Resolve call mdtokens when making tier 1 inline observations by tannergooding · Pull Request #50675 · dotnet/runtime

tannergooding · 2021-04-03T01:29:47Z

This updates Compiler::fgFindJumpTargets to support resolving call tokens when making inline observations for Tier 1.

A test of csc compiling a simple HelloWorld program (csc.dll Program.cs /r:System.Runtime.dll /r:System.Console.dll /r:System.Private.Corelib.dll) with COMPlus_ReadyToRun=0 and COMPlus_TieredCompilation=0 resulted in the following impact:

Compiler::fgFindJumpTarget goes from 0.48% to 0.63% of instructions retired (28.5m to 39.25m)
CEEInfo::resolveToken goes from 0.32% to 0.34% of instructions retired (19m to 21m)
CEEInfo::getCallInfo goes from 0.36% to 0.32% of instructions retired (21.25m to 19.75m)

As per the flame graph (see below), the majority of time is still spent in the backend. There was no measured impact for tier 0 methods.

This change will allow us to make potential improvements to inlining for methods that involve many intrinsic calls. Some potential metrics we could use to influence inlining include:

How many calls are intrinsic (Named or Otherwise)
How many calls are hwintrinsic and therefore are not calls at all, but are effectively "simple ops"
How many calls are CSE or constant folding candidates (such as Math calls)
How many calls are "constants" like Isa.IsSupported
How many calls are potential constants, like string.Length or Type.op_Equality checks

We can place this behind a DOTNET_* (COMPlus_*) switch if there are any remaining concerns around impact.

…argets when in Tier1 making inline observations

…True/False as constants in fgFindJumpTargets

tannergooding · 2021-04-03T01:31:50Z

This change produced zero PMI diffs for frameworks, benchmarks and tests. #50426 contains more details that led up to this PR.

CC. @jkotas, @AndyAyersMS

tannergooding · 2021-04-05T16:32:34Z

Any other feedback/changes needed here?

AndyAyersMS

No -- looks good.

…shim_mono # By Aaron Robinson (10) and others # Via GitHub * upstream/main: (108 commits) [mbr] Add Apple sample (dotnet#50740) make EstablishProxyTunnelAsync throw on failure status code from proxy (dotnet#50763) Improve RGB Min Max evaluation performance by using 2 or 3 comparison… (dotnet#50622) [mono] More domain cleanups (dotnet#50479) Fix Crossgen2 of PlatformDefaultMemberFunction methods and calls. (dotnet#50754) Disable EventSource generator in design-time builds (dotnet#50741) Fix X509 test failures on Android (dotnet#50301) Do not confuse fgDispBasicBlocks in fgMorphBlocks (dotnet#50703) Enforce 64KB event payload size limit on EventPipe (dotnet#50600) Reorganize CoreCLR native build to reduce CMake reconfigures when the build system is untouched (dotnet#49906) [mbr] Turn on hot reload for iOS, tvOS and MacCatalyst (dotnet#50458) improve connection scavenge logic by doing zero-byte read (dotnet#50545) Resolve call mdtokens when making tier 1 inline observations (dotnet#50675) Annotate APIs in System.Private.Xml (dotnet#49682) Support compiling against OpenSSL 3 headers Change Configuration.Json to use a regular Dictionary. (dotnet#50611) Remove unused BigNumFromBinary P/Invoke (dotnet#50670) Make Ninja the default CMake generator on Windows for the repo (dotnet#49715) [AppleAppBuilder] Entitlements to run tests on catalyst using the JIT (dotnet#50637) [mono] Fix delegate invokes to dynamic methods in mixed mode. (dotnet#50547) ... # Conflicts: # src/mono/dlls/mscordbi/CMakeLists.txt

tannergooding · 2021-04-03T01:29:47Z

This updates Compiler::fgFindJumpTargets to support resolving call tokens when making inline observations for Tier 1.

A test of csc compiling a simple HelloWorld program (csc.dll Program.cs /r:System.Runtime.dll /r:System.Console.dll /r:System.Private.Corelib.dll) with COMPlus_ReadyToRun=0 and COMPlus_TieredCompilation=0 resulted in the following impact:

Compiler::fgFindJumpTarget goes from 0.48% to 0.63% of instructions retired (28.5m to 39.25m)
CEEInfo::resolveToken goes from 0.32% to 0.34% of instructions retired (19m to 21m)
CEEInfo::getCallInfo goes from 0.36% to 0.32% of instructions retired (21.25m to 19.75m)

As per the flame graph (see below), the majority of time is still spent in the backend. There was no measured impact for tier 0 methods.

This change will allow us to make potential improvements to inlining for methods that involve many intrinsic calls. Some potential metrics we could use to influence inlining include:

How many calls are intrinsic (Named or Otherwise)
How many calls are hwintrinsic and therefore are not calls at all, but are effectively "simple ops"
How many calls are CSE or constant folding candidates (such as Math calls)
How many calls are "constants" like Isa.IsSupported
How many calls are potential constants, like string.Length or Type.op_Equality checks

We can place this behind a DOTNET_* (COMPlus_*) switch if there are any remaining concerns around impact.

…argets when in Tier1 making inline observations

…True/False as constants in fgFindJumpTargets

tannergooding · 2021-04-03T01:31:50Z

This change produced zero PMI diffs for frameworks, benchmarks and tests. #50426 contains more details that led up to this PR.

CC. @jkotas, @AndyAyersMS

tannergooding · 2021-04-05T16:32:34Z

Any other feedback/changes needed here?

