I've been attempting to use ReadFileScatter today in my code (which sounds like exactly what I need), so far without a lot of luck. Google'ing the internet for what goes wrong doesn't give me much insight.
The documentation states:
The array must contain enough elements to store nNumberOfBytesToRead bytes of data, plus one element for the terminating NULL. For example, if there are 40 KB to be read and the page size is 4 KB, the array must have 11 elements that includes 10 for the data and one for the NULL.
Each buffer must be at least the size of a system memory page and must be aligned on a system memory page size boundary. The system reads one system memory page of data into each buffer.
The function stores the data in the buffers in sequential order. For example, it stores data into the first buffer, then into the second buffer, and so on until each buffer is filled and all the data is stored, or there are no more buffers.
So far I've been attempting to do just that. I allocated a bunch of bytes using VirtualAlloc (which ensures the page boundary constraint), add a terminator NULL to the list, ensure the data on disk is on the system boundary (and implicitly a disk sector size boundary) as well and issue the call.
Without further due, here's the minimum test case in C++:
// Setup: c:\tmp\test.dat is a file with at least 12K of stuff.
// I attempt to read page 2/3, e.g. offset [4096-4096+8192>
// TEST:
SYSTEM_INFO systemInfo;
GetSystemInfo(&systemInfo);
auto pageSize = systemInfo.dwPageSize;
std::cout << "Page size: "<< pageSize << std::endl;
// Allocate 2 pages that are aligned with one in the middle:
auto buffer = reinterpret_cast<char*>(VirtualAlloc(NULL, pageSize * 3, MEM_RESERVE | MEM_COMMIT, PAGE_READWRITE));
// Create read buffer:
std::vector<FILE_SEGMENT_ELEMENT> elements;
{
FILE_SEGMENT_ELEMENT element1;
element1.Buffer = buffer;
elements.push_back(element1);
}
{
FILE_SEGMENT_ELEMENT element2;
element2.Buffer = buffer + pageSize * 2;
elements.push_back(element2);
}
{
FILE_SEGMENT_ELEMENT terminator;
terminator.Buffer = NULL;
elements.push_back(terminator);
}
// [..] Physical sector size is normally checked as well. In my case it's 512 bytes,
// so I guess that's irrelevant here.
//
// Open file:
auto fileHandle = CreateFile(
"c:\\tmp\\test.dat",
GENERIC_READ | GENERIC_WRITE,
FILE_SHARE_READ,
NULL,
OPEN_ALWAYS,
FILE_FLAG_NO_BUFFERING | FILE_FLAG_WRITE_THROUGH,
NULL);
auto err = GetLastError();
if (err != ERROR_ALREADY_EXISTS && err != ERROR_SUCCESS)
{
throw std::exception(); // FIXME.
}
OVERLAPPED overlapped;
memset(&overlapped, 0, sizeof(OVERLAPPED));
LARGE_INTEGER tmp;
tmp.QuadPart = 4096; // Read from disk page 1
overlapped.Offset = tmp.LowPart;
overlapped.OffsetHigh = tmp.HighPart;
overlapped.hEvent = CreateEvent(NULL, TRUE, FALSE, NULL);
auto succes = ReadFileScatter(fileHandle, elements.data(), pageSize * 2, NULL, &overlapped);
err = GetLastError();
if (!succes && err != ERROR_IO_PENDING && err != ERROR_SUCCESS)
{
throw std::exception(); // The call always ends up here with error 87: Invalid parameter
}
WaitForSingleObject(overlapped.hEvent, INFINITE);
std::cout << "Call succeeded!" << std::endl;
// FIXME: Proper exception handling.
// Clean up:
VirtualFree(buffer, pageSize * 3, MEM_DECOMMIT | MEM_RELEASE);
CloseHandle(overlapped.hEvent);
CloseHandle(fileHandle);
In the code, the error is noted with the comment // The call always ends up here with error 87: Invalid parameter. However, as far as I can see, I check all the boxes that they describe on MSDN... so...
