Defined invscalv, invscalm, invscald operations. (#661)

Details: - Defined invert-scale (invscal) operation on vectors (level-1v), matrices (level-1m), and diagonals (level-1d). - Added test modules for invscalv and invscalm to the testsuite. - Updated BLISObjectAPI.md and BLISTypedAPI.md API documentation to reflect the new operations. Also updated KernelsHowTo.md accordingly. - Renamed 'beta' to 'alpha' in scalv and scalm testsuite modules (and input.operations files) so that the parameter name matches the parameter used in the documentation.
flame · Sep 8, 2022 · 4afe0cf · 4afe0cf
1 parent a87eae2
commit 4afe0cf
Show file tree

Hide file tree

Showing 50 changed files with 1,070 additions and 88 deletions.
diff --git a/docs/BLISObjectAPI.md b/docs/BLISObjectAPI.md
@@ -41,11 +41,11 @@
 This index provides a quick way to jump directly to the description for each operation discussed later in the [Computational function reference](BLISObjectAPI.md#computational-function-reference) section:
 
   * **[Level-1v](BLISObjectAPI.md#level-1v-operations)**: Operations on vectors:
-    * [addv](BLISObjectAPI.md#addv), [amaxv](BLISObjectAPI.md#amaxv), [axpyv](BLISObjectAPI.md#axpyv), [axpbyv](BLISObjectAPI.md#axpbyv), [copyv](BLISObjectAPI.md#copyv), [dotv](BLISObjectAPI.md#dotv), [dotxv](BLISObjectAPI.md#dotxv), [invertv](BLISObjectAPI.md#invertv), [scal2v](BLISObjectAPI.md#scal2v), [scalv](BLISObjectAPI.md#scalv), [setv](BLISObjectAPI.md#setv), [setrv](BLISObjectAPI.md#setrv), [setiv](BLISObjectAPI.md#setiv), [subv](BLISObjectAPI.md#subv), [swapv](BLISObjectAPI.md#swapv), [xpbyv](BLISObjectAPI.md#xpbyv)
+    * [addv](BLISObjectAPI.md#addv), [amaxv](BLISObjectAPI.md#amaxv), [axpyv](BLISObjectAPI.md#axpyv), [axpbyv](BLISObjectAPI.md#axpbyv), [copyv](BLISObjectAPI.md#copyv), [dotv](BLISObjectAPI.md#dotv), [dotxv](BLISObjectAPI.md#dotxv), [invertv](BLISObjectAPI.md#invertv), [invscalv](BLISObjectAPI.md#invscalv), [scalv](BLISObjectAPI.md#scalv), [scal2v](BLISObjectAPI.md#scal2v), [setv](BLISObjectAPI.md#setv), [setrv](BLISObjectAPI.md#setrv), [setiv](BLISObjectAPI.md#setiv), [subv](BLISObjectAPI.md#subv), [swapv](BLISObjectAPI.md#swapv), [xpbyv](BLISObjectAPI.md#xpbyv)
   * **[Level-1d](BLISObjectAPI.md#level-1d-operations)**: Element-wise operations on matrix diagonals:
-    * [addd](BLISObjectAPI.md#addd), [axpyd](BLISObjectAPI.md#axpyd), [copyd](BLISObjectAPI.md#copyd), [invertd](BLISObjectAPI.md#invertd), [scald](BLISObjectAPI.md#scald), [scal2d](BLISObjectAPI.md#scal2d), [setd](BLISObjectAPI.md#setd), [setid](BLISObjectAPI.md#setid), [shiftd](BLISObjectAPI.md#shiftd), [subd](BLISObjectAPI.md#subd), [xpbyd](BLISObjectAPI.md#xpbyd)
+    * [addd](BLISObjectAPI.md#addd), [axpyd](BLISObjectAPI.md#axpyd), [copyd](BLISObjectAPI.md#copyd), [invertd](BLISObjectAPI.md#invertd), [invscald](BLISObjectAPI.md#invscald), [scald](BLISObjectAPI.md#scald), [scal2d](BLISObjectAPI.md#scal2d), [setd](BLISObjectAPI.md#setd), [setid](BLISObjectAPI.md#setid), [shiftd](BLISObjectAPI.md#shiftd), [subd](BLISObjectAPI.md#subd), [xpbyd](BLISObjectAPI.md#xpbyd)
   * **[Level-1m](BLISObjectAPI.md#level-1m-operations)**: Element-wise operations on matrices:
-    * [addm](BLISObjectAPI.md#addm), [axpym](BLISObjectAPI.md#axpym), [copym](BLISObjectAPI.md#copym), [scalm](BLISObjectAPI.md#scalm), [scal2m](BLISObjectAPI.md#scal2m), [setm](BLISObjectAPI.md#setm), [setrm](BLISObjectAPI.md#setrm), [setim](BLISObjectAPI.md#setim), [subm](BLISObjectAPI.md#subm)
+    * [addm](BLISObjectAPI.md#addm), [axpym](BLISObjectAPI.md#axpym), [copym](BLISObjectAPI.md#copym), [invscalm](BLISObjectAPI.md#invscalm), [scalm](BLISObjectAPI.md#scalm), [scal2m](BLISObjectAPI.md#scal2m), [setm](BLISObjectAPI.md#setm), [setrm](BLISObjectAPI.md#setrm), [setim](BLISObjectAPI.md#setim), [subm](BLISObjectAPI.md#subm)
   * **[Level-1f](BLISObjectAPI.md#level-1f-operations)**: Fused operations on multiple vectors:
     * [axpy2v](BLISObjectAPI.md#axpy2v), [dotaxpyv](BLISObjectAPI.md#dotaxpyv), [axpyf](BLISObjectAPI.md#axpyf), [dotxf](BLISObjectAPI.md#dotxf), [dotxaxpyf](BLISObjectAPI.md#dotxaxpyf)
   * **[Level-2](BLISObjectAPI.md#level-2-operations)**: Operations with one matrix and (at least) one vector operand:
@@ -845,6 +845,24 @@ Invert all elements of an _n_-length vector `x`.
 