AndyAyersMS

No -- looks good.

…shim_mono # By Aaron Robinson (10) and others # Via GitHub * upstream/main: (108 commits) [mbr] Add Apple sample (dotnet#50740) make EstablishProxyTunnelAsync throw on failure status code from proxy (dotnet#50763) Improve RGB Min Max evaluation performance by using 2 or 3 comparison… (dotnet#50622) [mono] More domain cleanups (dotnet#50479) Fix Crossgen2 of PlatformDefaultMemberFunction methods and calls. (dotnet#50754) Disable EventSource generator in design-time builds (dotnet#50741) Fix X509 test failures on Android (dotnet#50301) Do not confuse fgDispBasicBlocks in fgMorphBlocks (dotnet#50703) Enforce 64KB event payload size limit on EventPipe (dotnet#50600) Reorganize CoreCLR native build to reduce CMake reconfigures when the build system is untouched (dotnet#49906) [mbr] Turn on hot reload for iOS, tvOS and MacCatalyst (dotnet#50458) improve connection scavenge logic by doing zero-byte read (dotnet#50545) Resolve call mdtokens when making tier 1 inline observations (dotnet#50675) Annotate APIs in System.Private.Xml (dotnet#49682) Support compiling against OpenSSL 3 headers Change Configuration.Json to use a regular Dictionary. (dotnet#50611) Remove unused BigNumFromBinary P/Invoke (dotnet#50670) Make Ninja the default CMake generator on Windows for the repo (dotnet#49715) [AppleAppBuilder] Entitlements to run tests on catalyst using the JIT (dotnet#50637) [mono] Fix delegate invokes to dynamic methods in mixed mode. (dotnet#50547) ... # Conflicts: # src/mono/dlls/mscordbi/CMakeLists.txt

tannergooding · 2021-04-03T01:29:47Z

This updates Compiler::fgFindJumpTargets to support resolving call tokens when making inline observations for Tier 1.

A test of csc compiling a simple HelloWorld program (csc.dll Program.cs /r:System.Runtime.dll /r:System.Console.dll /r:System.Private.Corelib.dll) with COMPlus_ReadyToRun=0 and COMPlus_TieredCompilation=0 resulted in the following impact:

Compiler::fgFindJumpTarget goes from 0.48% to 0.63% of instructions retired (28.5m to 39.25m)
CEEInfo::resolveToken goes from 0.32% to 0.34% of instructions retired (19m to 21m)
CEEInfo::getCallInfo goes from 0.36% to 0.32% of instructions retired (21.25m to 19.75m)

As per the flame graph (see below), the majority of time is still spent in the backend. There was no measured impact for tier 0 methods.

This change will allow us to make potential improvements to inlining for methods that involve many intrinsic calls. Some potential metrics we could use to influence inlining include:

How many calls are intrinsic (Named or Otherwise)
How many calls are hwintrinsic and therefore are not calls at all, but are effectively "simple ops"
How many calls are CSE or constant folding candidates (such as Math calls)
How many calls are "constants" like Isa.IsSupported
How many calls are potential constants, like string.Length or Type.op_Equality checks

We can place this behind a DOTNET_* (COMPlus_*) switch if there are any remaining concerns around impact.

…argets when in Tier1 making inline observations

…True/False as constants in fgFindJumpTargets

tannergooding · 2021-04-03T01:31:50Z

This change produced zero PMI diffs for frameworks, benchmarks and tests. #50426 contains more details that led up to this PR.

CC. @jkotas, @AndyAyersMS

tannergooding · 2021-04-05T16:32:34Z

Any other feedback/changes needed here?

AndyAyersMS

No -- looks good.

…shim_mono # By Aaron Robinson (10) and others # Via GitHub * upstream/main: (108 commits) [mbr] Add Apple sample (dotnet#50740) make EstablishProxyTunnelAsync throw on failure status code from proxy (dotnet#50763) Improve RGB Min Max evaluation performance by using 2 or 3 comparison… (dotnet#50622) [mono] More domain cleanups (dotnet#50479) Fix Crossgen2 of PlatformDefaultMemberFunction methods and calls. (dotnet#50754) Disable EventSource generator in design-time builds (dotnet#50741) Fix X509 test failures on Android (dotnet#50301) Do not confuse fgDispBasicBlocks in fgMorphBlocks (dotnet#50703) Enforce 64KB event payload size limit on EventPipe (dotnet#50600) Reorganize CoreCLR native build to reduce CMake reconfigures when the build system is untouched (dotnet#49906) [mbr] Turn on hot reload for iOS, tvOS and MacCatalyst (dotnet#50458) improve connection scavenge logic by doing zero-byte read (dotnet#50545) Resolve call mdtokens when making tier 1 inline observations (dotnet#50675) Annotate APIs in System.Private.Xml (dotnet#49682) Support compiling against OpenSSL 3 headers Change Configuration.Json to use a regular Dictionary. (dotnet#50611) Remove unused BigNumFromBinary P/Invoke (dotnet#50670) Make Ninja the default CMake generator on Windows for the repo (dotnet#49715) [AppleAppBuilder] Entitlements to run tests on catalyst using the JIT (dotnet#50637) [mono] Fix delegate invokes to dynamic methods in mixed mode. (dotnet#50547) ... # Conflicts: # src/mono/dlls/mscordbi/CMakeLists.txt

tannergooding · 2021-04-03T01:29:47Z

This updates Compiler::fgFindJumpTargets to support resolving call tokens when making inline observations for Tier 1.

