cve-2024-49998
Vulnerability from cvelistv5
Published
2024-10-21 18:02
Modified
2025-01-09 15:35
Severity ?
Summary
In the Linux kernel, the following vulnerability has been resolved: net: dsa: improve shutdown sequence Alexander Sverdlin presents 2 problems during shutdown with the lan9303 driver. One is specific to lan9303 and the other just happens to reproduce there. The first problem is that lan9303 is unique among DSA drivers in that it calls dev_get_drvdata() at "arbitrary runtime" (not probe, not shutdown, not remove): phy_state_machine() -> ... -> dsa_user_phy_read() -> ds->ops->phy_read() -> lan9303_phy_read() -> chip->ops->phy_read() -> lan9303_mdio_phy_read() -> dev_get_drvdata() But we never stop the phy_state_machine(), so it may continue to run after dsa_switch_shutdown(). Our common pattern in all DSA drivers is to set drvdata to NULL to suppress the remove() method that may come afterwards. But in this case it will result in an NPD. The second problem is that the way in which we set dp->conduit->dsa_ptr = NULL; is concurrent with receive packet processing. dsa_switch_rcv() checks once whether dev->dsa_ptr is NULL, but afterwards, rather than continuing to use that non-NULL value, dev->dsa_ptr is dereferenced again and again without NULL checks: dsa_conduit_find_user() and many other places. In between dereferences, there is no locking to ensure that what was valid once continues to be valid. Both problems have the common aspect that closing the conduit interface solves them. In the first case, dev_close(conduit) triggers the NETDEV_GOING_DOWN event in dsa_user_netdevice_event() which closes user ports as well. dsa_port_disable_rt() calls phylink_stop(), which synchronously stops the phylink state machine, and ds->ops->phy_read() will thus no longer call into the driver after this point. In the second case, dev_close(conduit) should do this, as per Documentation/networking/driver.rst: | Quiescence | ---------- | | After the ndo_stop routine has been called, the hardware must | not receive or transmit any data. All in flight packets must | be aborted. If necessary, poll or wait for completion of | any reset commands. So it should be sufficient to ensure that later, when we zeroize conduit->dsa_ptr, there will be no concurrent dsa_switch_rcv() call on this conduit. The addition of the netif_device_detach() function is to ensure that ioctls, rtnetlinks and ethtool requests on the user ports no longer propagate down to the driver - we're no longer prepared to handle them. The race condition actually did not exist when commit 0650bf52b31f ("net: dsa: be compatible with masters which unregister on shutdown") first introduced dsa_switch_shutdown(). It was created later, when we stopped unregistering the user interfaces from a bad spot, and we just replaced that sequence with a racy zeroization of conduit->dsa_ptr (one which doesn't ensure that the interfaces aren't up).
References
416baaa9-dc9f-4396-8d5f-8c081fb06d67https://git.kernel.org/stable/c/6c24a03a61a245fe34d47582898331fa034b6ccdPatch
416baaa9-dc9f-4396-8d5f-8c081fb06d67https://git.kernel.org/stable/c/87bd909a7014e32790e8c759d5b7694a95778ca5
416baaa9-dc9f-4396-8d5f-8c081fb06d67https://git.kernel.org/stable/c/ab5d3420a1120950703dbdc33698b28a6ebc3d23Patch
416baaa9-dc9f-4396-8d5f-8c081fb06d67https://git.kernel.org/stable/c/b4a65d479213fe84ecb14e328271251eebe69492Patch
Impacted products
Vendor Product Version
Linux Linux Version: ff45899e732e57088985e3a497b1d9100571c0f5
Version: ee534378f00561207656663d93907583958339ae
Version: ee534378f00561207656663d93907583958339ae
Version: ee534378f00561207656663d93907583958339ae
 Create a notification for this product.
   Linux Linux Version: 5.17
 Create a notification for this product.
