Current test based on the fact below in my previous mail
Current Linux page allocation fairly provides pages for every process, since swap daemon only is started when memory is low, so when it starts to scan active_list, the private pages of processes are messed up with each other, vmscan.c:shrink_list() is the only approach to attach disk swap page to page on active_list, as the result, all private pages lost their affinity on swap partition. I will give a testlater...
Three testcases are imitated here 1) matrix: Some softwares do a lots of matrix arithmetic in their PrivateVMA, in fact, the type is much better to pps than Linux. 2) c: C malloc uses an arithmetic just like slab, so when an application resume from swap partition, let's supposed it touches three variables whose sizes are different, in the result, it should touch three pages and the three pages are closed with each other but aren't continual, I also imitate a case that if the application starts full-speed running later (touch more pages). 3) destruction: Typically, if an application resumes due to user clicks the close button, it totally visits all its private data to execute object destruction. Test stepping 1) run ./entry and say y to it, maybe need root right. 2) wait a moment until echo 'primarywait'. 3) swapoff -a && swapon -a. 4) ./hog until count = 10. 5) 'cat primary entry secondary > /dev/null' 6) 'cat /proc/vmstat' several times and record 'pswpin' field when it's stable. 7) type `1', `2' or `3' to 3 testcases, answer `2' to start fullspeed testcase. 8) record new 'pswpin' field. 9) which is better? see the 'pswpin' increment. pswpin is increased in mm/page_io.c:swap_readpage. Test stepping purposes 1) Step 1, 'entry' wakes up 'primary' and 'secondary' simultaneously, every time 'primary' allocates a page, 'secondary' inserts some pages into active_list closed to it. 1) Step 3, we should re-allocate swap pages. 2) Step 4, flush 'entry primary secondary' to swap partition. 3) Step 5, make file content 'entry primary secondary' present in memory. Testcases are done in vmware virtual machine 5.5, 32M memory. If you argue my circumstance, do your testcases following the steps advised 1) Run multiple memory-consumer together, make them pause at a point. (So mess up all private pages in pg_active list). 2) Flush them to swap partition. 3) Wake up one of them, let it run full-speed for a while, record pswpin of /proc/vmstat. 4) Invalidate all readaheaded pages. 5) Wake up another, repeat the test. 6) It's also good if you can record hard LED twinking:) Maybe your test resumes all memory-consumers together, so Linux readaheads some pages close to page-fault page but are belong to other processes, I think. By the way, what's linux-mm mail, it ins't in Documentation/SubmitPatches. In fact, you will find Linux column makes hard LED twinking per 5 seconds. ----------------------------- Linux pps matrix 5241 1597 5322 1620 (81) (23) c 8028 1937 8095 1954 fullspeed 8313 1964 (67) (17) (218) (10) destruction 9461 4445 9825 4484 (364) (39) Comment secondary.c:memset clause, so 'secondary' won't interrupt page-allocation in 'primary'. ----------------------------- Linux pps matrix 207 38 256 59 (49) (21) c 1273 347 1341 383 fullspeed 1362 386 (68) (36) (21) (3) destruction 2435 1178 2513 1246 (78) (68) entry.c ----------------------------- #include <sys/wait.h> #include <sys/types.h> #include <sys/ipc.h> #include <sys/sem.h> #include <sys/stat.h> #include <unistd.h> #include <stdlib.h> #include <stdio.h> #include <string.h> pid_t pids[4]; int sem_set; siginfo_t si; int main(int argc, char **argv) { int i, data; unsigned short init_data[4] = { 1, 1, 1, 1 }; if ((sem_set = semget(123321, 4, IPC_CREAT)) == -1) goto failed; if (semctl(sem_set, 0, SETALL, &init_data) == -1) goto failed; pid_t pid = vfork(); if (pid == -1) { goto failed; } else if (pid == 0) { if (execlp("./primary", NULL) == -1) goto failed; } else { pids[0] = pid; } pid = vfork(); if (pid == -1) { goto failed; } else if (pid == 0) { if (execlp("./secondary", NULL) == -1) goto failed; } else { pids[1] = pid; } printf("entry:continue?