A test of csc compiling a simple HelloWorld program (csc.dll Program.cs /r:System.Runtime.dll /r:System.Console.dll /r:System.Private.Corelib.dll) with COMPlus_ReadyToRun=0 and COMPlus_TieredCompilation=0 resulted in the following impact:

Compiler::fgFindJumpTarget goes from 0.48% to 0.63% of instructions retired (28.5m to 39.25m)
CEEInfo::resolveToken goes from 0.32% to 0.34% of instructions retired (19m to 21m)
CEEInfo::getCallInfo goes from 0.36% to 0.32% of instructions retired (21.25m to 19.75m)

As per the flame graph (see below), the majority of time is still spent in the backend. There was no measured impact for tier 0 methods.

This change will allow us to make potential improvements to inlining for methods that involve many intrinsic calls. Some potential metrics we could use to influence inlining include:

How many calls are intrinsic (Named or Otherwise)
How many calls are hwintrinsic and therefore are not calls at all, but are effectively "simple ops"
How many calls are CSE or constant folding candidates (such as Math calls)
How many calls are "constants" like Isa.IsSupported
How many calls are potential constants, like string.Length or Type.op_Equality checks

We can place this behind a DOTNET_* (COMPlus_*) switch if there are any remaining concerns around impact.

…argets when in Tier1 making inline observations

…True/False as constants in fgFindJumpTargets

tannergooding · 2021-04-03T01:31:50Z

This change produced zero PMI diffs for frameworks, benchmarks and tests. #50426 contains more details that led up to this PR.

CC. @jkotas, @AndyAyersMS

tannergooding · 2021-04-05T16:32:34Z

Any other feedback/changes needed here?

AndyAyersMS

No -- looks good.

…shim_mono # By Aaron Robinson (10) and others # Via GitHub * upstream/main: (108 commits) [mbr] Add Apple sample (dotnet#50740) make EstablishProxyTunnelAsync throw on failure status code from proxy (dotnet#50763) Improve RGB Min Max evaluation performance by using 2 or 3 comparison… (dotnet#50622) [mono] More domain cleanups (dotnet#50479) Fix Crossgen2 of PlatformDefaultMemberFunction methods and calls. (dotnet#50754) Disable EventSource generator in design-time builds (dotnet#50741) Fix X509 test failures on Android (dotnet#50301) Do not confuse fgDispBasicBlocks in fgMorphBlocks (dotnet#50703) Enforce 64KB event payload size limit on EventPipe (dotnet#50600) Reorganize CoreCLR native build to reduce CMake reconfigures when the build system is untouched (dotnet#49906) [mbr] Turn on hot reload for iOS, tvOS and MacCatalyst (dotnet#50458) improve connection scavenge logic by doing zero-byte read (dotnet#50545) Resolve call mdtokens when making tier 1 inline observations (dotnet#50675) Annotate APIs in System.Private.Xml (dotnet#49682) Support compiling against OpenSSL 3 headers Change Configuration.Json to use a regular Dictionary. (dotnet#50611) Remove unused BigNumFromBinary P/Invoke (dotnet#50670) Make Ninja the default CMake generator on Windows for the repo (dotnet#49715) [AppleAppBuilder] Entitlements to run tests on catalyst using the JIT (dotnet#50637) [mono] Fix delegate invokes to dynamic methods in mixed mode. (dotnet#50547) ... # Conflicts: # src/mono/dlls/mscordbi/CMakeLists.txt

tannergooding · 2021-04-03T01:29:47Z

This updates Compiler::fgFindJumpTargets to support resolving call tokens when making inline observations for Tier 1.

A test of csc compiling a simple HelloWorld program (csc.dll Program.cs /r:System.Runtime.dll /r:System.Console.dll /r:System.Private.Corelib.dll) with COMPlus_ReadyToRun=0 and COMPlus_TieredCompilation=0 resulted in the following impact:

Compiler::fgFindJumpTarget goes from 0.48% to 0.63% of instructions retired (28.5m to 39.25m)
CEEInfo::resolveToken goes from 0.32% to 0.34% of instructions retired (19m to 21m)
CEEInfo::getCallInfo goes from 0.36% to 0.32% of instructions retired (21.25m to 19.75m)

As per the flame graph (see below), the majority of time is still spent in the backend. There was no measured impact for tier 0 methods.

This change will allow us to make potential improvements to inlining for methods that involve many intrinsic calls. Some potential metrics we could use to influence inlining include:

How many calls are intrinsic (Named or Otherwise)
How many calls are hwintrinsic and therefore are not calls at all, but are effectively "simple ops"
How many calls are CSE or constant folding candidates (such as Math calls)
How many calls are "constants" like Isa.IsSupported
How many calls are potential constants, like string.Length or Type.op_Equality checks

We can place this behind a DOTNET_* (COMPlus_*) switch if there are any remaining concerns around impact.

…argets when in Tier1 making inline observations

…True/False as constants in fgFindJumpTargets

tannergooding · 2021-04-03T01:31:50Z

This change produced zero PMI diffs for frameworks, benchmarks and tests. #50426 contains more details that led up to this PR.

CC. @jkotas, @AndyAyersMS

tannergooding · 2021-04-05T16:32:34Z

Any other feedback/changes needed here?

AndyAyersMS

No -- looks good.

