Greg Kroah-Hartman, kes vastutab Linuxi tuuma stabiilsete ja \"staging\" harude toetamise eest ning on peamine hooldaja 16 tuuma alam s\u00fcsteemi, esitles ettekannet Rust Week 2026 konverentsil, kus kirjeldas, kuidas Rust keel aitab v\u00e4ltida tuumas tekkivaid haavatavusi, mis tulenevad t\u00fc\u00fcpilistest C-keele arendajate vigadest m\u00e4lu, lukustuse, vigade k\u00e4sitlemise ja ebausaldusv\u00e4\u00e4rsete andmete t\u00f6\u00f6tlemise alal. Rusti peamiseks eelduseks peetakse selliste vigade avastamist kompileerimise etapis, mitte inimeste koodi \u00fclevaatamisel. Sellegipoolest ei peeta Rusti universaalseks lahenduseks, mis suudaks k\u00f5ik probleemid k\u00f5rvaldada, ning keegi ei plaani tuuma Rusti kirjutada: oodatakse Rusti j\u00e4rkj\u00e4rgulist rakendamist uute draiverite ja alam s\u00fcsteemide kaudu.<\/p>\n
As an example of kernel errors that could be avoided with Rust, an issue in the Bluetooth subsystem that went unnoticed for 15 years and a problem in the Xen hypervisor were mentioned. In the first case, a developer dereferenced a pointer without checking, and in the second case, a lock was forgotten in the error handling code.
\nAccording to Greg, most errors in the kernel are caused by such minor issues that accumulate over time and surface as vulnerabilities. In Rust, many of these problems are prevented by the compiler. For instance, Rust abstractions for locks in the kernel only allow access to internal pointers of structures after acquiring the corresponding lock, which is released automatically. Without acquiring the lock, it is impossible to access the structure pointers in Rust. <\/p>\n
Greg believes that such capabilities in Rust would have prevented 60% of the errors detected in the kernel, and the checks performed by the compiler would save maintainers time from having to discuss error handling correctness and the justification for placing locks in the right spots with the authors. Moreover, the introduction of Rust support has already positively influenced the C code in the kernel by tidying up the C code and interfaces, as well as borrowing some development techniques (for example, locks with limited scope have been implemented).<\/p>\n
T\u00e4nu t\u00fcpihalduss\u00fcsteemile, mis tagab kindlate reeglite j\u00e4rgimise, ja pideva integreerimise s\u00fcsteemide rakendamisele, mis kontrollivad koodi kogumise ajal, saavad muutuste vaatlejad Rustis keskenduda t\u00f6\u00f6loogika kontrollimisele, mitte ressurssidega manipuleerimise j\u00e4lgimisele. Rusti kasutamine v\u00f5imaldab ka andmete, mis tulevad varustuselt v\u00f5i v\u00e4liste s\u00fcsteemidelt, eest hoolikamalt suhtuda. Seda saavutatakse, eristades usaldusv\u00e4\u00e4rseid ja usaldamatuid andmeid t\u00fcpihaldustasandil: arendaja peab lihtsalt anal\u00fc\u00fcsima, kui liikuda usaldamatust olekust usaldatavasse seisundisse.<\/p>\n
Viimasel ajal avaldab Linuxi tuumaturvalisuse meeskond iga p\u00e4ev umbes 13 haavatavuse raporteid, mis tundub varasemate haavatavuste tuvastamise d\u00fcnaamikaga v\u00f5rreldes m\u00f5istuse maalt v\u00e4lja minevana (n\u00e4iteks eilse p\u00e4eva jooksul avaldati 277 haavatavuse raporteid tuumas). Gregi arvates on Rusti kasutamine \u00fcks t\u00f5husamaid viise v\u00e4hendada tuumas vigu, mis on p\u00f5hjustatud traditsioonilistest vigadest, mis on seotud vigade k\u00e4sitlemise ja ressursside haldamisega. Rusti toetus on tuumas juba katsetamisest kaugemale j\u00f5udnud ja m\u00f6\u00f6dunud aasta l\u00f5pus tunnustati seda tuuma p\u00f5hifunktsioonina. <\/p>\n
![\"M\u00e4ngi](\"https:\/\/img.youtube.com\/vi\/0vhGWclF7LU\/hqdefault.jpg\")
\r\n