shm_map (9)
Leading comments
Copyright (c) 2011 Hudson River Trading LLC Written by: John H. Baldwin <jhb@FreeBSD.org> All rights reserved. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the...
NAME
shm_map , shm_unmap - map shared memory objects into the kernel's address spaceSYNOPSIS
In sys/types.h In sys/mman.h Ft int Fn shm_map struct file *fp size_t size off_t offset void **memp Ft int Fn shm_unmap struct file *fp void *mem size_t sizeDESCRIPTION
The Fn shm_map and Fn shm_unmap functions provide an API for mapping shared memory objects into the kernel. Shared memory objects are created by shm_open2. These objects can then be passed into the kernel via file descriptors.A shared memory object cannot be shrunk while it is mapped into the kernel. This is to avoid invalidating any pages that may be wired into the kernel's address space. Shared memory objects can still be grown while mapped into the kernel.
To simplify the accounting needed to enforce the above requirement, callers of this API are required to unmap the entire region mapped by Fn shm_map when calling Fn shm_unmap . Unmapping only a portion of the region is not permitted.
The Fn shm_map function locates the shared memory object associated with the open file Fa fp . It maps the region of that object described by Fa offset and Fa size into the kernel's address space. If it succeeds, Fa *memp will be set to the start of the mapping. All pages for the range will be wired into memory upon successful return.
The Fn shm_unmap function unmaps a region previously mapped by Fn shm_map . The Fa mem argument should match the value previously returned in Fa *memp , and the Fa size argument should match the value passed to Fn shm_map .
Note that Fn shm_map will not hold an extra reference on the open file Fa fp for the lifetime of the mapping. Instead, the calling code is required to do this if it wishes to use Fn shm_unmap on the region in the future.
RETURN VALUES
The Fn shm_map and Fn shm_unmap functions return zero on success or an error on failure.EXAMPLES
The following function accepts a file descriptor for a shared memory object. It maps the first sixteen kilobytes of the object into the kernel, performs some work on that address, and then unmaps the address before returning.
int
shm_example(int fd)
{
struct file *fp;
void *mem;
int error;
error = fget(curthread, fd, CAP_MMAP, &fp);
if (error)
return (error);
error = shm_map(fp, 16384, 0, &mem);
if (error) {
fdrop(fp, curthread);
return (error);
}
/* Do something with 'mem'. */
error = shm_unmap(fp, mem, 16384);
fdrop(fp, curthread);
return (error);
}
ERRORS
The Fn shm_map function returns the following errors on failure:- Bq Er EINVAL
- The open file Fa fp is not a shared memory object.
- Bq Er EINVAL
- The requested region described by Fa offset and Fa size extends beyond the end of the shared memory object.
- Bq Er ENOMEM
- Insufficient address space was available.
- Bq Er EACCES
- The shared memory object could not be mapped due to a protection error.
- Bq Er EINVAL
- The shared memory object could not be mapped due to some other VM error.
The Fn shm_unmap function returns the following errors on failure:
- Bq Er EINVAL
- The open file Fa fp is not a shared memory object.
- Bq Er EINVAL
- The address range described by Fa mem and Fa size is not a valid address range.
- Bq Er EINVAL
- The address range described by Fa mem and Fa size is not backed by the shared memory object associated with the open file Fa fp , or the address range does not cover the entire mapping of the object.