…shim_mono # By Aaron Robinson (10) and others # Via GitHub * upstream/main: (108 commits) [mbr] Add Apple sample (dotnet#50740) make EstablishProxyTunnelAsync throw on failure status code from proxy (dotnet#50763) Improve RGB Min Max evaluation performance by using 2 or 3 comparison… (dotnet#50622) [mono] More domain cleanups (dotnet#50479) Fix Crossgen2 of PlatformDefaultMemberFunction methods and calls. (dotnet#50754) Disable EventSource generator in design-time builds (dotnet#50741) Fix X509 test failures on Android (dotnet#50301) Do not confuse fgDispBasicBlocks in fgMorphBlocks (dotnet#50703) Enforce 64KB event payload size limit on EventPipe (dotnet#50600) Reorganize CoreCLR native build to reduce CMake reconfigures when the build system is untouched (dotnet#49906) [mbr] Turn on hot reload for iOS, tvOS and MacCatalyst (dotnet#50458) improve connection scavenge logic by doing zero-byte read (dotnet#50545) Resolve call mdtokens when making tier 1 inline observations (dotnet#50675) Annotate APIs in System.Private.Xml (dotnet#49682) Support compiling against OpenSSL 3 headers Change Configuration.Json to use a regular Dictionary. (dotnet#50611) Remove unused BigNumFromBinary P/Invoke (dotnet#50670) Make Ninja the default CMake generator on Windows for the repo (dotnet#49715) [AppleAppBuilder] Entitlements to run tests on catalyst using the JIT (dotnet#50637) [mono] Fix delegate invokes to dynamic methods in mixed mode. (dotnet#50547) ... # Conflicts: # src/mono/dlls/mscordbi/CMakeLists.txt

tannergooding · 2021-04-03T01:29:47Z

This updates Compiler::fgFindJumpTargets to support resolving call tokens when making inline observations for Tier 1.

A test of csc compiling a simple HelloWorld program (csc.dll Program.cs /r:System.Runtime.dll /r:System.Console.dll /r:System.Private.Corelib.dll) with COMPlus_ReadyToRun=0 and COMPlus_TieredCompilation=0 resulted in the following impact:

Compiler::fgFindJumpTarget goes from 0.48% to 0.63% of instructions retired (28.5m to 39.25m)
CEEInfo::resolveToken goes from 0.32% to 0.34% of instructions retired (19m to 21m)
CEEInfo::getCallInfo goes from 0.36% to 0.32% of instructions retired (21.25m to 19.75m)

As per the flame graph (see below), the majority of time is still spent in the backend. There was no measured impact for tier 0 methods.

This change will allow us to make potential improvements to inlining for methods that involve many intrinsic calls. Some potential metrics we could use to influence inlining include:

How many calls are intrinsic (Named or Otherwise)
How many calls are hwintrinsic and therefore are not calls at all, but are effectively "simple ops"
How many calls are CSE or constant folding candidates (such as Math calls)
How many calls are "constants" like Isa.IsSupported
How many calls are potential constants, like string.Length or Type.op_Equality checks

We can place this behind a DOTNET_* (COMPlus_*) switch if there are any remaining concerns around impact.

…argets when in Tier1 making inline observations

…True/False as constants in fgFindJumpTargets

tannergooding · 2021-04-03T01:31:50Z

This change produced zero PMI diffs for frameworks, benchmarks and tests. #50426 contains more details that led up to this PR.

CC. @jkotas, @AndyAyersMS

tannergooding · 2021-04-05T16:32:34Z

Any other feedback/changes needed here?

AndyAyersMS

No -- looks good.

…shim_mono # By Aaron Robinson (10) and others # Via GitHub * upstream/main: (108 commits) [mbr] Add Apple sample (dotnet#50740) make EstablishProxyTunnelAsync throw on failure status code from proxy (dotnet#50763) Improve RGB Min Max evaluation performance by using 2 or 3 comparison… (dotnet#50622) [mono] More domain cleanups (dotnet#50479) Fix Crossgen2 of PlatformDefaultMemberFunction methods and calls. (dotnet#50754) Disable EventSource generator in design-time builds (dotnet#50741) Fix X509 test failures on Android (dotnet#50301) Do not confuse fgDispBasicBlocks in fgMorphBlocks (dotnet#50703) Enforce 64KB event payload size limit on EventPipe (dotnet#50600) Reorganize CoreCLR native build to reduce CMake reconfigures when the build system is untouched (dotnet#49906) [mbr] Turn on hot reload for iOS, tvOS and MacCatalyst (dotnet#50458) improve connection scavenge logic by doing zero-byte read (dotnet#50545) Resolve call mdtokens when making tier 1 inline observations (dotnet#50675) Annotate APIs in System.Private.Xml (dotnet#49682) Support compiling against OpenSSL 3 headers Change Configuration.Json to use a regular Dictionary. (dotnet#50611) Remove unused BigNumFromBinary P/Invoke (dotnet#50670) Make Ninja the default CMake generator on Windows for the repo (dotnet#49715) [AppleAppBuilder] Entitlements to run tests on catalyst using the JIT (dotnet#50637) [mono] Fix delegate invokes to dynamic methods in mixed mode. (dotnet#50547) ... # Conflicts: # src/mono/dlls/mscordbi/CMakeLists.txt

tannergooding · 2021-04-03T01:29:47Z

This updates Compiler::fgFindJumpTargets to support resolving call tokens when making inline observations for Tier 1.

