AuxXxilium/linux_dsm_epyc7002
1
0
Fork 0
mirror of https://github.com/AuxXxilium/linux_dsm_epyc7002.git synced 2024-12-28 11:18:45 +07:00
Code Issues Actions Packages Projects Releases Wiki Activity
e725f6456f
linux_dsm_epyc7002/arch/m68k/Kconfig.cpu

507 lines
12 KiB
Plaintext
Raw Normal View History

m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
comment "Processor Type"
m68k/Kconfig: Separate classic m68k and coldfire early While you can build multiplatform kernels for machines with classic m68k processors, you cannot mix support for classic m68k and coldfire processors. To avoid such hybrid kernels, introduce CONFIG_M68KCLASSIC as an antipole for CONFIG_COLDFIRE, and make all specific processor support depend on one of them. All classic m68k machine support also needs to depend on this. The defaults (CONFIG_M68KCLASSIC if MMU, CONFIG_COLDFIRE if !MMU) are chosen such to make most of the existing configs build and work. Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-12-27 02:32:02 +07:00
choice
prompt "CPU family support"
default M68KCLASSIC if MMU
default COLDFIRE if !MMU
help
The Freescale (was Motorola) M68K family of processors implements
the full 68000 processor instruction set.
Fix typo in various Kconfig file Correct spelling typo in various Kconfig file. Signed-off-by: Masanari Iida <standby24x7@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-04-13 22:14:11 +07:00
The Freescale ColdFire family of processors is a modern derivative
m68k/Kconfig: Separate classic m68k and coldfire early While you can build multiplatform kernels for machines with classic m68k processors, you cannot mix support for classic m68k and coldfire processors. To avoid such hybrid kernels, introduce CONFIG_M68KCLASSIC as an antipole for CONFIG_COLDFIRE, and make all specific processor support depend on one of them. All classic m68k machine support also needs to depend on this. The defaults (CONFIG_M68KCLASSIC if MMU, CONFIG_COLDFIRE if !MMU) are chosen such to make most of the existing configs build and work. Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-12-27 02:32:02 +07:00
of the 68000 processor family. They are mainly targeted at embedded
applications, and are all System-On-Chip (SOC) devices, as opposed
to stand alone CPUs. They implement a subset of the original 68000
processor instruction set.
If you anticipate running this kernel on a computer with a classic
MC68xxx processor, select M68KCLASSIC.
If you anticipate running this kernel on a computer with a ColdFire
processor, select COLDFIRE.
config M68KCLASSIC
bool "Classic M68K CPU family support"
config COLDFIRE
bool "Coldfire CPU family support"
gpiolib/arches: Centralise bolierplate asm/gpio.h Rather than requiring architectures that use gpiolib but don't have any need to define anything custom to copy an asm/gpio.h provide a Kconfig symbol which architectures must select in order to include gpio.h and for other architectures just provide the trivial implementation directly. This makes it much easier to do gpiolib updates and is also a step towards making gpiolib APIs available on every architecture. For architectures with existing boilerplate code leave a stub header in place which warns on direct inclusion of asm/gpio.h and includes linux/gpio.h to catch code that's doing this. Direct inclusion of asm/gpio.h has long been deprecated. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Jonas Bonn <jonas@southpole.se> Acked-by: Tony Luck <tony.luck@intel.com> Acked-by: Linus Walleij <linus.walleij@linaro.org> Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
2012-04-15 16:52:54 +07:00
select ARCH_HAVE_CUSTOM_GPIO_H
m68k/Kconfig: Separate classic m68k and coldfire early While you can build multiplatform kernels for machines with classic m68k processors, you cannot mix support for classic m68k and coldfire processors. To avoid such hybrid kernels, introduce CONFIG_M68KCLASSIC as an antipole for CONFIG_COLDFIRE, and make all specific processor support depend on one of them. All classic m68k machine support also needs to depend on this. The defaults (CONFIG_M68KCLASSIC if MMU, CONFIG_COLDFIRE if !MMU) are chosen such to make most of the existing configs build and work. Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-12-27 02:32:02 +07:00
select CPU_HAS_NO_BITFIELDS
select CPU_HAS_NO_MULDIV64
select GENERIC_CSUM
m68k: do away with ARCH_REQUIRE_GPIOLIB Replace "select ARCH_REQUIRE_GPIOLIB" with "select GPIOLIB" as this can now be selected directly. Cc: Michael Büsch <m@bues.ch> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: linux-m68k@lists.linux-m68k.org Acked-by: Greg Ungerer <gerg@linux-m68k.org> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-19 16:17:49 +07:00
select GPIOLIB
m68knommu: select CONFIG_HAVE_CLK for ColdFire CPU types The ColdFire CPU sub-arch has kernel clk code support, so select CONFIG_HAVE_CLK. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2012-08-03 13:08:36 +07:00
select HAVE_CLK
m68k/Kconfig: Separate classic m68k and coldfire early While you can build multiplatform kernels for machines with classic m68k processors, you cannot mix support for classic m68k and coldfire processors. To avoid such hybrid kernels, introduce CONFIG_M68KCLASSIC as an antipole for CONFIG_COLDFIRE, and make all specific processor support depend on one of them. All classic m68k machine support also needs to depend on this. The defaults (CONFIG_M68KCLASSIC if MMU, CONFIG_COLDFIRE if !MMU) are chosen such to make most of the existing configs build and work. Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-12-27 02:32:02 +07:00
endchoice
if M68KCLASSIC
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
config M68000
m68knommu: allow for configuration of true 68000 based systems Allow the M68000 option to be user configurable, for systems based on the original stand alone 68000 CPU. Signed-off-by: Luis Alves <ljalvs@gmail.com> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2012-10-24 06:45:23 +07:00
bool "MC68000"
m68knommu: disable MC68000 cpu target when MMU is selected As pointed out by Geert, MC68000 target needs to be disabled when MMU support is enabled. From Geert: This needs a "depends on !MMU". Else allmodconfig will select it, causing -m68000 to be passed to the assembler, which may break the build depending on your version of binutils, a.o. arch/m68k/kernel/entry.S:186: Error: invalid instruction for this architecture; needs 68020 or higher (68020 [68k, 68ec020], 68030 [68ec030], 68040 [68ec040], 68060 [68ec060]) -- statement `bfextu %sp@(50){#0,#4},%d0' ignored arch/m68k/kernel/entry.S:211: Error: invalid operand mode for this architecture; needs 68020 or higher -- statement `jbsr @(sys_call_table,%d0:l:4)@(0)' ignored Cfr. http://kisskb.ellerman.id.au/kisskb/buildresult/7416877/ Signed-off-by: Luis Alves <ljalvs@gmail.com> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2012-10-26 03:01:16 +07:00
depends on !MMU
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
select CPU_HAS_NO_BITFIELDS
m68k: handle presence of 64bit mul/div instructions cleanly The traditional 68000 processors and the newer reduced instruction set ColdFire processors do not support the 32*32->64 multiply or the 64/32->32 divide instructions. This is not a difference based on the presence of a hardware MMU or not. Create a new config symbol to mark that a CPU type doesn't support the longer multiply/divide instructions. Use this then as a basis for using the fast 64bit based divide (in div64.h) and for linking in the extra libgcc functions that may be required (mulsi3, divsi3, etc). Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
2011-11-11 12:13:08 +07:00
select CPU_HAS_NO_MULDIV64
m68k: Introduce config option CPU_HAS_NO_UNALIGNED Use CONFIG_CPU_HAS_NO_UNALIGNED instead of open coding CONFIG_M68000 || CONFIG_COLDFIRE Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Acked-by: Greg Ungerer<gerg@uclinux.org>
2012-06-07 00:37:52 +07:00
select CPU_HAS_NO_UNALIGNED
m68k: simpler m68k and ColdFire CPU's can use generic csum code We have two implementations of the IP checksuming code for the m68k arch. One uses the more advanced instructions available in 68020 and above processors, the other uses the simpler instructions available on the original 68000 processors and the modern ColdFire processors. This simpler code is pretty much the same as the generic lib implementation of the IP csum functions. So lets just switch over to using that. That means we can completely remove the checksum_no.c file, and only have the local fast code used for the more complex 68k CPU family members. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-10-18 12:49:19 +07:00
select GENERIC_CSUM
lib/GCD.c: use binary GCD algorithm instead of Euclidean The binary GCD algorithm is based on the following facts: 1. If a and b are all evens, then gcd(a,b) = 2 * gcd(a/2, b/2) 2. If a is even and b is odd, then gcd(a,b) = gcd(a/2, b) 3. If a and b are all odds, then gcd(a,b) = gcd((a-b)/2, b) = gcd((a+b)/2, b) Even on x86 machines with reasonable division hardware, the binary algorithm runs about 25% faster (80% the execution time) than the division-based Euclidian algorithm. On platforms like Alpha and ARMv6 where division is a function call to emulation code, it's even more significant. There are two variants of the code here, depending on whether a fast __ffs (find least significant set bit) instruction is available. This allows the unpredictable branches in the bit-at-a-time shifting loop to be eliminated. If fast __ffs is not available, the "even/odd" GCD variant is used. I use the following code to benchmark: #include <stdio.h> #include <stdlib.h> #include <stdint.h> #include <string.h> #include <time.h> #include <unistd.h> #define swap(a, b) \ do { \ a ^= b; \ b ^= a; \ a ^= b; \ } while (0) unsigned long gcd0(unsigned long a, unsigned long b) { unsigned long r; if (a < b) { swap(a, b); } if (b == 0) return a; while ((r = a % b) != 0) { a = b; b = r; } return b; } unsigned long gcd1(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); for (;;) { a >>= __builtin_ctzl(a); if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd2(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; for (;;) { while (!(a & r)) a >>= 1; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } unsigned long gcd3(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); if (b == 1) return r & -r; for (;;) { a >>= __builtin_ctzl(a); if (a == 1) return r & -r; if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd4(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; if (b == r) return r; for (;;) { while (!(a & r)) a >>= 1; if (a == r) return r; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } static unsigned long (*gcd_func[])(unsigned long a, unsigned long b) = { gcd0, gcd1, gcd2, gcd3, gcd4, }; #define TEST_ENTRIES (sizeof(gcd_func) / sizeof(gcd_func[0])) #if defined(__x86_64__) #define rdtscll(val) do { \ unsigned long __a,__d; \ __asm__ __volatile__("rdtsc" : "=a" (__a), "=d" (__d)); \ (val) = ((unsigned long long)__a) | (((unsigned long long)__d)<<32); \ } while(0) static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { unsigned long long start, end; unsigned long long ret; unsigned long gcd_res; rdtscll(start); gcd_res = gcd(a, b); rdtscll(end); if (end >= start) ret = end - start; else ret = ~0ULL - start + 1 + end; *res = gcd_res; return ret; } #else static inline struct timespec read_time(void) { struct timespec time; clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &time); return time; } static inline unsigned long long diff_time(struct timespec start, struct timespec end) { struct timespec temp; if ((end.tv_nsec - start.tv_nsec) < 0) { temp.tv_sec = end.tv_sec - start.tv_sec - 1; temp.tv_nsec = 1000000000ULL + end.tv_nsec - start.tv_nsec; } else { temp.tv_sec = end.tv_sec - start.tv_sec; temp.tv_nsec = end.tv_nsec - start.tv_nsec; } return temp.tv_sec * 1000000000ULL + temp.tv_nsec; } static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { struct timespec start, end; unsigned long gcd_res; start = read_time(); gcd_res = gcd(a, b); end = read_time(); *res = gcd_res; return diff_time(start, end); } #endif static inline unsigned long get_rand() { if (sizeof(long) == 8) return (unsigned long)rand() << 32 | rand(); else return rand(); } int main(int argc, char **argv) { unsigned int seed = time(0); int loops = 100; int repeats = 1000; unsigned long (*res)[TEST_ENTRIES]; unsigned long long elapsed[TEST_ENTRIES]; int i, j, k; for (;;) { int opt = getopt(argc, argv, "n:r:s:"); /* End condition always first */ if (opt == -1) break; switch (opt) { case 'n': loops = atoi(optarg); break; case 'r': repeats = atoi(optarg); break; case 's': seed = strtoul(optarg, NULL, 10); break; default: /* You won't actually get here. */ break; } } res = malloc(sizeof(unsigned long) * TEST_ENTRIES * loops); memset(elapsed, 0, sizeof(elapsed)); srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); /* Do we have args? */ unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); unsigned long long min_elapsed[TEST_ENTRIES]; for (k = 0; k < repeats; k++) { for (i = 0; i < TEST_ENTRIES; i++) { unsigned long long tmp = benchmark_gcd_func(gcd_func[i], a, b, &res[j][i]); if (k == 0 || min_elapsed[i] > tmp) min_elapsed[i] = tmp; } } for (i = 0; i < TEST_ENTRIES; i++) elapsed[i] += min_elapsed[i]; } for (i = 0; i < TEST_ENTRIES; i++) printf("gcd%d: elapsed %llu\n", i, elapsed[i]); k = 0; srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); for (i = 1; i < TEST_ENTRIES; i++) { if (res[j][i] != res[j][0]) break; } if (i < TEST_ENTRIES) { if (k == 0) { k = 1; fprintf(stderr, "Error:\n"); } fprintf(stderr, "gcd(%lu, %lu): ", a, b); for (i = 0; i < TEST_ENTRIES; i++) fprintf(stderr, "%ld%s", res[j][i], i < TEST_ENTRIES - 1 ? ", " : "\n"); } } if (k == 0) fprintf(stderr, "PASS\n"); free(res); return 0; } Compiled with "-O2", on "VirtualBox 4.4.0-22-generic #38-Ubuntu x86_64" got: zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 10174 gcd1: elapsed 2120 gcd2: elapsed 2902 gcd3: elapsed 2039 gcd4: elapsed 2812 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9309 gcd1: elapsed 2280 gcd2: elapsed 2822 gcd3: elapsed 2217 gcd4: elapsed 2710 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9589 gcd1: elapsed 2098 gcd2: elapsed 2815 gcd3: elapsed 2030 gcd4: elapsed 2718 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9914 gcd1: elapsed 2309 gcd2: elapsed 2779 gcd3: elapsed 2228 gcd4: elapsed 2709 PASS [akpm@linux-foundation.org: avoid #defining a CONFIG_ variable] Signed-off-by: Zhaoxiu Zeng <zhaoxiu.zeng@gmail.com> Signed-off-by: George Spelvin <linux@horizon.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 07:03:57 +07:00
select CPU_NO_EFFICIENT_FFS
m68k: Add <asm/hash.h> This provides a multiply by constant GOLDEN_RATIO_32 = 0x61C88647 for the original mc68000, which lacks a 32x32-bit multiply instruction. Yes, the amount of optimization effort put in is excessive. :-) Shift-add chain found by Yevgen Voronenko's Hcub algorithm at http://spiral.ece.cmu.edu/mcm/gen.html Signed-off-by: George Spelvin <linux@sciencehorizons.net> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Greg Ungerer <gerg@linux-m68k.org> Cc: Andreas Schwab <schwab@linux-m68k.org> Cc: Philippe De Muyter <phdm@macq.eu> Cc: linux-m68k@lists.linux-m68k.org
2016-05-26 22:36:19 +07:00
select HAVE_ARCH_HASH
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
The Freescale (was Motorola) 68000 CPU is the first generation of
the well known M68K family of processors. The CPU core as well as
being available as a stand alone CPU was also used in many
System-On-Chip devices (eg 68328, 68302, etc). It does not contain
a paging MMU.