What am I doing wrong here?
We are trying to initialize disk with the properties of some existing disk on Windows server 2008/2012 through a C++ program.
We are using DeviceIoControl() method and IOCTL_DISK_CREATE_DISK, IOCTL_DISK_SET_DRIVE_LAYOUT_EX, IOCTL_DISK_SET_PARTITION_INFO_EX codes from Disk management control codes to make the disk available for use.
Got the following code snippet by searching a bit
//To open the drive
hDevice = CreateFile( TEXT("\\\\.\\PhysicalDrive7"),
GENERIC_READ | GENERIC_WRITE, // no access to the drive
FILE_SHARE_READ | FILE_SHARE_WRITE, // share mode
NULL, // default security attributes
OPEN_EXISTING, // disposition
0, // file attributes
NULL); // do not copy file attributes
CREATE_DISK dsk;
dsk.PartitionStyle = PARTITION_STYLE_MBR; //It can also be PARTITION_STYLE_GPT
dsk.Mbr.Signature = 1;
// Initialize disk
bResult = DeviceIoControl( hDevice, // device to be queried
IOCTL_DISK_CREATE_DISK, // operation to perform
&dsk, sizeof(dsk),
NULL, 0, // no output buffer
&junk, // # bytes returned
NULL
);
LARGE_INTEGER lgPartitionSize;
lgPartitionSize.QuadPart = (1024 * 1024 * 1024);
DWORD dwDriverLayoutInfoExLen = sizeof (DRIVE_LAYOUT_INFORMATION_EX) + 3 * sizeof(PARTITION_INFORMATION_EX);
DRIVE_LAYOUT_INFORMATION_EX *pdg = (DRIVE_LAYOUT_INFORMATION_EX *)new BYTE[dwDriverLayoutInfoExLen];
SecureZeroMemory(pdg, dwDriverLayoutInfoExLen);
pdg->PartitionStyle = PARTITION_STYLE_MBR;
pdg->PartitionCount = 1;
pdg->Mbr.Signature = 1;
pdg->PartitionEntry[0].PartitionStyle = PARTITION_STYLE_MBR;
pdg->PartitionEntry[0].StartingOffset.QuadPart = 1048576;
pdg->PartitionEntry[0].PartitionLength.QuadPart = lgPartitionSize.QuadPart * 200;
pdg->PartitionEntry[0].PartitionNumber = 1;
pdg->PartitionEntry[0].RewritePartition = TRUE;
pdg->PartitionEntry[0].Mbr.PartitionType = PARTITION_NTFT; // PARTITION_IFS (NTFS partition or logical drive)
pdg->PartitionEntry[0].Mbr.BootIndicator = TRUE;
pdg->PartitionEntry[0].Mbr.RecognizedPartition = 1;
pdg->PartitionEntry[0].Mbr.HiddenSectors = 32256 / 512;
// Partition a disk
bResult = DeviceIoControl( hDevice, // device to be queried
IOCTL_DISK_SET_DRIVE_LAYOUT_EX, // operation to perform
pdg, sizeof DRIVE_LAYOUT_INFORMATION_EX, //output buffer
NULL, 0, // no output buffer
&junk, // # bytes returned
NULL
);
bResult = DeviceIoControl( hDevice,
IOCTL_DISK_UPDATE_PROPERTIES,
NULL, 0, NULL, 0, &junk, NULL);
PARTITION_INFORMATION_EX dskinfo;
PARTITION_INFORMATION_MBR mbrinfo;
mbrinfo.PartitionType = PARTITION_NTFT;
mbrinfo.HiddenSectors = (32256 / 512);
mbrinfo.BootIndicator = 1;
mbrinfo.RecognizedPartition = 1;
dskinfo.PartitionStyle = PARTITION_STYLE_MBR;
dskinfo.StartingOffset.QuadPart = 1048576;//0;
dskinfo.PartitionLength.QuadPart = lgPartitionSize.QuadPart * 200;
dskinfo.PartitionNumber = 1;
dskinfo.RewritePartition = TRUE;
dskinfo.Mbr = mbrinfo;
bResult = DeviceIoControl( hDevice, // device to be queried
IOCTL_DISK_SET_PARTITION_INFO_EX, // operation to perform
&dskinfo, sizeof(dskinfo), // output buffer
NULL, 0, // no output buffer
&junk, // # bytes returned
NULL
);
All the calls to DeviceIoControl() are getting succeeded except the last one with IOCTL_DISK_SET_PARTITION_INFO_EX code with error 1 (i.e Incorrect function). What could be the reason for this?