 ---
 
+#### invscalv
+```c
+void bli_invscalv
+     (
+       obj_t*  alpha,
+       obj_t*  x
+     );
+```
+Perform
+```
+  x := ( 1.0 / conj?(alpha) ) * x
+```
+where `x` is a vector of length _n_, and `alpha` is a scalar.
+
+Observed object properties: `conj?(alpha)`.
+
+---
+
 #### scalv
 ```c
 void bli_scalv
@@ -1049,6 +1067,19 @@ Observed object properties: `diagoff(A)`.
 
 ---
 
+#### invscald
+```c
+void bli_invscald
+     (
+       obj_t*  alpha,
+       obj_t*  a
+     );
+```
+
+Observed object properties: `conj?(alpha)`, `diagoff(A)`.
+
+---
+
 #### scald
 ```c
 void bli_scald
@@ -1213,6 +1244,24 @@ Observed object properties: `diagoff(A)`, `diag(A)`, `uplo(A)`, `trans?(A)`.
 
 ---
 
+#### invscalm
+```c
+void bli_invscalm
+     (
+       obj_t*  alpha,
+       obj_t*  a
+     );
+```
+Perform
+```
+  A := ( 1.0 / conj?(alpha) ) * A
+```
+where `A` is an _m x n_ matrix stored as a dense matrix, or lower- or upper-triangular/trapezoidal matrix with arbitrary diagonal offset. If `uplo(A)` indicates lower or upper storage, only that part of matrix `A` will be updated.
+
+Observed object properties: `conj?(alpha)`, `diagoff(A)`, `uplo(A)`.
+
+---
+
 #### scalm
 ```c
 void bli_scalm

diff --git a/docs/BLISTypedAPI.md b/docs/BLISTypedAPI.md
@@ -36,11 +36,11 @@
 This index provides a quick way to jump directly to the description for each operation discussed later in the [Computational function reference](BLISTypedAPI.md#computational-function-reference) section:
 
   * **[Level-1v](BLISTypedAPI.md#level-1v-operations)**: Operations on vectors:
-    * [addv](BLISTypedAPI.md#addv), [amaxv](BLISTypedAPI.md#amaxv), [axpyv](BLISTypedAPI.md#axpyv), [axpbyv](BLISTypedAPI.md#axpbyv), [copyv](BLISTypedAPI.md#copyv), [dotv](BLISTypedAPI.md#dotv), [dotxv](BLISTypedAPI.md#dotxv), [invertv](BLISTypedAPI.md#invertv), [scal2v](BLISTypedAPI.md#scal2v), [scalv](BLISTypedAPI.md#scalv), [setv](BLISTypedAPI.md#setv), [subv](BLISTypedAPI.md#subv), [swapv](BLISTypedAPI.md#swapv), [xpbyv](BLISTypedAPI.md#xpbyv)
+    * [addv](BLISTypedAPI.md#addv), [amaxv](BLISTypedAPI.md#amaxv), [axpyv](BLISTypedAPI.md#axpyv), [axpbyv](BLISTypedAPI.md#axpbyv), [copyv](BLISTypedAPI.md#copyv), [dotv](BLISTypedAPI.md#dotv), [dotxv](BLISTypedAPI.md#dotxv), [invertv](BLISTypedAPI.md#invertv), [invscalv](BLISTypedAPI.md#invscalv), [scalv](BLISTypedAPI.md#scalv), [scal2v](BLISTypedAPI.md#scal2v), [setv](BLISTypedAPI.md#setv), [subv](BLISTypedAPI.md#subv), [swapv](BLISTypedAPI.md#swapv), [xpbyv](BLISTypedAPI.md#xpbyv)
