Add more GPUProcess release logging to facilitate debugging.
Created attachment 426420 [details] Patch
Comment on attachment 426420 [details] Patch Is it ever a good idea to log pointer values in a release build?
(In reply to Darin Adler from comment #2) > Comment on attachment 426420 [details] > Patch > > Is it ever a good idea to log pointer values in a release build? This is our common pattern thus far. E.g.: #define RELEASE_LOG_IF_ALLOWED(channel, fmt, ...) RELEASE_LOG_IF(isAlwaysOnLoggingAllowed(), channel, "%p - [pageProxyID=%" PRIu64 ", webPageID=%" PRIu64 ", PID=%i] WebPageProxy::" fmt, this, m_identifier.toUInt64(), m_webPageID.toUInt64(), m_process->processIdentifier(), ##__VA_ARGS__) #define RELEASE_LOG_ERROR_IF_ALLOWED(channel, fmt, ...) RELEASE_LOG_ERROR_IF(isAlwaysOnLoggingAllowed(), channel, "%p - [pageProxyID=%" PRIu64 ", webPageID=%" PRIu64 ", PID=%i] WebPageProxy::" fmt, this, m_identifier.toUInt64(), m_webPageID.toUInt64(), m_process->processIdentifier(), ##__VA_ARGS__)
Committed r276257 (236739@main): <https://commits.webkit.org/236739@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 426420 [details].
rdar://76849207