Repository navigation
Conversation
In MMU mode, shmem_init() allocates POSIX shared memory via shm_open() to back LKL physical memory. Two issues currently exist: 1. Because POSIX shared memory has kernel persistence, /dev/shm/lkl_phys_mem_<pid> remains alive after the process exits or crashes, leading to SHM exhaustion under automated testing or fuzzing. Fix this by unlinking the shm object immediately after shm_open(). 2. shared_mem_fd is currently never closed. If an application repeatedly boots and halts LKL within the same host process, each subsequent shmem_init() overwrites shared_mem_fd without closing the previous one, leaking both host file descriptors and physical memory across LKL restarts. Furthermore, missing FD_CLOEXEC causes child processes spawned via fork/exec to inherit the memory backing descriptor. Fix these by: - Calling shm_unlink() right after shm_open() and setting FD_CLOEXEC. - Defensively closing any existing shared_mem_fd in shmem_init(). - Adding a shmem_cleanup() host operation in lkl_host_operations, called from free_mem() during lkl_sys_halt(), to properly close shared_mem_fd when LKL shuts down. Closes: lkl#629 Signed-off-by: Cheng Lingfei <chenglingfei@foxmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
In MMU mode, shmem_init() allocates POSIX shared memory via shm_open() to back LKL physical memory. Two issues currently exist:
Because POSIX shared memory has kernel persistence, /dev/shm/lkl_phys_mem_ remains alive after the process exits or crashes, leading to SHM exhaustion under automated testing or fuzzing. Fix this by unlinking the shm object immediately after shm_open().
shared_mem_fd is currently never closed. If an application repeatedly boots and halts LKL within the same host process, each subsequent shmem_init() overwrites shared_mem_fd without closing the previous one, leaking both host file descriptors and physical memory across LKL restarts. Furthermore, missing FD_CLOEXEC causes child processes spawned via fork/exec to inherit the memory backing descriptor.
Fix these by:
Closes: #629