PR/57689: RVP: getcwd() not overridable with -D_FORTIFY_SOURCE

bsd.own.mk: rename GCC_NO_* to CC_WNO_*

Rename compiler-warning-disable variables from
GCC_NO_warning
to
CC_WNO_warning
where warning is the full warning name as used by the compiler.

GCC_NO_IMPLICIT_FALLTHRU is CC_WNO_IMPLICIT_FALLTHROUGH

Using the convention CC_compilerflag, where compilerflag
is based on the full compiler flag name.

introduce some common variables for use in GCC warning disables:

GCC_NO_FORMAT_TRUNCATION -Wno-format-truncation (GCC 7/8)
GCC_NO_STRINGOP_TRUNCATION -Wno-stringop-truncation (GCC 8)
GCC_NO_STRINGOP_OVERFLOW -Wno-stringop-overflow (GCC 8)
GCC_NO_CAST_FUNCTION_TYPE -Wno-cast-function-type (GCC 8)

use these to turn off warnings for most GCC-8 complaints. many
of these are false positives, most of the real bugs are already
commited, or are yet to come.


we plan to introduce versions of (some?) of these that use the
"-Wno-error=" form, which still displays the warnings but does
not make it an error, and all of the above will be re-considered
as either being "fix me" (warning still displayed) or "warning
is wrong."

convert HAVE_GCC == 7 to HAVE_GCC >= 7.

- tests explicitly check failure modes, ignore those warnings

branches: 1.7.24;
add stp*cpy() tests

branches: 1.6.6;
Update LLVM/Clang snapshot to r164464. This adopts the GCC options for
the SSP parameters and a not-yet-default rewrite of SROA.

branches: 1.5.2;
Disable new -Wstring-plus-int warning where needed for now.
Adjust various places that add GCC-only options to check for the active
compiler first.

branches: 1.4.2;
reinstate ssp:raw test, slightly differently.. this one is built
with -fstack-protector-all and tests that the compiler built in
stack protection works, by poking directly outside the buffer

don't pass -fstack-protector for these, the tests are actually for the
__*_chk functions included by -D_FORTIFY_SOURCE=2, and if the compiler
provides additional protection, it can mask broken functions..

(none of the _chk functions will return or process any data if
the length given is known to be larger than the buffer)

branches: 1.2.2;
The ssp/raw test is useless. Remove it.

branches: 1.1.2;
Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

PR/57689: RVP: getcwd() not overridable with -D_FORTIFY_SOURCE

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

PR/51893: Ngie Cooper: prevent toolchain from optimizing out memset call

branches: 1.1.28; 1.1.32;
Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

make sum volatile to avoid printing

gcc optimizes away the side effect free call to the "poke" function which
is supposed to smash our stack in this test case.
So: make it print out some dummy value based on the smashing changes.

branches: 1.4.2;
gcc seems to pad small character arrays, use an array of ints

while here, make the poke function non-static, so that it doesnt
get optimised away

reinstate ssp:raw test, slightly differently.. this one is built
with -fstack-protector-all and tests that the compiler built in
stack protection works, by poking directly outside the buffer

The ssp/raw test is useless. Remove it.

branches: 1.1.2;
Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

PR/51894: Ngie Cooper: Use /dev/zero to read instead of stdin

branches: 1.1.28; 1.1.32;
Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

branches: 1.1.2; 1.1.4; 1.1.8;
add stp*cpy() tests

branches: 1.2.2; 1.2.4; 1.2.8;
fix lossage with gcc-4.5

add stp*cpy() tests

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Add __printflike attribution to use vprintf and friends with an argument
as format string.

branches: 1.2.6;
allocate the buffer at the same function level as the compiler so that
the compiler knows to emit the right _chk call.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

Add __printflike attribution to use vprintf and friends with an argument
as format string.

branches: 1.2.6;
allocate the buffer at the same function level as the compiler so that
the compiler knows to emit the right _chk call.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.

PR/57689: RVP: getcwd() not overridable with -D_FORTIFY_SOURCE

add stp*cpy() tests

branches: 1.6.2;
In the strcat test, smash the stack more severely (this all may depend
on alignment and stack frame details).
The gcc folks disagree with this test in general:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59750

branches: 1.5.2; 1.5.8;
reinstate ssp:raw test, slightly differently.. this one is built
with -fstack-protector-all and tests that the compiler built in
stack protection works, by poking directly outside the buffer

branches: 1.4.2;
The ssp/raw test is useless. Remove it.

Disable the "raw" test case. The test is bogus, and did not work before
conversion from src/regress/ to atf

branches: 1.2.2;
Since atf_check doesn't deal with a pipeline, modify the tests to pipe
the output of the echo commands to atf_check itself. This fixes the
three new ssp failures.

Note that the raw, strcpy, vsprintf, and vsnprintf tests still fail, but
they failed in the old 'regress' style, too.

Migrate the ssp tests from regress to atf.

Of the 17 tests, 4 fail in both the regress and atf style. 3 additional
tests fails in atf that did NOT fail in regress: gets, fgets, and read.
I will investigate why, and update when I can, but it is still useful to
get these tests into the new format where they can be exercised.