config MCPU32
bool
select CPU_HAS_NO_BITFIELDS
m68k: CPU32 does not support unaligned accesses Hence select CPU_HAS_NO_UNALIGNED Reported-by: Philippe De Muyter <phdm@macqel.be> Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Acked-by: Greg Ungerer<gerg@uclinux.org>
2012-06-07 00:39:39 +07:00
select CPU_HAS_NO_UNALIGNED
lib/GCD.c: use binary GCD algorithm instead of Euclidean The binary GCD algorithm is based on the following facts: 1. If a and b are all evens, then gcd(a,b) = 2 * gcd(a/2, b/2) 2. If a is even and b is odd, then gcd(a,b) = gcd(a/2, b) 3. If a and b are all odds, then gcd(a,b) = gcd((a-b)/2, b) = gcd((a+b)/2, b) Even on x86 machines with reasonable division hardware, the binary algorithm runs about 25% faster (80% the execution time) than the division-based Euclidian algorithm. On platforms like Alpha and ARMv6 where division is a function call to emulation code, it's even more significant. There are two variants of the code here, depending on whether a fast __ffs (find least significant set bit) instruction is available. This allows the unpredictable branches in the bit-at-a-time shifting loop to be eliminated. If fast __ffs is not available, the "even/odd" GCD variant is used. I use the following code to benchmark: #include <stdio.h> #include <stdlib.h> #include <stdint.h> #include <string.h> #include <time.h> #include <unistd.h> #define swap(a, b) \ do { \ a ^= b; \ b ^= a; \ a ^= b; \ } while (0) unsigned long gcd0(unsigned long a, unsigned long b) { unsigned long r; if (a < b) { swap(a, b); } if (b == 0) return a; while ((r = a % b) != 0) { a = b; b = r; } return b; } unsigned long gcd1(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); for (;;) { a >>= __builtin_ctzl(a); if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd2(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; for (;;) { while (!(a & r)) a >>= 1; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } unsigned long gcd3(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); if (b == 1) return r & -r; for (;;) { a >>= __builtin_ctzl(a); if (a == 1) return r & -r; if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd4(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; if (b == r) return r; for (;;) { while (!(a & r)) a >>= 1; if (a == r) return r; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } static unsigned long (*gcd_func[])(unsigned long a, unsigned long b) = { gcd0, gcd1, gcd2, gcd3, gcd4, }; #define TEST_ENTRIES (sizeof(gcd_func) / sizeof(gcd_func[0])) #if defined(__x86_64__) #define rdtscll(val) do { \ unsigned long __a,__d; \ __asm__ __volatile__("rdtsc" : "=a" (__a), "=d" (__d)); \ (val) = ((unsigned long long)__a) | (((unsigned long long)__d)<<32); \ } while(0) static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { unsigned long long start, end; unsigned long long ret; unsigned long gcd_res; rdtscll(start); gcd_res = gcd(a, b); rdtscll(end); if (end >= start) ret = end - start; else ret = ~0ULL - start + 1 + end; *res = gcd_res; return ret; } #else static inline struct timespec read_time(void) { struct timespec time; clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &time); return time; } static inline unsigned long long diff_time(struct timespec start, struct timespec end) { struct timespec temp; if ((end.tv_nsec - start.tv_nsec) < 0) { temp.tv_sec = end.tv_sec - start.tv_sec - 1; temp.tv_nsec = 1000000000ULL + end.tv_nsec - start.tv_nsec; } else { temp.tv_sec = end.tv_sec - start.tv_sec; temp.tv_nsec = end.tv_nsec - start.tv_nsec; } return temp.tv_sec * 1000000000ULL + temp.tv_nsec; } static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { struct timespec start, end; unsigned long gcd_res; start = read_time(); gcd_res = gcd(a, b); end = read_time(); *res = gcd_res; return diff_time(start, end); } #endif static inline unsigned long get_rand() { if (sizeof(long) == 8) return (unsigned long)rand() << 32 | rand(); else return rand(); } int main(int argc, char **argv) { unsigned int seed = time(0); int loops = 100; int repeats = 1000; unsigned long (*res)[TEST_ENTRIES]; unsigned long long elapsed[TEST_ENTRIES]; int i, j, k; for (;;) { int opt = getopt(argc, argv, "n:r:s:"); /* End condition always first */ if (opt == -1) break; switch (opt) { case 'n': loops = atoi(optarg); break; case 'r': repeats = atoi(optarg); break; case 's': seed = strtoul(optarg, NULL, 10); break; default: /* You won't actually get here. */ break; } } res = malloc(sizeof(unsigned long) * TEST_ENTRIES * loops); memset(elapsed, 0, sizeof(elapsed)); srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); /* Do we have args? */ unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); unsigned long long min_elapsed[TEST_ENTRIES]; for (k = 0; k < repeats; k++) { for (i = 0; i < TEST_ENTRIES; i++) { unsigned long long tmp = benchmark_gcd_func(gcd_func[i], a, b, &res[j][i]); if (k == 0 || min_elapsed[i] > tmp) min_elapsed[i] = tmp; } } for (i = 0; i < TEST_ENTRIES; i++) elapsed[i] += min_elapsed[i]; } for (i = 0; i < TEST_ENTRIES; i++) printf("gcd%d: elapsed %llu\n", i, elapsed[i]); k = 0; srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); for (i = 1; i < TEST_ENTRIES; i++) { if (res[j][i] != res[j][0]) break; } if (i < TEST_ENTRIES) { if (k == 0) { k = 1; fprintf(stderr, "Error:\n"); } fprintf(stderr, "gcd(%lu, %lu): ", a, b); for (i = 0; i < TEST_ENTRIES; i++) fprintf(stderr, "%ld%s", res[j][i], i < TEST_ENTRIES - 1 ? ", " : "\n"); } } if (k == 0) fprintf(stderr, "PASS\n"); free(res); return 0; } Compiled with "-O2", on "VirtualBox 4.4.0-22-generic #38-Ubuntu x86_64" got: zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 10174 gcd1: elapsed 2120 gcd2: elapsed 2902 gcd3: elapsed 2039 gcd4: elapsed 2812 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9309 gcd1: elapsed 2280 gcd2: elapsed 2822 gcd3: elapsed 2217 gcd4: elapsed 2710 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9589 gcd1: elapsed 2098 gcd2: elapsed 2815 gcd3: elapsed 2030 gcd4: elapsed 2718 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9914 gcd1: elapsed 2309 gcd2: elapsed 2779 gcd3: elapsed 2228 gcd4: elapsed 2709 PASS [akpm@linux-foundation.org: avoid #defining a CONFIG_ variable] Signed-off-by: Zhaoxiu Zeng <zhaoxiu.zeng@gmail.com> Signed-off-by: George Spelvin <linux@horizon.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 07:03:57 +07:00
select CPU_NO_EFFICIENT_FFS
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
The Freescale (was then Motorola) CPU32 is a CPU core that is
based on the 68020 processor. For the most part it is used in
System-On-Chip parts, and does not contain a paging MMU.
config M68020
bool "68020 support"
depends on MMU
m68k: move CONFIG_FPU set to per-CPU configuration Move the selection of CONFIG_FPU to each CPU type configuration. Currently for m68k we have a global set of CONFIG_FPU based on if CONFIG_MMU is enabled or not. There is at least one CPU family we support (m5441x) that has an MMU but has no FPU hardware. So we need to be able to have CONFIG_MMU set and CONFIG_FPU not set. Whether we build for a CPU with MMU enabled or not doesn't change the fact that it has FPU hardware support. Our current non-MMU builds have never had CONIG_FPU enabled - and in fact the kernel will not compile with that set and CONFIG_MMU not set at the moment. It is easy enough to fix this - but it would involve a structure change to sigcontext.h, and that is a user space exported header (so ABI change). This change makes no configuration visible changes, and all configs end up with the same configuration settings as before. This change based on changes and discussion from Yannick Gicquel <yannick.gicquel@open.eurogiciel.org>. Signed-off-by: Greg Ungerer <gerg@linux-m68k.org>
2016-08-24 10:32:01 +07:00
select FPU
m68k: modify user space access functions to support ColdFire CPUs Modify the user space access functions to support the ColdFire V4e cores running with MMU enabled. The ColdFire processors do not support the "moves" instruction used by the traditional 680x0 processors for moving data into and out of another address space. They only support the notion of a single address space, and you use the usual "move" instruction to access that. Create a new config symbol (CONFIG_CPU_HAS_ADDRESS_SPACES) to mark the CPU types that support separate address spaces, and thus also support the sfc/dfc registers and the "moves" instruction that go along with that. The code is almost identical for user space access, so lets just use a define to choose either the "move" or "moves" in the assembler code. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Matt Waddel <mwaddel@yahoo.com> Acked-by: Kurt Mahan <kmahan@xmission.com> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
2011-10-14 11:43:30 +07:00
select CPU_HAS_ADDRESS_SPACES
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
If you anticipate running this kernel on a computer with a MC68020
processor, say Y. Otherwise, say N. Note that the 68020 requires a
68851 MMU (Memory Management Unit) to run Linux/m68k, except on the
Sun 3, which provides its own version.
config M68030
bool "68030 support"
depends on MMU && !MMU_SUN3
m68k: move CONFIG_FPU set to per-CPU configuration Move the selection of CONFIG_FPU to each CPU type configuration. Currently for m68k we have a global set of CONFIG_FPU based on if CONFIG_MMU is enabled or not. There is at least one CPU family we support (m5441x) that has an MMU but has no FPU hardware. So we need to be able to have CONFIG_MMU set and CONFIG_FPU not set. Whether we build for a CPU with MMU enabled or not doesn't change the fact that it has FPU hardware support. Our current non-MMU builds have never had CONIG_FPU enabled - and in fact the kernel will not compile with that set and CONFIG_MMU not set at the moment. It is easy enough to fix this - but it would involve a structure change to sigcontext.h, and that is a user space exported header (so ABI change). This change makes no configuration visible changes, and all configs end up with the same configuration settings as before. This change based on changes and discussion from Yannick Gicquel <yannick.gicquel@open.eurogiciel.org>. Signed-off-by: Greg Ungerer <gerg@linux-m68k.org>
2016-08-24 10:32:01 +07:00
select FPU
m68k: modify user space access functions to support ColdFire CPUs Modify the user space access functions to support the ColdFire V4e cores running with MMU enabled. The ColdFire processors do not support the "moves" instruction used by the traditional 680x0 processors for moving data into and out of another address space. They only support the notion of a single address space, and you use the usual "move" instruction to access that. Create a new config symbol (CONFIG_CPU_HAS_ADDRESS_SPACES) to mark the CPU types that support separate address spaces, and thus also support the sfc/dfc registers and the "moves" instruction that go along with that. The code is almost identical for user space access, so lets just use a define to choose either the "move" or "moves" in the assembler code. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Matt Waddel <mwaddel@yahoo.com> Acked-by: Kurt Mahan <kmahan@xmission.com> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
2011-10-14 11:43:30 +07:00
select CPU_HAS_ADDRESS_SPACES
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
If you anticipate running this kernel on a computer with a MC68030
processor, say Y. Otherwise, say N. Note that a MC68EC030 will not
work, as it does not include an MMU (Memory Management Unit).