A test of csc compiling a simple HelloWorld program (csc.dll Program.cs /r:System.Runtime.dll /r:System.Console.dll /r:System.Private.Corelib.dll) with COMPlus_ReadyToRun=0 and COMPlus_TieredCompilation=0 resulted in the following impact:

Compiler::fgFindJumpTarget goes from 0.48% to 0.63% of instructions retired (28.5m to 39.25m)
CEEInfo::resolveToken goes from 0.32% to 0.34% of instructions retired (19m to 21m)
CEEInfo::getCallInfo goes from 0.36% to 0.32% of instructions retired (21.25m to 19.75m)

As per the flame graph (see below), the majority of time is still spent in the backend. There was no measured impact for tier 0 methods.

This change will allow us to make potential improvements to inlining for methods that involve many intrinsic calls. Some potential metrics we could use to influence inlining include:

How many calls are intrinsic (Named or Otherwise)
How many calls are hwintrinsic and therefore are not calls at all, but are effectively "simple ops"
How many calls are CSE or constant folding candidates (such as Math calls)
How many calls are "constants" like Isa.IsSupported
How many calls are potential constants, like string.Length or Type.op_Equality checks

We can place this behind a DOTNET_* (COMPlus_*) switch if there are any remaining concerns around impact.

…argets when in Tier1 making inline observations

…True/False as constants in fgFindJumpTargets

tannergooding · 2021-04-03T01:31:50Z

This change produced zero PMI diffs for frameworks, benchmarks and tests. #50426 contains more details that led up to this PR.

CC. @jkotas, @AndyAyersMS

tannergooding · 2021-04-05T16:32:34Z

Any other feedback/changes needed here?

AndyAyersMS

No -- looks good.

…shim_mono # By Aaron Robinson (10) and others # Via GitHub * upstream/main: (108 commits) [mbr] Add Apple sample (dotnet#50740) make EstablishProxyTunnelAsync throw on failure status code from proxy (dotnet#50763) Improve RGB Min Max evaluation performance by using 2 or 3 comparison… (dotnet#50622) [mono] More domain cleanups (dotnet#50479) Fix Crossgen2 of PlatformDefaultMemberFunction methods and calls. (dotnet#50754) Disable EventSource generator in design-time builds (dotnet#50741) Fix X509 test failures on Android (dotnet#50301) Do not confuse fgDispBasicBlocks in fgMorphBlocks (dotnet#50703) Enforce 64KB event payload size limit on EventPipe (dotnet#50600) Reorganize CoreCLR native build to reduce CMake reconfigures when the build system is untouched (dotnet#49906) [mbr] Turn on hot reload for iOS, tvOS and MacCatalyst (dotnet#50458) improve connection scavenge logic by doing zero-byte read (dotnet#50545) Resolve call mdtokens when making tier 1 inline observations (dotnet#50675) Annotate APIs in System.Private.Xml (dotnet#49682) Support compiling against OpenSSL 3 headers Change Configuration.Json to use a regular Dictionary. (dotnet#50611) Remove unused BigNumFromBinary P/Invoke (dotnet#50670) Make Ninja the default CMake generator on Windows for the repo (dotnet#49715) [AppleAppBuilder] Entitlements to run tests on catalyst using the JIT (dotnet#50637) [mono] Fix delegate invokes to dynamic methods in mixed mode. (dotnet#50547) ... # Conflicts: # src/mono/dlls/mscordbi/CMakeLists.txt

tannergooding · 2021-04-03T01:29:47Z

This updates Compiler::fgFindJumpTargets to support resolving call tokens when making inline observations for Tier 1.

A test of csc compiling a simple HelloWorld program (csc.dll Program.cs /r:System.Runtime.dll /r:System.Console.dll /r:System.Private.Corelib.dll) with COMPlus_ReadyToRun=0 and COMPlus_TieredCompilation=0 resulted in the following impact:

Compiler::fgFindJumpTarget goes from 0.48% to 0.63% of instructions retired (28.5m to 39.25m)
CEEInfo::resolveToken goes from 0.32% to 0.34% of instructions retired (19m to 21m)
CEEInfo::getCallInfo goes from 0.36% to 0.32% of instructions retired (21.25m to 19.75m)

As per the flame graph (see below), the majority of time is still spent in the backend. There was no measured impact for tier 0 methods.

This change will allow us to make potential improvements to inlining for methods that involve many intrinsic calls. Some potential metrics we could use to influence inlining include:

How many calls are intrinsic (Named or Otherwise)
How many calls are hwintrinsic and therefore are not calls at all, but are effectively "simple ops"
How many calls are CSE or constant folding candidates (such as Math calls)
How many calls are "constants" like Isa.IsSupported
How many calls are potential constants, like string.Length or Type.op_Equality checks

We can place this behind a DOTNET_* (COMPlus_*) switch if there are any remaining concerns around impact.

…argets when in Tier1 making inline observations

…True/False as constants in fgFindJumpTargets

tannergooding · 2021-04-03T01:31:50Z

This change produced zero PMI diffs for frameworks, benchmarks and tests. #50426 contains more details that led up to this PR.

CC. @jkotas, @AndyAyersMS

tannergooding · 2021-04-05T16:32:34Z

Any other feedback/changes needed here?

AndyAyersMS

No -- looks good.