If we comment out the last call, the disk is being initialized as raw disk, But this won't meet our requirements.
The above sample is for MBR partition style only. We could not find any sample for GPT,... styles. Please give a link if someone is aware of one.
You're using the wrong structure type with IOCTL_DISK_SET_PARTITION_INFO_EX. It takes a SET_PARTITION_INFORMATION_EX structure, not a PARTITION_INFORMATION_EX structure.
You probably don't need to use IOCTL_DISK_SET_PARTITION_INFO_EX, since it just sets the partition type, which should have already been set with IOCTL_DISK_SET_DRIVE_LAYOUT_EX. Unfortunately you've used it to set the wrong partition type. NTFS partitions have the partition type PARTITION_IFS.
Multiplying lgPartitionSize by 200 is almost certainly wrong. If lgPartitionSize is supposed the size in sectors then you need to multiply this by the sector size of the disk. The sector size of hard drives used to be always 512 bytes (0x200 bytes), but modern drives use a 4096 byte sector size.
Correctly creating partition tables is not easy, and mindlessly copying other people's code like you've done isn't going to work. Even after you fix the problems I've mentioned above you're still likely to encounter other issues. You really need to understand the all the restrictions placed on how partitions should be laid out.
You might want to consider using the diskpart command to programically initialize disks instead of C++ code.
I use function CreateFileMapping MapViewOfFile UnmapViewOfFile to create a shared block on the disk. Can I placement new a class object on this block?
I use VS2003 IDE.
try
{
Sphere *pData = m_pBVH->GetFirstHalfData();
Sphere *p = new(pData)Sphere(center, radius, index);
}
catch (std::exception& e)
{
// Can't catch
}
Placement new error:
#ifdef _DEBUG
#define new DEBUG_NEW
#endif
At the Debug Mode, placement new operator may cause a syntax error. It's likely because of the code above, anotate it and try again!
At Release mode, it's OK, but throws exception which is not std::bad_alloc, can't figure out which it is. Maybe there is something with the way shared memory is built. My shared memory creation code is as bellow:
CBVHVertexBuffer::CBVHVertexBuffer(const CString &file, unsigned int size)
{
this->file = file;
this->size = size;
fileHandle = INVALID_HANDLE_VALUE;
mapHandle = INVALID_HANDLE_VALUE;
mapLength = 0;
start = NULL;
end = NULL;
head = NULL;
tail = NULL;
fileHandle = ::CreateFile(
file,
GENERIC_READ | GENERIC_WRITE,
0,
NULL,
CREATE_ALWAYS,
FILE_FLAG_SEQUENTIAL_SCAN | FILE_FLAG_DELETE_ON_CLOSE,
NULL);
if (fileHandle != INVALID_HANDLE_VALUE)
{
//::SetFilePointer(fileHandle, size * pickingVertexSize, NULL, FILE_BEGIN);
//::SetEndOfFile(fileHandle);
//mapLength = ::GetFileSize(fileHandle, NULL);
//mapLength = size * pickingVertexSize;
mapHandle = ::CreateFileMapping(fileHandle, NULL, PAGE_READWRITE, 0, size * pickingVertexSize, NULL);
mapLength = ::GetFileSize(fileHandle, NULL);
start = (Sphere*)::MapViewOfFile(mapHandle, FILE_MAP_WRITE, 0, 0, 0);
}
end = start + size;
Clear();
}
I would advice to only use Plain Old Data (POD) in your mapped file, because pointers can become invalid when mapping and unmapping files
In principle, you can use placement new operator (void* operator new (std::size_t size, void* ptr) throw();) to create an object in an already allocated memory buffer.
