Legacy ImageMagick Discussions Archive

It seems that is not a bug. The IM 6.9.9 configured with "--enable-shared --with-modules" is much faster than "--enable-static". Now I want to know how to make the static IM as fast as shared one.

I go on my test and find that IM6.9.9 is much slower than IM6.7.2. I do nothing but read a jpeg image from memory and destory it with 7-threads.
It seems that IM6.9.9 would use much more time in GetMagickInfo() so that the Semaphore Lock would occupy more time. The following code would use almost ...

Yeah, the #750 and #751 pull request is created by me.

I found the bug now. A thread safe issue in MagickWand/wand.c.
WandExport size_t AcquireWandId(void)
{
static size_t
id = 0;

if (wand_semaphore == (SemaphoreInfo *) NULL)
ActivateSemaphoreInfo(&wand_semaphore);
LockSemaphoreInfo(wand_semaphore);
if (wand_ids == (SplayTreeInfo *) NULL)
wand ...

When using the 7 threads in a 24 cores machine, the CPU usage would become really low after running a long time. It is about 400% at first, could become 200% in the final.
Here is my test program, I run it in CentOS 6.5 several times and the problem occured everytime. the ImageMagick version is 6.9 ...