This patch will bring the <meminfo> API into line with
that proc(5) document. There were several undocumented
fields that were not noted and these two were omitted:
. 'MmapCopy' was conditional on the #define CONFIG_MMU
. 'Quicklists' depends on the #define CONFIG_QUICKLIST
And we're about to get the following new field in 4.20
which will be represented, at least, in that proc.txt:
. 'KReclaimable' will include SReclaimable plus others
Signed-off-by: Jim Warner <james.warner@comcast.net>
If we ever were to eliminate the procps.h header file,
as discussed in the thread referenced below, then that
would impair the current XTRA_PROCPS_DEBUG provisions.
The only remaining way to verify result types would be
to explicitly include that <proc/xtra-procps-debug.h>.
So, this commit will once again enable the ./configure
provision for defining the -DXTRA_PROCPS_DEBUG option.
Reference(s):
https://www.freelists.org/post/procps/newlib-Qualys-patches,6
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch is the first of three implementing a newlib
branch version of that Jan Rybar master merge request.
With this series we'll ultimately extend 'EXE' support
to both ps and top (plus, everyone else who wants it).
Reference(s):
. master branch merge request
https://gitlab.com/procps-ng/procps/merge_requests/66
Signed-off-by: Jim Warner <james.warner@comcast.net>
Following that patch referenced below, the top SUPGRPS
field would produce a segmentation fault and ps SUPGRP
would often show "(null)". Such problems resulted from
some faulty logic in the status2proc() routine dealing
with 'Groups' (supgid) which served as a source field.
For many processes the original code produced an empty
string which prevented conversion to the expected "-".
Moreover, prior to release 3.3.15 such an empty string
will become 0 after strtol() which pwcache_get_group()
translates to 'root' yielding very misleading results.
So, now we'll check for empty '/proc/#/status/Groups:'
fields & consistently provide a "-" value for callers.
[ we'll also protect against future problems in that ]
[ new qualys logic by always ensuring valid 'supgrp' ]
[ pointers - logic which revealed our original flaw! ]
Reference(s):
. original qualys patch
0071-proc-readproc.c-Harden-supgrps_from_supgids.patch
Signed-off-by: Jim Warner <james.warner@comcast.net>
This refactor was done in response to the Qualys patch
referenced below, which deals with some 'readeither()'
flaws under the master branch. Under our newlib branch
those flaws mostly disappear since the function is now
private. But without a redesign the #define is broken.
When the #define FALSE_THREADS is active, some special
strings showing "[ duplicate ENUM ]" will appear under
each child thread. Note that the real reason for those
appearing isn't being exercised, only their mechanics.
In reality, they only show when a user duplicates such
enums in a results stack & only 1 instance can own it.
Reference(s):
. original qualys patch
0084-proc-readproc.c-Work-around-a-design-flaw-in-readeit.patch
. QUICK_THREADS became FALSE_THREADS
commit c546d9dd44
Signed-off-by: Jim Warner <james.warner@comcast.net>
In the new library 'cmd' is dynamically allocated just
like 'cmdline'. This will align us with the ref below.
Reference(s):
. master branch increase to 64
commit 2cfdbbe897
Signed-off-by: Jim Warner <james.warner@comcast.net>
readeither() caches (in new_p) a pointer to the proc_t of a task-group
leader, but readeither()'s callers can do pretty much anything with the
proc_t structure passed to and/or returned by this function. For
example, they can 1/ free it or 2/ recycle it (by passing it to
readeither() as x).
1/ leads to a use-after-free, and 2/ leads to unexpected behavior when
taskreader()/simple_readtask() is called with new_p equal to x (this is
not a theoretical flaw: 2/ happens in readproctab3() when want_task()
returns false and p is a group leader).
As a workaround, we keep a copy of new_p's first member (tid) in static
storage, and the next times we enter readeither() we check this "canary"
against the tid in new_p: if they differ, we reset new_p to NULL, which
forces the allocation of a new proc_t (the new "leader", or reference).
This always detects 2/ (because free_acquired(x,1) memsets x and hence
new_p); always detects 1/ if freed via free_acquired() and/or freeproc()
(very likely, otherwise memory may be leaked); probably detects 1/ even
if freed directly via free() (because the canary is the first member of
proc_t, likely to be overwritten by free()); but can not detect 1/ if
free() does not write to new_p's chunk at all.
Moreover, accessing new_p->tid to check the canary in case 1/ is itself
a use-after-free, so a better long-term solution should be implemented
at some point (we wanted to avoid intrusive and backward-incompatible
changes in this library function, hence this imperfect workaround).
---------------------------- adapted for newlib branch
. adapted via 'patch' (rejected due to 'xcalloc' ref)
. with loss of both readproctab functions, most no longer true
Signed-off-by: Jim Warner <james.warner@comcast.net>
If QUICK_THREADS is not defined (it is not by default, but most
distributions enable it) and task_dir_missing is true (only on very old
kernels), then readtask() forgets to reset some of the struct proc_t t's
members, which later results in double-free()s in free_acquired().
For now, we simply synchronized the list of members to be reset with the
list of members freed in free_acquired().
---------------------------- adapted for newlib branch
. now 'cmd' is also dynamic
. just synchronized with those freed in free_acquired
. QUICK_THREADS is now FALSE_THREADS, serving different purpose
. entire patch will be effectively reverted with upcoming refactor
Signed-off-by: Jim Warner <james.warner@comcast.net>
Replace xmalloc() with xcalloc().
---------------------------- adapted for newlib branch
. trade xcalloc() for calloc()
. thus we must account for potential ENOMEM
Signed-off-by: Jim Warner <james.warner@comcast.net>
Replace memcpy+strcpy with snprintf.
---------------------------- adapted for newlib branch
. adapted via 'patch' (without rejections)
Signed-off-by: Jim Warner <james.warner@comcast.net>
Check the return value of snprintf(), otherwise dst may point
out-of-bounds when it reaches the end of the dst_buffer (the snprintf()
always returns 1 in that case, even if there is not enough space left),
and vMAX becomes negative and is passed to snprintf() as a size_t.
---------------------------- adapted for newlib branch
. adapted via 'patch (without rejections)
Signed-off-by: Jim Warner <james.warner@comcast.net>
This detects an integer overflow of "strlen + 1", prevents an integer
overflow of "tot + adj + (2 * pSZ)", and avoids calling snprintf with a
string longer than INT_MAX. Truncate rather than fail, since the callers
do not expect a failure of this function.