However, creating objects in file mapping regions should be applied for POD objects. Any raw pointer member (and virtual memory table if any) will not be recommended to be stored in mapped file, because they will be invalidated as soon as the mapping is done in a different address space. You should also take care to manage the memory allocation in this case.
I recommend to take a look at boost interprocess library (they have pretty good support for memory mapped files, including management of the memory buffers - allocation algorithms, object tracking and a pointer-like class for linking objects address-space independent (offset_ptr)).
I'm trying to use CreateFileMapping and OpenFileMapping to share memory between processes. This isn't working as I want it to - OpenFileMapping returns null and GetLastError is 5 - access denied. Any ideas what I am doing wrong? Name is something like MemoryTest.
Edit:
using CreateFileMapping both times I can read the data written in the other process. The reason this is a problem is that I get Error 183 - memory area already exists. However, it still returns a handle to the existing memory.
var map_handle = CreateFileMapping(INVALID_HANDLE_VALUE, NULL, PAGE_READWRITE, 0, sizeof(int), name.c_str());
....
var handle = MapViewOfFile(map_handle, FILE_MAP_ALL_ACCESS , 0, 0, 0)
*handle = 10;
UnMapViewOfFile(map_handle);
getchar();
Other process:
var map_handle = OpenFileMapping(PAGE_READWRITE, false, name.c_str())
....
var handle = MapViewOfFile(map_handle, FILE_MAP_ALL_ACCESS , 0, 0, 0) //returns null
var out = *handle;
getchar();
This works for the second process though:
var map_handle = CreateFileMapping(INVALID_HANDLE_VALUE, NULL, PAGE_READWRITE, 0, sizeof(int), name.c_str());
....
var handle = MapViewOfFile(map_handle, FILE_MAP_ALL_ACCESS , 0, 0, 0) //returns null
var out = *handle;
getchar();
Simple things to be aware of from the very start:
Error code 5: ERROR_ACCESS_DENIED "Access is denied."
Error code 183: ERROR_ALREADY_EXISTS "Cannot create a file when that file already exists."
ERROR_ALREADY_EXISTS is a documented behavior and is an indication of scenario that you do receive handle, but it is a handle to already existing object, not created.
The problem with not working OpenFileMapping is around its first argument: the API function expects values/flags from another enumeration, it takes FILE_MAP_* values and not PAGE_*. Incorrect argument results in failure to open you the mapping you want.
In case someone else needed, in my case the error has nothing to do with the access to the file, it's with the size provided to the CreateFileMapping, after spending hours with a similar error I'd to use a working sample posted somewhere else and line by line compare what was the difference.
If you don't know the size of the file when executing the CreateFileMapping you need to use 0, this will tell the API to use the file size of the mapped file. Most of the answers in SO around this are wrong and people is not bothering testing what is the problem about, I wasted hours reading other posts with similar suggestions.
To solve the problem the code should look like this:
var map_handle = CreateFileMapping(INVALID_HANDLE_VALUE, NULL, PAGE_READWRITE, 0, 0, name.c_str());
Hope this saves hours to other fellow developers.
I am interested in the prospect of using memory mapped IO, preferably
exploiting the facilities in boost::interprocess for cross-platform
support, to map non-contiguous system-page-size blocks in a file into
a contiguous address space in memory.
A simplified concrete scenario:
I've a number of 'plain-old-data' structures, each of a fixed length
(less than the system page size.) These structures are concatenated
into a (very long) stream with the type & location of structures
determined by the values of those structures that proceed them in the
stream. I'm aiming to minimize latency and maximize throughput in a
demanding concurrent environment.
