Merge branch 'main' into amore

.github/workflows/gmp-sec.yml

@@ -18,19 +18,16 @@ jobs:
         config:
           - {
             name: "Windows Latest - MinGW",
-            artifact: "windows-mingw.tar.xz",
             os: windows-latest,
             cc: "gcc"
           }
           - {
             name: "Ubuntu Latest - GCC",
-            artifact: "linux-gcc.tar.xz",
             os: ubuntu-latest,
             cc: "gcc",
           }
           - {
             name: "Ubuntu Latest - Clang",
-            artifact: "linux-clang.tar.xz",
             os: ubuntu-latest,
             cc: "clang",
           }
@@ -39,6 +36,11 @@ jobs:
             os: macos-latest,
             cc: "clang",
           }
+
+    defaults:
+      run:
+        shell: ${{ matrix.config.os == 'windows-latest' && 'msys2 {0}' || 'bash' }}
+
     steps:
       - uses: actions/checkout@v4
 
@@ -66,11 +68,10 @@ jobs:
 
       - name: Run CMake (MingW)
         if: ${{ (runner.os == 'Windows') && (matrix.config.cc == 'gcc') }}
-        shell: bash
         run: |
           mkdir build
           cd build
-          cmake -DSEED= -DBENCH=0 -DARITH=gmp-sec -G "MinGW Makefiles" ..
+          cmake -DSEED= -DBENCH=0 -DARITH=gmp-sec -G "Unix Makefiles" ..
 
       - name: Run CMake (standard)
         if: ${{ !(runner.os == 'Windows') }}

.github/workflows/gmp.yml

@@ -18,27 +18,30 @@ jobs:
         config:
           - {
             name: "Windows Latest - MinGW",
-            artifact: "windows-mingw.tar.xz",
             os: windows-latest,
             cc: "gcc"
           }
           - {
             name: "Ubuntu Latest - GCC",
-            artifact: "linux-gcc.tar.xz",
             os: ubuntu-latest,
             cc: "gcc",
           }
           - {
             name: "Ubuntu Latest - Clang",
-            artifact: "linux-clang.tar.xz",
             os: ubuntu-latest,
             cc: "clang",
+
           }
           - {
             name: "MacOS Latest",
             os: macos-latest,
             cc: "clang",
           }
+
+    defaults:
+      run:
+        shell: ${{ matrix.config.os == 'windows-latest' && 'msys2 {0}' || 'bash' }}
+
     steps:
       - uses: actions/checkout@v4
 
@@ -54,27 +57,25 @@ jobs:
         if: ${{ (runner.os == 'Windows') && (matrix.config.cc == 'gcc') }}
         uses: msys2/setup-msys2@v2
         with:
-          msystem: MINGW64
+          msystem: UCRT64
           install: >-
-            gmp
-            gmp-devel
             git
             base-devel
             gcc
             cmake
+            gmp
+            gmp-devel
           update: true
 
       - name: Run CMake (MingW)
         if: ${{ (runner.os == 'Windows') && (matrix.config.cc == 'gcc') }}
-        shell: bash
         run: |
           mkdir build
           cd build
-          cmake -DSEED= -DBENCH=0 -DARITH=gmp -G "MinGW Makefiles" ..
+          cmake -DSEED= -DBENCH=0 -DARITH=gmp -G "Unix Makefiles" ..
 
       - name: Run CMake (standard)
         if: ${{ !(runner.os == 'Windows') }}
-        shell: bash
         run: |
           mkdir build
           cd build

CMakeLists.txt

@@ -269,6 +269,11 @@ else()
 	set(CFLAGS "$ENV{CFLAGS}" CACHE STRING "User-chosen compiler flags.")
 endif()
 
