CVE-2024-57884
mm: vmscan: account for free pages to prevent infinite Loop in throttle_direct_reclaim()
Severity Score
Exploit Likelihood
Affected Versions
Public Exploits
0Exploited in Wild
-Decision
Descriptions
In the Linux kernel, the following vulnerability has been resolved: mm: vmscan: account for free pages to prevent infinite Loop in throttle_direct_reclaim() The task sometimes continues looping in throttle_direct_reclaim() because
allow_direct_reclaim(pgdat) keeps returning false. #0 [ffff80002cb6f8d0] __switch_to at ffff8000080095ac #1 [ffff80002cb6f900] __schedule at ffff800008abbd1c #2 [ffff80002cb6f990] schedule at ffff800008abc50c #3 [ffff80002cb6f9b0] throttle_direct_reclaim at ffff800008273550 #4 [ffff80002cb6fa20] try_to_free_pages at ffff800008277b68 #5 [ffff80002cb6fae0] __alloc_pages_nodemask at ffff8000082c4660 #6 [ffff80002cb6fc50] alloc_pages_vma at ffff8000082e4a98 #7 [ffff80002cb6fca0] do_anonymous_page at ffff80000829f5a8 #8 [ffff80002cb6fce0] __handle_mm_fault at ffff8000082a5974 #9 [ffff80002cb6fd90] handle_mm_fault at ffff8000082a5bd4 At this point, the pgdat contains the following two zones: NODE: 4 ZONE: 0 ADDR: ffff00817fffe540 NAME: "DMA32" SIZE: 20480 MIN/LOW/HIGH: 11/28/45 VM_STAT: NR_FREE_PAGES: 359 NR_ZONE_INACTIVE_ANON: 18813 NR_ZONE_ACTIVE_ANON: 0 NR_ZONE_INACTIVE_FILE: 50 NR_ZONE_ACTIVE_FILE: 0 NR_ZONE_UNEVICTABLE: 0 NR_ZONE_WRITE_PENDING: 0 NR_MLOCK: 0 NR_BOUNCE: 0 NR_ZSPAGES: 0 NR_FREE_CMA_PAGES: 0 NODE: 4 ZONE: 1 ADDR: ffff00817fffec00 NAME: "Normal" SIZE: 8454144 PRESENT: 98304 MIN/LOW/HIGH: 68/166/264 VM_STAT: NR_FREE_PAGES: 146 NR_ZONE_INACTIVE_ANON: 94668 NR_ZONE_ACTIVE_ANON: 3 NR_ZONE_INACTIVE_FILE: 735 NR_ZONE_ACTIVE_FILE: 78 NR_ZONE_UNEVICTABLE: 0 NR_ZONE_WRITE_PENDING: 0 NR_MLOCK: 0 NR_BOUNCE: 0 NR_ZSPAGES: 0 NR_FREE_CMA_PAGES: 0 In allow_direct_reclaim(), while processing ZONE_DMA32, the sum of
inactive/active file-backed pages calculated in zone_reclaimable_pages()
based on the result of zone_page_state_snapshot() is zero. Additionally, since this system lacks swap, the calculation of inactive/
active anonymous pages is skipped. crash> p nr_swap_pages nr_swap_pages = $1937 = { counter = 0 } As a result, ZONE_DMA32 is deemed unreclaimable and skipped, moving on to
the processing of the next zone, ZONE_NORMAL, despite ZONE_DMA32 having
free pages significantly exceeding the high watermark. The problem is that the pgdat->kswapd_failures hasn't been incremented. crash> px ((struct pglist_data *) 0xffff00817fffe540)->kswapd_failures $1935 = 0x0 This is because the node deemed balanced. The node balancing logic in
balance_pgdat() evaluates all zones collectively. If one or more zones
(e.g., ZONE_DMA32) have enough free pages to meet their watermarks, the
entire node is deemed balanced. This causes balance_pgdat() to exit early
before incrementing the kswapd_failures, as it considers the overall
memory state acceptable, even though some zones (like ZONE_NORMAL) remain
under significant pressure. The patch ensures that zone_reclaimable_pages() includes free pages
(NR_FREE_PAGES) in its calculation when no other reclaimable pages are
available (e.g., file-backed or anonymous pages). This change prevents
zones like ZONE_DMA32, which have sufficient free pages, from being
mistakenly deemed unreclaimable. By doing so, the patch ensures proper
