similar to: [LLVMdev] [NVPTX] llc -march=nvptx64 -mcpu=sm_20 generates invalid zero align for device function params

Dear all, I'm attaching a patch that should fix the issue mentioned above. It simply makes the same check seen in the same file for global variables: emitPTXAddressSpace(PTy->getAddressSpace(), O); if (GVar->getAlignment() == 0) O << " .align " << (int) TD->getPrefTypeAlignment(ETy); else O << " .align " <<

I'm building this with llvm-c, and accessing these intrinsics via calling the intrinsic as if it were a function. class F_SREG<string OpStr, NVPTXRegClass regclassOut, Intrinsic IntOp> : NVPTXInst<(outs regclassOut:$dst), (ins), OpStr, [(set regclassOut:$dst, (IntOp))]>; def INT_PTX_SREG_TID_X : F_SREG<"mov.u32 \t$dst, %tid.x;",

Hi Timothy, I'm not sure what you mean by this working for other intrinsics, but in this case, I think you want the intrinsic name llvm.nvvm.read.ptx.sreg.tid.x. For me, this looks like: %x = call i32 @llvm.nvvm.read.ptx.sreg.tid.x() Pete On Fri, Mar 1, 2013 at 11:51 AM, Timothy Baldridge <tbaldridge at gmail.com> wrote: > I'm building this with llvm-c, and accessing these

The identifier INT_PTX_SREG_TID_X is the name of an instruction as the back-end sees it, and has very little to do with the name you should use in your IR. Your best bet is to look at the include/llvm/IR/IntrinsicsNVVM.td file and see the definitions for each intrinsic. Then, the name mapping is just: int_foo_bar -> llvm.foo.bar() int_ prefix becomes llvm., and all underscores turn into

Ok, as I said, the most precise way to figure out what's wrong is to emit LLVM IR first (use clang -emit-llvm ...) and check out how it differs from working examples, for instance, nvptx regression tests. ----- Original message ----- > I'm building this with llvm-c, and accessing these intrinsics via calling > the intrinsic as if it were a function. > > class F_SREG<string

I've written a compiler that outputs PTX code, the result seems fairly reasonable, but I'm not sure the intrinsics are getting compiled correctly. In addition, when I try load the module using CUDA, I get an error: CUDA_ERROR_NO_BINARY_FOR_GPU. I'm running this on a 2012 MBP with a 640M GPU. PTX Code (for a mandelbrot calculation): // // Generated by LLVM NVPTX Back-End //

Timothy, Those calls to compute grid intrinsics are definitely wrong. In ptx code they should end up into reading special registers, rather than function calls. Try to take some working example and figure out the LLVM IR differences between it and the result of your compiler. - D. ----- Original message ----- > I've written a compiler that outputs PTX code, the result seems fairly >

Hello, I am the developer of Clover, and so much activity about OpenCL these days is really exciting. Here is my point of view, mainly on Clover and how the projects could use each other. Clover is made in a way that allow a certain level of modularity. Although POCL would be very difficult to merge into Clover (or Clover into POCL), as these two projects are nearly exactly doing the same things

Hello, Before a couple of days ago I started /usr/local/sbin/cupsd manualy from console. When I press CTRL+C to interrupt a program, the system change runlevel and going to reboot. This was on FreeBSD V5.3, today I installed fresh new 5.4 but the problem is the same. cups-base-1.1.23.0_3 -- Todor Dragnev <todor.dragnev@gmail.com>

It's been a while and I finally had the time to look into this. What I did was to build a custom AliasAnalysis pass, as Chris suggested, that returns AliasAnalysis::Mod for values passed to the sample function in the sret spot, and NoModRef for all other values. I'm also returning AliasAnalysis::AccessesArguments in the pass' getModRefBehavior methods. However, I haven't been

