AWARE [SYSTEMS] Imaging expertise for the Delphi developer
AWare Systems, Imaging expertise for the Delphi developer, Home TIFF and LibTiff Mailing List Archive

LibTiff Mailing List

TIFF and LibTiff Mailing List Archive
January 2016

Previous Thread
Next Thread

Previous by Thread
Next by Thread

Previous by Date
Next by Date

Contact

The TIFF Mailing List Homepage
This list is run by Frank Warmerdam
Archive maintained by AWare Systems



Valid HTML 4.01!



Thread

2016.01.25 18:25 "OpenMP enabled libtiff", by Aaron Boxer
2016.01.25 19:05 "Re: OpenMP enabled libtiff", by Aaron Boxer
2016.01.25 19:17 "Re: OpenMP enabled libtiff", by <kandel3@illinois.edu>
2016.01.25 19:18 "Re: OpenMP enabled libtiff", by Bob Friesenhahn
2016.01.25 20:21 "Re: OpenMP enabled libtiff", by Aaron Boxer
2016.01.25 20:45 "Re: OpenMP enabled libtiff", by Bob Friesenhahn
2016.01.25 23:27 "Re: OpenMP enabled libtiff", by <kandel3@illinois.edu>
2016.01.26 09:12 "Re: OpenMP enabled libtiff", by <jcupitt@gmail.com>
2016.01.26 13:45 "Re: OpenMP enabled libtiff", by Aaron Boxer
2016.01.26 14:46 "Re: OpenMP enabled libtiff", by Olivier Paquet
2016.01.26 16:01 "Re: OpenMP enabled libtiff", by Aaron Boxer
2016.01.26 17:10 "Re: OpenMP enabled libtiff", by Bob Friesenhahn
2016.01.26 19:32 "Re: OpenMP enabled libtiff", by Aaron Boxer
2016.01.26 19:29 "Re: OpenMP enabled libtiff", by Aaron Boxer
2016.01.26 20:36 "Re: OpenMP enabled libtiff", by Bob Friesenhahn
2016.01.27 18:32 "Re: OpenMP enabled libtiff", by Aaron Boxer
2016.01.27 19:55 "Re: OpenMP enabled libtiff", by Bob Friesenhahn
2016.01.29 13:28 "Re: OpenMP enabled libtiff", by Aaron Boxer
2016.01.29 14:46 "Re: OpenMP enabled libtiff", by Bob Friesenhahn
2016.01.29 15:48 "Re: OpenMP enabled libtiff", by Aaron Boxer
2016.01.29 15:49 "Re: OpenMP enabled libtiff", by Aaron Boxer
2016.01.29 18:12 "Re: OpenMP enabled libtiff", by Bob Friesenhahn
2016.01.26 09:36 "Re: OpenMP enabled libtiff", by Mat Maher
2016.01.26 09:52 "Re: OpenMP enabled libtiff", by Even Rouault
2016.01.26 14:22 "Re: OpenMP enabled libtiff", by Fred Rothganger
2016.01.27 18:51 "Re: OpenMP enabled libtiff", by Larry Gritz

2016.01.29 13:28 "Re: OpenMP enabled libtiff", by Aaron Boxer

On Wed, Jan 27, 2016 at 2:55 PM, Bob Friesenhahn <
bfriesen@simple.dallas.tx.us> wrote:

> On Wed, 27 Jan 2016, Aaron Boxer wrote:
>
>>
>> Thanks, Bob. So, are you saying that as TIFF is currently designed,
>> memory mapping is beneficial  ? Because of the random access ?
>> This was my experience on windows : turning off memory mapping when using
>> libtiff degraded performance.
>>
>
> The memory mapping does eliminate system call and access time overheads
> associated with seeking, as long as data in the same MMU page (e.g. 4k
> block) has been touched before.  This is why libtiff memory maps for reads
> by default under Unix.
>
> Before making comprehensive decisions about degraded performance, try the
> use case where a large sequence of files is accessed in order (each
> accessed once for each traversal) and the total amount of file data is much
> larger than the amount of memory in your system.  In that case, do you see
> the same win or do you now see a penalty which looks very much like what
> happens when the system runs out of memory ("swapping")?
>

Thanks. This is exactly the work flow I am interested in : tens of
thousands of ~10 MB TIFFs, accessed sequentially.  Only 4 files would be
open at a given time. So, even though the files get closed, are you
suggesting I will see swapping in this case ?


>
> My other question is : why is unix and windows treated differently in
>> tifflib?  mmap call allows sharing of mapping between different file
>> handles, while on windows this is turned off. I think it would be nice to
>> have on windows.
>>
>
> I don't know why this is.  Memory mapping works fine for reads under
> Windows although the maximum contigious map size is likely to be smaller
> than on Unix type systems.  It does not work so well for writes due to
> stupidity and missing certain useful POSIX functions like ftruncate().
> Many functions which might otherwise be useful do not support large files
> or behave in much stupider ways (e.g. actually write zeros to the
> filesystem rather than create a hole) than on Unix type systems.
>
>
Interesting, Unfortunately windows is still a necessary evil.

Aaron



>
> Bob
> --
> Bob Friesenhahn
> bfriesen@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/
> GraphicsMagick Maintainer,    http://www.GraphicsMagick.org/
>