CVE-2026-98118: netfs: Fix readahead synchronisation issues by loading all folios upfront
In the Linux kernel, the following vulnerability has been resolved:
netfs: Fix readahead synchronisation issues by loading all folios upfront
There are some synchronisation issues that derive from the app thread adding more folios to the rolling buffer whilst the collector thread is looking at them or trying to clear them, such as determining the setting of frontfolioorder when the next folio hasn't been added yet,
The reason for the rolling buffer approach is that loading the buffer upfront and then dropping all the refs just acquired is quite a slow operation, and loading progressively allows some of the cost to be deferred until after at least some of the I/O is started.
Instead, a better way is to load all the folios into the rolling buffer upfront - and then drop the refs later, once the I/O is in progress. (Even better would be for the refs not to be there at all.)
Fix this by changing the rolling buffer loader to load all the folios selected by the VM for readahead upfront into the folio queue. The folio queue is allocated a batch worth at a time as we don't know how many folios are involved (the readaheadcontrol struct, alas, has a page count, not a folio count).
The folio refs acquired from readahead are then dropped in bulk once the first subrequest is dispatched as it's quite a slow operation. The collector waits for NETFSRREQNEEDPUTRAREFS to be cleared so that it doesn't unlock folios before the xarray has been scanned for them.
This simplifies the buffer handling later and isn't noticeably slower as the xarray doesn't need to be modified and the folios are all already pre-locked.
Affected Software
Remediation
Recommended actions to resolve this vulnerability, in priority order.
- Compensating control
Change the rolling buffer loader to load all folios selected for readahead into the folio queue upfront, then drop the readahead-acquired folio references in bulk once I/O is in progress.
Event History
Frequently Asked Questions
Which systems are affected by this issue?
The issue is in the Linux kernel netfs readahead path. Systems using the affected kernel code for network filesystem readahead are the relevant exposure population; the provided data does not identify specific distributions, kernel versions, or filesystem configurations.
What condition leads to the synchronization problem?
The problem arises when an application thread adds folios to a rolling buffer while a collector thread is examining or clearing that buffer. This can cause synchronization errors, including attempts to determine front_folio_order before the next folio has been added.
What does the fix change?
The fix loads all folios selected by the VM for readahead into the folio queue before processing begins, rather than progressively adding them while collection may be underway. References acquired during loading are dropped later after I/O is in progress.