…shim_mono # By Aaron Robinson (10) and others # Via GitHub * upstream/main: (108 commits) [mbr] Add Apple sample (dotnet#50740) make EstablishProxyTunnelAsync throw on failure status code from proxy (dotnet#50763) Improve RGB Min Max evaluation performance by using 2 or 3 comparison… (dotnet#50622) [mono] More domain cleanups (dotnet#50479) Fix Crossgen2 of PlatformDefaultMemberFunction methods and calls. (dotnet#50754) Disable EventSource generator in design-time builds (dotnet#50741) Fix X509 test failures on Android (dotnet#50301) Do not confuse fgDispBasicBlocks in fgMorphBlocks (dotnet#50703) Enforce 64KB event payload size limit on EventPipe (dotnet#50600) Reorganize CoreCLR native build to reduce CMake reconfigures when the build system is untouched (dotnet#49906) [mbr] Turn on hot reload for iOS, tvOS and MacCatalyst (dotnet#50458) improve connection scavenge logic by doing zero-byte read (dotnet#50545) Resolve call mdtokens when making tier 1 inline observations (dotnet#50675) Annotate APIs in System.Private.Xml (dotnet#49682) Support compiling against OpenSSL 3 headers Change Configuration.Json to use a regular Dictionary. (dotnet#50611) Remove unused BigNumFromBinary P/Invoke (dotnet#50670) Make Ninja the default CMake generator on Windows for the repo (dotnet#49715) [AppleAppBuilder] Entitlements to run tests on catalyst using the JIT (dotnet#50637) [mono] Fix delegate invokes to dynamic methods in mixed mode. (dotnet#50547) ... # Conflicts: # src/mono/dlls/mscordbi/CMakeLists.txt

tannergooding added 2 commits April 1, 2021 20:54

Add some basic handling to allow resolving call tokens in fgFindJumpT…
95f10fc
…argets when in Tier1 making inline observations

Don't count prefix opcodes as instructions and handle NI_IsSupported_…
4d4cdc3
…True/False as constants in fgFindJumpTargets

tannergooding requested a review from AndyAyersMS April 3, 2021 01:29

ghost added the area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI label Apr 3, 2021

jkotas reviewed Apr 3, 2021
View reviewed changes

Comment threadsrc/coreclr/jit/fgbasic.cpp Outdated

tannergooding added 2 commits April 2, 2021 19:28

Update src/coreclr/jit/fgbasic.cpp
c633127

Applying formatting patch
9796b1a

AndyAyersMS approved these changes Apr 5, 2021
View reviewed changes

tannergooding merged commit e9e8bc7 into dotnet:mainApr 5, 2021

tannergooding mentioned this pull request Apr 5, 2021
Consider adding a cache of well known tokens to the JIT #50426
Closed

runfoappBot mentioned this pull request Apr 5, 2021
Helix DNS failures causing work item failures in CI #50750
Closed

EgorBo mentioned this pull request Apr 6, 2021
JIT: Improve inliner (constant feeds constant test, foldable calls) #50784
Closed

DrewScoggins mentioned this pull request Apr 9, 2021
Regression in Crossgen2 Throughput #51006
Closed

JulieLeeMSFT assigned tannergooding Apr 15, 2021

JulieLeeMSFT added this to the 6.0.0 milestone Apr 15, 2021

ghost locked as resolved and limited conversation to collaborators May 15, 2021

tannergooding deleted the resolve-tokens-inline branch July 1, 2025 14:39

tannergooding added 2 commits April 1, 2021 20:54

Add some basic handling to allow resolving call tokens in fgFindJumpT…
95f10fc
…argets when in Tier1 making inline observations

Don't count prefix opcodes as instructions and handle NI_IsSupported_…
4d4cdc3
…True/False as constants in fgFindJumpTargets

tannergooding requested a review from AndyAyersMS April 3, 2021 01:29

ghost added the area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI label Apr 3, 2021

jkotas reviewed Apr 3, 2021
View reviewed changes

Comment threadsrc/coreclr/jit/fgbasic.cpp Outdated

tannergooding added 2 commits April 2, 2021 19:28

Update src/coreclr/jit/fgbasic.cpp
c633127

Applying formatting patch
9796b1a

AndyAyersMS approved these changes Apr 5, 2021
View reviewed changes

tannergooding merged commit e9e8bc7 into dotnet:mainApr 5, 2021

tannergooding mentioned this pull request Apr 5, 2021
Consider adding a cache of well known tokens to the JIT #50426
Closed

runfoappBot mentioned this pull request Apr 5, 2021
Helix DNS failures causing work item failures in CI #50750
Closed

EgorBo mentioned this pull request Apr 6, 2021
JIT: Improve inliner (constant feeds constant test, foldable calls) #50784
Closed

DrewScoggins mentioned this pull request Apr 9, 2021
Regression in Crossgen2 Throughput #51006
Closed

JulieLeeMSFT assigned tannergooding Apr 15, 2021

JulieLeeMSFT added this to the 6.0.0 milestone Apr 15, 2021

ghost locked as resolved and limited conversation to collaborators May 15, 2021

tannergooding deleted the resolve-tokens-inline branch July 1, 2025 14:39

tannergooding added 2 commits April 1, 2021 20:54

Add some basic handling to allow resolving call tokens in fgFindJumpT…
95f10fc
…argets when in Tier1 making inline observations

