BSD Commit Log Search

   cut.c: minor KNF and other formatting tweaks (NFCI)

   s/liner/linear/ in comment.

   Remove a global that is not, in fact, used in init_main.c, despite what
   the comment claims.

   Consolidate the declaration of phys_map.

   libc: fix undefined behavior in int64_to_float128

   When negating a signed integer, the integer must not be LLONG_MIN, as
   negating that value would not change the sign. GCC's -ftrapv mode
   detects cases like this reliably.

   In this case, on sparc64 with -O2 (but not -O0 or -O1),
   countLeadingZeros64(LLONG_MIN) returned 48 instead of the desired 0, and
   int64_to_float128(LLONG_MIN) returned -0x1p+14 instead of the desired
   -0x1p+63.

   The same pattern is used in several other conversion functions. These
   will be fixed in a follow-up commit once the lint tests for msg_380 and
   msg_381 on sparc64 work again.

   cut.1: add cross-reference to colrm(1)

   Consolidate the declaration of some globals.  More work to be done here.

   Align loading of physical memory with virt68k, and stop keeping two
   copies of this information (one in the phys_seg_list[] and one in
   mem_clusters[]; the latter is now gone).

   Use the data structure from <m68k/seglist.h> to describe the memory
   configuration extracted from linux bootinfo.  Align virt68k's memory
   loading with mvme68k (which will also get some changes in a follow-on
   commit).  This will be used as a template to eventually merge this
   code for all m68k platforms.

   PR 60154: do not print arbitrary control characters when printing
   file names while checking hashes with the -c option.

   Move seglist.h into a common m68k location.

   Remove unnecessary include.

   tests/libc: add function names in tests for long double

   Update for moved tag for ticket #217

   Additionally pull up the following, requested by christos in ticket #217:

        usr.bin/telnet/commands.c       1.82

   fix inverted logic

   tests/libc: test long double operations more thoroughly

   For lint, the tests msg_380.c and msg_381.c fail, which is probably due
   to a bug in the softfloat implementation.

   The tests are intended to yield the same results on all platforms, as
   they do not exercise any tricky rounding details.

   Ammend ticket #1249 for additional pullups

   Additionally pull up the following, requested by christos in ticket #1249:

        usr.bin/telnet/commands.c       1.82

   fix inverted logic.

   Use the common auto-vector interrupt stub.

   Re-factor the simple "basic 4BSD disklabel scheme" into kern/subr_disk_4bsd.c,
   the implementation itself lifted verbatim from hp300.  virt68k differed
   only by RCS IDs, news68k differed in older copyright notice plus some
   whitespace differences in 2 diffrent source lines.

   Only expose block_userspace_access on the kernels it's required for
   (assuming it still is).

   If BUS_DMA_24BIT is defined, clear it from the flags passed to
   _bus_dmamem_alloc_common(), since we've already constrained the
   allocation based on the window configured for the selected VME
   address space.

   mvme68k_bus_dma_tag -> m68k_bus_dma_tag

   (forgot to commit this file in m68k bus_dma consolidation.)

   new openssh out

   ld_virtio.c: work around a qemu bug

   qemu vhost-user-blk-server.c advertizes VIRTIO_BLK_F_SIZE_MAX
   with zero VIRTIO_BLK_CONFIG_SIZE_MAX. it doesn't make sense.

   references:
   https://github.com/qemu/qemu/blob/b6a7d06213e5d2f7d124d16418bc289c4a8a4b82/block/export/vhost-user-blk-server.c#L122
   https://github.com/qemu/qemu/blob/b6a7d06213e5d2f7d124d16418bc289c4a8a4b82/block/export/vhost-user-blk-server.c#L243

   machine_userret -> userret

   Merge mvme68k and virt68k bus_dma into the common m68k bus_dma and
   remove their local copies.  Minor cleanup tweaks to other platforms
   that already use the common version.

   Welcome to 11.0_RC3!

   Tickets #238 - #250

   Pull up following revision(s) (requested by yamt in ticket #241):
        etc/etc.amd64/MAKEDEV.conf: revision 1.38
   amd64 MAKEDEV: create 8 nvme (nvme0-nvme7) by default
   why 8?
   - match ld, sd, wd
   - my machine has 6 of them