In the Linux kernel, the following vulnerability has been resolved:<\/p>\n
pstore\/ram: fix buffer overflow in persistent_ram_save_old()<\/p>\n
persistent_ram_save_old() can be called multiple times for the same
\npersistent_ram_zone (e.g., via ramoops_pstore_read -> ramoops_get_next_prz
\nfor PSTORE_TYPE_DMESG records).<\/p>\n
Currently, the function only allocates prz->old_log when it is NULL,
\nbut it unconditionally updates prz->old_log_size to the current buffer
\nsize and then performs memcpy_fromio() using this new size. If the
\nbuffer size has grown since the first allocation (which can happen
\nacross different kernel boot cycles), this leads to:<\/p>\n
1. A heap buffer overflow (OOB write) in the memcpy_fromio() calls
\n2. A subsequent OOB read when ramoops_pstore_read() accesses the buffer
\n using the incorrect (larger) old_log_size<\/p>\n
The KASAN splat would look similar to:
\n BUG: KASAN: slab-out-of-bounds in ramoops_pstore_read+0x…
\n Read of size N at addr … by task …<\/p>\n
The conditions are likely extremely hard to hit:<\/p>\n
0. Crash with a ramoops write of less-than-record-max-size bytes.
\n 1. Reboot: ramoops registers, pstore_get_records(0) reads old crash,
\n allocates old_log with size X
\n 2. Crash handler registered, timer started (if pstore_update_ms >= 0)
\n 3. Oops happens (non-fatal, system continues)
\n 4. pstore_dump() writes oops via ramoops_pstore_write() size Y (>X)
\n 5. pstore_new_entry = 1, pstore_timer_kick() called
\n 6. System continues running (not a panic oops)
\n 7. Timer fires after pstore_update_ms milliseconds
\n 8. pstore_timefunc() \u2192 schedule_work() \u2192 pstore_dowork() \u2192 pstore_get_records(1)
\n 9. ramoops_get_next_prz() \u2192 persistent_ram_save_old()
\n 10. buffer_size() returns Y, but old_log is X bytes
\n 11. Y > X: memcpy_fromio() overflows heap<\/p>\n
Requirements:
\n – a prior crash record exists that did not fill the record size
\n (almost impossible since the crash handler writes as much as it
\n can possibly fit into the record, capped by max record size and
\n the kmsg buffer almost always exceeds the max record size)
\n – pstore_update_ms >= 0 (disabled by default)
\n – Non-fatal oops (system survives)<\/p>\n
Free and reallocate the buffer when the new size differs from the
\npreviously allocated size. This ensures old_log always has sufficient
\nspace for the data being copied.<\/p>\n","protected":false},"excerpt":{"rendered":"
In the Linux kernel, the following vulnerability has been resolved:<\/p>\n
pstore\/ram: fix buffer overflow in persistent_ram_save_old()<\/p>\n
persistent_ram_save_old() can be called multiple times for the same
\np<\/p>\n","protected":false},"template":"","meta":[],"categories":[],"tags":[],"class_list":["post-15469","vuln","type-vuln","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/vuln\/15469","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/vuln"}],"about":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/types\/vuln"}],"wp:attachment":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=15469"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=15469"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=15469"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}