From gentryx at gmx.de Thu May 8 19:04:55 2008 From: gentryx at gmx.de (Andreas =?iso-8859-1?Q?Sch=E4fer?=) Date: Fri, 9 May 2008 02:04:55 +0200 Subject: [Users] PUGH Questions Message-ID: <20080509000455.GA25396@wintermute> Hi, I'm currently playing around with Cactus' Wave Toy demo to see what Cactus can do (and how fast). AFAICS Wave Toy is parallelized by the PUGH thorn. I'm a bit unsure about some details, so it would be nice if you could answer some of my questions: * In a cluster with heterogeneous machines, can PUGH perform automatic static load balancing to give faster machines more grid points? * Can PUGH perform dynamic load balancing? * When I use numbers of nodes that are not powers of 2, the domain decomposition sometimes degrades. For instance for 31 nodes a cube would be decomposed into 31 thin slices, leading to a rather bad surface to volume ratio. Is there any way to avoid such situations? (using only powers of two as numbers of nodes is not an option ;-) ) * Are alternatives to PUGH available that perform better in regards to the questions above? Thanks -Andreas -- ============================================ Andreas Sch?fer Cluster and Metacomputing Working Group Friedrich-Schiller-Universit?t Jena, Germany PGP/GPG key via keyserver I'm a bright... http://www.the-brights.net ============================================ (\___/) (+'.'+) (")_(") This is Bunny. Copy and paste Bunny into your signature to help him gain world domination! -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: not available Url : http://www.cactuscode.org/pipermail/users/attachments/20080509/9f94854f/attachment.bin From frank.loeffler at aei.mpg.de Thu May 8 19:44:52 2008 From: frank.loeffler at aei.mpg.de (Frank Loeffler) Date: Thu, 8 May 2008 19:44:52 -0500 Subject: [Users] PUGH Questions In-Reply-To: <20080509000455.GA25396@wintermute> References: <20080509000455.GA25396@wintermute> Message-ID: <20080509004451.GZ12358@numrel07.cct.lsu.edu> Hi, On Fri, May 09, 2008 at 02:04:55AM +0200, Andreas Sch?fer wrote: > * In a cluster with heterogeneous machines, can PUGH perform automatic > static load balancing to give faster machines more grid points? No. You could distribute the data by hand probably, but AFAIK PUGH cannot do that automatically. PUGH does not know about the machine specifics like available memory. > * Can PUGH perform dynamic load balancing? No. > * When I use numbers of nodes that are not powers of 2, the domain > decomposition sometimes degrades. For instance for 31 nodes a cube > would be decomposed into 31 thin slices, leading to a rather bad > surface to volume ratio. Is there any way to avoid such situations? > (using only powers of two as numbers of nodes is not an option ;-) ) PUGH will (by default) try to devide the domain evenly to all processors with the restriction that you cannot have a processor domain edge like this: domain 1 ----------+---------- domain 2 | domain 3 | Because 31 is a prime number, PUGH has no other option as to devide it like you described. If you choose 30, it could devide e.g. into 2*3*5. > * Are alternatives to PUGH available that perform better in regards to > the questions above? I do not know one from the top of my head, but that does not mean that there couldn't exist one. However, you might want to have a look into Carpet (www.carpetcode.org), even if you do not intend to use mesh refinement. I am not sure if Carpet could e.g. handle the last question better. Erik Schnetter would be the right person to ask. Frank From schnetter at cct.lsu.edu Thu May 8 22:35:43 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Thu, 8 May 2008 22:35:43 -0500 Subject: [Users] PUGH Questions In-Reply-To: <20080509004451.GZ12358@numrel07.cct.lsu.edu> References: <20080509000455.GA25396@wintermute> <20080509004451.GZ12358@numrel07.cct.lsu.edu> Message-ID: On May 8, 2008, at 19:44:52, Frank Loeffler wrote: > Hi, > > On Fri, May 09, 2008 at 02:04:55AM +0200, Andreas Sch?fer wrote: >> * In a cluster with heterogeneous machines, can PUGH perform >> automatic >> static load balancing to give faster machines more grid points? > > No. You could distribute the data by hand probably, but AFAIK PUGH > cannot do that automatically. PUGH does not know about the machine > specifics like available memory. > >> * Can PUGH perform dynamic load balancing? > > No. > >> * When I use numbers of nodes that are not powers of 2, the domain >> decomposition sometimes degrades. For instance for 31 nodes a cube >> would be decomposed into 31 thin slices, leading to a rather bad >> surface to volume ratio. Is there any way to avoid such situations? >> (using only powers of two as numbers of nodes is not an option ;-) ) > > PUGH will (by default) try to devide the domain evenly to all > processors with the restriction that you cannot have a processor > domain > edge like this: > domain 1 > ----------+---------- > domain 2 | domain 3 > | > Because 31 is a prime number, PUGH has no other option as to devide it > like you described. If you choose 30, it could devide e.g. into 2*3*5. > >> * Are alternatives to PUGH available that perform better in regards >> to >> the questions above? > > I do not know one from the top of my head, but that does not mean that > there couldn't exist one. However, you might want to have a look into > Carpet (www.carpetcode.org), even if you do not intend to use mesh > refinement. I am not sure if Carpet could e.g. handle the last > question > better. Erik Schnetter would be the right person to ask. Yes, Carpet can handle the case where the number of processors is not a power of 2. What kindsw of clusters are you considering? Is the compute hardware different, or do you refer e.g. to the network topology? Would your problems be compute bound, memory bound, communication bound, or I/O bound? Do you expect the hardware performance to change at run time, or be approximately constant over a run? Or do you rather expect the computational load to be different in different parts of the simulation domain? Carpet assigns to each process a part of the problem that corresponds to the number of threads running in this process, assuming that each thread runs at the same speed. Handling heterogeneous machines statically would be rather easy to implement. For example, the thorn LSUThorns/DGEMM runs a short CPU benchmark at startup, which could be used to determine the compute power of a process. Changing the load distribution (even dynamically) is not difficult; the difficult part is determining _how_ to change it, since different parts of the evolution algorithm may run with differing speeds on the different processors. I do not typically use heterogeneous machines; if you are interested, we could offer an API that lets people dynamically set the relative performance of each MPI process, and Carpet would the change the load distribution according to these performances. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080508/ba608c39/attachment.bin From gentryx at gmx.de Fri May 9 01:43:40 2008 From: gentryx at gmx.de (Andreas =?iso-8859-1?Q?Sch=E4fer?=) Date: Fri, 9 May 2008 08:43:40 +0200 Subject: [Users] PUGH Questions In-Reply-To: References: <20080509000455.GA25396@wintermute> <20080509004451.GZ12358@numrel07.cct.lsu.edu> Message-ID: <20080509064340.GA31862@wintermute> Hi, Thanks for your quick replies! On 22:35 Thu 08 May , Erik Schnetter wrote: > Yes, Carpet can handle the case where the number of processors is > not a power of 2. What kind of domain decomposition are you using? Space filling curves, ParMetis, recursive coordinate bisection, something completely different? I had a look at the documentation available at http://www.carpetcode.org/ but was too blind to find any information on it. ^^ > What kindsw of clusters are you considering? Is the compute hardware > different, or do you refer e.g. to the network topology? Would your > problems be compute bound, memory bound, communication bound, or I/O bound? > Do you expect the hardware performance to change at run time, or be > approximately constant over a run? Or do you rather expect the > computational load to be different in different parts of the simulation > domain? Short answer: Yes, all of them. ^^ One of my goals is to find out how good Cactus does perform in a multi-cluster setup, so you'd have both, heterogeneous machines and networks. The reason behind this is that we've written a parallel library for time discrete simulations on structured grids is geared towards grids and multi-clusters. Cactus and this library have some crucial functionality in common and since Cactus is a corner stone when it comes to computer based simulations, I thought comparing the lib to Cactus would give readers a rough idea how to judge it. Applications could be anything, from communication bound (always ugly on heterogeneous networks) to compute bound (hence the load balancing question), but with IO and memory bound apps playing a minor role (ATM). For instance one of our apps, a simulation for cooling molten metal alloys, exhibits high load imbalances depending on the grid points' states: we see computational hot spots on the face border between molten and liquid metal and significantly less complex updates in all liquid and all solid areas. Since the face border is constantly moving as the simulation progresses and the metal solidifies, hot spots move through the grid, new hot spots arise and old ones disappear. > Changing the load distribution (even dynamically) is not difficult; > the difficult part is determining _how_ to change it, since > different parts of the evolution algorithm may run with differing > speeds on the different processors. One method we use is to hook in at the ghost zone communication. Assuming the following pseudo code... for each timestep update inner ghost zones send/recv ghost zones asynchronously update kernel wait for ghost zone communication end ...you could divide the time taken for the updates by the total time needed for one time step to get an estimate for a nodes utilization. If it is high on one node and low on its neighbors, they should share some of their load. Since it is not clear in advance which new grid point distribution leads to which new load distribution, a diffusion based redistribution has proven to be quite efficient for us. > I do not typically use heterogeneous machines; if you are interested, we > could offer an API that lets people dynamically set the relative > performance of each MPI process, and Carpet would the change the load > distribution according to these performances. The way I understand it, this means that a) it could be done easily, but hasn't yet been a use case and b) it is not yet implemented. Is that right? Thanks -Andreas -- ============================================ Andreas Sch?fer Cluster and Metacomputing Working Group Friedrich-Schiller-Universit?t Jena, Germany PGP/GPG key via keyserver I'm a bright... http://www.the-brights.net ============================================ (\___/) (+'.'+) (")_(") This is Bunny. Copy and paste Bunny into your signature to help him gain world domination! -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: not available Url : http://www.cactuscode.org/pipermail/users/attachments/20080509/4c5c494b/attachment.bin From schnetter at cct.lsu.edu Fri May 9 15:48:36 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Fri, 9 May 2008 15:48:36 -0500 Subject: [Users] PUGH Questions In-Reply-To: <20080509064340.GA31862@wintermute> References: <20080509000455.GA25396@wintermute> <20080509004451.GZ12358@numrel07.cct.lsu.edu> <20080509064340.GA31862@wintermute> Message-ID: <2F735EF4-620F-47DE-BDEE-056F92C3DD0D@cct.lsu.edu> On May 9, 2008, at 01:43:40, Andreas Sch?fer wrote: > On 22:35 Thu 08 May , Erik Schnetter wrote: >> Yes, Carpet can handle the case where the number of processors is >> not a power of 2. > > What kind of domain decomposition are you using? Space filling curves, > ParMetis, recursive coordinate bisection, something completely > different? I had a look at the documentation available at > http://www.carpetcode.org/ but was too blind to find any information > on it. ^^ PUGH decomposes the number of processors into three integer factors, taking the number of grid points in each dimension into account. This leads to the restrictions which Frank L?ffler described earlier. A specialised version of PUGH was indeed used some time ago in a heterogeneous, multi-machine setup. See e.g. the section "Gordon Bell Prize for Supercomputing (SC2001)" on . Sadly, these updates to PUGH are not included in the main Cactus distribution; they also hid some of the network latency. Carpet decomposes sets of domains onto sets of processors. Carpet decomposes the domain into cuboid regions, hence space-filling curves would not work. It uses a hierarchical method, similar to coordinate bisection. Instead of bisecting, Carpet n-sects, where n is chosen depending on the number of processors. Each dimension is n-sected only once. While bisection leads to a binary tree, n-section leads to a wider and more shallow tree. When the load per grid point is approximately equal, then n-section is a simpler algorithm. If the load per grid point can vary greatly, then n-section becomes too inflexible, and bisection would become appropriate. > One of my goals is to find out how > good Cactus does perform in a multi-cluster setup, so you'd have both, > heterogeneous machines and networks. The reason behind this is that > we've written a parallel library for time discrete simulations on > structured grids is geared towards grids and multi-clusters. Cactus > and this library have some crucial functionality in common and since > Cactus is a corner stone when it comes to computer based simulations, Thank you for the flowers. Do you have a pointer to web pages describing your library or your project? > I thought comparing the lib to Cactus would give readers a rough idea > how to judge it. We are looking into multi-cluster simulations, but have recently only considered heterogeneous sets of clusters. >> Changing the load distribution (even dynamically) is not difficult; >> the difficult part is determining _how_ to change it, since >> different parts of the evolution algorithm may run with differing >> speeds on the different processors. > > One method we use is to hook in at the ghost zone > communication. Assuming the following pseudo code... > > for each timestep > update inner ghost zones > send/recv ghost zones asynchronously > update kernel > wait for ghost zone communication > end > > ...you could divide the time taken for the updates by the total time > needed for one time step to get an estimate for a nodes > utilization. If it is high on one node and low on its neighbors, they > should share some of their load. Since it is not clear in advance > which new grid point distribution leads to which new load > distribution, a diffusion based redistribution has proven to be quite > efficient for us. Yes, that would be a good approach. I was referring to the fact that one may do multi-physics simulations (we often couple spacetime and hydrodynamics evolution, and also run some analysis tools at run time), and one has then to find an overall balance, which is not necessarily a good balance for each individual system. >> I do not typically use heterogeneous machines; if you are >> interested, we >> could offer an API that lets people dynamically set the relative >> performance of each MPI process, and Carpet would the change the load >> distribution according to these performances. > > The way I understand it, this means that a) it could be done easily, > but hasn't yet been a use case and b) it is not yet implemented. Is > that right? This is both correct. It is also (c) I find this topic interesting, especially since the workload in different parts of the domain is probably going to become less equally distributed in the future for our applications in relativistic astrophysics. You are probably aware of Ulrich Sperhake at FSU who is using Cactus and Carpet to simulate astrophysical systems, in particular binary black holes; see , or for a recent publication containing pointers describing their LEAN code. