R600/SI: Add assembler support for FLAT instructions

[oota-llvm.git] / docs / R600Usage.rst

diff --git a/docs/R600Usage.rst b/docs/R600Usage.rst
index 48a30c8a8dd6ba24d588c7d622915b24bf4e2dd4..9bd16f460986faae9a8a39817062f5d3daf5f92c 100644 (file)
--- a/docs/R600Usage.rst
+++ b/docs/R600Usage.rst
@@ -6,22 +6,56 @@ Introduction
 ============
 
 The R600 back-end provides ISA code generation for AMD GPUs, starting with
-the R600 family up until the current Sea Islands (GCN Gen 2).
+the R600 family up until the current Volcanic Islands (GCN Gen 3).
 
 
 Assembler
 =========
 
-The assembler is currently a work in progress and not yet complete.  Below
-are the currently supported features.
+The assembler is currently considered experimental.
+
+For syntax examples look in test/MC/R600.
+
+Below some of the currently supported features (modulo bugs).  These
+all apply to the Southern Islands ISA, Sea Islands and Volcanic Islands
+are also supported but may be missing some instructions and have more bugs:
+
+DS Instructions
+---------------
+All DS instructions are supported.
+
+FLAT Instructions
+------------------
+These instructions are only present in the Sea Islands and Volcanic Islands
+instruction set.  All FLAT instructions are supported for these architectures
+
+MUBUF Instructions
+------------------
+All non-atomic MUBUF instructions are supported.
+
+SMRD Instructions
+-----------------
+Only the s_load_dword* SMRD instructions are supported.
+
+SOP1 Instructions
+-----------------
+All SOP1 instructions are supported.
+
+SOP2 Instructions
+-----------------
+All SOP2 instructions are supported.
+
+SOPC Instructions
+-----------------
+All SOPC instructions are supported.
 
 SOPP Instructions
 -----------------
 
-Unless otherwise mentioned, all SOPP instructions that with an operand
-accept a integer operand(s) only.  No verification is performed on the
-operands, so it is up to the programmer to be familiar with the range
-or acceptable values.
+Unless otherwise mentioned, all SOPP instructions that have one or more
+operands accept integer operands only.  No verification is performed
+on the operands, so it is up to the programmer to be familiar with the
+range or acceptable values.
 
 s_waitcnt
 ^^^^^^^^^
@@ -41,3 +75,20 @@ wait for.
    // Wait for vmcnt counter to be 1.
    s_waitcnt vmcnt(1)
 
+VOP1, VOP2, VOP3, VOPC Instructions
+-----------------------------------
+
+All 32-bit and 64-bit encodings should work.
+
+The assembler will automatically detect which encoding size to use for
+VOP1, VOP2, and VOPC instructions based on the operands.  If you want to force
+a specific encoding size, you can add an _e32 (for 32-bit encoding) or
+_e64 (for 64-bit encoding) suffix to the instruction.  Most, but not all
+instructions support an explicit suffix.  These are all valid assembly
+strings:
+
+.. code-block:: nasm
+
+   v_mul_i32_i24 v1, v2, v3
+   v_mul_i32_i24_e32 v1, v2, v3
+   v_mul_i32_i24_e64 v1, v2, v3