similar to: [LLVMdev] Metadata

On Jan 4, 2010, at 3:14 PM, David Greene wrote: > Is there some documentation about metadata and how to use it > somewhere? > The doxygen-generated stuff obviously doesn't specify things like how > to create custom metadata kinds and all that jazz. The metadata design and APIs are still evolving. I wouldn't recommend building anything on it just yet, but I'm hoping it

On Monday 04 January 2010 19:22, Chris Lattner wrote: > On Jan 4, 2010, at 3:14 PM, David Greene wrote: > > Is there some documentation about metadata and how to use it > > somewhere? > > The doxygen-generated stuff obviously doesn't specify things like how > > to create custom metadata kinds and all that jazz. > > The metadata design and APIs are still

The current specification of the behavior of the !nontemporal attribute in LLVM, and the __builtin_nontemporal_* functions in Clang, is rather spartan and underspecified. In effect, it says the following things: * Atomic !nontemporal has no defined semantics * !nontemporal may use special instructions to save cache bandwidth, such as "MOVNT" on x86. What is crucially lacking

________________________________ From: llvm-dev <llvm-dev-bounces at lists.llvm.org> on behalf of JF Bastien via llvm-dev <llvm-dev at lists.llvm.org> Sent: Tuesday, April 28, 2020 4:54 PM To: Cranmer, Joshua <joshua.cranmer at intel.com> Cc: llvm-dev at lists.llvm.org <llvm-dev at lists.llvm.org> Subject: Re: [llvm-dev] Nontemporal memory accesses and fences I see

Right now the MemSDNode keeps a volatile bit in the SubclassData to mark volatile memory operations. We have some changes we'd like to push back that adds a NonTemporal flag to MemSDNode to mark instructions where movnt (on x86) and other goodness can happen (we'll also add the TableGen patterns to properly select movnt). In our tree we simply added another flag to the MemSDNode

On Jul 31, 2009, at 11:26 AM, David Greene wrote: > Right now the MemSDNode keeps a volatile bit in the SubclassData to > mark > volatile memory operations. > > We have some changes we'd like to push back that adds a NonTemporal > flag > to MemSDNode to mark instructions where movnt (on x86) and other > goodness > can happen (we'll also add the TableGen

Devang's work on debug info prompted this, thoughts welcome: http://nondot.org/sabre/LLVMNotes/ExtensibleMetadata.txt -Chris

On Sep 11, 2009, at 9:57 AM, Chris Lattner wrote: > Devang's work on debug info prompted this, thoughts welcome: > http://nondot.org/sabre/LLVMNotes/ExtensibleMetadata.txt The document mentions "instructions" a lot. We'll want to be able to apply metadata to ConstantExprs as well at least, if not also Arguments (think noalias) and other stuff, so it seems best to just

Dan Gohman wrote: > On Sep 11, 2009, at 9:57 AM, Chris Lattner wrote: > > >> Devang's work on debug info prompted this, thoughts welcome: >> http://nondot.org/sabre/LLVMNotes/ExtensibleMetadata.txt > > The document mentions "instructions" a lot. We'll want to be able to > apply metadata to ConstantExprs as well at least, if not also Arguments >

On Thursday 11 February 2010 14:05:21 David Greene wrote: > Either ParseLoad and probably other instructions need to look for metadata > explicitly or ParseOptionalCommaAlign needs to know about general metadata. > > My inkling is to fix ParseOptionalCommaAlign. Sound reasonable? Well, that's a rat's nest. I backed up and thought maybe I have the metadata syntax wrong. So

On Feb 11, 2010, at 12:50 PM, David Greene wrote: > On Thursday 11 February 2010 14:05:21 David Greene wrote: > >> Either ParseLoad and probably other instructions need to look for metadata >> explicitly or ParseOptionalCommaAlign needs to know about general metadata. >> >> My inkling is to fix ParseOptionalCommaAlign. Sound reasonable? > > Well, that's