---------------------------- adapted for newlib branch
. logic is now in pids.c
. former 'vectorize_this_str' is now 'pids_vectorize_this'
Signed-off-by: Jim Warner <james.warner@comcast.net>
1/ Prevent an out-of-bounds write if sz is 0.
2/ Limit sz to INT_MAX, because the return value is an int, not an
unsigned int (and because if INT_MAX is equal to SSIZE_MAX, man 2 read
says "If count is greater than SSIZE_MAX, the result is unspecified.")
3/ Always null-terminate dst (unless sz is 0), because a return value of
0 because of an open() error (for example) is indistinguishable from a
return value of 0 because of an empty file.
4/ Use an unsigned int for i (just like n), not an int.
5/ Check for snprintf() truncation.
---------------------------- adapted for newlib branch
. adapted via 'patch (without rejections)
Signed-off-by: Jim Warner <james.warner@comcast.net>
Note: this is by far the most important and complex patch of the whole
series, please review it carefully; thank you very much!
For this patch, we decided to keep the original function's design and
skeleton, to avoid regressions and behavior changes, while fixing the
various bugs and overflows. And like the "Harden file2str()" patch, this
patch does not fail when about to overflow, but truncates instead: there
is information available about this process, so return it to the caller;
also, we used INT_MAX as a limit, but a lower limit could be used.
The easy changes:
- Replace sprintf() with snprintf() (and check for truncation).
- Replace "if (n == 0 && rbuf == 0)" with "if (n <= 0 && tot <= 0)" and
do break instead of return: it simplifies the code (only one place to
handle errors), and also guarantees that in the while loop either n or
tot is > 0 (or both), even if n is reset to 0 when about to overflow.
- Remove the "if (n < 0)" block in the while loop: it is (and was) dead
code, since we enter the while loop only if n >= 0.
- Rewrite the missing-null-terminator detection: in the original
function, if the size of the file is a multiple of 2047, a null-
terminator is appended even if the file is already null-terminated.
- Replace "if (n <= 0 && !end_of_file)" with "if (n < 0 || tot <= 0)":
originally, it was equivalent to "if (n < 0)", but we added "tot <= 0"
to handle the first break of the while loop, and to guarantee that in
the rest of the function tot is > 0.
- Double-force ("belt and suspenders") the null-termination of rbuf:
this is (and was) essential to the correctness of the function.
- Replace the final "while" loop with a "for" loop that behaves just
like the preceding "for" loop: in the original function, this would
lead to unexpected results (for example, if rbuf is |\0|A|\0|, this
would return the array {"",NULL} but should return {"","A",NULL}; and
if rbuf is |A|\0|B| (should never happen because rbuf should be null-
terminated), this would make room for two pointers in ret, but would
write three pointers to ret).
The hard changes:
- Prevent the integer overflow of tot in the while loop, but unlike
file2str(), file2strvec() cannot let tot grow until it almost reaches
INT_MAX, because it needs more space for the pointers: this is why we
introduced ARG_LEN, which also guarantees that we can add "align" and
a few sizeof(char*)s to tot without overflowing.
- Prevent the integer overflow of "tot + c + align": when INT_MAX is
(almost) reached, we write the maximal safe amount of pointers to ret
(ARG_LEN guarantees that there is always space for *ret = rbuf and the
NULL terminator).
---------------------------- adapted for newlib branch
. there were many formatting differences
. i introduced several myself (especially comments)
. stdlib 'realloc' used, not that home grown xrealloc
. stdlib 'realloc' required extra 'return NULL' statement
Signed-off-by: Jim Warner <james.warner@comcast.net>
1/ Replace sprintf() with snprintf() (and check for truncation).
2/ Prevent an integer overflow of ub->siz. The "tot_read--" is needed to
avoid an off-by-one overflow in "ub->buf[tot_read] = '\0'". It is safe
to decrement tot_read here, because we know that tot_read is equal to
ub->siz (and ub->siz is very large).
We believe that truncation is a better option than failure (implementing
failure instead should be as easy as replacing the "tot_read--" with
"tot_read = 0").
---------------------------- adapted for newlib branch
. no real changes, patch refused due to mem alloc & failure return
Signed-off-by: Jim Warner <james.warner@comcast.net>
1/ Use a "size_t num" instead of an "unsigned num" (also, do not store
the return value of sscanf() into num, it was unused anyway).
2/ Check the return value of strchr() and strrchr().
3/ Never jump over the terminating null byte with "S = tmp + 2".
---------------------------- adapted for newlib branch
. newlib doesn't use that 'unlikely' crap
. the cmd field is now also dynamic (like cmdline)
. thus we must account for potential ENOMEM
Signed-off-by: Jim Warner <james.warner@comcast.net>
1/ Prevent an integer overflow of t.
2/ Avoid an infinite loop if s contains characters other than comma,
spaces, +, -, and digits.
3/ Handle all possible return values of snprintf().
---------------------------- adapted for newlib branch
. we can't use xrealloc(), so we use realloc() instead
. and must account for a mem failure via a return of 1
Signed-off-by: Jim Warner <james.warner@comcast.net>
1/ Do not read past the terminating null byte when hashing the name.
2/ S[x] is used as an index, but S is "char *S" (signed) and hence may
index the array out-of-bounds. Bit-mask S[x] with 127 (the array has 128
entries).
3/ Use a size_t for j, not an int (strlen() returns a size_t).
Notes:
- These are (mostly) theoretical problems, because the contents of
/proc/PID/status are (mostly) trusted.
- The "name" member of the status_table_struct has 8 bytes, and
"RssShmem" occupies exactly 8 bytes, which means that "name" is not
null-terminated. This is fine right now, because status2proc() uses
memcmp(), not strcmp(), but it is worth mentioning.
---------------------------- adapted for newlib branch
. newlib doesn't use that 'unlikely' crap
. newlib also had a '#ifdef FALSE_THREADS'
Signed-off-by: Jim Warner <james.warner@comcast.net>
This function is unused (SIGNAL_STRING is defined by default, and if it
is not, procps does not compile -- for example, there is no "outbuf" in
help_pr_sig()) but fix it anyway. There are two bugs:
- it accepts non-hexadecimal characters (anything >= 0x30);
- "(c - (c>0x57) ? 0x57 : 0x30)" is always equal to 0x57.