config M68040
bool "68040 support"
depends on MMU && !MMU_SUN3
m68k: move CONFIG_FPU set to per-CPU configuration Move the selection of CONFIG_FPU to each CPU type configuration. Currently for m68k we have a global set of CONFIG_FPU based on if CONFIG_MMU is enabled or not. There is at least one CPU family we support (m5441x) that has an MMU but has no FPU hardware. So we need to be able to have CONFIG_MMU set and CONFIG_FPU not set. Whether we build for a CPU with MMU enabled or not doesn't change the fact that it has FPU hardware support. Our current non-MMU builds have never had CONIG_FPU enabled - and in fact the kernel will not compile with that set and CONFIG_MMU not set at the moment. It is easy enough to fix this - but it would involve a structure change to sigcontext.h, and that is a user space exported header (so ABI change). This change makes no configuration visible changes, and all configs end up with the same configuration settings as before. This change based on changes and discussion from Yannick Gicquel <yannick.gicquel@open.eurogiciel.org>. Signed-off-by: Greg Ungerer <gerg@linux-m68k.org>
2016-08-24 10:32:01 +07:00
select FPU
m68k: modify user space access functions to support ColdFire CPUs Modify the user space access functions to support the ColdFire V4e cores running with MMU enabled. The ColdFire processors do not support the "moves" instruction used by the traditional 680x0 processors for moving data into and out of another address space. They only support the notion of a single address space, and you use the usual "move" instruction to access that. Create a new config symbol (CONFIG_CPU_HAS_ADDRESS_SPACES) to mark the CPU types that support separate address spaces, and thus also support the sfc/dfc registers and the "moves" instruction that go along with that. The code is almost identical for user space access, so lets just use a define to choose either the "move" or "moves" in the assembler code. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Matt Waddel <mwaddel@yahoo.com> Acked-by: Kurt Mahan <kmahan@xmission.com> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
2011-10-14 11:43:30 +07:00
select CPU_HAS_ADDRESS_SPACES
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
If you anticipate running this kernel on a computer with a MC68LC040
or MC68040 processor, say Y. Otherwise, say N. Note that an
MC68EC040 will not work, as it does not include an MMU (Memory
Management Unit).
config M68060
bool "68060 support"
depends on MMU && !MMU_SUN3
m68k: move CONFIG_FPU set to per-CPU configuration Move the selection of CONFIG_FPU to each CPU type configuration. Currently for m68k we have a global set of CONFIG_FPU based on if CONFIG_MMU is enabled or not. There is at least one CPU family we support (m5441x) that has an MMU but has no FPU hardware. So we need to be able to have CONFIG_MMU set and CONFIG_FPU not set. Whether we build for a CPU with MMU enabled or not doesn't change the fact that it has FPU hardware support. Our current non-MMU builds have never had CONIG_FPU enabled - and in fact the kernel will not compile with that set and CONFIG_MMU not set at the moment. It is easy enough to fix this - but it would involve a structure change to sigcontext.h, and that is a user space exported header (so ABI change). This change makes no configuration visible changes, and all configs end up with the same configuration settings as before. This change based on changes and discussion from Yannick Gicquel <yannick.gicquel@open.eurogiciel.org>. Signed-off-by: Greg Ungerer <gerg@linux-m68k.org>
2016-08-24 10:32:01 +07:00
select FPU
m68k: modify user space access functions to support ColdFire CPUs Modify the user space access functions to support the ColdFire V4e cores running with MMU enabled. The ColdFire processors do not support the "moves" instruction used by the traditional 680x0 processors for moving data into and out of another address space. They only support the notion of a single address space, and you use the usual "move" instruction to access that. Create a new config symbol (CONFIG_CPU_HAS_ADDRESS_SPACES) to mark the CPU types that support separate address spaces, and thus also support the sfc/dfc registers and the "moves" instruction that go along with that. The code is almost identical for user space access, so lets just use a define to choose either the "move" or "moves" in the assembler code. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Matt Waddel <mwaddel@yahoo.com> Acked-by: Kurt Mahan <kmahan@xmission.com> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
2011-10-14 11:43:30 +07:00
select CPU_HAS_ADDRESS_SPACES
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
If you anticipate running this kernel on a computer with a MC68060
processor, say Y. Otherwise, say N.
config M68328
bool "MC68328"
depends on !MMU
select M68000
help
Motorola 68328 processor support.
config M68EZ328
bool "MC68EZ328"
depends on !MMU
select M68000
help
Motorola 68EX328 processor support.
config M68VZ328
bool "MC68VZ328"
depends on !MMU
select M68000
help
Motorola 68VZ328 processor support.
m68k/Kconfig: Separate classic m68k and coldfire early While you can build multiplatform kernels for machines with classic m68k processors, you cannot mix support for classic m68k and coldfire processors. To avoid such hybrid kernels, introduce CONFIG_M68KCLASSIC as an antipole for CONFIG_COLDFIRE, and make all specific processor support depend on one of them. All classic m68k machine support also needs to depend on this. The defaults (CONFIG_M68KCLASSIC if MMU, CONFIG_COLDFIRE if !MMU) are chosen such to make most of the existing configs build and work. Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-12-27 02:32:02 +07:00
endif # M68KCLASSIC
if COLDFIRE
m68knommu: make ColdFire SoC selection a choice It would be nice if we could support multiple ColdFire SoC types in a single binary - but currently the code simply does not support it. Change the SoC selection config options to be a choice instead of individual selectable entries. This fixes problems with building allnoconfig, and means that a sane linux kernel is generated for a single ColdFire SoC type. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
2015-07-07 12:44:02 +07:00
choice
prompt "ColdFire SoC type"
default M520x
help
Select the type of ColdFire System-on-Chip (SoC) that you want
to build for.
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
config M5206
bool "MCF5206"
depends on !MMU
select COLDFIRE_SW_A7
select HAVE_MBAR
lib/GCD.c: use binary GCD algorithm instead of Euclidean The binary GCD algorithm is based on the following facts: 1. If a and b are all evens, then gcd(a,b) = 2 * gcd(a/2, b/2) 2. If a is even and b is odd, then gcd(a,b) = gcd(a/2, b) 3. If a and b are all odds, then gcd(a,b) = gcd((a-b)/2, b) = gcd((a+b)/2, b) Even on x86 machines with reasonable division hardware, the binary algorithm runs about 25% faster (80% the execution time) than the division-based Euclidian algorithm. On platforms like Alpha and ARMv6 where division is a function call to emulation code, it's even more significant. There are two variants of the code here, depending on whether a fast __ffs (find least significant set bit) instruction is available. This allows the unpredictable branches in the bit-at-a-time shifting loop to be eliminated. If fast __ffs is not available, the "even/odd" GCD variant is used. I use the following code to benchmark: #include <stdio.h> #include <stdlib.h> #include <stdint.h> #include <string.h> #include <time.h> #include <unistd.h> #define swap(a, b) \ do { \ a ^= b; \ b ^= a; \ a ^= b; \ } while (0) unsigned long gcd0(unsigned long a, unsigned long b) { unsigned long r; if (a < b) { swap(a, b); } if (b == 0) return a; while ((r = a % b) != 0) { a = b; b = r; } return b; } unsigned long gcd1(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); for (;;) { a >>= __builtin_ctzl(a); if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd2(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; for (;;) { while (!(a & r)) a >>= 1; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } unsigned long gcd3(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); if (b == 1) return r & -r; for (;;) { a >>= __builtin_ctzl(a); if (a == 1) return r & -r; if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd4(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; if (b == r) return r; for (;;) { while (!(a & r)) a >>= 1; if (a == r) return r; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } static unsigned long (*gcd_func[])(unsigned long a, unsigned long b) = { gcd0, gcd1, gcd2, gcd3, gcd4, }; #define TEST_ENTRIES (sizeof(gcd_func) / sizeof(gcd_func[0])) #if defined(__x86_64__) #define rdtscll(val) do { \ unsigned long __a,__d; \ __asm__ __volatile__("rdtsc" : "=a" (__a), "=d" (__d)); \ (val) = ((unsigned long long)__a) | (((unsigned long long)__d)<<32); \ } while(0) static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { unsigned long long start, end; unsigned long long ret; unsigned long gcd_res; rdtscll(start); gcd_res = gcd(a, b); rdtscll(end); if (end >= start) ret = end - start; else ret = ~0ULL - start + 1 + end; *res = gcd_res; return ret; } #else static inline struct timespec read_time(void) { struct timespec time; clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &time); return time; } static inline unsigned long long diff_time(struct timespec start, struct timespec end) { struct timespec temp; if ((end.tv_nsec - start.tv_nsec) < 0) { temp.tv_sec = end.tv_sec - start.tv_sec - 1; temp.tv_nsec = 1000000000ULL + end.tv_nsec - start.tv_nsec; } else { temp.tv_sec = end.tv_sec - start.tv_sec; temp.tv_nsec = end.tv_nsec - start.tv_nsec; } return temp.tv_sec * 1000000000ULL + temp.tv_nsec; } static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { struct timespec start, end; unsigned long gcd_res; start = read_time(); gcd_res = gcd(a, b); end = read_time(); *res = gcd_res; return diff_time(start, end); } #endif static inline unsigned long get_rand() { if (sizeof(long) == 8) return (unsigned long)rand() << 32 | rand(); else return rand(); } int main(int argc, char **argv) { unsigned int seed = time(0); int loops = 100; int repeats = 1000; unsigned long (*res)[TEST_ENTRIES]; unsigned long long elapsed[TEST_ENTRIES]; int i, j, k; for (;;) { int opt = getopt(argc, argv, "n:r:s:"); /* End condition always first */ if (opt == -1) break; switch (opt) { case 'n': loops = atoi(optarg); break; case 'r': repeats = atoi(optarg); break; case 's': seed = strtoul(optarg, NULL, 10); break; default: /* You won't actually get here. */ break; } } res = malloc(sizeof(unsigned long) * TEST_ENTRIES * loops); memset(elapsed, 0, sizeof(elapsed)); srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); /* Do we have args? */ unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); unsigned long long min_elapsed[TEST_ENTRIES]; for (k = 0; k < repeats; k++) { for (i = 0; i < TEST_ENTRIES; i++) { unsigned long long tmp = benchmark_gcd_func(gcd_func[i], a, b, &res[j][i]); if (k == 0 || min_elapsed[i] > tmp) min_elapsed[i] = tmp; } } for (i = 0; i < TEST_ENTRIES; i++) elapsed[i] += min_elapsed[i]; } for (i = 0; i < TEST_ENTRIES; i++) printf("gcd%d: elapsed %llu\n", i, elapsed[i]); k = 0; srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); for (i = 1; i < TEST_ENTRIES; i++) { if (res[j][i] != res[j][0]) break; } if (i < TEST_ENTRIES) { if (k == 0) { k = 1; fprintf(stderr, "Error:\n"); } fprintf(stderr, "gcd(%lu, %lu): ", a, b); for (i = 0; i < TEST_ENTRIES; i++) fprintf(stderr, "%ld%s", res[j][i], i < TEST_ENTRIES - 1 ? ", " : "\n"); } } if (k == 0) fprintf(stderr, "PASS\n"); free(res); return 0; } Compiled with "-O2", on "VirtualBox 4.4.0-22-generic #38-Ubuntu x86_64" got: zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 10174 gcd1: elapsed 2120 gcd2: elapsed 2902 gcd3: elapsed 2039 gcd4: elapsed 2812 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9309 gcd1: elapsed 2280 gcd2: elapsed 2822 gcd3: elapsed 2217 gcd4: elapsed 2710 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9589 gcd1: elapsed 2098 gcd2: elapsed 2815 gcd3: elapsed 2030 gcd4: elapsed 2718 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9914 gcd1: elapsed 2309 gcd2: elapsed 2779 gcd3: elapsed 2228 gcd4: elapsed 2709 PASS [akpm@linux-foundation.org: avoid #defining a CONFIG_ variable] Signed-off-by: Zhaoxiu Zeng <zhaoxiu.zeng@gmail.com> Signed-off-by: George Spelvin <linux@horizon.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 07:03:57 +07:00
select CPU_NO_EFFICIENT_FFS
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
Motorola ColdFire 5206 processor support.
