[NaCl SDK] Add realpath(3) implementation.

[NaCl SDK] Add realpath(3) implementation.

This really should be in nacl-newlib, but it's easier to land here now and
remove it later when it lands in newlib and rolls into the SDK.

This implementation does not support symlinks, but neither does nacl_io
currently.

BUG=none
R=sbc@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=272551

[Sam Clegg, 2014-05-21 23:39:21]

https://codereview.chromium.org/295933009/diff/1/native_client_sdk/src/librar...
File native_client_sdk/src/libraries/nacl_io/include/stdlib.h (right):

https://codereview.chromium.org/295933009/diff/1/native_client_sdk/src/librar...
native_client_sdk/src/libraries/nacl_io/include/stdlib.h:13: char*
realpath(const char* path, char* resolved_path);
Didn't we find that this was actually in newlib's headers?

https://codereview.chromium.org/295933009/diff/1/native_client_sdk/src/tests/...
File native_client_sdk/src/tests/nacl_io_test/syscalls_test.cc (right):

https://codereview.chromium.org/295933009/diff/1/native_client_sdk/src/tests/...
native_client_sdk/src/tests/nacl_io_test/syscalls_test.cc:45: ASSERT_EQ(0,
ki_mkdir("/foo", S_IREAD | S_IWRITE));
If you wrote these tests with everything happening under "/tmp" then you could
run the tests against linux's glibc to verify that your test cases are correct. 
(or maybe you did that already?).

[binji, 2014-05-21 23:56:01]

https://codereview.chromium.org/295933009/diff/1/native_client_sdk/src/librar...
File native_client_sdk/src/libraries/nacl_io/include/stdlib.h (right):

https://codereview.chromium.org/295933009/diff/1/native_client_sdk/src/librar...
native_client_sdk/src/libraries/nacl_io/include/stdlib.h:13: char*
realpath(const char* path, char* resolved_path);
On 2014/05/21 23:39:21, Sam Clegg wrote:
> Didn't we find that this was actually in newlib's headers?

It wasn't in my headers anywhere. I'm not sure why it was in yours.

https://codereview.chromium.org/295933009/diff/1/native_client_sdk/src/tests/...
File native_client_sdk/src/tests/nacl_io_test/syscalls_test.cc (right):

https://codereview.chromium.org/295933009/diff/1/native_client_sdk/src/tests/...
native_client_sdk/src/tests/nacl_io_test/syscalls_test.cc:45: ASSERT_EQ(0,
ki_mkdir("/foo", S_IREAD | S_IWRITE));
On 2014/05/21 23:39:21, Sam Clegg wrote:
> If you wrote these tests with everything happening under "/tmp" then you could
> run the tests against linux's glibc to verify that your test cases are
correct. 
> (or maybe you did that already?).

Hm, good idea. I thought I needed write access to root, but I just need absolute
paths. I'll give it a shot.

[binji, 2014-05-22 22:36:27]

ptal

[Sam Clegg, 2014-05-22 22:58:53]

lgtm

https://codereview.chromium.org/295933009/diff/40001/native_client_sdk/src/li...
File native_client_sdk/src/libraries/nacl_io/syscalls/mkdir.c (right):

https://codereview.chromium.org/295933009/diff/40001/native_client_sdk/src/li...
native_client_sdk/src/libraries/nacl_io/syscalls/mkdir.c:8: #if
defined(__native_client__)
Comment?

[binji, 2014-05-23 16:52:56]

https://codereview.chromium.org/295933009/diff/40001/native_client_sdk/src/li...
File native_client_sdk/src/libraries/nacl_io/syscalls/mkdir.c (right):

https://codereview.chromium.org/295933009/diff/40001/native_client_sdk/src/li...
native_client_sdk/src/libraries/nacl_io/syscalls/mkdir.c:8: #if
defined(__native_client__)
On 2014/05/22 22:58:54, Sam Clegg wrote:
> Comment?

Done.

[binji, 2014-05-23 16:58:59]

Committed patchset #5 manually as r272551 (presubmit successful).