Show details on NVD website

To clipboard 

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2024-49998",
                "options": [
                  {
                    "Exploitation": "none"
                  },
                  {
                    "Automatable": "no"
                  },
                  {
                    "Technical Impact": "partial"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2024-10-22T13:30:20.999100Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2024-10-22T13:38:41.547Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "defaultStatus": "unaffected",
          "product": "Linux",
          "programFiles": [
            "net/dsa/dsa.c"
          ],
          "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
          "vendor": "Linux",
          "versions": [
            {
              "lessThan": "87bd909a7014e32790e8c759d5b7694a95778ca5",
              "status": "affected",
              "version": "ff45899e732e57088985e3a497b1d9100571c0f5",
              "versionType": "git"
            },
            {
              "lessThan": "ab5d3420a1120950703dbdc33698b28a6ebc3d23",
              "status": "affected",
              "version": "ee534378f00561207656663d93907583958339ae",
              "versionType": "git"
            },
            {
              "lessThan": "b4a65d479213fe84ecb14e328271251eebe69492",
              "status": "affected",
              "version": "ee534378f00561207656663d93907583958339ae",
              "versionType": "git"
            },
            {
              "lessThan": "6c24a03a61a245fe34d47582898331fa034b6ccd",
              "status": "affected",
              "version": "ee534378f00561207656663d93907583958339ae",
              "versionType": "git"
            }
          ]
        },
        {
          "defaultStatus": "affected",
          "product": "Linux",
          "programFiles": [
            "net/dsa/dsa.c"
          ],
          "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
          "vendor": "Linux",
          "versions": [
            {
              "status": "affected",
              "version": "5.17"
            },
            {
              "lessThan": "5.17",
              "status": "unaffected",
              "version": "0",
              "versionType": "semver"
            },
            {
              "lessThanOrEqual": "5.15.*",
              "status": "unaffected",
              "version": "5.15.176",
              "versionType": "semver"
            },
            {
              "lessThanOrEqual": "6.10.*",
              "status": "unaffected",
              "version": "6.10.14",
              "versionType": "semver"
            },
            {
              "lessThanOrEqual": "6.11.*",
              "status": "unaffected",
              "version": "6.11.3",
              "versionType": "semver"
            },
            {
              "lessThanOrEqual": "*",
              "status": "unaffected",
              "version": "6.12",
