llvm dev - Aug 2021 - Semantics of udef values in PHI instructions

Hi,

Consider the following LLVM IR snippet. (also attached)

the value of %unify.phi is undefined if the loop body (while.body9)  is
executed, otherwise it has some specific value (%c.1 is a defined value).

However when I execute this IR (using lli), the value of %c.2 is always the
value of %c.1, even if the loop is executed multiple times.

What is the reason for this behavior?  how undefined values are handled in
LLVM and is this behaviour architecture dependent (depending on how undef
is handled in Codegen) ?

Thanks
Charitha

[image: Screen Shot 2021-08-14 at 9.36.28 PM.png]
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20210814/dec5dd7c/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screen Shot 2021-08-14 at 9.36.28 PM.png
Type: image/png
Size: 101653 bytes
Desc: not available
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20210814/dec5dd7c/attachment-0001.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test.ll.rtf
Type: text/rtf
Size: 6508 bytes
Desc: not available
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20210814/dec5dd7c/attachment-0001.bin>

Undef means it can take any value (except poison). Therefore LLVM replaces
phi(undef, %x) with %x, as it can fix the value of undef. This is optimization
is common and it’s not architecture dependent.

 

This optimization is actually *wrong* in case %x is poison, as then we are
replacing undef with poison for one of the predecessors (which is not correct).
But LLVM is still inconsistent in the treatment of undef and we can’t fix it
straight away.

 

Nuno

 

From: Charitha Saumya
Sent: 15 August 2021 02:53
To: via Llvm-dev <llvm-dev at lists.llvm.org>
Subject: [llvm-dev] Semantics of udef values in PHI instructions

 

Hi,

 

Consider the following LLVM IR snippet. (also attached)

 

the value of %unify.phi is undefined if the loop body (while.body9)  is
executed, otherwise it has some specific value (%c.1 is a defined value).

 

However when I execute this IR (using lli), the value of %c.2 is always the
value of %c.1, even if the loop is executed multiple times.

 

What is the reason for this behavior?  how undefined values are handled in LLVM
and is this behaviour architecture dependent (depending on how undef is handled
in Codegen) ?

 

Thanks

Charitha

 



 

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20210816/0e775129/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.png
Type: image/png
Size: 207504 bytes
Desc: not available
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20210816/0e775129/attachment-0001.png>