Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 12 years ago.
Improve this question
This may be slightly OT, but I was wondering why having a process which heavily uses IO (say cp big file from one location to the other on the same disk) slows everything down, even processes which are mostly CPU bound. I noticed that on both OS I heavily use (mac os x and linux).
In particular, I wonder why multi-core does not really help here: is it a hardware limitation for commodity hardware (disk controller, etc...), an os limitation, or is there something inherently hard into allocating the right resources (scheduling) ?
It could be a limitation of the current scheduler. Google "Galbraith's sched:autogroup patch" or "linux miracle patch" (yes really!). There's apparently a 200-line patch in the process of being refined and merged which adds group scheduling, about which Linus says:
I'm also very happy with just what it does to interactive performance.
Admittedly, my "testcase" is really trivial (reading email in a
web-browser, scrolling around a bit, while doing a "make -j64" on the
kernel at the same time), but it's a test-case that is very relevant
for me. And it is a huge improvement.
Before-and-after videos here.
Because, copying a large file (bigger than the available buffer cache) usually involves bringing it through the buffer cache, which generally causes less recently-used pages to be thrown out, which must then be brought back in.
Other processes which are doing tiny small amounts of occasional IO (say just stat'ing a directory) then get their caches all blown away and must do physical reads to bring those pages back in.
Hopefully this can get fixed by a copy-command which can detect this kind of thing and advise the kernel accordingly (e.g. with posix_fadvise) so that a large one-off bulk transfer of a file which does not need to be subsequently read does not completely discard all clean pages from the buffer cache, which now normally mostly happens.
A high rate of IO operations usually means a high rate of interrupts that must be serviced by the CPU, which takes CPU time.
In the case of cp, it also uses a considerable amount of the available memory bandwidth, as each block of data is copied to and from userspace. This will also tend to eject data required by other processes from the CPUs caches and TLB, which will slow down other processes as they take cache misses.
Also, would you know a way to validate your hypothesis on linux, e.g. number of interrupts while doing IO intensive operations.
To do with interrupts, I'm guessing that caf's hypothesis is:
many interrupts per second;
interrupts are serviced by any/all CPUs;
therefore, interrupts flush the CPU caches.
The statistics you'd need to test that would be the number of interrupts per second per CPU.
I don't know whether it's possible to tie interrupts to a single CPU: see http://www.google.com/#q=cpu+affinity+interrupt for further details.
Here's something I don't understand (this is the first time I've looked at this question): perfmon on my laptop (running Windows Vista) is showing 2000 interrupts/second (1000 on each core) when it's almost idle (doing nothing but displaying perfmon). I can't imagine which device is generating 2000 interrupts/second, and I would have thought that's enough to blow away the CPU caches (my guess is that the CPU quantum for a busy thread is something like 50 msec). It's also showing an average of 350 DPCs/sec.
Do high end hardware suffer from similar issues ?
One type of hardware difference might be the disk hardware and disk device driver, generating more or fewer interrupts and/or other contentions.
Related
With a 1270v3 and a single thread app I'm at the end of performance but when I watch monitoring tools like atop I don't understand how this whole stuff works. I tried to find a nice article about this sort of topic but they either have been explained in a language I don't understand or are not about the stuff I would like to know. I hope it is alright to ask this kind of stuff here.
From my understanding a single-thread app does only use one thread for all/most of the work. So the performance is defined by the single-thread power of the CPU.
A moment before I wrote this question I played around with CPU-frequency and noticed that although there are only two instances of the app running the usage is shared across all cores.
So I assume that the thread jumps around between these cores.
So I set the CPU scaling to performance with cpufreq-set -g performance. The result was that all CPU cores/threads stayed at about 2GHz like it was before besides one that is permanently on 3.5GHz (100%). As I only changed the scaling for one core, why is the usage still shared across all cores? I mean the app is running at about 300%, why doesn't it stick to the CPU core with the 100%?