              "versionType": "original_commit_for_fix"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "In the Linux kernel, the following vulnerability has been resolved:\n\nnet: dsa: improve shutdown sequence\n\nAlexander Sverdlin presents 2 problems during shutdown with the\nlan9303 driver. One is specific to lan9303 and the other just happens\nto reproduce there.\n\nThe first problem is that lan9303 is unique among DSA drivers in that it\ncalls dev_get_drvdata() at \"arbitrary runtime\" (not probe, not shutdown,\nnot remove):\n\nphy_state_machine()\n-\u003e ...\n   -\u003e dsa_user_phy_read()\n      -\u003e ds-\u003eops-\u003ephy_read()\n         -\u003e lan9303_phy_read()\n            -\u003e chip-\u003eops-\u003ephy_read()\n               -\u003e lan9303_mdio_phy_read()\n                  -\u003e dev_get_drvdata()\n\nBut we never stop the phy_state_machine(), so it may continue to run\nafter dsa_switch_shutdown(). Our common pattern in all DSA drivers is\nto set drvdata to NULL to suppress the remove() method that may come\nafterwards. But in this case it will result in an NPD.\n\nThe second problem is that the way in which we set\ndp-\u003econduit-\u003edsa_ptr = NULL; is concurrent with receive packet\nprocessing. dsa_switch_rcv() checks once whether dev-\u003edsa_ptr is NULL,\nbut afterwards, rather than continuing to use that non-NULL value,\ndev-\u003edsa_ptr is dereferenced again and again without NULL checks:\ndsa_conduit_find_user() and many other places. In between dereferences,\nthere is no locking to ensure that what was valid once continues to be\nvalid.\n\nBoth problems have the common aspect that closing the conduit interface\nsolves them.\n\nIn the first case, dev_close(conduit) triggers the NETDEV_GOING_DOWN\nevent in dsa_user_netdevice_event() which closes user ports as well.\ndsa_port_disable_rt() calls phylink_stop(), which synchronously stops\nthe phylink state machine, and ds-\u003eops-\u003ephy_read() will thus no longer\ncall into the driver after this point.\n\nIn the second case, dev_close(conduit) should do this, as per\nDocumentation/networking/driver.rst:\n\n| Quiescence\n| ----------\n|\n| After the ndo_stop routine has been called, the hardware must\n| not receive or transmit any data.  All in flight packets must\n| be aborted. If necessary, poll or wait for completion of\n| any reset commands.\n\nSo it should be sufficient to ensure that later, when we zeroize\nconduit-\u003edsa_ptr, there will be no concurrent dsa_switch_rcv() call\non this conduit.\n\nThe addition of the netif_device_detach() function is to ensure that\nioctls, rtnetlinks and ethtool requests on the user ports no longer\npropagate down to the driver - we\u0027re no longer prepared to handle them.\n\nThe race condition actually did not exist when commit 0650bf52b31f\n(\"net: dsa: be compatible with masters which unregister on shutdown\")\nfirst introduced dsa_switch_shutdown(). It was created later, when we\nstopped unregistering the user interfaces from a bad spot, and we just\nreplaced that sequence with a racy zeroization of conduit-\u003edsa_ptr\n(one which doesn\u0027t ensure that the interfaces aren\u0027t up)."
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2025-01-09T15:35:28.672Z",
        "orgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
        "shortName": "Linux"
      },
      "references": [
        {
          "url": "https://git.kernel.org/stable/c/87bd909a7014e32790e8c759d5b7694a95778ca5"
        },
        {
          "url": "https://git.kernel.org/stable/c/ab5d3420a1120950703dbdc33698b28a6ebc3d23"
        },
        {
          "url": "https://git.kernel.org/stable/c/b4a65d479213fe84ecb14e328271251eebe69492"
        },
        {
          "url": "https://git.kernel.org/stable/c/6c24a03a61a245fe34d47582898331fa034b6ccd"
        }
      ],