On Sep 11, 2009, at 7:17 PM, Nick Lewycky wrote: > Dan Gohman wrote: >> On Sep 11, 2009, at 9:57 AM, Chris Lattner wrote: >>> Devang's work on debug info prompted this, thoughts welcome: >>> http://nondot.org/sabre/LLVMNotes/ExtensibleMetadata.txt >> The document mentions "instructions" a lot. We'll want to be able to >> apply metadata to

Chris Lattner wrote: > > On Sep 11, 2009, at 7:17 PM, Nick Lewycky wrote: > >> Dan Gohman wrote: >>> On Sep 11, 2009, at 9:57 AM, Chris Lattner wrote: >>>> Devang's work on debug info prompted this, thoughts welcome: >>>> http://nondot.org/sabre/LLVMNotes/ExtensibleMetadata.txt >>> The document mentions "instructions" a lot.

On Fri, Sep 11, 2009 at 9:57 AM, Chris Lattner <clattner at apple.com> wrote: > Devang's work on debug info prompted this, thoughts welcome: > http://nondot.org/sabre/LLVMNotes/ExtensibleMetadata.txt Here is partial initial implementation. Thoughts ? - Devang -------------- next part -------------- A non-text attachment was scrubbed... Name: mdn.diff Type: application/octet-stream

On Sep 11, 2009, at 2:11 PM, Dan Gohman wrote: > > On Sep 11, 2009, at 9:57 AM, Chris Lattner wrote: > > >> Devang's work on debug info prompted this, thoughts welcome: >> http://nondot.org/sabre/LLVMNotes/ExtensibleMetadata.txt > > The document mentions "instructions" a lot. We'll want to be able to > apply metadata to ConstantExprs as well at

On Thu, Feb 11, 2010 at 8:50 PM, David Greene <dag at cray.com> wrote: > > So I tried a bunch of things: > > %r8 = load <2 x double>* %r6, align 16, metadata !"nontemporal" > %r8 = load <2 x double>* %r6, align 16, metadata !nontemporal > %r8 = load <2 x double>* %r6, align 16, !{ metadata !"nontemporal" } > %r8 = load <2 x

I also want a way to add target specific flag to a SDNode (which should be transferred to MachineInstr). For example, on x86 lots of opcodes have *lock* variants. Right now, these are separate instructions. I'd prefer to make it into a target specific flag that can be toggled by some sort of post-isel action routine. One way to handle this might be to expand the use of SubclassData.

On Tue, Jan 12, 2016 at 09:20:06AM -0800, Linus Torvalds wrote: > On Tue, Jan 12, 2016 at 5:57 AM, Michael S. Tsirkin <mst at redhat.com> wrote: > > #ifdef xchgrz > > /* same as xchg but poking at gcc red zone */ > > #define barrier() do { int ret; asm volatile ("xchgl %0, -4(%%" SP ");": "=r"(ret) :: "memory", "cc"); }

On Tue, Jan 12, 2016 at 09:20:06AM -0800, Linus Torvalds wrote: > On Tue, Jan 12, 2016 at 5:57 AM, Michael S. Tsirkin <mst at redhat.com> wrote: > > #ifdef xchgrz > > /* same as xchg but poking at gcc red zone */ > > #define barrier() do { int ret; asm volatile ("xchgl %0, -4(%%" SP ");": "=r"(ret) :: "memory", "cc"); }

On Tue, Sep 15, 2009 at 11:37 PM, Devang Patel <devang.patel at gmail.com> wrote: > On Fri, Sep 11, 2009 at 9:57 AM, Chris Lattner <clattner at apple.com> wrote: >> Devang's work on debug info prompted this, thoughts welcome: >> http://nondot.org/sabre/LLVMNotes/ExtensibleMetadata.txt > > Here is partial initial implementation. > Thoughts ? setHasMetadata