Displaying 20 results from an estimated 21 matches for "comiss".
Did you mean:
ucomiss
2013 Aug 29
1
[LLVMdev] Ordered / Unordered FP compare are not handled properly on X86
...ot;:
fcmp une == unordered not equal == A is NaN, B is NaN, or A != B
fcmp one == ordered and equal == A is not NaN, B is not NaN, and A != B
Both of these can be implemented with the x86 unordered comparison
instruction by performing certain tests on the EFLAGS result. But if
we used the comiss instruction (and exceptions weren't masked) then
we'd take an exception if either A or B was a NaN.
Look at the C file I've attached, where I start by unmasking
exceptions and then perform what's essentially an "one" comparison.
LLVM implements this with ucomiss, and gets...
2008 May 27
3
[LLVMdev] Float compare-for-equality and select optimization opportunity
...C syntax the
code looks like this:
float x, y;
int a, b, c
if(x == y) // Rotate the integers
{
int t;
t = a;
a = b;
b = c;
c = t;
}
This is the resulting x86 assembly code:
movss xmm0,dword ptr [ecx+4]
ucomiss xmm0,dword ptr [ecx+8]
sete al
setnp dl
test dl,al
mov edx,edi
cmovne edx,ecx
cmovne ecx,esi
cmovne esi,edi
While I'm pleasantly surprised that my branch does get turned into several
select operations as intended (cmov - condi...
2008 May 27
1
[LLVMdev] Float compare-for-equality andselect optimizationopportunity
...e online demo, and it uses FCMP_OEQ.
No, if you have:
x = NaN
y = NaN
then the comparison:
(x == y) is false.
Which is what your seeing from your first post and is the standard IEEE
expected behavior.
Why I expected your min/max question to be related, consider the flags of
'comiss, ucomiss, etc.' :
ZPC
unordered 111
greater than 000
less than 001
equal 100
Try the following C program with gcc, first with no options and then with
--ffinite-math-only (or --ffast-math)
-----------
#defi...
2008 May 27
1
[LLVMdev] Float compare-for-equality and select optimizationopportunity
Both ZF and PF will be set if unordered, so the code below is IEEE
correct...you want to generate 'fcmp ueq' instead of 'fcmp oqe'
This is the resulting x86 assembly code:
movss xmm0,dword ptr [ecx+4]
ucomiss xmm0,dword ptr [ecx+8]
sete al
setnp dl
test dl,al
mov edx,edi
cmovne edx,ecx
cmovne ecx,esi
cmovne esi,edi
While I'm pleasantly surprised that my branch does get turned into several
select operations as intended (cmov - condi...
2008 Mar 20
7
Download Problems with lustre 1.6.4.3
Hello,
I''ve tried just some minutes ago to download the sourcecode of lustre from
[1], but it failed with this error:
An error occurred while processing your request.
Reference #50.663007d9.1206002944.16ba5fe1
Does anybody have some hints how to get the lustre source code then? I need
this tarball for the debian packaging.
If someone could point me to an alternative download
2008 Dec 22
2
How can I avoid nested 'for' loops or quicken the process?
Hi All,
I'm still pretty new to using R - and I was hoping I might be able to get
some advice as to how to use 'apply' or a similar function instead of using
nested for loops.
Right now I have a script which uses nested for loops similar to this:
i <- 1
for(a in Alpha) { for (b in Beta) { for (c in Gamma) { for (d in Delta) {
for (e in Epsilon)
{
Output[i] <-
2008 May 27
0
[LLVMdev] Float compare-for-equality and select optimizationopportunity
...v] Float compare-for-equality and select
optimizationopportunity
Both ZF and PF will be set if unordered, so the code below is IEEE
correct...you want to generate 'fcmp ueq' instead of 'fcmp oqe'
This is the resulting x86 assembly code:
movss xmm0,dword ptr [ecx+4]
ucomiss xmm0,dword ptr [ecx+8]
sete al
setnp dl
test dl,al
mov edx,edi
cmovne edx,ecx
cmovne ecx,esi
cmovne esi,edi
While I'm pleasantly surprised that my branch does get turned into several
select operations as intended (cmov - condi...