---------------------------- adapted for newlib branch
. newlib doesn't use that 'unlikely' crap
Signed-off-by: Jim Warner <james.warner@comcast.net>
This solves several problems:
1/ outbuf[1] was written to, but not outbuf[0], which was left
uninitialized (well, SECURE_ESCAPE_ARGS() already fixes this, but do it
explicitly as well); we know it is safe to write one byte to outbuf,
because SECURE_ESCAPE_ARGS() guarantees it.
2/ If bytes was 1, the write to outbuf[1] was an off-by-one overflow.
3/ Do not call escape_str() with a 0 bufsize if bytes == overhead.
4/ Prevent various buffer overflows if bytes <= overhead.
Simply rearrange the old comparisons. The new comparisons are safe,
because we know from previous checks that:
1/ wlen > 0
2/ my_cells < *maxcells (also: my_cells >= 0 and *maxcells > 0)
3/ len > 1
4/ my_bytes+1 < bufsize (also: my_bytes >= 0 and bufsize > 0)
This should never happen, because wcwidth() is called only if iswprint()
returns nonzero. But belt-and-suspenders, and make it visually clear
(very important for the next patch).
The SECURE_ESCAPE_ARGS() macro solves several potential problems
(although we found no problematic calls to the escape*() functions in
procps's code-base, but had to thoroughly review every call; and this is
library code):
1/ off-by-one overflows if the size of the destination buffer is 0;
2/ buffer overflows if this size (or "maxroom") is negative;
3/ integer overflows (for example, "*maxcells+1");
4/ always null-terminate the destination buffer (unless its size is 0).
---------------------------- adapted for newlib branch
. the escape.c now has just a single exported function
. thus SECURE_ESCAPE_ARGS() is needed in only 2 places
. unlike that original patch, macro is executed 1 time
( not like 'escape_command' calling 'escape_strlist' )
( which might then call 'escape_str' multiple times! )
Signed-off-by: Jim Warner <james.warner@comcast.net>
In the human_readable case; otherwise the strcat() that follows may
append bytes to the previous contents of buf.
Also, slightly enlarge buf, as it was a bit too tight.
Could also replace all sprintf()s with snprintf()s, but all the calls
here output a limited number of characters, so they should be safe.
---------------------------- adapted for newlib branch
. the source file is now proc/uptime.c
. function is now named 'procps_uptime_sprint()'
. new human readable function 'procps_uptime_sprint_short()'
. both were already initialized, so just raised size of 2 buffers
Signed-off-by: Jim Warner <james.warner@comcast.net>
In proc/slab.c, functions parse_slabinfo20() and parse_slabinfo11(),
sscanf() might overflow curr->name, because "String input conversions
store a terminating null byte ('\0') to mark the end of the input; the
maximum field width does not include this terminator."
Add one byte to name[] for this terminator.
---------------------------- adapted for newlib branch
. file is now proc/slabinfo.c (not .h)
. manifest constant renamed SLABINFO_NAME_LEN
. older parse_slabinfo11() function no longer present
Signed-off-by: Jim Warner <james.warner@comcast.net>
When 'newlib' was introduced, in the commit referenced
below, the use of that glibc '__BEGIN_DECLS' macro was
standardized. However, as issue #88 revealed, this may
result in a fatal build error with other environments.
So, this patch just trades that macro for the standard
'#ifdef __cplusplus' conventions (thus avoiding use of
all those '#include <features.h>' directives as well).
Reference(s):
. newlib introduced
commit a410e236ab
. procps-ng-3.3.13 issue
https://gitlab.com/procps-ng/procps/issues/88
. some additional discussion
https://www.freelists.org/post/procps/PATCH-Replace-glibcspecific-macros-in-procnumah,1
. musl wiki (see: sys/cdefs.h error messages)
https://wiki.musl-libc.org/faq.html
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch simply eliminates that glibc specific macro
from all header files which contain no public callable
functions. After all, if user code can't link to them,
then protection from C++ name mangling is unnecessary.
[ we also remove any related '#include <features.h>' ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
Now that the procio logic was removed from the library
we must move the header file, lest we break make dist.
In the process, we will relocate that source file too.
[ we'll take a slightly different approach than that ]
[ used under the master branch by exploiting those 2 ]
[ non-library directories 'include' and 'lib', while ]
[ avoiding any sysctl hard coded function prototype. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
The procio functions that were in the library have been
moved into sysctl. sysctl is not linked to libprocps in
newlib and none of the other procps binaries would need
to read/write large data to the procfs.
References:
be6b048a41
to be able to read and write large buffers below /proc.
The buffers and file offsets are handled dynamically
on the required buffer size at read, that is lseek(2)
is used to determine this size. Large buffers at
write are split at a delimeter into pieces and also
lseek(2) is used to write each of them.
Signed-off-by: Werner Fink <werner@suse.de>
With the documentation update in the commit referenced
below, we should also account for such threads as they
will already be represented in the task/thread totals.
[ and do it in a way that might avoid future changes ]
Reference(s):
commit 91df65b9e7
Signed-off-by: Jim Warner <james.warner@comcast.net>
This removes the following error by stating the task ID can only be 10
characters wide, as it is an integer.
proc/readproc.c: In function ‘simple_nexttid’:
proc/readproc.c:1185:46: warning: ‘%s’ directive output may be truncated writing up to 255 bytes into a region of size between 41 and 51 [-Wformat-truncation=]
snprintf(path, PROCPATHLEN, "/proc/%d/task/%s", p->tgid, ent->d_name);
^~
proc/readproc.c:1185:3: note: ‘snprintf’ output between 14 and 279 bytes into a destination of size 64
snprintf(path, PROCPATHLEN, "/proc/%d/task/%s", p->tgid, ent->d_name);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This commit removes some obsolete parameter validation
code which was needed back when certain functions were
public, called directly by users (1st/2nd generation).
Now that they're static they can be safely eliminated.
Signed-off-by: Jim Warner <james.warner@comcast.net>
With older library logic having been modified to avoid
using those potentially deadly alloc.h routines, while
improving 'errno' handling, we're ready to standardize
and enhance newlib's approach to any potential errors.
In so doing, we'll establish the following objectives:
. . . . . . . . . . . . . functions returning an 'int'
. an error will be indicated by a negative number that
is always the inverse of some well known errno.h value
. . . . . . . . . . . functions returning an 'address'
. any error will be indicated by a NULL return pointer
with the actual reason found in the formal errno value
And, when errno is manipulated directly we will strive
to do so whenever possible within those routines which
have been declared with PROCPS_EXPORT. In other words,
in the user callable functions defined in source last.
