February 2008 - Darwin-dev - lists.apple.com

USB device prevented from mounting
by Nick Rogers 20 Feb '08

20 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: shm_open and EMFILE
by David McGavran 18 Feb '08

18 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Cheers Dave On Feb 18, 2008, at 8:37 PM, Terry Lambert wrote: -- Terry Like I said asking for help always let you find the answer. Debug in XCode 10240 Run in XCode 2200 something like that Finder or LSLaunchApplication 253 Is that correct? Cheers Dave On Feb 16, 2008, at 10:40 PM, David McGavran wrote: Cheers Dave On Feb 16, 2008, at 7:17 AM, Terry Lambert wrote: stdin, stdout, stderr = 3 10237 + 3 = 10240 -- Terry Ok I guess this is good news then :) Thanks! I love it when I am wrong :) Cheers Dave On Feb 15, 2008, at 2:05 AM, Terry Lambert wrote: You are mistaken. man 2 setrlimit _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/dmcgavra%40adobe.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Interesting but it is 100% consistent. If I debug from xcode getrlimit returns 10240. If I log from launching on the finder it says -1 for max and I can open 253. If I run from xcode it is 2000 something or other. In any case using setrlimit fixes it to be consistent. I would also think XCode shouldn't do this. Maybe I am inheriting their value because they are launching us? I don't know what Xcode does here, but I'd be surprised if it did this to you, since you more or less want it to emulate the runtime environment you get normally. If it doesn't, you will want to report it. My guess before your second message was that it was the client that was limiting the connections to the server, but that assumes a single client process. Additionally, if it were on the same machine, I'd expect closer to 5120, given half the decriptors would go to each process. On Feb 16, 2008, at 2:28 PM, David McGavran <dmcgavra(a)adobe.com> wrote: Let me guess XCode sets your limits much higher than running from the finder? Here are my guess: I never enabled the setrlimit code because when I tested it in the debugger it wasn't doing anything, because I was already at max. Ok, I am officially confused. So hopefully I will solve it as soon as I write this email ;) If I launch my app (which is a server) and just loop on creating files I get 10237 which is cool. If I launch my app and have a client connect to it using NSConnection and loop on creating files I get 253, very consistently. Any ideas why that would be? Are there any debugging tricks for this? If I look in activity monitor it is showing the 253 files but not too much else. There are some descriptors used up for normal housekeeping. Also realize that for ach one you have open, that's a file, socket, pipe, etc. that you are not going to be able to open. On Feb 14, 2008, at 11:46 PM, David McGavran wrote: I did a much more simple test and was able to hit 10237 which is much closer to the max of 10240 as I hoped. There must be something else wrong, I will have to dig deeper. On Feb 14, 2008, at 11:43 AM, David McGavran wrote: I would guess this has been asked and answered so any pointers are fine. I tried to check the digest but didn't find the answer. I am trying to setup a lot of mem mapped files. But at or around 240 I get EMFILE which man page says means I have too many open. If I am not mistaken on 10.5 at least we are allowed 12000 or so open files? Why can I only open 240 or so memory mapped files? The limit on the number of files for a process is by default 256, as a soft constraint, and the value of maxfiles, as a hard constraint. The value of maxfiles represents the maximum number of files permitted an individual process. In addition, there is a hard constraint based on whatever your parent process decides to constrain you to; the default value of this is "unlimited" (by which it means "whatever the current sysctl-based hard limits are"). Once lowered the value of the hard limit cannot be reaised except by a process with root privilege. The value of maxfiles is reported by the sysctlbyname() value for "kern.maxfilesperproc"; this in turn is constrained by the the sysctlbyname() value for "kern.maxfiles", which is a compile time system limit. Note that the use of actual file descriptors to implement POSIX shared memory is an implementation detail, so all of the above is potentially subject to change in its relevency if that implementation detail ever changes. What can I do to open more? Also note that other administrative constraints other than number of available file descriptors (such as available kernel memory) will potentially also cause your attempt to ail (e.g. ENOMEM, ENOSPC, SPERM, EACCES, etc.). -- Terry This email sent to dmcgavra(a)adobe.com This email sent to site_archiver(a)lists.apple.com

