sanchayanmaity/minix
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity
minix/servers/is/dmp_kernel.c
Ben Gras 50e2064049 No more intel/minix segments. 
This commit removes all traces of Minix segments (the text/data/stack
memory map abstraction in the kernel) and significance of Intel segments
(hardware segments like CS, DS that add offsets to all addressing before
page table translation). This ultimately simplifies the memory layout
and addressing and makes the same layout possible on non-Intel
architectures.

There are only two types of addresses in the world now: virtual
and physical; even the kernel and processes have the same virtual
address space. Kernel and user processes can be distinguished at a
glance as processes won't use 0xF0000000 and above.

No static pre-allocated memory sizes exist any more.

Changes to booting:
        . The pre_init.c leaves the kernel and modules exactly as
          they were left by the bootloader in physical memory
        . The kernel starts running using physical addressing,
          loaded at a fixed location given in its linker script by the
          bootloader.  All code and data in this phase are linked to
          this fixed low location.
        . It makes a bootstrap pagetable to map itself to a
          fixed high location (also in linker script) and jumps to
          the high address. All code and data then use this high addressing.
        . All code/data symbols linked at the low addresses is prefixed by
          an objcopy step with __k_unpaged_*, so that that code cannot
          reference highly-linked symbols (which aren't valid yet) or vice
          versa (symbols that aren't valid any more).
        . The two addressing modes are separated in the linker script by
          collecting the unpaged_*.o objects and linking them with low
          addresses, and linking the rest high. Some objects are linked
          twice, once low and once high.
        . The bootstrap phase passes a lot of information (e.g. free memory
          list, physical location of the modules, etc.) using the kinfo
          struct.
        . After this bootstrap the low-linked part is freed.
        . The kernel maps in VM into the bootstrap page table so that VM can
          begin executing. Its first job is to make page tables for all other
          boot processes. So VM runs before RS, and RS gets a fully dynamic,
          VM-managed address space. VM gets its privilege info from RS as usual
          but that happens after RS starts running.
        . Both the kernel loading VM and VM organizing boot processes happen
	  using the libexec logic. This removes the last reason for VM to
	  still know much about exec() and vm/exec.c is gone.

Further Implementation:
        . All segments are based at 0 and have a 4 GB limit.
        . The kernel is mapped in at the top of the virtual address
          space so as not to constrain the user processes.
        . Processes do not use segments from the LDT at all; there are
          no segments in the LDT any more, so no LLDT is needed.
        . The Minix segments T/D/S are gone and so none of the
          user-space or in-kernel copy functions use them. The copy
          functions use a process endpoint of NONE to realize it's
          a physical address, virtual otherwise.
        . The umap call only makes sense to translate a virtual address
          to a physical address now.
        . Segments-related calls like newmap and alloc_segments are gone.
        . All segments-related translation in VM is gone (vir2map etc).
        . Initialization in VM is simpler as no moving around is necessary.
        . VM and all other boot processes can be linked wherever they wish
          and will be mapped in at the right location by the kernel and VM
          respectively.

Other changes:
        . The multiboot code is less special: it does not use mb_print
          for its diagnostics any more but uses printf() as normal, saving
          the output into the diagnostics buffer, only printing to the
          screen using the direct print functions if a panic() occurs.
        . The multiboot code uses the flexible 'free memory map list'
          style to receive the list of free memory if available.
        . The kernel determines the memory layout of the processes to
          a degree: it tells VM where the kernel starts and ends and
          where the kernel wants the top of the process to be. VM then
          uses this entire range, i.e. the stack is right at the top,
          and mmap()ped bits of memory are placed below that downwards,
          and the break grows upwards.