[ But, that won't always be possible. In particular, ]
[ all the 'read_failed' functions will sometimes set ]
[ 'errno' so that they can serve callers returning a ]
[ NULL or an int without duplicating a lot of logic. ]
[ Also, that includes one subordinate function which ]
[ was called by 'read_failed' in the <slabinfo> API. ]
------------------------------------------------------
Along the way, several additional miscellaneous issues
were addressed. They're listed here now for posterity.
. the '-1' return value passed outside the library was
eliminated since it would erroneously equate to -EPERM
. the stacks_fetch functions in <diskstats> and <stat>
weren't checked for their possible minus return values
. hash create was not checked in <meminfo> or <vmstat>
. fixed 'new' function faulty parm check in <slabinfo>
Signed-off-by: Jim Warner <james.warner@comcast.net>
While that old master branch library may utilize those
memory allocation functions found in the alloc module,
it was inappropriate for this newlib branch to subject
callers to a stderr message followed by an early exit.
Of course, the old libprocps offered a message handler
override provision (xalloc_err_handler) but that, too,
would seem to be inappropriate for our modern library.
[ remember the battles fought with that damn libnuma ]
So, this commit will tweak those old inherited sources
setting the stage for standardized return values/errno
settings in connection with a memory allocation error.
------------------------------------------------------
Along the way, we'll address the following miscellany:
. Completely eliminate usage of anything from alloc.h.
This, of course, entails our own error checking of the
alternative allocation calls from stdlib.h & string.h.
. Eliminate use of the strdup function where possible,
as with 'procps_uptime' and 'procps_loadavg' routines.
. Whack some obsolete code (getslabinfo) in sysinfo.c.
Signed-off-by: Jim Warner <james.warner@comcast.net>
. ensure whitespace exists between the code & comments
[ changing txt slightly keeps right margin alignment ]
. strive for more consistency with some comment styles
[ don't use C '/*' style where C++ '//' style exists ]
. removed the instance of double space in 1 assignment
[ still striving for consistency in whitespace usage ]
. fixed comment relating to number of 'derived fields'
[ the <meminfo> api recently added one new such enum ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch just rearranges 1 item in the stacks_extent
struct to make it equivalent to all the other modules.
Signed-off-by: Jim Warner <james.warner@comcast.net>
As an oversight, delta values for SWAP amounts weren't
included in the <meminfo> API. Since any runtime costs
of including them only amount to slightly more storage
this commit will simply correct the earlier oversight.
Signed-off-by: Jim Warner <james.warner@comcast.net>
When supplying the -p command to uptime, it does not display any
sections where the value is less than 1; however, after a reboot, this
causes the command to just output "up". Showing 0 minutes when the
system has been up for less than a minute makes it clear a reboot just
occurred.
References:
commit 325d68b7c3
Signed-off-by: Craig Small <csmall@enc.com.au>
Awhile back, that former QUICK_THREADS #define evolved
into the development (only) FALSE_THREADS which can be
used to ensure a 'duplicate ENUM' convention is output
when certain string fields can't be easily duplicated.
Unfortunately, that original implementation was marred
with zeros being displayed for /proc/$$/meminfo fields
in all the child threads for a multi-threaded process.
So this commit corrects that zero memory field buglet.
Reference(s):
. QUICK_THREADS becomes FALSE_THREADS
commit c546d9dd44
Signed-off-by: Jim Warner <james.warner@comcast.net>
The <meminfo> module attempted to duplicate the former
sysinfo memory calculations wherein 'SReclaimable' was
added to 'Cached' for the 'kb_main_cached' equivalent.
But, this original approach was wrong for two reasons.
1. The addition occurred too late to impact the 'USED'
calculation which could then cause an underflow in the
top memory display if 'SReclaimable' was heavily used.
2. In changing the actual /proc/meminfo 'Cached' value
it meant that users could not rely on that proc(5) man
page when interpreting the MEMINFO_MEM_CACHED results.
So this commit adds a new enumerator for the inclusive
cached amount plus repositions the calculation so that
a MEMINFO_MEM_USED result will exclude 'SReclaimable'.
Signed-off-by: Jim Warner <james.warner@comcast.net>
In response to that suggestion referenced below, these
changes allow display of task/thread level NUMA nodes.
Currently, only the 'top' program offers any NUMA type
support and it is limited to the Summary Area display.
With this commit both the 'top' and 'ps' programs will
be able to display NUMA nodes associated with threads.
Reference(s):
https://gitlab.com/procps-ng/procps/issues/58
Signed-off-by: Jim Warner <james.warner@comcast.net>
Reference(s):
proc/readproc.c: In function 'statm2proc'
proc/readproc.c:600:9: warning: variable 'num' set but not used [-Wunused-but-set-variable]
proc/stat.c: In function 'stat_derive_unique':
proc/stat.c:429:1: warning: no return statement in function returning non-void [-Wreturn-type]
ps/parser.c: In function 'arg_type':
ps/parser.c:1098:3: warning: this 'if' clause does not guard... [-Wmisleading-indentation]
ps/parser.c:1099:34: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the 'if'
lib/signals.c: In function 'strtosig':
lib/signals.c:243:9: warning: this 'if' clause does not guard... [-Wmisleading-indentation]
lib/signals.c:245:13: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the 'if'
slabtop.c: In function 'print_summary':
slabtop.c:223:29: warning: unused variable 'stats' [-Wunused-variable]
watch.c: In function 'process_ansi':
watch.c:232:5: warning: this 'if' clause does not guard... [-Wmisleading-indentation]
watch.c:235:2: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the 'if'
Signed-off-by: Jim Warner <james.warner@comcast.net>
When those standardized 'derived' TIC enumerators were
introduced, a problem with potential DELTA distortions
was also introduced when toggling cpus offline/online.
It has always been true that the 1st (summary) line in
/proc/stat will experience a decrease in total tics if
a new cpu is brought online. Such decreases are mostly
due to reductions in 'idle' and 'iowait' tics. Exactly
why such a counterintuitive phenomenon should occur is
a mystery, but this has been acknowledged in proc.txt.
A separate potential distortion arises with individual
cpus. And, here it extends to both bringing processors
online plus taking them offline too. When that happens
the order of the cpus array tracking is upset, placing
the 'new' values in some other processor's array slot.
