Bug 186757 – Properly check the sscanf return value

• Component: WebKitGTK
• Status: RESOLVED FIXED

[Tomas Popela]

Found by Coverity scan:

5. webkitgtk-2.20.3/Source/WebCore/page/linux/ResourceUsageThreadLinux.cpp:61: check_return: Calling "sscanf" without checking return value (as is done elsewhere 10 out of 12 times).
6. webkitgtk-2.20.3/Source/JavaScriptCore/jsc.cpp:2099: example_checked: Example 1: "sscanf(timeoutString, "%lf", &s_desiredTimeout)" has its value checked in "sscanf(timeoutString, "%lf", &s_desiredTimeout) != 1".
7. webkitgtk-2.20.3/Source/WTF/wtf/NumberOfCores.cpp:55: example_checked: Example 2: "sscanf(coresEnv, "%u", &numberOfCores)" has its value checked in "sscanf(coresEnv, "%u", &numberOfCores) == 1".
8. webkitgtk-2.20.3/Source/WTF/wtf/linux/MemoryFootprintLinux.cpp:86: example_checked: Example 3: "sscanf(buffer, "Private_Dirty: %lu", &privateDirtyInKB)" has its value checked in "sscanf(buffer, "Private_Dirty: %lu", &privateDirtyInKB) == 1".
9. webkitgtk-2.20.3/Source/JavaScriptCore/runtime/Options.cpp:250: example_assign: Example 4: Assigning: "scanResult" = return value from "sscanf(p, " %u:%u", &this->m_lowLimit, &this->m_highLimit)".
10. webkitgtk-2.20.3/Source/JavaScriptCore/runtime/Options.cpp:252: example_checked: Example 4 (cont.): "scanResult" has its value checked in "scanResult".
11. webkitgtk-2.20.3/Source/JavaScriptCore/runtime/Options.cpp:107: example_checked: Example 5: "sscanf(string, "%lf", value)" has its value checked in "sscanf(string, "%lf", value) == 1".
#    59|       unsigned long long ioWait, irq, softIrq, steal, guest, guestnice;
#    60|       ioWait = irq = softIrq = steal = guest = guestnice = 0;
#    61|->     sscanf(buffer, "cpu  %16llu %16llu %16llu %16llu %16llu %16llu %16llu %16llu %16llu %16llu",
#    62|           &userTime, &niceTime, &systemTime, &idleTime, &ioWait, &irq, &softIrq, &steal, &guest, &guestnice);
#    63|

[Tomas Popela]

Created attachment 342924 [details]
Patch

[Tomas Popela]

Comment on attachment 342924 [details]
Patch

Clearing flags on attachment: 342924

Committed r232929: <https://trac.webkit.org/changeset/232929>

[Tomas Popela]

All reviewed patches have been landed.  Closing bug.

[Darin Adler]

Comment on attachment 342924 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=342924&action=review

> Source/WebCore/page/linux/ResourceUsageThreadLinux.cpp:64
> +    if (retVal < 10 || retVal == EOF) {

I think this should just be:

    if (retVal != 10)

[Tomas Popela]

(In reply to Darin Adler from comment #4)
> I think this should just be:
> 
>     if (retVal != 10)

That's true Darin! I will change it.

[Tomas Popela]

Committed r233142: <https://trac.webkit.org/changeset/233142>