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080509/b13cde87/attachment-0001.bin From gentryx at gmx.de Sat May 10 11:38:21 2008 From: gentryx at gmx.de (Andreas =?iso-8859-1?Q?Sch=E4fer?=) Date: Sat, 10 May 2008 18:38:21 +0200 Subject: [Users] PUGH Questions In-Reply-To: <2F735EF4-620F-47DE-BDEE-056F92C3DD0D@cct.lsu.edu> References: <20080509000455.GA25396@wintermute> <20080509004451.GZ12358@numrel07.cct.lsu.edu> <20080509064340.GA31862@wintermute> <2F735EF4-620F-47DE-BDEE-056F92C3DD0D@cct.lsu.edu> Message-ID: <20080510163821.GA20427@neuromancer> Heya, On 15:48 Fri 09 May , Erik Schnetter wrote: > A specialised version of PUGH was indeed used some time ago in a > heterogeneous, multi-machine setup. See e.g. the section "Gordon Bell > Prize for Supercomputing (SC2001)" on > . Sadly, these updates to PUGH are > not included in the main Cactus distribution; they also hid some of the > network latency. Very interesting! Somehow I must have missed this paper. I'm a bit curious about the proposed domain decomposition. In Fig. 2 the processor topology seems to form a lengthy cuboid of the dimensions 5x12x25, with subsystems being split at the Z axis. Isn't this prone to degenerated surfaces? If the simulation grid was equally spaced in all dimensions, then the smaller systems of 5x12x2 CPUs would have the same surface as the larger ones, leading to an increased "outside communication" to load ratio. We're currently working on structured grids only (just like PUGH?), and the problems above led us to consider different domain decomposition schemes. The Z curve seems to be quite good at volume to surface ratios, and graceful in terms of load balancing (see below for the problems we had with bisection), but we're digging into this just now and don't have final results yet. > Carpet n-sects, where n is chosen depending on the > number of processors. Each dimension is n-sected only once. While > bisection leads to a binary tree, n-section leads to a wider and more > shallow tree. How exactly do you choose n? Does n have to be a divisor of the number of processors? If you n-sec each dimension only once, couldn't this also lead to a degenerated surface to volume ratio? I'm not sure how "each dimension is n-sected only once" affects this, but using bisections we've observed a nasty effect during load balancing that we call "flip over": Assume we had four machines and would like to do a weighted bisection with the weights [0.51, 0.25, 0.12, 0.12]. Then we would end up with a bisection equivalent to the following: (assuming that we start with a 1x1 square and always split along the longest axis, numbers correspond to the machine's position in the weight vector) -------------- | | | | | | 2|3 | | | | | | 0 ------- | | | | | 1 | | | | -------------- (please forgive me my ASCII art) Now, if we then decided to change the weights slightly to [0.49, 0.25, 0.13, 0.13] (e.g. because we observed a load imbalance), we'd end up with the following bisection... -------------- | | 2 | | ------- | | 3 | | 0 ------- | | | | | 1 | | | | -------------- ...which requires the nodes 2 and 3 to communicate ~50% of their share of the simulation grid. That's quite expensive for such a small change in the weights vector. Of course it could be argued that the chosen example with its strange weights is rather arcane, but it's easy to see that a) it could be quickly generalized to one with more nodes and an almost even weight distribution, b) it could hit not only two single nodes, but even whole groups of nodes where some nodes would have to exchange 100% of their grid region, and c) there are similar cases for n-sections (when dimensions are divided multiple times). Actually, these flip overs are most common for setups with many homogeneous nodes (a power of two, preferably). Could the example above happen when balancing the load with Carpet? If not, how does the n-section work exactly? > Do you have a pointer to web pages describing your library or your project? Not yet, I'm currently writing a paper for the ParSim (at the EuroPVM/MPI) and we'll put up the code with some docs later on, but the paper comes first. ^^ > Yes, that would be a good approach. I was referring to the fact that one > may do multi-physics simulations (we often couple spacetime and > hydrodynamics evolution, and also run some analysis tools at run time), and > one has then to find an overall balance, which is not necessarily a good > balance for each individual system. Ok, I get it. > You are probably aware of Ulrich Sperhake at FSU who is using Cactus and > Carpet to simulate astrophysical systems, in particular binary black holes; Aware yes, although we're not currently working with them. Maybe this will change in the course of Jena's campus grid. We'll see. (-8 Cheers -Andreas -- ============================================ Andreas Sch?fer Cluster and Metacomputing Working Group Friedrich-Schiller-Universit?t Jena, Germany PGP/GPG key via keyserver I'm a bright... http://www.the-brights.net ============================================ (\___/) (+'.'+) (")_(") This is Bunny. Copy and paste Bunny into your signature to help him gain world domination! -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: not available Url : http://www.cactuscode.org/pipermail/users/attachments/20080510/215be427/attachment.bin From babiuc at marshall.edu Wed May 14 08:59:30 2008 From: babiuc at marshall.edu (Babiuc-Hamilton, Maria) Date: Wed, 14 May 2008 09:59:30 -0400 Subject: [Users] problems with the Absoft compiler on Mac Message-ID: <31D46DC505870A4C980A8F38BB2DBD0E4E52F7209A@MUXC09.marshall.edu> Hello, I am trying to compile and run the wavetoy on a Mac laptop, with OSX version 10.4.11 and Intel Core 2 Duo processor, and Absoft10.1 fortran compiler, but I run into troubles. The problem is: ________________________________________________________________________ make[3]: *** No rule to make target `ScalarBoundary.c.o', needed by `make.checked'. Stop. make[2]: *** [make.checked] Error 2 I am using the config file that I found on the Cactus web page: CC=gcc CXX=g++ F77=/Applications/Absoft10.1/bin/f77 F90=/Applications/Absoft10.1/bin/f95 -Dminimise_outstanding_communications=moc_toolong -Duse_collective_communication_buffers=uccb_toolong Nevertheless, at the configuration, I still have the complain: Unknown MacOS X f90 compiler. Please add appropriate information to /Users/mariababiuc/Cactus/lib/make/known-architectures/darwin and send the updated file to CactusMaint We will try anyway ... I am attaching the configuration and compile log files. Please help, Maria -------------- next part -------------- A non-text attachment was scrubbed... Name: mac_compile Type: application/octet-stream Size: 5932 bytes Desc: mac_compile Url : http://www.cactuscode.org/pipermail/users/attachments/20080514/6194de2e/attachment.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: mac_config Type: application/octet-stream Size: 4823 bytes Desc: mac_config Url : http://www.cactuscode.org/pipermail/users/attachments/20080514/6194de2e/attachment-0001.obj From schnetter at cct.lsu.edu Wed May 14 12:11:14 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Wed, 14 May 2008 12:11:14 -0500 Subject: [Users] problems with the Absoft compiler on Mac In-Reply-To: <31D46DC505870A4C980A8F38BB2DBD0E4E52F7209A@MUXC09.marshall.edu> References: <31D46DC505870A4C980A8F38BB2DBD0E4E52F7209A@MUXC09.marshall.edu> Message-ID: On May 14, 2008, at 08:59:30, Babiuc-Hamilton, Maria wrote: > Hello, > > I am trying to compile and run the wavetoy on a Mac laptop, with OSX > version 10.4.11 and Intel Core 2 Duo processor, and Absoft10.1 > fortran compiler, but I run into troubles. > The problem is: > ________________________________________________________________________ > make[3]: *** No rule to make target `ScalarBoundary.c.o', needed by > `make.checked'. Stop. > make[2]: *** [make.checked] Error 2 > > I am using the config file that I found on the Cactus web page: > > CC=gcc > CXX=g++ > F77=/Applications/Absoft10.1/bin/f77 > F90=/Applications/Absoft10.1/bin/f95 > -Dminimise_outstanding_communications=moc_toolong > -Duse_collective_communication_buffers=uccb_toolong Maria, you are receiving errors: Could not parse configuration line 5... '-Dminimise_outstanding_communications=moc_toolong' Could not parse configuration line 6... '-Duse_collective_communication_buffers=uccb_toolong' I believe that the two -D options should be on the same line as the F90 setting. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080514/11e3d9ee/attachment.bin From babiuc at marshall.edu Wed May 14 14:26:29 2008 From: babiuc at marshall.edu (Babiuc-Hamilton, Maria) Date: Wed, 14 May 2008 15:26:29 -0400 Subject: [Users] problems with the Absoft compiler on Mac In-Reply-To: References: <31D46DC505870A4C980A8F38BB2DBD0E4E52F7209A@MUXC09.marshall.edu>, Message-ID: <31D46DC505870A4C980A8F38BB2DBD0E4E52F720A5@MUXC09.marshall.edu> Hello Erik, Yes, you are right. That was silly from my part. I corrected it. However, the compilation is still stopped by the error: Checking status of thorn Boundary ________________________________________________________________________ if [ ! -d ./ ] ; then mkdir -p ./ ; fi cd ./ ; make CCTK_TARGET=make.checked TOP=/Users/mariababiuc/Cactus/configs/wavetoy CONFIG=/Users/mariababiuc/Cactus/configs/wavetoy/config-data SRCDIR=/Users/mariababiuc/Cactus/arrangements/CactusBase/Boundary/src THORN=Boundary USESTHORNS="" -f /Users/mariababiuc/Cactus/lib/make/make.subdir make[3]: *** No rule to make target `ScalarBoundary.c.o', needed by `make.checked'. Stop. make[2]: *** [make.checked] Error 2 make[1]: *** [/Users/mariababiuc/Cactus/configs/wavetoy/lib/libthorn_Boundary.a] Error 2 There are some error in the config.log file, which is attached. The problem seems to be with gcc. Would you please take a look? Thanks, Maria ________________________________________ From: users-bounces at cactuscode.org [users-bounces at cactuscode.org] On Behalf Of Erik Schnetter [schnetter at cct.lsu.edu] Sent: Wednesday, May 14, 2008 1:11 PM To: Cactus users mailing list Subject: Re: [Users] problems with the Absoft compiler on Mac On May 14, 2008, at 08:59:30, Babiuc-Hamilton, Maria wrote: > Hello, > > I am trying to compile and run the wavetoy on a Mac laptop, with OSX > version 10.4.11 and Intel Core 2 Duo processor, and Absoft10.1 > fortran compiler, but I run into troubles. > The problem is: > ________________________________________________________________________ > make[3]: *** No rule to make target `ScalarBoundary.c.o', needed by > `make.checked'. Stop. > make[2]: *** [make.checked] Error 2 > > I am using the config file that I found on the Cactus web page: > > CC=gcc > CXX=g++ > F77=/Applications/Absoft10.1/bin/f77 > F90=/Applications/Absoft10.1/bin/f95 > -Dminimise_outstanding_communications=moc_toolong > -Duse_collective_communication_buffers=uccb_toolong Maria, you are receiving errors: Could not parse configuration line 5... '-Dminimise_outstanding_communications=moc_toolong' Could not parse configuration line 6... '-Duse_collective_communication_buffers=uccb_toolong' I believe that the two -D options should be on the same line as the F90 setting. