CVS log for src/lib/libkvm/kvm_proc.c

Up to [cvs.NetBSD.org] / src / lib / libkvm

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

remove unused.

avoid various use-after-free issues.

create a ptrdiff_t offset between the start of an allocation region and
some interesting pointer, so it can be adjusted with this offset after
realloc() returns.

found by GCC 12.

lib: remove CONSTCOND comment

Since 2021-01-31, lint doesn't need it anymore for the common pattern of
'do ... while (0)'.

read the ps arguments from the correct place.

Get rid of usrstack/USRSTACK. Document that the old version of getargv is
broken because of ASLR.

Match the declaration in <sys/kauth.h>

Merge changes from current as of 20200406

Redo previous more conventionally.  Requested by kre@.

Mask out always zero bits off pg->phys_addr.

some ports need <sys/wait.h>

catch up with p_xstat split.

Rebase to HEAD as of a few days ago.

sync with head.

for a reference, the tree before this commit was tagged
as yamt-pagecache-tag8.

this commit was splitted into small chunks to avoid
a limitation of cvs.  ("Protocol error: too many arguments")

Remove the #include <sys/user.h> from all of libkvm.
sys/user.h is a stub that just #includes sys/pcb.h.
There are no 'struct pcb' anywhere in here, so I'm extremely doubtful
any of the builds will fail.
OTOH it might be relying on a header that pcb.h includes.
In any case i386 and amd64 build.

sync with head

Pull up following revision(s) (requested by martin in ticket #310):
	sys/kern/kern_proc.c: revision 1.184
	lib/libkvm/kvm_proc.c: revision 1.89
Measure kinfo_proc2::p_vm_vsize in pages, as it was always documented.
This value seems to never have been used anywhere.
This makes it consistent with it's cousin p_vm_msize (which is in pages as
well and has several uses).

Measure kinfo_proc2::p_vm_vsize in pages, as it was always documented.
This value seems to never have been used anywhere.
This makes it consistent with it's cousin p_vm_msize (which is in pages as
well and has several uses).

Fix build.

Define KREAD() inside kvm_private.h, for reusability.

Change kvm_pa2off() and kvm_kvatop() prototypes (private to kvm(3)):

-int     _kvm_kvatop(kvm_t *, u_long, u_long *);
-off_t   _kvm_pa2off(kvm_t *, u_long);
+int     _kvm_kvatop(kvm_t *, vaddr_t, paddr_t *);
+off_t   _kvm_pa2off(kvm_t *, paddr_t);

Basically, use vaddr_t for VA and paddr_t for PA. In addition, for variables
representing addresses, use paddr_t or vaddr_t, depending on the context.

For most arches, vaddr_t and paddr_t are equivalent to unsigned long. However,
the change was needed for exotic situations, like i386 PAE, were unsigned long
is not suitable for PA which are 64 bits long. As this required a complete
change of the function prototypes, all arches had to be adapted accordingly.

Core files from before this commit should still work with the new code; I did
not see any direct dependency between core's structure and kvatop/pa2off.

The change was compile tested for all arches, as it impacts all of them.

See also:

http://mail-index.netbsd.org/current-users/2010/09/07/msg014249.html

Ansify and KNF all functions within kvm(3). No objection on current-users@.

Used as ground for u_long => vaddr_t/paddr_t replacements in kvm(3)
private functions.

Compile tested for all arches. See also:

http://mail-index.netbsd.org/current-users/2010/09/07/msg014249.html

Remove uarea swap-out functionality:

- Addresses the issue described in PR/38828.
- Some simplification in threading and sleepq subsystems.
- Eliminates pmap_collect() and, as a side note, allows pmap optimisations.
- Eliminates XS_CTL_DATA_ONSTACK in scsipi code.
- Avoids few scans on LWP list and thus potentially long holds of proc_lock.
- Cuts ~1.5k lines of code.  Reduces amd64 kernel size by ~4k.
- Removes __SWAP_BROKEN cases.

Tested on x86, mips, acorn32 (thanks <mpumford>) and partly tested on
acorn26 (thanks to <bjh21>).

Discussed on <tech-kern>, reviewed by <ad>.

kvm_getprocs actually works with KVM_NO_FILES.

Sync with HEAD.

Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html

Pull up following revision(s) (requested by mrg in ticket #622):
	bin/csh/csh.1: revision 1.46
	bin/csh/func.c: revision 1.37
	bin/ps/print.c: revision 1.111
	bin/ps/ps.c: revision 1.74
	bin/sh/miscbltin.c: revision 1.38
	bin/sh/sh.1: revision 1.92 via patch
	external/bsd/top/dist/machine/m_netbsd.c: revision 1.7
	lib/libkvm/kvm_proc.c: revision 1.82
	sys/arch/mips/mips/cpu_exec.c: revision 1.55
	sys/compat/darwin/darwin_exec.c: revision 1.57
	sys/compat/ibcs2/ibcs2_exec.c: revision 1.73
	sys/compat/irix/irix_resource.c: revision 1.15
	sys/compat/linux/arch/amd64/linux_exec_machdep.c: revision 1.16
	sys/compat/linux/arch/i386/linux_exec_machdep.c: revision 1.12
	sys/compat/linux/common/linux_limit.h: revision 1.5
	sys/compat/osf1/osf1_resource.c: revision 1.14
	sys/compat/svr4/svr4_resource.c: revision 1.18
	sys/compat/svr4_32/svr4_32_resource.c: revision 1.17
	sys/kern/exec_subr.c: revision 1.62
	sys/kern/init_sysctl.c: revision 1.160
	sys/kern/kern_exec.c: revision 1.288
	sys/kern/kern_resource.c: revision 1.151
	sys/sys/param.h: patch
	sys/sys/resource.h: revision 1.31
	sys/sys/sysctl.h: revision 1.184
	sys/uvm/uvm_extern.h: revision 1.153
	sys/uvm/uvm_glue.c: revision 1.136
	sys/uvm/uvm_mmap.c: revision 1.128
	usr.bin/systat/ps.c: revision 1.32
- - add new RLIMIT_AS (aka RLIMIT_VMEM) resource that limits the total
address space available to processes.  this limit exists in most other
modern unix variants, and like most of them, our defaults are unlimited.
remove the old mmap / rlimit.datasize hack.
- - adds the VMCMD_STACK flag to all the stack-creation vmcmd callers.
it is currently unused, but was added a few years ago.
- - add a pair of new process size values to kinfo_proc2{}. one is the
total size of the process memory map, and the other is the total size
adjusted for unused stack space (since most processes have a lot of
this...)
- - patch sh, and csh to notice RLIMIT_AS.  (in some cases, the alias
RLIMIT_VMEM was already present and used if availble.)
- - patch ps, top and systat to notice the new k_vm_vsize member of
kinfo_proc2{}.
- - update irix, svr4, svr4_32, linux and osf1 emulations to support
this information.  (freebsd could be done, but that it's best left
as part of the full-update of compat/freebsd.)
this addresses PR 7897.  it also gives correct memory usage values,
which have never been entirely correct (since mmap), and have been
very incorrect since jemalloc() was enabled.
tested on i386 and sparc64, build tested on several other platforms.
thanks to many folks for feedback and testing but most espcially
chuq and yamt for critical suggestions that lead to this patch not
having a special ugliness i wasn't happy with anyway :-)

- add new RLIMIT_AS (aka RLIMIT_VMEM) resource that limits the total
address space available to processes.  this limit exists in most other
modern unix variants, and like most of them, our defaults are unlimited.
remove the old mmap / rlimit.datasize hack.

- adds the VMCMD_STACK flag to all the stack-creation vmcmd callers.
it is currently unused, but was added a few years ago.

- add a pair of new process size values to kinfo_proc2{}. one is the
total size of the process memory map, and the other is the total size
adjusted for unused stack space (since most processes have a lot of
this...)

- patch sh, and csh to notice RLIMIT_AS.  (in some cases, the alias
RLIMIT_VMEM was already present and used if availble.)

- patch ps, top and systat to notice the new k_vm_vsize member of
kinfo_proc2{}.

- update irix, svr4, svr4_32, linux and osf1 emulations to support
this information.  (freebsd could be done, but that it's best left
as part of the full-update of compat/freebsd.)


this addresses PR 7897.  it also gives correct memory usage values,
which have never been entirely correct (since mmap), and have been
very incorrect since jemalloc() was enabled.

tested on i386 and sparc64, build tested on several other platforms.

thanks to many folks for feedback and testing but most espcially
chuq and yamt for critical suggestions that lead to this patch not
having a special ugliness i wasn't happy with anyway :-)

tdev is unsigned

cast dev_t's to int32_t because XXX: sysctl can't change size. This should
be not a problem since tty's live in the low part of the dev range.