+# Remove annoying flags when compiler cannot compute buffer sizes.
+if (CMAKE_C_COMPILER_ID STREQUAL "GNU")
+	set(CFLAGS "${CFLAGS} -Wno-stringop-overflow -Wno-stringop-overread")
+endif()
+
 if(AUSAN)
 	set(DFLAGS "${DFLAGS} -ggdb -fsanitize=address -fsanitize=undefined")
 endif()

bench/bench_cp.c

@@ -1342,9 +1342,9 @@ static void bbs(void) {
 
 static int cls(void) {
 	int i, code = RLC_ERR;
-	bn_t r, t, u, v, _v[4];
-	g1_t a, A, b, B, c, _A[4], _B[4];
-	g2_t x, y, z, _z[4];
+	bn_t r, t, u, v, vs[4];
+	g1_t a, A, b, B, c, As[4], Bs[4];
+	g2_t x, y, z, zs[4];
 	uint8_t m[5] = { 0, 1, 2, 3, 4 };
 	const uint8_t *ms[5] = {m, m, m, m, m};
 	const size_t ls[5] = {sizeof(m), sizeof(m), sizeof(m), sizeof(m), sizeof(m)};
@@ -1362,10 +1362,10 @@ static int cls(void) {
 	g2_null(y);
 	g2_null(z);
 	for (i = 0; i < 4; i++) {
-		bn_null(_v[i]);
-		g1_null(_A[i]);
-		g1_null(_B[i]);
-		g2_null(_z[i]);
+		bn_null(vs[i]);
+		g1_null(As[i]);
+		g1_null(Bs[i]);
+		g2_null(zs[i]);
 	}
 
 	bn_new(r);
@@ -1381,10 +1381,10 @@ static int cls(void) {
 	g2_new(y);
 	g2_new(z);
 	for (i = 0; i < 4; i++) {
-		bn_new(_v[i]);
-		g1_new(_A[i]);
-		g1_new(_B[i]);
-		g2_new(_z[i]);
+		bn_new(vs[i]);
+		g1_new(As[i]);
+		g1_new(Bs[i]);
+		g2_new(zs[i]);
 	}
 
 	BENCH_RUN("cp_cls_gen") {
@@ -1413,15 +1413,15 @@ static int cls(void) {
 	} BENCH_END;
 
 	BENCH_RUN("cp_clb_gen (5)") {
-		BENCH_ADD(cp_clb_gen(t, u, _v, x, y, _z, 5));
+		BENCH_ADD(cp_clb_gen(t, u, vs, x, y, zs, 5));
 	} BENCH_END;
 
 	BENCH_RUN("cp_clb_sig (5)") {
-		BENCH_ADD(cp_clb_sig(a, _A, b, _B, c, ms, ls, t, u, _v, 5));
+		BENCH_ADD(cp_clb_sig(a, As, b, Bs, c, ms, ls, t, u, vs, 5));
 	} BENCH_END;
 
 	BENCH_RUN("cp_clb_ver (5)") {
-		BENCH_ADD(cp_clb_ver(a, _A, b, _B, c, ms, ls, x, y, _z, 5));
+		BENCH_ADD(cp_clb_ver(a, As, b, Bs, c, ms, ls, x, y, zs, 5));
 	} BENCH_END;
 
 	bn_free(r);
@@ -1437,18 +1437,18 @@ static int cls(void) {
 	g2_free(y);
 	g2_free(z);
 	for (i = 0; i < 4; i++) {
-		bn_free(_v[i]);
-		g1_free(_A[i]);
-		g1_free(_B[i]);
-		g2_free(_z[i]);
+		bn_free(vs[i]);
+		g1_free(As[i]);
+		g1_free(Bs[i]);
+		g2_free(zs[i]);
 	}
 	return code;
 }
 
 static void pss(void) {
-	bn_t ms[10], n, u, v, _v[10];
+	bn_t ms[10], n, u, v, vs[10];
 	g1_t a, b;
-	g2_t g, x, y, _y[10];
+	g2_t g, x, y, ys[10];
 
 	bn_null(n);
 	bn_null(u);
@@ -1470,12 +1470,12 @@ static void pss(void) {
 	g1_get_ord(n);
 	for (int i = 0; i < 10; i++) {
 		bn_null(ms[i]);
-		bn_null(_v[i]);
-		g2_null(_y[i]);
+		bn_null(vs[i]);
+		g2_null(ys[i]);
 		bn_new(ms[i]);
 		bn_rand_mod(ms[i], n);
-		bn_new(_v[i]);
-		g2_new(_y[i]);
+		bn_new(vs[i]);
+		g2_new(ys[i]);
 	}
 
