<?xml version="1.0" encoding="utf-8"?>
<feed xml:lang="en" xmlns="http://www.w3.org/2005/Atom"><title>Recent posts to Discussion</title><link href="https://sourceforge.net/p/fastmm/discussion/" rel="alternate"/><link href="https://sourceforge.net/p/fastmm/discussion/feed.atom" rel="self"/><id>https://sourceforge.net/p/fastmm/discussion/</id><updated>2018-11-27T17:25:39.675000Z</updated><subtitle>Recent posts to Discussion</subtitle><entry><title>Delphi 2009</title><link href="https://sourceforge.net/p/fastmm/discussion/443400/thread/fad3b1b074/?limit=25#13d8" rel="alternate"/><published>2018-11-27T17:25:39.675000Z</published><updated>2018-11-27T17:25:39.675000Z</updated><author><name>Ivan Ivanov</name><uri>https://sourceforge.net/u/ironcrossbg/</uri></author><id>https://sourceforge.netdc054c30ba164147df209d65d6bf4b1692203bd9</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Is there a way to force FastMM to release freed memory back to the OS? Using Delphi 2009 with buildin FastMM. Tried to change the build configuration to Release instead of Debug, but FastMM still keeps the memory instead of releasing it back. Back in Delphi 7, when you free something, it insta released it back to the OS. I know it might not be good for performance.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Splash screen</title><link href="https://sourceforge.net/p/fastmm/discussion/443399/thread/58bf13bb/?limit=25#63af" rel="alternate"/><published>2018-02-08T22:06:20.051000Z</published><updated>2018-02-08T22:06:20.051000Z</updated><author><name>Jeffrey Eib</name><uri>https://sourceforge.net/u/jjeib213/</uri></author><id>https://sourceforge.netef4401bd9e00484191c167c3eddaba325535acde</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Sorry. The &amp;gt;&amp;gt; did not display as I expected. The access violation error occurs in the call to LogStackTrace.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Splash screen</title><link href="https://sourceforge.net/p/fastmm/discussion/443399/thread/58bf13bb/?limit=25#5c0e" rel="alternate"/><published>2018-02-08T22:05:08.378000Z</published><updated>2018-02-08T22:05:08.378000Z</updated><author><name>Jeffrey Eib</name><uri>https://sourceforge.net/u/jjeib213/</uri></author><id>https://sourceforge.net5340abfad4117708319e3cc253024b739d69d74b</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;I have an application that takes some time to open due to significant preloading of lists from a database. I implemented a splash screen, opening it in the project source file. This is the code:&lt;/p&gt;
&lt;p&gt;begin&lt;br/&gt;
  Application.Initialize;&lt;br/&gt;
  Application.MainFormOnTaskbar := True;&lt;/p&gt;
&lt;p&gt;frm_SplashScreen := Tfrm_SplashScreen.Create(Application);&lt;br/&gt;
  frm_SplashScreen.ProgramName := 'Garrett Customer Relations';&lt;br/&gt;
  frm_SplashScreen.CopyrightYear := 2016;&lt;br/&gt;
  frm_SplashScreen.SplashImageRes := 'SPLASHIMAGE';&lt;br/&gt;
  frm_SplashScreen.ProgressMessage := 'Loading';&lt;/p&gt;
&lt;p&gt;Application.CreateForm(Tdat_LinerData, dat_LinerData);  // DB unit&lt;br/&gt;
  Application.CreateForm(Tdat_CoverData, dat_CoverData);  // DB unit&lt;br/&gt;
  Application.CreateForm(Tfrm_CustomerRelations, frm_CustomerRelations);  // Main Form&lt;br/&gt;
     .&lt;br/&gt;
     .&lt;br/&gt;
  frm_SplashScreen.Hide;&lt;br/&gt;
  frm_SplashScreen.Free;&lt;br/&gt;
  frm_SplashScreen := nil;&lt;/p&gt;