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: config.log Type: application/octet-stream Size: 21923 bytes Desc: config.log Url : http://www.cactuscode.org/pipermail/users/attachments/20080514/46814911/attachment-0001.obj From schnetter at cct.lsu.edu Wed May 14 15:07:42 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Wed, 14 May 2008 15:07:42 -0500 Subject: [Users] problems with the Absoft compiler on Mac In-Reply-To: <31D46DC505870A4C980A8F38BB2DBD0E4E52F720A5@MUXC09.marshall.edu> References: <31D46DC505870A4C980A8F38BB2DBD0E4E52F7209A@MUXC09.marshall.edu>, <31D46DC505870A4C980A8F38BB2DBD0E4E52F720A5@MUXC09.marshall.edu> Message-ID: Maria, your configure output contains the steps that configure runs, but not the results that configure finds. Can you send the screen output instead? -erik On May 14, 2008, at 14:26:29, Babiuc-Hamilton, Maria wrote: > Hello Erik, > > Yes, you are right. That was silly from my part. I corrected it. > However, the compilation is still stopped by the error: > > Checking status of thorn Boundary > ________________________________________________________________________ > if [ ! -d ./ ] ; then mkdir -p ./ ; fi > cd ./ ; make CCTK_TARGET=make.checked TOP=/Users/mariababiuc/Cactus/ > configs/wavetoy CONFIG=/Users/mariababiuc/Cactus/configs/wavetoy/ > config-data SRCDIR=/Users/mariababiuc/Cactus/arrangements/CactusBase/ > Boundary/src THORN=Boundary USESTHORNS="" -f /Users/mariababiuc/ > Cactus/lib/make/make.subdir > make[3]: *** No rule to make target `ScalarBoundary.c.o', needed by > `make.checked'. Stop. > make[2]: *** [make.checked] Error 2 > make[1]: *** [/Users/mariababiuc/Cactus/configs/wavetoy/lib/ > libthorn_Boundary.a] Error 2 > > There are some error in the config.log file, which is attached. The > problem seems to be with gcc. Would you please take a look? > > Thanks, > Maria > ________________________________________ > From: users-bounces at cactuscode.org [users-bounces at cactuscode.org] On > Behalf Of Erik Schnetter [schnetter at cct.lsu.edu] > Sent: Wednesday, May 14, 2008 1:11 PM > To: Cactus users mailing list > Subject: Re: [Users] problems with the Absoft compiler on Mac > > On May 14, 2008, at 08:59:30, Babiuc-Hamilton, Maria wrote: > >> Hello, >> >> I am trying to compile and run the wavetoy on a Mac laptop, with OSX >> version 10.4.11 and Intel Core 2 Duo processor, and Absoft10.1 >> fortran compiler, but I run into troubles. >> The problem is: >> ________________________________________________________________________ >> make[3]: *** No rule to make target `ScalarBoundary.c.o', needed by >> `make.checked'. Stop. >> make[2]: *** [make.checked] Error 2 >> >> I am using the config file that I found on the Cactus web page: >> >> CC=gcc >> CXX=g++ >> F77=/Applications/Absoft10.1/bin/f77 >> F90=/Applications/Absoft10.1/bin/f95 >> -Dminimise_outstanding_communications=moc_toolong >> -Duse_collective_communication_buffers=uccb_toolong > > Maria, > > you are receiving errors: > > Could not parse configuration line 5... > '-Dminimise_outstanding_communications=moc_toolong' > Could not parse configuration line 6... > '-Duse_collective_communication_buffers=uccb_toolong' > > I believe that the two -D options should be on the same line as the > F90 setting. > > -erik > > -- > Erik Schnetter http://www.cct.lsu.edu/~eschnett/ > > My email is as private as my paper mail. I therefore support > encrypting > and signing email messages. Get my PGP key from www.keyserver.net. > > > > _______________________________________________ > Users mailing list > Users at cactuscode.org > http://www.cactuscode.org/mailman/listinfo/users -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080514/69527baf/attachment.bin From babiuc at marshall.edu Wed May 14 15:27:21 2008 From: babiuc at marshall.edu (Babiuc-Hamilton, Maria) Date: Wed, 14 May 2008 16:27:21 -0400 Subject: [Users] problems with the Absoft compiler on Mac In-Reply-To: References: <31D46DC505870A4C980A8F38BB2DBD0E4E52F7209A@MUXC09.marshall.edu>, <31D46DC505870A4C980A8F38BB2DBD0E4E52F720A5@MUXC09.marshall.edu>, Message-ID: <31D46DC505870A4C980A8F38BB2DBD0E4E52F720A7@MUXC09.marshall.edu> Hi, Yes. Here is the screen output: ________________________________________________________________________ Cactus - version: 4.0.b16 Reconfiguring wavetoy. Adding configuration options from '/Users/mariababiuc/Cactus/configs/wavetoy/config-info'... Setting CC to 'gcc' Setting CXX to 'g++' Setting F77 to '/Applications/Absoft10.1/bin/f77' Setting F90 to '/Applications/Absoft10.1/bin/f95' Setting FPPFLAGS to '-Dminimise_outstanding_communications=moc_toolong -Duse_collective_communication_buffers=uccb_toolong' End of options from '/Users/mariababiuc/Cactus/configs/wavetoy/config-info'. Adding configuration options from user defaults... End of options from user defaults. creating cache ./config.cache checking host system type... i386-apple-darwin8.11.1 checking whether make sets ${MAKE}... yes checking whether the C compiler (gcc ) works... yes checking whether the C compiler (gcc ) is a cross-compiler... no checking whether we are using GNU C... yes checking whether the C++ compiler (g++ ) works... yes checking whether the C++ compiler (g++ ) is a cross-compiler... no checking whether we are using GNU C++... yes checking for ranlib... ranlib checking for cpp... /usr/bin/cpp checking for cpp... /usr/bin/cpp checking for perl... perl checking for ar... ar checking for mkdir... mkdir checking for bash... /bin/sh Unknown MacOS X f90 compiler. Please add appropriate information to /Users/mariababiuc/Cactus/lib/make/known-architectures/darwin and send the updated file to CactusMaint We will try anyway ... checking how to run the C preprocessor... /usr/bin/cpp checking for ANSI C header files... yes checking whether byte ordering is bigendian... no checking size of long long... 8 checking size of long int... 4 checking size of int... 4 checking size of short int... 2 checking size of long double... 16 checking size of double... 8 checking size of float... 4 checking size of char *... 4 checking for the null device... /dev/null checking for gethostbyname... yes checking whether mode_t is defined... yes checking for availability of gettimeofday timing... yes checking if gettimeofday needs timezone... yes checking for availability of getrusage timing... yes checking for availability of _ftime timing... no checking for time.h... yes checking for sys/time.h... yes checking for sys/types.h... yes checking for unistd.h... yes checking for string.h... yes checking for assert.h... yes checking for tgmath.h... yes checking for sys/stat.h... yes checking for getopt.h... yes checking for dirent.h... yes checking for c_asm.h... no checking for intrinsics.h... no checking for mach/mach_time.h... yes checking for regex.h... yes checking for sys/filio.h... yes checking for sys/ioctl.h... yes checking for sys/socket.h... yes checking for netinet/in.h... yes checking for netdb.h... yes checking for arpa/inet.h... yes checking for winsock2.h... no checking for crypt.h... no checking for signal.h... yes checking for malloc.h... no checking whether time.h and sys/time.h may both be included... yes checking whether socklen_t is defined... yes checking whether SOCKET is defined... no checking for hrtime_t... no checking for gethrtime... no checking for read_real_time... no checking for time_base_to_time... no checking for clock_gettime... no checking for mach_absolute_time... yes checking for _rtc intrinsic... no checking for mallinfo... no checking for mallopt... no checking whether M_MMAP_THRESHOLD is defined... no checking if __int64 is defined... no checking for vector... yes checking for vector.h... yes checking for getopt_long_only... yes checking for working const... yes checking for C inline... inline checking for C restrict... __restrict__ checking for C++ restrict... __restrict__ checking for C bool... no checking for CXX bool... yes checking for Fortran REAL*4... yes checking for Fortran REAL*8... yes checking for Fortran REAL*16... yes checking for Fortran COMPLEX*8... yes checking for Fortran COMPLEX*16... yes checking for Fortran COMPLEX*32... yes checking for crypt... yes checking for finite... yes checking for isnan... yes checking for isinf... yes checking for mkstemp... yes checking for va_copy... yes creating cctk_Archdefs.h creating make.arch.defn creating cctk_Extradefs.h creating make.extra.defn cctk_Archdefs.h is unchanged make.arch.defn is unchanged cctk_Extradefs.h is unchanged make.extra.defn is unchanged updating cache ./config.cache creating ./config.status creating make.config.defn creating make.config.deps creating make.config.rules creating cctk_Config.h cctk_Config.h is unchanged Determining number of fortran underscores... Compiling test file with /Applications/Absoft10.1/bin/f77 ... Lowercase - One trailing underscore Compiling test file with /Applications/Absoft10.1/bin/f95 ... Lowercase - No trailing underscore ________________________________________________________________________ Use make wavetoy to build the configuration. ________________________________________ From: users-bounces at cactuscode.org [users-bounces at cactuscode.org] On Behalf Of Erik Schnetter [schnetter at cct.lsu.edu] Sent: Wednesday, May 14, 2008 4:07 PM To: Cactus users mailing list Subject: Re: [Users] problems with the Absoft compiler on Mac Maria, your configure output contains the steps that configure runs, but not the results that configure finds. Can you send the screen output instead? -erik On May 14, 2008, at 14:26:29, Babiuc-Hamilton, Maria wrote: > Hello Erik, > > Yes, you are right. That was silly from my part. I corrected it. > However, the compilation is still stopped by the error: > > Checking status of thorn Boundary > ________________________________________________________________________ > if [ ! -d ./ ] ; then mkdir -p ./ ; fi > cd ./ ; make CCTK_TARGET=make.checked TOP=/Users/mariababiuc/Cactus/ > configs/wavetoy CONFIG=/Users/mariababiuc/Cactus/configs/wavetoy/ > config-data SRCDIR=/Users/mariababiuc/Cactus/arrangements/CactusBase/ > Boundary/src THORN=Boundary USESTHORNS="" -f /Users/mariababiuc/ > Cactus/lib/make/make.subdir > make[3]: *** No rule to make target `ScalarBoundary.c.o', needed by > `make.checked'. Stop. > make[2]: *** [make.checked] Error 2 > make[1]: *** [/Users/mariababiuc/Cactus/configs/wavetoy/lib/ > libthorn_Boundary.a] Error 2 > > There are some error in the config.log file, which is attached. The > problem seems to be with gcc. Would you please take a look? > > Thanks, > Maria > ________________________________________ > From: users-bounces at cactuscode.org [users-bounces at cactuscode.org] On > Behalf Of Erik Schnetter [schnetter at cct.lsu.edu] > Sent: Wednesday, May 14, 2008 1:11 PM > To: Cactus users mailing list > Subject: Re: [Users] problems with the Absoft compiler on Mac > > On May 14, 2008, at 08:59:30, Babiuc-Hamilton, Maria wrote: > >> Hello, >> >> I am trying to compile and run the wavetoy on a Mac laptop, with OSX >> version 10.4.11 and Intel Core 2 Duo processor, and Absoft10.1 >> fortran compiler, but I run into troubles. >> The problem is: >> ________________________________________________________________________ >> make[3]: *** No rule to make target `ScalarBoundary.c.o', needed by >> `make.checked'. Stop. >> make[2]: *** [make.checked] Error 2 >> >> I am using the config file that I found on the Cactus web page: >> >> CC=gcc >> CXX=g++ >> F77=/Applications/Absoft10.1/bin/f77 >> F90=/Applications/Absoft10.1/bin/f95 >> -Dminimise_outstanding_communications=moc_toolong >> -Duse_collective_communication_buffers=uccb_toolong > > Maria, > > you are receiving errors: > > Could not parse configuration line 5... > '-Dminimise_outstanding_communications=moc_toolong' > Could not parse configuration line 6... > '-Duse_collective_communication_buffers=uccb_toolong' > > I believe that the two -D options should be on the same line as the > F90 setting. > > -erik > > -- > Erik Schnetter http://www.cct.lsu.edu/~eschnett/ > > My email is as private as my paper mail. I therefore support > encrypting > and signing email messages. Get my PGP key from www.keyserver.net. > > > > _______________________________________________ > Users mailing list > Users at cactuscode.org > http://www.cactuscode.org/mailman/listinfo/users -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. From babiuc at marshall.edu Wed May 14 16:12:56 2008 From: babiuc at marshall.edu (Babiuc-Hamilton, Maria) Date: Wed, 14 May 2008 17:12:56 -0400 Subject: [Users] Problems with the Intel compiler on Mac Message-ID: <31D46DC505870A4C980A8F38BB2DBD0E4E52F720A8@MUXC09.marshall.edu> Hello, The same problem [No rule to make target `ScalarBoundary.c.o', needed by `make.checked'] appears at the compilation of the wavetoy, when I am using the Intel compilers for mac. The config screen output is: ________________________________________________________________________ Cactus - version: 4.0.b16 Reconfiguring wavetoy. Adding configuration options from user defaults... Setting CC to 'icc' Setting CXX to 'icpc' Setting F77 to 'ifort' Setting F90 to 'ifort' Setting FPPFLAGS to '-traditional' End of options from user defaults. creating cache ./config.cache checking host system type... i386-apple-darwin8.11.1 checking whether make sets ${MAKE}... yes checking whether the C compiler (icc ) works... yes checking whether the C compiler (icc ) is a cross-compiler... no checking whether we are using GNU C... yes checking whether the C++ compiler (icpc ) works... yes checking whether the C++ compiler (icpc ) is a cross-compiler... no checking whether we are using GNU C++... yes checking for ranlib... ranlib checking for cpp... /usr/bin/cpp checking for cpp... /usr/bin/cpp checking for perl... perl checking for ar... ar checking for mkdir... mkdir checking for bash... /bin/sh checking how to run the C preprocessor... /usr/bin/cpp checking for ANSI C header files... yes checking whether byte ordering is bigendian... no checking size of long long... 8 checking size of long int... 4 checking size of int... 4 checking size of short int... 2 checking size of long double... 16 checking size of double... 8 checking size of float... 4 checking size of char *... 