1 0

Re: shm_open and EMFILE
by Terry Lambert 18 Feb '08

18 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -- Terry On Feb 16, 2008, at 2:28 PM, David McGavran <dmcgavra(a)adobe.com> wrote: Like I said asking for help always let you find the answer. Debug in XCode 10240 Run in XCode 2200 something like that Finder or LSLaunchApplication 253 Is that correct? Cheers Dave On Feb 16, 2008, at 10:40 PM, David McGavran wrote: Cheers Dave On Feb 16, 2008, at 7:17 AM, Terry Lambert wrote: stdin, stdout, stderr = 3 10237 + 3 = 10240 -- Terry Ok I guess this is good news then :) Thanks! I love it when I am wrong :) Cheers Dave On Feb 15, 2008, at 2:05 AM, Terry Lambert wrote: You are mistaken. man 2 setrlimit _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… I don't know what Xcode does here, but I'd be surprised if it did this to you, since you more or less want it to emulate the runtime environment you get normally. If it doesn't, you will want to report it. My guess before your second message was that it was the client that was limiting the connections to the server, but that assumes a single client process. Additionally, if it were on the same machine, I'd expect closer to 5120, given half the decriptors would go to each process. Let me guess XCode sets your limits much higher than running from the finder? Here are my guess: I never enabled the setrlimit code because when I tested it in the debugger it wasn't doing anything, because I was already at max. Ok, I am officially confused. So hopefully I will solve it as soon as I write this email ;) If I launch my app (which is a server) and just loop on creating files I get 10237 which is cool. If I launch my app and have a client connect to it using NSConnection and loop on creating files I get 253, very consistently. Any ideas why that would be? Are there any debugging tricks for this? If I look in activity monitor it is showing the 253 files but not too much else. There are some descriptors used up for normal housekeeping. Also realize that for ach one you have open, that's a file, socket, pipe, etc. that you are not going to be able to open. On Feb 14, 2008, at 11:46 PM, David McGavran wrote: I did a much more simple test and was able to hit 10237 which is much closer to the max of 10240 as I hoped. There must be something else wrong, I will have to dig deeper. On Feb 14, 2008, at 11:43 AM, David McGavran wrote: I would guess this has been asked and answered so any pointers are fine. I tried to check the digest but didn't find the answer. I am trying to setup a lot of mem mapped files. But at or around 240 I get EMFILE which man page says means I have too many open. If I am not mistaken on 10.5 at least we are allowed 12000 or so open files? Why can I only open 240 or so memory mapped files? The limit on the number of files for a process is by default 256, as a soft constraint, and the value of maxfiles, as a hard constraint. The value of maxfiles represents the maximum number of files permitted an individual process. In addition, there is a hard constraint based on whatever your parent process decides to constrain you to; the default value of this is "unlimited" (by which it means "whatever the current sysctl-based hard limits are"). Once lowered the value of the hard limit cannot be reaised except by a process with root privilege. The value of maxfiles is reported by the sysctlbyname() value for "kern.maxfilesperproc"; this in turn is constrained by the the sysctlbyname() value for "kern.maxfiles", which is a compile time system limit. Note that the use of actual file descriptors to implement POSIX shared memory is an implementation detail, so all of the above is potentially subject to change in its relevency if that implementation detail ever changes. What can I do to open more? Also note that other administrative constraints other than number of available file descriptors (such as available kernel memory) will potentially also cause your attempt to ail (e.g. ENOMEM, ENOSPC, SPERM, EACCES, etc.). -- Terry _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/dmcgavra% 40adobe.com This email sent to dmcgavra(a)adobe.com This email sent to site_archiver(a)lists.apple.com