But even if we were to occupy the same slot, the issue
regarding reductions in 'idle' & 'iowait' still apply.
In all cases, when a DELTA field was found to be minus
it was forced to zero via the 'TICsetH' macro. However
the 'derived' calculations are subject to new forms of
distortion with their own DELTA values. For example we
could find DELTA_SUM_USER + DELTA_SUM_SYSTEM exceeding
DELTA_SUM_TOTAL, an illogical/inappropriate condition.
So this commit moves former protections for individual
cpus to the stat_derive_unique() function and modifies
it to also extend protections to the 'derived' values.
In the process we now protect the cpu 'summary' counts
which were unfortunately previously overlooked (oops).
Reference(s):
. 'derived' types introduced
commit 2c86c4984a
Since its introduction, our evolved /proc/stat API has
relied on a static buffer of 8192 bytes. This approach
is probably Ok for other /proc files but it would only
accommodate around 100 processors. If such a threshold
were exceeded then this interface could never succeed.
Now days 100 processors doesn't seem at all excessive.
So this commit trades that static buffer for a dynamic
self-tuning one. And since so much former top CPU code
was already rolled into this module, we just stole the
already proven top dynamic buffer management code too.
[ this also meant switching low level unbuffered I/O ]
[ calls to standard library buffered I/O calls. that ]
[ is exactly what <slabinfo> and <diskstats> employ. ]
Reference(s):
. 1st gen readstat introduction
commit a410e236ab
Signed-off-by: Jim Warner <james.warner@comcast.net>
With the addition of those new derived SUM values, any
CPUs taken offline or brought online would distort the
historical (delta) results. So this patch just forces
a history reset when such transitions are encountered.
Reference(s):
. derived SUM provisions introduced
commit 2c86c4984a
Signed-off-by: Jim Warner <james.warner@comcast.net>
For each of those interfaces employing a priming read,
all the other 'read' functions begin with the module's
name except this guy which began with 'read_slabinfo'.
Now, they'll all begin with their module name then end
the same with a '_read_failed' boolean hinting suffix.
Signed-off-by: Jim Warner <james.warner@comcast.net>
Upon reflection, at the point where the 'priming read'
was introduced, any possibility of history distortions
was also eliminated. This was true because all of the
'old' (zeroed) data will have been replaced with 'new'
data whenever a user finally calls get, select & reap.
Thus, any DELTA values will automatically reflect that
interval between 'new' and subsequent retrieval calls.
[ diskstats didn't actually employ a 1st time switch ]
[ like the others so we have changed a comment only. ]
[ but that module will retain something similar used ]
[ inside node_update whenever a new node is created. ]
Reference(s):
. priming read added to slabinfo
commit 5d5a52a380
. priming read added to diskstats
commit ecd64f4445
. priming read added to meminfo, stat, vmstat
commit 1a2b62c779
Signed-off-by: Jim Warner <james.warner@comcast.net>
This commit arose out of the discussion (and research)
surrounding the issue cited below. It is an attempt to
consolidate and standardize the calculation of jiffies
categories (e.g. 'idle', 'busy', etc.) once & for all.
Also included is the enum STAT_TIC_NUM_CONTRIBUTORS in
case anyone, in the future, decides to calculate usage
based upon elapsed time * Hz (like top does in process
level %CPU stats). In such an event, a total number of
CPUs or NUMA Nodes would be needed for proper scaling.
Reference(s):
https://gitlab.com/procps-ng/procps/issues/48
Signed-off-by: Jim Warner <james.warner@comcast.net>
Unlike the ps kludge under the master branch to ensure
that namespaces appear the same under both 32 & 64-bit
models, this newlib branch already used a proper type.
However source data still carried the original type as
'signed long' versus that more proper 'unsigned long'.
So, this patch makes sources & destinations identical.
Reference(s):
. master branch ps kludge
commit c41c614b0c
Signed-off-by: Jim Warner <james.warner@comcast.net>
This commit will resolve the RedHat Bugzilla #1322111.
[ import from identical commit against master branch ]
[ but without trailing whitespace, thank you so much ]
Imported by: Jim Warner <james.warner@comcast.net>
In each module employing a priming read at 'new' time,
should that read fail, a call to 'unref' will be made.
However, there is a hidden dependency that these calls
must never occur before the context 'refcount' was set
due to the way an 'unref' conditional was constructed.
So this commit just ensures that 'unref' will function
as expected, even if called with a 'refcount' of zero.
Signed-off-by: Jim Warner <james.warner@comcast.net>
Removing the Item_table 'stat' oldflags for WCHAN_ADDR
was wrong since that 'stat' field is not a constant 0.
Rather, it could assume these 3 values: -1, 0, and +1.
I have not been able to pin down a '-1' result, but it
probably means some sort of permission error (-EPERM).
The '1' or '0' values were supposed to distinguish the
tasks that were or were not blocked (whether there was
a wchan address). However, in practice there is little
correlation between those values and availability of a
kernel symbol in /proc/$$/wchan (perhaps due to race).
Anyway, the real point is that a 'stat' wchan does not
now intentionally contain an address. Thus, outputting
'ffffff', '-' or '1' in programs like ps is senseless.
So this patch just eliminates PIDS_WCHAN_ADDR from our
item enumerators leaving only the PIDS_WCHAN_NAME guy.
Now the new library can't be blamed for bad addresses!
Reference(s):
. removed Item_table 'oldflags'
commit c4aa6c0ab4
. linux removal of wchan addresses
commit b2f73922d119686323f14fbbe46587f863852328
Signed-off-by: Jim Warner <james.warner@comcast.net>
It seems inappropriate to blindly include fields known
to always be zero in our brand new library. Therefore,
this patch removes support for three such enumerators.
[ that stat 'it_real_value' (PIDS_ALARM) field could ]
[ have been made obsolete before a linux 2.6 release ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
This commit just contains some tweaks to comments plus
a few adjustments to whitespace for alignment purposes
and a normalization of the header inclusion #define's.
[ plus a spelling error in one header file was fixed ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
When this module was upgraded to 3rd generation in the
patch referenced below, numa node support was migrated
from the top program into newlib. The 'guest_nice' and
'guest' tics were overlooked as top did not need them.
So, this commit corrects that oversight and achieves a
proper symmetry between the cpu & numa jiffies counts.