config M5206e
bool "MCF5206e"
depends on !MMU
select COLDFIRE_SW_A7
select HAVE_MBAR
lib/GCD.c: use binary GCD algorithm instead of Euclidean The binary GCD algorithm is based on the following facts: 1. If a and b are all evens, then gcd(a,b) = 2 * gcd(a/2, b/2) 2. If a is even and b is odd, then gcd(a,b) = gcd(a/2, b) 3. If a and b are all odds, then gcd(a,b) = gcd((a-b)/2, b) = gcd((a+b)/2, b) Even on x86 machines with reasonable division hardware, the binary algorithm runs about 25% faster (80% the execution time) than the division-based Euclidian algorithm. On platforms like Alpha and ARMv6 where division is a function call to emulation code, it's even more significant. There are two variants of the code here, depending on whether a fast __ffs (find least significant set bit) instruction is available. This allows the unpredictable branches in the bit-at-a-time shifting loop to be eliminated. If fast __ffs is not available, the "even/odd" GCD variant is used. I use the following code to benchmark: #include <stdio.h> #include <stdlib.h> #include <stdint.h> #include <string.h> #include <time.h> #include <unistd.h> #define swap(a, b) \ do { \ a ^= b; \ b ^= a; \ a ^= b; \ } while (0) unsigned long gcd0(unsigned long a, unsigned long b) { unsigned long r; if (a < b) { swap(a, b); } if (b == 0) return a; while ((r = a % b) != 0) { a = b; b = r; } return b; } unsigned long gcd1(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); for (;;) { a >>= __builtin_ctzl(a); if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd2(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; for (;;) { while (!(a & r)) a >>= 1; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } unsigned long gcd3(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); if (b == 1) return r & -r; for (;;) { a >>= __builtin_ctzl(a); if (a == 1) return r & -r; if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd4(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; if (b == r) return r; for (;;) { while (!(a & r)) a >>= 1; if (a == r) return r; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } static unsigned long (*gcd_func[])(unsigned long a, unsigned long b) = { gcd0, gcd1, gcd2, gcd3, gcd4, }; #define TEST_ENTRIES (sizeof(gcd_func) / sizeof(gcd_func[0])) #if defined(__x86_64__) #define rdtscll(val) do { \ unsigned long __a,__d; \ __asm__ __volatile__("rdtsc" : "=a" (__a), "=d" (__d)); \ (val) = ((unsigned long long)__a) | (((unsigned long long)__d)<<32); \ } while(0) static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { unsigned long long start, end; unsigned long long ret; unsigned long gcd_res; rdtscll(start); gcd_res = gcd(a, b); rdtscll(end); if (end >= start) ret = end - start; else ret = ~0ULL - start + 1 + end; *res = gcd_res; return ret; } #else static inline struct timespec read_time(void) { struct timespec time; clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &time); return time; } static inline unsigned long long diff_time(struct timespec start, struct timespec end) { struct timespec temp; if ((end.tv_nsec - start.tv_nsec) < 0) { temp.tv_sec = end.tv_sec - start.tv_sec - 1; temp.tv_nsec = 1000000000ULL + end.tv_nsec - start.tv_nsec; } else { temp.tv_sec = end.tv_sec - start.tv_sec; temp.tv_nsec = end.tv_nsec - start.tv_nsec; } return temp.tv_sec * 1000000000ULL + temp.tv_nsec; } static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { struct timespec start, end; unsigned long gcd_res; start = read_time(); gcd_res = gcd(a, b); end = read_time(); *res = gcd_res; return diff_time(start, end); } #endif static inline unsigned long get_rand() { if (sizeof(long) == 8) return (unsigned long)rand() << 32 | rand(); else return rand(); } int main(int argc, char **argv) { unsigned int seed = time(0); int loops = 100; int repeats = 1000; unsigned long (*res)[TEST_ENTRIES]; unsigned long long elapsed[TEST_ENTRIES]; int i, j, k; for (;;) { int opt = getopt(argc, argv, "n:r:s:"); /* End condition always first */ if (opt == -1) break; switch (opt) { case 'n': loops = atoi(optarg); break; case 'r': repeats = atoi(optarg); break; case 's': seed = strtoul(optarg, NULL, 10); break; default: /* You won't actually get here. */ break; } } res = malloc(sizeof(unsigned long) * TEST_ENTRIES * loops); memset(elapsed, 0, sizeof(elapsed)); srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); /* Do we have args? */ unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); unsigned long long min_elapsed[TEST_ENTRIES]; for (k = 0; k < repeats; k++) { for (i = 0; i < TEST_ENTRIES; i++) { unsigned long long tmp = benchmark_gcd_func(gcd_func[i], a, b, &res[j][i]); if (k == 0 || min_elapsed[i] > tmp) min_elapsed[i] = tmp; } } for (i = 0; i < TEST_ENTRIES; i++) elapsed[i] += min_elapsed[i]; } for (i = 0; i < TEST_ENTRIES; i++) printf("gcd%d: elapsed %llu\n", i, elapsed[i]); k = 0; srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); for (i = 1; i < TEST_ENTRIES; i++) { if (res[j][i] != res[j][0]) break; } if (i < TEST_ENTRIES) { if (k == 0) { k = 1; fprintf(stderr, "Error:\n"); } fprintf(stderr, "gcd(%lu, %lu): ", a, b); for (i = 0; i < TEST_ENTRIES; i++) fprintf(stderr, "%ld%s", res[j][i], i < TEST_ENTRIES - 1 ? ", " : "\n"); } } if (k == 0) fprintf(stderr, "PASS\n"); free(res); return 0; } Compiled with "-O2", on "VirtualBox 4.4.0-22-generic #38-Ubuntu x86_64" got: zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 10174 gcd1: elapsed 2120 gcd2: elapsed 2902 gcd3: elapsed 2039 gcd4: elapsed 2812 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9309 gcd1: elapsed 2280 gcd2: elapsed 2822 gcd3: elapsed 2217 gcd4: elapsed 2710 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9589 gcd1: elapsed 2098 gcd2: elapsed 2815 gcd3: elapsed 2030 gcd4: elapsed 2718 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9914 gcd1: elapsed 2309 gcd2: elapsed 2779 gcd3: elapsed 2228 gcd4: elapsed 2709 PASS [akpm@linux-foundation.org: avoid #defining a CONFIG_ variable] Signed-off-by: Zhaoxiu Zeng <zhaoxiu.zeng@gmail.com> Signed-off-by: George Spelvin <linux@horizon.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 07:03:57 +07:00
select CPU_NO_EFFICIENT_FFS
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
Motorola ColdFire 5206e processor support.
config M520x
bool "MCF520x"
depends on !MMU
select GENERIC_CLOCKEVENTS
select HAVE_CACHE_SPLIT
help
Freescale Coldfire 5207/5208 processor support.
config M523x
bool "MCF523x"
depends on !MMU
select GENERIC_CLOCKEVENTS
select HAVE_CACHE_SPLIT
select HAVE_IPSBAR
help
Freescale Coldfire 5230/1/2/4/5 processor support
config M5249
bool "MCF5249"
depends on !MMU
select COLDFIRE_SW_A7
select HAVE_MBAR
lib/GCD.c: use binary GCD algorithm instead of Euclidean The binary GCD algorithm is based on the following facts: 1. If a and b are all evens, then gcd(a,b) = 2 * gcd(a/2, b/2) 2. If a is even and b is odd, then gcd(a,b) = gcd(a/2, b) 3. If a and b are all odds, then gcd(a,b) = gcd((a-b)/2, b) = gcd((a+b)/2, b) Even on x86 machines with reasonable division hardware, the binary algorithm runs about 25% faster (80% the execution time) than the division-based Euclidian algorithm. On platforms like Alpha and ARMv6 where division is a function call to emulation code, it's even more significant. There are two variants of the code here, depending on whether a fast __ffs (find least significant set bit) instruction is available. This allows the unpredictable branches in the bit-at-a-time shifting loop to be eliminated. If fast __ffs is not available, the "even/odd" GCD variant is used. I use the following code to benchmark: #include <stdio.h> #include <stdlib.h> #include <stdint.h> #include <string.h> #include <time.h> #include <unistd.h> #define swap(a, b) \ do { \ a ^= b; \ b ^= a; \ a ^= b; \ } while (0) unsigned long gcd0(unsigned long a, unsigned long b) { unsigned long r; if (a < b) { swap(a, b); } if (b == 0) return a; while ((r = a % b) != 0) { a = b; b = r; } return b; } unsigned long gcd1(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); for (;;) { a >>= __builtin_ctzl(a); if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd2(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; for (;;) { while (!(a & r)) a >>= 1; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } unsigned long gcd3(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); if (b == 1) return r & -r; for (;;) { a >>= __builtin_ctzl(a); if (a == 1) return r & -r; if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd4(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; if (b == r) return r; for (;;) { while (!(a & r)) a >>= 1; if (a == r) return r; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } static unsigned long (*gcd_func[])(unsigned long a, unsigned long b) = { gcd0, gcd1, gcd2, gcd3, gcd4, }; #define TEST_ENTRIES (sizeof(gcd_func) / sizeof(gcd_func[0])) #if defined(__x86_64__) #define rdtscll(val) do { \ unsigned long __a,__d; \ __asm__ __volatile__("rdtsc" : "=a" (__a), "=d" (__d)); \ (val) = ((unsigned long long)__a) | (((unsigned long long)__d)<<32); \ } while(0) static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { unsigned long long start, end; unsigned long long ret; unsigned long gcd_res; rdtscll(start); gcd_res = gcd(a, b); rdtscll(end); if (end >= start) ret = end - start; else ret = ~0ULL - start + 1 + end; *res = gcd_res; return ret; } #else static inline struct timespec read_time(void) { struct timespec time; clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &time); return time; } static inline unsigned long long diff_time(struct timespec start, struct timespec end) { struct timespec temp; if ((end.tv_nsec - start.tv_nsec) < 0) { temp.tv_sec = end.tv_sec - start.tv_sec - 1; temp.tv_nsec = 1000000000ULL + end.tv_nsec - start.tv_nsec; } else { temp.tv_sec = end.tv_sec - start.tv_sec; temp.tv_nsec = end.tv_nsec - start.tv_nsec; } return temp.tv_sec * 1000000000ULL + temp.tv_nsec; } static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { struct timespec start, end; unsigned long gcd_res; start = read_time(); gcd_res = gcd(a, b); end = read_time(); *res = gcd_res; return diff_time(start, end); } #endif static inline unsigned long get_rand() { if (sizeof(long) == 8) return (unsigned long)rand() << 32 | rand(); else return rand(); } int main(int argc, char **argv) { unsigned int seed = time(0); int loops = 100; int repeats = 1000; unsigned long (*res)[TEST_ENTRIES]; unsigned long long elapsed[TEST_ENTRIES]; int i, j, k; for (;;) { int opt = getopt(argc, argv, "n:r:s:"); /* End condition always first */ if (opt == -1) break; switch (opt) { case 'n': loops = atoi(optarg); break; case 'r': repeats = atoi(optarg); break; case 's': seed = strtoul(optarg, NULL, 10); break; default: /* You won't actually get here. */ break; } } res = malloc(sizeof(unsigned long) * TEST_ENTRIES * loops); memset(elapsed, 0, sizeof(elapsed)); srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); /* Do we have args? */ unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); unsigned long long min_elapsed[TEST_ENTRIES]; for (k = 0; k < repeats; k++) { for (i = 0; i < TEST_ENTRIES; i++) { unsigned long long tmp = benchmark_gcd_func(gcd_func[i], a, b, &res[j][i]); if (k == 0 || min_elapsed[i] > tmp) min_elapsed[i] = tmp; } } for (i = 0; i < TEST_ENTRIES; i++) elapsed[i] += min_elapsed[i]; } for (i = 0; i < TEST_ENTRIES; i++) printf("gcd%d: elapsed %llu\n", i, elapsed[i]); k = 0; srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); for (i = 1; i < TEST_ENTRIES; i++) { if (res[j][i] != res[j][0]) break; } if (i < TEST_ENTRIES) { if (k == 0) { k = 1; fprintf(stderr, "Error:\n"); } fprintf(stderr, "gcd(%lu, %lu): ", a, b); for (i = 0; i < TEST_ENTRIES; i++) fprintf(stderr, "%ld%s", res[j][i], i < TEST_ENTRIES - 1 ? ", " : "\n"); } } if (k == 0) fprintf(stderr, "PASS\n"); free(res); return 0; } Compiled with "-O2", on "VirtualBox 4.4.0-22-generic #38-Ubuntu x86_64" got: zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 10174 gcd1: elapsed 2120 gcd2: elapsed 2902 gcd3: elapsed 2039 gcd4: elapsed 2812 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9309 gcd1: elapsed 2280 gcd2: elapsed 2822 gcd3: elapsed 2217 gcd4: elapsed 2710 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9589 gcd1: elapsed 2098 gcd2: elapsed 2815 gcd3: elapsed 2030 gcd4: elapsed 2718 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9914 gcd1: elapsed 2309 gcd2: elapsed 2779 gcd3: elapsed 2228 gcd4: elapsed 2709 PASS [akpm@linux-foundation.org: avoid #defining a CONFIG_ variable] Signed-off-by: Zhaoxiu Zeng <zhaoxiu.zeng@gmail.com> Signed-off-by: George Spelvin <linux@horizon.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 07:03:57 +07:00
select CPU_NO_EFFICIENT_FFS
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
Motorola ColdFire 5249 processor support.
