freebsd-skq/sys/i386/linux/imgact_linux.c
John Baldwin 91d5354a2c Locking for the per-process resource limits structure.
- struct plimit includes a mutex to protect a reference count.  The plimit
  structure is treated similarly to struct ucred in that is is always copy
  on write, so having a reference to a structure is sufficient to read from
  it without needing a further lock.
- The proc lock protects the p_limit pointer and must be held while reading
  limits from a process to keep the limit structure from changing out from
  under you while reading from it.
- Various global limits that are ints are not protected by a lock since
  int writes are atomic on all the archs we support and thus a lock
  wouldn't buy us anything.
- All accesses to individual resource limits from a process are abstracted
  behind a simple lim_rlimit(), lim_max(), and lim_cur() API that return
  either an rlimit, or the current or max individual limit of the specified
  resource from a process.
- dosetrlimit() was renamed to kern_setrlimit() to match existing style of
  other similar syscall helper functions.
- The alpha OSF/1 compat layer no longer calls getrlimit() and setrlimit()
  (it didn't used the stackgap when it should have) but uses lim_rlimit()
  and kern_setrlimit() instead.
- The svr4 compat no longer uses the stackgap for resource limits calls,
  but uses lim_rlimit() and kern_setrlimit() instead.
- The ibcs2 compat no longer uses the stackgap for resource limits.  It
  also no longer uses the stackgap for accessing sysctl's for the
  ibcs2_sysconf() syscall but uses kernel_sysctl() instead.  As a result,
  ibcs2_sysconf() no longer needs Giant.
- The p_rlimit macro no longer exists.

Submitted by:	mtm (mostly, I only did a few cleanups and catchups)
Tested on:	i386
Compiled on:	alpha, amd64
2004-02-04 21:52:57 +00:00