&lt;p&gt;Application.Run;&lt;br/&gt;
end.&lt;/p&gt;
&lt;p&gt;When I run the application and close it, I get an access violation error in the FastMM4 unit at the line mark by &amp;gt;&amp;gt; below:&lt;/p&gt;
&lt;div class="codehilite"&gt;&lt;pre&gt;&lt;span&gt;&lt;/span&gt;if APointer.AllocationStackTrace[0] &amp;lt;&amp;gt; 0 then
begin
  LMsgPtr := AppendStringToBuffer(ThreadIDAtAllocMsg, LMsgPtr, Length(ThreadIDAtAllocMsg));
  LMsgPtr := NativeUIntToHexBuf(APointer.AllocatedByThread, LMsgPtr);
  LMsgPtr := AppendStringToBuffer(StackTraceMsg, LMsgPtr, Length(StackTraceMsg));
&lt;/pre&gt;&lt;/div&gt;


&lt;blockquote&gt;
&lt;blockquote&gt;
&lt;div class="codehilite"&gt;&lt;pre&gt;&lt;span&gt;&lt;/span&gt; LMsgPtr := LogStackTrace(@APointer.AllocationStackTrace, StackTraceDepth, LMsgPtr);
end;
&lt;/pre&gt;&lt;/div&gt;


&lt;/blockquote&gt;
&lt;/blockquote&gt;
&lt;p&gt;I think that this is being caused by my unusual method of instantiating the splash screen  in the .dpr file before the call to Application.Run command, but I have no way of finding out for sure.  Has anybody else seen this behavior? Am I correct in my assumption of the cause of the error?&lt;/p&gt;
&lt;p&gt;Thanks in advance for your response.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Performance (again)</title><link href="https://sourceforge.net/p/fastmm/discussion/443399/thread/2eb69931/?limit=25#5934" rel="alternate"/><published>2017-07-14T05:46:12.259000Z</published><updated>2017-07-14T05:46:12.259000Z</updated><author><name>Maxim Masiutin</name><uri>https://sourceforge.net/u/maximmasiutin/</uri></author><id>https://sourceforge.netd1d8b32887769dc84067733baa3805135f217d67</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;FastMM4, by default, on thread contention, when one thread cannot acquire access to data, locked by another thread, calls Windows API function Sleep(0), and then, if the lock is still not available enters a loop by calling Sleep(1) after each check of the lock.&lt;/p&gt;
&lt;p&gt;Each call to Sleep(0) experiences the expensive cost of a context switch, which can be 10000+ cycles; it also suffers the cost of ring 3 to ring 0 transitions, which can be 1000+ cycles. As about Sleep(1) – besides the costs associated with Sleep(0) – it also delays execution by at least 1 millisecond, ceding control to other threads, and, if there are no threads waiting to be executed by a physical CPU core, puts the core into sleep, effectively reducing CPU usage and power consumption. &lt;/p&gt;
&lt;p&gt;This way of acquiring locks can be improved by using &lt;code&gt;pause&lt;/code&gt;-based spin-locks.&lt;/p&gt;
&lt;p&gt;A better way would have been a spin-lock of about 5000 &lt;code&gt;pause&lt;/code&gt; instructions, and, if the lock was still busy, calling SwitchToThread() API call. If &lt;code&gt;pause&lt;/code&gt; is not available (on very old processors with no SSE2 support) or SwitchToThread() API call was not available (on very old Windows versions, prior to Windows 2000), the best solution would be to utilize EnterCriticalSection / LeaveCriticalSection, that don’t have latency associated by Sleep(1), and which also very effectively cedes control of the CPU core to other threads.&lt;/p&gt;