4 checking for the null device... /dev/null checking for gethostbyname... yes checking whether mode_t is defined... yes checking for availability of gettimeofday timing... yes checking if gettimeofday needs timezone... yes checking for availability of getrusage timing... yes checking for availability of _ftime timing... no checking for time.h... yes checking for sys/time.h... yes checking for sys/types.h... yes checking for unistd.h... yes checking for string.h... yes checking for assert.h... yes checking for tgmath.h... yes checking for sys/stat.h... yes checking for getopt.h... yes checking for dirent.h... yes checking for c_asm.h... no checking for intrinsics.h... no checking for mach/mach_time.h... yes checking for regex.h... yes checking for sys/filio.h... yes checking for sys/ioctl.h... yes checking for sys/socket.h... yes checking for netinet/in.h... yes checking for netdb.h... yes checking for arpa/inet.h... yes checking for winsock2.h... no checking for crypt.h... no checking for signal.h... yes checking for malloc.h... no checking whether time.h and sys/time.h may both be included... yes checking whether socklen_t is defined... yes checking whether SOCKET is defined... no checking for hrtime_t... no checking for gethrtime... no checking for read_real_time... no checking for time_base_to_time... no checking for clock_gettime... no checking for mach_absolute_time... yes checking for _rtc intrinsic... no checking for mallinfo... no checking for mallopt... no checking whether M_MMAP_THRESHOLD is defined... no checking if __int64 is defined... yes checking for vector... yes checking for vector.h... yes checking for getopt_long_only... yes checking for working const... yes checking for C inline... inline checking for C restrict... __restrict__ checking for C++ restrict... __restrict__ checking for C bool... no checking for CXX bool... yes checking for Fortran REAL*4... yes checking for Fortran REAL*8... yes checking for Fortran REAL*16... yes checking for Fortran COMPLEX*8... yes checking for Fortran COMPLEX*16... yes checking for Fortran COMPLEX*32... yes checking for crypt... yes checking for finite... yes checking for isnan... yes ^[[Bchecking for isinf... yes checking for mkstemp... yes checking for va_copy... yes creating cctk_Archdefs.h creating make.arch.defn creating cctk_Extradefs.h creating make.extra.defn cctk_Archdefs.h is unchanged make.arch.defn is unchanged cctk_Extradefs.h is unchanged make.extra.defn is unchanged updating cache ./config.cache creating ./config.status creating make.config.defn creating make.config.deps creating make.config.rules creating cctk_Config.h Determining number of fortran underscores... Compiling test file with ifort -align -w95 ... Lowercase - One trailing underscore Compiling test file with ifort -align -w95 ... Lowercase - One trailing underscore From babiuc at marshall.edu Fri May 16 16:19:37 2008 From: babiuc at marshall.edu (Babiuc-Hamilton, Maria) Date: Fri, 16 May 2008 17:19:37 -0400 Subject: [Users] (no subject) Message-ID: <31D46DC505870A4C980A8F38BB2DBD0E4E52F720AB@MUXC09.marshall.edu> Hello, Does anybody know how to compile cactus on a Mac OS X 10.4, with either the intel r Absoft compilers? Even if I remove all the thorn, I still get an ' no rule to make target' error: Checking status of thorn Cactus ________________________________________________________________________ make[3]: *** No rule to make target `IOMethods.c.o', needed by `make.checked'. Stop. make[2]: *** [IO/make.checked] Error 2 Cheers, Maria From yye00 at cct.lsu.edu Fri May 16 16:23:34 2008 From: yye00 at cct.lsu.edu (Yaakoub El Khamra) Date: Fri, 16 May 2008 16:23:34 -0500 Subject: [Users] (no subject) In-Reply-To: <31D46DC505870A4C980A8F38BB2DBD0E4E52F720AB@MUXC09.marshall.edu> References: <31D46DC505870A4C980A8F38BB2DBD0E4E52F720AB@MUXC09.marshall.edu> Message-ID: <1210973015.3853.5.camel@localhost.localdomain> Do a make -realclean and try again. usually that takes care of the .c.o errors. On Fri, 2008-05-16 at 17:19 -0400, Babiuc-Hamilton, Maria wrote: > Hello, > > Does anybody know how to compile cactus on a Mac OS X 10.4, with either the intel r Absoft compilers? Even if I remove all the thorn, I still get an ' no rule to make target' error: > > > Checking status of thorn Cactus > ________________________________________________________________________ > make[3]: *** No rule to make target `IOMethods.c.o', needed by `make.checked'. Stop. > make[2]: *** [IO/make.checked] Error 2 > > Cheers, > Maria > > _______________________________________________ > Users mailing list > Users at cactuscode.org > http://www.cactuscode.org/mailman/listinfo/users > From babiuc at marshall.edu Fri May 16 16:26:30 2008 From: babiuc at marshall.edu (Babiuc-Hamilton, Maria) Date: Fri, 16 May 2008 17:26:30 -0400 Subject: [Users] (no subject) In-Reply-To: <1210973015.3853.5.camel@localhost.localdomain> References: <31D46DC505870A4C980A8F38BB2DBD0E4E52F720AB@MUXC09.marshall.edu> <1210973015.3853.5.camel@localhost.localdomain> Message-ID: <31D46DC505870A4C980A8F38BB2DBD0E4E5335F5F1@MUXC09.marshall.edu> Thanks, but I'm afraid I did several times the realclean and it still does not want to work. Dp you have any other suggestions that I might try? Thanks, Maria -----Original Message----- From: users-bounces at cactuscode.org [mailto:users-bounces at cactuscode.org] On Behalf Of Yaakoub El Khamra Sent: Friday, May 16, 2008 5:24 PM To: Cactus users mailing list Subject: Re: [Users] (no subject) Do a make -realclean and try again. usually that takes care of the .c.o errors. On Fri, 2008-05-16 at 17:19 -0400, Babiuc-Hamilton, Maria wrote: > Hello, > > Does anybody know how to compile cactus on a Mac OS X 10.4, with either the intel r Absoft compilers? Even if I remove all the thorn, I still get an ' no rule to make target' error: > > > Checking status of thorn Cactus > ________________________________________________________________________ > make[3]: *** No rule to make target `IOMethods.c.o', needed by `make.checked'. Stop. > make[2]: *** [IO/make.checked] Error 2 > > Cheers, > Maria > > _______________________________________________ > Users mailing list > Users at cactuscode.org > http://www.cactuscode.org/mailman/listinfo/users > _______________________________________________ Users mailing list Users at cactuscode.org http://www.cactuscode.org/mailman/listinfo/users From schnetter at cct.lsu.edu Fri May 16 17:13:37 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Fri, 16 May 2008 17:13:37 -0500 Subject: [Users] Problems with the Intel compiler on Mac In-Reply-To: <31D46DC505870A4C980A8F38BB2DBD0E4E52F720A8@MUXC09.marshall.edu> References: <31D46DC505870A4C980A8F38BB2DBD0E4E52F720A8@MUXC09.marshall.edu> Message-ID: Maria, this error indicates that make does not know how to compile a C programme. Is there anything else happening out of the ordinary? For example, does make abort for some reason, and you restart it, and it continues a bit further? Are there other errors or warnings which appear? Can you build with the option SILENT=no? That is, you create a new configuration, and then call make with make SILENT=no And please send us more of the screen output. So far you have only shown us part of the output; you may be accidentally hiding part of the error message, or some information that may point us to the proble. -erik On May 14, 2008, at 16:12:56, Babiuc-Hamilton, Maria wrote: > Hello, > > The same problem [No rule to make target `ScalarBoundary.c.o', > needed by `make.checked'] appears at the compilation of the wavetoy, > when I am using the Intel compilers for mac. > The config screen output is: > ________________________________________________________________________ > Cactus - version: 4.0.b16 > Reconfiguring wavetoy. > Adding configuration options from user defaults... > Setting CC to 'icc' > Setting CXX to 'icpc' > Setting F77 to 'ifort' > Setting F90 to 'ifort' > Setting FPPFLAGS to '-traditional' > End of options from user defaults. > creating cache ./config.cache > checking host system type... i386-apple-darwin8.11.1 > checking whether make sets ${MAKE}... yes > checking whether the C compiler (icc ) works... yes > checking whether the C compiler (icc ) is a cross-compiler... no > checking whether we are using GNU C... yes > checking whether the C++ compiler (icpc ) works... yes > checking whether the C++ compiler (icpc ) is a cross-compiler... no > checking whether we are using GNU C++... yes > checking for ranlib... ranlib > checking for cpp... /usr/bin/cpp > checking for cpp... /usr/bin/cpp > checking for perl... perl > checking for ar... ar > checking for mkdir... mkdir > checking for bash... /bin/sh > checking how to run the C preprocessor... /usr/bin/cpp > checking for ANSI C header files... yes > checking whether byte ordering is bigendian... no > checking size of long long... 8 > checking size of long int... 4 > checking size of int... 4 > checking size of short int... 2 > checking size of long double... 16 > checking size of double... 8 > checking size of float... 4 > checking size of char *... 4 > checking for the null device... /dev/null > checking for gethostbyname... yes > checking whether mode_t is defined... yes > checking for availability of gettimeofday timing... yes > checking if gettimeofday needs timezone... yes > checking for availability of getrusage timing... yes > checking for availability of _ftime timing... no > checking for time.h... yes > checking for sys/time.h... yes > checking for sys/types.h... yes > checking for unistd.h... yes > checking for string.h... yes > checking for assert.h... yes > checking for tgmath.h... yes > checking for sys/stat.h... yes > checking for getopt.h... yes > checking for dirent.h... yes > checking for c_asm.h... no > checking for intrinsics.h... no > checking for mach/mach_time.h... yes > checking for regex.h... yes > checking for sys/filio.h... yes > checking for sys/ioctl.h... yes > checking for sys/socket.h... yes > checking for netinet/in.h... yes > checking for netdb.h... yes > checking for arpa/inet.h... yes > checking for winsock2.h... no > checking for crypt.h... no > checking for signal.h... yes > checking for malloc.h... no > checking whether time.h and sys/time.h may both be included... yes > checking whether socklen_t is defined... yes > checking whether SOCKET is defined... no > checking for hrtime_t... no > checking for gethrtime... no > checking for read_real_time... no > checking for time_base_to_time... no > checking for clock_gettime... no > checking for mach_absolute_time... yes > checking for _rtc intrinsic... no > checking for mallinfo... no > checking for mallopt... no > checking whether M_MMAP_THRESHOLD is defined... no > checking if __int64 is defined... yes > checking for vector... yes > checking for vector.h... yes > checking for getopt_long_only... yes > checking for working const... yes > checking for C inline... inline > checking for C restrict... __restrict__ > checking for C++ restrict... __restrict__ > checking for C bool... no > checking for CXX bool... yes > checking for Fortran REAL*4... yes > checking for Fortran REAL*8... yes > checking for Fortran REAL*16... yes > checking for Fortran COMPLEX*8... yes > checking for Fortran COMPLEX*16... yes > checking for Fortran COMPLEX*32... yes > checking for crypt... yes > checking for finite... yes > checking for isnan... yes > ^[[Bchecking for isinf... yes > checking for mkstemp... yes > checking for va_copy... yes > creating cctk_Archdefs.h > creating make.arch.defn > creating cctk_Extradefs.h > creating make.extra.defn > cctk_Archdefs.h is unchanged > make.arch.defn is unchanged > cctk_Extradefs.h is unchanged > make.extra.defn is unchanged > updating cache ./config.cache > creating ./config.status > creating make.config.defn > creating make.config.deps > creating make.config.rules > creating cctk_Config.h > Determining number of fortran underscores... > Compiling test file with ifort -align -w95 ... > Lowercase - One trailing underscore > Compiling test file with ifort -align -w95 ... > Lowercase - One trailing underscore > > _______________________________________________ > Users mailing list > Users at cactuscode.org > http://www.cactuscode.org/mailman/listinfo/users > -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080516/95285eca/attachment.bin From frank.loeffler at aei.mpg.de Sat May 17 03:27:13 2008 From: frank.loeffler at aei.mpg.de (Frank Loeffler) Date: Sat, 17 May 2008 03:27:13 -0500 Subject: [Users] (no subject) In-Reply-To: <31D46DC505870A4C980A8F38BB2DBD0E4E52F720AB@MUXC09.marshall.edu> References: <31D46DC505870A4C980A8F38BB2DBD0E4E52F720AB@MUXC09.marshall.edu> Message-ID: <20080517082713.GF28333@numrel07.cct.lsu.edu> On Fri, May 16, 2008 at 05:19:37PM -0400, Babiuc-Hamilton, Maria wrote: > Does anybody know how to compile cactus on a Mac OS X 10.4, with either the intel r Absoft compilers? Even if I remove all the thorn, I still get an ' no rule to make target' error: Can you configure a clean (new) configuration and look whether the compiler is recognized? You could send the stdout during the configure stage. Look for lines like: checking whether the C compiler (/usr/local/compilers/intel_cce_10.1.013_x64/bin/icc -openmp -g -debug all -align -std=gnu99 -ansi_alias -traceback -Wl,-rpath,/usr/local/packages/numrel/gsl-1.8/lib -Wl,-rpath,/usr/local/packages/numrel/hdf5-1.8.0-beta5/lib -Wl,-rpath,/usr/local/compilers/intel_cce_10.1.013_x64/lib) works... yes checking whether the C compiler (/usr/local/compilers/intel_cce_10.1.013_x64/bin/icc -openmp -g -debug all -align -std=gnu99 -ansi_alias -traceback -Wl,-rpath,/usr/local/packages/numrel/gsl-1.8/lib -Wl,-rpath,/usr/local/packages/numrel/hdf5-1.8.0-beta5/lib -Wl,-rpath,/usr/local/compilers/intel_cce_10.1.013_x64/lib) is a cross-compiler... no checking whether we are using GNU C... yes checking whether the C++ compiler (/usr/local/compilers/intel_cce_10.1.013_x64/bin/icpc -openmp -g -debug all -align -restrict -traceback -Wl,-rpath,/usr/local/packages/numrel/gsl-1.8/lib -Wl,-rpath,/usr/local/packages/numrel/hdf5-1.8.0-beta5/lib -Wl,-rpath,/usr/local/compilers/intel_cce_10.1.013_x64/lib) works... yes checking whether the C++ compiler (/usr/local/compilers/intel_cce_10.1.013_x64/bin/icpc -openmp -g -debug all -align -restrict -traceback -Wl,-rpath,/usr/local/packages/numrel/gsl-1.8/lib -Wl,-rpath,/usr/local/packages/numrel/hdf5-1.8.0-beta5/lib -Wl,-rpath,/usr/local/compilers/intel_cce_10.1.013_x64/lib) is a cross-compiler... no checking whether we are using GNU C++... yes Frank From schnetter at cct.lsu.edu Sat May 17 16:56:46 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Sat, 17 May 2008 16:56:46 -0500 Subject: [Users] PUGH Questions In-Reply-To: <20080510163821.GA20427@neuromancer> References: <20080509000455.GA25396@wintermute> <20080509004451.GZ12358@numrel07.cct.lsu.edu> <20080509064340.GA31862@wintermute> <2F735EF4-620F-47DE-BDEE-056F92C3DD0D@cct.lsu.edu> <20080510163821.GA20427@neuromancer> Message-ID: On May 10, 2008, at 11:38:21, Andreas Sch?fer wrote: > On 15:48 Fri 09 May , Erik Schnetter wrote: >> >> Carpet n-sects, where n is chosen depending on the >> number of processors. Each dimension is n-sected only once. While >> bisection leads to a binary tree, n-section leads to a wider and more >> shallow tree. > > How exactly do you choose n? Does n have to be a divisor of the number > of processors? If you n-sec each dimension only once, couldn't this > also lead to a degenerated surface to volume ratio? We choose n such that the ratio of n to the total number of processors is approximately the same as the ratio of the side length to the total number of grid points. Each sub-block is then n-sected independently. The surface-to-volume ratio should stay small, except if there are only very few processors or if the domain has a degenerated shape. For example, the following domain shape could be generated on five processors: AAABBB AAABBB CCDDEE CCDDEE CCDDEE where the letter describes which processor owns a grid point. This could arise by first 2-secting in the vertical direction, and then 2- secting and 3-secting the upper and lower parts, respectively. > I'm not sure how "each dimension is n-sected only once" affects this, > but using bisections we've observed a nasty effect during load > balancing that we call "flip over": > Could the example above happen when balancing the load with Carpet? If > not, how does the n-section work exactly? We are cheating: during load balancing, we re-distribute the grid completely. This works for us since we need to redistribute only every N time steps, and (after some optimisations) the time spent redistributing the whole grid does not seem to be too expensive for us. Having said this, we are just now defining a benchmark to actually measure this under "realistic" conditions instead of just examining timing information in production runs. