test from sanitycheck() to validatemakeparams(). If MKUNPRIVED is set
in mk.conf (and not on the build.sh command line) then it's not yet
available for use in the sanitycheck function.
Also move some other code for consistency.
before actually updating anything, since changing the db while the query
that extracts the list of files is still in progress results in
the extraction query failing before it finds everything.
floating point error when raising a signal.
Multiple errors bits can easily be set if different sub-fields of an
xmm simd operation generate different errors.
Replace the hand generated 28 lines that initialised fpetable[]
with some pre-processor expressions.
The latter was verified to give the same table entries.
There is no need to save the fpu state here, and definitely no need
to initialise the fpu.
The code is running with interrupts disabled having trapped on either
an x87 instruction (the one after the one that generated the error)
or on an SSE (etc) instruction that caused the error.
So all it needs to do it obtain the 'error' bits from the relevant
status register, clear the bits, and then raise any signal.
The signal code will save the fp state if the signal itself isn't masked.
It also passes the FP state to the signal handler - which can modify it.
(I suspect that wasn't thecase when this code was written.)
Seems to work for both 64bit and 32bit 'divide by zero' errors.
For the xmm trap, the xmm registers are updated for the result of the
instruction, but the trap returns to re-execute the instruction!
This makes it difficult for the signal handler to do anything sensible.
I've also changed the code to only use unmasked error bits when deciding
the signal code.
The atf_arch and atf_machine configuration variables were removed from
atf-0.19 without me realizing that some tests were querying them directly.
Instead of reintroducing those variables, just rely on compiler builtins
as many other tests already do.
Should fix PR bin/48582.
into the hashlist and discards the now unneeded node it will raise a
panic "dead but not clean".
Reorder the initialization and use ungetnewvnode() to discard the node.
sysctl_copyin fails, 'tmp' is not initialized. This bug is
harmless since only the return value will be different;
it does not expose kernel memory unless diagnostic is enabled.
ok agc@ martin@
inside the ucontext structure passed to signal handlers to modify the
xmm registers.
This should make the code compile - I'm not at all sure it works as expected,
the interactions between FP and signal handlers aren't at all clear.
AFAICT the FP state is saved on the user stack when the handler is called,
however the FP trap code can already done odd things to the FPU....
- memory leaks in virtual host plugged
- ensure hr_host is only the host/port part when the request
contains the hostname in the URI not Host: header.
also update the references to the old http/1.1 draft rev 06
to RFC 2616 (fortunately, most sections hadn't moved.)
This has been tested with both MKKYUA=no (the default) and MKKYUA=yes, as
well as with a non-clean tree and a clean build. Hopefully I got the
details right... but they were tricky!
The main change here is that the atf-config, atf-report, atf-run and
atf-version tools no longer depend on libatf-c nor libatf-c++. Instead,
they depend on an internal libtools.a that contains code specifically
for these tools and nothing else, making them self-contained.
Changes in version 0.19
***********************
Experimental version released on February 7th, 2014.
This is the last release to bundle the code for the deprecated tools.
The next release will drop their code and will stop worrying about
backwards compatibility between the ATF libraries and what the old tools
may or may not support.
If you still require the old tools for some reason, grab a copy of the
'tools' directory now. The code in this directory is standalone and
does not depend on any internal details of atf-c++ any longer.
* Various fixes and improvements to support running as part of the FreeBSD
test suite.
* Project hosting moved from Google Code (as a subproject of Kyua) to
GitHub (as a first-class project). The main reason for the change is
the suppression of binary downloads in Google Code on Jan 15th, 2014.
See https://github.com/jmmv/atf/
* Removed builtin help from atf-sh(1) and atf-check(1) for simplicity
reasons. In other words, their -h option is gone.
* Moved the code of the deprecated tools into a 'tools' directory and
completely decoupled their code from the internals of atf-c++. The
reason for this is to painlessly allow a third-party to maintain a
copy of these tools after we delete them because upcoming changes to
atf-c++ would break the stale tools.
Changes in version 0.18
***********************
Experimental version released on November 16th, 2013.
* Issue 45: Added require.memory support in atf-run for FreeBSD.
* Fixed an issue with the handling of cin with libc++.
* Issue 64: Fixed various mandoc formatting warnings.
* NetBSD PR bin/48284: Made atf-check flush its progress message to
stdout so that an interrupted test case always shows the last message
being executed.
* NetBSD PR bin/48285: Fixed atf_check examples in atf-sh-api(3).
- Delete some unnecessary files.
- Print out both the list of deleted files and added files in the import.
Useful when adjusting file lists and the reachover Makefiles.
- Adjust the syntax - remove "inet" keyword in favour of more explicit
"inet4" for the address family. Consistent with "inet6" for IPv6.
- Adjust and improve the man page a little bit.
definitions match those of i386.
Mostly just structure and field renames, in addition:
1) process_xmm_to_s87() and process_s87_to_xmm() moved into
x86/convert_xmm_s87.c so they can be used by amd64's netbsd32 code.
2) The linux signal code simplified to use a structure copy for ths fxsave
data - it matches the hardware definition and won't change.
The relevant bit is always masked out higher up.
The only place I can find a reference to a 'datachain' error is for the
weitek 1167 fpu. Even the 8087 has the corresponding bit of the x87
status reserved.
Quite why it has appeared here is anybodies guess.