      "title": "net: dsa: improve shutdown sequence",
      "x_generator": {
        "engine": "bippy-5f407fcff5a0"
      }
    }
  },
  "cveMetadata": {
    "assignerOrgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
    "assignerShortName": "Linux",
    "cveId": "CVE-2024-49998",
    "datePublished": "2024-10-21T18:02:38.316Z",
    "dateReserved": "2024-10-21T12:17:06.057Z",
    "dateUpdated": "2025-01-09T15:35:28.672Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.1",
  "vulnerability-lookup:meta": {
    "nvd": "{\"cve\":{\"id\":\"CVE-2024-49998\",\"sourceIdentifier\":\"416baaa9-dc9f-4396-8d5f-8c081fb06d67\",\"published\":\"2024-10-21T18:15:19.907\",\"lastModified\":\"2025-01-09T16:16:20.520\",\"vulnStatus\":\"Modified\",\"cveTags\":[],\"descriptions\":[{\"lang\":\"en\",\"value\":\"In the Linux kernel, the following vulnerability has been resolved:\\n\\nnet: dsa: improve shutdown sequence\\n\\nAlexander Sverdlin presents 2 problems during shutdown with the\\nlan9303 driver. One is specific to lan9303 and the other just happens\\nto reproduce there.\\n\\nThe first problem is that lan9303 is unique among DSA drivers in that it\\ncalls dev_get_drvdata() at \\\"arbitrary runtime\\\" (not probe, not shutdown,\\nnot remove):\\n\\nphy_state_machine()\\n-\u003e ...\\n   -\u003e dsa_user_phy_read()\\n      -\u003e ds-\u003eops-\u003ephy_read()\\n         -\u003e lan9303_phy_read()\\n            -\u003e chip-\u003eops-\u003ephy_read()\\n               -\u003e lan9303_mdio_phy_read()\\n                  -\u003e dev_get_drvdata()\\n\\nBut we never stop the phy_state_machine(), so it may continue to run\\nafter dsa_switch_shutdown(). Our common pattern in all DSA drivers is\\nto set drvdata to NULL to suppress the remove() method that may come\\nafterwards. But in this case it will result in an NPD.\\n\\nThe second problem is that the way in which we set\\ndp-\u003econduit-\u003edsa_ptr = NULL; is concurrent with receive packet\\nprocessing. dsa_switch_rcv() checks once whether dev-\u003edsa_ptr is NULL,\\nbut afterwards, rather than continuing to use that non-NULL value,\\ndev-\u003edsa_ptr is dereferenced again and again without NULL checks:\\ndsa_conduit_find_user() and many other places. In between dereferences,\\nthere is no locking to ensure that what was valid once continues to be\\nvalid.\\n\\nBoth problems have the common aspect that closing the conduit interface\\nsolves them.\\n\\nIn the first case, dev_close(conduit) triggers the NETDEV_GOING_DOWN\\nevent in dsa_user_netdevice_event() which closes user ports as well.\\ndsa_port_disable_rt() calls phylink_stop(), which synchronously stops\\nthe phylink state machine, and ds-\u003eops-\u003ephy_read() will thus no longer\\ncall into the driver after this point.\\n\\nIn the second case, dev_close(conduit) should do this, as per\\nDocumentation/networking/driver.rst:\\n\\n| Quiescence\\n| ----------\\n|\\n| After the ndo_stop routine has been called, the hardware must\\n| not receive or transmit any data.  All in flight packets must\\n| be aborted. If necessary, poll or wait for completion of\\n| any reset commands.\\n\\nSo it should be sufficient to ensure that later, when we zeroize\\nconduit-\u003edsa_ptr, there will be no concurrent dsa_switch_rcv() call\\non this conduit.\\n\\nThe addition of the netif_device_detach() function is to ensure that\\nioctls, rtnetlinks and ethtool requests on the user ports no longer\\npropagate down to the driver - we\u0027re no longer prepared to handle them.