-
Notifications
You must be signed in to change notification settings - Fork 14.3k
[MachineVerifier] Improve checks for G_INSERT_SUBVECTOR. #109209
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Changes from all commits
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -11,9 +11,11 @@ body: | | |
%1:_(<vscale x 2 x s32>) = G_IMPLICIT_DEF | ||
%2:_(<vscale x 1 x s32>) = G_IMPLICIT_DEF | ||
|
||
; CHECK: generic instruction must use register operands | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Is this supposed to say There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. anything that has a type declared in the .td automatically generates a check that the operand is a register. This handles a significant portion of the generic instructions like G_ADD, G_FADD without writing any custom code. I was going to remove the error from G_INSERT_SUBVECTOR and G_EXTRACT_SUBVECTOR, but we need the call to |
||
; CHECK: G_INSERT_SUBVECTOR first source must be a register | ||
%3:_(<vscale x 2 x s32>) = G_INSERT_SUBVECTOR 1, %2, 0 | ||
|
||
; CHECK: generic instruction must use register operands | ||
; CHECK: G_INSERT_SUBVECTOR second source must be a register | ||
%4:_(<vscale x 2 x s32>) = G_INSERT_SUBVECTOR %1, 1, 0 | ||
|
||
|
@@ -23,18 +25,18 @@ body: | | |
; CHECK: Destination type must be a vector | ||
%6:_(s32) = G_INSERT_SUBVECTOR %1, %2, 0 | ||
|
||
; CHECK: First source must be a vector | ||
; CHECK: Type mismatch in generic instruction | ||
%7:_(<vscale x 2 x s32>) = G_INSERT_SUBVECTOR %0, %2, 0 | ||
|
||
; CHECK: Second source must be a vector | ||
%8:_(<vscale x 2 x s32>) = G_INSERT_SUBVECTOR %1, %0, 0 | ||
|
||
; CHECK: Destination type must match the first source vector type | ||
; CHECK: Type mismatch in generic instruction | ||
%9:_(<vscale x 2 x s32>) = G_INSERT_SUBVECTOR %2, %1, 0 | ||
|
||
%10:_(<vscale x 1 x s16>) = G_IMPLICIT_DEF | ||
|
||
; CHECK: Element type of source vectors must be the same | ||
; CHECK: Element type of vectors must be the same | ||
%11:_(<vscale x 2 x s32>) = G_INSERT_SUBVECTOR %1, %10, 0 | ||
|
||
%12:_(<vscale x 4 x s32>) = G_IMPLICIT_DEF | ||
|
@@ -43,5 +45,23 @@ body: | | |
%13:_(<vscale x 4 x s32>) = G_INSERT_SUBVECTOR %12, %1, 3 | ||
|
||
; CHECK: Index must be a multiple of the second source vector's minimum vector length | ||
%13:_(<vscale x 4 x s32>) = G_INSERT_SUBVECTOR %12, %1, 1 | ||
%14:_(<vscale x 4 x s32>) = G_INSERT_SUBVECTOR %12, %1, 1 | ||
|
||
%15:_(<vscale x 4 x s32>) = G_IMPLICIT_DEF | ||
|
||
; CHECK: Second source must be smaller than destination vector | ||
%16:_(<vscale x 2 x s32>) = G_INSERT_SUBVECTOR %1, %15, 0 | ||
|
||
; CHECK: Subvector type and index must not cause insert to overrun the vector being inserted into | ||
%17:_(<vscale x 4 x s32>) = G_INSERT_SUBVECTOR %12, %1, 4 | ||
|
||
%18:_(<vscale x 3 x s32>) = G_IMPLICIT_DEF | ||
|
||
; CHECK: Subvector type and index must not cause insert to overrun the vector being inserted into | ||
%19:_(<vscale x 3 x s32>) = G_INSERT_SUBVECTOR %18, %1, 2 | ||
|
||
%20:_(<2 x s32>) = G_IMPLICIT_DEF | ||
|
||
; CHECK: Vector types must both be fixed or both be scalable | ||
%21:_(<vscale x 1 x s32>) = G_INSERT_SUBVECTOR %12, %20, 2 | ||
... |
Uh oh!
There was an error while loading. Please reload this page.