1 0

Re: Ascertaining dirty pages
by Vincent Coetzee 18 Feb '08

18 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi Steven, Thanks anyway Vincent On 18 Feb 2008, at 3:38 PM, Steven Bytnar wrote: man vmmap vmmap has been around since at least 10.2. Here's an example of the output: % vmmap -pages screen Virtual Memory Map of process 5755 (screen) Output report format: 2.2 -- 32-bit process ==== Legend SM=sharing mode: COW=copy_on_write PRV=private NUL=empty ALI=aliased SHM=shared ZER=zero_filled S/A=shared_alias REGION TYPE [ VIRTUAL] =========== [ =======] MALLOC [ 2582] STACK GUARD [ 14344] Stack [ 2048] __DATA [ 139] __LINKEDIT [ 1149] __PAGEZERO [ 1] __TEXT [ 596] shared memory [ 2] --Steve Dear List, Thanks in advance Vince _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/sbytnar%40kagi.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… I am aware of vmmap, but perhaps I did not make the nature of my problem very clear. My application needs to know about these pages, and needs to be able to access information about the pages. Additionally vmmap does not tell one which pages are dirty, ,it just tells you about virtual memory allocated to a process. I need to know on a page by page basis (i.e. currently 4096 bytes per page) which pages have been written to and the exact address of the changes within the page. ==== Non-writable regions for process 5755 __PAGEZERO 00000000-00001000 [ 1] ---/--- SM=NUL .../bin/screen __TEXT 00001000-00046000 [ 69] r-x/rwx SM=COW .../bin/screen __LINKEDIT 0004e000-00051000 [ 3] r--/rwx SM=COW .../bin/screen STACK GUARD 00051000-00052000 [ 1] ---/rwx SM=NUL STACK GUARD 00053000-00054000 [ 1] ---/rwx SM=NUL STACK GUARD 00068000-00069000 [ 1] ---/rwx SM=NUL STACK GUARD 0006a000-0006b000 [ 1] ---/rwx SM=NUL __TEXT 8fe00000-8fe31000 [ 49] r-x/rwx SM=COW / usr/lib/dyld __LINKEDIT 8fe68000-8fe76000 [ 14] r--/rwx SM=COW / usr/lib/dyld __TEXT 91a5c000-91bf6000 [ 410] r-x/r-x SM=COW ...tem.B.dylib __TEXT 93735000-93767000 [ 50] r-x/r-x SM=COW ...s.5.4.dylib __TEXT 93ff4000-94000000 [ 12] r-x/r-x SM=COW ...c_s.1.dylib __TEXT 9611c000-96122000 [ 6] r-x/r-x SM=COW ...mon.A.dylib __LINKEDIT 96e6e000-972da000 [ 1132] r--/r-- SM=COW ...mon.A.dylib STACK GUARD bc000000-bf800000 [ 14336] ---/rwx SM=NUL STACK GUARD fffec000-ffff0000 [ 4] ---/rwx SM=NUL shared memory ffff8000-ffffa000 [ 2] r-x/r-x SM=SHM ==== Writable regions for process 5755 __DATA 00046000-0004e000 [ 8] rw-/rwx SM=COW .../bin/screen MALLOC (freed?) 00052000-00053000 [ 1] rw-/rwx SM=PRV MALLOC_LARGE 00054000-00068000 [ 20] rw-/rwx SM=COW ...ne_0x100000 MALLOC (freed?) 00069000-0006a000 [ 1] rw-/rwx SM=COW MALLOC_TINY 00100000-00200000 [ 256] rw-/rwx SM=PRV ...ne_0x100000 MALLOC_TINY 00200000-00300000 [ 256] rw-/rwx SM=COW zone_0x0 MALLOC_SMALL 00800000-01000000 [ 2048] rw-/rwx SM=PRV ...ne_0x100000 __DATA 8fe31000-8fe68000 [ 55] rw-/rwx SM=COW / usr/lib/dyld __DATA a0336000-a0377000 [ 65] rw-/rwx SM=COW ...tem.B.dylib __DATA a0767000-a0770000 [ 9] rw-/rwx SM=COW ...s.5.4.dylib __DATA a0890000-a0891000 [ 1] rw-/rwx SM=COW ...c_s.1.dylib __DATA a0c02000-a0c03000 [ 1] rw-/rwx SM=COW ...mon.A.dylib Stack bf800000-c0000000 [ 2048] rw-/rwx SM=COW thread 0 ==== Summary for process 5755 ReadOnly portion of Libraries: Total=6980K resident=6700K(96%) swapped_out_or_unallocated=280K(4%) Writable regions: Total=18.3M written=224K(1%) resident=356K(2%) swapped_out=0K(0%) unallocated=18.0M(98%) On Mon, Feb 18, 2008 at 01:43:00PM +0200, Vincent Coetzee wrote: I am hoping someone on these lists can assist me with my little problem. If the problem is not appropriate to these lists, could someone please point me in the right direction. I have a requirement to know which memory pages (allocated by my application) have been written to during the execution of my application, I also need to know the addresses of the areas of the pages that have been changed. I have spent hours perusing the BSD and Darwin documents without finding anything that seems to allow for this. Does anyone know how to do this or can suggest something please, I am at my wits end. This email sent to sbytnar(a)kagi.com This email sent to site_archiver(a)lists.apple.com

