Bump jax to 0.7.0.dev20250703+cd1b9520b

.dep-versions

@@ -1,7 +1,7 @@
 # Always update the version check in catalyst.__init__ when changing the JAX version.
 # To update JAX version alongside compatible dependency tags, run the following script:
 # python3 .github/workflows/set_dep_versions.py {JAX_version}
-jax=0.6.2
+jax=cd1b9520bd2e1b0dd451281f06e98f030347aa8b
 stablehlo=0a4440a5c8de45c4f9649bf3eb4913bf3f97da0d
 llvm=113f01aa82d055410f22a9d03b3468fa68600589
 enzyme=v0.0.203

frontend/catalyst/__init__.py

@@ -68,6 +68,10 @@
 sys.modules["mlir_quantum.ir"] = __import__("jaxlib.mlir.ir").mlir.ir
 sys.modules["mlir_quantum._mlir_libs"] = __import__("jaxlib.mlir._mlir_libs").mlir._mlir_libs
 
+from catalyst.jax_extras.patches import patch_primitives
+
+patch_primitives()
+
 from catalyst import debug, logging, passes
 from catalyst.api_extensions import *
 from catalyst.api_extensions import __all__ as _api_extension_list

frontend/catalyst/jax_extras/lowering.py

@@ -23,8 +23,6 @@
 from jax._src.effects import ordered_effects as jax_ordered_effects
 from jax._src.interpreters.mlir import _module_name_regex
 from jax._src.sharding_impls import AxisEnv, ReplicaAxisContext
-from jax._src.source_info_util import new_name_stack
-from jax._src.util import wrap_name
 from jax.extend.core import ClosedJaxpr
 from jax.interpreters.mlir import (
     AxisContext,
@@ -72,15 +70,13 @@ def jaxpr_to_mlir(func_name, jaxpr):
         nrep = jaxpr_replicas(jaxpr)
         effects = jax_ordered_effects.filter_in(jaxpr.effects)
         axis_context = ReplicaAxisContext(AxisEnv(nrep, (), ()))
-        name_stack = new_name_stack(wrap_name("ok", "jit"))
         module, context = custom_lower_jaxpr_to_module(
             func_name="jit_" + func_name,
             module_name=func_name,
             jaxpr=jaxpr,
             effects=effects,
             platform="cpu",
             axis_context=axis_context,
-            name_stack=name_stack,
         )
 
     return module, context
@@ -95,7 +91,6 @@ def custom_lower_jaxpr_to_module(
     effects,
     platform: str,
     axis_context: AxisContext,
-    name_stack,
     replicated_args=None,
     arg_shardings=None,
     result_shardings=None,
@@ -142,27 +137,32 @@ def custom_lower_jaxpr_to_module(
         # XLA computation preserves the module name.
         module_name = _module_name_regex.sub("_", module_name)
         ctx.module.operation.attributes["sym_name"] = ir.StringAttr.get(module_name)
+
+        # Use main_function=False to preserve the function name (e.g., "jit_func")
+        # instead of renaming it to "main"
         lower_jaxpr_to_fun(
             ctx,
             func_name,
             jaxpr,
             effects,
-            public=True,
+            main_function=False,
             replicated_args=replicated_args,
             arg_shardings=arg_shardings,
             result_shardings=result_shardings,
-            name_stack=name_stack,
         )
 
+        # Set the entry point function visibility to public and other functions to internal
         worklist = [*ctx.module.body.operations]
         while worklist:
             op = worklist.pop()
             func_name = str(op.name)
-            is_entry_point = func_name.startswith('"jit_')
-            if is_entry_point:
-                continue
             if isinstance(op, FuncOp):
-                op.attributes["llvm.linkage"] = ir.Attribute.parse("#llvm.linkage<internal>")
+                if func_name.startswith('"jit_'):
+                    # Keep entry point functions public
+                    op.attributes["sym_visibility"] = ir.StringAttr.get("public")
+                else:
+                    # Set non-entry functions to internal linkage
+                    op.attributes["llvm.linkage"] = ir.Attribute.parse("#llvm.linkage<internal>")
             if isinstance(op, ModuleOp):
                 worklist += [*op.body.operations]
 

frontend/catalyst/jax_extras/patches.py

@@ -40,6 +40,7 @@
     "_no_clean_up_dead_vars",
     "_gather_shape_rule_dynamic",
     "gather2_p",
+    "patch_primitives",
 )
 