Furthermore as I noticed that only one of the CPU's got scaled up I looked into the help page and found -r which should scale all cores with the performance settings. Unfortunately nothing does change. (Is this a bug in Ubuntu 1404?) So I used -c with the number 8 (8 threads) -> didn't work. 4 -> works but only scales 2 cores out of 8. 7 -> scaled 4 cores. So I'm wondering, does this not support hyper-threading or is the whole program that buggy?
However as I understand it, the CPU's with the max frequency together with the thread jump around in the monitoring tools as they display the average the usage, which than looks like shared. Did I figure this right?
Would forcing one cpu to 3.5GHz and forcing the app to this core improve performance or is all the stuff I'm wondering about only about avg calculation between the data they show each second.
If so am I right that I should run best with cpufreq-set -c 7 -g performance if power consumption doesn't matter?
Thanks for reading so far, I hope you have a moment to help me understand the whole thing.
Atop example screenshots:
http://i.imgur.com/VFEBvLx.png
http://i.imgur.com/cBKOnJM.png
http://i.imgur.com/bgQfwfU.png
I believe a lot of your confusion has to do with the fuzzy mapping of the capabilities of cpufreq to the actual capabilities of the hardware.
Here’s a description of what is taking place on the HW and in the OS.
A processor is a collection of cores on the same silicon substrate. The cores are what we used to call CPUs with some enhancements. CPUs now have the capability of running multiple HW threads (hyperthreading), each hardware thread being equivalent to one of the old type CPUs. Putting this all together, the 1270v3 is a quad core (if I recall correctly), meaning it has 4 cores on the same silicon substrate. Each core can support two HW threads, each HW thread being equivalent to what the OS calls a CPU (and I’ll call a virtual CPU). So from the OS perspective, the 1270v3 has 8 (virtual) CPUs.
The OS doesn’t see packages, cores or HW threads. It sees CPUs, and to it there appear to be 8 of them.
To further complicate the issue, modern processors have various HW supporting power saving states called P-states, C-states and package C-states. Why do I mention these? It’s because they are intimately associated with the frequency of the processor. And cpufreq professes to provide the user with control over the processor’s frequency.
Now, I’m not familiar with cpufreq outside of reading the manpage and other material on the web. From my reading, it has a lot of idiosyncrasies, so I’ll talk about what it is doing from a broad perspective.
In a general sense, cpufreq has a lot of generic capability that may or may not be supported by the HW or the kernel. This is confusing because it looks like the functionality is there but then things don’t happen as you would expect. For example, cpufreq gives the impression that you can set each CPU’s frequency independently. In reality, on a hyperthreaded system, two “CPUs” are associated with each core and must have the same frequency.
A lot of the functionality that cpufreq is supposed to control is associated with the power efficiency characteristics of the processor, but again, its mapping to the processor’s actual hardware capabilities is incomplete and misleading. Though cpufreq seems to allow you to set max and min frequencies, the processor hardware doesn’t work this way. In modern Intel processors, such as the 1270v3, there are something called P-states. These P-states are frequency-voltage pairs that slow down a processor’s frequency (and drop its voltage) to reduce the processor’s power consumption at the cost of the application taking longer to run. These frequency-voltage pairings aren’t arbitrary though cpufreq gives the impression that they are.
What does this all mean? In addition to the thread migration issues that the commenter mentioned, cpufreq isn’t going to behave the way you expect because it appears to have capability that it actually doesn’t, and the capability that it does actually have maps only roughly to the actual capabilities of the HW and OS.
I embedded some further comments in your text.
With a 1270v3 and a single thread app I'm at the end of performance but when I watch monitoring tools like atop I don't understand how this whole stuff works. I tried to find a nice article about this sort of topic but they either have been explained in a language I don't understand or are not about the stuff I would like to know. I hope it is alright to ask this kind of stuff here.