I can read this data very effectively by memory-mapping it in blocks
of at least twice the system-page-size... and establishing a new
mapping immediately having read a structure extending beyond the
penultimate system-page-boundary. This allows the code that interacts
with the plain-old-data structures to be blissfully unaware that these
structures are memory mapped... and, for example, could compare two
different structures using memcmp() directly without having to care
about page boundaries.
Where things get interesting is with respect to updating these data
streams... while they're being (concurrently) read. The strategy I'd
like to use is inspired by 'Copy On Write' on a system-page-size
granularity... essentially writing 'overlay-pages' - allowing one
process to read the old data while another reads the updated data.
While managing which overlay pages to use, and when, isn't necessarily
trivial... that's not my main concern. My main concern is that I may
have a structure spanning pages 4 and 5, then update a
structure wholly contained in page 5... writing the new page in
location 6... leaving page 5 to be 'garbage collected' when it is
determined to be no-longer reachable. This means that, if I map page
4 into location M, I need to map page 6 into memory location
M+page_size... in order to be able to reliably process structures that
cross page boundaries using existing (non-memory-mapping-aware) functions.
I'm trying to establish the best strategy, and I'm hampered by
documentation I feel is incomplete. Essentially, I need to decouple
allocation of address space from memory mapping into that address
space. With mmap(), I'm aware that I can use MAP_FIXED - if I wish to
explicitly control the mapping location... but I'm unclear how I
should reserve address space in order to do this safely. Can I map
/dev/zero for two pages without MAP_FIXED, then use MAP_FIXED twice to
map two pages into that allocated space at explicit VM addresses? If
so, should I call munmap() three times too? Will it leak resources
and/or have any other untoward overhead? To make the issue even more
complex, I'd like comparable behaviour on Windows... is there any way
to do this? Are there neat solutions if I were to compromise my
cross-platform ambitions?
--
Thanks for your answer, Mahmoud... I've read, and think I've understood that code... I've compiled it under Linux and it behaves as you suggest.
My main concerns are with line 62 - using MAP_FIXED. It makes some assumptions about mmap, which I've been unable to confirm when I read the documentation I can find. You're mapping the 'update' page into the same address space as mmap() returned initially - I assume that this is 'correct' - i.e. not something that just happens to work on Linux? I'd also need to assume that it works cross-platform for file-mappings as well as anonymous mappings.
The sample definitely moves me forwards... documenting that what I ultimately need is probably achievable with mmap() on Linux - at least. What I'd really like is a pointer to documentation that shows that the MAP_FIXED line will work as the sample demonstrates... and, idealy, a transformation from the Linux/Unix specific mmap() to a platform independent (Boost::interprocess) approach.
Your question is a little confusing. From what I understood, this code will do what you need:
#define PAGESIZE 4096
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/mman.h>
#include <errno.h>
#include <sys/types.h>
#include <fcntl.h>
#include <unistd.h>
#include <assert.h>
struct StoredObject
{
int IntVal;
char StrVal[25];
};
int main(int argc, char **argv)
{
int fd = open("mmapfile", O_RDWR | O_CREAT | O_TRUNC, (mode_t) 0600);
//Set the file to the size of our data (2 pages)
lseek(fd, PAGESIZE*2 - 1, SEEK_SET);
write(fd, "", 1); //The final byte
unsigned char *mapPtr = (unsigned char *) mmap(0, PAGESIZE * 2, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
struct StoredObject controlObject;
controlObject.IntVal = 12;
strcpy(controlObject.StrVal, "Mary had a little lamb.\n");
struct StoredObject *mary1;
mary1 = (struct StoredObject *)(mapPtr + PAGESIZE - 4); //Will fall on the boundary between first and second page
memcpy(mary1, &controlObject, sizeof(StoredObject));
printf("%d, %s", mary1->IntVal, mary1->StrVal);
//Should print "12, Mary had a little lamb.\n"
struct StoredObject *john1;
john1 = mary1 + 1; //Comes immediately after mary1 in memory; will start and end in the second page
memcpy(john1, &controlObject, sizeof(StoredObject));
john1->IntVal = 42;
strcpy(john1->StrVal, "John had a little lamb.\n");
printf("%d, %s", john1->IntVal, john1->StrVal);
//Should print "12, Mary had a little lamb.\n"
//Make sure the data's on the disk, as this is the initial, "read-only" data
msync(mapPtr, PAGESIZE * 2, MS_SYNC);
//This is the inital data set, now in memory, loaded across two pages
//At this point, someone could be reading from there. We don't know or care.