&lt;p&gt;I have modified FastMM4 to use a new approach to waiting for a lock: CriticalSections instead of Sleep(). With these options, the Sleep() will never be used but EnterCriticalSection / LeaveCriticalSection will be used instead. Testing has shown that the approach of using CriticalSections instead of Sleep (which was used by default before in FastMM4) provides significant gain in situations when the number of threads working with the memory manager is the same or higher than the number of physical cores. The gain is even more evident on computers with multiple physical CPUs and Non-Uniform Memory Access (NUMA). I have implemented compile-time options to take away the original FastMM4 approach of using Sleep(InitialSleepTime) and then Sleep(AdditionalSleepTime) (or Sleep(0) and Sleep(1)) and replace them with EnterCriticalSection / LeaveCriticalSection to save valuable CPU cycles wasted by Sleep(0) and to improve speed (reduce latency) that was affected each time by at least 1 millisecond by Sleep(1), because the Critical Sections are much more CPU-friendly and have definitely lower latency than Sleep(1).&lt;/p&gt;
&lt;p&gt;When these options are enabled, FastMM4-AVX it checks:&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;whether the CPU supports SSE2 and thus the "pause" instruction, and&lt;/li&gt;
&lt;li&gt;whether the operating system has the SwitchToThread() API call, and,&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;and in this case uses "pause" spin-loop for 5000 iterations and then SwitchToThread() instead of critical sections; If a CPU doesn't have the "pause" instrcution or Windows doesn't have the SwitchToThread() API function, it will use EnterCriticalSection/LeaveCriticalSection.&lt;br/&gt;
I have made available the fork called FastMM4-AVX at &lt;a href="https://github.com/maximmasiutin/FastMM4" rel="nofollow"&gt;https://github.com/maximmasiutin/FastMM4&lt;/a&gt;&lt;/p&gt;
&lt;p&gt;Here are the comparison of the Original FastMM4 version 4.992, with default options compiled for Win64 by Delphi 10.2 Tokyo (Release with Optimization), and the current FastMM4-AVX branch. Under some scenarios, the FastMM4-AVX branch is more than twice as fast comparing to the Original FastMM4. The tests have been run on two different computers: one under Xeon E6-2543v2 with 2 CPU sockets, each has 6 physical cores (12 logical threads) - with only 5 physical core per socket enabled for the test application. Another test was done under a i7-7700K CPU.&lt;/p&gt;
&lt;p&gt;Used the "Multi-threaded allocate, use and free" and "NexusDB" test cases from the FastCode Challenge Memory Manager test suite, modified to run under 64-bit.&lt;/p&gt;
&lt;div class="codehilite"&gt;&lt;pre&gt;                     Xeon E6-2543v2 2*CPU     i7-7700K CPU
                    (allocated 20 logical  (allocated 8 logical
                     threads, 10 physical   threads, 4 physical
                     cores, NUMA)           cores)

                    Orig.  AVX-br.  Ratio   Orig.  AVX-br. Ratio
                    ------  -----  ------   -----  -----  ------
02-threads realloc   96552  59951  62.09%   65213  49471  75.86%
04-threads realloc   97998  39494  40.30%   64402  47714  74.09%
08-threads realloc   98325  33743  34.32%   64796  58754  90.68%
16-threads realloc  116708  45855  39.29%   71457  60173  84.21%
16-threads realloc  116273  45161  38.84%   70722  60293  85.25%
31-threads realloc  122528  53616  43.76%   70939  62962  88.76%
64-threads realloc  137661  54330  39.47%   73696  64824  87.96%
NexusDB 02 threads  122846  90380  73.72%   79479  66153  83.23%