   * **[Level-1d](BLISTypedAPI.md#level-1d-operations)**: Element-wise operations on matrix diagonals:
-    * [addd](BLISTypedAPI.md#addd), [axpyd](BLISTypedAPI.md#axpyd), [copyd](BLISTypedAPI.md#copyd), [invertd](BLISTypedAPI.md#invertd), [scald](BLISTypedAPI.md#scald), [scal2d](BLISTypedAPI.md#scal2d), [setd](BLISTypedAPI.md#setd), [setid](BLISTypedAPI.md#setid), [shiftd](BLISTypedAPI.md#shiftd), [subd](BLISTypedAPI.md#subd), [xpbyd](BLISTypedAPI.md#xpbyd)
+    * [addd](BLISTypedAPI.md#addd), [axpyd](BLISTypedAPI.md#axpyd), [copyd](BLISTypedAPI.md#copyd), [invertd](BLISTypedAPI.md#invertd), [invscald](BLISTypedAPI.md#invscald), [scald](BLISTypedAPI.md#scald), [scal2d](BLISTypedAPI.md#scal2d), [setd](BLISTypedAPI.md#setd), [setid](BLISTypedAPI.md#setid), [shiftd](BLISTypedAPI.md#shiftd), [subd](BLISTypedAPI.md#subd), [xpbyd](BLISTypedAPI.md#xpbyd)
   * **[Level-1m](BLISTypedAPI.md#level-1m-operations)**: Element-wise operations on matrices:
-    * [addm](BLISTypedAPI.md#addm), [axpym](BLISTypedAPI.md#axpym), [copym](BLISTypedAPI.md#copym), [scalm](BLISTypedAPI.md#scalm), [scal2m](BLISTypedAPI.md#scal2m), [setm](BLISTypedAPI.md#setm), [subm](BLISTypedAPI.md#subm)
+    * [addm](BLISTypedAPI.md#addm), [axpym](BLISTypedAPI.md#axpym), [copym](BLISTypedAPI.md#copym), [invscalm](BLISTypedAPI.md#invscalm), [scalm](BLISTypedAPI.md#scalm), [scal2m](BLISTypedAPI.md#scal2m), [setm](BLISTypedAPI.md#setm), [subm](BLISTypedAPI.md#subm)
   * **[Level-1f](BLISTypedAPI.md#level-1f-operations)**: Fused operations on multiple vectors:
     * [axpy2v](BLISTypedAPI.md#axpy2v), [dotaxpyv](BLISTypedAPI.md#dotaxpyv), [axpyf](BLISTypedAPI.md#axpyf), [dotxf](BLISTypedAPI.md#dotxf), [dotxaxpyf](BLISTypedAPI.md#dotxaxpyf)
   * **[Level-2](BLISTypedAPI.md#level-2-operations)**: Operations with one matrix and (at least) one vector operand:
@@ -369,6 +369,24 @@ Invert all elements of an _n_-length vector `x`.
 
 ---
 
+#### invscalv
+```c
+void bli_?invscalv
+     (
+       conj_t  conjalpha,
+       dim_t   n,
+       ctype*  alpha,
+       ctype*  x, inc_t incx
+     );
+```
+Perform
+```
+  x := ( 1.0 / conjalpha(alpha) ) * x
+```
+where `x` is a vector of length _n_, and `alpha` is a scalar.
+
+---
+
 #### scalv
 ```c
 void bli_?scalv
@@ -548,6 +566,21 @@ void bli_?invertd
 