//We want to modify john1, but don't want to write over the existing data
//Easy as pie.
//This is the shadow map. COW-like optimization will take place:
//we'll map the entire address space from the shared source, then overlap with a new map to modify
//This is mapped anywhere, letting the system decide what address we'll be using for the new data pointer
unsigned char *mapPtr2 = (unsigned char *) mmap(0, PAGESIZE * 2, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
//Map the second page on top of the first mapping; this is the one that we're modifying. It is *not* backed by disk
unsigned char *temp = (unsigned char *) mmap(mapPtr2 + PAGESIZE, PAGESIZE, PROT_READ | PROT_WRITE, MAP_SHARED | MAP_FIXED | MAP_ANON, 0, 0);
if (temp == MAP_FAILED)
{
printf("Fixed map failed. %s", strerror(errno));
}
assert(temp == mapPtr2 + PAGESIZE);
//Make a copy of the old data that will later be changed
memcpy(mapPtr2 + PAGESIZE, mapPtr + PAGESIZE, PAGESIZE);
//The two address spaces should still be identical until this point
assert(memcmp(mapPtr, mapPtr2, PAGESIZE * 2) == 0);
//We can now make our changes to the second page as needed
struct StoredObject *mary2 = (struct StoredObject *)(((unsigned char *)mary1 - mapPtr) + mapPtr2);
struct StoredObject *john2 = (struct StoredObject *)(((unsigned char *)john1 - mapPtr) + mapPtr2);
john2->IntVal = 52;
strcpy(john2->StrVal, "Mike had a little lamb.\n");
//Test that everything worked OK
assert(memcmp(mary1, mary2, sizeof(struct StoredObject)) == 0);
printf("%d, %s", john2->IntVal, john2->StrVal);
//Should print "52, Mike had a little lamb.\n"
//Now assume our garbage collection routine has detected that no one is using the original copy of the data
munmap(mapPtr, PAGESIZE * 2);
mapPtr = mapPtr2;
//Now we're done with all our work and want to completely clean up
munmap(mapPtr2, PAGESIZE * 2);
close(fd);
return 0;
}
My modified answer should address your safety concerns.
Only use MAP_FIXED on the second mmap call (like I have above). The cool thing about MAP_FIXED is that it lets you overwrite an existing mmap address section. It'll unload the range you're overlapping and replace it with your new mapped content:
MAP_FIXED
[...] If the memory
region specified by addr and len overlaps pages of any existing
mapping(s), then the overlapped part of the existing mapping(s) will be
discarded. [...]
This way, you let the OS take care of finding a contiguous memory block of hundreds of megs for you (never call MAP_FIXED on address you don't know for sure isn't available). Then you call MAP_FIXED on a subsection of that now-mapped huge space with the data that you will be modifying. Tada.