node balancing, avoids masking pressure on other zones like ZONE_NORMAL,
and prevents infinite loops in throttle_direct_reclaim() caused by
allow_direct_reclaim(pgdat) repeatedly returning false. The kernel hangs due to a task stuck in throttle_direct_reclaim(), caused
by a node being incorrectly deemed balanced despite pressure in certain
zones, such as ZONE_NORMAL. This issue arises from
zone_reclaimable_pages
---truncated---
En el kernel de Linux, se ha resuelto la siguiente vulnerabilidad: mm: vmscan: tiene en cuenta las páginas libres para evitar un bucle infinito en throttle_direct_reclaim() La tarea a veces continúa en bucle en throttle_direct_reclaim() porque allow_direct_reclaim(pgdat) sigue devolviendo falso. #0 [ffff80002cb6f8d0] __switch_to en ffff8000080095ac #1 [ffff80002cb6f900] __schedule en ffff800008abbd1c #2 [ffff80002cb6f990] schedule en ffff800008abc50c #3 [ffff80002cb6f9b0] throttle_direct_reclaim en ffff800008273550 #4 [ffff80002cb6fa20] try_to_free_pages en ffff800008277b68 #5 [ffff80002cb6fae0] __alloc_pages_nodemask en ffff8000082c4660 #6 [ffff80002cb6fc50] alloc_pages_vma en ffff8000082e4a98 #7 [ffff80002cb6fca0] do_anonymous_page en ffff80000829f5a8 #8 [ffff80002cb6fce0] __handle_mm_fault en ffff8000082a5974 #9 [ffff80002cb6fd90] handle_mm_fault en ffff8000082a5bd4 En este punto, el pgdat contiene las siguientes dos zonas: NODO: 4 ZONA: 0 DIRECCIÓN: ffff00817fffe540 NOMBRE: "DMA32" TAMAÑO: 20480 MÍN./BAJO/ALTO: 11/28/45 ESTADÍSTICA DE VM: NR_PÁGINAS_LIBRES: 359 NR_ZONA_INACTIVA_ANON: 18813 NR_ZONA_ACTIVA_ANON: 0 NR_ZONA_ARCHIVO_INACTIVO: 50 NR_ZONA_ARCHIVO_ACTIVO: 0 NR_ZONA_UNEVICTABLE: 0 NR_ZONA_ESCRITURA_PENDIENTE: 0 NR_MLOCK: 0 NR_BOUNCE: 0 NR_ZSPAGES: 0 NR_PÁGINAS_CMA_LIBRES: 0 NODO: 4 ZONA: 1 DIRECCIÓN: ffff00817fffec00 NOMBRE: "Normal" TAMAÑO: 8454144 PRESENTE: 98304 MÍN./BAJO/ALTO: 68/166/264 ESTADÍSTICO_VM: NR_PÁGINAS_LIBRES: 146 NR_ZONE_INACTIVE_ANON: 94668 NR_ZONE_ACTIVE_ANON: 3 NR_ZONE_INACTIVE_FILE: 735 NR_ZONE_ACTIVE_FILE: 78 NR_ZONE_UNEVICTABLE: 0 NR_ZONE_WRITE_PENDING: 0 NR_MLOCK: 0 NR_BOUNCE: 0 NR_ZSPAGES: 0 NR_FREE_CMA_PAGES: 0 En allow_direct_reclaim(), mientras se procesa ZONE_DMA32, la suma de páginas inactivas/activas respaldadas por archivos calculada en zone_reclaimable_pages() en función del resultado de zone_page_state_snapshot() es cero. Además, dado que este sistema carece de intercambio, se omite el cálculo de páginas anónimas inactivas/activas. crash> p nr_swap_pages nr_swap_pages = $1937 = { counter = 0 } Como resultado, ZONE_DMA32 se considera irrecuperable y se omite, pasando al procesamiento de la siguiente zona, ZONE_NORMAL, a pesar de que ZONE_DMA32 tiene páginas libres que exceden significativamente la marca de agua alta. El problema es que pgdat->kswapd_failures no se ha incrementado. crash> px ((struct pglist_data *) 0xffff00817fffe540)->kswapd_failures $1935 = 0x0 Esto se debe a que el nodo se considera equilibrado. La lógica de equilibrio de nodos en balance_pgdat() evalúa todas las zonas colectivamente. Si una o más zonas (por ejemplo, ZONE_DMA32) tienen suficientes páginas libres para cumplir con sus marcas de agua, todo el nodo se considera equilibrado. Esto hace que balance_pgdat() salga antes de incrementar kswapd_failures, ya que considera que el estado general de la memoria es aceptable, aunque algunas