Other Consequences:
        . Every process gets its own page table as address spaces
          can't be separated any more by segments.
        . As all segments are 0-based, there is no distinction between
          virtual and linear addresses, nor between userspace and
          kernel addresses.
        . Less work is done when context switching, leading to a net
          performance increase. (8% faster on my machine for 'make servers'.)
	. The layout and configuration of the GDT makes sysenter and syscall
	  possible.
2012-07-15 22:30:15 +02:00

382 lines
12 KiB
C
Raw Blame History

/* Debugging dump procedures for the kernel. */
#include "inc.h"
#include <timers.h>
#include <machine/interrupt.h>
#include <minix/endpoint.h>
#include <minix/sysutil.h>
#include <minix/sys_config.h>
#include "kernel/const.h"
#include "kernel/config.h"
#include "kernel/debug.h"
#include "kernel/type.h"
#include "kernel/proc.h"
#include "kernel/ipc.h"
#define LINES 22
#define PRINTRTS(rp) { \
char *procname = ""; \
printf(" %s", p_rts_flags_str(rp->p_rts_flags)); \
if (rp->p_rts_flags & RTS_SENDING) \
procname = proc_name(_ENDPOINT_P(rp->p_sendto_e)); \
else if (rp->p_rts_flags & RTS_RECEIVING) \
procname = proc_name(_ENDPOINT_P(rp->p_getfrom_e)); \
printf(" %-7.7s", procname); \
}
static int pagelines;
#define PROCLOOP(rp, oldrp) \
pagelines = 0; \
for (rp = oldrp; rp < END_PROC_ADDR; rp++) { \
oldrp = BEG_PROC_ADDR; \
if (isemptyp(rp)) continue; \
if (++pagelines > LINES) { oldrp = rp; printf("--more--\n"); break; }\
if (proc_nr(rp) == IDLE) printf("(%2d) ", proc_nr(rp)); \
else if (proc_nr(rp) < 0) printf("[%2d] ", proc_nr(rp)); \
else printf(" %2d ", proc_nr(rp));
#define click_to_round_k(n) \
((unsigned) ((((unsigned long) (n) << CLICK_SHIFT) + 512) / 1024))
/* Declare some local dump procedures. */
static char *proc_name(int proc_nr);
static char *s_traps_str(int flags);
static char *s_flags_str(int flags);
static char *p_rts_flags_str(int flags);
/* Some global data that is shared among several dumping procedures.
* Note that the process table copy has the same name as in the kernel
* so that most macros and definitions from proc.h also apply here.
*/
struct proc proc[NR_TASKS + NR_PROCS];
struct priv priv[NR_SYS_PROCS];
struct boot_image image[NR_BOOT_PROCS];
/*===========================================================================*
* kmessages_dmp *
*===========================================================================*/
void kmessages_dmp()
{
struct kmessages kmess; /* get copy of kernel messages */
char print_buf[_KMESS_BUF_SIZE+1]; /* this one is used to print */
int start; /* calculate start of messages */
int r;
/* Try to get a copy of the kernel messages. */
if ((r = sys_getkmessages(&kmess)) != OK) {
printf("IS: warning: couldn't get copy of kmessages: %d\n", r);
return;
}
/* Try to print the kernel messages. First determine start and copy the
* buffer into a print-buffer. This is done because the messages in the
* copy may wrap (the kernel buffer is circular).
*/
start = ((kmess.km_next + _KMESS_BUF_SIZE) - kmess.km_size) % _KMESS_BUF_SIZE;
r = 0;
while (kmess.km_size > 0) {
print_buf[r] = kmess.km_buf[(start+r) % _KMESS_BUF_SIZE];
r ++;
kmess.km_size --;
}
print_buf[r] = 0; /* make sure it terminates */
printf("Dump of all messages generated by the kernel.\n\n");
printf("%s", print_buf); /* print the messages */
}
/*===========================================================================*
* monparams_dmp *
*===========================================================================*/
void monparams_dmp()
{
char val[MULTIBOOT_PARAM_BUF_SIZE];
char *e;
int r;
/* Try to get a copy of the boot monitor parameters. */
if ((r = sys_getmonparams(val, sizeof(val))) != OK) {
printf("IS: warning: couldn't get copy of monitor params: %d\n", r);