On Windows, something like this should work (I'm on a Mac at the moment, so untested):
int main(int argc, char **argv)
{
HANDLE hFile = CreateFile(L"mmapfile", GENERIC_READ | GENERIC_WRITE, FILE_SHARE_READ, NULL, CREATE_ALWAYS, FILE_ATTRIBUTE_NORMAL, NULL);
//Set the file to the size of our data (2 pages)
SetFilePointer(hFile, PAGESIZE*2 - 1, 0, FILE_BEGIN);
DWORD bytesWritten = -1;
WriteFile(hFile, "", 1, &bytesWritten, NULL);
HANDLE hMap = CreateFileMapping(hFile, NULL, PAGE_READWRITE, 0, PAGESIZE * 2, NULL);
unsigned char *mapPtr = (unsigned char *) MapViewOfFile(hMap, FILE_MAP_READ | FILE_MAP_WRITE, 0, 0, PAGESIZE * 2);
struct StoredObject controlObject;
controlObject.IntVal = 12;
strcpy(controlObject.StrVal, "Mary had a little lamb.\n");
struct StoredObject *mary1;
mary1 = (struct StoredObject *)(mapPtr + PAGESIZE - 4); //Will fall on the boundary between first and second page
memcpy(mary1, &controlObject, sizeof(StoredObject));
printf("%d, %s", mary1->IntVal, mary1->StrVal);
//Should print "12, Mary had a little lamb.\n"
struct StoredObject *john1;
john1 = mary1 + 1; //Comes immediately after mary1 in memory; will start and end in the second page
memcpy(john1, &controlObject, sizeof(StoredObject));
john1->IntVal = 42;
strcpy(john1->StrVal, "John had a little lamb.\n");
printf("%d, %s", john1->IntVal, john1->StrVal);
//Should print "12, Mary had a little lamb.\n"
//Make sure the data's on the disk, as this is the initial, "read-only" data
//msync(mapPtr, PAGESIZE * 2, MS_SYNC);
//This is the inital data set, now in memory, loaded across two pages
//At this point, someone could be reading from there. We don't know or care.
//We want to modify john1, but don't want to write over the existing data
//Easy as pie.
//This is the shadow map. COW-like optimization will take place:
//we'll map the entire address space from the shared source, then overlap with a new map to modify
//This is mapped anywhere, letting the system decide what address we'll be using for the new data pointer
unsigned char *reservedMem = (unsigned char *) VirtualAlloc(NULL, PAGESIZE * 2, MEM_RESERVE, PAGE_READWRITE);
HANDLE hMap2 = CreateFileMapping(hFile, NULL, PAGE_READWRITE, 0, PAGESIZE, NULL);
unsigned char *mapPtr2 = (unsigned char *) MapViewOfFileEx(hMap2, FILE_MAP_READ | FILE_MAP_WRITE, 0, 0, PAGESIZE, reservedMem);
//Map the second page on top of the first mapping; this is the one that we're modifying. It is *not* backed by disk
unsigned char *temp = (unsigned char *) MapViewOfFileEx(hMap2, FILE_MAP_READ | FILE_MAP_WRITE, 0, 0, PAGESIZE, reservedMem + PAGESIZE);
if (temp == NULL)
{
printf("Fixed map failed. 0x%x\n", GetLastError());
return -1;
}
assert(temp == mapPtr2 + PAGESIZE);
//Make a copy of the old data that will later be changed
memcpy(mapPtr2 + PAGESIZE, mapPtr + PAGESIZE, PAGESIZE);
//The two address spaces should still be identical until this point
assert(memcmp(mapPtr, mapPtr2, PAGESIZE * 2) == 0);
//We can now make our changes to the second page as needed
struct StoredObject *mary2 = (struct StoredObject *)(((unsigned char *)mary1 - mapPtr) + mapPtr2);
struct StoredObject *john2 = (struct StoredObject *)(((unsigned char *)john1 - mapPtr) + mapPtr2);
john2->IntVal = 52;
strcpy(john2->StrVal, "Mike had a little lamb.\n");
//Test that everything worked OK
assert(memcmp(mary1, mary2, sizeof(struct StoredObject)) == 0);
printf("%d, %s", john2->IntVal, john2->StrVal);
//Should print "52, Mike had a little lamb.\n"
//Now assume our garbage collection routine has detected that no one is using the original copy of the data
//munmap(mapPtr, PAGESIZE * 2);
mapPtr = mapPtr2;
//Now we're done with all our work and want to completely clean up
//munmap(mapPtr2, PAGESIZE * 2);
//close(fd);
return 0;
}
but I'm unclear how I should reserve address space in order to do this safely
That's going to vary by OS, but a little digging on msdn for mmap (I started with "xp mmap" on the msdn search) shows Microsoft have their usual VerboseAndHelpfullyCapitalizedNames for (the many) functions that implement pieces of mmap. Both the file- and anonymous- mappers can handle fixed-address requests just the same as any POSIX-2001 system can, i.e. if something else in your address space is talking to the kernel, you get to sort it out. No way I'm going to touch "safely", there's no such thing as "safely" with code you're wanting to port to unspecified platforms. You're going to have to build your own pool of pre-mapped anonymous memory that you can unmap and parcel out later under your own control.
