search for: sigpending

Kay Sievers wrote: > Hi, > this simple program compiled with klibc: > > #include <stdio.h> > int main(int argc, char* argv[]) > { > setenv("TESTENV", "happy", 1); > printf("TESTENV='%s'\n", getenv("TESTENV")); > } > > prints: > > ./env > TESTENV='(null)' > >

...preempt_stop #define resume_kernel restore_nocheck @@ -161,7 +162,7 @@ ret_from_intr: testl $(VM_MASK | 3), %eax jz resume_kernel ENTRY(resume_userspace) - cli # make sure we don't miss an interrupt + CLI # make sure we don't miss an interrupt # setting need_resched or sigpending # between sampling and the iret movl TI_flags(%ebp), %ecx @@ -172,7 +173,7 @@ ENTRY(resume_userspace) #ifdef CONFIG_PREEMPT ENTRY(resume_kernel) - cli + CLI cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ? jnz restore_nocheck need_resched: @@ -193,7 +194,7 @@ ENTRY(sysente...

...preempt_stop #define resume_kernel restore_nocheck @@ -161,7 +162,7 @@ ret_from_intr: testl $(VM_MASK | 3), %eax jz resume_kernel ENTRY(resume_userspace) - cli # make sure we don't miss an interrupt + CLI # make sure we don't miss an interrupt # setting need_resched or sigpending # between sampling and the iret movl TI_flags(%ebp), %ecx @@ -172,7 +173,7 @@ ENTRY(resume_userspace) #ifdef CONFIG_PREEMPT ENTRY(resume_kernel) - cli + CLI cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ? jnz restore_nocheck need_resched: @@ -193,7 +194,7 @@ ENTRY(sysente...

...preempt_stop #define resume_kernel restore_nocheck @@ -148,7 +149,7 @@ ret_from_intr: testl $(VM_MASK | 3), %eax jz resume_kernel ENTRY(resume_userspace) - cli # make sure we don't miss an interrupt + CLI # make sure we don't miss an interrupt # setting need_resched or sigpending # between sampling and the iret movl TI_flags(%ebp), %ecx @@ -159,7 +160,7 @@ ENTRY(resume_userspace) #ifdef CONFIG_PREEMPT ENTRY(resume_kernel) - cli + CLI cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ? jnz restore_nocheck need_resched: @@ -179,7 +180,7 @@ need_resched:...

...preempt_stop #define resume_kernel restore_nocheck @@ -148,7 +149,7 @@ ret_from_intr: testl $(VM_MASK | 3), %eax jz resume_kernel ENTRY(resume_userspace) - cli # make sure we don't miss an interrupt + CLI # make sure we don't miss an interrupt # setting need_resched or sigpending # between sampling and the iret movl TI_flags(%ebp), %ecx @@ -159,7 +160,7 @@ ENTRY(resume_userspace) #ifdef CONFIG_PREEMPT ENTRY(resume_kernel) - cli + CLI cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ? jnz restore_nocheck need_resched: @@ -179,7 +180,7 @@ need_resched:...

How can I tell if this is normal behaviour? Oracle imports are horribly slow, an order of magnitude slower than on the same hardware with a slower disk array and Solaris 9. What I can look for to see where the problem lies? The server is 99% idle right now, with one database running. Each sample is about 5 seconds. I''ve tried setting kernel parameters despite the docs saying that

...r/klibc/SYSCALLS.def b/usr/klibc/SYSCALLS.def index 16faabcb..0768df0d 100644 --- a/usr/klibc/SYSCALLS.def +++ b/usr/klibc/SYSCALLS.def @@ -212,7 +212,7 @@ ssize_t sendfile64,sendfile::sendfile(int, int, off_t *, size_t, off_t); int rt_sigsuspend::__rt_sigsuspend(const sigset_t *, size_t); int rt_sigpending::__rt_sigpending(sigset_t *, size_t); int rt_sigprocmask::__rt_sigprocmask(int, const sigset_t *, sigset_t *, size_t); -<sparc64> void rt_sigreturn::__sigreturn(); +<sparc,sparc64> void rt_sigreturn::__sigreturn(); #else int sigaction::__sigaction(int, const struct sigaction *, struc...