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080517/e80cd80f/attachment.bin From schnetter at cct.lsu.edu Sat May 17 22:48:32 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Sat, 17 May 2008 22:48:32 -0500 Subject: [Users] using PETSc with the appropriate benchmark In-Reply-To: References: <6404F96D-2EC4-401F-A2F7-BC11650E3017@cct.lsu.edu> Message-ID: <148D2758-FC60-4423-8AB8-7AE3D8375D87@cct.lsu.edu> On Apr 14, 2008, at 16:06:00, Avi Purkayastha wrote: >> The problem size is chosen by the parameter global_nsize e.g. in >> CactusWave/IDScalarWaveElliptic/par/source_petsc.par. > > btw, I noticed that was no CactusWave thorn in the Cactus source that > I have. How does one get it separately? Avi, I just noticed this question of yours. Your Cactus source code is probably the benchmarking source code which contains only those thorns necessary for the benchmarks, and which therefore do not include CactusWave. If you check out Cactus as e.g. described on , then you will also download some sample thorns, such as CactusWave. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080517/a1702489/attachment.bin From hinder at gravity.psu.edu Mon May 19 12:11:20 2008 From: hinder at gravity.psu.edu (Ian Hinder) Date: Mon, 19 May 2008 13:11:20 -0400 Subject: [Users] ScheduleADMBase Message-ID: <4831B4B8.6030805@gravity.psu.edu> Hi, Having just updated my Cactus tree, I am having trouble compiling. I am getting lots of warnings like: /home/hinder/xirel/configs/xirel4/bindings/Schedule/ScheduleADMBase.c(60): warning #810: conversion from "int *" to "int" may lose significant bits cctkschedulei_tlevelarray /* Array of timelevel data for storage groups */, ^ /home/hinder/xirel/configs/xirel4/bindings/Schedule/ScheduleADMBase.c(61): warning #167: argument of type "char *" is incompatible with parameter of type "const int *" "global"); and a fatal error /home/hinder/xirel/configs/xirel4/bindings/Schedule/ScheduleADMBase.c(79): error #165: too few arguments in function call cctkschedulei_tlevelarray /* Array of timelevel data for storage groups */); ^ I have tried doing a -rebuild and a -realclean, and neither helped. Any ideas? -- Ian Hinder hinder at gravity.psu.edu http://www.gravity.psu.edu/~hinder From jtao at cct.lsu.edu Mon May 19 13:06:24 2008 From: jtao at cct.lsu.edu (Jian Tao) Date: Mon, 19 May 2008 13:06:24 -0500 Subject: [Users] ScheduleADMBase In-Reply-To: <4831B4B8.6030805@gravity.psu.edu> References: <4831B4B8.6030805@gravity.psu.edu> Message-ID: <4831C1A0.50203@cct.lsu.edu> Hi Ian, I haven't seen any compilation errors with the latest version of ADMBase. My guess is that either your version of ADMBase is not up to date or your "cvs update" messed up the code a bit. What does cvs diff -D "1 minute ago" show ? Regards, Jian Ian Hinder wrote: > Hi, > > Having just updated my Cactus tree, I am having trouble compiling. I am > getting lots of warnings like: > > /home/hinder/xirel/configs/xirel4/bindings/Schedule/ScheduleADMBase.c(60): > warning #810: conversion from "int *" to "int" may lose significant bits > cctkschedulei_tlevelarray /* Array of > timelevel data for storage groups */, > ^ > > /home/hinder/xirel/configs/xirel4/bindings/Schedule/ScheduleADMBase.c(61): > warning #167: argument of type "char *" is incompatible with parameter > of type "const int *" > "global"); > > > > > and a fatal error > > /home/hinder/xirel/configs/xirel4/bindings/Schedule/ScheduleADMBase.c(79): > error #165: too few arguments in function call > cctkschedulei_tlevelarray /* Array of > timelevel data for storage groups */); > > ^ > > I have tried doing a -rebuild and a -realclean, and neither helped. Any > ideas? > From schnetter at cct.lsu.edu Mon May 19 13:38:34 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Mon, 19 May 2008 13:38:34 -0500 Subject: [Users] ScheduleADMBase In-Reply-To: <4831B4B8.6030805@gravity.psu.edu> References: <4831B4B8.6030805@gravity.psu.edu> Message-ID: On May 19, 2008, at 12:11:20, Ian Hinder wrote: > I have tried doing a -rebuild and a -realclean, and neither helped. > Any > ideas? It could be that your version of the flesh is inconsistent. For example, if you updated the src directory but not the lib directory, then the auto-generated C files do not match the routines in the src directory any more. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080519/83b677fc/attachment.bin From hinder at gravity.psu.edu Mon May 19 13:46:34 2008 From: hinder at gravity.psu.edu (Ian Hinder) Date: Mon, 19 May 2008 14:46:34 -0400 Subject: [Users] ScheduleADMBase In-Reply-To: References: <4831B4B8.6030805@gravity.psu.edu> Message-ID: <4831CB0A.9020601@gravity.psu.edu> Erik Schnetter wrote: > On May 19, 2008, at 12:11:20, Ian Hinder wrote: > >> I have tried doing a -rebuild and a -realclean, and neither helped. Any >> ideas? > > > It could be that your version of the flesh is inconsistent. For > example, if you updated the src directory but not the lib directory, > then the auto-generated C files do not match the routines in the src > directory any more. Yes - that is exactly what I did :) After Jian's email, I realized what I had done and updated the lib directory. Everything works now. Thanks guys! In future, I will use "make cvsupdate" rather than manually running cvs... -- Ian Hinder hinder at gravity.psu.edu http://www.gravity.psu.edu/~hinder From hinder at gravity.psu.edu Mon May 19 15:28:55 2008 From: hinder at gravity.psu.edu (Ian Hinder) Date: Mon, 19 May 2008 16:28:55 -0400 Subject: [Users] Timestamp consistency check Message-ID: <4831E307.9050005@gravity.psu.edu> Hi, I have recently had a problem with a fileserver which has its time set incorrectly (6 minutes into the past). The CST was looping indefinitely. Would it be possible to add a little check into Cactus somewhere which ensured that when a file is created, it gets a timestamp which is not before the time it was created, and an error or warning printed if there is evidence that the fileserver time is incorrect? -- Ian Hinder hinder at gravity.psu.edu http://www.gravity.psu.edu/~hinder From schnetter at cct.lsu.edu Mon May 19 16:34:48 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Mon, 19 May 2008 16:34:48 -0500 Subject: [Users] Timestamp consistency check In-Reply-To: <4831E307.9050005@gravity.psu.edu> References: <4831E307.9050005@gravity.psu.edu> Message-ID: On May 19, 2008, at 15:28:55, Ian Hinder wrote: > Hi, > > I have recently had a problem with a fileserver which has its time set > incorrectly (6 minutes into the past). The CST was looping > indefinitely. Would it be possible to add a little check into Cactus > somewhere which ensured that when a file is created, it gets a > timestamp > which is not before the time it was created, and an error or warning > printed if there is evidence that the fileserver time is incorrect? Usually, make does not compare file timestamps to the current time, but only compares file timestamps to other file timestamps. This prevents most of such problems since only the file server's internal consistency is needed, not its absolute consistency with other machines. This fails if someone uses the command "touch", which (a) creates a file and then (b) gives the file the current time of the machine where make is running, not the file server's current time. It is better to create files with commands like : > file echo > file instead of touch. I see there is at least one "touch" command in the linux configure script. Did Cactus really loop in the CST stage, or was it in the configuration state? The CST stage interprets the thorns' ccl files, whereas the configure stage uses autoconf and friends. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080519/813319f2/attachment.bin From hinder at gravity.psu.edu Mon May 19 18:35:31 2008 From: hinder at gravity.psu.edu (Ian Hinder) Date: Mon, 19 May 2008 19:35:31 -0400 Subject: [Users] Timestamp consistency check In-Reply-To: References: <4831E307.9050005@gravity.psu.edu> Message-ID: <48320EC3.4030601@gravity.psu.edu> Erik Schnetter wrote: > On May 19, 2008, at 15:28:55, Ian Hinder wrote: > >> Hi, >> >> I have recently had a problem with a fileserver which has its time set >> incorrectly (6 minutes into the past). The CST was looping >> indefinitely. Would it be possible to add a little check into Cactus >> somewhere which ensured that when a file is created, it gets a timestamp >> which is not before the time it was created, and an error or warning >> printed if there is evidence that the fileserver time is incorrect? > > > Usually, make does not compare file timestamps to the current time, but > only compares file timestamps to other file timestamps. This prevents > most of such problems since only the file server's internal consistency > is needed, not its absolute consistency with other machines. Ah - yes that makes sense. In my case, I think the problem might have been related to the ThornList file. This was being copied from another system, and its timestamp would have been preserved. So maybe the thornlist was being seen as newer than some other files in the configuration every time. The test case that would catch this is to create a file, and determine what time the fileserver thinks it is, and then to make sure that the thornlist is not newer than that. But that catches a very specific case in a roundabout way, so maybe it is not the right solution. A generic test to verify the consistency of the fileserver's clock with the current system's current time would work better. > This fails if someone uses the command "touch", which (a) creates a file > and then (b) gives the file the current time of the machine where make > is running, not the file server's current time. It is better to create > files with commands like > > : > file > echo > file Actually, the test case that I constructed was to touch a file and observe that the timestamp was in the past of a "date" command run before the touch. So this suggests that the fileserver *is* determining the timestamp, even with "touch". Maybe the fileserver refuses to make a timestamp in the past, even if asked to by the client? > instead of touch. I see there is at least one "touch" command in the > linux configure script. Did Cactus really loop in the CST stage, or was > it in the configuration state? The CST stage interprets the thorns' ccl > files, whereas the configure stage uses autoconf and friends. It was during a make rather than a make -config, so it was definitely not the configuration stage. -- Ian Hinder hinder at gravity.psu.edu http://www.gravity.psu.edu/~hinder From schnetter at cct.lsu.edu Mon May 19 20:27:25 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Mon, 19 May 2008 20:27:25 -0500 Subject: [Users] Timestamp consistency check In-Reply-To: <48320EC3.4030601@gravity.psu.edu> References: <4831E307.9050005@gravity.psu.edu> <48320EC3.4030601@gravity.psu.edu> Message-ID: On May 19, 2008, at 18:35:31, Ian Hinder wrote: > Erik Schnetter wrote: >> On May 19, 2008, at 15:28:55, Ian Hinder wrote: >> >>> Hi, >>> >>> I have recently had a problem with a fileserver which has its time >>> set >>> incorrectly (6 minutes into the past). The CST was looping >>> indefinitely. Would it be possible to add a little check into >>> Cactus >>> somewhere which ensured that when a file is created, it gets a >>> timestamp >>> which is not before the time it was created, and an error or warning >>> printed if there is evidence that the fileserver time is incorrect? Usually, make itself should at least complain (but maybe not abort). >> Usually, make does not compare file timestamps to the current time, >> but >> only compares file timestamps to other file timestamps. This >> prevents >> most of such problems since only the file server's internal >> consistency >> is needed, not its absolute consistency with other machines. > > Ah - yes that makes sense. In my case, I think the problem might have > been related to the ThornList file. This was being copied from > another > system, and its timestamp would have been preserved. Normally, copying a file