similar to: [LLVMdev] function pass to visit BB bottom-up order

Hi,      I am interested in getting llvm IR only for a subset of the input source code - basically starting from a specified top level function, including all its callee functions (recursively).      For example, in the following code, I am interested in a command like "extract -top_function blah()" that will create an llvm IR with just blah() and foo().       int foo() {          

fixing missing subject line in original email. ----- Forwarded Message ----- From: Ananth Durbha <avdurbha at yahoo.com> To: llvm list <llvmdev at cs.uiuc.edu> Sent: Tuesday, August 9, 2011 11:38 PM Subject: Hi,      I am interested in getting llvm IR only for a subset of the input source code - basically starting from a specified top level function, including all its callee

                              The getInitializer() method of a GlobalVariable returns 4 pointers like below. [4 x i32*] [i32* getelementptr inbounds ([256 x i32]* @CRC24ATable, i32 0, i32 0), i32* getelementptr inbounds ([256 x i32]* @CRC24BTable, i32 0, i32 0), i32* getelementptr inbounds ([256 x i32]* @CRC16Table, i32 0, i32 0), i32* getelementptr inbounds ([256 x i32]* @CRC8Table, i32 0,

Hi, I''m trying to setup xen on a server with bridge networking. I have RedHat as my dom0. I have two Ubuntu instances and one CentOS instance as domU''s. All of them (dom0 and domU) use static ip addresses. The configuration files for domU''s specify in the vif line what ip address to be assigned to that particular domU. It works fine for my ubuntu domU''s. But

The reverse postorder iterator does what you want. It's defined in ADT/PostOrderIterator.h, and is used like this: ReversePostOrderTraversal<Function*> RPOT(&F); for (ReversePostOrderTraversal<Function*>::rpo_iterator RI = RPOT.begin(), RE = RPOT.end(); RI != RE; ++RI) <work here> Creating a ReversePostOrderTraversal is not cheap, because it first has to

Hello, When I install guest OS on xen using "xm create /path/of/file/ubuntu.cfg -c" , I am getting an error message *"Error*: Device 0 (*vif*) could not be connected. Hotplug scripts not working <http://markmail.org/message/24aj3qs6ddrtwdks>" . I also saw that hotplugs and udev are prerequisites to be installed before compiling xen. I could not find hotplugs in Yast.

Hello, When I install guest OS on xen using "xm create /path/of/file/ubuntu.cfg -c" , I am getting an error message *"Error*: Device 0 (*vif*) could not be connected. Hotplug scripts not working <http://markmail.org/message/24aj3qs6ddrtwdks>" . I also saw that hotplugs and udev are prerequisites to be installed before compiling xen. I could not find hotplugs in Yast.

Hi, I am facing the below error while compiling dom0. Please guide. root@cbananth:/usr/src/xen-3.3.1# make linux-2.6-xen0-config CONFIGMODE=menuconfig make -f buildconfigs/mk.linux-2.6-xen0 config make[1]: Entering directory `/usr/src/xen-3.3.1'' set -e ; \ if [ ! -e linux-2.6.18-xen.hg/.hg ] ; then \ __repo=$(sh buildconfigs/select-repository linux-2.6.18-xen.hg .:..) ; \

Hi, I''m trying to create OpenSolaris (2008.11) domU on RedHat Linux dom0. I''m using a file backed VBD. I used the iso image to first launch and install OpenSolaris into the file VBD. Once that process is complete, I''m trying to run it using the config file below: name = ''opensolaris'' vcpus = 1 memory = "512" vif =

Hi, I am using KVM hypervisor on Ubuntu 12.04 (libvirt-0.9.8). The call to list the interfaces seem to be failing on this version of libvirt as seen below. virsh # iface-list error: Failed to list active interfaces error: this function is not supported by the connection driver: virConnectNumOfInterfaces I have seen this issue with libvirt 0.9.2 as well, which seemed to be working in 0.9.4

Too obvious! Thanks Dan On Apr 13, 2009, at 7:58 PM, John Mosby wrote: > po_iterator (ADT/PostOrderIterator.h) ? > > On Mon, Apr 13, 2009 at 5:40 PM, Daniel M Gessel <gessel at apple.com> > wrote: > That looks like it does a preorder depth first traversal (I think). > I'm looking for postorder. Is there a trivial transform between the > two? (I don't know

---------- Forwarded message ---------- From: Ryan Taylor <ryta1203 at gmail.com> Date: Mon, Nov 21, 2011 at 10:30 AM Subject: Re: [LLVMdev] Order of Basic Blocks To: Benjamin Kramer <benny.kra at googlemail.com> This worked, though the RPO_iterator apparently wasn't what I was looking for anyways, it seems it doesn't rreally go top->down. I have a simple example code,

Chris, attached you will find a patch implementing the approach I've posted yesterday. I don't want the topic fall by the wayside*. *Note that this patch here contains the df_iterator changes only, some (IMHO non-critical but needed) po_iterator changes are still in the pipeline and posted by me in another thread. I hope, the current approach fits all needs. Chris Lattner schrieb:

That looks like it does a preorder depth first traversal (I think). I'm looking for postorder. Is there a trivial transform between the two? (I don't know one.) Am I wrong about the preorder/postorder? If not, anything lurking for postorder traversal? Thanks, Dan On Apr 13, 2009, at 2:35 PM, Chris Lattner wrote: > > On Apr 13, 2009, at 11:20 AM, Daniel M Gessel wrote:

Just adding a few more details to the previous mail so that I can avoid some trouble for you folks. 1. I have done a /execname != "dtrace"/ to avoid dtrace itself causing more syscalls to be fired. 2. I have also supressed the printing on the terminal which makes dtrace slower. And what happens is this: dtrace -qn ''syscall::read:entry /execname != "dtrace"/ {

On Jun 15, 2009, at 4:33 AM, Olaf Krzikalla wrote: > While trying to eleminate as much std::tr1::function as possible I > stumbled over a design flaw in llvm::df_iterator. Ok. > However if fn replaces childrens of a just processed statement > (which happens a lot), the iteration may crash. Looking at > df_iterator reveals the reason: the first child of a particular >

Hi @llvm, below is a copy of a message I sent to clang a hour before. I guess it's more appropriate here. --snip-- While trying to eleminate as much std::tr1::function as possible I stumbled over a design flaw in llvm::df_iterator. Consider the following code: void for_all_stmts(Stmt* S, const std::tr1::function<void(Stmt*)>& fn) { if (S) { fn(S); for

Here is an updated patch for shrink wrapping with: - spills/restores done with stack slot stores/loads - stack adjustment removed - refactoring (but still in need of more) - spill/restore insertion code unified with spill/restore placement code Documentation available here<http://wiki.github.com/jdmdj/llvm-work/shrink-wrapping-work> illustrates shrink wrapping with loops and discusses a

Sorry, forgot to add group to CC. On Sun, Nov 20, 2011 at 6:14 PM, Ryan Taylor <ryta1203 at gmail.com> wrote: > Cameron, > > To make it less vague, I would ideally like to traverse top down. I > believe what you suggested is what I want, I will have a look at it. > > Currently, I am iterating over the BBs in a Function, so > Function::iterator BBitr=F->begin(),

po_iterator (ADT/PostOrderIterator.h) ? On Mon, Apr 13, 2009 at 5:40 PM, Daniel M Gessel <gessel at apple.com> wrote: > That looks like it does a preorder depth first traversal (I think). > I'm looking for postorder. Is there a trivial transform between the > two? (I don't know one.) > > Am I wrong about the preorder/postorder? If not, anything lurking for >