\\n\\nThe race condition actually did not exist when commit 0650bf52b31f\\n(\\\"net: dsa: be compatible with masters which unregister on shutdown\\\")\\nfirst introduced dsa_switch_shutdown(). It was created later, when we\\nstopped unregistering the user interfaces from a bad spot, and we just\\nreplaced that sequence with a racy zeroization of conduit-\u003edsa_ptr\\n(one which doesn\u0027t ensure that the interfaces aren\u0027t up).\"},{\"lang\":\"es\",\"value\":\"En el kernel de Linux, se ha resuelto la siguiente vulnerabilidad: net: dsa: mejorar la secuencia de apagado Alexander Sverdlin presenta 2 problemas durante el apagado con el controlador lan9303. Uno es espec\u00edfico de lan9303 y el otro simplemente se reproduce all\u00ed. El primer problema es que lan9303 es \u00fanico entre los controladores DSA en el sentido de que llama a dev_get_drvdata() en un \\\"tiempo de ejecuci\u00f3n arbitrario\\\" (no sondeo, no apagado, no eliminaci\u00f3n): phy_state_machine() -\u0026gt; ... -\u0026gt; dsa_user_phy_read() -\u0026gt; ds-\u0026gt;ops-\u0026gt;phy_read() -\u0026gt; lan9303_phy_read() -\u0026gt; chip-\u0026gt;ops-\u0026gt;phy_read() -\u0026gt; lan9303_mdio_phy_read() -\u0026gt; dev_get_drvdata() Pero nunca detenemos phy_state_machine(), por lo que puede continuar ejecut\u00e1ndose despu\u00e9s de dsa_switch_shutdown(). Nuestro patr\u00f3n com\u00fan en todos los controladores DSA es establecer drvdata en NULL para suprimir el m\u00e9todo remove() que puede venir despu\u00e9s. Pero en este caso resultar\u00e1 en un NPD. El segundo problema es que la forma en que establecemos dp-\u0026gt;conduit-\u0026gt;dsa_ptr = NULL; es concurrente con el procesamiento de paquetes de recepci\u00f3n. dsa_switch_rcv() verifica una vez si dev-\u0026gt;dsa_ptr es NULL, pero despu\u00e9s, en lugar de continuar usando ese valor no NULL, dev-\u0026gt;dsa_ptr se desreferencia una y otra vez sin verificaciones NULL: dsa_conduit_find_user() y muchos otros lugares. Entre desreferencias, no hay bloqueo para asegurar que lo que era v\u00e1lido una vez contin\u00faa siendo v\u00e1lido. Ambos problemas tienen el aspecto com\u00fan de que cerrar la interfaz del conducto los resuelve. En el primer caso, dev_close(conduit) activa el evento NETDEV_GOING_DOWN en dsa_user_netdevice_event() que tambi\u00e9n cierra los puertos de usuario. dsa_port_disable_rt() llama a phylink_stop(), que detiene sincr\u00f3nicamente la m\u00e1quina de estado de phylink, y ds-\u0026gt;ops-\u0026gt;phy_read() ya no llamar\u00e1 al controlador despu\u00e9s de este punto. En el segundo caso, dev_close(conduit) deber\u00eda hacer esto, seg\u00fan Documentation/networking/driver.rst: | Quiescence | ---------- | | Despu\u00e9s de que se haya llamado a la rutina ndo_stop, el hardware no debe recibir ni transmitir ning\u00fan dato. Todos los paquetes en tr\u00e1nsito deben ser abortados. Si es necesario, sondee o espere a que se completen los comandos de reinicio. Por lo tanto, deber\u00eda ser suficiente para garantizar que m\u00e1s adelante, cuando pongamos a cero conduit-\u0026gt;dsa_ptr, no habr\u00e1 ninguna llamada dsa_switch_rcv() concurrente en este conducto. La adici\u00f3n de la funci\u00f3n netif_device_detach() es para garantizar que las solicitudes ioctls, rtnetlinks y ethtool en los puertos de usuario ya no se propaguen al controlador; ya no estamos preparados para manejarlas. La