return;
}
/* Append new lines to the result. */
e = val;
do {
e += strlen(e);
*e++ = '\n';
} while (*e != 0);
/* Finally, print the result. */
printf("Dump of kernel environment strings set by boot monitor.\n");
printf("\n%s\n", val);
}
/*===========================================================================*
* irqtab_dmp *
*===========================================================================*/
void irqtab_dmp()
{
int i,r;
struct irq_hook irq_hooks[NR_IRQ_HOOKS];
int irq_actids[NR_IRQ_VECTORS];
struct irq_hook *e; /* irq tab entry */
if ((r = sys_getirqhooks(irq_hooks)) != OK) {
printf("IS: warning: couldn't get copy of irq hooks: %d\n", r);
return;
}
if ((r = sys_getirqactids(irq_actids)) != OK) {
printf("IS: warning: couldn't get copy of irq mask: %d\n", r);
return;
}
#if 0
printf("irq_actids:");
for (i= 0; i<NR_IRQ_VECTORS; i++)
printf(" [%d] = 0x%08x", i, irq_actids[i]);
printf("\n");
#endif
printf("IRQ policies dump shows use of kernel's IRQ hooks.\n");
printf("-h.id- -proc.nr- -irq nr- -policy- -notify id- -masked-\n");
for (i=0; i<NR_IRQ_HOOKS; i++) {
e = &irq_hooks[i];
printf("%3d", i);
if (e->proc_nr_e==NONE) {
printf(" <unused>\n");
continue;
}
printf("%10d ", e->proc_nr_e);
printf(" (%02d) ", e->irq);
printf(" %s", (e->policy & IRQ_REENABLE) ? "reenable" : " - ");
printf(" %4lu", e->notify_id);
if (irq_actids[e->irq] & e->id)
printf(" masked");
printf("\n");
}
printf("\n");
}
/*===========================================================================*
* image_dmp *
*===========================================================================*/
void image_dmp()
{
int m, r;
struct boot_image *ip;
if ((r = sys_getimage(image)) != OK) {
printf("IS: warning: couldn't get copy of image table: %d\n", r);
return;
}
printf("Image table dump showing all processes included in system image.\n");
printf("---name- -nr- flags -stack-\n");
for (m=0; m<NR_BOOT_PROCS; m++) {
ip = &image[m];
printf("%8s %4d\n", ip->proc_name, ip->proc_nr);
}
printf("\n");
}
/*===========================================================================*
* kenv_dmp *
*===========================================================================*/
void kenv_dmp()
{
struct kinfo kinfo;
struct machine machine;
int r;
if ((r = sys_getkinfo(&kinfo)) != OK) {
printf("IS: warning: couldn't get copy of kernel info struct: %d\n", r);
return;
}
if ((r = sys_getmachine(&machine)) != OK) {
printf("IS: warning: couldn't get copy of kernel machine struct: %d\n", r);
return;
}
printf("Dump of kinfo structure.\n\n");
printf("Kernel info structure:\n");
printf("- nr_procs: %3u\n", kinfo.nr_procs);
printf("- nr_tasks: %3u\n", kinfo.nr_tasks);
printf("- release: %.6s\n", kinfo.release);
printf("- version: %.6s\n", kinfo.version);
printf("\n");
}
/*===========================================================================*
* s_flags_str *
*===========================================================================*/
static char *s_flags_str(int flags)
{
static char str[10];
str[0] = (flags & PREEMPTIBLE) ? 'P' : '-';
str[1] = (flags & BILLABLE) ? 'B' : '-';
str[2] = (flags & DYN_PRIV_ID) ? 'D' : '-';
str[3] = (flags & SYS_PROC) ? 'S' : '-';
str[4] = (flags & CHECK_IO_PORT) ? 'I' : '-';
str[5] = (flags & CHECK_IRQ) ? 'Q' : '-';
str[6] = (flags & CHECK_MEM) ? 'M' : '-';
str[7] = '\0';
return str;
}
/*===========================================================================*
* s_traps_str *
*===========================================================================*/
static char *s_traps_str(int flags)
{
static char str[10];
str[0] = (flags & (1 << SEND)) ? 'S' : '-';
str[1] = (flags & (1 << SENDA)) ? 'A' : '-';
str[2] = (flags & (1 << RECEIVE)) ? 'R' : '-';
str[3] = (flags & (1 << SENDREC)) ? 'B' : '-';
str[4] = (flags & (1 << NOTIFY)) ? 'N' : '-';
str[5] = '\0';
return str;
}