does not preserve the time stamp. If the thorn list is copied or created anew, it should be tagged with the current time of the file server. Did you copy the thorn list in a special way, e.g. explicitly preserving its time stamp? > So maybe the > thornlist was being seen as newer than some other files in the > configuration every time. The test case that would catch this is to > create a file, and determine what time the fileserver thinks it is, > and > then to make sure that the thornlist is not newer than that. Creating a file tags it with the file server's current time. This is also the only way to query the file server about its idea of the current time. > But that > catches a very specific case in a roundabout way, so maybe it is not > the > right solution. A generic test to verify the consistency of the > fileserver's clock with the current system's current time would work > better. That is possible. One can create a file with the file server's current time and one with the local machine's current time, e.g. with : > server touch local >> This fails if someone uses the command "touch", which (a) creates a >> file >> and then (b) gives the file the current time of the machine where >> make >> is running, not the file server's current time. It is better to >> create >> files with commands like >> >> : > file >> echo > file > > Actually, the test case that I constructed was to touch a file and > observe that the timestamp was in the past of a "date" command run > before the touch. So this suggests that the fileserver *is* > determining > the timestamp, even with "touch". That is very fishy. It would also be easy to detect. > Maybe the fileserver refuses to make > a timestamp in the past, even if asked to by the client? It shouldn't. It doesn't on some LONI machines I just tested (touch - t 01010101 file). Can you try the touch again and send a screen shot? -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080519/efae238c/attachment.bin From babiuc at marshall.edu Thu May 22 09:18:16 2008 From: babiuc at marshall.edu (Babiuc-Hamilton, Maria) Date: Thu, 22 May 2008 10:18:16 -0400 Subject: [Users] cactus compilation problem on mac Message-ID: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7888@MUXC09.marshall.edu> Hello, I was able to compile the wavetoy, and I do not know exactly how it did finally work. My guess is that a modification in the file make.config.rules.in was for the better. Now, though, I cannot compile the thorn AEIThorns/Fortran, which is essential for my code. I checked both the absoft and intel fortran compilers with a simple fortran porgram and they work fine. The error is: > Creating cactus_nulltest in /Users/mariababiuc/Cactus/exe from > CactusEinstein/ADMBase AEIThorns/AEILocalInterp CactusBase/Boundary > CactusBase/CartGrid3D CactusBase/CoordBase AEIThorns/Fortran > CactusBase/IOASCII CactusBase/IOBasic CactusPUGHIO/IOHDF5 > CactusPUGHIO/IOHDF5Util CactusBase/IOUtil CactusBase/LocalInterp > CactusBase/LocalReduce CactusBase/MoL CactusUtils/NaNChecker > CactusPUGH/PUGH CactusPUGH/PUGHInterp CactusPUGH/PUGHReduce > CactusPUGH/PUGHSlab CactusEinstein/SpaceMask CactusEinstein/StaticConformal > CactusBase/SymBase CactusBase/Time datetime=`/bin/date > +'%Y-%m-%dT%H:%M:%S%z' 2>&1` ; \ > if [ $? != 0 ]; then datetime=`/bin/date` ; fi ; \ > gcc -O2 -DCCTK_COMPILE_DATETIME="$datetime" -c -o > "/Users/mariababiuc/Cactus/configs/nulltest/datestamp.o" > -I"/Users/mariababiuc/Cactus/src/include" > -I"/Users/mariababiuc/Cactus/configs/nulltest/config-data" > "/Users/mariababiuc/Cactus/src/datestamp.c" if [ ! -d > /Users/mariababiuc/Cactus/exe ]; then mkdir -p > /Users/mariababiuc/Cactus/exe ; fi g++ -o > "/Users/mariababiuc/Cactus/exe/cactus_nulltest" -O2 > "/Users/mariababiuc/Cactus/configs/nulltest/datestamp.o" > -L/Users/mariababiuc/Cactus/configs/nulltest/lib -lthorn_Cactus > -lthorn_CactusBindings -lthorn_IOHDF5 -lthorn_IOHDF5Util -lthorn_MoL > -lthorn_NaNChecker -lthorn_PUGHReduce -lthorn_PUGHInterp -lthorn_IOBasic > -lthorn_Time -lthorn_SymBase -lthorn_LocalReduce -lthorn_LocalInterp > -lthorn_Fortran -lthorn_Boundary -lthorn_AEILocalInterp -lthorn_IOASCII > -lthorn_IOUtil -lthorn_StaticConformal -lthorn_ADMBase -lthorn_PUGHSlab > -lthorn_PUGH -lthorn_SpaceMask -lthorn_CartGrid3D -lthorn_CoordBase > -lthorn_Cactus -lthorn_CactusBindings -L/usr/local/hdf5/lib -lhdf5 -lz -lm > -lm /usr/bin/ld: Undefined symbols: > __absoft_casgn > collect2: ld returned 1 exit status > make[1]: *** [/Users/mariababiuc/Cactus/exe/cactus_nulltest] Error 1 ________________________________________ Any ideas? Thanks, Maria From babiuc at marshall.edu Thu May 22 09:24:14 2008 From: babiuc at marshall.edu (Babiuc-Hamilton, Maria) Date: Thu, 22 May 2008 10:24:14 -0400 Subject: [Users] cactus compilation problem on bigben Message-ID: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7889@MUXC09.marshall.edu> Hello, I have a cactus compilation oroblem on the Pittsburgh supercomputer bigben, after updating to the current cvs version of cactus. The problem is somehow similar to the one I had with the mac, and I am wondering if a change in make.config.rules.in would fix it. The output of the eror is: Checking status of thorn Boundary ________________________________________________________________________ /opt/xt-pe/1.5.47/bin/snos64/cc: INFO: catamount target is being used /opt/xt-pe/1.5.47/bin/snos64/cc: INFO: catamount target is being used /opt/xt-pe/1.5.47/bin/snos64/cc: INFO: catamount target is being used /opt/xt-pe/1.5.47/bin/snos64/cc: INFO: catamount target is being used /opt/xt-pe/1.5.47/bin/snos64/cc: INFO: catamount target is being used /opt/xt-pe/1.5.47/bin/snos64/cc: INFO: catamount target is being used /opt/xt-pe/1.5.47/bin/snos64/cc: INFO: catamount target is being used /opt/xt-pe/1.5.47/bin/snos64/cc: INFO: catamount target is being used /opt/xt-pe/1.5.47/bin/snos64/cc: INFO: catamount target is being used /opt/xt-pe/1.5.47/bin/snos64/cc: INFO: catamount target is being used Preprocessing /usr/users/7/babiuc/Cactus/arrangements/CactusBase/Boundary/src/ScalarBoundary.c Compiling /usr/users/7/babiuc/Cactus/arrangements/CactusBase/Boundary/src/ScalarBoundary.c /opt/xt-pe/1.5.47/bin/snos64/cc: INFO: catamount target is being used pgcc-Error-Unknown switch: -pipe pgcc-Error-Unknown switch: -std=gnu99 gmake[3]: *** [ScalarBoundary.c.o] Error 1 gmake[2]: *** [make.checked] Error 2 gmake[1]: *** [/usr/users/7/babiuc/Cactus/configs/wavetoy/lib/libthorn_Boundary.a] Error 2 gmake: *** [wavetoy] Error 2 The configuration output looks like this: Reconfiguring wavetoy. Adding configuration options from user defaults... Setting MACHINE_ARCH to 'x86_64' Setting CC to 'cc' Setting CXX to 'CC' Setting F90 to 'ftn' Setting F77 to 'ftn' Setting CPP to '/usr/bin/cpp' Setting FPP to '/usr/bin/cpp' Setting MPI to 'CUSTOM' Setting AR to 'ar' Setting RANLIB to 'ranlib' Setting ENDIAN to 'big' Setting SIZEOF_SHORT_INT to '2' Setting SIZEOF_INT to '4' Setting SIZEOF_LONG_INT to '8' Setting SIZEOF_LONG_LONG to '8' Setting SIZEOF_FLOAT to '4' Setting SIZEOF_DOUBLE to '8' Setting SIZEOF_LONG_DOUBLE to '8' Setting SIZEOF_POINTER to '8' Setting NULL_DEVICE to ''/dev/null'' Setting RANLIBFLAGS to '' Setting USE_RANLIB to 'yes' End of options from user defaults. creating cache ./config.cache checking host system type... x86_64-unknown-linux-gnu checking whether make sets ${MAKE}... yes checking whether the C compiler (cc ) works... yes checking whether the C compiler (cc ) is a cross-compiler... yes checking whether we are using GNU C... no checking whether the C++ compiler (CC ) works... yes checking whether the C++ compiler (CC ) is a cross-compiler... yes checking whether we are using GNU C++... no checking for perl... perl checking for mkdir... mkdir checking for bash... /bin/sh Setting CACHELINE_BYTES to 64 Cross compiling - unable to determine cache size Don't know details for cross-compilation to x86_64 checking how to run the C preprocessor... /usr/bin/cpp checking for ANSI C header files... yes checking size of long long... 8 checking size of long int... 8 checking size of int... 4 checking size of short int... 2 checking size of long double... 8 checking size of double... 8 checking size of float... 4 checking size of char *... 8 checking for gethostbyname... no checking for gethostbyname in library nsl... no checking whether mode_t is defined... no checking for availability of gettimeofday timing... no checking for availability of getrusage timing... no checking for availability of _ftime timing... no checking for time.h... no checking for sys/time.h... no checking for sys/types.h... no checking for unistd.h... no checking for string.h... no checking for assert.h... no checking for tgmath.h... no checking for sys/stat.h... no checking for getopt.h... no checking for dirent.h... no checking for c_asm.h... no checking for intrinsics.h... no checking for mach/mach_time.h... no checking for regex.h... no checking for sys/filio.h... no checking for sys/ioctl.h... no checking for sys/socket.h... no checking for netinet/in.h... no checking for netdb.h... no checking for arpa/inet.h... no checking for winsock2.h... no checking for crypt.h... no checking for signal.h... no checking for malloc.h... no checking whether time.h and sys/time.h may both be included... no checking whether socklen_t is defined... no checking whether SOCKET is defined... no checking for hrtime_t... no checking for gethrtime... no checking for read_real_time... no checking for time_base_to_time... no checking for clock_gettime... no checking for mach_absolute_time... no checking for _rtc intrinsic... no checking for mallinfo... no checking for mallopt... nochecking whether M_MMAP_THRESHOLD is defined... no checking if __int64 is defined... no checking for vector... yes checking for vector.h... no checking for getopt_long_only... no checking for working const... no checking for C inline... no checking for C restrict... no checking for C++ restrict... __restrict__ checking for C bool... no checking for CXX bool... yes checking for Fortran REAL*4... yes checking for Fortran REAL*8... yes checking for Fortran REAL*16... yes checking for Fortran COMPLEX*8... yes checking for Fortran COMPLEX*16... yes checking for Fortran COMPLEX*32... yes checking for crypt... no checking for crypt in library crypt... no checking for finite... no checking for finite in library m... no checking for isnan... no checking for isnan in library m... no checking for isinf... no checking for isinf in library m... no checking for mkstemp... no checking for mkstemp in library c... no checking for va_copy... no Configuring with MPI Custom MPI selected creating cctk_Archdefs.h creating make.arch.defn creating cctk_Extradefs.h creating make.extra.defn updating cache ./config.cache creating ./config.status creating make.config.defn creating make.config.deps creating make.config.rules creating cctk_Config.h Determining number of fortran underscores... Compiling test file with ftn -Mx,125,0x200 ... /opt/xt-pe/1.5.47/bin/snos64/ftn: INFO: catamount target is being used Lowercase - One trailing underscore Compiling test file with ftn -Mx,125,0x200 ... /opt/xt-pe/1.5.47/bin/snos64/ftn: INFO: catamount target is being used Lowercase - One trailing underscore Thanks, Maria From schnetter at cct.lsu.edu Fri May 23 07:15:39 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Fri, 23 May 2008 14:15:39 +0200 Subject: [Users] cactus compilation problem on bigben In-Reply-To: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7889@MUXC09.marshall.edu> References: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7889@MUXC09.marshall.edu> Message-ID: <3E2E2AA9-B6A6-43F6-A74E-ABEC98408135@cct.lsu.edu> On May 22, 2008, at 16:24:14, Babiuc-Hamilton, Maria wrote: > Hello, > > I have a cactus compilation oroblem on the Pittsburgh supercomputer > bigben, after updating to the current cvs version of cactus. > The problem is somehow similar to the one I had with the mac, and I > am wondering if a change in make.config.rules.in would fix it. Maria, the configure stage seems to pick up the compiler options "-pipe - std=gnu99" somewhere. This should not be the case for a Catamount system. Your output contains the line "Reconfiguring wavetoy." Can you instead configure from scratch? I have not used Big Ben myself, so I cannot really help you. Its architecture is quite different from other machines. I hope that someone else can help you. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080523/37193dec/attachment-0001.bin From schnetter at cct.lsu.edu Fri May 23 07:18:19 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Fri, 23 May 2008 14:18:19 +0200 Subject: [Users] cactus compilation problem on mac In-Reply-To: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7888@MUXC09.marshall.edu> References: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7888@MUXC09.marshall.edu> Message-ID: <63987D1B-8304-4556-B743-1DCB3C3BC7F0@cct.lsu.edu> On May 22, 2008, at 16:18:16, Babiuc-Hamilton, Maria wrote: > Hello, > > > I was able to compile the wavetoy, and I do not know exactly how it > did finally work. My guess is that a modification in the file > make.config.rules.in was for the better. > Now, though, I cannot compile the thorn AEIThorns/Fortran, which is > essential for my code. I checked both the absoft and intel fortran > compilers with a simple fortran porgram and they work fine. The > error is: > > >> Creating cactus_nulltest in /Users/mariababiuc/Cactus/exe from >> CactusEinstein/ADMBase AEIThorns/AEILocalInterp CactusBase/Boundary >> CactusBase/CartGrid3D CactusBase/CoordBase AEIThorns/Fortran >> CactusBase/IOASCII CactusBase/IOBasic CactusPUGHIO/IOHDF5 >> CactusPUGHIO/IOHDF5Util CactusBase/IOUtil CactusBase/LocalInterp >> CactusBase/LocalReduce CactusBase/MoL CactusUtils/NaNChecker >> CactusPUGH/PUGH CactusPUGH/PUGHInterp CactusPUGH/PUGHReduce >> CactusPUGH/PUGHSlab CactusEinstein/SpaceMask CactusEinstein/ >> StaticConformal >> CactusBase/SymBase CactusBase/Time datetime=`/bin/date >> +'%Y-%m-%dT%H:%M:%S%z' 2>&1` ; \ >> if [ $? != 0 ]; then datetime=`/bin/date` ; fi ; \ >> gcc -O2 -DCCTK_COMPILE_DATETIME="$datetime" -c -o >> "/Users/mariababiuc/Cactus/configs/nulltest/datestamp.o" >> -I"/Users/mariababiuc/Cactus/src/include" >> -I"/Users/mariababiuc/Cactus/configs/nulltest/config-data" >> "/Users/mariababiuc/Cactus/src/datestamp.c" if [ ! -d >> /Users/mariababiuc/Cactus/exe ]; then mkdir -p >> /Users/mariababiuc/Cactus/exe ; fi g++ -o >> "/Users/mariababiuc/Cactus/exe/cactus_nulltest" -O2 >> "/Users/mariababiuc/Cactus/configs/nulltest/datestamp.o" >> -L/Users/mariababiuc/Cactus/configs/nulltest/lib -lthorn_Cactus >> -lthorn_CactusBindings -lthorn_IOHDF5 -lthorn_IOHDF5Util -lthorn_MoL >> -lthorn_NaNChecker -lthorn_PUGHReduce -lthorn_PUGHInterp - >> lthorn_IOBasic >> -lthorn_Time -lthorn_SymBase -lthorn_LocalReduce -lthorn_LocalInterp >> -lthorn_Fortran -lthorn_Boundary -lthorn_AEILocalInterp - >> lthorn_IOASCII >> -lthorn_IOUtil -lthorn_StaticConformal -lthorn_ADMBase - >> lthorn_PUGHSlab >> -lthorn_PUGH -lthorn_SpaceMask -lthorn_CartGrid3D -lthorn_CoordBase >> -lthorn_Cactus -lthorn_CactusBindings -L/usr/local/hdf5/lib -lhdf5 - >> lz -lm >> -lm /usr/bin/ld: Undefined symbols: >> __absoft_casgn >> collect2: ld returned 1 exit status >> make[1]: *** [/Users/mariababiuc/Cactus/exe/cactus_nulltest] Error 1 > ________________________________________ > > Any ideas? The routine absoft_casgn is not found. It is probably defined in one of Absoft's libraries. You need to find out which library this is, and add this library explicitly, probably using LDFLAGS. For example, I use the setting LDFLAGS = -lgfortran for gcc43 on my system. The Unix command "nm" lists the routines which are defined in a library. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080523/20994696/attachment.bin From dlleuz at xmission.com Wed May 28 16:27:42 2008 From: dlleuz at xmission.com (rsoares) Date: Wed, 28 May 2008 15:27:42 -0600 Subject: [Users] HDF Output for Wavedemo Message-ID: <483DCE4E.9070804@xmission.com> Hi, Running Wavedemo generates several .xg .asc .jpeg files. 1.) How do you specify output to be generated in hdf format rather than .xg etc or in addition. 2.) the .xg files for Wavedemo are large and don't load into xgraph. Can xgraph be configured to accept these? Thanks for your time. Sincerely, R.Soares From babiuc at marshall.edu Wed May 28 17:17:44 2008 From: babiuc at marshall.edu (Babiuc-Hamilton, Maria) Date: Wed, 28 May 2008 18:17:44 -0400 Subject: [Users] cactus compilation problem on bigben In-Reply-To: <3E2E2AA9-B6A6-43F6-A74E-ABEC98408135@cct.lsu.edu> References: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7889@MUXC09.marshall.edu>, <3E2E2AA9-B6A6-43F6-A74E-ABEC98408135@cct.lsu.edu> Message-ID: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7896@MUXC09.marshall.edu> Hello, It looks to me that the catamount system is not properly recognized and the architecture is default to linux, which is the only one that has this compiler option. Bigben is a Cray XT3, so the problem might be in the test that is done to recognize the architecture, at the configuration time. Is there a way to fix this problem? Maria ________________________________________ From: users-bounces at cactuscode.org [users-bounces at cactuscode.org] On Behalf Of Erik Schnetter [schnetter at cct.lsu.edu] Sent: Friday, May 23, 2008 8:15 AM To: Cactus users mailing list Subject: Re: [Users] cactus compilation problem on bigben On May 22, 2008, at 16:24:14, Babiuc-Hamilton, Maria wrote: > Hello, > > I have a cactus compilation oroblem on the Pittsburgh supercomputer > bigben, after updating to the current cvs version of cactus. > The problem is somehow similar to the one I had with the mac, and I > am wondering if a change in make.config.rules.in would fix it. Maria, the configure stage seems to pick up the compiler options "-pipe - std=gnu99" somewhere. This should not be the case for a Catamount system. Your output contains the line "Reconfiguring wavetoy." Can you instead configure from scratch? I have not used Big Ben myself, so I cannot really help you. Its architecture is quite different from other machines. I hope that someone else can help you. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. From tradke at aei.mpg.de Thu May 29 05:20:23 2008 From: tradke at aei.mpg.de (Thomas Radke) Date: Thu, 29 May 2008 12:20:23 +0200 Subject: [Users] cactus compilation problem on bigben In-Reply-To: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7896@MUXC09.marshall.edu> References: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7889@MUXC09.marshall.edu>, <3E2E2AA9-B6A6-43F6-A74E-ABEC98408135@cct.lsu.edu> <31D46DC505870A4C980A8F38BB2DBD0E4E544E7896@MUXC09.marshall.edu> Message-ID: <483E8367.9060400@aei.mpg.de> Babiuc-Hamilton, Maria wrote: > Hello, > > It looks to me that the catamount system is not properly recognized and the architecture is default to linux, which is the only one that has this compiler option. > Bigben is a Cray XT3, so the problem might be in the test that is done to recognize the architecture, at the configuration time. > Is there a way to fix this problem? Hi Maria, can you send us the output of the Cactus configuration stage so that we can check whether the architecture isn't properly recognised ? -- Cheers, Thomas. From yye00 at cct.lsu.edu Thu May 29 07:27:58 2008 From: yye00 at cct.lsu.edu (Yaakoub El Khamra) Date: Thu, 29 May 2008 07:27:58 -0500 Subject: [Users] cactus compilation problem on bigben In-Reply-To: <483E8367.9060400@aei.mpg.de> References: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7889@MUXC09.marshall.edu> , <3E2E2AA9-B6A6-43F6-A74E-ABEC98408135@cct.lsu.edu> <31D46DC505870A4C980A8F38BB2DBD0E4E544E7896@MUXC09.marshall.edu> <483E8367.9060400@aei.mpg.de> Message-ID: <1212064078.23342.5.camel@localhost.localdomain> Hi Maria Out of curiosity where are you configuring Cactus on the head nodes or compute nodes? I remember having to cross compile on BigBen. On Thu, 2008-05-29 at 12:20 +0200, Thomas Radke wrote: > Babiuc-Hamilton, Maria wrote: > > Hello, > > > > It looks to me that the catamount system is not properly recognized and the architecture is default to linux, which is the only one that has this compiler option. > > Bigben is a Cray XT3, so the problem might be in the test that is done to recognize the architecture, at the configuration time. > > Is there a way to fix this problem? > > Hi Maria, > > can you send us the output of the Cactus configuration stage so that we > can check whether the architecture isn't properly recognised ? > From yye00 at cct.lsu.edu Thu May 29 07:46:38 2008 From: yye00 at cct.lsu.edu (Yaakoub El Khamra) Date: Thu, 29 May 2008 07:46:38 -0500 Subject: [Users] HDF Output for Wavedemo In-Reply-To: <483DCE4E.9070804@xmission.com> References: <483DCE4E.9070804@xmission.com> Message-ID: <1212065198.23342.22.camel@localhost.localdomain> Greetings > 1.) How do you specify output to be generated in hdf format rather than > .xg etc or in addition. You will need to do the following: 1) Make sure you have the CactusPUGHIO/IOHDF5 and CactusPUGHIO/IOHDF5Util in arrangements, if you don't you need to check them out 2) Add the CactusPUGHIO/IOHDF5 and CactusPUGHIO/IOHDF5Util to your configuration's thornlist (edit the file configs//ThornList 3) Compile the configuration 4) Edit the parameter file you want to use with the executable you just compiled, add IOHDF5 and IOHDF5Util to the list of active thorns 5) Add the parameters that control the HDF5 IO to the parameter file, things like: iohdf5::out_dir="/scratch/" iohdf5::out_vars="phi" iohdf5::out_every = 10 You can find a list of parameters that you can add by looking at the param.ccl file in the IOHDF5 thorn. The visualization How-To is helpful if you want to check it out: http://www.cactuscode.org/Documentation/visualizationHowTo.txt For other sample parameter files containing HDF5 you can have a look at the ones we use for IO benchmarking http://www.cactuscode.org/Benchmarks/benchio_hdf5 > 2.) the .xg files for Wavedemo are large and don't load into xgraph. > Can xgraph be configured to accept these? Not sure if xgraph can be configured but for somewhat large data we tend to use gnuplot for ASCII and for really large data we use HDF5. You might want to check out the visualization webpage for Cactus: http://www.cactuscode.org/Visualization/ > > Thanks for your time. > > Sincerely, > > R.Soares > _______________________________________________ > Users mailing list > Users at cactuscode.org > http://www.cactuscode.org/mailman/listinfo/users > From babiuc at marshall.edu Thu May 29 08:08:35 2008 From: babiuc at marshall.edu (Babiuc-Hamilton, Maria) Date: Thu, 29 May 2008 09:08:35 -0400 Subject: [Users] cactus compilation problem on bigben In-Reply-To: <483E8367.9060400@aei.mpg.de> References: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7889@MUXC09.marshall.edu>, <3E2E2AA9-B6A6-43F6-A74E-ABEC98408135@cct.lsu.edu> <31D46DC505870A4C980A8F38BB2DBD0E4E544E7896@MUXC09.marshall.edu>, <483E8367.9060400@aei.mpg.de> Message-ID: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7898@MUXC09.marshall.edu> Hello, Sure. The screen output for the configuration stage is: Cactus - version: 4.0.b16 Reconfiguring wavetoy. Adding configuration options from user defaults... Setting MACHINE_ARCH to 'x86_64' Setting CC to 'cc' Setting CXX to 'CC' Setting F90 to 'ftn' Setting F77 to 'ftn' Setting CPP to '/usr/bin/cpp' Setting FPP to '/usr/bin/cpp' Setting MPI to 'CUSTOM' Setting AR to 'ar' Setting RANLIB to 'ranlib' Setting ENDIAN to 'big' Setting SIZEOF_SHORT_INT to '2' Setting SIZEOF_INT to '4' Setting SIZEOF_LONG_INT to '8' Setting SIZEOF_LONG_LONG to '8' Setting SIZEOF_FLOAT to '4' Setting SIZEOF_DOUBLE to '8' Setting SIZEOF_LONG_DOUBLE to '8' Setting SIZEOF_POINTER to '8' Setting NULL_DEVICE to ''/dev/null'' Setting RANLIBFLAGS to '' Setting USE_RANLIB to 'yes' End of options from user defaults. creating cache ./config.cache checking host system type... x86_64-unknown-linux-gnu checking whether make sets ${MAKE}... yes checking whether the C compiler (cc ) works... yes checking whether the C compiler (cc ) is a cross-compiler... yes checking whether we are using GNU C... no checking whether the C++ compiler (CC ) works... yes checking whether the C++ compiler (CC ) is a cross-compiler... yes checking whether we are using GNU C++... no checking for perl... perl checking for mkdir... mkdir checking for bash... /bin/sh Setting CACHELINE_BYTES to 64 Cross compiling - unable to determine cache size Don't know details for cross-compilation to x86_64 checking how to run the C preprocessor... /usr/bin/cpp checking for ANSI C header files... yes checking size of long long... 8 checking size of long int... 8 checking size of int... 4 checking size of short int... 2 checking size of long double... 8 checking size of double... 8 checking size of float... 4 checking size of char *... 8 checking for gethostbyname... no checking for gethostbyname in library nsl... no checking whether mode_t is defined... no checking for availability of gettimeofday timing... no checking for availability of getrusage timing... no checking for availability of _ftime timing... no checking for time.h... no checking for sys/time.h... no checking for sys/types.h... no checking for unistd.h... no checking for string.h... no checking for assert.h... no checking for tgmath.h... no checking for sys/stat.h... no checking for getopt.h... no checking for dirent.h... no checking for c_asm.h... no checking for intrinsics.h... no checking for mach/mach_time.h... no checking for regex.h... no checking for sys/filio.h... no checking for sys/ioctl.h... no checking for sys/socket.h... no checking for netinet/in.h... no checking for netdb.h... no checking for arpa/inet.h... no checking for winsock2.h... no checking for crypt.h... no checking for signal.h... no checking for malloc.h... no checking whether time.h and sys/time.h may both be included... no checking whether socklen_t is defined... no checking whether SOCKET is defined... no checking for hrtime_t... no checking for gethrtime... no checking for read_real_time... no checking for time_base_to_time... no checking for clock_gettime... no checking for mach_absolute_time... no checking for _rtc intrinsic... no checking for mallinfo... no checking for mallopt... no checking whether M_MMAP_THRESHOLD is defined... no checking if __int64 is defined... no checking for vector... yes checking for vector.h... no checking for getopt_long_only... no checking for working const... no checking for C inline... no checking for C restrict... no checking for C++ restrict... __restrict__ checking for C bool... no checking for CXX bool... yes checking for Fortran REAL*4... yes checking for Fortran REAL*8... yes checking for Fortran REAL*16... yes checking for Fortran COMPLEX*8... yes checking for Fortran COMPLEX*16... yes checking for Fortran COMPLEX*32... yes checking for crypt... no checking for crypt in library crypt... no checking for finite... no checking for finite in library m... no checking for isnan... no checking for isnan in library m... no checking for isinf... no checking for isinf in library m... no checking for mkstemp... no checking for mkstemp in library c... no checking for va_copy... no Configuring with MPI Custom MPI selected creating cctk_Archdefs.h creating make.arch.defn creating cctk_Extradefs.h creating make.extra.defn cctk_Archdefs.h is unchanged make.arch.defn is unchanged cctk_Extradefs.h is unchanged make.extra.defn is unchanged updating cache ./config.cache creating ./config.status creating make.config.defn creating make.config.deps creating make.config.rules creating cctk_Config.h Determining number of fortran underscores... Compiling test file with ftn -Mx,125,0x200 ... /opt/xt-pe/1.5.47/bin/snos64/ftn: INFO: catamount target is being used Lowercase - One trailing underscore Compiling test file with ftn -Mx,125,0x200 ... /opt/xt-pe/1.5.47/bin/snos64/ftn: INFO: catamount target is being used Lowercase - One trailing underscore Best, Maria ________________________________________ From: users-bounces at cactuscode.org [users-bounces at cactuscode.org] On Behalf Of