From my understanding a single-thread app does only use one thread for all/most of the work. [Yes, but this doesn’t mean that the thread is locked to a specific virtual CPU or core.] So the performance is defined by the single-thread power of the CPU. [It’s not that simple. The OS migrates threads around, it has its own maintenance processes, etc] A moment before I wrote this question I played around with CPU-frequency and noticed that although there are only two instances of the app running the usage is shared across all cores. So I assume that the thread jumps around between these cores. So I set the CPU scaling to performance with cpufreq-set -g performance. The result was that all CPU cores/threads stayed at about 2GHz like it was before besides one that is permanently on 3.5GHz (100%). As I only changed the scaling for one core, why is the usage still shared across all cores? I mean the app is running at about 300%, why doesn't it stick to the CPU core with the 100%? [Since I can’t see what you are observing, I don’t really understand what you are asking.]
Furthermore as I noticed that only one of the CPU's got scaled up I looked into the help page and found -r which should scale all cores with the performance settings. Unfortunately nothing does change. (Is this a bug in Ubuntu 1404?) So I used -c with the number 8 (8 threads) -> didn't work. 4 -> works but only scales 2 cores out of 8. 7 -> scaled 4 cores. [I haven’t used cpufreq so can’t directly speak to its behavior, but the manpage implies that “-c ” refers to a specific virtual CPU and not the number of virtual CPUs.] So I'm wondering, does this not support hyper-threading or is the whole program that buggy? [Again, I’m not sure from your explanation what you are doing, but the n->n/2 behavior makes sense to me. You can change the frequency of a core but since each core has two hyperthreads/virtual CPUs, two of those virtual CPUs must scale together.]
However as I understand it, the CPU's with the max frequency together with the thread jump around in the monitoring tools as they display the average the usage, which than looks like shared. Did I figure this right? [Again, I’m not sure what you are observing. Both physically and in atop, the CPU designation should not change, meaning CPU001 will always refer to the same virtual CPU. The core with the max frequency shouldn’t physically jump around, though the user thread may. Something to note is that monitoring tools can be pretty heavy users of the CPU. This heavy usage can make figuring out your processor usage difficult if it causes threads to jump around to different virtual CPUs.]
Would forcing one cpu to 3.5GHz and forcing the app to this core improve performance or is all the stuff I'm wondering about only about avg calculation between the data they show each second. [I found a pretty good explanation of atop with a lot of helpful screen shots: http://www.unixmen.com/linux-basics-monitor-system-resources-processes-using-atop/] If so am I right that I should run best with cpufreq-set -c 7 -g performance if power consumption doesn't matter? [It all depends upon what other processes are running on your system. If your system is mostly idle except for your processes, then forcing a core to a certain frequency won’t make a difference. [I’m suspicious of what a “governor” does. The language appears to refer to power-efficiency/performance (“balanced”, “powersave”, “performance”, etc) but the details don’t match the capability of today’s hardware.]
Thanks for reading so far, I hope you have a moment to help me
Here is my system based on Linux2.6.32.12:
1 It contains 20 processes which occupy a lot of usr cpu
2 It needs to write data on rate 100M/s to disk and those data would not be used recently.
What I expect:
It can run steadily and disk I/O would not affect my system.
My problem:
At the beginning, the system run as I thought. But as the time passed, Linux would cache a lot data for the disk I/O, that lead to physical memory reducing. At last, there will be not enough memory, then Linux will swap in/out my processes. It will cause I/O problem that a lot cpu time was used to I/O.
What I have try:
I try to solved the problem, by "fsync" everytime I write a large block.But the physical memory is still decreasing while cached increasing.
How to stop page cache here, it's useless for me
More infomation:
When Top show free 46963m, all is well including cpu %wa is low and vmstat shows no si or so.
When Top show free 273m, %wa is so high which affect my processes and vmstat shows a lot si and so.
I'm not sure that changing something will affect overall performance.
Maybe you might use posix_fadvise(2) and sync_file_range(2) in your program (and more rarely fsync(2) or fdatasync(2) or sync(2) or syncfs(2), ...). Also look at madvise(2), mlock(2) and munlock(2), and of course mmap(2) and munmap(2). Perhaps ionice(1) could help.
In the reader process, you might perhaps use readhahead(2) (perhaps in a separate thread).