Pull up following revision(s) (requested by cegger in ticket #146):
	lib/libkvm/kvm_proc.c: revision 1.79
check error case for kvm_getlwps or SIGSEGV occurs due to dereferecing a
NULL pointer.
return NULL if there are no threads or proceed further through all loops.
Fixes PR 40004. Ok rmind@

check error case for kvm_getlwps or SIGSEGV occurs due to dereferecing a NULL pointer.
return NULL if there are no threads or proceed further through all loops.
Fixes PR 40004. Ok rmind@

sync with head.

Remove clause 3 and 4 from TNF licenses

sync with HEAD

PR/37962: Andrew Doran: libkvm is busted due to kauth_cred redefinition

Sync structure with kernel version.

Handle reading from raw disk devices.

sync with HEAD

sync with kernel.

sync with HEAD

Catch up with scheduling changes in the kernel.

'true' is a C99 keyword.

treat ESRCH as a soft error. From Anon Ymous

- Fix kvm_getlwps not to spit errors if we lose the race:
	- if the process does not exist anymore, return NULL
	- if the process has created more lwps retry
	XXX[1]: We should fix sysctl_kern_lwp() to return the number of lwps it
		needs in oldlenp when ENOMEM. The we can avoid calling sysctl
		twice (by starting let's say with always 1 lwp, or 10).
	XXX[2]: We should fix kvm_getlwps to never spit errors.
- Silence new lint warnings

fix for yamt-idlelwp

Set  kp2p->p_ru = 0 in the kproc2 structure.
It was only ever non-zero for zombies, and now doesn't exist.
The 'pstats' field has been valid for zombies for a while (it isn't in
the uarea any more).
Fixes part of build breakage.

Revert the kauth_impl.h change. Elad is going to maintain this. Asked by core@

Update only user code that uses the types of the e_[pu]cred members
of struct kinfo_proc.

Pull up following revision(s) (requested by chs in ticket #421):
	lib/libkvm/kvm_proc.c: revision 1.64
fill in p_stat for crashdumps like the kernel does for the sysctl.

Pull up following revision(s) (requested by elad in ticket #415):
	lib/libkvm/kvm_proc.c: revision 1.65
PR/35506: Chuck Silvers: ps doesn't work on crash dumps, libkvm doesn't
know about kauth
Implemented _kvm_convertcred() to handle copying kauth_cred_t to userspace
and converting it to the appropriate 'struct eproc' members ('pcred' and
'ucred' respectively).
Okay mlelstv@.

Sync with kernel changes introduced by merging the newlock2 branch.

PR/35506: Chuck Silvers: ps doesn't work on crash dumps, libkvm doesn't
know about kauth

Implemented _kvm_convertcred() to handle copying kauth_cred_t to userspace
and converting it to the appropriate 'struct eproc' members ('pcred' and
'ucred' respectively).

Okay mlelstv@.

fill in p_stat for crashdumps like the kernel does for the sysctl.

kvm_getproc2: retry on ENOMEM.
it happens when the number of processes are changed in the mean time.

#include a necessary header directly, rather than via user.h.

1. Eliminate some unnecessary to kvm_{m,re}alloc.
2. Don't malloc/free procbase/procbase2/lwpbase continuously. Keep track
   of the size, and only do it if necessary.
3. Write a macro to malloc/realloc and set the size of members so that it
   is done correctly. Previous open coded version in kvm_file.c always
   set the length, which is incorrect.
4. Remove bogus check against INT_MAX.
5. use NULL to initialize pointers instead of 0.

as swap is now optional, ENXIO on /dev/drum is normal.

Fix fallout from vm_anon changes

Fix typos.

Use new PTRTOUINT64() macro instead of local PTRTOINT64() macro.

zero out nl, from Todd Miller. thanks!

available, not avaliable. From miod@openbsd.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22280, verified by myself.

remove scanning of deadproc, since zombproc should contain all the processes
in deadproc according to the comment in kern_proc.c. deadproc is not a
struct proclist anymore, so it would require more work to scan anyway, and
this is unnecessary.

fix LP64 and other warnings

Alternative pid/proc allocater, removes all searches associated with pid
lookup and allocation, and any dependency on NPROC or MAXUSERS.
NO_PID changed to -1 (and renamed NO_PGID) to remove artificial limit
on PID_MAX.
As discussed on tech-kern.

Add p_svuid and p_svgid to kinfo_proc2.  Populate them in the kernel
and in libkvm.  Then teach ps how to show them to you.

Also, teach ps how to show the names for all the uids, the rest of the
group numbers, and the "group access list".

