minix/servers/vm/mem_file.c
Ben Gras 565f13088f make vfs & filesystems use failable copying
Change the kernel to add features to vircopy and safecopies so that
transparent copy fixing won't happen to avoid deadlocks, and such copies
fail with EFAULT.

Transparently making copying work from filesystems (as normally done by
the kernel & VM when copying fails because of missing/readonly memory)
is problematic as it can happen that, for file-mapped ranges, that that
same filesystem that is blocked on the copy request is needed to satisfy
the memory range, leading to deadlock. Dito for VFS itself, if done with
a blocking call.

This change makes the copying done from a filesystem fail in such cases
with EFAULT by VFS adding the CPF_TRY flag to the grants. If a FS call
fails with EFAULT, VFS will then request the range to be made available
to VM after the FS is unblocked, allowing it to be used to satisfy the
range if need be in another VFS thread.

Similarly, for datacopies that VFS itself does, it uses the failable
vircopy variant and callers use a wrapper that talk to VM if necessary
to get the copy to work.

	. kernel: add CPF_TRY flag to safecopies
	. kernel: only request writable ranges to VM for the
	  target buffer when copying fails
	. do copying in VFS TRY-first
	. some fixes in VM to build SANITYCHECK mode
	. add regression test for the cases where
	  - a FS system call needs memory mapped in a process that the
	    FS itself must map.
	  - such a range covers more than one file-mapped region.
	. add 'try' mode to vircopy, physcopy
	. add flags field to copy kernel call messages
	. if CP_FLAG_TRY is set, do not transparently try
	  to fix memory ranges
	. for use by VFS when accessing user buffers to avoid
	  deadlock
	. remove some obsolete backwards compatability assignments
        . VFS: let thread scheduling work for VM requests too
          Allows VFS to make calls to VM while suspending and resuming
          the currently running thread. Does currently not work for the
          main thread.
        . VM: add fix memory range call for use by VFS