\n"); getchar(); init_data[0] = init_data[1] = 0; if (semctl(sem_set, 0, SETALL, &init_data) == -1) goto failed; sleep(30000); exit(EXIT_SUCCESS); failed: perror("entry errno:"); exit(EXIT_FAILURE); } primary.c ----------------------------- #include <sys/mman.h> #include <signal.h> #include <sys/sem.h> #include <stdlib.h> #include <stdio.h> #include <string.h> int cont = 0; void* addr; int VMA_SIZE = 32; void init_mmap() { int i, j, a, b; addr = mmap(NULL, 4096 * VMA_SIZE, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANON, 0, 0); printf("primary:addr:%x\n", (unsigned long) addr); if (addr == MAP_FAILED) goto failed; for (i = 0; i < VMA_SIZE; i++) { memset((unsigned char*) addr + i * 4096, i, 1); sleep(2); } return; failed: perror("primary errno:"); exit(EXIT_FAILURE); } #define TOUCH_IT(value) \ if (*((unsigned char*) addr + value * 4096) != value) \ printf("BUG, memory corrupt!%d\n", value); #define WEAK_RANDOM(from, to) \ range = to - from; \ range_temp = rand() % range; \ TOUCH_IT(range_temp) #define STRONG_RANDOM \ WEAK_RANDOM(0, VMA_SIZE) void fullspeed_case() { int i, a; for (i = 0; i < VMA_SIZE / 2; i++) { a = rand() % VMA_SIZE; TOUCH_IT(i) sleep(5); } } void c_case() { int from, to, i, a, range, range_temp; for (i = 0; i < 3; i++) { from = rand() % VMA_SIZE; if (from + 8 >= VMA_SIZE) from = VMA_SIZE - 8; to = from + 8; WEAK_RANDOM(from, to) sleep(5); } STRONG_RANDOM sleep(5); STRONG_RANDOM } void destruction_case() { int i, a; for (i = 0; i < VMA_SIZE; i++) { a = rand() % VMA_SIZE; TOUCH_IT(i) sleep(5); } } void matrix_case() { int base, index, i, a, range, range_temp, temp; for (i = 0; i < 3; i++) { base = rand() % VMA_SIZE; if (base + 8 >= VMA_SIZE) base = VMA_SIZE - 8; index = rand() % 8; index = (index == 7 ? index-- : index); temp = base + index; TOUCH_IT(temp) sleep(5); temp++; TOUCH_IT(temp) sleep(5); } STRONG_RANDOM sleep(5); STRONG_RANDOM } int main(int argc, char **argv) { int a; int sem_set; struct sembuf sem_buf; if ((sem_set = semget(123321, 4, 0)) == -1) goto failed; sem_buf.sem_num = 0; sem_buf.sem_op = 0; sem_buf.sem_flg = 0; if (semop(sem_set, &sem_buf, 1) == -1) goto failed; init_mmap(); printf("primarywait\n"); scanf("%d", &a); printf("primaryresume\n"); switch (a) { case 1: printf("matrix_case\n"); matrix_case(); break; case 2: printf("c_case\n"); c_case(); break; case 3: printf("destruction_case\n"); destruction_case(); break; } if (a == 2) { printf("primaryfullspeed\n"); scanf("%d", &a); fullspeed_case(); } printf("primarydone\n"); sleep(30000); exit(EXIT_SUCCESS); failed: perror("primary errno:"); exit(EXIT_FAILURE); } secondary.c ----------------------------- #include <sys/mman.h> #include <sys/sem.h> #include <stdlib.h> #include <stdio.h> #include <string.h> int main(int argc, char **argv) { char c; int i, j, sem_set; int a, b; void* addr; struct sembuf sem_buf; if ((sem_set = semget(123321, 4, 0)) == -1) goto failed; sem_buf.sem_num = 1; sem_buf.sem_op = 0; sem_buf.sem_flg = 0; if (semop(sem_set, &sem_buf, 1) == -1) goto failed; addr = mmap(NULL, 4096 * 32 * 66, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANON, 0, 0); printf("secondaryaddr%x\n", addr); if (addr == MAP_FAILED) goto failed; for (i = 0; i < 32 * 66; i++) { memset((unsigned char*) addr + i * 4096, i, 1); if (i % 32 == 31) sleep(1); } sleep(30000); exit(EXIT_SUCCESS); failed: perror("secondary errno:"); exit(EXIT_FAILURE); } - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/