I tested the windows code from #Mahmoud, well actually I tested the following similar code, and it doesn't work (the Linux code works.) If you uncomment VirtualFree, it will work. As mentioned in my comment above, on windows you can reserve the address space with VirtualAlloc, but you can't use MapViewOfFileEx with an already mapped address, so you need to VirtualFree it first. Then there's a race condition where another thread can grab the memory address before you do, so you have to do everything in a loop, e.g. try up to 1000 times and then give up.
package main
import (
"fmt"
"os"
"syscall"
)
func main() {
const size = 1024 * 1024
file, err := os.Create("foo.dat")
if err != nil {
panic(err)
}
if err := file.Truncate(size); err != nil {
panic(err)
}
const MEM_COMMIT = 0x1000
addr, err := virtualAlloc(0, size, MEM_COMMIT, protReadWrite)
if err != nil {
panic(err)
}
fd, err := syscall.CreateFileMapping(
syscall.Handle(file.Fd()),
nil,
uint32(protReadWrite),
0,
uint32(size),
nil,
)
//if err := virtualFree(addr); err != nil {
// panic(err)
//}
base, err := mapViewOfFileEx(fd, syscall.FILE_MAP_READ|syscall.FILE_MAP_WRITE, 0, 0, size, addr)
if base == 0 {
panic("mapViewOfFileEx returned 0")
}
if err != nil {
panic(err)
}
fmt.Println("success!")
}
type memProtect uint32
const (
protReadOnly memProtect = 0x02
protReadWrite memProtect = 0x04
protExecute memProtect = 0x20
protAll memProtect = 0x40
)
var (
modkernel32 = syscall.MustLoadDLL("kernel32.dll")
procMapViewOfFileEx = modkernel32.MustFindProc("MapViewOfFileEx")
procVirtualAlloc = modkernel32.MustFindProc("VirtualAlloc")
procVirtualFree = modkernel32.MustFindProc("VirtualFree")
procVirtualProtect = modkernel32.MustFindProc("VirtualProtect")
)
func mapViewOfFileEx(handle syscall.Handle, prot memProtect, offsetHigh uint32, offsetLow uint32, length uintptr, target uintptr) (addr uintptr, err error) {
r0, _, e1 := syscall.Syscall6(procMapViewOfFileEx.Addr(), 6, uintptr(handle), uintptr(prot), uintptr(offsetHigh), uintptr(offsetLow), length, target)
addr = uintptr(r0)
if addr == 0 {
if e1 != 0 {
err = error(e1)
} else {
err = syscall.EINVAL
}
}
return addr, nil
}
func virtualAlloc(addr, size uintptr, allocType uint32, prot memProtect) (mem uintptr, err error) {
r0, _, e1 := syscall.Syscall6(procVirtualAlloc.Addr(), 4, addr, size, uintptr(allocType), uintptr(prot), 0, 0)
mem = uintptr(r0)
if e1 != 0 {
return 0, error(e1)
}
return mem, nil
}
func virtualFree(addr uintptr) error {
const MEM_RELEASE = 0x8000
_, _, e1 := syscall.Syscall(procVirtualFree.Addr(), 3, addr, 0, MEM_RELEASE)
if e1 != 0 {
return error(e1)
}
return nil
}