Upgrading your kernel (to a 3.6 or better) could certainly help: Linux has improved significantly on these points since 2.6.32 which is really old.
To drop pagecache you can do the following:
"echo 1 > /proc/sys/vm/drop_caches"
drop_caches are usually 0. And, can be changed as per need. As you've identified yourself, that you need to free pagecache, so this is how to do it. You can also take a look at dirty_writeback_centisecs (and it's related tunables)(http://lxr.linux.no/linux+*/Documentation/sysctl/vm.txt#L129) to make quick writeback, but note it might have consequences, as it calls up kernel flasher thread to write out dirty pages. Also, note the uses of dirty_expire_centices, which defines how much time some data needs to be eligible for writeout.
I'm writing lots and lots of data that will not be read again for weeks - as my program runs the amount of free memory on the machine (displayed with 'free' or 'top') drops very quickly, the amount of memory my app uses does not increase - neither does the amount of memory used by other processes.
This leads me to believe the memory is being consumed by the filesystems cache - since I do not intend to read this data for a long time I'm hoping to bypass the systems buffers, such that my data is written directly to disk. I dont have dreams of improving perf or being a super ninja, my hope is to give a hint to the filesystem that I'm not going to be coming back for this memory any time soon, so dont spend time optimizing for those cases.
On Windows I've faced similar problems and fixed the problem using FILE_FLAG_NO_BUFFERING|FILE_FLAG_WRITE_THROUGH - the machines memory was not consumed by my app and the machine was more usable in general. I'm hoping to duplicate the improvements I've seen but on Linux. On Windows there is the restriction of writing in sector sized pieces, I'm happy with this restriction for the amount of gain I've measured.
is there a similar way to do this in Linux?
The closest equivalent to the Windows flags you mention I can think of is to open your file with the open(2) flags O_DIRECT | O_SYNC:
O_DIRECT (Since Linux 2.4.10)
Try to minimize cache effects of the I/O to and from this file. In
general this will degrade performance, but it is useful in special
situations, such as when applications do their own caching. File I/O
is done directly to/from user space buffers. The O_DIRECT flag on its
own makes at an effort to transfer data synchronously, but does not
give the guarantees of the O_SYNC that data and necessary metadata are
transferred. To guarantee synchronous I/O the O_SYNC must be used in
addition to O_DIRECT. See NOTES below for further discussion.
A semantically similar (but deprecated) interface for block devices is
described in raw(8).
Granted, trying to do research on this flag to confirm it's what you want I found this interesting piece telling you that unbuffered I/O is a bad idea, Linus describing it as "brain damaged". According to that you should be using madvise() instead to tell the kernel how to cache pages. YMMV.
You can use O_DIRECT, but in that case you need to do the block IO yourself; you must write in multiples of the FS block size and on block boundaries (it is possible that it is not mandatory but if you do not its performance will suck x1000 because every unaligned write will need a read first).
Another much less impacting way of stopping your blocks using up the OS cache without using O_DIRECT, is to use posix_fadvise(fd, offset,len, POSIX_FADV_DONTNEED). Under Linux 2.6 kernels which support it, this immediately discards (clean) blocks from the cache. Of course you need to use fdatasync() or such like first, otherwise the blocks may still be dirty and hence won't be cleared from the cache.
It is probably a bad idea of fdatasync() and posix_fadvise( ... POSIX_FADV_DONTNEED) after every write, but instead wait until you've done a reasonable amount (50M, 100M maybe).
So in short
after every (significant chunk) of writes,
Call fdatasync followed by posix_fadvise( ... POSIX_FADV_DONTNEED)
This will flush the data to disc and immediately remove them from the OS cache, leaving space for more important things.
Some users have found that things like fast-growing log files can easily blow "more useful" stuff out of the disc cache, which reduces cache hits a lot on a box which needs to have a lot of read cache, but also writes logs quickly. This is the main motivation for this feature.