2013 Aug 29
0
[LLVMdev] Ordered / Unordered FP compare are not handled properly on X86
But this is another case. LLVM IR distinguishes between ordered and unordered compare and X86 backend has appropriate instructions.
But during DAG selection we just lose this information and always generate unordered fcmp.
I.e. in case of ordered fcmp the vcomiss should be generated, and in case of unordered - vucomiss.
- Elena
-----Original Message-----
From: Dr D. Chisnall [mailto:dc552 at hermes.cam.ac.uk] On Behalf Of David Chisnall
Sent: Thursday, August 29, 2013 10:50
To: Tim Northover
Cc: Demikhovsky, Elena; llvmdev at cs.uiuc.edu
Subject: Re: [...
2013 Aug 29
2
[LLVMdev] Ordered / Unordered FP compare are not handled properly on X86
....p.northover at gmail.com> wrote:
> If so, a compare that used that instruction would have to become more
> like an "invoke" with a landingpad for the exception and so on,
> wouldn't it? The current fcmp can already distinguish between ordered
> and unordered, because ucomiss provides that information.
There are currently lots of limitations in the expressiveness of LLVM IR for floating point operations (e.g. distinguishing between trapping and non-trapping cases and representing the floating point environment). If anyone wants to fully implement the floating point pa...
2013 May 20
2
[LLVMdev] VCOMISS instruction in X86
Hi,
I'm looking at scalar and packed instructions in X86.
The instruction VCOMISS is scalar. May I remove SSEPackedSingle/SSEPackedDouble domain from it?
defm VUCOMISS : sse12_ord_cmp<0x2E, FR32, X86cmp, f32, f32mem, loadf32,
"ucomiss", SSEPackedSingle>, TB, VEX, VEX_LIG;
defm VUCOMISD : sse12_ord_cmp<0x2E, FR64, X86cmp,...
2019 Jan 22
2
dbtool --cross-ncs and undeletable errors..
...and dc01 - Used to be a 4-Dc setup but 02 and 03
are gone), I've noticed the following errors which I am unable to fix.. Any hints?
* Basic dbcheck is clean.
[root at dc00 ~]# samba-tool dbcheck
Checking 327 objects
Checked 327 objects (0 errors)
* Cross-NCS shows two errors related to a de-comissionned DC (dc02) and
cannot auto-fix this.. How do I fix those errors?
[root at dc00 ~]# samba-tool dbcheck --cross-ncs --fix --yes
Checking 3574 objects
ERROR: no target object found for GUID component for link fromServer in
object
CN=4b3f95b1-7774-42cf-8bc0-755c7d29f6cc,CN=LostAndFoundConfig,C...
2019 Jan 22
2
dbtool --cross-ncs and undeletable errors..
...following errors which I am unable
>> to fix.. Any hints?
>>
>> * Basic dbcheck is clean.
>>
>> [root at dc00 ~]# samba-tool dbcheck
>> Checking 327 objects
>> Checked 327 objects (0 errors)
>>
>> * Cross-NCS shows two errors related to a de-comissionned DC (dc02)
>> and cannot auto-fix this.. How do I fix those errors?
>>
>> [root at dc00 ~]# samba-tool dbcheck --cross-ncs --fix --yes
>> Checking 3574 objects
>> ERROR: no target object found for GUID component for link fromServer
>> in object
>> CN...
2009 Aug 26
4
Fw: app_swift issue
...web far-and-wide for a solution. I eventually contacted Darren Sessions -- who was maintaining that plug-in -- about a month ago. He was involved in another project and said he might be able get to it after a few weeks. But, since then, his website http://www.darrensessions.com/ has gone out of comission.
I think we may be on our own on this one!