zonas (como ZONE_NORMAL) permanezcan bajo una presión significativa. El parche garantiza que zone_reclaimable_pages() incluya páginas libres (NR_FREE_PAGES) en su cálculo cuando no haya otras páginas recuperables disponibles (por ejemplo, páginas anónimas o respaldadas por archivos). Este cambio evita que zonas como ZONE_DMA32, que tienen suficientes páginas libres, se consideren por error no recuperables. Al hacerlo, el parche garantiza un equilibrio adecuado de los nodos, evita enmascarar la presión en otras zonas como ZONE_NORMAL y evita bucles infinitos en throttle_direct_reclaim() causados por allow_direct_reclaim(pgdat) que devuelve falso repetidamente. El núcleo se cuelga debido a una tarea atascada en throttle_direct_reclaim(), causada por un nodo que se considera incorrectamente equilibrado a pesar de la presión en ciertas zonas, como ZONE_NORMAL. Este problema surge de zone_reclaimable_pages ---truncado---
In the Linux kernel, the following vulnerability has been resolved: mm: vmscan: account for free pages to prevent infinite Loop in throttle_direct_reclaim() The task sometimes continues looping in throttle_direct_reclaim() because allow_direct_reclaim(pgdat) keeps returning false. #0 [ffff80002cb6f8d0] __switch_to at ffff8000080095ac #1 [ffff80002cb6f900] __schedule at ffff800008abbd1c #2 [ffff80002cb6f990] schedule at ffff800008abc50c #3 [ffff80002cb6f9b0] throttle_direct_reclaim at ffff800008273550 #4 [ffff80002cb6fa20] try_to_free_pages at ffff800008277b68 #5 [ffff80002cb6fae0] __alloc_pages_nodemask at ffff8000082c4660 #6 [ffff80002cb6fc50] alloc_pages_vma at ffff8000082e4a98 #7 [ffff80002cb6fca0] do_anonymous_page at ffff80000829f5a8 #8 [ffff80002cb6fce0] __handle_mm_fault at ffff8000082a5974 #9 [ffff80002cb6fd90] handle_mm_fault at ffff8000082a5bd4 At this point, the pgdat contains the following two zones: NODE: 4 ZONE: 0 ADDR: ffff00817fffe540 NAME: "DMA32" SIZE: 20480 MIN/LOW/HIGH: 11/28/45 VM_STAT: NR_FREE_PAGES: 359 NR_ZONE_INACTIVE_ANON: 18813 NR_ZONE_ACTIVE_ANON: 0 NR_ZONE_INACTIVE_FILE: 50 NR_ZONE_ACTIVE_FILE: 0 NR_ZONE_UNEVICTABLE: 0 NR_ZONE_WRITE_PENDING: 0 NR_MLOCK: 0 NR_BOUNCE: 0 NR_ZSPAGES: 0 NR_FREE_CMA_PAGES: 0 NODE: 4 ZONE: 1 ADDR: ffff00817fffec00 NAME: "Normal" SIZE: 8454144 PRESENT: 98304 MIN/LOW/HIGH: 68/166/264 VM_STAT: NR_FREE_PAGES: 146 NR_ZONE_INACTIVE_ANON: 94668 NR_ZONE_ACTIVE_ANON: 3 NR_ZONE_INACTIVE_FILE: 735 NR_ZONE_ACTIVE_FILE: 78 NR_ZONE_UNEVICTABLE: 0 NR_ZONE_WRITE_PENDING: 0 NR_MLOCK: 0 NR_BOUNCE: 0 NR_ZSPAGES: 0 NR_FREE_CMA_PAGES: 0 In allow_direct_reclaim(), while processing ZONE_DMA32, the sum of inactive/active file-backed pages calculated in zone_reclaimable_pages() based on the result of zone_page_state_snapshot() is zero. Additionally, since this system lacks swap, the calculation of inactive/ active anonymous pages is skipped. crash> p nr_swap_pages nr_swap_pages = $1937 = { counter = 0 } As a result, ZONE_DMA32 is deemed unreclaimable and skipped, moving on to the processing of the next zone, ZONE_NORMAL, despite ZONE_DMA32 having free pages significantly exceeding the high watermark. The problem is that the pgdat->kswapd_failures hasn't been incremented. crash> px ((struct pglist_data *) 0xffff00817fffe540)->kswapd_failures $1935 = 0x0 This is because the node deemed balanced. The node balancing logic in