However, like any optimisation
a) You're not going to need it so
b) Do not do it (yet)
as my program runs the amount of free memory on the machine drops very quickly
Why is this a problem? Free memory is memory that isn't serving any useful purpose. When it's used to cache data, at least there is a chance it will be useful.
If one of your programs requests more memory, file caches will be the first thing to go. Linux knows that it can re-read that data from disk whenever it wants, so it will just reap the memory and give it a new use.
It's true that Linux by default waits around 30 seconds (this is what the value used to be anyhow) before flushing writes to disk. You can speed this up with a call to fsync(). But once the data has been written to disk, there's practically zero cost to keeping a cache of the data in memory.
Seeing as you write to the file and don't read from it, Linux will probably guess that this data is the best to throw out, in preference to other cached data. So don't waste effort trying to optimise unless you've confirmed that it's a performance problem.
Azul Systems has an appliance that supports thousands of cache coherent CPUs. I would love insight into what changes would need to occur to an operating system in order to schedule thousands of simultaneously running threads.
Scheduling thousands of threads is not a big deal, but scheduling them on hundreds of CPUs is. What you need, first and foremost, is very fine-grained locking, or, better yet, lock-free data structures and algorithms. You just can't afford to let 200 CPUs waiting while one CPU executes a critical section.
You're asking for possible changes to the OS, so I presume there's a significant engineering team behind this effort.
There are also a few pieces of clarififying info that would help define the problem parameters:
How much IPC (inter process communication) do you need?
Do they really have to be threads, or can they be processes?
If they're processes, is it okay if the have to talk to each other through sockets, and not by using shared memory?
What is the memory architecture? Are you straight SMP with 1024 cores, or is there some other NUMA (Non-Uniform Memory Architecture) or MMP going on here? What are your page tables like?
Knowing only the very smallest of info about Azul systems, I would guess that you have very little IPC, and that a simple "run one kernel per core" model might actually work out just fine. If processes need to talk to each other, then they can create sockets and transfer data that way. Does your hardware support this model? (You would likely end up needing one IP address per core as well, and at 1024 IP addrs, this might be troublesome, although they could all be NAT'd, and maybe it's not such a big deal). If course, this model would lead to some inefficiencies, like extra page tables, and a fair bit of RAM overhead, and may even not be supported by your hardware system.
Even if "1 kernel per core" doesn't work, you could likely run 1024/8 kernels, and be just fine, letting each kernel control 8 physical CPUs.
That said, if you wanted to run 1 thread per core in a traditional SMP machine with 1024 cores (and only a few physical CPUs) then I would expect that the old fashioned O(1) scheduler is what you'd want. It's likely that your CPU[0] will end up nearly 100% in kernel and doing interrupt handling, but that's just fine for this use case, unless you need more than 1 core to handle your workload.
Making Linux scale has been a long and ongoing project. The first multiprocessor capable Linux kernel had a single lock protecting the entire kernel (the Big Kernel Lock, BKL), which was simple, but limited scalability.
Subsequently the locking has been made more fine-grained, i.e. there are many locks (thousands?), each covering only a small portion of data. However, there are limits to how far this can be taken, as fine-grained locking tends to be complicated, and the locking overhead starts to eat up the performance benefit, especially considering that most multi-CPU Linux systems have relatively few CPU's.
Another thing, is that as far as possible the kernel uses per-cpu data structures. This is very important, as it avoids the cache coherency performance issues with shared data, and of course there is no locking overhead. E.g. every CPU runs its own process scheduler, requiring only occasional global synchronization.
Also, some algorithms are chosen with scalability in mind. E.g. some read-mostly data is protected by Read-Copy-Update (RCU) instead of traditional mutexes; this allows readers to proceed during a concurrent update.
As for memory, Linux tries hard to allocate memory from the same NUMA node as where the process is running. This provides better memory bandwidth and latency for the applications.
My uneducated guess would be that there is a run-queue per processor and a work-stealing algorithm when a processor is idle. I could see this working in an M:N model, where there is a single process per cpu and light-weight processes as the work items. This would then feel similar to a work-stealing threadpool, such as the one in Java-7's fork-join library.
