Merge pull request #95 from oscarbenjamin/pr_dlog

fix(fmpz_mod): avoid using malloc/free

Author: oscarbenjamin

src/flint/types/fmpz_mod.pxd

@@ -8,12 +8,14 @@ from flint.flintlib.fmpz_mod cimport (
 cdef class fmpz_mod_ctx:
     cdef fmpz_mod_ctx_t val
     cdef bint _is_prime
-    cdef fmpz_mod_discrete_log_pohlig_hellman_t *L
+    cdef bint _init_L
+    cdef fmpz_mod_discrete_log_pohlig_hellman_t L
+
     cdef set_any_as_fmpz_mod(self, fmpz_t val, obj)
     cdef any_as_fmpz_mod(self, obj)
-    cdef _precompute_dlog_prime(self)
+    cdef discrete_log_pohlig_hellman_run(self, fmpz_t x, fmpz_t y)
 
 cdef class fmpz_mod(flint_scalar):
     cdef fmpz_mod_ctx ctx
     cdef fmpz_t val
-    cdef fmpz_t *x_g
+    cdef fmpz_t x_g

src/flint/types/fmpz_mod.pyx

@@ -13,6 +13,7 @@ from flint.flintlib.fmpz cimport(
     fmpz_divexact,
     fmpz_gcd,
     fmpz_is_one,
+    fmpz_is_zero,
     fmpz_randm
 )
 from flint.flintlib.fmpz cimport fmpz_mod as fmpz_type_mod
@@ -41,14 +42,12 @@ cdef class fmpz_mod_ctx:
         cdef fmpz one = fmpz.__new__(fmpz)
         fmpz_one(one.val)
         fmpz_mod_ctx_init(self.val, one.val)
-        self.L = NULL
+        fmpz_mod_discrete_log_pohlig_hellman_clear(self.L)
         self._is_prime = 0
-        
 
     def __dealloc__(self):
         fmpz_mod_ctx_clear(self.val)
-        if self.L:
-            fmpz_mod_discrete_log_pohlig_hellman_clear(self.L[0])
+        fmpz_mod_discrete_log_pohlig_hellman_clear(self.L)
 
     def __init__(self, mod):
         # Ensure modulus is fmpz type
@@ -137,18 +136,15 @@ cdef class fmpz_mod_ctx:
 
         return res
 
-    cdef _precompute_dlog_prime(self):
-        """
-        Initalise the dlog data, all discrete logs are solved with an 
-        internally chosen base `y`
-        """
-        self.L = <fmpz_mod_discrete_log_pohlig_hellman_t *>libc.stdlib.malloc(
-            cython.sizeof(fmpz_mod_discrete_log_pohlig_hellman_struct)
-        )
-        fmpz_mod_discrete_log_pohlig_hellman_init(self.L[0])
-        fmpz_mod_discrete_log_pohlig_hellman_precompute_prime(
-            self.L[0], self.val.n
-        )
+    cdef discrete_log_pohlig_hellman_run(self, fmpz_t x, fmpz_t y):
+        # First, Ensure that L has performed precomputations This generates a
+        # base which is a primative root, and used as the base in
+        # fmpz_mod_discrete_log_pohlig_hellman_run
+        if not self._init_L:
+            fmpz_mod_discrete_log_pohlig_hellman_precompute_prime(self.L, self.val.n)
+            self._init_L = True
+
+        fmpz_mod_discrete_log_pohlig_hellman_run(x, self.L, y)
 
     cdef set_any_as_fmpz_mod(self, fmpz_t val, obj):
         # Try and convert obj to fmpz
@@ -235,13 +231,11 @@ cdef class fmpz_mod(flint_scalar):
 
     def __cinit__(self):
         fmpz_init(self.val)
-        self.x_g = NULL
+        fmpz_init(self.x_g)
 
     def __dealloc__(self):
         fmpz_clear(self.val)
-        if self.x_g:
-            fmpz_clear(self.x_g[0])
-            libc.stdlib.free(self.x_g)
+        fmpz_clear(self.x_g)
 
     def __init__(self, val, ctx):
         if not typecheck(ctx, fmpz_mod_ctx):
@@ -354,33 +348,20 @@ cdef class fmpz_mod(flint_scalar):
         if a is NotImplemented:
             raise TypeError(f"Cannot solve the discrete log with {type(a)} as input")
 
-        # First, Ensure that self.ctx.L has performed precomputations
-        # This generates a `y` which is a primative root, and used as
-        # the base in `fmpz_mod_discrete_log_pohlig_hellman_run`
-        if not self.ctx.L:
-            self.ctx._precompute_dlog_prime()
-        
         # Solve the discrete log for the chosen base and target
         # g = y^x_g and  a = y^x_a
         # We want to find x such that a = g^x =>
         # (y^x_a) = (y^x_g)^x => x = (x_a / x_g) mod (p-1)
 
         # For repeated calls to discrete_log, it's more efficient to
         # store x_g rather than keep computing it
-        if not self.x_g:
-            self.x_g = <fmpz_t *>libc.stdlib.malloc(
-                cython.sizeof(fmpz_t)
-            )
-            fmpz_mod_discrete_log_pohlig_hellman_run(
-                self.x_g[0], self.ctx.L[0], self.val
-            )
+        if fmpz_is_zero(self.x_g):
+            self.ctx.discrete_log_pohlig_hellman_run(self.x_g, self.val)
 
         # Then we need to compute x_a which will be different for each call
         cdef fmpz_t x_a
         fmpz_init(x_a)
-        fmpz_mod_discrete_log_pohlig_hellman_run(
-            x_a, self.ctx.L[0], (<fmpz_mod>a).val
-        )
+        self.ctx.discrete_log_pohlig_hellman_run(x_a, (<fmpz_mod>a).val)
 
         # If g is not a primative root, then x_g and pm1 will share
         # a common factor. We can use this to compute the order of
@@ -390,14 +371,14 @@ cdef class fmpz_mod(flint_scalar):
         fmpz_init(g_order)
         fmpz_init(x_g)
 
-        fmpz_gcd(g, self.x_g[0], self.ctx.L[0].pm1)
+        fmpz_gcd(g, self.x_g, self.ctx.L.pm1)
         if not fmpz_is_one(g):
-            fmpz_divexact(x_g, self.x_g[0], g)
+            fmpz_divexact(x_g, self.x_g, g)
             fmpz_divexact(x_a, x_a, g)
-            fmpz_divexact(g_order, self.ctx.L[0].pm1, g)
+            fmpz_divexact(g_order, self.ctx.L.pm1, g)
         else:
-            fmpz_set(g_order, self.ctx.L[0].pm1)
-            fmpz_set(x_g, self.x_g[0])
+            fmpz_set(g_order, self.ctx.L.pm1)
+            fmpz_set(x_g, self.x_g)
 
         # Finally, compute output exponent by computing
         # (x_a / x_g) mod g_order