PVOs, and so the modified state of the page can no longer be communicated
to the VM layer, causing pages not to be flushed to swap when needed, in
turn causing memory corruption. Also make several correctness adjustments
to I-Cache synchronization and TLB invalidation for 64-bit Book-S CPUs.
Obtained from: projects/ppc64
Discussed with: grehan
MFC after: 2 weeks
This patch basically gives us the best of both worlds. Instead of
forcing the compiler to emulate GNU-style inline semantics even though
we're using ISO C99, it will only use GNU-style inlining when the
compiler is configured that way (__GNUC_GNU_INLINE__).
Tested by: jhb
NO_WCAST_ALIGN. The headers of the standard C++ library are
not 64-bit clean and trigger the warning. This prevents use
of WARNS>=4 on ia64 for example.
Getting the little-endian PCI bus working on the big-endian CPU proved to be
quite challenging. We let the PCI devices be mapped in the "match byte lanes"
address window. This is where they are mapped by the CFE and DMA transfers
generated to or from addresses within this window are not subject to automatic
byte-swapping.
However any access by the driver to memory-mapped pci space is redirected
via the "match bit lanes" address window. We get the benefit of automatic
byte swapping through this address window and drivers don't need to change
to deal with CPU big-endianness.
about to be extracted already exists. The question, and interpretation
of the response is deliberately compatible with Info-Zip.
This change was originally obtained from NetBSD, but has three changes:
- better compatibility with Info-Zip in the handling of ^D
- Use getdelim() rather than getline()
- bug fix: != changed to == in the "file rename" code
I suspect the latter is also a bug in NetBSD, but I can't easily confirm
this.
PR: bin/143307
Reviewed by: rdivacky (change to unzip.c only)
Obtained from: NetBSD src/usr.bin/unzip/unzip.c 1.8
MFC after: 1 month
module. With r203732 it became apparent that creating the sysctl nodes
twice causes at least a warning, however the whole code shouldn't be
present twice in the first place.
Discussed with: rmacklem
the static TLS model, which is fundamentally different from the dynamic
TLS model. The consequence was data corruption. Limit the attribute to
i386 and amd64.
o Use obj->tlsinitsize to determine whether there's initialized data.
o If obj->tlssize > obj->tlsinitsize, then bzero uninitialized data.
o Don't exclude variant I from the work-around in free_tls_offset().
o uses v4 firmware instead of v3. A port will be committed to create
the bwn firmware module.
o supports B/G and LP(low power) PHYs.
o supports 32 / 64 bits DMA operations.
o tested on big / little endian machines so should work on all
architectures.
It'd not connected to the build until the firmware port is committed.