250 lines
7.1 KiB
C

/*-
* Copyright (c) 1994-1996 Søren Schmidt
* All rights reserved.
*
* Based heavily on /sys/kern/imgact_aout.c which is:
* Copyright (c) 1993, David Greenman
*
* 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
* in this position and unchanged.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. The name of the author may not be used to endorse or promote products
* derived from this software without specific prior written permission
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/exec.h>
#include <sys/imgact.h>
#include <sys/imgact_aout.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/mman.h>
#include <sys/mutex.h>
#include <sys/proc.h>
#include <sys/resourcevar.h>
#include <sys/vnode.h>
#include <vm/vm.h>
#include <vm/vm_kern.h>
#include <vm/vm_param.h>
#include <vm/pmap.h>
#include <vm/vm_map.h>
#include <vm/vm_extern.h>
#include <i386/linux/linux.h>
static int exec_linux_imgact(struct image_params *iparams);
static int
exec_linux_imgact(struct image_params *imgp)
{
const struct exec *a_out = (const struct exec *) imgp->image_header;
struct vmspace *vmspace;
vm_offset_t vmaddr;
unsigned long virtual_offset, file_offset;
vm_offset_t buffer;
unsigned long bss_size;
struct thread *td = curthread;
int error;
if (((a_out->a_magic >> 16) & 0xff) != 0x64)
return -1;
/*
* Set file/virtual offset based on a.out variant.
*/
switch ((int)(a_out->a_magic & 0xffff)) {
case 0413:
virtual_offset = 0;
file_offset = 1024;
break;
case 0314:
virtual_offset = 4096;
file_offset = 0;
break;
default:
return (-1);
}
bss_size = round_page(a_out->a_bss);
#ifdef DEBUG
printf("imgact: text: %08lx, data: %08lx, bss: %08lx\n",
(u_long)a_out->a_text, (u_long)a_out->a_data, bss_size);
#endif
/*
* Check various fields in header for validity/bounds.
*/
if (a_out->a_entry < virtual_offset ||
a_out->a_entry >= virtual_offset + a_out->a_text ||
a_out->a_text & PAGE_MASK || a_out->a_data & PAGE_MASK)
return (-1);
/* text + data can't exceed file size */
if (a_out->a_data + a_out->a_text > imgp->attr->va_size)
return (EFAULT);
/*
* text/data/bss must not exceed limits
*/
PROC_LOCK(imgp->proc);
if (a_out->a_text > maxtsiz ||
a_out->a_data + bss_size > lim_cur(imgp->proc, RLIMIT_DATA)) {
PROC_UNLOCK(imgp->proc);
return (ENOMEM);
}
PROC_UNLOCK(imgp->proc);
VOP_UNLOCK(imgp->vp, 0, td);
/* copy in arguments and/or environment from old process */
error = exec_extract_strings(imgp);
if (error)
goto fail;
/*
* Destroy old process VM and create a new one (with a new stack)
*/
exec_new_vmspace(imgp, &linux_sysvec);
vmspace = imgp->proc->p_vmspace;
/*
* Check if file_offset page aligned,.
* Currently we cannot handle misalinged file offsets,
* and so we read in the entire image (what a waste).
*/
if (file_offset & PAGE_MASK) {
#ifdef DEBUG
printf("imgact: Non page aligned binary %lu\n", file_offset);
#endif
/*
* Map text+data+bss read/write/execute
*/
vmaddr = virtual_offset;
error = vm_map_find(&vmspace->vm_map, NULL, 0, &vmaddr,
a_out->a_text + a_out->a_data + bss_size, FALSE,
VM_PROT_ALL, VM_PROT_ALL, 0);
if (error)
goto fail;
error = vm_mmap(kernel_map, &buffer,
round_page(a_out->a_text + a_out->a_data + file_offset),
VM_PROT_READ, VM_PROT_READ, 0,
(caddr_t) imgp->vp, trunc_page(file_offset));
if (error)
goto fail;
error = copyout((void *)(uintptr_t)(buffer + file_offset),
(void *)vmaddr, a_out->a_text + a_out->a_data);
vm_map_remove(kernel_map, buffer,
buffer + round_page(a_out->a_text + a_out->a_data + file_offset));
if (error)
goto fail;
/*
* remove write enable on the 'text' part
*/
error = vm_map_protect(&vmspace->vm_map,
vmaddr,
vmaddr + a_out->a_text,
VM_PROT_EXECUTE|VM_PROT_READ,
TRUE);
if (error)
goto fail;
}
else {
#ifdef DEBUG
printf("imgact: Page aligned binary %lu\n", file_offset);
#endif
/*
* Map text+data read/execute
*/
vmaddr = virtual_offset;
error = vm_mmap(&vmspace->vm_map, &vmaddr,
a_out->a_text + a_out->a_data,
VM_PROT_READ | VM_PROT_EXECUTE,
VM_PROT_ALL,
MAP_PRIVATE | MAP_FIXED,
(caddr_t)imgp->vp, file_offset);
if (error)
goto fail;
#ifdef DEBUG
printf("imgact: startaddr=%08lx, length=%08lx\n",
(u_long)vmaddr, a_out->a_text + a_out->a_data);
#endif
/*
* allow read/write of data
*/
error = vm_map_protect(&vmspace->vm_map,
vmaddr + a_out->a_text,
vmaddr + a_out->a_text + a_out->a_data,
VM_PROT_ALL,
FALSE);
if (error)
goto fail;
/*
* Allocate anon demand-zeroed area for uninitialized data
*/
if (bss_size != 0) {
vmaddr = virtual_offset + a_out->a_text + a_out->a_data;
error = vm_map_find(&vmspace->vm_map, NULL, 0, &vmaddr,
bss_size, FALSE, VM_PROT_ALL, VM_PROT_ALL, 0);
if (error)
goto fail;
#ifdef DEBUG
printf("imgact: bssaddr=%08lx, length=%08lx\n",
(u_long)vmaddr, bss_size);
#endif
}
/* Indicate that this file should not be modified */
mp_fixme("Unlocked v_flag access");
imgp->vp->v_vflag |= VV_TEXT;
}
/* Fill in process VM information */
vmspace->vm_tsize = round_page(a_out->a_text) >> PAGE_SHIFT;
vmspace->vm_dsize = round_page(a_out->a_data + bss_size) >> PAGE_SHIFT;
vmspace->vm_taddr = (caddr_t)(void *)(uintptr_t)virtual_offset;
vmspace->vm_daddr = (caddr_t)(void *)(uintptr_t)
(virtual_offset + a_out->a_text);
/* Fill in image_params */
imgp->interpreted = 0;
imgp->entry_addr = a_out->a_entry;
imgp->proc->p_sysent = &linux_sysvec;
fail:
vn_lock(imgp->vp, LK_EXCLUSIVE | LK_RETRY, td);
return (error);
}
/*
* Tell kern_execve.c about it, with a little help from the linker.
*/
static struct execsw linux_execsw = { exec_linux_imgact, "linux a.out" };
EXEC_SET(linuxaout, linux_execsw);