 	BENCH_RUN("cp_pss_gen") {
@@ -1491,15 +1491,15 @@ static void pss(void) {
 	} BENCH_END;
 
 	BENCH_RUN("cp_psb_gen (10)") {
-		BENCH_ADD(cp_psb_gen(u, _v, g, x, _y, 10));
+		BENCH_ADD(cp_psb_gen(u, vs, g, x, ys, 10));
 	} BENCH_END;
 
 	BENCH_RUN("cp_psb_sig (10)") {
-		BENCH_ADD(cp_psb_sig(a, b, ms, u, _v, 10));
+		BENCH_ADD(cp_psb_sig(a, b, ms, u, vs, 10));
 	} BENCH_END;
 
 	BENCH_RUN("cp_psb_ver (10)") {
-		BENCH_ADD(cp_psb_ver(a, b, ms, g, x, _y, 10));
+		BENCH_ADD(cp_psb_ver(a, b, ms, g, x, ys, 10));
 	} BENCH_END;
 
 	bn_free(u);
@@ -1511,17 +1511,17 @@ static void pss(void) {
 	g2_free(y);
 	for (int i = 0; i < 10; i++) {
 		bn_free(ms[i]);
-		bn_free(_v[i]);
-		g1_free(_y[i]);
+		bn_free(vs[i]);
+		g1_free(ys[i]);
 	}
 }
 
 #if defined(WITH_MPC)
 
 static void mpss(void) {
-	bn_t m[2], n, u[2], v[2], ms[10][2], _v[10][2];
+	bn_t m[2], n, u[2], v[2], ms[10][2], vs[10][2];
 	g1_t g, s[2];
-	g2_t h, x[2], y[2], _y[10][2];
+	g2_t h, x[2], y[2], ys[10][2];
 	gt_t r[2];
 	mt_t tri[3][2];
 	pt_t t[2];
@@ -1560,12 +1560,12 @@ static void mpss(void) {
 		g1_get_ord(n);
 		for (int j = 0; j < 10; j++) {
 			bn_null(ms[j][i]);
-			bn_null(_v[j][i]);
-			g2_null(_y[j][i]);
+			bn_null(vs[j][i]);
+			g2_null(ys[j][i]);
 			bn_new(ms[j][i]);
 			bn_rand_mod(ms[j][i], n);
-			bn_new(_v[j][i]);
-			g2_new(_y[j][i]);
+			bn_new(vs[j][i]);
+			g2_new(ys[j][i]);
 		}
 	}
 
@@ -1610,23 +1610,23 @@ static void mpss(void) {
 	mpc_mt_gen(tri[2], n);
 
 	BENCH_RUN("cp_mpsb_gen (10)") {
-		BENCH_ADD(cp_mpsb_gen(u, _v, h, x, _y, 10));
+		BENCH_ADD(cp_mpsb_gen(u, vs, h, x, ys, 10));
 	} BENCH_END;
 
 	BENCH_RUN("cp_mpsb_bct (10)") {
-		BENCH_ADD(cp_mpsb_bct(x, _y, 10));
+		BENCH_ADD(cp_mpsb_bct(x, ys, 10));
 	} BENCH_END;
 
 	BENCH_RUN("cp_mpsb_sig (10)") {
-		BENCH_ADD(cp_mpsb_sig(g, s, ms, u, _v, tri[0], tri[1], 10));
+		BENCH_ADD(cp_mpsb_sig(g, s, ms, u, vs, tri[0], tri[1], 10));
 	} BENCH_DIV(2);
 
