Is it possible (on iPhone/iPod Touch) for a file written like this:
if (FILE* file = fopen(filename, "wb")) {
fwrite(buf, buf_size, 1, file);
fclose(file);
}
to get corrupted, e.g. when app is forced to terminate?
From what I know fwrite should be an atomic operation, so when I write whole file with one instruction no corruption should occure. I could not find any information on the net that would say otherwise.
Since when fwrite is atomic? I can't find any references. Anyway, even if fwrite can be atomic, the time between fopen and fwrite is not, so if your app is forced to terminate between those times, you'll get an empty file.
As you're writing for iPhoneOS, you can use -[NSData writeToFile:atomically:] to ensure the whole open-write-close procedure is atomic (it works by writing to a temporary file, then replace the original one.)
You could make things easier for yourself and write the data using the NSData class that has a writeToFile:atomically: method waiting for you. Wrapping the raw buffer with NSData is not hard, there are the dataWithBytes:length or dataWithBytesNoCopy:length:freeWhenDone: initializers.
Data written with fwrite is buffered. So a sudden termination might not flush the buffers. fclose will flush the buffer but this does not implicate that the bytes are also written to the disk (due to OS level caches) AFAIK.
Related
Does read() system call causes the kernel device driver code to be executed?
I mean, when I want to read files from disk or from any driver, I use read() system call. Is that system call must use driver code to complete this task?
Thanks.
Yes, the read() uses standard file descriptors to read files from fixed offsets in files and of fixed bytes. The standard file descriptors are not files on the disc, but rather associated with a different device, the terminal device. Hence the drivers must be in use for the files to be accessed.
See man read for its usage and C prototype.
For more info about file handling use this link, particularly the "Reviewing Open Files" section for your particular query.
So, I am currently learning about the INode file system and am asked to write a simple file system using Inodes.
So far, I understand that there is an INode table that has a mapping from INode-> Data blocks through direct/indirect pointers.
Let's assume data gets written into a file, the data is stored into two blocks. Let's say each block is 512bytes, and the file takes one full block, and only 200 bytes of the second block. What happens with the rest of the space in that data block? Is it reserved for that file only or do other files use this block?
Depending on the file system, usually and most likely this area is now lost. I think the Reiser File System actually reclaimed this area, but I could be wrong.
Creating your own File System can be a challenging experience, but also an enjoyable experience. I have created a few myself and worked on another. If you are creating your own file system, you can have it do whatever you wish.
Look at the bottom of this page for a few that I am working on/with. The LeanFS in particular, uses Inodes as well. The SFS is a very simple file system. Each is well documented so that you can research and decide what you would like to do.
Does
pcap_t *pcap_open_offline(const char *fname, char *errbuf)
from libpcap read the whole pcap file into memory? If not so, I have to use tcpslice or similar tools to split pcap file up?
Thanks.
A strange way of wording your question, but I'll try and answer what I can.
pcap_open_offline() takes a .dump file (or similarly named output from tcpdump, tcpslice, or libpcap's pcap_dump_open() + pcap_dump() functions) as an input.
This file is exactly the same in format and function as a live trace of a network device, IE, you can use this pcap_t object in pcap_next, pcap_loop, etc.
Altering a dump file in any way (IE, stripping information or parsing out only what you want with tcpslice or wireshark) will render it unreadable by pcap_open_offline(), as it will not be formatted in the manner of a live packet trace.
However, it does not load the entire file at any one time into memory. It streams the file, as you would stream packets from a live trace.
To summarize: pcap_open_live() opens an unaltered tcpdump/tcpslice dump and reads it like a live stream. It does not load the entire file into its memory, as dumps can get quite large! Instead it just goes through the file only loading one packet's worth of the file at a time.
I have implemented a log file that will be storing the cpu and memory state of a process after every minute.I have limited the maximum size of the file to 3MB (thats enough for my purpose).
The script will be called by a cron job after every minute and the script will log the details for that minute and will rename the file as "Log_.log".
When the size reaches "3MB - 100 bytes" I reset the file pointer to point to the begining and will overwrite the first entry in the log file and will now rename the file as "Log_<0+some offset>.log".
As I am renaming the file after every minute to update the file pointer position, is it a good/efficient way ?
I do not want to maintain more than one log file for this purpose.
Another option for me is to maintain the file pointer position in a file ,but ....another file !! not interested in maintaining one if this option is good :)
Thanks in Advance.
Are you an engineer? This is a nice example of some simple task, solved by a perfectly working but overly complex solution.
Unless the content you put in takes exactly as many bytes as the content you take out, writing "in" a file will actually cause the whole following part after your writing position to be rewritten to disk. Append is much cheaper.
Renaming the file to store the pointer works - but it's not very elegant, and makes stuff more complex (for one, your process needs write rights to the directory in which the file resides - else just write access to two files is sufficient)
Unless disk space is an issue (and really, it rarely is), your approach is less efficient than say, append everything to a file, and rotate the file when it reaches its maximum size. This way you always have the last 3MB of logs available, and maximum 3MB more in your current file. It will make parsing the file a lot easier too, instead of recalculating the entire pointer position thing.
Update to answer your comment:
Renaming a file every minute (or even every second) shouldn't slow down your system significantly, don't worry about that.
Our concerns are mainly with "why you think you need to rename the file". It's not better technically, it's not better from a logical point of view, it makes a lot of other (future) tasks harder. You could store the file pointer in a seperate file, or at the end of your file, and there are better^H^H^H^H^H^H simpler solutions that don't require the file pointer at all.
I'm confused why you would rename your file. What does this accomplish?
Are the log entries fixed size? Or variable size?
If the entries are fixed size, then there is no trouble in re-writing the existing file from the start: you won't ever have incomplete entries in your file, and if you are writing a counter or timestamps to the file, it should be clear where the 'cursor' is located.
If the entries are variable size, then you should probably not begin re-writing the file from the beginning without somehow making it clear where the 'cursor' is located in the file, and write code that is resilient to reading truncated log entries.
Can you re-use existing tools such as RRDtool?
I have a situation where I can save a post-processing pass through the audio by taking some manipulated buffer from the end of the track and writing them to the beginning of my output file.
I originally thought I could do this by resetting the write pointer using ExtAudioFileSeek, and was about to implement it when I saw this line in the docs
Ensure that the file you are seeking in is open for reading only. This function’s behavior with files open for writing is undefined.
Now I know I could close the file for writing then reopen it, but the process is a little more complicated than that. Part of the manipulation I am doing is reading from buffers that are in the file I am writing to. The overall process looks like this:
Read buffers from the end of the read file
Read buffers from the beginning of the write file
Process the buffers
Write the buffers back to the beginning of the write file, overwriting the buffers I read in step 2
Logically, this can be done in 1 pass no problem. Programmatically, how can I achieve the same thing without corrupting my data, becoming less-efficient (opposite of my goal) or potentially imploding the universe?
Yes, using a single audio file for both reading and writing may, as you put it, implode the universe, or at least lead to other nastiness. I think that the key to solving this problem is in step 4, where you should write the output to a new file instead of trying to "recycle" the initial write file. After your processing is complete, you can simply scrap the intermediate write file.
Or have I misunderstood the problem?
Oh, and also, you should use ExtAudioFileWriteAsync instead of ExtAudioFileWrite for your writes if you are doing this in realtime. Otherwise the I/O load will cause audio dropouts.