balance_pgdat() evaluates all zones collectively. If one or more zones (e.g., ZONE_DMA32) have enough free pages to meet their watermarks, the entire node is deemed balanced. This causes balance_pgdat() to exit early before incrementing the kswapd_failures, as it considers the overall memory state acceptable, even though some zones (like ZONE_NORMAL) remain under significant pressure. The patch ensures that zone_reclaimable_pages() includes free pages (NR_FREE_PAGES) in its calculation when no other reclaimable pages are available (e.g., file-backed or anonymous pages). This change prevents zones like ZONE_DMA32, which have sufficient free pages, from being mistakenly deemed unreclaimable. By doing so, the patch ensures proper node balancing, avoids masking pressure on other zones like ZONE_NORMAL, and prevents infinite loops in throttle_direct_reclaim() caused by allow_direct_reclaim(pgdat) repeatedly returning false. The kernel hangs due to a task stuck in throttle_direct_reclaim(), caused by a node being incorrectly deemed balanced despite pressure in certain zones, such as ZONE_NORMAL. This issue arises from zone_reclaimable_pages ---truncated---
CVSS Scores
SSVC
- Decision:-
Timeline
- 2025-01-11 CVE Reserved
- 2025-01-15 CVE Published
- 2025-01-15 CVE Updated
- 2025-01-16 EPSS Updated
- ---------- Exploited in Wild
- ---------- KEV Due Date
- ---------- First Exploit
CWE
CAPEC
References (8)
URL | Tag | Source |
---|---|---|
https://git.kernel.org/stable/c/5a1c84b404a7176b8b36e2a0041b6f0adb3151a3 | Vuln. Introduced |
URL | Date | SRC |
---|
URL | Date | SRC |
---|
Affected Vendors, Products, and Versions
Vendor | Product | Version | Other | Status | ||||||
---|---|---|---|---|---|---|---|---|---|---|
Vendor | Product | Version | Other | Status | <-- --> | Vendor | Product | Version | Other | Status |
Linux Search vendor "Linux" | Linux Kernel Search vendor "Linux" for product "Linux Kernel" | >= 4.8 < 5.4.289 Search vendor "Linux" for product "Linux Kernel" and version " >= 4.8 < 5.4.289" | en |
Affected
| ||||||
Linux Search vendor "Linux" | Linux Kernel Search vendor "Linux" for product "Linux Kernel" | >= 4.8 < 5.10.233 Search vendor "Linux" for product "Linux Kernel" and version " >= 4.8 < 5.10.233" | en |
Affected
| ||||||
Linux Search vendor "Linux" | Linux Kernel Search vendor "Linux" for product "Linux Kernel" | >= 4.8 < 5.15.176 Search vendor "Linux" for product "Linux Kernel" and version " >= 4.8 < 5.15.176" | en |
Affected
| ||||||
Linux Search vendor "Linux" | Linux Kernel Search vendor "Linux" for product "Linux Kernel" | >= 4.8 < 6.1.124 Search vendor "Linux" for product "Linux Kernel" and version " >= 4.8 < 6.1.124" | en |
Affected
| ||||||
Linux Search vendor "Linux" | Linux Kernel Search vendor "Linux" for product "Linux Kernel" | >= 4.8 < 6.6.70 Search vendor "Linux" for product "Linux Kernel" and version " >= 4.8 < 6.6.70" | en |
Affected
| ||||||
Linux Search vendor "Linux" | Linux Kernel Search vendor "Linux" for product "Linux Kernel" | >= 4.8 < 6.12.9 Search vendor "Linux" for product "Linux Kernel" and version " >= 4.8 < 6.12.9" | en |
Affected
| ||||||
Linux Search vendor "Linux" | Linux Kernel Search vendor "Linux" for product "Linux Kernel" | >= 4.8 < 6.13-rc6 Search vendor "Linux" for product "Linux Kernel" and version " >= 4.8 < 6.13-rc6" | en |
Affected
|