...r/klibc/SYSCALLS.def b/usr/klibc/SYSCALLS.def index 16faabcb..0768df0d 100644 --- a/usr/klibc/SYSCALLS.def +++ b/usr/klibc/SYSCALLS.def @@ -212,7 +212,7 @@ ssize_t sendfile64,sendfile::sendfile(int, int, off_t *, size_t, off_t); int rt_sigsuspend::__rt_sigsuspend(const sigset_t *, size_t); int rt_sigpending::__rt_sigpending(sigset_t *, size_t); int rt_sigprocmask::__rt_sigprocmask(int, const sigset_t *, sigset_t *, size_t); -<sparc64> void rt_sigreturn::__sigreturn(); +<sparc,sparc64> void rt_sigreturn::__sigreturn(); #else int sigaction::__sigaction(int, const struct sigaction *, struc...

...n - 1; + + while ( n-- ) { + if ( *sp == (unsigned char)c ) + return (void *)sp; + sp--; + } + + return NULL; +} --- klibc-0.194/klibc/Makefile.orig 2005-01-06 23:09:57.423628424 -0700 +++ klibc-0.194/klibc/Makefile 2005-01-06 23:09:48.031056312 -0700 @@ -30,7 +30,7 @@ sigaction.o sigpending.o sigprocmask.o sigsuspend.o \ brk.o sbrk.o malloc.o realloc.o calloc.o mmap.o \ memcpy.o memcmp.o memset.o memccpy.o memmem.o memswap.o \ - memmove.o memchr.o \ + memmove.o memchr.o memrchr.o \ strcasecmp.o strncasecmp.o strndup.o strerror.o \ strcat.o strchr.o strcmp.o strcpy....

...gnal(int, __sighandler_t); __extern __sighandler_t bsd_signal(int, __sighandler_t); +/* make sure classic code works */ +#define signal bsd_signal __extern int sigaction(int, const struct sigaction *, struct sigaction *); __extern int sigprocmask(int, const sigset_t *, sigset_t *); __extern int sigpending(sigset_t *);

Hi Andrew, This series of patches lays the basic ground work for the paravirtualized kernel patches coming later on. I think this lot is ready for the rough-and-tumble world of the -mm tree. The main change from the last posting is that all the page-table related patches have been moved out, and will be posted separately. Also, the off-by-one in reserving the top of address space has been

Hi Andrew, This series of patches lays the basic ground work for the paravirtualized kernel patches coming later on. I think this lot is ready for the rough-and-tumble world of the -mm tree. The main change from the last posting is that all the page-table related patches have been moved out, and will be posted separately. Also, the off-by-one in reserving the top of address space has been

...r/klibc/SYSCALLS.def b/usr/klibc/SYSCALLS.def index 461d9bf6..30ff58b8 100644 --- a/usr/klibc/SYSCALLS.def +++ b/usr/klibc/SYSCALLS.def @@ -212,7 +212,6 @@ ssize_t sendfile64,sendfile::sendfile(int, int, off_t *, size_t, off_t); int rt_sigsuspend::__rt_sigsuspend(const sigset_t *, size_t); int rt_sigpending::__rt_sigpending(sigset_t *, size_t); int rt_sigprocmask::__rt_sigprocmask(int, const sigset_t *, sigset_t *, size_t); -<sparc64> void rt_sigreturn::__sigreturn(); #else int sigaction::__sigaction(int, const struct sigaction *, struct sigaction *); int sigpending(sigset_t *); @@ -229,6 +2...

...eax movb CS(%esp), %al - testl $(VM_MASK | 3), %eax - jz resume_kernel + andb $SEGMENT_RPL_MASK, %al + cmpl $SEGMENT_RPL_MASK, %eax + jb resume_kernel # returning to kernel or vm86-space ENTRY(resume_userspace) cli # make sure we don't miss an interrupt # setting need_resched or sigpending @@ -386,17 +388,14 @@ syscall_badsys: /* put ESP to the proper location */ \ movl %eax, %esp; #define UNWIND_ESPFIX_STACK \ - pushl %eax; \ - movl %ss, %eax; \ - /* see if on 16bit stack */ \ - cmpw $__ESPFIX_SS, %ax; \ + COMPARE_SEGMENT_REG(__ESPFIX_SS, %ss); \ jne 28f; \ - movl $__KERNEL_D...