Hi all, I'm currently building a DSL for a computer graphics project that is not unlike NVIDIA's Cg. I have an intrinsic with the following signature float4 sample(texture tex, float2 coords); that is translated to this LLVM IR code: declare void @"sample"(%float4* noalias nocapture sret, %texture, $float2) nounwind readnone The type float4 is basically an array of four

Consider the following C code: typedef __attribute__(( ext_vector_type(2) )) float float2; typedef __attribute__(( ext_vector_type(2) )) __attribute__(( aligned(4) )) float float2_align2; void foo(void) { const float * p; size_t offset; float2 tmp = *((float2_align2 *)(p+offset)); } When compiled with clang ‹emit-llvm I get: define void @foo() { entry: %p = alloca float*, align 4

For reasons unknown, any connections to localhost -- tcp, icmp, or udp -- are all originating from my external interface, rl0: $ telnet localhost 25 Trying ::1... Trying 127.0.0.1... telnet: connect to address 127.0.0.1: Can't assign requested address telnet: Unable to connect to remote host IPFW log: Jul 16 12:46:43 octo ipfw: 100 Accept TCP 192.168.1.119:1434 127.0.0.1:25 out via rl0

On Fri, Jul 18, 2008 at 6:45 AM, Benedict Gaster <benedict.gaster at amd.com> wrote: > Consider the following C code: > > typedef __attribute__(( ext_vector_type(2) )) float float2; > typedef __attribute__(( ext_vector_type(2) )) __attribute__(( aligned(4) )) AFAIK, the aligned attribute doesn't do anything on a typedef of anything other than a struct/union type in either

hi all, i got flooded by these msgs like 1000+ lines, any idea? my kernel is dated Nov-30 FreeBSD 4.9-stable # tail -f /var/log/messages Jan 18 19:43:23 xb /kernel: arp: 202.79.180.1 moved from 00:04:5a:49:eb:74 to 00:50:0f:4f:c0:00 on rl0 Jan 18 19:45:06 xb /kernel: arp: 202.79.180.1 moved from 00:50:0f:4f:c0:00 to 00:04:5a:49:eb:74 on rl0 Jan 18 19:45:18 xb /kernel: arp: 202.79.180.1 moved from

Hi Stephan, > intrinsic float4 sample(int tex, float2 tc); > > float4 main(int tex, float2 tc) > { > float4 x = sample(tex, tc); > return 0.0; > } without additional information it would be wrong to remove the call to sample because it might write to a global variable. > As you can see, the call to the sample function is still present, > although the actual value

Hi, I have a dial up box (1.4) and another as an dns server. The default route on 1.4 is for the dial out for the other hosts. If I want to establish an connection (http) from the dialout box I can''t establish it (Network unreachable). The other hosts are configured with an default gateway 192.168.1.4 and have no such problems. Routing tables Internet: Destination Gateway

I have a problem I can not identify. Two firewalls with OpenBSD 2.8 sshd version OpenSSH_2.3.0 I do a ssh root at 195.84.181.91 -v SSH Version OpenSSH_2.3.0, protocol versions 1.5/2.0. Compiled with SSL (0x0090581f). debug: Reading configuration data /etc/ssh_config debug: ssh_connect: getuid 0 geteuid 0 anon 0 debug: Connecting to 195.84.181.91 [195.84.181.91] port 22. debug: Allocated

Hi. We just opened a gaming center and have chosen to run a FreeBsd box for our firewall. IPFW is configured at it's very basic running natd through rl0 and allowing any to any connections from the lan to the outer world. Natd controls access to the lan. We have a 6.0 mb/s ADSL net connection for all the gaming clients to use, however if a gamer starts downloading a file, that file

Hi, i've got this problem with my jail and i'm abolutly lost as in the why of it. I previously posted this on comp.unix.bsd.freebsd.misc but i was advised to send here I was unable to find help on google :( To resume quick, when i'm in a jail, netstat doesn't work properly. Hopefully i have provided sufficient information for anyone willing to help me :p First of all, my system :