If you really want to know, go pick up Solaris Internals or dig into the Solaris kernel code. I'm still reading Design & Impl of FreeBSD, with Solaris Internals being the next on my list, so all I can do is make wild guesses atm.
I am pretty sure that the SGI Altix we have at work, (which does ccNUMA) uses special hardware for cache coherency.
There is a huge overhead connected to hold 4mb cache per core coherent. It's unlikely to happen in software only.
in an array of 256 cpus you would need 768mb ram just to hold the cache-invalidation bits.
12mb cache / 128 bytes per cache line * 256² cores.
Modifying the OS is one thing, but using unchanged application code is a waste of hardware. When going over some limit (depending on the hardware), the effort to keep coherency and synchronization in order to execute generic code is simply too much. You can do it, but it will be very expensive.
From the OS side you'll need complex affinity model, i.e. not to jump CPUs just because yours is busy. Scheduling threads based on hardware topology - cooperating threads on CPUs that are "close" to minimize penalties. Simple work stealing is not a good solution, you must consider topology. One solution is hierarchical work stealing - steal work by distance, divide topology to sectors and try to steal from closest first.
Touching a bit the lock issue; you'll still use spin-locks nd such, but using totally different implementations. This is probably the most patented field in CS these days.
But, again, you will need to program specifically for such massive scale. Or you'll simply under-use it. No automatic "parallelizers" will do it for you.
The easiest way to do this is to bind each process/thread to a few CPUS, and then only those CPUs would have to compete for a lock on that thread. Obviously, there would need to be some way to move threads around to even out the load, but on a NUMA architecture, you have to minimize this as much as possible.
Even on dual-core intel systems, I'm pretty sure that Linux can already handle "Thousands" of threads with native posix threads.
(Glibc and the kernel both need to be configured to support this, however, but I believe most systems these days have that by default now).
I'm working on an embedded Linux project that interfaces an ARM9 to a hardware video encoder chip, and writes the video out to SD card or USB stick. The software architecture involves a kernel driver that reads data into a pool of buffers, and a userland app that writes the data to a file on the mounted removable device.
I am finding that above a certain data rate (around 750kbyte/sec) I start to see the userland video-writing app stalling for maybe half a second, about every 5 seconds. This is enough to cause the kernel driver to run out of buffers - and even if I could increase the number of buffers, the video data has to be synchronised (ideally within 40ms) with other things that are going on in real time. Between these 5 second "lag spikes", the writes complete well within 40ms (as far as the app is concerned - I appreciate they're buffered by the OS)
I think this lag spike is to do with the way Linux is flushing data out to disk - I note that pdflush is designed to wake up every 5s, my understanding is that this would be what does the writing. As soon as the stall is over the userland app is able to quickly service and write the backlog of buffers (that didn't overflow).
I think the device I'm writing to has reasonable ultimate throughput: copying a 15MB file from a memory fs and waiting for sync to complete (and the usb stick's light to stop flashing) gave me a write speed of around 2.7MBytes/sec.
I'm looking for two kinds of clues:
How can I stop the bursty writing from stalling my app - perhaps process priorities, realtime patches, or tuning the filesystem code to write continuously rather than burstily?
How can I make my app(s) aware of what is going on with the filesystem in terms of write backlog and throughput to the card/stick? I have the ability to change the video bitrate in the hardware codec on the fly which would be much better than dropping frames, or imposing an artificial cap on maximum allowed bitrate.
Some more info: this is a 200MHz ARM9 currently running a Montavista 2.6.10-based kernel.
Updates:
Mounting the filesystem SYNC causes throughput to be much too poor.
The removable media is FAT/FAT32 formatted and must be as the purpose of the design is that the media can be plugged into any Windows PC and read.
Regularly calling sync() or fsync() say, every second causes regular stalls and unacceptably poor throughput
I am using write() and open(O_WRONLY | O_CREAT | O_TRUNC) rather than fopen() etc.