m68knommu: Add support for the Coldfire 5251/5253 Basic support for the Coldfire 5251/5253. Signed-off-by: Steven king <sfking@fdwdc.com> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2012-06-05 22:23:08 +07:00
config M525x
bool "MCF525x"
depends on !MMU
select COLDFIRE_SW_A7
select HAVE_MBAR
lib/GCD.c: use binary GCD algorithm instead of Euclidean The binary GCD algorithm is based on the following facts: 1. If a and b are all evens, then gcd(a,b) = 2 * gcd(a/2, b/2) 2. If a is even and b is odd, then gcd(a,b) = gcd(a/2, b) 3. If a and b are all odds, then gcd(a,b) = gcd((a-b)/2, b) = gcd((a+b)/2, b) Even on x86 machines with reasonable division hardware, the binary algorithm runs about 25% faster (80% the execution time) than the division-based Euclidian algorithm. On platforms like Alpha and ARMv6 where division is a function call to emulation code, it's even more significant. There are two variants of the code here, depending on whether a fast __ffs (find least significant set bit) instruction is available. This allows the unpredictable branches in the bit-at-a-time shifting loop to be eliminated. If fast __ffs is not available, the "even/odd" GCD variant is used. I use the following code to benchmark: #include <stdio.h> #include <stdlib.h> #include <stdint.h> #include <string.h> #include <time.h> #include <unistd.h> #define swap(a, b) \ do { \ a ^= b; \ b ^= a; \ a ^= b; \ } while (0) unsigned long gcd0(unsigned long a, unsigned long b) { unsigned long r; if (a < b) { swap(a, b); } if (b == 0) return a; while ((r = a % b) != 0) { a = b; b = r; } return b; } unsigned long gcd1(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); for (;;) { a >>= __builtin_ctzl(a); if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd2(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; for (;;) { while (!(a & r)) a >>= 1; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } unsigned long gcd3(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); if (b == 1) return r & -r; for (;;) { a >>= __builtin_ctzl(a); if (a == 1) return r & -r; if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd4(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; if (b == r) return r; for (;;) { while (!(a & r)) a >>= 1; if (a == r) return r; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } static unsigned long (*gcd_func[])(unsigned long a, unsigned long b) = { gcd0, gcd1, gcd2, gcd3, gcd4, }; #define TEST_ENTRIES (sizeof(gcd_func) / sizeof(gcd_func[0])) #if defined(__x86_64__) #define rdtscll(val) do { \ unsigned long __a,__d; \ __asm__ __volatile__("rdtsc" : "=a" (__a), "=d" (__d)); \ (val) = ((unsigned long long)__a) | (((unsigned long long)__d)<<32); \ } while(0) static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { unsigned long long start, end; unsigned long long ret; unsigned long gcd_res; rdtscll(start); gcd_res = gcd(a, b); rdtscll(end); if (end >= start) ret = end - start; else ret = ~0ULL - start + 1 + end; *res = gcd_res; return ret; } #else static inline struct timespec read_time(void) { struct timespec time; clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &time); return time; } static inline unsigned long long diff_time(struct timespec start, struct timespec end) { struct timespec temp; if ((end.tv_nsec - start.tv_nsec) < 0) { temp.tv_sec = end.tv_sec - start.tv_sec - 1; temp.tv_nsec = 1000000000ULL + end.tv_nsec - start.tv_nsec; } else { temp.tv_sec = end.tv_sec - start.tv_sec; temp.tv_nsec = end.tv_nsec - start.tv_nsec; } return temp.tv_sec * 1000000000ULL + temp.tv_nsec; } static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { struct timespec start, end; unsigned long gcd_res; start = read_time(); gcd_res = gcd(a, b); end = read_time(); *res = gcd_res; return diff_time(start, end); } #endif static inline unsigned long get_rand() { if (sizeof(long) == 8) return (unsigned long)rand() << 32 | rand(); else return rand(); } int main(int argc, char **argv) { unsigned int seed = time(0); int loops = 100; int repeats = 1000; unsigned long (*res)[TEST_ENTRIES]; unsigned long long elapsed[TEST_ENTRIES]; int i, j, k; for (;;) { int opt = getopt(argc, argv, "n:r:s:"); /* End condition always first */ if (opt == -1) break; switch (opt) { case 'n': loops = atoi(optarg); break; case 'r': repeats = atoi(optarg); break; case 's': seed = strtoul(optarg, NULL, 10); break; default: /* You won't actually get here. */ break; } } res = malloc(sizeof(unsigned long) * TEST_ENTRIES * loops); memset(elapsed, 0, sizeof(elapsed)); srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); /* Do we have args? */ unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); unsigned long long min_elapsed[TEST_ENTRIES]; for (k = 0; k < repeats; k++) { for (i = 0; i < TEST_ENTRIES; i++) { unsigned long long tmp = benchmark_gcd_func(gcd_func[i], a, b, &res[j][i]); if (k == 0 || min_elapsed[i] > tmp) min_elapsed[i] = tmp; } } for (i = 0; i < TEST_ENTRIES; i++) elapsed[i] += min_elapsed[i]; } for (i = 0; i < TEST_ENTRIES; i++) printf("gcd%d: elapsed %llu\n", i, elapsed[i]); k = 0; srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); for (i = 1; i < TEST_ENTRIES; i++) { if (res[j][i] != res[j][0]) break; } if (i < TEST_ENTRIES) { if (k == 0) { k = 1; fprintf(stderr, "Error:\n"); } fprintf(stderr, "gcd(%lu, %lu): ", a, b); for (i = 0; i < TEST_ENTRIES; i++) fprintf(stderr, "%ld%s", res[j][i], i < TEST_ENTRIES - 1 ? ", " : "\n"); } } if (k == 0) fprintf(stderr, "PASS\n"); free(res); return 0; } Compiled with "-O2", on "VirtualBox 4.4.0-22-generic #38-Ubuntu x86_64" got: zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 10174 gcd1: elapsed 2120 gcd2: elapsed 2902 gcd3: elapsed 2039 gcd4: elapsed 2812 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9309 gcd1: elapsed 2280 gcd2: elapsed 2822 gcd3: elapsed 2217 gcd4: elapsed 2710 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9589 gcd1: elapsed 2098 gcd2: elapsed 2815 gcd3: elapsed 2030 gcd4: elapsed 2718 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9914 gcd1: elapsed 2309 gcd2: elapsed 2779 gcd3: elapsed 2228 gcd4: elapsed 2709 PASS [akpm@linux-foundation.org: avoid #defining a CONFIG_ variable] Signed-off-by: Zhaoxiu Zeng <zhaoxiu.zeng@gmail.com> Signed-off-by: George Spelvin <linux@horizon.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 07:03:57 +07:00
select CPU_NO_EFFICIENT_FFS
m68knommu: Add support for the Coldfire 5251/5253 Basic support for the Coldfire 5251/5253. Signed-off-by: Steven king <sfking@fdwdc.com> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2012-06-05 22:23:08 +07:00
help
Freescale (Motorola) Coldfire 5251/5253 processor support.
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
config M5271
bool "MCF5271"
depends on !MMU
select M527x
select HAVE_CACHE_SPLIT
select HAVE_IPSBAR
select GENERIC_CLOCKEVENTS
help
Freescale (Motorola) ColdFire 5270/5271 processor support.
config M5272
bool "MCF5272"
depends on !MMU
select COLDFIRE_SW_A7
select HAVE_MBAR
lib/GCD.c: use binary GCD algorithm instead of Euclidean The binary GCD algorithm is based on the following facts: 1. If a and b are all evens, then gcd(a,b) = 2 * gcd(a/2, b/2) 2. If a is even and b is odd, then gcd(a,b) = gcd(a/2, b) 3. If a and b are all odds, then gcd(a,b) = gcd((a-b)/2, b) = gcd((a+b)/2, b) Even on x86 machines with reasonable division hardware, the binary algorithm runs about 25% faster (80% the execution time) than the division-based Euclidian algorithm. On platforms like Alpha and ARMv6 where division is a function call to emulation code, it's even more significant. There are two variants of the code here, depending on whether a fast __ffs (find least significant set bit) instruction is available. This allows the unpredictable branches in the bit-at-a-time shifting loop to be eliminated. If fast __ffs is not available, the "even/odd" GCD variant is used. I use the following code to benchmark: #include <stdio.h> #include <stdlib.h> #include <stdint.h> #include <string.h> #include <time.h> #include <unistd.h> #define swap(a, b) \ do { \ a ^= b; \ b ^= a; \ a ^= b; \ } while (0) unsigned long gcd0(unsigned long a, unsigned long b) { unsigned long r; if (a < b) { swap(a, b); } if (b == 0) return a; while ((r = a % b) != 0) { a = b; b = r; } return b; } unsigned long gcd1(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); for (;;) { a >>= __builtin_ctzl(a); if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd2(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; for (;;) { while (!(a & r)) a >>= 1; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } unsigned long gcd3(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); if (b == 1) return r & -r; for (;;) { a >>= __builtin_ctzl(a); if (a == 1) return r & -r; if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd4(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; if (b == r) return r; for (;;) { while (!(a & r)) a >>= 1; if (a == r) return r; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } static unsigned long (*gcd_func[])(unsigned long a, unsigned long b) = { gcd0, gcd1, gcd2, gcd3, gcd4, }; #define TEST_ENTRIES (sizeof(gcd_func) / sizeof(gcd_func[0])) #if defined(__x86_64__) #define rdtscll(val) do { \ unsigned long __a,__d; \ __asm__ __volatile__("rdtsc" : "=a" (__a), "=d" (__d)); \ (val) = ((unsigned long long)__a) | (((unsigned long long)__d)<<32); \ } while(0) static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { unsigned long long start, end; unsigned long long ret; unsigned long gcd_res; rdtscll(start); gcd_res = gcd(a, b); rdtscll(end); if (end >= start) ret = end - start; else ret = ~0ULL - start + 1 + end; *res = gcd_res; return ret; } #else static inline struct timespec read_time(void) { struct timespec time; clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &time); return time; } static inline unsigned long long diff_time(struct timespec start, struct timespec end) { struct timespec temp; if ((end.tv_nsec - start.tv_nsec) < 0) { temp.tv_sec = end.tv_sec - start.tv_sec - 1; temp.tv_nsec = 1000000000ULL + end.tv_nsec - start.tv_nsec; } else { temp.tv_sec = end.tv_sec - start.tv_sec; temp.tv_nsec = end.tv_nsec - start.tv_nsec; } return temp.tv_sec * 1000000000ULL + temp.tv_nsec; } static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { struct timespec start, end; unsigned long gcd_res; start = read_time(); gcd_res = gcd(a, b); end = read_time(); *res = gcd_res; return diff_time(start, end); } #endif static inline unsigned long get_rand() { if (sizeof(long) == 8) return (unsigned long)rand() << 32 | rand(); else return rand(); } int main(int argc, char **argv) { unsigned int seed = time(0); int loops = 100; int repeats = 1000; unsigned long (*res)[TEST_ENTRIES]; unsigned long long elapsed[TEST_ENTRIES]; int i, j, k; for (;;) { int opt = getopt(argc, argv, "n:r:s:"); /* End condition always first */ if (opt == -1) break; switch (opt) { case 'n': loops = atoi(optarg); break; case 'r': repeats = atoi(optarg); break; case 's': seed = strtoul(optarg, NULL, 10); break; default: /* You won't actually get here. */ break; } } res = malloc(sizeof(unsigned long) * TEST_ENTRIES * loops); memset(elapsed, 0, sizeof(elapsed)); srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); /* Do we have args? */ unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); unsigned long long min_elapsed[TEST_ENTRIES]; for (k = 0; k < repeats; k++) { for (i = 0; i < TEST_ENTRIES; i++) { unsigned long long tmp = benchmark_gcd_func(gcd_func[i], a, b, &res[j][i]); if (k == 0 || min_elapsed[i] > tmp) min_elapsed[i] = tmp; } } for (i = 0; i < TEST_ENTRIES; i++) elapsed[i] += min_elapsed[i]; } for (i = 0; i < TEST_ENTRIES; i++) printf("gcd%d: elapsed %llu\n", i, elapsed[i]); k = 0; srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); for (i = 1; i < TEST_ENTRIES; i++) { if (res[j][i] != res[j][0]) break; } if (i < TEST_ENTRIES) { if (k == 0) { k = 1; fprintf(stderr, "Error:\n"); } fprintf(stderr, "gcd(%lu, %lu): ", a, b); for (i = 0; i < TEST_ENTRIES; i++) fprintf(stderr, "%ld%s", res[j][i], i < TEST_ENTRIES - 1 ? ", " : "\n"); } } if (k == 0) fprintf(stderr, "PASS\n"); free(res); return 0; } Compiled with "-O2", on "VirtualBox 4.4.0-22-generic #38-Ubuntu x86_64" got: zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 10174 gcd1: elapsed 2120 gcd2: elapsed 2902 gcd3: elapsed 2039 gcd4: elapsed 2812 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9309 gcd1: elapsed 2280 gcd2: elapsed 2822 gcd3: elapsed 2217 gcd4: elapsed 2710 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9589 gcd1: elapsed 2098 gcd2: elapsed 2815 gcd3: elapsed 2030 gcd4: elapsed 2718 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9914 gcd1: elapsed 2309 gcd2: elapsed 2779 gcd3: elapsed 2228 gcd4: elapsed 2709 PASS [akpm@linux-foundation.org: avoid #defining a CONFIG_ variable] Signed-off-by: Zhaoxiu Zeng <zhaoxiu.zeng@gmail.com> Signed-off-by: George Spelvin <linux@horizon.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 07:03:57 +07:00
select CPU_NO_EFFICIENT_FFS
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
Motorola ColdFire 5272 processor support.
config M5275
bool "MCF5275"
depends on !MMU
select M527x
select HAVE_CACHE_SPLIT
select HAVE_IPSBAR
select GENERIC_CLOCKEVENTS
help
Freescale (Motorola) ColdFire 5274/5275 processor support.
config M528x
bool "MCF528x"
depends on !MMU
select GENERIC_CLOCKEVENTS
select HAVE_CACHE_SPLIT
select HAVE_IPSBAR
help
Motorola ColdFire 5280/5282 processor support.