Reference(s):
. 3rd gen redesign, numa support imported
commit abc71a46ad
Signed-off-by: Jim Warner <james.warner@comcast.net>
A priming read at 'new' time in that <slabinfo> module
was important so that permission problems are detected
early. Plus, it also had the potential of making delta
values valid when 'get' or 'select' were first called.
It is for that latter reason that such a read was also
incorporated in the <diskstats> module 'new' function.
No other module, however, employed such priming reads.
This patch just brings those potential benefits to all
of our other newlib modules with the exception of that
<pids> guy. That module is, of necessity, sufficiently
different from those others to justify such exclusion.
Not only are there precious few DELTA enums in <pids>,
but the costs of a priming read would be much greater.
[ otherwise, these newly added priming reads have no ]
[ measurable negative impact on performance/timings. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
The <slabinfo> header provides 3 groups of enumerators
with prefixes of SLABINFO, SLABS & SLABNODE. The first
is strictly user oriented & isn't supported internally
by any structure. The other two, however, have structs
associated with 'em but, unfortunately, 1 is misnamed.
The 'struct slabs_node' is associated with 'nodes' and
supports the enumerators with the SLABNODE prefix. But
the 'struct slabs_hist' was associated with 'hist' yet
supports those enumerators with just the SLABS prefix.
We do not care very much what some structure is called
but we do care about an identifier used manipulate it.
This patch will trade the 'hist' identifier associated
with 'struct slabs_hist' for a more congruous 'slabs'.
[ it's awful when the author can't remember what the ]
[ true meaning of an identifier is after creating it ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
If, in fact, numa nodes are dynamic (like that current
total of on-line cpus) the existing logic was lacking.
It included an early return before checking the total.
So, this commit ensures that the nodes total is always
set or updated consistently in only a single function.
There's no need to set it at the time 'new' is called.
[ and since under our existing code this nodes total ]
[ could never possibly have been zero, the erroneous ]
[ test (with the early return) has now been whacked! ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
Still unhappy with a minor memory leak associated with
libnuma, I experimented with omitting the dlclose that
was issued at module's end. For some reason which will
remain a mystery, the valgrind leak then went bye-bye.
So this patch just omits one use of dlclose and relies
on whatever kernel magic is at work to free the memory
when each process ends. We kept, however, the original
code (now commented-out) to serve as a future caution.
There remains one potential (but unlikely) dlclose use
near the original dlopen. But there will be no leak as
that 'numa_node_of_cpu' will not yet have been called.
This seems to be the culprit that triggers such leaks.
None of this libnuma shit would likely have come close
to hitting our fan had the numa developers provided us
with 'new' and 'unref' functions like our newlib does.
Signed-off-by: Jim Warner <james.warner@comcast.net>
Recent profiling and timings have resulted in improved
newlib performance. This patch completes that process.
It just normalizes naming conventions employed for all
allocations involving reaped stacks & history support.
The modules offering a 'reap' function will also offer
the now standardized corresponding STACKS_INCR define.
The modules which provide dynamic history support will
now have a separate #define called NEWOLD_INCR used in
allocations/reallocations. And, while values currently
are set equal to that STACKS_INCR value, in the future
some reason for divorcing those two may be discovered.
----------------------------- for future reference ---
In those modules which contain the STACKS_INCR #define
it is tempting to specify a large value so as to avoid
repeated calls to malloc/realloc. However, in doing so
an extra runtime price will be paid in 'cleanup_stack'
calls with any iterative programs like top or slabtop.
So, with the current values a balance has been sought.
Signed-off-by: Jim Warner <james.warner@comcast.net>
Profiling revealed a large amount of time spent in the
'escape_str_utf8' function (escape.c) with both of our
NLS branches (newlib and master). That same result was
not seen under an ancient top-3.2.8 program & library.
Well, the 3.2.8 result was ultimately explained by the
absence of a 'setlocale', necessary under NLS support.
Thus, when that ancient library tested for locale, all
it got was 'ANSI_...' & assumed 'UTF-8' wasn't active.
But after a hack to that ancient code to place it on a
par with newlib/master, I still found cost differences
that led me to revisit an old change referenced below.
It turns out that 'iswprint' costs far more than would
a call of 'isprint', even with the extra support code.
So this commit just reverts that five year old change.
Reference(s):
commit 7b0fc19e9d
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch contains the following miscellaneous stuff:
. The pids_stacks_fetch() routine might call for newly
allocated stacks to be itemized. However, that job was
already tended to by the pids_stacks_alloc() function.
So, this patch just eliminates a redundant invocation.
------------------------------------------------------
. The concept of 'dirty_stacks' has not kept pace with
the evolving stacks implementation. Originally, stacks
were considered dirty only if free() of dynamic memory
was needed before refreshing any single result struct.
Later, with the introduction of the 'extra' item and a
promise to reset it to zero, 'dirty' was much broader.
So, this patch just treats the dirty flg as others do.
------------------------------------------------------
. Lastly, a word or three about performance & timings.
Tuning efforts concentrated on the <pids> API and top.
And unless an oldlib equivalent to the preceding patch
is applied (favoring stat vs. status), newlib top will
often outperform the oldlib version (obviously wrong).
So assuming /proc/stat is preferred in both libraries,
generally speaking, a cpu and elapsed time increase of
1-5% was found for this new stacks oriented interface.
Of course, there's also an increased memory footprint.
There are some occasions, however, when the newlib top
is at a substantial disadvantage. For example if WCHAN
or TTY is displayed, such items will be present in all
newlib reaped stacks (i.e. every process). But old top
would only incur such overhead with displayable tasks.
Thus, oldlib top could outperform newlib by up to 25%,
for example, if only fields requiring NO library flags
were displayed. However, such a scenario is not likely
since only GID, UID, PID, TGID & WCHAN would be shown.
In the usual case, that overhead associated with WCHAN
and/or TTY is overshadowed by other top runtime costs.
All in all a pleasing outcome I deem quite acceptable.
------------------------------------------------------
Signed-off-by: Jim Warner <james.warner@comcast.net>
Long ago, in a galaxy far away (when top was in charge
of library FILL flgs) /proc/status was to be preferred
over /proc/stat if a field could be satisfied by both.
This was done to avoid costly 64-bit math emulation in
a 32-bit application due to 'unsigned long long' data.
Well it's time to acknowledge the prevalence of 64-bit
platforms. And in such an environment the cost picture
has shifted significantly. It now costs 14 times (wow)
as much to access /proc/status compared to /proc/stat.