condici\u00f3n de ejecuci\u00f3n en realidad no exist\u00eda cuando el commit 0650bf52b31f (\\\"net: dsa: sea compatible con los maestros que cancelan el registro al apagar\\\") introdujo por primera vez dsa_switch_shutdown(). Se cre\u00f3 m\u00e1s tarde, cuando dejamos de cancelar el registro de las interfaces de usuario desde un lugar incorrecto y simplemente reemplazamos esa secuencia con una puesta a cero de ejecuci\u00f3n de conduit-\u0026gt;dsa_ptr (que no garantiza que las interfaces no est\u00e9n activas).\"}],\"metrics\":{\"cvssMetricV31\":[{\"source\":\"nvd@nist.gov\",\"type\":\"Primary\",\"cvssData\":{\"version\":\"3.1\",\"vectorString\":\"CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:N/A:H\",\"baseScore\":4.7,\"baseSeverity\":\"MEDIUM\",\"attackVector\":\"LOCAL\",\"attackComplexity\":\"HIGH\",\"privilegesRequired\":\"LOW\",\"userInteraction\":\"NONE\",\"scope\":\"UNCHANGED\",\"confidentialityImpact\":\"NONE\",\"integrityImpact\":\"NONE\",\"availabilityImpact\":\"HIGH\"},\"exploitabilityScore\":1.0,\"impactScore\":3.6}]},\"weaknesses\":[{\"source\":\"nvd@nist.gov\",\"type\":\"Primary\",\"description\":[{\"lang\":\"en\",\"value\":\"CWE-367\"},{\"lang\":\"en\",\"value\":\"CWE-476\"}]}],\"configurations\":[{\"nodes\":[{\"operator\":\"OR\",\"negate\":false,\"cpeMatch\":[{\"vulnerable\":true,\"criteria\":\"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*\",\"versionStartIncluding\":\"5.15.155\",\"versionEndExcluding\":\"5.16\",\"matchCriteriaId\":\"5E1DEA80-F82B-404D-9BFF-B251D28EECE1\"},{\"vulnerable\":true,\"criteria\":\"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*\",\"versionStartIncluding\":\"5.16.10\",\"versionEndExcluding\":\"5.17\",\"matchCriteriaId\":\"EC68CAC9-3619-42EC-A9B1-3CB0ACC2EE27\"},{\"vulnerable\":true,\"criteria\":\"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*\",\"versionStartIncluding\":\"5.17\",\"versionEndExcluding\":\"6.10.14\",\"matchCriteriaId\":\"76DE2BAD-A9E3-4933-B8B4-9ED1212F1DF1\"},{\"vulnerable\":true,\"criteria\":\"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*\",\"versionStartIncluding\":\"6.11\",\"versionEndExcluding\":\"6.11.3\",\"matchCriteriaId\":\"54D9C704-D679-41A7-9C40-10A6B1E7FFE9\"},{\"vulnerable\":true,\"criteria\":\"cpe:2.3:o:linux:linux_kernel:6.12:rc1:*:*:*:*:*:*\",\"matchCriteriaId\":\"7F361E1D-580F-4A2D-A509-7615F73167A1\"}]}]}],\"references\":[{\"url\":\"https://git.kernel.org/stable/c/6c24a03a61a245fe34d47582898331fa034b6ccd\",\"source\":\"416baaa9-dc9f-4396-8d5f-8c081fb06d67\",\"tags\":[\"Patch\"]},{\"url\":\"https://git.kernel.org/stable/c/87bd909a7014e32790e8c759d5b7694a95778ca5\",\"source\":\"416baaa9-dc9f-4396-8d5f-8c081fb06d67\"},{\"url\":\"https://git.kernel.org/stable/c/ab5d3420a1120950703dbdc33698b28a6ebc3d23\",\"source\":\"416baaa9-dc9f-4396-8d5f-8c081fb06d67\",\"tags\":[\"Patch\"]},{\"url\":\"https://git.kernel.org/stable/c/b4a65d479213fe84ecb14e328271251eebe69492\",\"source\":\"416baaa9-dc9f-4396-8d5f-8c081fb06d67\",\"tags\":[\"Patch\"]}]}}",
    "vulnrichment": {