NexusDB 04 threads  122131  53103  43.77%   69183  43001  62.16%
NexusDB 08 threads  124419  40914  32.88%   64977  33609  51.72%
NexusDB 12 threads  181239  55818  30.80%   83983  44658  53.18%
NexusDB 16 threads  135211  62044  43.61%   59917  32463  54.18%
NexusDB 31 threads  134815  48132  33.46%   54686  31184  57.02%
NexusDB 64 threads  187094  57672  30.25%   63089  41955  66.50%
&lt;/pre&gt;&lt;/div&gt;


&lt;p&gt;You can find better tests of the memory manager in the FastCode challenge test suite at &lt;a href="http://fastcode.sourceforge.net/"&gt;http://fastcode.sourceforge.net/&lt;/a&gt;&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Zeroize freed memory</title><link href="https://sourceforge.net/p/fastmm/discussion/443399/thread/97d1841e/?limit=25#c7c1" rel="alternate"/><published>2017-06-21T01:15:03.581000Z</published><updated>2017-06-21T01:15:03.581000Z</updated><author><name>Maxim Masiutin</name><uri>https://sourceforge.net/u/maximmasiutin/</uri></author><id>https://sourceforge.nete44f8c15b8d89796254ca99eb66ec50cefad1c92</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;What do you think about an option to fill freed (not actually used) pages with zero bytes? This may improve performance under Windows, and also under VMWare and other virtual machine environments? For example, VMWare and HyperV calculate hash of memory pages, and, if the contents is the same, mark this page as "shared"  inside a virtual machine and between virtual machines on the same host, until the page is modified. It effectively decreases memory consumption. Windows does the same - it handles zero pages differently, treating them as free.&lt;/p&gt;
&lt;p&gt;We could have FastMM automatically fill memory with zeros when we call FreeMem/ReallocMem. As an alternative option, we could have a function (like we now have ScanMemoryPoolForCorruption) that zeroizes empty memory. Of course, this function have to be thread-safe.&lt;/p&gt;
&lt;p&gt;The drawback of filling memory with zeros is touching the memory, which might have already been turned into virtual, thus issuing page faults.&lt;/p&gt;
&lt;p&gt;If we manage to fill 4-K pages completely with zeros, the hypervisor or Windows will explicitly mark it as a zero page. But even partial zeroizing may be beneficial, since the hypervisor may compress pages using LZ or similar algorithms to save physical memory.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Duplicate allocation number</title><link href="https://sourceforge.net/p/fastmm/discussion/443400/thread/539e6538/?limit=25#1442" rel="alternate"/><published>2017-02-15T04:55:17.756000Z</published><updated>2017-02-15T04:55:17.756000Z</updated><author><name>Andrei Korshunov</name><uri>https://sourceforge.net/u/andkors/</uri></author><id>https://sourceforge.net1e6ff65a714628bf3778dcd1613066bea0083ec9</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;A similar problem.&lt;br/&gt;
In search of memory leaks write a memory dump every 10 minutes in a separate file.&lt;br/&gt;
The analysis found that blocks with numbers 4563031, 4562241, 4561344, 4561553 duplicate.&lt;/p&gt;
&lt;p&gt;Written by a special program for log analysis, which examines the number of consecutive log into the database.&lt;/p&gt;
&lt;p&gt;Recording logs do a timer in the study on memory loss program:&lt;/p&gt;