1 0

Re: Ascertaining dirty pages
by Steven Bytnar 18 Feb '08

18 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mutt/1.4i man vmmap vmmap has been around since at least 10.2. Here's an example of the output: % vmmap -pages screen Virtual Memory Map of process 5755 (screen) Output report format: 2.2 -- 32-bit process ==== Non-writable regions for process 5755 __PAGEZERO 00000000-00001000 [ 1] ---/--- SM=NUL .../bin/screen __TEXT 00001000-00046000 [ 69] r-x/rwx SM=COW .../bin/screen __LINKEDIT 0004e000-00051000 [ 3] r--/rwx SM=COW .../bin/screen STACK GUARD 00051000-00052000 [ 1] ---/rwx SM=NUL STACK GUARD 00053000-00054000 [ 1] ---/rwx SM=NUL STACK GUARD 00068000-00069000 [ 1] ---/rwx SM=NUL STACK GUARD 0006a000-0006b000 [ 1] ---/rwx SM=NUL __TEXT 8fe00000-8fe31000 [ 49] r-x/rwx SM=COW /usr/lib/dyld __LINKEDIT 8fe68000-8fe76000 [ 14] r--/rwx SM=COW /usr/lib/dyld __TEXT 91a5c000-91bf6000 [ 410] r-x/r-x SM=COW ...tem.B.dylib __TEXT 93735000-93767000 [ 50] r-x/r-x SM=COW ...s.5.4.dylib __TEXT 93ff4000-94000000 [ 12] r-x/r-x SM=COW ...c_s.1.dylib __TEXT 9611c000-96122000 [ 6] r-x/r-x SM=COW ...mon.A.dylib __LINKEDIT 96e6e000-972da000 [ 1132] r--/r-- SM=COW ...mon.A.dylib STACK GUARD bc000000-bf800000 [ 14336] ---/rwx SM=NUL STACK GUARD fffec000-ffff0000 [ 4] ---/rwx SM=NUL shared memory ffff8000-ffffa000 [ 2] r-x/r-x SM=SHM ==== Writable regions for process 5755 __DATA 00046000-0004e000 [ 8] rw-/rwx SM=COW .../bin/screen MALLOC (freed?) 00052000-00053000 [ 1] rw-/rwx SM=PRV MALLOC_LARGE 00054000-00068000 [ 20] rw-/rwx SM=COW ...ne_0x100000 MALLOC (freed?) 00069000-0006a000 [ 1] rw-/rwx SM=COW MALLOC_TINY 00100000-00200000 [ 256] rw-/rwx SM=PRV ...ne_0x100000 MALLOC_TINY 00200000-00300000 [ 256] rw-/rwx SM=COW zone_0x0 MALLOC_SMALL 00800000-01000000 [ 2048] rw-/rwx SM=PRV ...ne_0x100000 __DATA 8fe31000-8fe68000 [ 55] rw-/rwx SM=COW /usr/lib/dyld __DATA a0336000-a0377000 [ 65] rw-/rwx SM=COW ...tem.B.dylib __DATA a0767000-a0770000 [ 9] rw-/rwx SM=COW ...s.5.4.dylib __DATA a0890000-a0891000 [ 1] rw-/rwx SM=COW ...c_s.1.dylib __DATA a0c02000-a0c03000 [ 1] rw-/rwx SM=COW ...mon.A.dylib Stack bf800000-c0000000 [ 2048] rw-/rwx SM=COW thread 0 ==== Legend SM=sharing mode: COW=copy_on_write PRV=private NUL=empty ALI=aliased SHM=shared ZER=zero_filled S/A=shared_alias ==== Summary for process 5755 ReadOnly portion of Libraries: Total=6980K resident=6700K(96%) swapped_out_or_unallocated=280K(4%) Writable regions: Total=18.3M written=224K(1%) resident=356K(2%) swapped_out=0K(0%) unallocated=18.0M(98%) REGION TYPE [ VIRTUAL] =========== [ =======] MALLOC [ 2582] STACK GUARD [ 14344] Stack [ 2048] __DATA [ 139] __LINKEDIT [ 1149] __PAGEZERO [ 1] __TEXT [ 596] shared memory [ 2] --Steve On Mon, Feb 18, 2008 at 01:43:00PM +0200, Vincent Coetzee wrote: > Dear List, > > I am hoping someone on these lists can assist me with my little > problem. If the problem is not appropriate to these lists, could > someone please point me in the right direction. I have a requirement > to know which memory pages (allocated by my application) have been > written to during the execution of my application, I also need to know > the addresses of the areas of the pages that have been changed. I have > spent hours perusing the BSD and Darwin documents without finding > anything that seems to allow for this. Does anyone know how to do this > or can suggest something please, I am at my wits end. > > Thanks in advance > > Vince > _______________________________________________ > Do not post admin requests to the list. They will be ignored. > Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) > Help/Unsubscribe/Update your Subscription: > http://lists.apple.com/mailman/options/darwin-kernel/sbytnar%40kagi.com > > This email sent to sbytnar(a)kagi.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Ascertaining dirty pages
by Vincent Coetzee 18 Feb '08