Don't count prefix opcodes as instructions and handle NI_IsSupported_…
4d4cdc3
…True/False as constants in fgFindJumpTargets

tannergooding requested a review from AndyAyersMS April 3, 2021 01:29

ghost added the area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI label Apr 3, 2021

jkotas reviewed Apr 3, 2021
View reviewed changes

Comment threadsrc/coreclr/jit/fgbasic.cpp Outdated

tannergooding added 2 commits April 2, 2021 19:28

Update src/coreclr/jit/fgbasic.cpp
c633127

Applying formatting patch
9796b1a

AndyAyersMS approved these changes Apr 5, 2021
View reviewed changes

tannergooding merged commit e9e8bc7 into dotnet:mainApr 5, 2021

tannergooding mentioned this pull request Apr 5, 2021
Consider adding a cache of well known tokens to the JIT #50426
Closed

runfoappBot mentioned this pull request Apr 5, 2021
Helix DNS failures causing work item failures in CI #50750
Closed

EgorBo mentioned this pull request Apr 6, 2021
JIT: Improve inliner (constant feeds constant test, foldable calls) #50784
Closed

DrewScoggins mentioned this pull request Apr 9, 2021
Regression in Crossgen2 Throughput #51006
Closed

JulieLeeMSFT assigned tannergooding Apr 15, 2021

JulieLeeMSFT added this to the 6.0.0 milestone Apr 15, 2021

ghost locked as resolved and limited conversation to collaborators May 15, 2021

tannergooding deleted the resolve-tokens-inline branch July 1, 2025 14:39

tannergooding added 2 commits April 1, 2021 20:54

Add some basic handling to allow resolving call tokens in fgFindJumpT…
95f10fc
…argets when in Tier1 making inline observations

Don't count prefix opcodes as instructions and handle NI_IsSupported_…
4d4cdc3
…True/False as constants in fgFindJumpTargets

tannergooding requested a review from AndyAyersMS April 3, 2021 01:29

ghost added the area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI label Apr 3, 2021

jkotas reviewed Apr 3, 2021
View reviewed changes

Comment threadsrc/coreclr/jit/fgbasic.cpp Outdated

tannergooding added 2 commits April 2, 2021 19:28

Update src/coreclr/jit/fgbasic.cpp
c633127

Applying formatting patch
9796b1a

AndyAyersMS approved these changes Apr 5, 2021
View reviewed changes

tannergooding merged commit e9e8bc7 into dotnet:mainApr 5, 2021

tannergooding mentioned this pull request Apr 5, 2021
Consider adding a cache of well known tokens to the JIT #50426
Closed

runfoappBot mentioned this pull request Apr 5, 2021
Helix DNS failures causing work item failures in CI #50750
Closed

EgorBo mentioned this pull request Apr 6, 2021
JIT: Improve inliner (constant feeds constant test, foldable calls) #50784
Closed

DrewScoggins mentioned this pull request Apr 9, 2021
Regression in Crossgen2 Throughput #51006
Closed

JulieLeeMSFT assigned tannergooding Apr 15, 2021

JulieLeeMSFT added this to the 6.0.0 milestone Apr 15, 2021

ghost locked as resolved and limited conversation to collaborators May 15, 2021

tannergooding deleted the resolve-tokens-inline branch July 1, 2025 14:39

tannergooding added 2 commits April 1, 2021 20:54

Add some basic handling to allow resolving call tokens in fgFindJumpT…
95f10fc
…argets when in Tier1 making inline observations

Don't count prefix opcodes as instructions and handle NI_IsSupported_…
4d4cdc3
…True/False as constants in fgFindJumpTargets

tannergooding requested a review from AndyAyersMS April 3, 2021 01:29

ghost added the area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI label Apr 3, 2021

jkotas reviewed Apr 3, 2021
View reviewed changes

Comment threadsrc/coreclr/jit/fgbasic.cpp Outdated

tannergooding added 2 commits April 2, 2021 19:28

Update src/coreclr/jit/fgbasic.cpp
c633127

Applying formatting patch
9796b1a

AndyAyersMS approved these changes Apr 5, 2021
View reviewed changes

tannergooding merged commit e9e8bc7 into dotnet:mainApr 5, 2021

tannergooding mentioned this pull request Apr 5, 2021
Consider adding a cache of well known tokens to the JIT #50426
Closed

runfoappBot mentioned this pull request Apr 5, 2021
Helix DNS failures causing work item failures in CI #50750
Closed

EgorBo mentioned this pull request Apr 6, 2021
JIT: Improve inliner (constant feeds constant test, foldable calls) #50784
Closed

DrewScoggins mentioned this pull request Apr 9, 2021
Regression in Crossgen2 Throughput #51006
Closed

JulieLeeMSFT assigned tannergooding Apr 15, 2021

JulieLeeMSFT added this to the 6.0.0 milestone Apr 15, 2021

ghost locked as resolved and limited conversation to collaborators May 15, 2021

tannergooding deleted the resolve-tokens-inline branch July 1, 2025 14:39

tannergooding added 2 commits April 1, 2021 20:54

Add some basic handling to allow resolving call tokens in fgFindJumpT…
95f10fc
…argets when in Tier1 making inline observations

Don't count prefix opcodes as instructions and handle NI_IsSupported_…
4d4cdc3
…True/False as constants in fgFindJumpTargets

tannergooding requested a review from AndyAyersMS April 3, 2021 01:29

ghost added the area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI label Apr 3, 2021

jkotas reviewed Apr 3, 2021
View reviewed changes

Comment threadsrc/coreclr/jit/fgbasic.cpp Outdated

tannergooding added 2 commits April 2, 2021 19:28

Update src/coreclr/jit/fgbasic.cpp
c633127

Applying formatting patch
9796b1a

AndyAyersMS approved these changes Apr 5, 2021
View reviewed changes

tannergooding merged commit e9e8bc7 into dotnet:mainApr 5, 2021

tannergooding mentioned this pull request Apr 5, 2021
Consider adding a cache of well known tokens to the JIT #50426
Closed

runfoappBot mentioned this pull request Apr 5, 2021
Helix DNS failures causing work item failures in CI #50750
Closed

