Erroneous VM_MINOR_FAULT return from filemap_fault?
There appears to be a path through filemap_fault such that the function returns VM_MINOR_FAULT, even though disk IO was performed. Suppose that the page in question is not already in the page cache (find_lock_page returns NULL) and we trip the MMAP_LOTSAMISS logic, ending up at no_cached_page. Immediately page_cache_read is called, but ret was never set to VM_MAJOR_FAULT (still at the default of VM_MINOR_FAULT). Control jumps back to retry_find, find_lock_page returns the page, and eventually the function returns VM_MINOR_FAULT.
Is there some assumption I'm missing where the execution path I summarized couldn't actually happen? Perhaps this behavior is intentional? Certainly it seems that if IO is performed, the fault should be considered a major fault.
Signature · 9 lines
-Steven Walter <[email protected]> Freedom is the freedom to say that 2 + 2 = 4 B2F1 0ECC E605 7321 E818 7A65 FC81 9777 DC28 9E8F -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"[email protected]"> [email protected] </a>