similar to: [LLVMdev] Matching icmp/fcmp in a back-end

Hi all, I am trying to write a back-end for LLVM where any instruction may take any type of data. I am looking for the output to be of the format: inst.type reg1,reg2 etc. Where inst is the instruction, e.g. mov and type is data-type e.g. f32 etc. I tried creating a back-end with a register class which could take i32 and f32: def GPRegs : RegisterClass <"Test",

Hi all, I was wondering if it is possible to stop floating-point constants being converted to use the constant pool? As for our back-end we would like to be able to treat floating point constants the same way integer constants are treated instead of having to go via the constant pool. Thanks for your help, Rob - This message is subject to Imagination Technologies' e-mail terms:

Hi, I was wondering if it was possible to detect if a function is recursive in a back-end. For instance, I'd like to be able to say: "If this function we are about to call is recursive, store the return address to the stack, if it isn't we don't need a stack so do nothing". Does anyone know if this is possible? Thanks, Rob - This message is subject to Imagination

I know, it's a strange question but I want install Wine on windows Why? Because of compatibility, I use XP and I want use win 3.x applications, and vista applications. I think wine it's a possibility, I use it on linux but I don't want reboot all time. Je sais, c'est un question bizarre mais je veux installer Wine sous windows Pourquoi? ? cause se la compatibilit?, j'utilise

I have a program that sets up a server on UDP port 59; it runs with no problem under windows XP. Under wine it fails, the log shows "Permission denied" for the WS_bind function. Has anyone else seen this? Does anyone know how to work around this? trace:winsock:WSAStartup succeeded trace:winsock:WS_socket af=2 type=1 protocol=0 trace:winsock:WSASocketA af=2 type=1 protocol=0

> On Apr 29, 2015, at 2:33 PM, Matt Arsenault <arsenm2 at gmail.com> wrote: > >> On Apr 29, 2015, at 10:06 AM, Silviu Baranga <Silviu.Baranga at arm.com <mailto:Silviu.Baranga at arm.com>> wrote: >> >> Note that dividing by an integer constant should be a cheap operation >> compared to FP multiplication and comparison as this would get lowered to a

Hi, I'm trying expand the Float2Int pass in order to make it able to optimize expressions like f * x > y, where x and y are integers (we'll assume unsigned for simplicity) and f is a floating point constant. The optimization would convert the expression to something like: (a * x)/b > y where a and b are integers guessed by the compiler (currently using continued

On Jul 3, 2008, at 3:07 PM, Richard Osborne wrote: >> >> This seems to break the convention. It should be the responsibility >> of the caller to further legalize the results. >> >> Evan > That makes sense. In that case I believe > SelectionDAGLegalize::LegalizeSetCCOperands > should be legalizing the result. The description of this function > says

Evan Cheng wrote: > On Jul 3, 2008, at 3:07 PM, Richard Osborne wrote: > >>> This seems to break the convention. It should be the responsibility >>> of the caller to further legalize the results. >>> >>> Evan >>> >> That makes sense. In that case I believe >> SelectionDAGLegalize::LegalizeSetCCOperands >> should be

I am having a little trouble with the fcmp one instruction on doubles only. For ordered comparisons, the LLVM manual states that true should be returned iff neither operands is QNAN. ( http://llvm.org/docs/LangRef.html#i_fcmp) If I do fcmp one which includes one or both operands as a NaN, the result is expected to be 0 then. If I run the bitcode with lli (JIT off), no problem. If I use the

On Mon, Jun 12, 2017 at 7:30 PM, Langley, Robert <Robert.Langley at ventura.org > wrote: > As far as the peer status (and I now remember seeing this earlier) the > issue appears to be that the host name for gsaov07 is attempting to resolve > over the wrong network for gluster "ent...." and not "stor.local". > So, it may be as simple as removing gsaov07 as a

As far as the peer status (and I now remember seeing this earlier) the issue appears to be that the host name for gsaov07 is attempting to resolve over the wrong network for gluster "ent...." and not "stor.local". So, it may be as simple as removing gsaov07 as a peer, then probing over the correct network. I'll follow up with the Engine log. Sent using OWA for iPhone

On Jun 23, 2008, at 5:35 AM, Richard Osborne wrote: > I'm trying to write a backend for a target with no hardware floating > point support. I've added a single i32 register class. I'm wanting all > floating point operations to be lowered to library function calls. For > the most part LLVM seems to get this right. For example > > define double @div(double %a, double

I'm trying to write a backend for a target with no hardware floating point support. I've added a single i32 register class. I'm wanting all floating point operations to be lowered to library function calls. For the most part LLVM seems to get this right. For example define double @div(double %a, double %b) { %result = fdiv double %a, %b ret double %result } is expanded to a

On Mon, 12 Jun 2017 at 17:40, Langley, Robert <Robert.Langley at ventura.org> wrote: > Thank you for your response. There has been no change of IP addresses. And > I have tried restarting the glusterd service multiple times. > I am using fully qualified names with a DNS server that has forward and > reverse setup. > Something I had noticed is that, with oVirt, communication

On Jun 25, 2008, at 5:13 AM, Richard Osborne wrote: > Evan Cheng wrote: >> On Jun 23, 2008, at 5:35 AM, Richard Osborne wrote: >> >>> I'm trying to write a backend for a target with no hardware floating >>> point support. I've added a single i32 register class. I'm wanting >>> all >>> floating point operations to be lowered to

I have the following in extentions.conf: exten => 2111,1,Dial(SIP/2111 at gw1.langley) exten => 2111,2,Voicemail(u2111) exten => 2111,3,Hangup exten => 2111,100,Voicemail(b2111) exten => 2111,101,Hangup I have the following in sip.conf: ; Cisco 1750 [gw1.langley] type=friend host=172.16.17.1 context=default canreinvite=no Like the ATA, lots of stuff doesn't work on the 1750

Thank you for your response. There has been no change of IP addresses. And I have tried restarting the glusterd service multiple times. I am using fully qualified names with a DNS server that has forward and reverse setup. Something I had noticed is that, with oVirt, communication is being attempted over the VM network, rather than storage network. Which is not how the bricks are defined. Not sure

On Mon, Jun 12, 2017 at 6:41 PM, Atin Mukherjee <amukherj at redhat.com> wrote: > > On Mon, 12 Jun 2017 at 17:40, Langley, Robert <Robert.Langley at ventura.org> > wrote: > >> Thank you for your response. There has been no change of IP addresses. >> And I have tried restarting the glusterd service multiple times. >> I am using fully qualified names with a

Evan Cheng wrote: > On Jun 23, 2008, at 5:35 AM, Richard Osborne wrote: > >> I'm trying to write a backend for a target with no hardware floating >> point support. I've added a single i32 register class. I'm wanting all >> floating point operations to be lowered to library function calls. For >> the most part LLVM seems to get this right. For example