Change-Id: I295794269cea51a3163519a9cfe5901301d90b32
2014-07-28 17:05:14 +02:00

266 lines
7 KiB
C

/* This file implements the methods of memory-mapped files. */
#include <assert.h>
#include "proto.h"
#include "vm.h"
#include "region.h"
#include "glo.h"
#include "cache.h"
/* These functions are static so as to not pollute the
* global namespace, and are accessed through their function
* pointers.
*/
static void mappedfile_split(struct vmproc *vmp, struct vir_region *vr,
struct vir_region *r1, struct vir_region *r2);
static int mappedfile_unreference(struct phys_region *pr);
static int mappedfile_pagefault(struct vmproc *vmp, struct vir_region *region,
struct phys_region *ph, int write, vfs_callback_t callback, void *state,
int len, int *io);
static int mappedfile_sanitycheck(struct phys_region *pr, const char *file, int line);
static int mappedfile_writable(struct phys_region *pr);
static int mappedfile_copy(struct vir_region *vr, struct vir_region *newvr);
static int mappedfile_lowshrink(struct vir_region *vr, vir_bytes len);
static void mappedfile_delete(struct vir_region *region);
static int mappedfile_pt_flags(struct vir_region *vr);
struct mem_type mem_type_mappedfile = {
.name = "file-mapped memory",
.ev_unreference = mappedfile_unreference,
.ev_pagefault = mappedfile_pagefault,
.ev_sanitycheck = mappedfile_sanitycheck,
.ev_copy = mappedfile_copy,
.writable = mappedfile_writable,
.ev_split = mappedfile_split,
.ev_lowshrink = mappedfile_lowshrink,
.ev_delete = mappedfile_delete,
.pt_flags = mappedfile_pt_flags,
};
static int mappedfile_pt_flags(struct vir_region *vr){
#if defined(__arm__)
return ARM_VM_PTE_CACHED;
#else
return 0;
#endif
}
static int mappedfile_unreference(struct phys_region *pr)
{
assert(pr->ph->refcount == 0);
if(pr->ph->phys != MAP_NONE)
free_mem(ABS2CLICK(pr->ph->phys), 1);
return OK;
}
static int cow_block(struct vmproc *vmp, struct vir_region *region,
struct phys_region *ph, u16_t clearend)
{
int r;
if((r=mem_cow(region, ph, MAP_NONE, MAP_NONE)) != OK) {
printf("mappedfile_pagefault: COW failed\n");
return r;
}
/* After COW we are a normal piece of anonymous memory. */
ph->memtype = &mem_type_anon;
if(clearend) {
phys_bytes phaddr = ph->ph->phys, po = VM_PAGE_SIZE-clearend;
assert(clearend < VM_PAGE_SIZE);
phaddr += po;
if(sys_memset(NONE, 0, phaddr, clearend) != OK) {
panic("cow_block: clearend failed\n");
}
}
return OK;
}
static int mappedfile_pagefault(struct vmproc *vmp, struct vir_region *region,
struct phys_region *ph, int write, vfs_callback_t cb,
void *state, int statelen, int *io)
{
u32_t allocflags;
int procfd = region->param.file.fdref->fd;
allocflags = vrallocflags(region->flags);
assert(ph->ph->refcount > 0);
assert(region->param.file.inited);
assert(region->param.file.fdref);
assert(region->param.file.fdref->dev != NO_DEV);
/* Totally new block? Create it. */
if(ph->ph->phys == MAP_NONE) {
struct cached_page *cp;
u64_t referenced_offset =
region->param.file.offset + ph->offset;
if(region->param.file.fdref->ino == VMC_NO_INODE) {
cp = find_cached_page_bydev(region->param.file.fdref->dev,
referenced_offset, VMC_NO_INODE, 0, 1);
} else {
cp = find_cached_page_byino(region->param.file.fdref->dev,
region->param.file.fdref->ino, referenced_offset, 1);
}
if(cp) {
int result = OK;
pb_unreferenced(region, ph, 0);
pb_link(ph, cp->page, ph->offset, region);
if(roundup(ph->offset+region->param.file.clearend,
VM_PAGE_SIZE) >= region->length) {
result = cow_block(vmp, region, ph,
region->param.file.clearend);
} else if(result == OK && write) {
result = cow_block(vmp, region, ph, 0);
}
return result;
}
if(!cb) {
#if 0
printf("VM: mem_file: no callback, returning EFAULT\n");
#endif
sys_diagctl_stacktrace(vmp->vm_endpoint);
return EFAULT;
}
if(vfs_request(VMVFSREQ_FDIO, procfd, vmp, referenced_offset,
VM_PAGE_SIZE, cb, NULL, state, statelen) != OK) {
printf("VM: mappedfile_pagefault: vfs_request failed\n");
return ENOMEM;
}
*io = 1;
return SUSPEND;
}
if(!write) {
#if 0
printf("mappedfile_pagefault: nonwrite fault?\n");
#endif
return OK;
}
return cow_block(vmp, region, ph, 0);
}
static int mappedfile_sanitycheck(struct phys_region *pr, const char *file, int line)
{
MYASSERT(usedpages_add(pr->ph->phys, VM_PAGE_SIZE) == OK);
return OK;
}
static int mappedfile_writable(struct phys_region *pr)
{
/* We are never writable. */
return 0;
}
int mappedfile_copy(struct vir_region *vr, struct vir_region *newvr)
{
assert(vr->param.file.inited);
mappedfile_setfile(newvr->parent, newvr, vr->param.file.fdref->fd,
vr->param.file.offset,
vr->param.file.fdref->dev, vr->param.file.fdref->ino,
vr->param.file.clearend, 0, 0);
assert(newvr->param.file.inited);
return OK;
}
int mappedfile_setfile(struct vmproc *owner,
struct vir_region *region, int fd, u64_t offset,
dev_t dev, ino_t ino, u16_t clearend, int prefill, int mayclosefd)
{
vir_bytes vaddr;
struct fdref *newref;
newref = fdref_dedup_or_new(owner, ino, dev, fd, mayclosefd);
assert(newref);
assert(!region->param.file.inited);
assert(dev != NO_DEV);
fdref_ref(newref, region);
region->param.file.offset = offset;
region->param.file.clearend = clearend;
region->param.file.inited = 1;
if(!prefill) return OK;
for(vaddr = 0; vaddr < region->length; vaddr+=VM_PAGE_SIZE) {
struct cached_page *cp = NULL;
struct phys_region *pr;
u64_t referenced_offset = offset + vaddr;
if(roundup(vaddr+region->param.file.clearend,
VM_PAGE_SIZE) >= region->length) {
break;
}
if(ino == VMC_NO_INODE) {
cp = find_cached_page_bydev(dev, referenced_offset,
VMC_NO_INODE, 0, 1);
} else {
cp = find_cached_page_byino(dev, ino,
referenced_offset, 1);
}
if(!cp) continue;
if(!(pr = pb_reference(cp->page, vaddr, region,
&mem_type_mappedfile))) {
printf("mappedfile_setfile: pb_reference failed\n");
break;
}
if(map_ph_writept(region->parent, region, pr) != OK) {
printf("mappedfile_setfile: map_ph_writept failed\n");
break;
}
}
return OK;
}
static void mappedfile_split(struct vmproc *vmp, struct vir_region *vr,
struct vir_region *r1, struct vir_region *r2)
{
assert(!r1->param.file.inited);
assert(!r2->param.file.inited);
assert(vr->param.file.inited);
assert(r1->length + r2->length == vr->length);
assert(vr->def_memtype == &mem_type_mappedfile);
assert(r1->def_memtype == &mem_type_mappedfile);
assert(r2->def_memtype == &mem_type_mappedfile);
r1->param.file = vr->param.file;
r2->param.file = vr->param.file;
fdref_ref(vr->param.file.fdref, r1);
fdref_ref(vr->param.file.fdref, r2);
r1->param.file.clearend = 0;
r2->param.file.offset += r1->length;
assert(r1->param.file.inited);
assert(r2->param.file.inited);
}
static int mappedfile_lowshrink(struct vir_region *vr, vir_bytes len)
{
assert(vr->param.file.inited);
vr->param.file.offset += len;
return OK;
}
static void mappedfile_delete(struct vir_region *region)
{
assert(region->def_memtype == &mem_type_mappedfile);
assert(region->param.file.inited);
assert(region->param.file.fdref);
fdref_deref(region);
region->param.file.inited = 0;
}