 ---
 
+#### invscald
+```c
+void bli_?invscald
+     (
+       conj_t  conjalpha,
+       doff_t  diagoffa,
+       dim_t   m,
+       dim_t   n,
+       ctype*  alpha,
+       ctype*  a, inc_t rsa, inc_t csa
+     );
+```
+
+---
+
 #### scald
 ```c
 void bli_?scald
@@ -737,6 +770,27 @@ where `B` is an _m x n_ matrix, `A` is stored as a dense matrix, or lower- or up
 
 ---
 
+#### invscalm
+```c
+void bli_?invscalm
+     (
+       conj_t  conjalpha,
+       doff_t  diagoffa,
+       uplo_t  uploa,
+       dim_t   m,
+       dim_t   n,
+       ctype*  alpha,
+       ctype*  a, inc_t rsa, inc_t csa
+     );
+```
+Perform
+```
+  A := ( 1.0 / conjalpha(alpha) ) * A
+```
+where `A` is an _m x n_ matrix stored as a dense matrix, or lower- or upper-triangular/trapezoidal matrix, as specified by `uploa`, with the diagonal offset of `A` specified by `diagoffa`. If `uploa` indicates lower or upper storage, only that part of matrix `A` will be updated.
+
+---
+
 #### scalm
 ```c
 void bli_?scalm

diff --git a/docs/KernelsHowTo.md b/docs/KernelsHowTo.md
@@ -22,11 +22,11 @@ One of the primary features of BLIS is that it provides a large set of dense lin
 
 Presently, BLIS supports several groups of operations:
   * **[Level-1v](BLISTypedAPI.md#level-1v-operations)**: Operations on vectors:
-    * [addv](BLISTypedAPI.md#addv), [amaxv](BLISTypedAPI.md#amaxv), [axpyv](BLISTypedAPI.md#axpyv), [copyv](BLISTypedAPI.md#copyv), [dotv](BLISTypedAPI.md#dotv), [dotxv](BLISTypedAPI.md#dotxv), [invertv](BLISTypedAPI.md#invertv), [scal2v](BLISTypedAPI.md#scal2v), [scalv](BLISTypedAPI.md#scalv), [setv](BLISTypedAPI.md#setv), [subv](BLISTypedAPI.md#subv), [swapv](BLISTypedAPI.md#swapv)
+    * [addv](BLISTypedAPI.md#addv), [amaxv](BLISTypedAPI.md#amaxv), [axpyv](BLISTypedAPI.md#axpyv), [copyv](BLISTypedAPI.md#copyv), [dotv](BLISTypedAPI.md#dotv), [dotxv](BLISTypedAPI.md#dotxv), [invertv](BLISTypedAPI.md#invertv), [invscalv](BLISTypedAPI.md#invscalv), [scalv](BLISTypedAPI.md#scalv), [scal2v](BLISTypedAPI.md#scal2v), [setv](BLISTypedAPI.md#setv), [subv](BLISTypedAPI.md#subv), [swapv](BLISTypedAPI.md#swapv)
   * **[Level-1d](BLISTypedAPI.md#level-1d-operations)**: Element-wise operations on matrix diagonals:
-    * [addd](BLISTypedAPI.md#addd), [axpyd](BLISTypedAPI.md#axpyd), [copyd](BLISTypedAPI.md#copyd), [invertd](BLISTypedAPI.md#invertd), [scald](BLISTypedAPI.md#scald), [scal2d](BLISTypedAPI.md#scal2d), [setd](BLISTypedAPI.md#setd), [setid](BLISTypedAPI.md#setid), [subd](BLISTypedAPI.md#subd)
+    * [addd](BLISTypedAPI.md#addd), [axpyd](BLISTypedAPI.md#axpyd), [copyd](BLISTypedAPI.md#copyd), [invertd](BLISTypedAPI.md#invertd), [invscald](BLISTypedAPI.md#invscald), [scald](BLISTypedAPI.md#scald), [scal2d](BLISTypedAPI.md#scal2d), [setd](BLISTypedAPI.md#setd), [setid](BLISTypedAPI.md#setid), [subd](BLISTypedAPI.md#subd)
   * **[Level-1m](BLISTypedAPI.md#level-1m-operations)**: Element-wise operations on matrices:
-    * [addm](BLISTypedAPI.md#addm), [axpym](BLISTypedAPI.md#axpym), [copym](BLISTypedAPI.md#copym), [scalm](BLISTypedAPI.md#scalm), [scal2m](BLISTypedAPI.md#scal2m), [setm](BLISTypedAPI.md#setm), [subm](BLISTypedAPI.md#subm)
+    * [addm](BLISTypedAPI.md#addm), [axpym](BLISTypedAPI.md#axpym), [copym](BLISTypedAPI.md#copym), [invscalm](BLISTypedAPI.md#invscalm), [scalm](BLISTypedAPI.md#scalm), [scal2m](BLISTypedAPI.md#scal2m), [setm](BLISTypedAPI.md#setm), [subm](BLISTypedAPI.md#subm)
   * **[Level-1f](BLISTypedAPI.md#level-1f-operations)**: Fused operations on multiple vectors:
     * [axpy2v](BLISTypedAPI.md#axpy2v), [dotaxpyv](BLISTypedAPI.md#dotaxpyv), [axpyf](BLISTypedAPI.md#axpyf), [dotxf](BLISTypedAPI.md#dotxf), [dotxaxpyf](BLISTypedAPI.md#dotxaxpyf)
   * **[Level-2](BLISTypedAPI.md#level-2-operations)**: Operations with one matrix and (at least) one vector operand:
@@ -81,6 +81,7 @@ BLIS supports the following 14 level-1v kernels. These kernels are used primaril
   * **dotv**: Performs a [dot product](BLISTypedAPI.md#dotv) where the output scalar is overwritten.
   * **dotxv**: Performs an [extended dot product](BLISTypedAPI.md#dotxv) operation where the dot product is first scaled and then accumulated into a scaled output scalar.
   * **invertv**: Performs an [element-wise vector inversion](BLISTypedAPI.md#invertv) operation.
+  * **invscalv**: Performs an [in-place (destructive) vector inverse-scaling](BLISTypedAPI.md#invscalv) operation.
   * **scalv**: Performs an [in-place (destructive) vector scaling](BLISTypedAPI.md#scalv) operation.
   * **scal2v**: Performs an [out-of-place (non-destructive) vector scaling](BLISTypedAPI.md#scal2v) operation.
   * **setv**: Performs a [vector broadcast](BLISTypedAPI.md#setv) operation.
@@ -184,6 +185,7 @@ datatype characters.
 | copyv            | `BLIS_COPYV_KER`      | `?copyv_ft`           |
 | dotxv            | `BLIS_DOTXV_KER`      | `?dotxv_ft`           |
 | invertv          | `BLIS_INVERTV_KER`    | `?invertv_ft`         |
+| invscalv         | `BLIS_INVSCALV_KER`   | `?invscalv_ft`        |
 | scalv            | `BLIS_SCALV_KER`      | `?scalv_ft`           |
 | scal2v           | `BLIS_SCAL2V_KER`     | `?scal2v_ft`          |
 | setv             | `BLIS_SETV_KER`       | `?setv_ft`            |
@@ -220,6 +222,7 @@ This section seeks to provide developers with a complete reference for each of t
     * [dotv](KernelsHowTo.md#dotv-kernel)
     * [dotxv](KernelsHowTo.md#dotxv-kernel)
     * [invertv](KernelsHowTo.md#invertv-kernel)
+    * [invscalv](KernelsHowTo.md#invscalv-kernel)
     * [scalv](KernelsHowTo.md#scalv-kernel)
     * [scal2v](KernelsHowTo.md#scal2v-kernel)
     * [setv](KernelsHowTo.md#setv-kernel)
@@ -929,6 +932,25 @@ This kernel inverts all elements of an _n_-length vector `x`.
 