Thomas Radke [tradke at aei.mpg.de] Sent: Thursday, May 29, 2008 6:20 AM To: Cactus users mailing list Subject: Re: [Users] cactus compilation problem on bigben Babiuc-Hamilton, Maria wrote: > Hello, > > It looks to me that the catamount system is not properly recognized and the architecture is default to linux, which is the only one that has this compiler option. > Bigben is a Cray XT3, so the problem might be in the test that is done to recognize the architecture, at the configuration time. > Is there a way to fix this problem? Hi Maria, can you send us the output of the Cactus configuration stage so that we can check whether the architecture isn't properly recognised ? -- Cheers, Thomas. _______________________________________________ Users mailing list Users at cactuscode.org http://www.cactuscode.org/mailman/listinfo/users From tradke at aei.mpg.de Thu May 29 09:01:25 2008 From: tradke at aei.mpg.de (Thomas Radke) Date: Thu, 29 May 2008 16:01:25 +0200 Subject: [Users] cactus compilation problem on bigben In-Reply-To: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7898@MUXC09.marshall.edu> References: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7889@MUXC09.marshall.edu>, <3E2E2AA9-B6A6-43F6-A74E-ABEC98408135@cct.lsu.edu> <31D46DC505870A4C980A8F38BB2DBD0E4E544E7896@MUXC09.marshall.edu>, <483E8367.9060400@aei.mpg.de> <31D46DC505870A4C980A8F38BB2DBD0E4E544E7898@MUXC09.marshall.edu> Message-ID: <483EB735.8070706@aei.mpg.de> Babiuc-Hamilton, Maria wrote: > Hello, > > Sure. The screen output for the configuration stage is: Hmm, it's a cross compiler but that looks okay to me so far. What happens when you try to build a configuration (with option 'SILENT=no') ? -- Cheers, Thomas. From jtao at cct.lsu.edu Thu May 29 12:40:49 2008 From: jtao at cct.lsu.edu (Jian Tao) Date: Thu, 29 May 2008 12:40:49 -0500 Subject: [Users] HDF Output for Wavedemo In-Reply-To: <483DCE4E.9070804@xmission.com> References: <483DCE4E.9070804@xmission.com> Message-ID: <483EEAA1.1060002@cct.lsu.edu> rsoares wrote: > Hi, > > Running Wavedemo generates several .xg .asc .jpeg files. > > 1.) How do you specify output to be generated in hdf format rather than > .xg etc or in addition. To output files in HDF format, you will need to add the thorn CactusPUGHIO/IOHDF5 (I assume you are using PUGH) and compile Cactus together with HDF5. To compile HDF5 with Cactus, you need to insert the following in your option file (from Cactus user manual): HDF5=YES HDF5_DIR=dir You will probably need to specify the directory for libz or libsz. They could be specified with LIBZ DIR =dir LIBSZ DIR =dir A sample parameter file for outputting in HDF5 format can be found at arrangements/CactusPUGHIO/IOHDF5/par/hdf5.par under your Cactus directory. There are several packages you can use to visualize the HDF5 files as shown at http://www.cactuscode.org/Visualization/ . > 2.) the .xg files for Wavedemo are large and don't load into xgraph. > Can xgraph be configured to accept these? You can try to output not so often by setting the parameter "IO::out_every" to a big number. to shrink the file size. Xgraph should work out of box. What is the error message you got when you used Xgraph ? The problem you encounter was likely caused by the big output file. For plotting ASCII files, gnuplot should be a better choice. You can find more info on how to use it on Cactus output files at http://www.cactuscode.org/Visualization/gnuPlot Regards, Jian > Thanks for your time. > > Sincerely, > > R.Soares > _______________________________________________ > Users mailing list > Users at cactuscode.org > http://www.cactuscode.org/mailman/listinfo/users > From babiuc at marshall.edu Thu May 29 13:40:22 2008 From: babiuc at marshall.edu (Babiuc-Hamilton, Maria) Date: Thu, 29 May 2008 14:40:22 -0400 Subject: [Users] cactus compilation problem on bigben In-Reply-To: <483EB735.8070706@aei.mpg.de> References: <31D46DC505870A4C980A8F38BB2DBD0E4E544E7889@MUXC09.marshall.edu>, <3E2E2AA9-B6A6-43F6-A74E-ABEC98408135@cct.lsu.edu> <31D46DC505870A4C980A8F38BB2DBD0E4E544E7896@MUXC09.marshall.edu>, <483E8367.9060400@aei.mpg.de> <31D46DC505870A4C980A8F38BB2DBD0E4E544E7898@MUXC09.marshall.edu>, <483EB735.8070706@aei.mpg.de> Message-ID: <31D46DC505870A4C980A8F38BB2DBD0E4E544E789A@MUXC09.marshall.edu> Hello, The last part of the build with option SILENT-no output is: current_wd=`pwd` ; cd /usr/users/7/babiuc/Cactus/configs/wavetoy/scratch ; cc -pipe -std=gnu99 -O3 -c -o $current_wd/ScalarBoundary.c.o $current_wd/ScalarBoundary.c -I"/usr/users/7/babiuc/Cactus/arrangements/CactusBase/Boundary/src" -I"/usr/users/7/babiuc/Cactus/arrangements/CactusBase/Boundary/src/include" -I"/usr/users/7/babiuc/Cactus/configs/wavetoy/config-data" -I"/usr/users/7/babiuc/Cactus/configs/wavetoy/bindings/include" -I"/usr/users/7/babiuc/Cactus/src/include" -I"/usr/users/7/babiuc/Cactus/arrangements" -I"/usr/users/7/babiuc/Cactus/configs/wavetoy/bindings/Configuration/Thorns" -I"/usr/users/7/babiuc/Cactus/arrangements/CactusBase/Boundary/src" -I"/usr/users/7/babiuc/Cactus/arrangements/CactusBase/Boundary/src/include" -DTHORN_IS_Boundary -DCCODE /opt/xt-pe/1.5.47/bin/snos64/cc: INFO: catamount target is being used pgcc-Error-Unknown switch: -pipe pgcc-Error-Unknown switch: -std=gnu99 gmake[4]: *** [ScalarBoundary.c.o] Error 1 gmake[4]: Leaving directory `/usr/users/7/babiuc/Cactus/configs/wavetoy/build/Boundary' gmake[3]: *** [make.checked] Error 2 gmake[3]: Leaving directory `/usr/users/7/babiuc/Cactus/configs/wavetoy/build/Boundary' gmake[2]: *** [/usr/users/7/babiuc/Cactus/configs/wavetoy/lib/libthorn_Boundary.a] Error 2 The whole output is attached. I would really appreciate help, because I need to finish some runs fast. Thanks, Maria ________________________________________ From: users-bounces at cactuscode.org [users-bounces at cactuscode.org] On Behalf Of Thomas Radke [tradke at aei.mpg.de] Sent: Thursday, May 29, 2008 10:01 AM To: Cactus users mailing list Subject: Re: [Users] cactus compilation problem on bigben Babiuc-Hamilton, Maria wrote: > Hello, > > Sure. The screen output for the configuration stage is: Hmm, it's a cross compiler but that looks okay to me so far. What happens when you try to build a configuration (with option 'SILENT=no') ? -- Cheers, Thomas. _______________________________________________ Users mailing list Users at cactuscode.org http://www.cactuscode.org/mailman/listinfo/users -------------- next part -------------- A non-text attachment was scrubbed... Name: bigben_build Type: application/octet-stream Size: 19260 bytes Desc: bigben_build Url : http://www.cactuscode.org/pipermail/users/attachments/20080529/9e4488b3/attachment-0001.obj From babiuc at marshall.edu Fri May 30 08:47:45 2008 From: babiuc at marshall.edu (Babiuc-Hamilton, Maria) Date: Fri, 30 May 2008 09:47:45 -0400 Subject: [Users] cactus compilation problem Message-ID: <31D46DC505870A4C980A8F38BB2DBD0E4E544E78A1@MUXC09.marshall.edu> Hello, The compilation problem in bigben was solved by Roberto Gomez, by changing the line: RCS file: /cactusdevcvs/Cactus/lib/make/known-architectures/linux,v retrieving revision 1.126 diff -r1.126 linux 345c345 < gcc | cc) --- > gcc ) Now it works. The only problem that I have is with the thorn SphericalSurface: Checking status of thorn SphericalSurface ________________________________________________________________________ /opt/xt-pe/1.5.47/bin/snos64/CC: INFO: catamount target is being used "/opt/xt-mpt/1.5.47/mpich2-64/P2/include/mpicxx.h", line 26: catastrophic error: #error directive: "SEEK_SET is #defined but must not be for the C++ binding of MPI" #error "SEEK_SET is #defined but must not be for the C++ binding of MPI" ^ 1 catastrophic error detected in the compilation of "/usr/users/7/babiuc/Cactus/arrangements/AEIThorns/SphericalSurface/src/setup.cc". Compilation terminated. gmake[3]: *** [setup.cc.d] Error 2 gmake[2]: *** [make.checked] Error 2 gmake[1]: *** [/usr/users/7/babiuc/Cactus/configs/newstest/lib/libthorn_SphericalSurface.a] Error 2 gmake: *** [newstest] Error 2 Is there any easy fix for it? About the compilation on mac, I did not find anywhere what library shall I add explicitly, using LDFLAGS? There are 11 routines that contain __absoft_casgn in Absoft10.1/lib and 8 in Absoft10.1/lib64. cheers, Maria ________________________________________ From: users-bounces at cactuscode.org [users-bounces at cactuscode.org] On Behalf Of Erik Schnetter [schnetter at cct.lsu.edu] Sent: Friday, May 23, 2008 8:18 AM To: Cactus users mailing list Subject: Re: [Users] cactus compilation problem on mac On May 22, 2008, at 16:18:16, Babiuc-Hamilton, Maria wrote: > Hello, > > > I was able to compile the wavetoy, and I do not know exactly how it > did finally work. My guess is that a modification in the file > make.config.rules.in was for the better. > Now, though, I cannot compile the thorn AEIThorns/Fortran, which is > essential for my code. I checked both the absoft and intel fortran > compilers with a simple fortran porgram and they work fine. The > error is: > > >> Creating cactus_nulltest in /Users/mariababiuc/Cactus/exe from >> CactusEinstein/ADMBase AEIThorns/AEILocalInterp CactusBase/Boundary >> CactusBase/CartGrid3D CactusBase/CoordBase AEIThorns/Fortran >> CactusBase/IOASCII CactusBase/IOBasic CactusPUGHIO/IOHDF5 >> CactusPUGHIO/IOHDF5Util CactusBase/IOUtil CactusBase/LocalInterp >> CactusBase/LocalReduce CactusBase/MoL CactusUtils/NaNChecker >> CactusPUGH/PUGH CactusPUGH/PUGHInterp CactusPUGH/PUGHReduce >> CactusPUGH/PUGHSlab CactusEinstein/SpaceMask CactusEinstein/ >> StaticConformal >> CactusBase/SymBase CactusBase/Time datetime=`/bin/date >> +'%Y-%m-%dT%H:%M:%S%z' 2>&1` ; \ >> if [ $? != 0 ]; then datetime=`/bin/date` ; fi ; \ >> gcc -O2 -DCCTK_COMPILE_DATETIME="$datetime" -c -o >> "/Users/mariababiuc/Cactus/configs/nulltest/datestamp.o" >> -I"/Users/mariababiuc/Cactus/src/include" >> -I"/Users/mariababiuc/Cactus/configs/nulltest/config-data" >> "/Users/mariababiuc/Cactus/src/datestamp.c" if [ ! -d >> /Users/mariababiuc/Cactus/exe ]; then mkdir -p >> /Users/mariababiuc/Cactus/exe ; fi g++ -o >> "/Users/mariababiuc/Cactus/exe/cactus_nulltest" -O2 >> "/Users/mariababiuc/Cactus/configs/nulltest/datestamp.o" >> -L/Users/mariababiuc/Cactus/configs/nulltest/lib -lthorn_Cactus >> -lthorn_CactusBindings -lthorn_IOHDF5 -lthorn_IOHDF5Util -lthorn_MoL >> -lthorn_NaNChecker -lthorn_PUGHReduce -lthorn_PUGHInterp - >> lthorn_IOBasic >> -lthorn_Time -lthorn_SymBase -lthorn_LocalReduce -lthorn_LocalInterp >> -lthorn_Fortran -lthorn_Boundary -lthorn_AEILocalInterp - >> lthorn_IOASCII >> -lthorn_IOUtil -lthorn_StaticConformal -lthorn_ADMBase - >> lthorn_PUGHSlab >> -lthorn_PUGH -lthorn_SpaceMask -lthorn_CartGrid3D -lthorn_CoordBase >> -lthorn_Cactus -lthorn_CactusBindings -L/usr/local/hdf5/lib -lhdf5 - >> lz -lm >> -lm /usr/bin/ld: Undefined symbols: >> __absoft_casgn >> collect2: ld returned 1 exit status >> make[1]: *** [/Users/mariababiuc/Cactus/exe/cactus_nulltest] Error 1 > ________________________________________ > > Any ideas? The routine absoft_casgn is not found. It is probably defined in one of Absoft's libraries. You need to find out which library this is, and add this library explicitly, probably using LDFLAGS. For example, I use the setting LDFLAGS = -lgfortran for gcc43 on my system. The Unix command "nm" lists the routines which are defined in a library. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. From schnetter at cct.lsu.edu Fri May 30 09:11:22 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Fri, 30 May 2008 16:11:22 +0200 Subject: [Users] cactus compilation problem In-Reply-To: <31D46DC505870A4C980A8F38BB2DBD0E4E544E78A1@MUXC09.marshall.edu> References: <31D46DC505870A4C980A8F38BB2DBD0E4E544E78A1@MUXC09.marshall.edu> Message-ID: <32B742EE-173B-4465-B16B-90D0B05C02E2@cct.lsu.edu> On May 30, 2008, at 15:47:45, Babiuc-Hamilton, Maria wrote: > Hello, > > The compilation problem in bigben was solved by Roberto Gomez, by > changing the line: > RCS file: /cactusdevcvs/Cactus/lib/make/known-architectures/linux,v > retrieving revision 1.126 > diff -r1.126 linux > 345c345 > < gcc | cc) > --- >> gcc ) > Now it works. Thanks for the pointer. > The only problem that I have is with the thorn SphericalSurface: > > Checking status of thorn SphericalSurface > ________________________________________________________________________ > /opt/xt-pe/1.5.47/bin/snos64/CC: INFO: catamount target is being used > "/opt/xt-mpt/1.5.47/mpich2-64/P2/include/mpicxx.h", line 26: > catastrophic error: > #error directive: "SEEK_SET is #defined but must not be for > the C++ > binding of MPI" > #error "SEEK_SET is #defined but must not be for the C++ binding of > MPI" This is a problem with the C++ bindings of the MPI library. It can probably be avoided by CPPFLAGS = -DMPICH_IGNORE_CXX_SEEK -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080530/1365cf80/attachment.bin From schnetter at cct.lsu.edu Fri May 30 10:45:55 2008 From: schnetter at cct.lsu.edu (Erik Schnetter) Date: Fri, 30 May 2008 17:45:55 +0200 Subject: [Users] cactus compilation problem In-Reply-To: <31D46DC505870A4C980A8F38BB2DBD0E4E544E78A1@MUXC09.marshall.edu> References: <31D46DC505870A4C980A8F38BB2DBD0E4E544E78A1@MUXC09.marshall.edu> Message-ID: <4B6A1291-D227-4AD3-868D-E63A4E8897C9@cct.lsu.edu> On May 30, 2008, at 15:47:45, Babiuc-Hamilton, Maria wrote: > Hello, > > The compilation problem in bigben was solved by Roberto Gomez, by > changing the line: > RCS file: /cactusdevcvs/Cactus/lib/make/known-architectures/linux,v > retrieving revision 1.126 > diff -r1.126 linux > 345c345 > < gcc | cc) > --- >> gcc ) > Now it works. Maria, this indicates that you configured Big Ben as a Linux machine. Big Ben is an XT3 machine, and I think that configuring it as xt3-cray may be more appropriate. -erik -- Erik Schnetter http://www.cct.lsu.edu/~eschnett/ My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from www.keyserver.net. -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: This is a digitally signed message part Url : http://www.cactuscode.org/pipermail/users/attachments/20080530/0f7a0a66/attachment.bin