/*===========================================================================*
* privileges_dmp *
*===========================================================================*/
void privileges_dmp()
{
register struct proc *rp;
static struct proc *oldrp = BEG_PROC_ADDR;
register struct priv *sp;
int r, i;
/* First obtain a fresh copy of the current process and system table. */
if ((r = sys_getprivtab(priv)) != OK) {
printf("IS: warning: couldn't get copy of system privileges table: %d\n", r);
return;
}
if ((r = sys_getproctab(proc)) != OK) {
printf("IS: warning: couldn't get copy of process table: %d\n", r);
return;
}
printf("-nr- -id- -name-- -flags- traps grants -ipc_to--"
" -kernel calls-\n");
PROCLOOP(rp, oldrp)
r = -1;
for (sp = &priv[0]; sp < &priv[NR_SYS_PROCS]; sp++)
if (sp->s_proc_nr == rp->p_nr) { r ++; break; }
if (r == -1 && !isemptyp(rp)) {
sp = &priv[USER_PRIV_ID];
}
printf("(%02u) %-7.7s %s %s %6d",
sp->s_id, rp->p_name,
s_flags_str(sp->s_flags), s_traps_str(sp->s_trap_mask),
sp->s_grant_entries);
for (i=0; i < NR_SYS_PROCS; i += BITCHUNK_BITS) {
printf(" %08x", get_sys_bits(sp->s_ipc_to, i));
}
printf(" ");
for (i=0; i < NR_SYS_CALLS; i += BITCHUNK_BITS) {
printf(" %08x", sp->s_k_call_mask[i/BITCHUNK_BITS]);
}
printf("\n");
}
}
/*===========================================================================*
* p_rts_flags_str *
*===========================================================================*/
static char *p_rts_flags_str(int flags)
{
static char str[10];
str[0] = (flags & RTS_PROC_STOP) ? 's' : '-';
str[1] = (flags & RTS_SENDING) ? 'S' : '-';
str[2] = (flags & RTS_RECEIVING) ? 'R' : '-';
str[3] = (flags & RTS_SIGNALED) ? 'I' : '-';
str[4] = (flags & RTS_SIG_PENDING) ? 'P' : '-';
str[5] = (flags & RTS_P_STOP) ? 'T' : '-';
str[6] = (flags & RTS_NO_PRIV) ? 'p' : '-';
str[7] = '\0';
return str;
}
/*===========================================================================*
* proctab_dmp *
*===========================================================================*/
#if (CHIP == INTEL)
void proctab_dmp()
{
/* Proc table dump */
register struct proc *rp;
static struct proc *oldrp = BEG_PROC_ADDR;
int r;
/* First obtain a fresh copy of the current process table. */
if ((r = sys_getproctab(proc)) != OK) {
printf("IS: warning: couldn't get copy of process table: %d\n", r);
return;
}
printf("\n-nr-----gen---endpoint-name--- -prior-quant- -user----sys-rtsflags-from/to-\n");
PROCLOOP(rp, oldrp)
printf(" %5d %10d ", _ENDPOINT_G(rp->p_endpoint), rp->p_endpoint);
printf("%-8.8s %5u %5u %6lu %6lu ",
rp->p_name,
rp->p_priority,
rp->p_quantum_size_ms,
rp->p_user_time, rp->p_sys_time);
PRINTRTS(rp);
printf("\n");
}
}
#endif /* (CHIP == INTEL) */
/*===========================================================================*
* procstack_dmp *
*===========================================================================*/
void procstack_dmp()
{
/* Proc table dump, with stack */
register struct proc *rp;
static struct proc *oldrp = BEG_PROC_ADDR;
int r;
/* First obtain a fresh copy of the current process table. */
if ((r = sys_getproctab(proc)) != OK) {
printf("IS: warning: couldn't get copy of process table: %d\n", r);
return;
}
printf("\n-nr-rts flags-- --stack--\n");
PROCLOOP(rp, oldrp)
PRINTRTS(rp);
sys_sysctl_stacktrace(rp->p_endpoint);
}
}
/*===========================================================================*
* proc_name *
*===========================================================================*/
static char *proc_name(proc_nr)
int proc_nr;
{
struct proc *p;
if (proc_nr == ANY) return "ANY";
if (proc_nr == NONE) return "NONE"; /* bogus */
if (proc_nr < -NR_TASKS || proc_nr >= NR_PROCS) return "BOGUS";
p = proc_addr(proc_nr);
if (isemptyp(p)) return "EMPTY"; /* bogus */
return p->p_name;
}
Reference in a new issue View git blame Copy permalink