 ---
 
+#### invscalv kernel
+```c
+void bli_?invscalv_<suffix>
+     (
+       conj_t           conjalpha,
+       dim_t            n,
+       ctype*  restrict alpha,
+       ctype*  restrict x, inc_t incx,
+       cntx_t* restrict cntx
+     )
+```
+This kernel performs the following operation:
+```
+  x := ( 1.0 / conjalpha(alpha) ) * x
+```
+where `x` is a vector of length _n_ stored with stride `incx` and `alpha` is a scalar.
+
+---
+
 #### scalv kernel
 ```c
 void bli_?scalv_<suffix>

diff --git a/frame/1/bli_l1v_check.c b/frame/1/bli_l1v_check.c
@@ -165,6 +165,7 @@ void PASTEMAC(opname,_check) \
 	bli_l1v_ax_check( alpha, x ); \
 }
 
+GENFRONT( invscalv )
 GENFRONT( scalv )
 GENFRONT( setv )
 

diff --git a/frame/1/bli_l1v_check.h b/frame/1/bli_l1v_check.h
@@ -140,6 +140,7 @@ void PASTEMAC(opname,_check) \
        const obj_t* x  \
      );
 
+GENTPROT( invscalv )
 GENTPROT( scalv )
 GENTPROT( setv )
 

diff --git a/frame/1/bli_l1v_fpa.c b/frame/1/bli_l1v_fpa.c
@@ -60,6 +60,7 @@ GENFRONT( scal2v )
 GENFRONT( dotv )
 GENFRONT( dotxv )
 GENFRONT( invertv )
+GENFRONT( invscalv )
 GENFRONT( scalv )
 GENFRONT( setv )
 GENFRONT( swapv )

diff --git a/frame/1/bli_l1v_fpa.h b/frame/1/bli_l1v_fpa.h
@@ -52,6 +52,7 @@ GENPROT( scal2v )
 GENPROT( dotv )
 GENPROT( dotxv )
 GENPROT( invertv )
+GENPROT( invscalv )
 GENPROT( scalv )
 GENPROT( setv )
 GENPROT( swapv )

diff --git a/frame/1/bli_l1v_ft.h b/frame/1/bli_l1v_ft.h
@@ -158,7 +158,7 @@ typedef void (*PASTECH3(ch,opname,EX_SUF,tsuf)) \
 
 INSERT_GENTDEF( invertv )
 
-// scalv, setv
+// invscalv, scalv, setv
 
 #undef  GENTDEF
 #define GENTDEF( ctype, ch, opname, tsuf ) \
@@ -172,6 +172,7 @@ typedef void (*PASTECH3(ch,opname,EX_SUF,tsuf)) \
        BLIS_TAPI_EX_PARAMS  \
      );
 
+INSERT_GENTDEF( invscalv )
 INSERT_GENTDEF( scalv )
 INSERT_GENTDEF( setv )
 

diff --git a/frame/1/bli_l1v_ft_ker.h b/frame/1/bli_l1v_ft_ker.h
@@ -161,7 +161,7 @@ typedef void (*PASTECH3(ch,opname,_ker,tsuf)) \
 
 INSERT_GENTDEF( invertv )
 
-// scalv, setv
+// invscalv, scalv, setv
 
 #undef  GENTDEF
 #define GENTDEF( ctype, ch, opname, tsuf ) \
@@ -175,6 +175,7 @@ typedef void (*PASTECH3(ch,opname,_ker,tsuf)) \
        cntx_t*         cntx  \
      );
 
+INSERT_GENTDEF( invscalv )
 INSERT_GENTDEF( scalv )
 INSERT_GENTDEF( setv )
 

diff --git a/frame/1/bli_l1v_ker.h b/frame/1/bli_l1v_ker.h
@@ -90,6 +90,12 @@ INSERT_GENTPROT_BASIC0( dotxv_ker_name )
 INSERT_GENTPROT_BASIC0( invertv_ker_name )
 
 
+#undef  GENTPROT
+#define GENTPROT INVSCALV_KER_PROT