&lt;div class="codehilite"&gt;&lt;pre&gt;&lt;span class="nt"&gt;procedure&lt;/span&gt; &lt;span class="nt"&gt;TFS_CL_Oper_fmMain&lt;/span&gt;&lt;span class="nc"&gt;.Timer1Timer&lt;/span&gt;&lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="nt"&gt;Sender&lt;/span&gt;&lt;span class="o"&gt;:&lt;/span&gt; &lt;span class="nt"&gt;TObject&lt;/span&gt;&lt;span class="o"&gt;);&lt;/span&gt;
&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="err"&gt;$&lt;/span&gt;&lt;span class="n"&gt;IFDEF&lt;/span&gt; &lt;span class="n"&gt;LOGGING&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="nt"&gt;var&lt;/span&gt;
  &lt;span class="nt"&gt;HostName&lt;/span&gt;&lt;span class="o"&gt;,&lt;/span&gt; &lt;span class="nt"&gt;FName&lt;/span&gt;&lt;span class="o"&gt;:&lt;/span&gt; &lt;span class="nt"&gt;String&lt;/span&gt;&lt;span class="o"&gt;;&lt;/span&gt;
&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="err"&gt;$&lt;/span&gt;&lt;span class="n"&gt;ENDIF&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="nt"&gt;begin&lt;/span&gt;
&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="err"&gt;$&lt;/span&gt;&lt;span class="n"&gt;IFDEF&lt;/span&gt; &lt;span class="n"&gt;LOGGING&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
  &lt;span class="nt"&gt;HostName&lt;/span&gt; &lt;span class="o"&gt;:=&lt;/span&gt; &lt;span class="s1"&gt;''&lt;/span&gt;&lt;span class="o"&gt;;&lt;/span&gt;
  &lt;span class="nt"&gt;if&lt;/span&gt; &lt;span class="nt"&gt;Tics&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nt"&gt;0&lt;/span&gt; &lt;span class="nt"&gt;then&lt;/span&gt;
  &lt;span class="nt"&gt;begin&lt;/span&gt;
    &lt;span class="nt"&gt;FName&lt;/span&gt; &lt;span class="o"&gt;:=&lt;/span&gt; &lt;span class="nt"&gt;ExtractFileDir&lt;/span&gt;&lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="nt"&gt;ParamStr&lt;/span&gt;&lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="nt"&gt;0&lt;/span&gt;&lt;span class="o"&gt;))&lt;/span&gt; &lt;span class="o"&gt;+&lt;/span&gt; &lt;span class="s1"&gt;'\OT_'&lt;/span&gt;&lt;span class="o"&gt;;&lt;/span&gt;
    &lt;span class="nt"&gt;If&lt;/span&gt; &lt;span class="nt"&gt;HostName&lt;/span&gt; &lt;span class="o"&gt;&amp;lt;&amp;gt;&lt;/span&gt; &lt;span class="s1"&gt;''&lt;/span&gt; &lt;span class="nt"&gt;then&lt;/span&gt;
      &lt;span class="nt"&gt;FName&lt;/span&gt; &lt;span class="o"&gt;:=&lt;/span&gt; &lt;span class="nt"&gt;FName&lt;/span&gt; &lt;span class="o"&gt;+&lt;/span&gt; &lt;span class="nt"&gt;HostName&lt;/span&gt; &lt;span class="o"&gt;+&lt;/span&gt; &lt;span class="s1"&gt;'_'&lt;/span&gt;&lt;span class="o"&gt;;&lt;/span&gt;
    &lt;span class="nt"&gt;SetMMLogFileName&lt;/span&gt;&lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="nt"&gt;PAnsiChar&lt;/span&gt;&lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="nt"&gt;AnsiString&lt;/span&gt;&lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="nt"&gt;FName&lt;/span&gt; &lt;span class="o"&gt;+&lt;/span&gt; &lt;span class="nt"&gt;IntToStr&lt;/span&gt;&lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="nt"&gt;CurNum&lt;/span&gt;&lt;span class="o"&gt;)&lt;/span&gt; &lt;span class="o"&gt;+&lt;/span&gt; &lt;span class="s1"&gt;'.Txt'&lt;/span&gt;&lt;span class="o"&gt;)));&lt;/span&gt;
    &lt;span class="nt"&gt;LogAllocatedBlocksToFile&lt;/span&gt;&lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="nt"&gt;0&lt;/span&gt;&lt;span class="o"&gt;,&lt;/span&gt; &lt;span class="o"&gt;$&lt;/span&gt;&lt;span class="nt"&gt;ffffffff&lt;/span&gt;&lt;span class="o"&gt;);&lt;/span&gt;