 	BENCH_RUN("cp_mpsb_ver (10)") {
-		BENCH_ADD(cp_mpsb_ver(r[1], g, s, ms, h, x[0], _y, NULL, tri[2], t, 10));
+		BENCH_ADD(cp_mpsb_ver(r[1], g, s, ms, h, x[0], ys, NULL, tri[2], t, 10));
 	} BENCH_DIV(2);
 
 	BENCH_RUN("cp_mpsb_ver (10,sk)") {
-		BENCH_ADD(cp_mpsb_ver(r[1], g, s, ms, h, x[0], _y, _v, tri[2], t, 10));
+		BENCH_ADD(cp_mpsb_ver(r[1], g, s, ms, h, x[0], ys, vs, tri[2], t, 10));
 	} BENCH_DIV(2);
 
 	bn_free(n);
@@ -1646,8 +1646,8 @@ static void mpss(void) {
 		pt_free(t[i]);
 		for (int j = 0; j < 10; j++) {
 			bn_free(ms[j][i]);
-			bn_free(_v[j][i]);
-			g2_free(_y[j][i]);
+			bn_free(vs[j][i]);
+			g2_free(ys[j][i]);
 		}
 	}
 }

bench/bench_fb.c

@@ -165,7 +165,7 @@ static void arith(void) {
 	fb_st t[RLC_FB_TABLE_MAX];
 	dv_t e;
 	bn_t f;
-	int bits;
+	uint_t bits;
 
 	fb_null(a);
 	fb_null(b);

bench/bench_fp.c

@@ -583,6 +583,15 @@ static void arith(void) {
 	BENCH_END;
 #endif
 
+#if FP_SMB == BINAR || !defined(STRIP)
+	BENCH_RUN("fp_smb_binar") {
+		fp_rand(a);
+		fp_sqr(a, a);
+		BENCH_ADD(fp_smb_binar(a));
+	}
+	BENCH_END;
+#endif
+
 #if FP_SMB == DIVST || !defined(STRIP)
 	BENCH_RUN("fp_smb_divst") {
 		fp_rand(a);

bench/bench_pc.c

@@ -692,6 +692,12 @@ static void arith(void) {
 	}
 	BENCH_END;
 
+	BENCH_RUN("gt_frb (1)") {
+		gt_rand(a);
+		BENCH_ADD(gt_frb(c, a, 1));
+	}
+	BENCH_END;
+
 	BENCH_RUN("gt_exp") {
 		gt_rand(a);
 		pc_get_ord(d);

bench/bench_rand.c

@@ -40,7 +40,7 @@
 #include <sys/stat.h>
 #include <fcntl.h>
 
-static void test_bytes(uint8_t *buf, int size, void *args) {
+static void test_bytes(uint8_t *buf, size_t size, void *args) {
 	int c, l, fd = *(int *)args;
 
 	if (fd == -1) {

cmake/gmp.cmake

@@ -49,9 +49,9 @@ find_path(GMP_INCLUDE_DIR gmp.h
 )
 
 if(STBIN)
-	find_library(GMP_LIBRARIES NAMES libgmp.a gmp.lib libgmp-10 libgmp gmp)
+	find_library(GMP_LIBRARIES NAMES libgmp.a libgmp.dll.a gmp.lib libgmp-10 libgmp gmp)
 else(STBIN)
-	find_library(GMP_LIBRARIES NAMES libgmp.so gmp.lib libgmp-10 libgmp gmp)
+	find_library(GMP_LIBRARIES NAMES libgmp.so libgmp.dll.so gmp.lib libgmp-10 libgmp gmp)
 endif(STBIN)
 
 if(GMP_INCLUDE_DIR AND GMP_LIBRARIES)
@@ -61,7 +61,7 @@ endif()
 if(GMP_FOUND)
 	message(STATUS "Configured GMP: -I${GMP_INCLUDE_DIR} -L${GMP_LIBRARIES}")
 else(GMP_FOUND)
-	message(STATUS "Could NOT find GMP")
+	message(STATUS "Could NOT find GMP!")
 endif(GMP_FOUND)
 
 mark_as_advanced(GMP_INCLUDE_DIR GMP_LIBRARIES)