+
+INSERT_GENTPROT_BASIC0( invscalv_ker_name )
+
+
 #undef  GENTPROT
 #define GENTPROT SCALV_KER_PROT
 

diff --git a/frame/1/bli_l1v_ker_prot.h b/frame/1/bli_l1v_ker_prot.h
@@ -139,6 +139,18 @@ void PASTEMAC(ch,opname) \
      ); \
 
 
+#define INVSCALV_KER_PROT( ctype, ch, opname ) \
+\
+void PASTEMAC(ch,opname) \
+     ( \
+       conj_t           conjalpha, \
+       dim_t            n, \
+       ctype*  restrict alpha, \
+       ctype*  restrict x, inc_t incx, \
+       cntx_t*          cntx  \
+     ); \
+
+
 #define SCALV_KER_PROT( ctype, ch, opname ) \
 \
 void PASTEMAC(ch,opname) \

diff --git a/frame/1/bli_l1v_oapi.c b/frame/1/bli_l1v_oapi.c
@@ -460,6 +460,7 @@ void PASTEMAC(opname,EX_SUF) \
 	); \
 }
 
+GENFRONT( invscalv )
 GENFRONT( scalv )
 GENFRONT( setv )
 

diff --git a/frame/1/bli_l1v_oapi.h b/frame/1/bli_l1v_oapi.h
@@ -147,6 +147,7 @@ BLIS_EXPORT_BLIS void PASTEMAC(opname,EX_SUF) \
        BLIS_OAPI_EX_PARAMS  \
      );
 
+GENTPROT( invscalv )
 GENTPROT( scalv )
 GENTPROT( setv )
 

diff --git a/frame/1/bli_l1v_tapi.c b/frame/1/bli_l1v_tapi.c
@@ -341,6 +341,7 @@ void PASTEMAC2(ch,opname,EX_SUF) \
 	); \
 }
 
+INSERT_GENTFUNC_BASIC( invscalv, BLIS_INVSCALV_KER )
 INSERT_GENTFUNC_BASIC( scalv, BLIS_SCALV_KER )
 INSERT_GENTFUNC_BASIC( setv,  BLIS_SETV_KER )
 

diff --git a/frame/1/bli_l1v_tapi.h b/frame/1/bli_l1v_tapi.h
@@ -163,6 +163,7 @@ BLIS_EXPORT_BLIS void PASTEMAC2(ch,opname,EX_SUF) \
        BLIS_TAPI_EX_PARAMS  \
      ); \
 
+INSERT_GENTPROT_BASIC0( invscalv )
 INSERT_GENTPROT_BASIC0( scalv )
 INSERT_GENTPROT_BASIC0( setv )
 

diff --git a/frame/1d/bli_l1d_check.c b/frame/1d/bli_l1d_check.c
@@ -98,6 +98,7 @@ void PASTEMAC(opname,_check) \
 	bli_l1d_ax_check( alpha, x ); \
 }
 
+GENFRONT( invscald )
 GENFRONT( scald )
 GENFRONT( setd )
 GENFRONT( setid )

diff --git a/frame/1d/bli_l1d_check.h b/frame/1d/bli_l1d_check.h
@@ -85,6 +85,7 @@ void PASTEMAC(opname,_check) \
        const obj_t*  x  \
     );
 
+GENTPROT( invscald )
 GENTPROT( scald )
 GENTPROT( setd )
 GENTPROT( setid )

diff --git a/frame/1d/bli_l1d_fpa.c b/frame/1d/bli_l1d_fpa.c
@@ -56,6 +56,7 @@ GENFRONT( subd )
 GENFRONT( axpyd )
 GENFRONT( scal2d )
 GENFRONT( invertd )
+GENFRONT( invscald )
 GENFRONT( scald )
 GENFRONT( setd )
 GENFRONT( setid )
-Original file line number
+Diff line change
@@ Expand Up / @@ -460,6 +460,7 @@ void PASTEMAC(opname,EX_SUF) \ @@
     	); \
     }
+    GENFRONT( invscalv )
     GENFRONT( scalv )
     GENFRONT( setv )
@@ Expand Down @@