    &lt;span class="nt"&gt;Inc&lt;/span&gt;&lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="nt"&gt;CurNum&lt;/span&gt;&lt;span class="o"&gt;);&lt;/span&gt;
  &lt;span class="nt"&gt;end&lt;/span&gt; &lt;span class="nt"&gt;else&lt;/span&gt;
    &lt;span class="nt"&gt;inc&lt;/span&gt;&lt;span class="o"&gt;(&lt;/span&gt;&lt;span class="nt"&gt;Tics&lt;/span&gt;&lt;span class="o"&gt;);&lt;/span&gt;
  &lt;span class="nt"&gt;if&lt;/span&gt; &lt;span class="nt"&gt;Tics&lt;/span&gt; &lt;span class="o"&gt;&amp;gt;=&lt;/span&gt; &lt;span class="nt"&gt;10&lt;/span&gt; &lt;span class="nt"&gt;then&lt;/span&gt;
    &lt;span class="nt"&gt;Tics&lt;/span&gt; &lt;span class="o"&gt;:=&lt;/span&gt; &lt;span class="nt"&gt;0&lt;/span&gt;&lt;span class="o"&gt;;&lt;/span&gt;
  &lt;span class="nt"&gt;Timer1&lt;/span&gt;&lt;span class="nc"&gt;.Interval&lt;/span&gt; &lt;span class="o"&gt;:=&lt;/span&gt; &lt;span class="nt"&gt;LoggingInterval&lt;/span&gt;&lt;span class="o"&gt;;&lt;/span&gt;
  &lt;span class="nt"&gt;if&lt;/span&gt; &lt;span class="nt"&gt;CurNum&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nt"&gt;8&lt;/span&gt; &lt;span class="nt"&gt;then&lt;/span&gt;
    &lt;span class="nt"&gt;Halt&lt;/span&gt;&lt;span class="o"&gt;;&lt;/span&gt;
&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="err"&gt;$&lt;/span&gt;&lt;span class="n"&gt;ENDIF&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="nt"&gt;end&lt;/span&gt;&lt;span class="o"&gt;;&lt;/span&gt;
&lt;/pre&gt;&lt;/div&gt;


&lt;p&gt;The result can be seen, the block number 4561344 in file OT_5.Txt (IdFyles = 6) is present 2 times with different class, and in OT_6.Txt file (IdFyles = 7) taken after 10 minutes - there is only one instance of this block. To complete the picture I put the log files themselves (OT_5.Txt file in the database has IdFyles = 6, OT_6.Txt has IdFyles = 7).&lt;/p&gt;
&lt;p&gt;How can it be?&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Disable/Enable FastMM4</title><link href="https://sourceforge.net/p/fastmm/discussion/443400/thread/dbe834a7/" rel="alternate"/><published>2016-11-22T18:22:37.589000Z</published><updated>2016-11-22T18:22:37.589000Z</updated><author><name>Dércio Luis Zanatta</name><uri>https://sourceforge.net/u/derciozanatta/</uri></author><id>https://sourceforge.net9c3d4bcd9d89d29768b51cc1ae0df2db1806d892</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Hello.&lt;br/&gt;
I have an application made in Delphi 7 and had several problems with "out of memory" After including the FastMM4.pas in the project the problem was solved 100%, but caused another problem. My application sends NFCe to the SEFAZ Web Service and after it includes the FASTMM4.Pas, I noticed that the instances of communication failure by Time out increase considerably. I would like to know if there is any way to disable / enable FastMM4 via command, because that way I could disable it before connecting to the WebService and activate it again after receiving the return.&lt;br/&gt;
It is possible  ?&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>When you run the exe file directly, detection dialog box does not appear.</title><link href="https://sourceforge.net/p/fastmm/discussion/443400/thread/59d5d6cb/" rel="alternate"/><published>2016-10-11T05:25:49.079000Z</published><updated>2016-10-11T05:25:49.079000Z</updated><author><name>Taro Tanaka</name><uri>https://sourceforge.net/u/tarotanaka/</uri></author><id>https://sourceforge.netc58e59f7c9a8e34a2341eeb21ba2e6b4aa92b968</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Mr. Muetze1. Thank you for advice.&lt;/p&gt;