      "containers": "{\"adp\": [{\"title\": \"CISA ADP Vulnrichment\", \"metrics\": [{\"other\": {\"type\": \"ssvc\", \"content\": {\"id\": \"CVE-2024-49998\", \"role\": \"CISA Coordinator\", \"options\": [{\"Exploitation\": \"none\"}, {\"Automatable\": \"no\"}, {\"Technical Impact\": \"partial\"}], \"version\": \"2.0.3\", \"timestamp\": \"2024-10-22T13:30:20.999100Z\"}}}], \"providerMetadata\": {\"orgId\": \"134c704f-9b21-4f2e-91b3-4a467353bcc0\", \"shortName\": \"CISA-ADP\", \"dateUpdated\": \"2024-10-22T13:30:24.226Z\"}}], \"cna\": {\"title\": \"net: dsa: improve shutdown sequence\", \"affected\": [{\"repo\": \"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git\", \"vendor\": \"Linux\", \"product\": \"Linux\", \"versions\": [{\"status\": \"affected\", \"version\": \"ff45899e732e57088985e3a497b1d9100571c0f5\", \"lessThan\": \"87bd909a7014e32790e8c759d5b7694a95778ca5\", \"versionType\": \"git\"}, {\"status\": \"affected\", \"version\": \"ee534378f00561207656663d93907583958339ae\", \"lessThan\": \"ab5d3420a1120950703dbdc33698b28a6ebc3d23\", \"versionType\": \"git\"}, {\"status\": \"affected\", \"version\": \"ee534378f00561207656663d93907583958339ae\", \"lessThan\": \"b4a65d479213fe84ecb14e328271251eebe69492\", \"versionType\": \"git\"}, {\"status\": \"affected\", \"version\": \"ee534378f00561207656663d93907583958339ae\", \"lessThan\": \"6c24a03a61a245fe34d47582898331fa034b6ccd\", \"versionType\": \"git\"}], \"programFiles\": [\"net/dsa/dsa.c\"], \"defaultStatus\": \"unaffected\"}, {\"repo\": \"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git\", \"vendor\": \"Linux\", \"product\": \"Linux\", \"versions\": [{\"status\": \"affected\", \"version\": \"5.17\"}, {\"status\": \"unaffected\", \"version\": \"0\", \"lessThan\": \"5.17\", \"versionType\": \"semver\"}, {\"status\": \"unaffected\", \"version\": \"5.15.176\", \"versionType\": \"semver\", \"lessThanOrEqual\": \"5.15.*\"}, {\"status\": \"unaffected\", \"version\": \"6.10.14\", \"versionType\": \"semver\", \"lessThanOrEqual\": \"6.10.*\"}, {\"status\": \"unaffected\", \"version\": \"6.11.3\", \"versionType\": \"semver\", \"lessThanOrEqual\": \"6.11.*\"}, {\"status\": \"unaffected\", \"version\": \"6.12\", \"versionType\": \"original_commit_for_fix\", \"lessThanOrEqual\": \"*\"}], \"programFiles\": [\"net/dsa/dsa.c\"], \"defaultStatus\": \"affected\"}], \"references\": [{\"url\": \"https://git.kernel.org/stable/c/87bd909a7014e32790e8c759d5b7694a95778ca5\"}, {\"url\": \"https://git.kernel.org/stable/c/ab5d3420a1120950703dbdc33698b28a6ebc3d23\"}, {\"url\": \"https://git.kernel.org/stable/c/b4a65d479213fe84ecb14e328271251eebe69492\"}, {\"url\": \"https://git.kernel.org/stable/c/6c24a03a61a245fe34d47582898331fa034b6ccd\"}], \"x_generator\": {\"engine\": \"bippy-5f407fcff5a0\"}, \"descriptions\": [{\"lang\": \"en\", \"value\": \"In the Linux kernel, the following vulnerability has been resolved:\\n\\nnet: dsa: improve shutdown sequence\\n\\nAlexander Sverdlin presents 2 problems during shutdown with the\\nlan9303 driver. One is specific to lan9303 and the other just happens\\nto reproduce there.\\n\\nThe first problem is that lan9303 is unique among DSA drivers in that it\\ncalls dev_get_drvdata() at \\\"arbitrary runtime\\\" (not probe, not shutdown,\\nnot remove):\\n\\nphy_state_machine()\\n-\u003e ...\\n   -\u003e dsa_user_phy_read()\\n      -\u003e ds-\u003eops-\u003ephy_read()\\n         -\u003e lan9303_phy_read()\\n            -\u003e chip-\u003eops-\u003ephy_read()\\n               -\u003e lan9303_mdio_phy_read()\\n                  -\u003e dev_get_drvdata()\\n\\nBut we never stop the phy_state_machine(), so it may continue to run\\nafter dsa_switch_shutdown(). Our common pattern in all DSA drivers is\\nto set drvdata to NULL to suppress the remove() method that may come\\nafterwards. But in this case it will result in an NPD.