config M5307
bool "MCF5307"
depends on !MMU
select COLDFIRE_SW_A7
select HAVE_CACHE_CB
select HAVE_MBAR
lib/GCD.c: use binary GCD algorithm instead of Euclidean The binary GCD algorithm is based on the following facts: 1. If a and b are all evens, then gcd(a,b) = 2 * gcd(a/2, b/2) 2. If a is even and b is odd, then gcd(a,b) = gcd(a/2, b) 3. If a and b are all odds, then gcd(a,b) = gcd((a-b)/2, b) = gcd((a+b)/2, b) Even on x86 machines with reasonable division hardware, the binary algorithm runs about 25% faster (80% the execution time) than the division-based Euclidian algorithm. On platforms like Alpha and ARMv6 where division is a function call to emulation code, it's even more significant. There are two variants of the code here, depending on whether a fast __ffs (find least significant set bit) instruction is available. This allows the unpredictable branches in the bit-at-a-time shifting loop to be eliminated. If fast __ffs is not available, the "even/odd" GCD variant is used. I use the following code to benchmark: #include <stdio.h> #include <stdlib.h> #include <stdint.h> #include <string.h> #include <time.h> #include <unistd.h> #define swap(a, b) \ do { \ a ^= b; \ b ^= a; \ a ^= b; \ } while (0) unsigned long gcd0(unsigned long a, unsigned long b) { unsigned long r; if (a < b) { swap(a, b); } if (b == 0) return a; while ((r = a % b) != 0) { a = b; b = r; } return b; } unsigned long gcd1(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); for (;;) { a >>= __builtin_ctzl(a); if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd2(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; for (;;) { while (!(a & r)) a >>= 1; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } unsigned long gcd3(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); if (b == 1) return r & -r; for (;;) { a >>= __builtin_ctzl(a); if (a == 1) return r & -r; if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd4(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; if (b == r) return r; for (;;) { while (!(a & r)) a >>= 1; if (a == r) return r; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } static unsigned long (*gcd_func[])(unsigned long a, unsigned long b) = { gcd0, gcd1, gcd2, gcd3, gcd4, }; #define TEST_ENTRIES (sizeof(gcd_func) / sizeof(gcd_func[0])) #if defined(__x86_64__) #define rdtscll(val) do { \ unsigned long __a,__d; \ __asm__ __volatile__("rdtsc" : "=a" (__a), "=d" (__d)); \ (val) = ((unsigned long long)__a) | (((unsigned long long)__d)<<32); \ } while(0) static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { unsigned long long start, end; unsigned long long ret; unsigned long gcd_res; rdtscll(start); gcd_res = gcd(a, b); rdtscll(end); if (end >= start) ret = end - start; else ret = ~0ULL - start + 1 + end; *res = gcd_res; return ret; } #else static inline struct timespec read_time(void) { struct timespec time; clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &time); return time; } static inline unsigned long long diff_time(struct timespec start, struct timespec end) { struct timespec temp; if ((end.tv_nsec - start.tv_nsec) < 0) { temp.tv_sec = end.tv_sec - start.tv_sec - 1; temp.tv_nsec = 1000000000ULL + end.tv_nsec - start.tv_nsec; } else { temp.tv_sec = end.tv_sec - start.tv_sec; temp.tv_nsec = end.tv_nsec - start.tv_nsec; } return temp.tv_sec * 1000000000ULL + temp.tv_nsec; } static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { struct timespec start, end; unsigned long gcd_res; start = read_time(); gcd_res = gcd(a, b); end = read_time(); *res = gcd_res; return diff_time(start, end); } #endif static inline unsigned long get_rand() { if (sizeof(long) == 8) return (unsigned long)rand() << 32 | rand(); else return rand(); } int main(int argc, char **argv) { unsigned int seed = time(0); int loops = 100; int repeats = 1000; unsigned long (*res)[TEST_ENTRIES]; unsigned long long elapsed[TEST_ENTRIES]; int i, j, k; for (;;) { int opt = getopt(argc, argv, "n:r:s:"); /* End condition always first */ if (opt == -1) break; switch (opt) { case 'n': loops = atoi(optarg); break; case 'r': repeats = atoi(optarg); break; case 's': seed = strtoul(optarg, NULL, 10); break; default: /* You won't actually get here. */ break; } } res = malloc(sizeof(unsigned long) * TEST_ENTRIES * loops); memset(elapsed, 0, sizeof(elapsed)); srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); /* Do we have args? */ unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); unsigned long long min_elapsed[TEST_ENTRIES]; for (k = 0; k < repeats; k++) { for (i = 0; i < TEST_ENTRIES; i++) { unsigned long long tmp = benchmark_gcd_func(gcd_func[i], a, b, &res[j][i]); if (k == 0 || min_elapsed[i] > tmp) min_elapsed[i] = tmp; } } for (i = 0; i < TEST_ENTRIES; i++) elapsed[i] += min_elapsed[i]; } for (i = 0; i < TEST_ENTRIES; i++) printf("gcd%d: elapsed %llu\n", i, elapsed[i]); k = 0; srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); for (i = 1; i < TEST_ENTRIES; i++) { if (res[j][i] != res[j][0]) break; } if (i < TEST_ENTRIES) { if (k == 0) { k = 1; fprintf(stderr, "Error:\n"); } fprintf(stderr, "gcd(%lu, %lu): ", a, b); for (i = 0; i < TEST_ENTRIES; i++) fprintf(stderr, "%ld%s", res[j][i], i < TEST_ENTRIES - 1 ? ", " : "\n"); } } if (k == 0) fprintf(stderr, "PASS\n"); free(res); return 0; } Compiled with "-O2", on "VirtualBox 4.4.0-22-generic #38-Ubuntu x86_64" got: zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 10174 gcd1: elapsed 2120 gcd2: elapsed 2902 gcd3: elapsed 2039 gcd4: elapsed 2812 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9309 gcd1: elapsed 2280 gcd2: elapsed 2822 gcd3: elapsed 2217 gcd4: elapsed 2710 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9589 gcd1: elapsed 2098 gcd2: elapsed 2815 gcd3: elapsed 2030 gcd4: elapsed 2718 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9914 gcd1: elapsed 2309 gcd2: elapsed 2779 gcd3: elapsed 2228 gcd4: elapsed 2709 PASS [akpm@linux-foundation.org: avoid #defining a CONFIG_ variable] Signed-off-by: Zhaoxiu Zeng <zhaoxiu.zeng@gmail.com> Signed-off-by: George Spelvin <linux@horizon.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 07:03:57 +07:00
select CPU_NO_EFFICIENT_FFS
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
Motorola ColdFire 5307 processor support.
config M532x
bool "MCF532x"
depends on !MMU
m68knommu: create and use a common M53xx ColdFire class of CPUs The current CONFIG_M532x support definitions are actually common to a larger set of version 3 ColdFire CPU types. In the future we want to add support for the 537x family. It is very similar to the 532x internally, and will be able to use most of the same definitions. Create a CONFIG_M53xx option that is enabled to support any of the common 532x and 537x CPU types. Convert the current users of CONFIG_M532x to use CONFIG_M53xx instead. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2012-11-05 09:01:38 +07:00
select M53xx
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
select HAVE_CACHE_CB
help
Freescale (Motorola) ColdFire 532x processor support.
m68knommu: add support for the ColdFire 537x family of CPUs The ColdFire 537x family is very similar internally to the ColdFire 532x family. So with just a little extra configuration we can configure and target them as well. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2012-11-05 12:32:15 +07:00
config M537x
bool "MCF537x"
depends on !MMU
select M53xx
select HAVE_CACHE_CB
help
Freescale ColdFire 537x processor support.
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
config M5407
bool "MCF5407"
depends on !MMU
select COLDFIRE_SW_A7
select HAVE_CACHE_CB
select HAVE_MBAR
lib/GCD.c: use binary GCD algorithm instead of Euclidean The binary GCD algorithm is based on the following facts: 1. If a and b are all evens, then gcd(a,b) = 2 * gcd(a/2, b/2) 2. If a is even and b is odd, then gcd(a,b) = gcd(a/2, b) 3. If a and b are all odds, then gcd(a,b) = gcd((a-b)/2, b) = gcd((a+b)/2, b) Even on x86 machines with reasonable division hardware, the binary algorithm runs about 25% faster (80% the execution time) than the division-based Euclidian algorithm. On platforms like Alpha and ARMv6 where division is a function call to emulation code, it's even more significant. There are two variants of the code here, depending on whether a fast __ffs (find least significant set bit) instruction is available. This allows the unpredictable branches in the bit-at-a-time shifting loop to be eliminated. If fast __ffs is not available, the "even/odd" GCD variant is used. I use the following code to benchmark: #include <stdio.h> #include <stdlib.h> #include <stdint.h> #include <string.h> #include <time.h> #include <unistd.h> #define swap(a, b) \ do { \ a ^= b; \ b ^= a; \ a ^= b; \ } while (0) unsigned long gcd0(unsigned long a, unsigned long b) { unsigned long r; if (a < b) { swap(a, b); } if (b == 0) return a; while ((r = a % b) != 0) { a = b; b = r; } return b; } unsigned long gcd1(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); for (;;) { a >>= __builtin_ctzl(a); if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd2(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; for (;;) { while (!(a & r)) a >>= 1; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } unsigned long gcd3(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); if (b == 1) return r & -r; for (;;) { a >>= __builtin_ctzl(a); if (a == 1) return r & -r; if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd4(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; if (b == r) return r; for (;;) { while (!(a & r)) a >>= 1; if (a == r) return r; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } static unsigned long (*gcd_func[])(unsigned long a, unsigned long b) = { gcd0, gcd1, gcd2, gcd3, gcd4, }; #define TEST_ENTRIES (sizeof(gcd_func) / sizeof(gcd_func[0])) #if defined(__x86_64__) #define rdtscll(val) do { \ unsigned long __a,__d; \ __asm__ __volatile__("rdtsc" : "=a" (__a), "=d" (__d)); \ (val) = ((unsigned long long)__a) | (((unsigned long long)__d)<<32); \ } while(0) static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { unsigned long long start, end; unsigned long long ret; unsigned long gcd_res; rdtscll(start); gcd_res = gcd(a, b); rdtscll(end); if (end >= start) ret = end - start; else ret = ~0ULL - start + 1 + end; *res = gcd_res; return ret; } #else static inline struct timespec read_time(void) { struct timespec time; clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &time); return time; } static inline unsigned long long diff_time(struct timespec start, struct timespec end) { struct timespec temp; if ((end.tv_nsec - start.tv_nsec) < 0) { temp.tv_sec = end.tv_sec - start.tv_sec - 1; temp.tv_nsec = 1000000000ULL + end.tv_nsec - start.tv_nsec; } else { temp.tv_sec = end.tv_sec - start.tv_sec; temp.tv_nsec = end.tv_nsec - start.tv_nsec; } return temp.tv_sec * 1000000000ULL + temp.tv_nsec; } static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { struct timespec start, end; unsigned long gcd_res; start = read_time(); gcd_res = gcd(a, b); end = read_time(); *res = gcd_res; return diff_time(start, end); } #endif static inline unsigned long get_rand() { if (sizeof(long) == 8) return (unsigned long)rand() << 32 | rand(); else return rand(); } int main(int argc, char **argv) { unsigned int seed = time(0); int loops = 100; int repeats = 1000; unsigned long (*res)[TEST_ENTRIES]; unsigned long long elapsed[TEST_ENTRIES]; int i, j, k; for (;;) { int opt = getopt(argc, argv, "n:r:s:"); /* End condition always first */ if (opt == -1) break; switch (opt) { case 'n': loops = atoi(optarg); break; case 'r': repeats = atoi(optarg); break; case 's': seed = strtoul(optarg, NULL, 10); break; default: /* You won't actually get here. */ break; } } res = malloc(sizeof(unsigned long) * TEST_ENTRIES * loops); memset(elapsed, 0, sizeof(elapsed)); srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); /* Do we have args? */ unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); unsigned long long min_elapsed[TEST_ENTRIES]; for (k = 0; k < repeats; k++) { for (i = 0; i < TEST_ENTRIES; i++) { unsigned long long tmp = benchmark_gcd_func(gcd_func[i], a, b, &res[j][i]); if (k == 0 || min_elapsed[i] > tmp) min_elapsed[i] = tmp; } } for (i = 0; i < TEST_ENTRIES; i++) elapsed[i] += min_elapsed[i]; } for (i = 0; i < TEST_ENTRIES; i++) printf("gcd%d: elapsed %llu\n", i, elapsed[i]); k = 0; srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); for (i = 1; i < TEST_ENTRIES; i++) { if (res[j][i] != res[j][0]) break; } if (i < TEST_ENTRIES) { if (k == 0) { k = 1; fprintf(stderr, "Error:\n"); } fprintf(stderr, "gcd(%lu, %lu): ", a, b); for (i = 0; i < TEST_ENTRIES; i++) fprintf(stderr, "%ld%s", res[j][i], i < TEST_ENTRIES - 1 ? ", " : "\n"); } } if (k == 0) fprintf(stderr, "PASS\n"); free(res); return 0; } Compiled with "-O2", on "VirtualBox 4.4.0-22-generic #38-Ubuntu x86_64" got: zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 10174 gcd1: elapsed 2120 gcd2: elapsed 2902 gcd3: elapsed 2039 gcd4: elapsed 2812 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9309 gcd1: elapsed 2280 gcd2: elapsed 2822 gcd3: elapsed 2217 gcd4: elapsed 2710 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9589 gcd1: elapsed 2098 gcd2: elapsed 2815 gcd3: elapsed 2030 gcd4: elapsed 2718 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9914 gcd1: elapsed 2309 gcd2: elapsed 2779 gcd3: elapsed 2228 gcd4: elapsed 2709 PASS [akpm@linux-foundation.org: avoid #defining a CONFIG_ variable] Signed-off-by: Zhaoxiu Zeng <zhaoxiu.zeng@gmail.com> Signed-off-by: George Spelvin <linux@horizon.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 07:03:57 +07:00
select CPU_NO_EFFICIENT_FFS
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
Motorola ColdFire 5407 processor support.