I can't immediately find anything online about the mentioned "Linux realtime filesystems". Links?
I hope this makes sense. First embedded Linux question on stackoverflow? :)
For the record, there turned out to be two main aspects that seem to have eliminated the problem in all but the most extreme cases. This system is still in development and hasn't been thoroughly torture-tested yet but is working fairly well (touch wood).
The big win came from making the userland writer app multi-threaded. It is the calls to write() that block sometimes: other processes and threads still run. So long as I have a thread servicing the device driver and updating frame counts and other data to sychronise with other apps that are running, the data can be buffered and written out a few seconds later without breaking any deadlines. I tried a simple ping-pong double buffer first but that wasn't enough; small buffers would be overwhelmed and big ones just caused bigger pauses while the filesystem digested the writes. A pool of 10 1MB buffers queued between threads is working well now.
The other aspect is keeping an eye on ultimate write throughput to physical media. For this I am keeping an eye on the stat Dirty: reported by /proc/meminfo. I have some rough and ready code to throttle the encoder if Dirty: climbs above a certain threshold, seems to vaguely work. More testing and tuning needed later. Fortunately I have lots of RAM (128M) to play with giving me a few seconds to see my backlog building up and throttle down smoothly.
I'll try to remember to pop back and update this answer if I find I need to do anything else to deal with this issue. Thanks to the other answerers.
I'll throw out some suggestions, advice is cheap.
make sure you are using a lower level API for writing to the disk, don't use user-mode caching functions like fopen, fread, fwrite use the lower level functions open, read, write.
pass the O_SYNC flag when you open the file, this will cause each write operation to block until written to disk, which will remove the bursty behavior of your writes...with the expense of each write being slower.
If you are doing reads/ioctls from a device to grab a chunk of video data, you may want to consider allocating a shared memory region between the application and kernel, otherwise you are getting hit with a bunch of copy_to_user calls when transferring video data buffers from kernel space to user space.
You may need to validate that your USB flash device is fast enough with sustained transfers to write the data.
Just a couple thoughts, hope this helps.
Here is some information about tuning pdflush for write-heavy operations.
Sounds like you're looking for linux realtime filesystems. Be sure to search Google et al for that.
XFS has a realtime option, though I haven't played with it.
hdparm might let you turn off the caching altogether.
Tuning the filesystem options (turn off all the extra unneeded file attributes) might reduce what you need to flush, thus speeding the flush. I doubt that'd help much, though.
But my suggestion would be to avoid using the stick as a filesystem at all and instead use it as a raw device. Stuff data on it like you would using 'dd'. Then elsewhere read that raw data and write it out after baking.
Of course, I don't know if that's an option for you.
Has a debugging aid, you could use strace to see what operations is taking time.
There might be some surprising thing with the FAT/FAT32.
Do you write into a single file, or in multiple file ?
You can make a reading thread, that will maintain a pool of video buffer ready to be written in a queue.
When a frame is received, it is added to the queue, and the writing thread is signaled
Shared data
empty_buffer_queue
ready_buffer_queue
video_data_ready_semaphore
Reading thread :
buf=get_buffer()
bufer_to_write = buf_dequeue(empty_buffer_queue)
memcpy(bufer_to_write, buf)
buf_enqueue(bufer_to_write, ready_buffer_queue)
sem_post(video_data_ready_semaphore)
Writing thread
sem_wait(vido_data_ready_semaphore)
bufer_to_write = buf_dequeue(ready_buffer_queue)
write_buffer
buf_enqueue(bufer_to_write, empty_buffer_queue)
If your writing threaded is blocked waiting for the kernel, this could work.
However, if you are blocked inside the kerne space, then thereis nothing much you can do, except looking for a more recent kernel than your 2.6.10
Without knowing more about your particular circumstances, I can only offer the following guesses:
Try using fsync()/sync() to force the kernel to flush data to the storage device more frequently. It sounds like the kernel buffers all your writes and then ties up the bus or otherwise stalls your system while performing the actual write. With careful calls to fsync() you can try to schedule writes over the system bus in a more fine grained way.