18 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dear List, Thanks in advance Vince _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… I am hoping someone on these lists can assist me with my little problem. If the problem is not appropriate to these lists, could someone please point me in the right direction. I have a requirement to know which memory pages (allocated by my application) have been written to during the execution of my application, I also need to know the addresses of the areas of the pages that have been changed. I have spent hours perusing the BSD and Darwin documents without finding anything that seems to allow for this. Does anyone know how to do this or can suggest something please, I am at my wits end. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Linker-fu
by Martin Costabel 18 Feb '08

18 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Mozilla Thunderbird 1.0.2 (Macintosh/20050317) Hi. I'm trying to port a general unix application (multimatlab) to osX where there is the folliwing error libmpich.dylib is input for the dynamic linke editor, is not relocatable by the static link editor again The application I'm trying to compile uses several librairies, most from mpich and matlab. Libmpich.dylib is supplied by matlab and file states it's a ppc bundle. If it's a bundle, you can't use it as dylib for linking. -- Martin _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Mathieu Béliveau wrote: This email sent to site_archiver(a)lists.apple.com

1 0

Linker-fu
by Mathieu Béliveau 17 Feb '08

17 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition; bh=uVcQuvvVhYmPWa61bKmMkQG86911FJ8tGN9oUmrv+OQ=; b=ZY8FXKJis10mXP8AjffBy1kcSiwlxMFaWHb1zU9zVmkutSgqdJb6W/USsxE42N7l8eEMrlpUgBCZ4PZnpYllAMNfC20PQtHdSX6g9znLfmZqwhNrVojjHFJjrbA+G3zhnqyhP3JKe66SDh6M6EY732C72wwWktGfDCIQvWk9mk8= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition; b=SBYU6tRpxLAR6glXiTaWeADdjstQpR2BHeUUcTqeIxSSaAIZbetzjbPZOhhWIsyzNk9XjY3GuhK4sGEtN+PxPVDPE2h9IO1qebLjPs+WuJ7/B6zOW480Zg0v9Ix1ykfnlHbO80cGtK5y/h4/qyrRQo6uY5qU/HNgeqcBeqx0UJw= Hi. I'm trying to port a general unix application (multimatlab) to osX where there is the folliwing error libmpich.dylib is input for the dynamic linke editor, is not relocatable by the static link editor again The application I'm trying to compile uses several librairies, most from mpich and matlab. Libmpich.dylib is supplied by matlab and file states it's a ppc bundle. As the provided makefile didn't took osX in account, I had to explicitly set the libs path and set the proper architecture. Concerning the different linker flags, unfortunately, the usual -bundle -flat_namespace, etc. won't change a thing, I still get the same error message. Any help would greatly be appreciated! Here is the make output showing the full compiler options passed to gcc: -> mexopts.sh sourced from directory (DIR = $MATLAB/bin) FILE = /Applications/matlab-new/bin/mexopts.sh ---------------------------------------------------------------- -> MATLAB = /Applications/matlab-new -> CC = gcc-4.0 -> CC flags: CFLAGS = -fno-common -no-cpp-precomp -fexceptions CDEBUGFLAGS = -g COPTIMFLAGS = -O3 -fno-loop-optimize -DNDEBUG CLIBS = -L/Applications/matlab-new/bin/mac -lmx -lmex -lmat -lstdc++ arguments = -DMX_COMPAT_32 -> CXX = g++-4.0 -> CXX flags: CXXFLAGS = -fno-common -no-cpp-precomp -fexceptions -arch ppc CXXDEBUGFLAGS = -g CXXOPTIMFLAGS = -O3 -fno-loop-optimize -DNDEBUG CXXLIBS = -L/Applications/matlab-new/bin/mac -lmx -lmex -lmat -lstdc++ arguments = -DMX_COMPAT_32 -> FC = g95 -> FC flags: FFLAGS = -fexceptions FDEBUGFLAGS = -g FOPTIMFLAGS = -O FLIBS = -L/Applications/matlab-new/bin/mac -lmx -lmex -lmat -L -lf95 arguments = -DMX_COMPAT_32 -> LD = gcc-4.0 -> Link flags: LDFLAGS = -bundle -Wl,-flat_namespace -undefined suppress -Wl,-exported_symbols_list,/Applications/matlab-new/extern/lib/mac/mexFunction.map LDDEBUGFLAGS = -g LDOPTIMFLAGS = -O LDEXTENSION = .mexmac arguments = -L/usr/local/mpich/lib/ -lmpich -lm -> LDCXX = -> Link flags: LDCXXFLAGS = LDCXXDEBUGFLAGS = LDCXXOPTIMFLAGS = LDCXXEXTENSION = arguments = -L/usr/local/mpich/lib/ -lmpich -lm ---------------------------------------------------------------- -> gcc-4.0 -c -I/usr/local/mpich -I/usr/local/mpich/include -I/usr/local/mpich/mpid/ch_p4/p4 -I/usr/local/mpich/mpid/ch_p4/p4/include -I/Applications/matlab-new/extern/include -DMATLAB_MEX_FILE -fno-common -no-cpp-precomp -fexceptions -DMX_COMPAT_32 -O3 -fno-loop-optimize -DNDEBUG MMAT.c MMAT.c:9: warning: initialization makes integer from pointer without a cast MMAT.c:9: warning: initialization makes integer from pointer without a cast MMAT.c:9: warning: initialization makes integer from pointer without a cast MMAT.c:9: warning: initialization makes integer from pointer without a cast -> gcc-4.0 -c -I/usr/local/mpich -I/usr/local/mpich/include -I/usr/local/mpich/mpid/ch_p4/p4 -I/usr/local/mpich/mpid/ch_p4/p4/include -I/Applications/matlab-new/extern/include -DMATLAB_MEX_FILE -fno-common -no-cpp-precomp -fexceptions -DMX_COMPAT_32 -O3 -fno-loop-optimize -DNDEBUG /Applications/matlab-new/extern/src/mexversion.c -> gcc-4.0 -O -bundle -Wl,-flat_namespace -undefined suppress -Wl,-exported_symbols_list,/Applications/matlab-new/extern/lib/mac/mexFunction.map -o MMAT.mexmac MMAT.o mexversion.o -L/usr/local/mpich/lib/ -lmpich -lm -L/Applications/matlab-new/bin/mac -lmx -lmex -lmat -lstdc++ /usr/bin/ld: /usr/local/mpich/lib//libmpich.dylib is input for the dynamic link editor, is not relocatable by the static link editor again collect2: ld returned 1 exit status mex: link of 'MMAT.mexmac' failed. make: *** [MMAT.] Error 1 Below, is the content of my make file: SHELL = /bin/sh ################################################### # # NOTE: # # Currently MultiMATLAB only works with MATLAB 6.x # ################################################### # # ACMC setup # MPI_HOME = /usr/local/mpich MATLAB_DIR := /Applications/matlab-new # # Griffith setup # #MPI_HOME = /usr/local/mpich-1.2.7 #MATLAB_DIR := /usr/local/matlab.6.5.1 # # JCU setup # #MPI_HOME := /sw/applications/mpich-1.2.5.2 #MATLAB_DIR := /sw/applications/matlab6.2/ # # All the needs changing should be above. # That is, don't touch below here unless # you know what you are doing. # MM_SRC := $(shell pwd) MM_HOME = $(patsubst %/src,%, $(MM_SRC)) COMM = $(MPI_HOME)/mpid/ch_p4/p4 CC = $(MATLAB_DIR)/bin/mex CLINKER = $(MATLAB_DIR)/bin/mex LIB_PATH = -L$(MPI_HOME)/lib/ LIB_LIST = -lmpich MPILIB = $(MPI_HOME)/lib/libmpich.a INCLUDE_DIR = -I$(MPI_HOME) -I$(MPI_HOME)/include -I$(COMM) -I$(COMM)/include MKDIR = mkdir RM = rm MV = mv CP = cp DOL = "$$" PLATFORM = $(shell uname -s) ifeq ($(findstring IRIX, $(PLATFORM)), IRIX) # ARCH is the matlab mex format. Type help mex in matlab to get the list of formats. ARCH = mexsg CFLAGS = -v -D_BSD_TYPES $(INCLUDE_DIR) endif ifeq ($(PLATFORM), Linux) ARCH = mexglx CFLAGS = -v -D_XOPEN_SOURCE=600 $(INCLUDE_DIR) endif ifeq ($(PLATFORM), SunOS) ARCH = mexsol CFLAGS = -v $(INCLUDE_DIR) endif CFLAGS = -v $(INCLUDE_DIR) MMAT_BASE = $(MM_HOME) SPMD_OBJS = Init.m \ ID.m \ Nproc.m \ Send.m \ Recv.m \ Probe.m \ Coord.m \ Gridsize.m \ Reduce.m \ Allreduce.m \ Barrier.m \ multimatlab.m \ Multimatlab.m \ MultiMATLAB.m \ Graphics.m \ PerformShift.m \ MMgrid.m \ MMwindow.m \ MMresetwindow.m \ MMsum.m \ MMmax.m \ MMmin.m \ MMprod.m \ MMbcast.m INTERACTIVE_OBJS = Eval.$(ARCH) \ Eval.m \ Quit.m \ Close.m \ Put.m \ Get.m \ pid.$(ARCH) \ Start.m \ Nproc.m \ multimatlab.m \ Multimatlab.m \ MultiMATLAB.m \ Graphics.m \ Init.m \ Max.m\ Min.m \ Window.m \ ResetWindow.m \ Prod.m \ Sum.m \ Grid.m \ Refresh.m \ Bcast.m \ mmchecknodes.m \ Distribute.m \ Shift.m \ DistributeMap.m \ Gather.m \ Collect.m REMOTE_OBJS = startup.m ERR_OBJS = Interrupt.m Abort.m .SUFFIXES: .$(ARCH) .c all: clean dirs mmat common interactive remote err cshrc dirs: mkdir $(MMAT_BASE)/common chmod 755 $(MMAT_BASE)/common mkdir $(MMAT_BASE)/interactive chmod 755 $(MMAT_BASE)/interactive mkdir $(MMAT_BASE)/remote chmod 755 $(MMAT_BASE)/remote mmat: MMAT.$(ARCH) cp MMAT.$(ARCH) $(MMAT_BASE)/common cp MMAT.$(ARCH) $(MMAT_BASE)/interactive chmod 555 $(MMAT_BASE)/common/MMAT.$(ARCH) chmod 555 $(MMAT_BASE)/interactive/MMAT.$(ARCH) common: $(SPMD_OBJS) cp $(SPMD_OBJS) $(MMAT_BASE)/common chmod a+r $(MMAT_BASE)/common/* interactive: $(INTERACTIVE_OBJS) cp $(INTERACTIVE_OBJS) $(MMAT_BASE)/interactive chmod a+r $(MMAT_BASE)/interactive/* remote: $(REMOTE_OBJS) cp $(REMOTE_OBJS) $(MMAT_BASE)/remote chmod a+r $(MMAT_BASE)/remote/* err: $(ERR_OBJS) cp $(ERR_OBJS) $(MMAT_BASE)/interactive chmod a+r $(MMAT_BASE)/interactive/* cshrc: @echo "Creating $(MM_HOME)/cshrc.mm" @echo "#" >> $(MM_HOME)/cshrc.mm @echo "# This file is machine made. Do not