EgorBo mentioned this pull request Apr 6, 2021
JIT: Improve inliner (constant feeds constant test, foldable calls) #50784
Closed

DrewScoggins mentioned this pull request Apr 9, 2021
Regression in Crossgen2 Throughput #51006
Closed

JulieLeeMSFT assigned tannergooding Apr 15, 2021

JulieLeeMSFT added this to the 6.0.0 milestone Apr 15, 2021

ghost locked as resolved and limited conversation to collaborators May 15, 2021

tannergooding deleted the resolve-tokens-inline branch July 1, 2025 14:39

tannergooding added 2 commits April 1, 2021 20:54

Add some basic handling to allow resolving call tokens in fgFindJumpT…
95f10fc
…argets when in Tier1 making inline observations

Don't count prefix opcodes as instructions and handle NI_IsSupported_…
4d4cdc3
…True/False as constants in fgFindJumpTargets

tannergooding requested a review from AndyAyersMS April 3, 2021 01:29

ghost added the area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI label Apr 3, 2021

jkotas reviewed Apr 3, 2021
View reviewed changes

Comment threadsrc/coreclr/jit/fgbasic.cpp Outdated

tannergooding added 2 commits April 2, 2021 19:28

Update src/coreclr/jit/fgbasic.cpp
c633127

Applying formatting patch
9796b1a

AndyAyersMS approved these changes Apr 5, 2021
View reviewed changes

tannergooding merged commit e9e8bc7 into dotnet:mainApr 5, 2021

tannergooding mentioned this pull request Apr 5, 2021
Consider adding a cache of well known tokens to the JIT #50426
Closed

runfoappBot mentioned this pull request Apr 5, 2021
Helix DNS failures causing work item failures in CI #50750
Closed

EgorBo mentioned this pull request Apr 6, 2021
JIT: Improve inliner (constant feeds constant test, foldable calls) #50784
Closed

DrewScoggins mentioned this pull request Apr 9, 2021
Regression in Crossgen2 Throughput #51006
Closed

JulieLeeMSFT assigned tannergooding Apr 15, 2021

JulieLeeMSFT added this to the 6.0.0 milestone Apr 15, 2021

ghost locked as resolved and limited conversation to collaborators May 15, 2021

tannergooding deleted the resolve-tokens-inline branch July 1, 2025 14:39

tannergooding added 2 commits April 1, 2021 20:54

Add some basic handling to allow resolving call tokens in fgFindJumpT…
95f10fc
…argets when in Tier1 making inline observations

Don't count prefix opcodes as instructions and handle NI_IsSupported_…
4d4cdc3
…True/False as constants in fgFindJumpTargets

tannergooding requested a review from AndyAyersMS April 3, 2021 01:29

ghost added the area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI label Apr 3, 2021

jkotas reviewed Apr 3, 2021
View reviewed changes

Comment threadsrc/coreclr/jit/fgbasic.cpp Outdated

tannergooding added 2 commits April 2, 2021 19:28

Update src/coreclr/jit/fgbasic.cpp
c633127

Applying formatting patch
9796b1a

AndyAyersMS approved these changes Apr 5, 2021
View reviewed changes

tannergooding merged commit e9e8bc7 into dotnet:mainApr 5, 2021

tannergooding mentioned this pull request Apr 5, 2021
Consider adding a cache of well known tokens to the JIT #50426
Closed

runfoappBot mentioned this pull request Apr 5, 2021
Helix DNS failures causing work item failures in CI #50750
Closed

EgorBo mentioned this pull request Apr 6, 2021
JIT: Improve inliner (constant feeds constant test, foldable calls) #50784
Closed

DrewScoggins mentioned this pull request Apr 9, 2021
Regression in Crossgen2 Throughput #51006
Closed

JulieLeeMSFT assigned tannergooding Apr 15, 2021

JulieLeeMSFT added this to the 6.0.0 milestone Apr 15, 2021

ghost locked as resolved and limited conversation to collaborators May 15, 2021

tannergooding deleted the resolve-tokens-inline branch July 1, 2025 14:39

Uh oh!

Conversation

tannergooding commented Apr 3, 2021

Uh oh!

tannergooding commented Apr 3, 2021

Uh oh!

Uh oh!

tannergooding commented Apr 5, 2021

Uh oh!

AndyAyersMS left a comment

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Uh oh!

Conversation

tannergooding commented Apr 3, 2021

Uh oh!

tannergooding commented Apr 3, 2021

Uh oh!

Uh oh!

tannergooding commented Apr 5, 2021

Uh oh!

AndyAyersMS left a comment

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Uh oh!

Conversation

tannergooding commented Apr 3, 2021

Uh oh!

tannergooding commented Apr 3, 2021

Uh oh!

Uh oh!

tannergooding commented Apr 5, 2021

Uh oh!

AndyAyersMS left a comment

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Uh oh!

Conversation

tannergooding commented Apr 3, 2021

Uh oh!

tannergooding commented Apr 3, 2021

Uh oh!

Uh oh!

tannergooding commented Apr 5, 2021

Uh oh!

AndyAyersMS left a comment

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants