The latter will result in a message that says "ASSERTION FAILED: kr = KERN_SUCCESS".
I considered ASSERT(false) (which seems equivalent to ASSERT_NOT_REACHED), but I felt that it was not as good as ASSERT(kr = KERN_SUCCESS). > Source/WebKit2/Platform/mac/SharedMemoryMac.cpp:200
WEBKIT ENCOUNTERED AN INTERNAL ERROR PATCH
So I'll take it out of this patch and will file a bug to cover the other cases. It's not generally used, we're transitioning to the os_log-based logging, almost everyone is now living on a system that supports os_log, etc. There are other instances of double-logging in: NetworkResourceLoader.cpp, WebLoaderStrategy.cpp, WebResourceLoader.cppĪfter having a chat with Chris about this double-logging, he felt that it would be OK to take the old logging out. I didn't want to upset the workflows of people who rely on any of that. The other logging shows up in syslog, the terminal, or Xcode console.
The RELEASE_LOG macros always log and will send their output to the Console app. The idea is to preserve previous logging that people may be using for pre-existing workflows. %s (%x)", nullptr, mach_error_string(kr), kr) > + RELEASE_LOG_ERROR(VirtualMemory, "%p - SharedMemory::allocate: Failed to allocate mach_vm_allocate shared memory (%zu bytes). %s (%x)", size, mach_error_string(kr), kr) > LOG_ERROR("Failed to allocate mach_vm_allocate shared memory (%zu bytes). View in context: > Source/WebKit2/Platform/mac/SharedMemoryMac.cpp:110
0 kommentar(er)