 
@@ -208,3 +209,75 @@ def _gather_shape_rule_dynamic(
     sharding_rule=_gather_sharding_rule,
     vma_rule=partial(standard_vma_rule, "gather"),
 )
+
+
+# TODO: Remove this patch when JAX/PennyLane are updated to use the new JAX 0.7+ API.
+def patch_primitives():
+    """Patch PennyLane/JAX primitives to make them compatible with JAX 0.7+.
+
+    JAX 0.7+ requires all primitive parameters to be hashable, but PennyLane
+    passes **lists** for some parameters like control_values, jaxpr_branches, etc.
+    This patch wraps the bind method to convert lists to tuples to make them hashable.
+    """
+
+    def make_hashable(value):
+        """Recursively convert lists to tuples to make them hashable."""
+        if isinstance(value, list):
+            return tuple(make_hashable(item) for item in value)
+        return value
+
+    try:
+        from pennylane.capture.primitives import ctrl_transform_prim
+        from pennylane.capture.primitives import cond_prim
+        from pennylane.ops.op_math.controlled import Controlled
+
+        original_ctrl_bind = ctrl_transform_prim.bind
+        original_cond_bind = cond_prim.bind
+        original_controlled_bind = Controlled._primitive.bind
+
+        def patched_ctrl_bind(*args, **kwargs):
+            # Convert control_values from list to tuple if present
+            if "control_values" in kwargs:
+                kwargs["control_values"] = make_hashable(kwargs["control_values"])
+            return original_ctrl_bind(*args, **kwargs)
+
+        def patched_cond_bind(*args, **kwargs):
+            # Convert list parameters to tuples
+            if "jaxpr_branches" in kwargs:
+                kwargs["jaxpr_branches"] = make_hashable(kwargs["jaxpr_branches"])
+            if "consts_slices" in kwargs:
+                kwargs["consts_slices"] = make_hashable(kwargs["consts_slices"])
+            return original_cond_bind(*args, **kwargs)
+
+        def patched_controlled_bind(*args, **kwargs):
+            # Convert control_values from list to tuple if present
+            if "control_values" in kwargs:
+                kwargs["control_values"] = make_hashable(kwargs["control_values"])
+            return original_controlled_bind(*args, **kwargs)
+
+        # Replace the bind method
+        ctrl_transform_prim.bind = patched_ctrl_bind
+        cond_prim.bind = patched_cond_bind
+        Controlled._primitive.bind = patched_controlled_bind
+
+    except ImportError:
+        pass
+
+    # patch DynamicJaxprTrace members: makevar and getvar
+    try:
+        from jax._src.interpreters import partial_eval as pe
+
+        def patched_makevar(self, tracer):
+            assert tracer.val is None, "a jaxpr variable must be created only once per tracer"
+            tracer.val = self.frame.newvar(tracer.aval)
+            return tracer.val
+
+        def patched_getvar(self, tracer):  # pylint: disable=unused-argument
+            if var := tracer.val:
+                return var
+            raise jax.core.escaped_tracer_error(tracer)
+
+        pe.DynamicJaxprTrace.makevar = patched_makevar
+        pe.DynamicJaxprTrace.getvar = patched_getvar
+    except ImportError:
+        pass

frontend/catalyst/jax_extras/tracing.py

@@ -427,9 +427,8 @@ def trace_to_jaxpr(
 def new_inner_tracer(trace: DynamicJaxprTrace, aval) -> DynamicJaxprTracer:
     """Create a JAX tracer tracing an abstract value ``aval`, without specifying its source
     primitive."""
-    dt = DynamicJaxprTracer(trace, aval, current_source_info())
-    trace.frame.tracers.append(dt)
-    trace.frame.tracer_to_var[id(dt)] = trace.frame.newvar(aval)
+    atom = trace.frame.newvar(aval)
+    dt = DynamicJaxprTracer(trace, aval, atom, line_info=current_source_info())
     return dt
 
 
@@ -907,12 +906,11 @@ def bind(self, *args, **params):
             # `abstract_eval` returned `out_type` calculated for empty constants.
             [],
             tracers,
-            maker=lambda a: DynamicJaxprTracer(trace, a, source_info),
+            maker=lambda aval: new_inner_tracer(trace, aval),
         )
 
-        invars = map(trace.getvar, tracers)
-        outvars = map(trace.makevar, out_tracers)
-
+        invars = map(lambda t: t.val, tracers)
+        outvars = map(lambda t: t.val, out_tracers)
         eqn = new_jaxpr_eqn(invars, outvars, self, params, [], source_info)
         trace.frame.add_eqn(eqn)
         return out_tracers if self.multiple_results else out_tracers.pop()