&lt;p&gt;Firm to confirm the contents of the FastMM4Options.inc, was changed as follows, now works correctly.&lt;/p&gt;
&lt;p&gt;1, add the following definition to the IDE&lt;br/&gt;
 borlndmmdll&lt;br/&gt;
 FullDebugMode&lt;br/&gt;
 debugdll&lt;br/&gt;
 dllforide&lt;br/&gt;
 LogErrorsToFile&lt;br/&gt;
 LogMemoryLeakDetailToFile&lt;/p&gt;
&lt;p&gt;2, change the definitions in FastMM4Options.inc to be the same as above&lt;br/&gt;
 {. $ Define borlndmmdll} -&amp;gt; {$ define borlndmmdll}&lt;br/&gt;
 {. $ Define FullDebugMode} -&amp;gt; {$ define FullDebugMode}&lt;br/&gt;
 {. $ Define debugdll} -&amp;gt; {$ define debugdll}&lt;br/&gt;
 {. $ Define dllforide} -&amp;gt; {$ define dllforide}&lt;/p&gt;
&lt;p&gt;Thank you!!&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>When you run the exe file directly, detection dialog box does not appear.</title><link href="https://sourceforge.net/p/fastmm/discussion/443400/thread/59d5d6cb/" rel="alternate"/><published>2016-10-08T10:24:26.186000Z</published><updated>2016-10-08T10:24:26.186000Z</updated><author><name>Muetze1</name><uri>https://sourceforge.net/u/muetze1/</uri></author><id>https://sourceforge.net8e0aa6244dc45b8b60019ad8b7cd6b64e7a3932a</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Hi!&lt;/p&gt;
&lt;p&gt;Please take a look in the FastMM4Options.inc for the possible options. Inside the file is a section "Memory Leak Reporting" where are multiple options to control the reporting. There are multiple options that can prevent the dialog to appear after running the application stand-alone:&lt;/p&gt;
&lt;p&gt;RequireIDEPresenceForLeakReporting&lt;br/&gt;
RequireDebuggerPresenceForLeakReporting&lt;br/&gt;
(RequireDebugInfoForLeakReporting)&lt;br/&gt;
ManualLeakReportingControl&lt;br/&gt;
HideMemoryLeakHintMessage&lt;/p&gt;
&lt;p&gt;I suggest to check the first two compiler defines as a first attempt and also check the FastMM4Options.inc for any undefine and other settings according to your project settings.&lt;/p&gt;
&lt;p&gt;Hope that helps &amp;amp; Greets,&lt;br/&gt;
Muetze1&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>When you run the exe file directly, detection dialog box does not appear.</title><link href="https://sourceforge.net/p/fastmm/discussion/443400/thread/59d5d6cb/" rel="alternate"/><published>2016-10-07T01:25:07.046000Z</published><updated>2016-10-07T01:25:07.046000Z</updated><author><name>Taro Tanaka</name><uri>https://sourceforge.net/u/tarotanaka/</uri></author><id>https://sourceforge.netecf17345f12c0105b1990af8e7e28b4290467525</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Create a program to the memory leak, and run from the IDE to detect memory leaks, detection dialog box will appear. However, when run by double-clicking the exe file, detection dialog does not appear. To also display the detection dialog when you run from the exe file is, do you need something special definition? Currently defines the following three.&lt;br/&gt;
  FullDebugMode&lt;br/&gt;
  LogMemoryLeakDetailToFile&lt;br/&gt;
  ClearLogFileOnStartup&lt;/p&gt;
&lt;p&gt;Development environment&lt;br/&gt;
  Windows 7 64bit&lt;br/&gt;
  C ++ Builder 5&lt;br/&gt;
  FastMM4991&lt;/p&gt;&lt;/div&gt;</summary></entry></feed>