It might make sense to structure the application in such a way that the encoding/capture (you didn't mention video capture, so I'm making an assumption here - you might want to add more information) task runs in its own thread and buffers its output in userland - then, a second thread can handle writing to the device. This will give you a smoothing buffer to allow the encoder to always finish its writes without blocking.
One thing that sounds suspicious is that you only see this problem at a certain data rate - if this really was a buffering issue, I'd expect the problem to happen less frequently at lower data rates, but I'd still expect to see this issue.
In any case, more information might prove useful. What's your system's architecture? (In very general terms.)
Given the additional information you provided, it sounds like the device's throughput is rather poor for small writes and frequent flushes. If you're sure that for larger writes you can get sufficient throughput (and I'm not sure that's the case, but the file system might be doing something stupid, like updating the FAT after every write) then having an encoding thread piping data to a writing thread with sufficient buffering in the writing thread to avoid stalls. I've used shared memory ring buffers in the past to implement this kind of scheme, but any IPC mechanism that would allow the writer to write to the I/O process without stalling unless the buffer is full should do the trick.
A useful Linux function and alternative to sync or fsync is sync_file_range. This lets you schedule data for writing without waiting for the in-kernel buffer system to get around to it.
To avoid long pauses, make sure your IO queue (for example: /sys/block/hda/queue/nr_requests) is large enough. That queue is where data goes in between being flushed from memory and arriving on disk.
Note that sync_file_range isn't portable, and is only available in kernels 2.6.17 and later.
I've been told that after the host sends a command, MMC and SD cards "must respond within 0 to 8 bytes".
However, the spec allows these cards to respond with "busy" until they have finished the operation, and apparently there is no limit to how long a card can claim to be busy (please, please tell me if there is such a limit).
I see that some low-cost flash chips such as the M25P80 have a guaranteed "maximum single-sector erase time" of 3 seconds, although typically it "only" requires 0.6 seconds.
That 0.6 seconds sounds suspiciously similar to your "stalling for maybe half a second".
I suspect the tradeoff between cheap, slow flash chips and expensive, fast flash chips has something to do with the wide variation in USB flash drive results:
http://www.testfreaks.com/blog/information/16gb-usb-drive-comparison-17-drives-compared/
http://www.tomshardware.com/reviews/data-transfer-run,1037-10.html
I've heard rumors that every time a flash sector is erased and then re-programmed, it takes a little bit longer than the last time.
So if you have a time-critical application, you may need to (a) test your SD cards and USB sticks to make sure they meet the minimum latency, bandwidth, etc. required by your application, and (b) peridically re-test or pre-emptively replace these memory devices.
Well obvious first, have you tried explicitly telling the file to flush? I also think there might be some ioctl you can use to do it, but I honestly haven't done much C/POSIX file programming.
Seeing you're on a Linux kernel you should be able to tune and rebuild the kernel to something that suits your needs better, eg. much more frequent but then also smaller flushes to the permanent storage.
A quick check in my man pages finds this:
SYNC(2) Linux Programmer’s Manual SYNC(2)
NAME
sync - commit buffer cache to disk
SYNOPSIS
#include <unistd.h>
void sync(void);
Feature Test Macro Requirements for glibc (see feature_test_macros(7)):
sync(): _BSD_SOURCE || _XOPEN_SOURCE >= 500
DESCRIPTION
sync() first commits inodes to buffers, and then buffers to disk.
ERRORS
This function is always successful.
Doing your own flush()ing sounds right to me - you want to be in control, not leave it to the vagaries of the generic buffer layer.
This may be obvious, but make sure you're not calling write() too often - make sure every write() has enough data to be written to make the syscall overhead worth it. Also, in the other direction, don't call it too seldom, or it'll block for long enough to cause a problem.
On a more difficult-to-reimplement track, have you tried switching to asynchronous i/o? Using aio you could fire off a write and hand it one set of buffers while you're sucking video data into the other set, and when the write finishes you switch sets of buffers.