config M547x
bool "MCF547x"
select M54xx
m68k: allow ColdFire 547x and 548x CPUs to be built with MMU enabled The ColdFire 547x and 548x CPUs have internal MMU hardware. All code to support this is now in, so we can build kernels with it enabled. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> Acked-by: Matt Waddel <mwaddel@yahoo.com> Acked-by: Kurt Mahan <kmahan@xmission.com>
2011-10-19 11:13:18 +07:00
select MMU_COLDFIRE if MMU
m68k: move CONFIG_FPU set to per-CPU configuration Move the selection of CONFIG_FPU to each CPU type configuration. Currently for m68k we have a global set of CONFIG_FPU based on if CONFIG_MMU is enabled or not. There is at least one CPU family we support (m5441x) that has an MMU but has no FPU hardware. So we need to be able to have CONFIG_MMU set and CONFIG_FPU not set. Whether we build for a CPU with MMU enabled or not doesn't change the fact that it has FPU hardware support. Our current non-MMU builds have never had CONIG_FPU enabled - and in fact the kernel will not compile with that set and CONFIG_MMU not set at the moment. It is easy enough to fix this - but it would involve a structure change to sigcontext.h, and that is a user space exported header (so ABI change). This change makes no configuration visible changes, and all configs end up with the same configuration settings as before. This change based on changes and discussion from Yannick Gicquel <yannick.gicquel@open.eurogiciel.org>. Signed-off-by: Greg Ungerer <gerg@linux-m68k.org>
2016-08-24 10:32:01 +07:00
select FPU if MMU
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
select HAVE_CACHE_CB
select HAVE_MBAR
lib/GCD.c: use binary GCD algorithm instead of Euclidean The binary GCD algorithm is based on the following facts: 1. If a and b are all evens, then gcd(a,b) = 2 * gcd(a/2, b/2) 2. If a is even and b is odd, then gcd(a,b) = gcd(a/2, b) 3. If a and b are all odds, then gcd(a,b) = gcd((a-b)/2, b) = gcd((a+b)/2, b) Even on x86 machines with reasonable division hardware, the binary algorithm runs about 25% faster (80% the execution time) than the division-based Euclidian algorithm. On platforms like Alpha and ARMv6 where division is a function call to emulation code, it's even more significant. There are two variants of the code here, depending on whether a fast __ffs (find least significant set bit) instruction is available. This allows the unpredictable branches in the bit-at-a-time shifting loop to be eliminated. If fast __ffs is not available, the "even/odd" GCD variant is used. I use the following code to benchmark: #include <stdio.h> #include <stdlib.h> #include <stdint.h> #include <string.h> #include <time.h> #include <unistd.h> #define swap(a, b) \ do { \ a ^= b; \ b ^= a; \ a ^= b; \ } while (0) unsigned long gcd0(unsigned long a, unsigned long b) { unsigned long r; if (a < b) { swap(a, b); } if (b == 0) return a; while ((r = a % b) != 0) { a = b; b = r; } return b; } unsigned long gcd1(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); for (;;) { a >>= __builtin_ctzl(a); if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd2(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; for (;;) { while (!(a & r)) a >>= 1; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } unsigned long gcd3(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); if (b == 1) return r & -r; for (;;) { a >>= __builtin_ctzl(a); if (a == 1) return r & -r; if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd4(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; if (b == r) return r; for (;;) { while (!(a & r)) a >>= 1; if (a == r) return r; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } static unsigned long (*gcd_func[])(unsigned long a, unsigned long b) = { gcd0, gcd1, gcd2, gcd3, gcd4, }; #define TEST_ENTRIES (sizeof(gcd_func) / sizeof(gcd_func[0])) #if defined(__x86_64__) #define rdtscll(val) do { \ unsigned long __a,__d; \ __asm__ __volatile__("rdtsc" : "=a" (__a), "=d" (__d)); \ (val) = ((unsigned long long)__a) | (((unsigned long long)__d)<<32); \ } while(0) static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { unsigned long long start, end; unsigned long long ret; unsigned long gcd_res; rdtscll(start); gcd_res = gcd(a, b); rdtscll(end); if (end >= start) ret = end - start; else ret = ~0ULL - start + 1 + end; *res = gcd_res; return ret; } #else static inline struct timespec read_time(void) { struct timespec time; clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &time); return time; } static inline unsigned long long diff_time(struct timespec start, struct timespec end) { struct timespec temp; if ((end.tv_nsec - start.tv_nsec) < 0) { temp.tv_sec = end.tv_sec - start.tv_sec - 1; temp.tv_nsec = 1000000000ULL + end.tv_nsec - start.tv_nsec; } else { temp.tv_sec = end.tv_sec - start.tv_sec; temp.tv_nsec = end.tv_nsec - start.tv_nsec; } return temp.tv_sec * 1000000000ULL + temp.tv_nsec; } static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { struct timespec start, end; unsigned long gcd_res; start = read_time(); gcd_res = gcd(a, b); end = read_time(); *res = gcd_res; return diff_time(start, end); } #endif static inline unsigned long get_rand() { if (sizeof(long) == 8) return (unsigned long)rand() << 32 | rand(); else return rand(); } int main(int argc, char **argv) { unsigned int seed = time(0); int loops = 100; int repeats = 1000; unsigned long (*res)[TEST_ENTRIES]; unsigned long long elapsed[TEST_ENTRIES]; int i, j, k; for (;;) { int opt = getopt(argc, argv, "n:r:s:"); /* End condition always first */ if (opt == -1) break; switch (opt) { case 'n': loops = atoi(optarg); break; case 'r': repeats = atoi(optarg); break; case 's': seed = strtoul(optarg, NULL, 10); break; default: /* You won't actually get here. */ break; } } res = malloc(sizeof(unsigned long) * TEST_ENTRIES * loops); memset(elapsed, 0, sizeof(elapsed)); srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); /* Do we have args? */ unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); unsigned long long min_elapsed[TEST_ENTRIES]; for (k = 0; k < repeats; k++) { for (i = 0; i < TEST_ENTRIES; i++) { unsigned long long tmp = benchmark_gcd_func(gcd_func[i], a, b, &res[j][i]); if (k == 0 || min_elapsed[i] > tmp) min_elapsed[i] = tmp; } } for (i = 0; i < TEST_ENTRIES; i++) elapsed[i] += min_elapsed[i]; } for (i = 0; i < TEST_ENTRIES; i++) printf("gcd%d: elapsed %llu\n", i, elapsed[i]); k = 0; srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); for (i = 1; i < TEST_ENTRIES; i++) { if (res[j][i] != res[j][0]) break; } if (i < TEST_ENTRIES) { if (k == 0) { k = 1; fprintf(stderr, "Error:\n"); } fprintf(stderr, "gcd(%lu, %lu): ", a, b); for (i = 0; i < TEST_ENTRIES; i++) fprintf(stderr, "%ld%s", res[j][i], i < TEST_ENTRIES - 1 ? ", " : "\n"); } } if (k == 0) fprintf(stderr, "PASS\n"); free(res); return 0; } Compiled with "-O2", on "VirtualBox 4.4.0-22-generic #38-Ubuntu x86_64" got: zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 10174 gcd1: elapsed 2120 gcd2: elapsed 2902 gcd3: elapsed 2039 gcd4: elapsed 2812 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9309 gcd1: elapsed 2280 gcd2: elapsed 2822 gcd3: elapsed 2217 gcd4: elapsed 2710 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9589 gcd1: elapsed 2098 gcd2: elapsed 2815 gcd3: elapsed 2030 gcd4: elapsed 2718 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9914 gcd1: elapsed 2309 gcd2: elapsed 2779 gcd3: elapsed 2228 gcd4: elapsed 2709 PASS [akpm@linux-foundation.org: avoid #defining a CONFIG_ variable] Signed-off-by: Zhaoxiu Zeng <zhaoxiu.zeng@gmail.com> Signed-off-by: George Spelvin <linux@horizon.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 07:03:57 +07:00
select CPU_NO_EFFICIENT_FFS
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
Freescale ColdFire 5470/5471/5472/5473/5474/5475 processor support.
config M548x
bool "MCF548x"
m68k: allow ColdFire 547x and 548x CPUs to be built with MMU enabled The ColdFire 547x and 548x CPUs have internal MMU hardware. All code to support this is now in, so we can build kernels with it enabled. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> Acked-by: Matt Waddel <mwaddel@yahoo.com> Acked-by: Kurt Mahan <kmahan@xmission.com>
2011-10-19 11:13:18 +07:00
select MMU_COLDFIRE if MMU
m68k: move CONFIG_FPU set to per-CPU configuration Move the selection of CONFIG_FPU to each CPU type configuration. Currently for m68k we have a global set of CONFIG_FPU based on if CONFIG_MMU is enabled or not. There is at least one CPU family we support (m5441x) that has an MMU but has no FPU hardware. So we need to be able to have CONFIG_MMU set and CONFIG_FPU not set. Whether we build for a CPU with MMU enabled or not doesn't change the fact that it has FPU hardware support. Our current non-MMU builds have never had CONIG_FPU enabled - and in fact the kernel will not compile with that set and CONFIG_MMU not set at the moment. It is easy enough to fix this - but it would involve a structure change to sigcontext.h, and that is a user space exported header (so ABI change). This change makes no configuration visible changes, and all configs end up with the same configuration settings as before. This change based on changes and discussion from Yannick Gicquel <yannick.gicquel@open.eurogiciel.org>. Signed-off-by: Greg Ungerer <gerg@linux-m68k.org>
2016-08-24 10:32:01 +07:00
select FPU if MMU
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
select M54xx
select HAVE_CACHE_CB
select HAVE_MBAR
lib/GCD.c: use binary GCD algorithm instead of Euclidean The binary GCD algorithm is based on the following facts: 1. If a and b are all evens, then gcd(a,b) = 2 * gcd(a/2, b/2) 2. If a is even and b is odd, then gcd(a,b) = gcd(a/2, b) 3. If a and b are all odds, then gcd(a,b) = gcd((a-b)/2, b) = gcd((a+b)/2, b) Even on x86 machines with reasonable division hardware, the binary algorithm runs about 25% faster (80% the execution time) than the division-based Euclidian algorithm. On platforms like Alpha and ARMv6 where division is a function call to emulation code, it's even more significant. There are two variants of the code here, depending on whether a fast __ffs (find least significant set bit) instruction is available. This allows the unpredictable branches in the bit-at-a-time shifting loop to be eliminated. If fast __ffs is not available, the "even/odd" GCD variant is used. I use the following code to benchmark: #include <stdio.h> #include <stdlib.h> #include <stdint.h> #include <string.h> #include <time.h> #include <unistd.h> #define swap(a, b) \ do { \ a ^= b; \ b ^= a; \ a ^= b; \ } while (0) unsigned long gcd0(unsigned long a, unsigned long b) { unsigned long r; if (a < b) { swap(a, b); } if (b == 0) return a; while ((r = a % b) != 0) { a = b; b = r; } return b; } unsigned long gcd1(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); for (;;) { a >>= __builtin_ctzl(a); if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd2(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; for (;;) { while (!(a & r)) a >>= 1; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } unsigned long gcd3(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; b >>= __builtin_ctzl(b); if (b == 1) return r & -r; for (;;) { a >>= __builtin_ctzl(a); if (a == 1) return r & -r; if (a == b) return a << __builtin_ctzl(r); if (a < b) swap(a, b); a -= b; } } unsigned long gcd4(unsigned long a, unsigned long b) { unsigned long r = a | b; if (!a || !b) return r; r &= -r; while (!(b & r)) b >>= 1; if (b == r) return r; for (;;) { while (!(a & r)) a >>= 1; if (a == r) return r; if (a == b) return a; if (a < b) swap(a, b); a -= b; a >>= 1; if (a & r) a += b; a >>= 1; } } static unsigned long (*gcd_func[])(unsigned long a, unsigned long b) = { gcd0, gcd1, gcd2, gcd3, gcd4, }; #define TEST_ENTRIES (sizeof(gcd_func) / sizeof(gcd_func[0])) #if defined(__x86_64__) #define rdtscll(val) do { \ unsigned long __a,__d; \ __asm__ __volatile__("rdtsc" : "=a" (__a), "=d" (__d)); \ (val) = ((unsigned long long)__a) | (((unsigned long long)__d)<<32); \ } while(0) static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { unsigned long long start, end; unsigned long long ret; unsigned long gcd_res; rdtscll(start); gcd_res = gcd(a, b); rdtscll(end); if (end >= start) ret = end - start; else ret = ~0ULL - start + 1 + end; *res = gcd_res; return ret; } #else static inline struct timespec read_time(void) { struct timespec time; clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &time); return time; } static inline unsigned long long diff_time(struct timespec start, struct timespec end) { struct timespec temp; if ((end.tv_nsec - start.tv_nsec) < 0) { temp.tv_sec = end.tv_sec - start.tv_sec - 1; temp.tv_nsec = 1000000000ULL + end.tv_nsec - start.tv_nsec; } else { temp.tv_sec = end.tv_sec - start.tv_sec; temp.tv_nsec = end.tv_nsec - start.tv_nsec; } return temp.tv_sec * 1000000000ULL + temp.tv_nsec; } static unsigned long long benchmark_gcd_func(unsigned long (*gcd)(unsigned long, unsigned long), unsigned long a, unsigned long b, unsigned long *res) { struct timespec start, end; unsigned long gcd_res; start = read_time(); gcd_res = gcd(a, b); end = read_time(); *res = gcd_res; return diff_time(start, end); } #endif static inline unsigned long get_rand() { if (sizeof(long) == 8) return (unsigned long)rand() << 32 | rand(); else return rand(); } int main(int argc, char **argv) { unsigned int seed = time(0); int loops = 100; int repeats = 1000; unsigned long (*res)[TEST_ENTRIES]; unsigned long long elapsed[TEST_ENTRIES]; int i, j, k; for (;;) { int opt = getopt(argc, argv, "n:r:s:"); /* End condition always first */ if (opt == -1) break; switch (opt) { case 'n': loops = atoi(optarg); break; case 'r': repeats = atoi(optarg); break; case 's': seed = strtoul(optarg, NULL, 10); break; default: /* You won't actually get here. */ break; } } res = malloc(sizeof(unsigned long) * TEST_ENTRIES * loops); memset(elapsed, 0, sizeof(elapsed)); srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); /* Do we have args? */ unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); unsigned long long min_elapsed[TEST_ENTRIES]; for (k = 0; k < repeats; k++) { for (i = 0; i < TEST_ENTRIES; i++) { unsigned long long tmp = benchmark_gcd_func(gcd_func[i], a, b, &res[j][i]); if (k == 0 || min_elapsed[i] > tmp) min_elapsed[i] = tmp; } } for (i = 0; i < TEST_ENTRIES; i++) elapsed[i] += min_elapsed[i]; } for (i = 0; i < TEST_ENTRIES; i++) printf("gcd%d: elapsed %llu\n", i, elapsed[i]); k = 0; srand(seed); for (j = 0; j < loops; j++) { unsigned long a = get_rand(); unsigned long b = argc > optind ? strtoul(argv[optind], NULL, 10) : get_rand(); for (i = 1; i < TEST_ENTRIES; i++) { if (res[j][i] != res[j][0]) break; } if (i < TEST_ENTRIES) { if (k == 0) { k = 1; fprintf(stderr, "Error:\n"); } fprintf(stderr, "gcd(%lu, %lu): ", a, b); for (i = 0; i < TEST_ENTRIES; i++) fprintf(stderr, "%ld%s", res[j][i], i < TEST_ENTRIES - 1 ? ", " : "\n"); } } if (k == 0) fprintf(stderr, "PASS\n"); free(res); return 0; } Compiled with "-O2", on "VirtualBox 4.4.0-22-generic #38-Ubuntu x86_64" got: zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 10174 gcd1: elapsed 2120 gcd2: elapsed 2902 gcd3: elapsed 2039 gcd4: elapsed 2812 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9309 gcd1: elapsed 2280 gcd2: elapsed 2822 gcd3: elapsed 2217 gcd4: elapsed 2710 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9589 gcd1: elapsed 2098 gcd2: elapsed 2815 gcd3: elapsed 2030 gcd4: elapsed 2718 PASS zhaoxiuzeng@zhaoxiuzeng-VirtualBox:~/develop$ ./gcd -r 500000 -n 10 gcd0: elapsed 9914 gcd1: elapsed 2309 gcd2: elapsed 2779 gcd3: elapsed 2228 gcd4: elapsed 2709 PASS [akpm@linux-foundation.org: avoid #defining a CONFIG_ variable] Signed-off-by: Zhaoxiu Zeng <zhaoxiu.zeng@gmail.com> Signed-off-by: George Spelvin <linux@horizon.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-21 07:03:57 +07:00
select CPU_NO_EFFICIENT_FFS
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
Freescale ColdFire 5480/5481/5482/5483/5484/5485 processor support.
