{"id":"AZL-104595","summary":"CVE-2026-98094 affecting package kernel 6.6.157.1-1","details":"In the Linux kernel, the following vulnerability has been resolved:\n\nstaging: fbtft: make dirty_lock IRQ-safe\n\nfbtft_mkdirty() can be reached from the fbcon rendering path while\nprocessing printk() in hardirq context. Meanwhile, dirty_lock is also\ntaken by fbtft_deferred_io() in workqueue context with local interrupts\nenabled.\n\nLockdep reports a possible IRQ lock inversion involving dirty_lock and\nconsole_owner. A hardirq can interrupt a CPU holding dirty_lock and\nenter the console rendering path, which can attempt to acquire\ndirty_lock again.\n\nThe following lockdep report was observed on an RK3566 system with\nCONFIG_PROVE_LOCKING enabled:\n\n  WARNING: possible irq lock inversion dependency detected\n  swapper/2/0 just changed the state of lock:\n  (console_owner){-...}-{0:0}\n  but this lock took another, HARDIRQ-unsafe lock in the past:\n  (&par-\u003edirty_lock){+.+.}-{2:2}\n\n  CPU0                    CPU1\n  ----                    ----\n  lock(&par-\u003edirty_lock);\n                         local_irq_disable();\n                         lock(console_owner);\n                         lock(&par-\u003edirty_lock);\n  \u003cInterrupt\u003e\n    lock(console_owner);\n\n  *** DEADLOCK ***\n\nUse spin_lock_irqsave() for fbtft_mkdirty() and spin_lock_irq() for\nfbtft_deferred_io(). They only access the dirty line range, so the\nIRQ-off regions remain short.","modified":"2026-09-26T14:16:15.556928530Z","published":"2026-09-25T11:17:39Z","upstream":["CVE-2026-98094"],"references":[{"type":"WEB","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-98094"}],"affected":[{"package":{"name":"kernel","ecosystem":"Azure Linux:3","purl":"pkg:rpm/azure-linux/kernel"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"6.6.157.1-1"}]}],"database_specific":{"source":"https://github.com/microsoft/AzureLinuxVulnerabilityData/blob/main/osv/AZL-104595.json"}}],"schema_version":"1.9.0"}