frontend/catalyst/jax_primitives.py

@@ -686,7 +686,7 @@ def _grad_lowering(ctx, *args, jaxpr, fn, grad_params):
         flat_output_types,
         ir.StringAttr.get(method),
         symbol_ref,
-        mlir.flatten_lowering_ir_args(args_and_consts),
+        mlir.flatten_ir_values(args_and_consts),
         diffArgIndices=diffArgIndices,
         finiteDiffParam=finiteDiffParam,
     ).results
@@ -763,7 +763,7 @@ def _value_and_grad_lowering(ctx, *args, jaxpr, fn, grad_params):
         gradient_result_types,
         ir.StringAttr.get(method),
         symbol_ref,
-        mlir.flatten_lowering_ir_args(func_args),
+        mlir.flatten_ir_values(func_args),
         diffArgIndices=ir.DenseIntElementsAttr.get(new_argnums),
         finiteDiffParam=ir.FloatAttr.get(ir.F64Type.get(mlir_ctx), h) if h else None,
     ).results
@@ -816,8 +816,8 @@ def _jvp_lowering(ctx, *args, jaxpr, fn, grad_params):
         flat_output_types[len(flat_output_types) // 2 :],
         ir.StringAttr.get(method),
         symbol_ref,
-        mlir.flatten_lowering_ir_args(func_args),
-        mlir.flatten_lowering_ir_args(tang_args),
+        mlir.flatten_ir_values(func_args),
+        mlir.flatten_ir_values(tang_args),
         diffArgIndices=ir.DenseIntElementsAttr.get(new_argnums),
         finiteDiffParam=ir.FloatAttr.get(ir.F64Type.get(mlir_ctx), h) if h else None,
     ).results
@@ -866,8 +866,8 @@ def _vjp_lowering(ctx, *args, jaxpr, fn, grad_params):
         vjp_result_types,
         ir.StringAttr.get(method),
         symbol_ref,
-        mlir.flatten_lowering_ir_args(func_args),
-        mlir.flatten_lowering_ir_args(cotang_args),
+        mlir.flatten_ir_values(func_args),
+        mlir.flatten_ir_values(cotang_args),
         diffArgIndices=ir.DenseIntElementsAttr.get(new_argnums),
         finiteDiffParam=ir.FloatAttr.get(ir.F64Type.get(mlir_ctx), h) if h else None,
     ).results
@@ -930,7 +930,7 @@ def _zne_lowering(ctx, *args, folding, jaxpr, fn):
     return ZneOp(
         flat_output_types,
         symbol_ref,
-        mlir.flatten_lowering_ir_args(args_and_consts),
+        mlir.flatten_ir_values(args_and_consts),
         _folding_attribute(ctx, folding),
         num_folds,
     ).results
@@ -1686,17 +1686,18 @@ def custom_measurement_staging_rule(
     else:
         out_shapes = tuple(core.DShapedArray(shape, dtype) for dtype in dtypes)
 
-    invars = [jaxpr_trace.getvar(obs)]
-    for dyn_dim in dynamic_shape:
-        invars.append(jaxpr_trace.getvar(dyn_dim))
+    invars = [obs.val] + [t.val for t in dynamic_shape]
 
     params = {"static_shape": static_shape}
 
-    out_tracers = tuple(pe.DynamicJaxprTracer(jaxpr_trace, out_shape) for out_shape in out_shapes)
+    out_tracers = tuple(
+        pe.DynamicJaxprTracer(jaxpr_trace, out_shape, jaxpr_trace.frame.newvar(out_shape))
+        for out_shape in out_shapes
+    )
 
     eqn = pe.new_jaxpr_eqn(
         invars,
-        [jaxpr_trace.makevar(out_tracer) for out_tracer in out_tracers],
+        [out_tracer.val for out_tracer in out_tracers],
         primitive,
         params,
         jax.core.no_effects,
@@ -2006,7 +2007,7 @@ def _cond_lowering(
     num_preds = len(branch_jaxprs) - 1
     preds = preds_and_branch_args_plus_consts[:num_preds]
     branch_args_plus_consts = preds_and_branch_args_plus_consts[num_preds:]
-    flat_args_plus_consts = mlir.flatten_lowering_ir_args(branch_args_plus_consts)
+    flat_args_plus_consts = mlir.flatten_ir_values(branch_args_plus_consts)
 
     # recursively lower if-else chains to nested IfOps
     def emit_branches(preds, branch_jaxprs, ip):
@@ -2106,7 +2107,7 @@ def _while_loop_lowering(
     preserve_dimensions: bool,
 ):
     loop_carry_types_plus_consts = [mlir.aval_to_ir_types(a)[0] for a in jax_ctx.avals_in]
-    flat_args_plus_consts = mlir.flatten_lowering_ir_args(iter_args_plus_consts)
+    flat_args_plus_consts = mlir.flatten_ir_values(iter_args_plus_consts)
     assert [val.type for val in flat_args_plus_consts] == loop_carry_types_plus_consts
 