m68knommu: Add support for the Coldfire m5441x. Add support for the Coldfire 5441x (54410/54415/54416/54417/54418). Currently we only support noMMU mode. It requires the PIT patch posted previously as it uses the PIT instead of the dma timer as a clock source so we can get all that GENERIC_CLOCKEVENTS goodness. It also adds some simple clk definitions and very simple minded power management. The gpio code is tweeked and some additional devices are added to devices.c. The Makefile uses -mv4e as apparently, the only difference a v4m (m5441x) and a v4e is the later has a FPU, which I don't think should matter to us in the kernel. Signed-off-by: Steven King <sfking@fdwdc.com> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2012-06-07 04:28:31 +07:00
config M5441x
bool "MCF5441x"
depends on !MMU
select GENERIC_CLOCKEVENTS
select HAVE_CACHE_CB
help
Freescale Coldfire 54410/54415/54416/54417/54418 processor support.
m68knommu: make ColdFire SoC selection a choice It would be nice if we could support multiple ColdFire SoC types in a single binary - but currently the code simply does not support it. Change the SoC selection config options to be a choice instead of individual selectable entries. This fixes problems with building allnoconfig, and means that a sane linux kernel is generated for a single ColdFire SoC type. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
2015-07-07 12:44:02 +07:00
endchoice
config M527x
bool
config M53xx
bool
config M54xx
bool
m68k/Kconfig: Separate classic m68k and coldfire early While you can build multiplatform kernels for machines with classic m68k processors, you cannot mix support for classic m68k and coldfire processors. To avoid such hybrid kernels, introduce CONFIG_M68KCLASSIC as an antipole for CONFIG_COLDFIRE, and make all specific processor support depend on one of them. All classic m68k machine support also needs to depend on this. The defaults (CONFIG_M68KCLASSIC if MMU, CONFIG_COLDFIRE if !MMU) are chosen such to make most of the existing configs build and work. Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-12-27 02:32:02 +07:00
endif # COLDFIRE
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
comment "Processor Specific Options"
config M68KFPU_EMU
arch/m68k: remove CONFIG_EXPERIMENTAL This config item has not carried much meaning for a while now and is almost always enabled by default. As agreed during the Linux kernel summit, remove it. CC: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
2012-10-24 03:01:27 +07:00
bool "Math emulation support"
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
depends on MMU
help
At some point in the future, this will cause floating-point math
instructions to be emulated by the kernel on machines that lack a
floating-point math coprocessor. Thrill-seekers and chronically
sleep-deprived psychotic hacker types can say Y now, everyone else
should probably wait a while.
config M68KFPU_EMU_EXTRAPREC
bool "Math emulation extra precision"
depends on M68KFPU_EMU
help
The fpu uses normally a few bit more during calculations for
correct rounding, the emulator can (often) do the same but this
extra calculation can cost quite some time, so you can disable
it here. The emulator will then "only" calculate with a 64 bit
mantissa and round slightly incorrect, what is more than enough
for normal usage.
config M68KFPU_EMU_ONLY
bool "Math emulation only kernel"
depends on M68KFPU_EMU
help
This option prevents any floating-point instructions from being
compiled into the kernel, thereby the kernel doesn't save any
floating point context anymore during task switches, so this
kernel will only be usable on machines without a floating-point
math coprocessor. This makes the kernel a bit faster as no tests
needs to be executed whether a floating-point instruction in the
kernel should be executed or not.
config ADVANCED
bool "Advanced configuration options"
depends on MMU
---help---
This gives you access to some advanced options for the CPU. The
defaults should be fine for most users, but these options may make
it possible for you to improve performance somewhat if you know what
you are doing.
Note that the answer to this question won't directly affect the
kernel: saying N will just cause the configurator to skip all
the questions about these options.
Most users should say N to this question.
config RMW_INSNS
bool "Use read-modify-write instructions"
depends on ADVANCED
---help---
This allows to use certain instructions that work with indivisible
read-modify-write bus cycles. While this is faster than the
workaround of disabling interrupts, it can conflict with DMA
( = direct memory access) on many Amiga systems, and it is also said
to destabilize other machines. It is very likely that this will
cause serious problems on any Amiga or Atari Medusa if set. The only
configuration where it should work are 68030-based Ataris, where it
apparently improves performance. But you've been warned! Unless you
really know what you are doing, say N. Try Y only if you're quite
adventurous.
config SINGLE_MEMORY_CHUNK
bool "Use one physical chunk of memory only" if ADVANCED && !SUN3
depends on MMU
default y if SUN3
select NEED_MULTIPLE_NODES
help
Ignore all but the first contiguous chunk of physical memory for VM
purposes. This will save a few bytes kernel size and may speed up
some operations. Say N if not sure.
config ARCH_DISCONTIGMEM_ENABLE
def_bool MMU && !SINGLE_MEMORY_CHUNK
config 060_WRITETHROUGH
bool "Use write-through caching for 68060 supervisor accesses"
depends on ADVANCED && M68060
---help---
The 68060 generally uses copyback caching of recently accessed data.
Copyback caching means that memory writes will be held in an on-chip
cache and only written back to memory some time later. Saying Y
here will force supervisor (kernel) accesses to use writethrough
caching. Writethrough caching means that data is written to memory
straight away, so that cache and memory data always agree.
Writethrough caching is less efficient, but is needed for some
drivers on 68060 based systems where the 68060 bus snooping signal
is hardwired on. The 53c710 SCSI driver is known to suffer from
this problem.
config M68K_L2_CACHE
bool
depends on MAC
default y
config NODES_SHIFT
int
default "3"
depends on !SINGLE_MEMORY_CHUNK
m68k: Move CPU_HAS_* config options They belong together with the CPU selection Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Acked-by: Greg Ungerer<gerg@uclinux.org>
2012-06-06 22:26:35 +07:00
config CPU_HAS_NO_BITFIELDS
bool
config CPU_HAS_NO_MULDIV64
bool
m68k: Introduce config option CPU_HAS_NO_UNALIGNED Use CONFIG_CPU_HAS_NO_UNALIGNED instead of open coding CONFIG_M68000 || CONFIG_COLDFIRE Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Acked-by: Greg Ungerer<gerg@uclinux.org>
2012-06-07 00:37:52 +07:00
config CPU_HAS_NO_UNALIGNED
bool
m68k: Move CPU_HAS_* config options They belong together with the CPU selection Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Acked-by: Greg Ungerer<gerg@uclinux.org>
2012-06-06 22:26:35 +07:00
config CPU_HAS_ADDRESS_SPACES
bool
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
config FPU
bool
config COLDFIRE_SW_A7
bool
config HAVE_CACHE_SPLIT
bool
config HAVE_CACHE_CB
bool
config HAVE_MBAR
bool
config HAVE_IPSBAR
bool
config CLOCK_FREQ
int "Set the core clock frequency"
m68knommu: improve the clock configuration defaults Create some intelligent default settings for each ColdFire SoC type in the configuration entry for CONFIG_CLOCK_FREQ. The ColdFire clock frequency is configurable at build time. There is a lot of variation in the frequency of operation on specific ColdFire based boards. But we can choose a default that matches the maximum frequency of clock operation for a particular ColdFire part. That is typically the most common clock setting. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
2015-07-07 12:01:53 +07:00
default "25000000" if M5206
default "54000000" if M5206e
default "166666666" if M520x
default "140000000" if M5249
default "150000000" if M527x || M523x
default "90000000" if M5307
default "50000000" if M5407
default "266000000" if M54xx
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
default "66666666"
m68knommu: force setting of CONFIG_CLOCK_FREQ for ColdFire It is possible to disable the clock selection at configuration time, but for ColdFire targets we always expect a clock frequency to be selected. This results in the following compile time error: CC arch/m68k/kernel/asm-offsets.s In file included from ./arch/m68k/include/asm/timex.h:14:0, from include/linux/timex.h:65, from include/linux/sched.h:19, from arch/m68k/kernel/asm-offsets.c:14: ./arch/m68k/include/asm/coldfire.h:25:2: error: #error "Don't know what your ColdFire CPU clock frequency is??" Remove CONFIG_CLOCK_SELECT completely and always enable CONFIG_CLOCK_FREQ for ColdFire. Signed-off-by: Greg Ungerer <gerg@uclinux.org> Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
2015-07-07 11:21:21 +07:00
depends on COLDFIRE
m68k: reorganize Kconfig options to improve mmu/non-mmu selections The current mmu and non-mmu Kconfig files can be merged to form a more general selection of options. The current break up of options is due to the simple brute force merge from the m68k and m68knommu arch directories. Many of the options are not at all specific to having the MMU enabled or not. They are actually associated with a particular CPU type or platform type. Ultimately as we support all processors with the MMU disabled we need many of these options to be selectable without the MMU option enabled. And likewise some of the ColdFire processors, which currently are only supported with the MMU disabled, do have MMU hardware, and will need to have options selected on CPU type, not MMU disabled. This patch removes the old mmu and non-mmu Kconfigs and instead breaks up the configuration into four areas: cpu, machine, bus, devices. The Kconfig.cpu lists all the options associated with selecting a CPU, and includes options specific to each CPU type as well. Kconfig.machine lists all options associated with selecting a machine type. Almost always the machines selectable is restricted by the chosen CPU. Kconfig.bus contains options associated with selecting bus types on the various machine types. That includes PCI bus, PCMCIA bus, etc. Kconfig.devices contains options for drivers and driver associated options. Signed-off-by: Greg Ungerer <gerg@uclinux.org>
2011-06-20 12:49:09 +07:00
help
Define the CPU clock frequency in use. This is the core clock
frequency, it may or may not be the same as the external clock
crystal fitted to your board. Some processors have an internal
PLL and can have their frequency programmed at run time, others
use internal dividers. In general the kernel won't setup a PLL
if it is fitted (there are some exceptions). This value will be
specific to the exact CPU that you are using.
config OLDMASK
bool "Old mask 5307 (1H55J) silicon"
depends on M5307
help
Build support for the older revision ColdFire 5307 silicon.
Specifically this is the 1H55J mask revision.
if HAVE_CACHE_SPLIT
choice
prompt "Split Cache Configuration"
default CACHE_I
config CACHE_I
bool "Instruction"
help
Use all of the ColdFire CPU cache memory as an instruction cache.
config CACHE_D
bool "Data"
help
Use all of the ColdFire CPU cache memory as a data cache.
config CACHE_BOTH
bool "Both"
help
Split the ColdFire CPU cache, and use half as an instruction cache
and half as a data cache.
endchoice
endif
if HAVE_CACHE_CB
choice
prompt "Data cache mode"
default CACHE_WRITETHRU
config CACHE_WRITETHRU
bool "Write-through"
help
The ColdFire CPU cache is set into Write-through mode.
config CACHE_COPYBACK
bool "Copy-back"
help
The ColdFire CPU cache is set into Copy-back mode.
endchoice
endif
Reference in New Issue Copy Permalink