Todd
----- Forwarded Message ----
From: Todd Fulton <todd.fulton at yahoo.com>
To: Todd Fulton <todd.fulton at yahoo.com>
Sent: Wednesday, August 26, 2009 10:59:34 AM
Subject: [asterisk-users] app_swift issue
Hello
I have installed...
2013 Aug 29
0
[LLVMdev] Ordered / Unordered FP compare are not handled properly on X86
On 29 August 2013 06:31, Demikhovsky, Elena <elena.demikhovsky at intel.com> wrote:
> Should I open a ticket for this?
I think he was saying this is intended behaviour. Isn't the difference
between ucomiss and comiss just whether an exception is raised for
NaN?
If so, a compare that used that instruction would have to become more
like an "invoke" with a landingpad for the exception and so on,
wouldn't it? The current fcmp can already distinguish between ordered
and unordered, because u...
2009 Mar 31
1
Intel Integrated Raid (iir) relevance
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
(It would be probably good idea to redirect this discussion to -stable@,
redirected)
Hi, Danny,
Danny Braniss wrote:
> It's no longer working (for me) under 7.2, and so far
> I am not getting any feedback, so since it seems that
> this particular hardware has reached EOL, I was wondering
> if,
> a) it's true,
> b) drop
2006 Jan 02
0
job: small image bank
...ide the country. There are
already 2 future phases set to happen - one for i18n and another for
adding e-commerce - for which there''ll be dedicated budgets for
programming. Also, there are plans to sell this image bank to one of
the blockbuster operations out there (corbis or getty), and a
comission deal might be worked out.
if you''re interested, please contact me at
email: trabalho at oliverbarnes dot com
msn messenger: oli_azevedo_bar at hotmail dot com
google talk: ol1barn3s
aim: oliver azevedo b
skype: digiberber
happy new year,
Oliver
2013 Aug 29
2
[LLVMdev] Ordered / Unordered FP compare are not handled properly on X86
...d / Unordered FP compare are not handled properly on X86
On Wed, Aug 28, 2013 at 2:16 AM, Demikhovsky, Elena <elena.demikhovsky at intel.com<mailto:elena.demikhovsky at intel.com>> wrote:
I found that there is no diff in code generator for Ordered / Unordered FP compare instructions.
FUCOMISS, FUCOMISD are generated in the both cases.
Yes. That's how fcmp is defined in LangRef.
-Eli
---------------------------------------------------------------------
Intel Israel (74) Limited
This e-mail and any attachments may contain confidential material for
the sole use of the intended recip...
2019 Jan 22
0
dbtool --cross-ncs and undeletable errors..
...d 03 are gone), I've noticed the following errors which I am unable
> to fix.. Any hints?
>
> * Basic dbcheck is clean.
>
> [root at dc00 ~]# samba-tool dbcheck
> Checking 327 objects
> Checked 327 objects (0 errors)
>
> * Cross-NCS shows two errors related to a de-comissionned DC (dc02)
> and cannot auto-fix this.. How do I fix those errors?
>
> [root at dc00 ~]# samba-tool dbcheck --cross-ncs --fix --yes
> Checking 3574 objects
> ERROR: no target object found for GUID component for link fromServer
> in object
> CN=4b3f95b1-7774-42cf-8bc0-755...
2019 Jan 22
0
dbtool --cross-ncs and undeletable errors..
...able to fix.. Any hints?
> >>
> >> * Basic dbcheck is clean.
> >>
> >> [root at dc00 ~]# samba-tool dbcheck
> >> Checking 327 objects
> >> Checked 327 objects (0 errors)
> >>
> >> * Cross-NCS shows two errors related to a de-comissionned DC (dc02)
> >> and cannot auto-fix this.. How do I fix those errors?
> >>
> >> [root at dc00 ~]# samba-tool dbcheck --cross-ncs --fix --yes
> >> Checking 3574 objects
> >> ERROR: no target object found for GUID component for link
> >> fro...
2006 Jan 23
9
Web Hosting Options?
Can someone recommend a good web host for ROR?
--
Posted via http://www.ruby-forum.com/.