...eax movb CS(%esp), %al - testl $(VM_MASK | 3), %eax - jz resume_kernel + andb $SEGMENT_RPL_MASK, %al + cmpl $SEGMENT_RPL_MASK, %eax + jb resume_kernel # returning to kernel or vm86-space ENTRY(resume_userspace) cli # make sure we don't miss an interrupt # setting need_resched or sigpending @@ -386,17 +388,14 @@ syscall_badsys: /* put ESP to the proper location */ \ movl %eax, %esp; #define UNWIND_ESPFIX_STACK \ - pushl %eax; \ - movl %ss, %eax; \ - /* see if on 16bit stack */ \ - cmpw $__ESPFIX_SS, %ax; \ + COMPARE_SEGMENT_REG(__ESPFIX_SS, %ss); \ jne 28f; \ - movl $__KERNEL_D...

...; also exit since it's the same thread group and it does: > > do_exit -> exit_files -> put_files_struct -> close_files -> fput Ah. thanks. I confused CLONE_FS in vhost_task_create() with CLONE_FILES. > > Also. Suppose that vhost_worker() dequeues SIGKILL and clears TIF_SIGPENDING. > > > > SIGSTOP, PTRACE_INTERRUPT, freezer can come and set TIF_SIGPENDING again. > > In this case the main for (;;) loop will spin without sleeping until > > vhost_task_should_stop() becomes true? > > I see. So I either have to be able to call get_signal after SIGKIL...

...INFO(%ebp) movl EFLAGS(%esp), %eax # mix EFLAGS and CS movb CS(%esp), %al - testl $(VM_MASK | 3), %eax - jz resume_kernel + andl $(VM_MASK | 3), %eax + cmpl $3, %eax + jb resume_kernel ENTRY(resume_userspace) CLI # make sure we don't miss an interrupt # setting need_resched or sigpending @@ -399,10 +400,11 @@ syscall_badsys: #define UNWIND_ESPFIX_STACK \ pushl %eax; \ movl %ss, %eax; \ + MASK_RPL(%ax); \ /* see if on 16bit stack */ \ cmpw $__ESPFIX_SS, %ax; \ jne 28f; \ - movl $__KERNEL_DS, %edx; \ + movl $__USER_DS, %edx; \ movl %edx, %ds; \ movl %edx, %es; \ /* s...

...INFO(%ebp) movl EFLAGS(%esp), %eax # mix EFLAGS and CS movb CS(%esp), %al - testl $(VM_MASK | 3), %eax - jz resume_kernel + andl $(VM_MASK | 3), %eax + cmpl $3, %eax + jb resume_kernel ENTRY(resume_userspace) CLI # make sure we don't miss an interrupt # setting need_resched or sigpending @@ -399,10 +400,11 @@ syscall_badsys: #define UNWIND_ESPFIX_STACK \ pushl %eax; \ movl %ss, %eax; \ + MASK_RPL(%ax); \ /* see if on 16bit stack */ \ cmpw $__ESPFIX_SS, %ax; \ jne 28f; \ - movl $__KERNEL_DS, %edx; \ + movl $__USER_DS, %edx; \ movl %edx, %ds; \ movl %edx, %es; \ /* s...

...mber > + * what was set. > + * > + * This thread won't actually exit until all > + * of the file descriptors are closed, and > + * the release function is called. > + */ > + dead = get_signal(&ksig); > + if (dead) > + clear_thread_flag(TIF_SIGPENDING); this can't be right or I am totally confused. Another signal_wake_up() can come right after clear(SIGPENDING). Again, I'll try to re-read this patch, but let me ask anyway... Do we have a plan B? I mean... iirc you have mentioned that you can change these code paths to do something l...

...ONG_LONG_PAIR(HI, LO) HI, LO +#endif + #endif /* _ENDIAN_H */ diff --git a/usr/klibc/Kbuild b/usr/klibc/Kbuild index 40d43c7..d3e2b9f 100644 --- a/usr/klibc/Kbuild +++ b/usr/klibc/Kbuild @@ -35,6 +35,7 @@ klib-y += vsnprintf.o snprintf.o vsprintf.o sprintf.o \ siglongjmp.o \ sigaction.o sigpending.o sigprocmask.o sigsuspend.o \ pselect.o ppoll.o \ + pread.o pwrite.o \ brk.o sbrk.o malloc.o realloc.o zalloc.o calloc.o \ mmap.o shm_open.o shm_unlink.o \ memcpy.o memcmp.o memset.o memccpy.o memmem.o memswap.o \ diff --git a/usr/klibc/SYSCALLS.def b/usr/klibc/SYSCALLS.def index...