In other words, even with '%llu' variables, a sscanf()
call in stat2proc() beats the pants off that home brew
gperf based hashing employed by the status2proc() guy.
In fact, status2proc incurs higher costs than found in
the most expensive aspect of top's forest view option.
Here's a gprof extract to illustrate the costs. It was
produced with an rcfile requiring fields from both the
/proc/stat & /proc/status pseudo files (among others).
There were 5000 iterations in each of 4 separate gprof
runs subsequently merged into 1 gmon.sum for analysis.
% self self
time seconds calls us/call name
----- ------- ------- ------- -----------
28.65 4.10 4689423 0.87 status2proc
26.14 3.74 40000 93.50 forest_adds
...
01.96 0.28 4689427 0.06 stat2proc
[ since forest_adds is recursive, the calls value is ]
[ the non-recursive #, its 'call graph' shows totals ]
Anyway, now that such cost is known this patch becomes
what is euphemistically known as the usual no-brainer.
[ jeeze, was it really this long between profilings? ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch will begin some refinements associated with
gprof. Initially, functions names have been changed to
help in identifying potential bottlenecks. This effort
also included the obscure set, free and sort routines.
Plus the following additional modifications were made:
. the stacks_alloc prologue was generalized plus added
to a couple of modules where it had not yet propagated
. a couple of the '// end ...' comments were corrected
. some functions have been formally tagged as 'inline'
Signed-off-by: Jim Warner <james.warner@comcast.net>
The immediately prior commit demonstrated how our APIs
might be expanded in at some point in the future while
maintaining binary compatibility in previous programs.
However, since we've yet to release the 1st version of
our new library, there's no need to violate alphabetic
ordering just yet. So, this patch restores that order.
Signed-off-by: Jim Warner <james.warner@comcast.net>
With the 4.8 kernel, 2 new fields will be added to the
meminfo pseudo file. This commit, soon to be replaced,
is intended as an example of how such changes might be
incorporated plus still maintain binary compatibility.
This actually goes further than is strictly necessary,
by retaining meminfo_item ordering for 'set' functions
and the creation of hash table entries. However, there
is only 1 true requirement, that of Item_table entries
which must always agree exactly with item enumerators.
All of the other changes could be done alphabetically.
Ok, so what happens when an old program encounters the
new expanded meminfo items? Well, if it was thoroughly
tested against an old library, it won't even see those
new fields. On the other hand, if it somehow exceeds a
previous MEMINFO_logical_end, then it will just get an
extra result structure or two, with no real harm done.
[ this patch is being replace by the very next patch ]
[ so that our iniitial newlib release can maintain a ]
[ strict alphabetic ordering in all areas initially! ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
When the numa stuff was imported from the top program,
that #include for dlopen() was not made conditional as
it should have been. Well, here it is being corrected.
Signed-off-by: Jim Warner <james.warner@comcast.net>
Rather than return a 0 result for all VAL type errors,
return what would have been without validation active.
This will enable a program like pgrep to still print a
result even though it used some incorrect type member.
With this commit, our VAL macro validations logic will
behave in exactly the same way as the GET validations.
While warning messages may be issued, except for a bad
enumerator, values will always be returned to callers.
Signed-off-by: Jim Warner <james.warner@comcast.net>
One ought not to assume that random memory access will
always succeed or, when it does, that an obviously bad
item enumerator will always be found at that location.
Thus, this patch corrects some really poor assumptions
associated with the 'xtra_procps_debug.h' header file.
[ and it does so in somewhat contorted ways so as to ]
[ avoid several darn gcc -Wnonnull warning messages! ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
When users call the native 'get' functions they have a
responsibility to check that the result struct address
was indeed returned. But when using those 'GET' macros
there was no protection for possible NULL dereference.
So this patch will add some protection for a potential
failure of an underlying 'get' function. And should it
occur then those 'GET' macros will just return a zero.
Plus, we'll also mirror that behavior in the debugging
header should the XTRA_PROCPS_DEBUG #define be active.
And, we might as well add a warning when invalid items
are passed to 'GET' macros, just like we do for 'VAL'.
[ lastly, we added the missing opening parens/braces ]
[ to 2 'GET' macros in that xtra-procps-debug.h file ]
[ which went unnoticed until the qa folks caught up. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
During development, we now have a means for validating
that a type referenced in application code matches the
actual type set by the library. The new feature can be
activated through either of the following two methods:
1) ./configure CFLAGS='-DXTRA_PROCPS_DEBUG' (all pgms)
2) an #include <proc/xtra-procps-debug.h> (single pgm)
[ in the future, one could add a formal configure.ac ]
[ provision. but for now a manual approach is safer. ]
Lastly, for any module which provides a sort function,
the handling for both 'noop' & 'extra' enumerators was
made consistent. Now, 'noop' is not sorted and 'extra'
will be sorted as that module's widest supported type.
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch will set the stage for validating the types
referenced in the result union. For now, the parameter
representing that 'info' structure will remain unused.
[ and while we're at it, let us correct a faulty GET ]
[ macro in the diskstats header. that puppy missed a ]
[ parm which ain't so good if that guy is ever used! ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
This commit attempts to minimize the variety of types
currently used. Plus, the following were also changed:
. the MEM fields were switched to parallel the VM guys
. PIDS_MEM_VIRT -> PIDS_MEM_VIRT_PGS
. PIDS_MEM_VIRT_KIB -> PIDS_MEM_VIRT
. made NICE 's_int' so that it then parallels PRIORITY
. change RTPRIO & SCHED_CLASS from 'ul_int' to 's_int'
. removed Item_table 'oldflags' for an obsoleted field
. PIDS_WCHAN_ADDR
. added calculations like TICS_ALL_C for the following
. PIDS_TICS_USER_C
. PIDS_TICS_SYSTEM_C
. these three new 'TICS' fields have been incorporated
. PIDS_TICS_BLKIO - jiffies spent in block i/o
. PIDS_TICS_GUEST - jiffies spent as a guest
. PIDS_TICS_GUEST_C - as above, includes dead children
. that PIDS_TICS_DELTA was renamed PIDS_TICS_ALL_DELTA
( so it did not hide between TICS_BLKIO & TICS_GUEST )
( and to make clearer what's included: utime + stime )
. eliminated 'sl_int' entirely from that result struct
[ often, the <pids> module changes necessitated that ]
[ readproc header and source files had to change too ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
With this patch, all of the modules which offer a reap
function (pids, diskstats, slabinfo and stat too) will
now also provide for sorting whatever had been reaped.