\\n\\nThe second problem is that the way in which we set\\ndp-\u003econduit-\u003edsa_ptr = NULL; is concurrent with receive packet\\nprocessing. dsa_switch_rcv() checks once whether dev-\u003edsa_ptr is NULL,\\nbut afterwards, rather than continuing to use that non-NULL value,\\ndev-\u003edsa_ptr is dereferenced again and again without NULL checks:\\ndsa_conduit_find_user() and many other places. In between dereferences,\\nthere is no locking to ensure that what was valid once continues to be\\nvalid.\\n\\nBoth problems have the common aspect that closing the conduit interface\\nsolves them.\\n\\nIn the first case, dev_close(conduit) triggers the NETDEV_GOING_DOWN\\nevent in dsa_user_netdevice_event() which closes user ports as well.\\ndsa_port_disable_rt() calls phylink_stop(), which synchronously stops\\nthe phylink state machine, and ds-\u003eops-\u003ephy_read() will thus no longer\\ncall into the driver after this point.\\n\\nIn the second case, dev_close(conduit) should do this, as per\\nDocumentation/networking/driver.rst:\\n\\n| Quiescence\\n| ----------\\n|\\n| After the ndo_stop routine has been called, the hardware must\\n| not receive or transmit any data.  All in flight packets must\\n| be aborted. If necessary, poll or wait for completion of\\n| any reset commands.\\n\\nSo it should be sufficient to ensure that later, when we zeroize\\nconduit-\u003edsa_ptr, there will be no concurrent dsa_switch_rcv() call\\non this conduit.\\n\\nThe addition of the netif_device_detach() function is to ensure that\\nioctls, rtnetlinks and ethtool requests on the user ports no longer\\npropagate down to the driver - we\u0027re no longer prepared to handle them.\\n\\nThe race condition actually did not exist when commit 0650bf52b31f\\n(\\\"net: dsa: be compatible with masters which unregister on shutdown\\\")\\nfirst introduced dsa_switch_shutdown(). It was created later, when we\\nstopped unregistering the user interfaces from a bad spot, and we just\\nreplaced that sequence with a racy zeroization of conduit-\u003edsa_ptr\\n(one which doesn\u0027t ensure that the interfaces aren\u0027t up).\"}], \"providerMetadata\": {\"orgId\": \"416baaa9-dc9f-4396-8d5f-8c081fb06d67\", \"shortName\": \"Linux\", \"dateUpdated\": \"2025-01-09T15:35:28.672Z\"}}}",
      "cveMetadata": "{\"cveId\": \"CVE-2024-49998\", \"state\": \"PUBLISHED\", \"dateUpdated\": \"2025-01-09T15:35:28.672Z\", \"dateReserved\": \"2024-10-21T12:17:06.057Z\", \"assignerOrgId\": \"416baaa9-dc9f-4396-8d5f-8c081fb06d67\", \"datePublished\": \"2024-10-21T18:02:38.316Z\", \"assignerShortName\": \"Linux\"}",
      "dataType": "CVE_RECORD",
      "dataVersion": "5.1"
    }
  }
}


Log in or create an account to share your comment.




Tags
Taxonomy of the tags.




Sightings

Author Source Type Date

Nomenclature

  • Seen: The vulnerability was mentioned, discussed, or seen somewhere by the user.
  • Confirmed: The vulnerability is confirmed from an analyst perspective.
  • Exploited: This vulnerability was exploited and seen by the user reporting the sighting.
  • Patched: This vulnerability was successfully patched by the user reporting the sighting.
  • Not exploited: This vulnerability was not exploited or seen by the user reporting the sighting.
  • Not confirmed: The user expresses doubt about the veracity of the vulnerability.
  • Not patched: This vulnerability was not successfully patched by the user reporting the sighting.