CVS log for src/common/lib/libutil/snprintb.c

Up to [cvs.NetBSD.org] / src / common / lib / libutil

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

libutil/snprintb: factor out common subexpression

GCC 10 was not able to optimize the code size on its own, so offer a
little help.

snprintb: reject empty bit descriptions and wrongly placed defaults

snprintb: reject combinations of 'f' with ':' as well as 'F' with '='

These combinations would lead to garbled output.

snprintb: reject empty descriptions

In cases where the snprintb output is garbled, it is not trustworthy, so
make sure the mistakes in the bitfmt are fixed early.

snprintb: remove redundant memset in kernel mode

The provided buffer is already null-terminated by finish_buffer, even in
error cases, there is no need to repeat the same work.

snprintb: mark the end of the buffer if the buffer is too small

This avoids confusion in case the buffer ends with an incomplete number.

snprintb: shrink code size, both in source and binary

snprintb: error out on unknown new-style formatting directive

snprintb: clean up

Use consistent data types for buffer positions and lengths, to avoid
type casts.

snprintb: allow non-ASCII descriptions on all platforms

Previously, these descriptions were only allowed on platforms where
plain 'char' is unsigned. On platforms where plain 'char' is signed,
they invoked undefined behavior or terminated the output early.

snprintb: always null-terminate output

Always null-terminate the output in the buffer, even in error cases. The
wording in the manual page has been promising this since 2008. For
snprintb_m, ensure that the output is terminated with two null
characters, to gracefully handle situations in which the caller does not
check whether snprintb returned an error.

If the buffer size is zero, allow the buffer to be a null pointer,
analogous to snprintf.

Fix an out-of-bounds memory read if the bitfmt ends with a '*' directive
(since today).

In the tests, merge the helper functions for snprintb, snprintb_m, as
they were similar enough.

Fix a few 'line_max exceeded' tests, ensuring that they output a '#'
marker, and that the 'complete' tests don't.

snprintb: error out on out-of-bounds bit shifts

Previously, these invoked undefined behavior, now they lead to an early
return. An example of out-of-bounds bit number is in SCZ_PCICTRL_BITS.
Bit fields that extend beyond the msb are still allowed.

Allow 'f' and 'F' to have fields that are 64 bits wide. This only makes
sense when the field starts at bit 0.

Remove the unused 'val_len', it was only needed before snprintb.c 1.20.

snprintb_m: limit length of generated strings

The manual page promises that the 'max' argument limits the length of
the generated strings, so don't generate any strings that are longer
than that limit. Mark incomplete strings with a trailing '#' to make
them easily recognizable.

snprintb: fix how named bit-fields are split into multiple lines

snprintb: convert macros to local functions

Let the compiler decide whether to inline the functions; allow stepping
through the code in a debugger.

snprintb: do not modify bufsize when producing multiple lines

snprintb: eliminate two local variables, allow zero-size buffer

Like snprintf, the buffer size may be zero. Unlike snprintf, the buffer
must not be NULL.

snprintb: use size_t for buffer sizes and positions

snprintb: fix integer overflow when writing past a small buffer

Previously, snprintb returned -1 in this case, assuming that the
snprintf used in FMTSTR validates against an overly large buffer size.

snprintb: use unsigned integers for parsing the bitfmt

snprintb: fix '=' and ':' for 8-bit values on platforms with signed char

Previously, '=' and ':' worked only on platforms where plain 'char' is
unsigned. On platforms where plain 'char' is signed, only 7-bit values
worked.

snprintb: eliminate a few local variables

snprintb: rename local variables

The single-letter variables 't', 's', 'l' and 'c' were too hard to
decipher.

The variable 'f_len' was used for two independent purposes.

Use a narrow scope for some variables, to avoid having to keep track of
22 individual variables at the same time.

No binary change.

snprintb: clean up

Remove redundant parentheses and casts.

Indent statement-like macros consistently, use separate lines for each
statement, add parentheses to macro definitions.

Remove CONSTCOND comments as lint doesn't need them anymore.