Use _kvm_freeprocs() rather than free() so that kvm_close() won't
free already free'ed region.

Cosmetic changes.

delint (unsigned long long) -> (unsigned long) casts.

Merge the nathanw_sa branch.

Fill in new kinfo_proc2 flags from core.

Use kvm_read() to get l.l_wmesg from a core file.

Freeing the kinfo_lwp structure is centralized.

Make libkvm able to cope with a LWPified kernel.

Add a kvm_getlwps() function to retrieve the LWPs of a specific process.

Pull up revision 1.45 (requested by jdolecek):
  Properly '\0'-terminate data from sysctl() which might not end
  properly if data would be longer than requested.  Fixes PR#11467,
  PR#12389 and possibly PR#12439.

file kvm_proc.c was added on branch nathanw_sa on 2002-04-23 20:10:20 +0000

kvm_doargv2(): make sure the string ends with '\0'; sysctl() might not
	properly end it if the data would be longer than the supplied
	buffer

This fixes lib/11467 by SAITOH Masanobu and lib/12389, maybe also
lib/12439.

Minor fix to kvm_doargv2() - if the kd->argv is reallocated, reinitialize
'ap' too. This doesn't seem to have any effect either way; seems like the
condition is never met?

adapt to latest struct proc, struct sigacts, struct sigctx changes

Pull up revision 1.42 (via patch, requested by he):
  Fix compilation problem after format string audit fixes.

Make this file compiles on alpha.

Pull up revision 1.41 (via patch, requested by sommerfeld):
  Format string audit.

format-string audit cleanups

revert part of previous; u_long -> uintptr_t now that uintptr_t is u_long.

de-lint. Most important problem was the cast from pointer to uintptr_t to
u_int64_t; changed the uintptr_t to u_long.

<vm/vm.h> -> <uvm/uvm_extern.h>

remove redundant vm includes

Sync w/ netbsd-1-5-base.

Use "NULL" instead of "0" in pointer assignments and comparisons.

p_schedflags doesn't exist anymore; just set it to zero for now,
although we'll be able to get the correct info through kmem
groveling later.

Add kvm interface to the new sysctls:
	kvm_getproc2() -> sysctl(KERN_PROC2)
	kvm_getargv2() -> sysctl(KERN_PROC_ARGS, KERN_PROC_ARGV)
	kvm_getenvv2() -> sysctl(KERN_PROC_ARGS, KERN_PROC_ENV)
Add new KVM_NO_FILES flag to kvm_open*() - set up enough state to
use the above calls without having to open any kernel files.

XXX: kvm_getprocs.3 and kvm_open.3 to be updated soon.

Fill in e_sid when building the eproc from a crash dump.

in kvm_getprocs(), check for error from kvm_deadprocs().

More trailing white space.

completely remove Mach VM support.  all that is left is the all the
header files as UVM still uses (most of) these.

catch up with the UVM amap changes.. tested by nathanw@mit.edu, thanks.

Remove lint

Adjust for the reaper process.

XXX Would really like to use proclists[] here, but not right now.

Assign my copyrights to TNF.

bzero->memset, bcopy->memcpy, bcmp->memcmp

Use pread(2) and pwrite(2) rather than lseek(2)/{read,write}(2).

add UVM support.

while it is still kinda broken, at least let it compile -DUVM...

remove obsolete register declarations

Fix compiler warnings.

use <sys/cdefs.h> __RCSID() macro

Instead of: USRSTACK, VM_MIN_ADDRESS, VM_MAXUSER_ADDRESS
Use these:  kd->usrstack, kd->min_uva, kd->max_uva

fix some printf() formats

RCS id police.

slight type cleanup.  add some prototypes for internal functions,
some 'const' poisoning, some lvalue lossage.

Fix up the last argv pointer correctly when terminating early.

Fix one more reallocing bug.

Fix up another pointer if we realloc the argument list.

Update copyright.

Use the new ps_strings format.  Find the arguments using their
addresses in the argv array, rather than guessing based on the location
of NULs.

The sparc port no longer needs special casing.

Eliminate _kvm_uvatop(), in favor of using the machine-independent VM structures.

fix type of variables, add headers, protos, where necessary.

Move the swap page buffer into struct __kvm.  Cache the page size during
kvm_open().  Fix an existing memory leak.

Don't use NBPG/PGOFSET/PGSHIFT. Use getpagesize() and calculate them.

Update to match include files.

update from trunk

Don't return garbage if there are no argument and/or environment strings.

Deal with shadow objects correctly.

new libkvm

Initial revision