mode defined
> unconditionally.

Done.

https://codereview.chromium.org/218633011/diff/1/components/nacl/loader/nacl_...
components/nacl/loader/nacl_listener.cc:363: args = NaClChromeMainArgsCreate();
On 2014/04/01 14:35:36, Mark Seaborn wrote:
> Combine with previous line?

Done.

https://codereview.chromium.org/218633011/diff/1/ppapi/native_client/src/trus...
File ppapi/native_client/src/trusted/plugin/plugin.cc (right):

https://codereview.chromium.org/218633011/diff/1/ppapi/native_client/src/trus...
ppapi/native_client/src/trusted/plugin/plugin.cc:349: const bool uses_irt =
!uses_nonsfi_mode;
On 2014/04/01 14:35:36, Mark Seaborn wrote:
> It might be slightly preferable to do this in nacl_process_host.cc instead,
i.e.
> 
>   params.uses_irt = uses_irt_ && !uses_nonsfi_mode_;
> 
> since this stops the renderer requesting a non-working combination.

Done.

[Mark Seaborn, 2014-04-03 14:07:54]

LGTM

[hamaji, 2014-04-04 03:18:42]

The CQ bit was checked by hamaji@chromium.org

[commit-bot: I haz the power, 2014-04-04 03:18:51]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/hamaji@chromium.org/218633011/20001

[commit-bot: I haz the power, 2014-04-04 06:24:55]

Change committed as 261691