No binary change.

snprintb: fix string termination (since today)

In the previous commit, I had accidentally only run the tests for
snprintb_m but not those for snprintb, thereby missing a newly
introduced bug that would not null-terminate the resulting strings.

Add more tests to cover similar situations in which the buffer is too
small to contain the complete output.

snprintb: rename buflen to bufsize, following the wording in snprintf

snprintb: fix out-of-bounds write

Ooops, restore accidently removed files from merge mishap

Sync with HEAD

Mostly merge changes from HEAD upto 20200411

Formalize that the printf formats should be uintmax_t so we can
uniformly use 'j' in the user-provided formatting strings instead
of depending on _LP64 to use 'll' or 'l' (and the PRI macros). The
alternative is to parse the printf format manually to determine
which modifier to apply which would make this transparent to the
user (they could still always use '%u' or '%x'), but that's too
painful.

PR/54300: Andreas Gustafsson: Remove useless case. should fix
lib/libutil/t_snprintb test regression on sparc

Sync with HEAD

snprintb(3) says that, in the new(?) Torek format, all fields specs end with \0
The F spec is one of those, it should be terminated with \0 just like all
the others (irrelevant that it has no extra data to delimit).

Fix <sys/mman.h> to define the snprintb() format string correctly (include
the missing \0's).   Fix the copy of that definition included into
snprintb(3) to match the updated mman.h version (ride the date bump
from the day before yesterday .. this is the same change, just corrected).

Undo the previous snprintb.c change ("off by one" fix) which was an
attempt to make the broken mman.h usage work (and did, but not the way
it should be done).   Also, after using the new * format (instead of only
when something has already matched) skip the associated data so we don't
attempt to interpret it as more field specifiers.  This func needs lots of TLC!

Fix the ATF tests for snprintb() to not assume that F format is really
exactly like f format, and has data after the field specifier.  It doesn't.
Add several more tests (including testing the '*' field operator
recently added).

Fix an off-by-one in the F modifier that ended up skipping the first : or =
entry, and add a * specifier for a "default case".

Sync with HEAD

Avoid undefined behavior in snprintb.c

Do not change the signedness bit with a left shift operation.
Switch to unsigned integer to prevent this.

snprintb.c:178:17, left shift of 1 by 31 places cannot be represented in type 'int'

Detected with micro-UBSan in the user mode.

doesn't work new format 'F' with a strings of length 1, or a string beginning with 'F'

Rebase to HEAD as of a few days ago.

Rebase.

fix lack of '>' pointed out in PR/48517

PR/48517 do not add extra '>' to string-converted value when nothing matches

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")

Now that all of the macros are enclosed in "do { ... } while 0" we can
remove unneded parenthesization of their invocations.

OK mrg@

Get the correct inttypes.h header, so this can build on !NetBSD.

apply some do { ... } while (0) so that macros aren't broken,
completing the previous change.

The PUT* things are macros.  Use { ... } to make sure they expand correctly.

Fix boundary condition in multi-line format.  When the '=' following a
field name exactly filled a line, the back-up logic failed and really
messed up the formatting!

If HAVE_SNPRINTB_M is defined, then do not define the snprintb() and
snprintb_m() functions.  This should allow this file to be compiled
from a tools build regardless of whether or not the host defines these
functions.

Use "unsigned int" instead of "u_int" to make it easier to steal.
u_int is out of fashion anyway...

sync with head

also handle the separator after the F field.

In the new format F did not print a separator as required.

Sync with HEAD.

Commit is split, to avoid a "too many arguments" protocol error.

Implement snprintb_m(3) to provide multi-line bit/bit-field decode.
Discussed on tech-kern.

fix -Wsign-compare issues

Include sys/param.h in the kernel to make the vax properly pick up
memset.  Otherwise vax/macros.h doesn't get included and the usual
lossage ensues.
(I bet you saw that one coming)

We don't need this for _STANDALONE, and it needs snprintf to work.

share with kernel; replaces bitmask_snprintf which was very broken:
- subtracted one from len without checking if len was 0
- checked for a minimum buffer size
- arguments order was inconsistent with other printf like functions
- no way to tell if it overflowed