edit." >> $(MM_HOME)/cshrc.mm @echo "# Users should include this file in their .cshrc" >> $(MM_HOME)/cshrc.mm @echo "#" >> $(MM_HOME)/cshrc.mm @echo >> $(MM_HOME)/cshrc.mm @echo "setenv MPI_HOME $(MPI_HOME)" >> $(MM_HOME)/cshrc.mm @echo "if (! "$(DOL)"?LD_LIBRARY_PATH) then" >> $(MM_HOME)/cshrc.mm @echo " setenv LD_LIBRARY_PATH" $(DOL)"MPI_HOME/lib" >> $(MM_HOME)/cshrc.mm @echo "else" >> $(MM_HOME)/cshrc.mm @echo " setenv LD_LIBRARY_PATH "$(DOL)"{LD_LIBRARY_PATH}:"$(DOL)"MPI_HOME/lib" >> $(MM_HOME)/cshrc.mm @echo "endif" >> $(MM_HOME)/cshrc.mm @echo >> $(MM_HOME)/cshrc.mm @echo "setenv MATLAB_DIR $(MATLAB_DIR)" >> $(MM_HOME)/cshrc.mm @echo "alias multimatlab "$(DOL)"MATLAB_DIR/bin/matlab -nojvm -nosplash" >> $(MM_HOME)/cshrc.mm @echo >> $(MM_HOME)/cshrc.mm @echo "setenv MM_HOME $(MM_HOME)" >> $(MM_HOME)/cshrc.mm @echo "setenv MATLABPATH "$(DOL)"MM_HOME/interactive:"$(DOL)"MM_HOME/common:"$(DOL)"MM_HOME/example:"$(DOL)"MM_HOME/example/Demo:"$(DOL)"MM_HOME/example/financial" >> $(MM_HOME)/cshrc.mm @echo >> $(MM_HOME)/cshrc.mm chmod 644 $(MM_HOME)/cshrc.mm #patch: # cp $(MM_HOME)/mpich_patch/*.o . #.c.o: mm.h $(MPI_HOME)/include/MPI.h $(MPILIB) # .c.$(ARCH): mm.h $(MPI_HOME)/include/mpi.h $(MPILIB) $(CLINKER) $(CFLAGS) $*.c $(LIB_PATH) $(LIB_LIST) -lm clean: $(RM) -f $(MMAT_BASE)/cshrc.mm $(RM) -f *.o *~ PI* *.$(ARCH) $(EXECS) $(RM) -f $(MMAT_BASE)/common/* $(MMAT_BASE)/interactive/* $(MMAT_BASE)/remote/* $(RM) -rf $(MMAT_BASE)/common $(MMAT_BASE)/interactive $(MMAT_BASE)/remote _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Darwin 9.2 Sources
by David M. Cotter 17 Feb '08

17 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Feb 14, 2008, at 4:06 PM, William Siegrist wrote: CF-476.10 has been posted. http://www.opensource.apple.com/darwinsource/10.5.2/CF-476.10/ Is CFNetwork no longer being released? _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… -Bill This email sent to site_archiver(a)lists.apple.com

1 0

Re: getcontext() -> Bus error?
by Amanda Walker 16 Feb '08

16 Feb '08

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com --Amanda _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… On Feb 16, 2008, at 4:23 PM, Edd Dawson wrote: I can see how what I did would be problematic for certain implementations. On the other hand I don't believe I'm doing anything funny as far as the POSIX standard is concerned. Well, as of the 2004 edition of 1003.1, strictly conforming POSIX applications can't use these functions at all, and are supposed to use the POSIX threads functions instead. Last I checked (late 2006), they were slated for removal altogether in the next rev of the standard (which should be happening right around now, if I recall correctly). As to why it works in Linux but not Darwin, my first guess would be "luck" :-). Using these functions has always been a fairly delicate matter. This email sent to site_archiver(a)lists.apple.com

1 0