[GTK][WPE] Fix BubblewrapLauncher clang warnings

• Component: WebKitGTK
• Status: RESOLVED FIXED

[Tomas Popela]

../../Source/WebKit/UIProcess/Launcher/glib/BubblewrapLauncher.cpp:474:37: warning: variable 'i' is uninitialized when used here [-Wuninitialized]
    for (size_t i; splitPaths.get()[i]; ++i)
                                    ^
../../Source/WebKit/UIProcess/Launcher/glib/BubblewrapLauncher.cpp:474:18: note: initialize the variable 'i' to silence this warning
    for (size_t i; splitPaths.get()[i]; ++i)
                 ^
                  = 0
../../Source/WebKit/UIProcess/Launcher/glib/BubblewrapLauncher.cpp:574:34: warning: missing field 'datum_b' initializer [-Wmissing-field-initializers]
    struct scmp_arg_cmp ttyArg = SCMP_A1(SCMP_CMP_EQ, (int)TIOCSTI);
                                 ^
/usr/include/seccomp.h:211:23: note: expanded from macro 'SCMP_A1'
#define SCMP_A1(...)            SCMP_CMP(1, __VA_ARGS__)
                                ^
/usr/include/seccomp.h:201:58: note: expanded from macro 'SCMP_CMP'
#define SCMP_CMP(...)           ((struct scmp_arg_cmp){__VA_ARGS__})
                                                                  ^

[Tomas Popela]

Created attachment 355886 [details]
Patch

[Michael Catanzaro]

Comment on attachment 355886 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=355886&action=review

> Source/WebKit/UIProcess/Launcher/glib/BubblewrapLauncher.cpp:474
> -    for (size_t i; splitPaths.get()[i]; ++i)
> +    for (size_t i = 0; splitPaths.get()[i]; ++i)

Ow!

> Source/WebKit/UIProcess/Launcher/glib/BubblewrapLauncher.cpp:574
> -    struct scmp_arg_cmp ttyArg = SCMP_A1(SCMP_CMP_EQ, (int)TIOCSTI);
> +    struct scmp_arg_cmp ttyArg = SCMP_A1(SCMP_CMP_EQ, (scmp_datum_t)TIOCSTI, 0);

I spent too long investigating why you have the trailing 0. Would be more clear if we add an unnecessary cast:

struct scmp_arg_cmp ttyArg = SCMP_A1(SCMP_CMP_EQ, (scmp_datum_t)TIOCSTI, (scmp_datum_t)0);

But now I notice that we are impermissibly using C-style casts here. How about:

struct scmp_arg_cmp ttyArg = SCMP_A1(SCMP_CMP_EQ, static_cast<scmp_datum_t>(TIOCSTI), static_cast<scmp_datum_t>(0));

Er, maybe too verbose. It's probably fine with just the first cast. But it should be a static_cast.

[Tomas Popela]

Comment on attachment 355886 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=355886&action=review

>> Source/WebKit/UIProcess/Launcher/glib/BubblewrapLauncher.cpp:574
>> +    struct scmp_arg_cmp ttyArg = SCMP_A1(SCMP_CMP_EQ, (scmp_datum_t)TIOCSTI, 0);
> 
> I spent too long investigating why you have the trailing 0. Would be more clear if we add an unnecessary cast:
> 
> struct scmp_arg_cmp ttyArg = SCMP_A1(SCMP_CMP_EQ, (scmp_datum_t)TIOCSTI, (scmp_datum_t)0);
> 
> But now I notice that we are impermissibly using C-style casts here. How about:
> 
> struct scmp_arg_cmp ttyArg = SCMP_A1(SCMP_CMP_EQ, static_cast<scmp_datum_t>(TIOCSTI), static_cast<scmp_datum_t>(0));
> 
> Er, maybe too verbose. It's probably fine with just the first cast. But it should be a static_cast.

True, that sounds a lot better (I don't know, why I didn't do that)! :)

[Tomas Popela]

Created attachment 355985 [details]
Patch for landing

[WebKit Commit Bot]

The commit-queue encountered the following flaky tests while processing attachment 355985 [details]:

workers/bomb.html bug 171985 (author: [e-mail])
The commit-queue is continuing to process your patch.

[WebKit Commit Bot]

The commit-queue encountered the following flaky tests while processing attachment 355985 [details]:

imported/w3c/web-platform-tests/css/css-shapes/shape-outside/values/shape-outside-inset-001.html bug 192139 (author: [e-mail])
imported/w3c/web-platform-tests/encoding/legacy-mb-korean/euc-kr/euckr-encode-form.html bug 192140 (author: [e-mail])
webgl/1.0.2/conformance/more/conformance/quickCheckAPI-A.html bug 192141 (author: [e-mail])
The commit-queue is continuing to process your patch.

[WebKit Commit Bot]

The commit-queue encountered the following flaky tests while processing attachment 355985 [details]:

workers/bomb.html bug 171985 (author: [e-mail])
The commit-queue is continuing to process your patch.

[WebKit Commit Bot]

Comment on attachment 355985 [details]
Patch for landing

Clearing flags on attachment: 355985

Committed r238663: <https://trac.webkit.org/changeset/238663>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.