It was easy to overlook a sort function for our <STAT>
guy given the paucity of CPUs on your typical personal
desktop or laptop. However, out in the world one might
find boxes with hundreds of CPUs plus many NUMA nodes.
Hey, who are we to disallow sorts on something another
person might see as useful under the above conditions?
And, there's always something to be said for symmetry.
[ of course, several minor tweaks were also included ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
. a more appropriate error return code was substituted
. a safer form of comparison is utilized in two places
. a STAT_VAL macro replaced by more proper MEMINFO_VAL
. several of the silly 'xerrx' formats had been missed
. a few whitespace changes have also been incorporated
Signed-off-by: Jim Warner <james.warner@comcast.net>
With this patch we've completed a progression toward a
standard approach to naming conventions which follows:
* Only functions will begin with that 'procps_' prefix
. ........................................... examples
. procps_vmstat_get ()
. procps_diskstats_select ()
- ----------------------------------------------------
* Exposed structures begin with the module/header name
. ........................................... examples
. struct pids_info
. struct stat_reaped
- ----------------------------------------------------
* Item enumerators begin like structs, but capitalized
. ........................................... examples
. VMSTAT_COMPACT_FAIL
. MEMINFO_DELTA_ACTIVE
[ slabinfo varies slightly due to some item variants ]
. SLABINFO_extra
. SLABS_SIZE_ACTIVE
. SLABNODE_OBJS_PER_SLAB
[ could cure with a prefix of SLABINFO, but too long ]
- ----------------------------------------------------
* Other enumerators work exactly like item enumerators
. ........................................... examples
. PIDS_SORT_ASCEND
. STAT_REAP_CPUS_AND_NODES
- ----------------------------------------------------
* Macros and constants begin just like the enumerators
. ........................................... examples
. #define SLABINFO_GET
. #define DISKSTATS_TYPE_DISK
- ----------------------------------------------------
Signed-off-by: Jim Warner <james.warner@comcast.net>
Many of our item enumerator identifiers are very long,
especially in that <VMSTAT> module. Additionally, they
all contain the exact same universal 'PROCPS_' prefix.
The origins for this are likely found in the desire to
avoid name clashes with other potential include files.
But with procps-ng newlib, we've probably gone way too
far. Did 'PROCPS_PIDS_TICS_SYSTEM' actually offer more
protection against clash than 'PIDS_TICS_SYSTEM' does?
I don't think so. Besides, no matter how big that name
becomes, one can never guarantee they'll never be some
clash. And, conversely, extremely short names will not
always create conflict. Of course, in either case when
some clash occurs, one can always #undef that problem.
Thus, this commit will eliminate that 'PROCPS_' prefix
making all of those enum identifiers a little shorter.
And, we'll still be well above some ridiculously short
(criminally short) names found in some common headers:
- - - - - - - - - - <term.h>
- 'tab', 'TTY', etc
- - - - - - - - - - - - - - - - <search.h>
- 'ENTER', ENTRY', 'FIND', etc
------------------------------------------------------
Finally, with this as a last of the wholesale changes,
we will have established the naming conventions below:
. only functions will begin with that 'procps_' prefix
. exposed structures begin with the module/header name
. item enumerators begin like structs, but capitalized
. other enumerators work exactly like item enumerators
. macros and constants begin just like the enumerators
------------------------------------------------------
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch attempts to standardize the naming of those
most important (declared not defined) context structs.
The present practice represents a hodge podge of names
only some of which reflect the source /proc file name.
And 2 of those file names embed a literal 'info' which
is likely the origin of that required parm identifier.
Now we'll append a universal '_info' to such structure
names, while including the names of those /proc pseudo
files where possible. In any case, that context struct
will *always* begin with the actual module/header file
name. And only the following two sound a little weird!
---------> 'meminfo_info' + 'slabinfo_info' <---------
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch will bring this interface up to our 3rd gen
standards. The following summarizes the major changes.
* New delta provisions have been added to most fields.
There are, of course, some fields for which a delta is
inappropriate. They include the identifying items such
as name, type, major and minor. Plus the io_inprogress
field which already acts, in effect, as a delta value.
* To provide delta support, dev_node historical values
have become persistent. By the same token, the library
must provide for future removal of disks/partitions. A
timestamp is used to detect 'stale' data which will be
deleted so as not to satisfy some get, select or reap.
* Such persistent support is provided by a linked list
which, by default, grows from the bottom down so as to
maintain compatibility with the /proc/diskstats order.
Initially, I was tempted to use the GNU tsearch (tree)
provisions until I discovered the overhead of building
that tree plus costs of a subsequent 'twalk'. Besides,
walking such a tree means retrieval order would differ
from an order required/expected by the vmstat program.
* The '/sys/block' directory is no longer scanned with
every refresh cycle. Rather, it's only accessed when a
node is first encountered. Then, that node's 'type' is
persistent for its lifetime like several other fields.
* A sort provision was included, at virtually no cost,
even though such a provision was not currently needed.
Signed-off-by: Jim Warner <james.warner@comcast.net>
Where possible, libprocps files convey the name of the
actual source pseudo file under the '/proc' directory.
This brings diskstats into line with such an approach.
Signed-off-by: Jim Warner <james.warner@comcast.net>
[ plus we also play catch up on some earlier changes ]
[ that impacted skill.c, after using --enable-skill! ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
A collection of miscellaneous code and comment tweaks.
[ such changes will stop when desk checking ends too ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
That #define QUICK_THREADS was impossible to implement
under the new <pids> interface. Plus it was also found
to distort some thread information (referenced below).
So, it's always been inactive under the newlib branch.
However, it will (with small changes) still serve some
useful purpose in our library. Now, when the redefined
FALSE_THREADS is active, those special strings showing
"[ duplicate ENUM ]" will appear for any child thread.
Note: the real reason for such strings appearing isn't
being exercised, only their mechanics. In actual usage
they are substituted when a user duplicates such items
in a results stack & only the 1st instance can own it.
With this patch, we are simply fooling the <pids> code
into thinking an item was duplicated via a NULL value.
Reference(s):
. from master branch
commit 25a6ecdbfb
Signed-off-by: Jim Warner <james.warner@comcast.net>