     # split the argument list into 3 separate groups

frontend/catalyst/jax_primitives_utils.py

@@ -143,11 +143,10 @@ def lower_callable_to_funcop(ctx, callable_, call_jaxpr, public=False):
     kwargs["name"] = name
     kwargs["jaxpr"] = call_jaxpr
     kwargs["effects"] = []
-    kwargs["name_stack"] = ctx.name_stack
 
-    # Make the visibility of the function public=True
+    # Make the visibility of the function main_function=True
     # to avoid elimination by the compiler
-    kwargs["public"] = public
+    kwargs["main_function"] = public
 
     func_op = mlir.lower_jaxpr_to_fun(**kwargs)
 
@@ -269,7 +268,7 @@ def create_call_op(ctx, func_op, *args):
     """Create a func::CallOp from JAXPR."""
     output_types = list(map(mlir.aval_to_ir_types, ctx.avals_out))
     flat_output_types = util.flatten(output_types)
-    mlir_args = mlir.flatten_lowering_ir_args(args)
+    mlir_args = mlir.flatten_ir_values(args)
     symbol_ref = get_symbolref(ctx, func_op)
     is_call_same_module = ctx.module_context.module.operation == func_op.parent
     constructor = CallOp if is_call_same_module else LaunchKernelOp

frontend/catalyst/jax_tracer.py

@@ -554,7 +554,7 @@ def bind_overwrite_classical_tracers(
         for i, t in enumerate(out_expanded_tracers):
             # We look for what were the previous output tracers.
             # If they haven't changed, then we leave them unchanged.
-            if trace.getvar(t) in jaxpr_variables:
+            if t.val in jaxpr_variables:
                 continue
 
             # If the variable cannot be found in the current frame
@@ -596,7 +596,7 @@ def bind_overwrite_classical_tracers(
             #            qrp2 = op.trace_quantum(ctx, device, trace, qrp, **kwargs)
             #
             # So it should be safe to cache the tracers as we are doing it.
-            eqn.outvars[i] = trace.getvar(t)
+            eqn.outvars[i] = t.val
 
         # Now, the output variables can be considered as part of the current frame.
         # This allows us to avoid importing all equations again next time.

frontend/scan.sh

@@ -0,0 +1,35 @@
+#!/bin/bash
+# 保存為 check_jax_migration.sh
+
+echo "🔍 JAX 0.7.0 遷移檢查"
+echo "===================="
+
+echo -e "\n❌ P0 - 必須修改的問題:"
+echo "----------------------"
+
+echo -e "\n1. DynamicJaxprTracer 構造函數:"
+rg "DynamicJaxprTracer\(" --type py -C 2
+
+echo -e "\n2. trace.getvar() 調用:"
+rg "\.getvar\(" --type py
+
+echo -e "\n3. trace.makevar() 調用:"
+rg "\.makevar\(" --type py
+
+echo -e "\n4. tracer_to_var 使用:"
+rg "tracer_to_var" --type py
+
+echo -e "\n5. frame.tracers 使用:"
+rg "frame\.tracers" --type py
+
+echo -e "\n\n⚠️  P1 - 需要檢查的代碼:"
+echo "----------------------"
+
+echo -e "\n6. 自定義 staging rules:"
+rg "staging_rule|process_primitive" --type py
+
+echo -e "\n7. JaxprStackFrame 使用:"
+rg "JaxprStackFrame" --type py
+
+echo -e "\n8. Tracer ID 使用:"
+rg "id\(.*tracer" --type py

frontend/test/pytest/from_plxpr/test_from_plxpr_qubit_handler.py

@@ -274,8 +274,8 @@ def test_simple_gate(self):
         # Also check with actual jaxpr variables
         with take_current_trace() as trace:
             gate_out_qubits = trace.frame.eqns[-1].outvars
-            assert trace.frame.tracer_to_var[id(qubit_handler[0])] == gate_out_qubits[0]
-            assert trace.frame.tracer_to_var[id(qubit_handler[1])] == gate_out_qubits[1]
+            assert qubit_handler[0].val == gate_out_qubits[0]
+            assert qubit_handler[1].val == gate_out_qubits[1]
 
     def test_iter(self):
         """Test __iter__ in the qreg manager"""
@@ -316,8 +316,8 @@ def test_chained_gate(self):
         # Also check with actual jaxpr variables
         with take_current_trace() as trace:
             gate_out_qubits = trace.frame.eqns[-1].outvars
-            assert trace.frame.tracer_to_var[id(qubit_handler[0])] == gate_out_qubits[0]
-            assert trace.frame.tracer_to_var[id(qubit_handler[1])] == gate_out_qubits[1]
+            assert qubit_handler[0].val == gate_out_qubits[0]
+            assert qubit_handler[1].val == gate_out_qubits[1]
 
     def test_insert_all_dangling_qubits(self):
         """