wsdisplay: update scr_flags properly even after WSDISPLAYIO_MODE_DUMBFB
Fixes PR/60948 (WSDISPLAYIO_GMODE returns wrong value once
MODE_DUMBFB is set).
Should be pulled up to netbsd-11 and netbsd-10.
atari: split out STE and PTE initialization into pmap_bootstrap1()
This prepares for the transition to the m68k NEW_PMAP.
Tested on TT030 (both RELOC enabled and disabled) and Milan.
atari: move crash dump functions to machdep.c as other m68k ports
For now header initialization is not changed, and using temporary
local function declarations until future cleanup on NEW_PMAP transition.
atari: add a missed out of bounds check in memory accouting for Milan
Fortunately currently harmless because usable_segs[NMEM_SEGS].start
is boot_segs[0].start and it's always zero.
Commit fixes provided by Chuck Cranor, thank you very much.
The fixes correct color handling glitches which were the result of
the terminal colour incorrectly being reset to the original pair not
the default color that had been set.
Install the manhtml obsolete set.
While here, remove some files from it that are not actually obsolete,
but in the comp set.
I don't understand the logic for putting files in one or the other,
someone who does: please review them.
Pass a number of bpf_insn to npf_bpf_compile, this should fix kern/60975.
Contrary to what the PR says, it broke loudly only in -current
when I tightened checks in bpfjit_generate_code on July 5th, 2026.
In stable releases it reads past the end of allocations but it generates
valid jitted code.
Why it works in stable releases:
- bpf_validate() checks that a rule ends with a valid RET and
that jumps don't jump past the last RET,
- bpfjit in stable releases doesn't check for a valid RET, it
relies on bpf_validate().
- bpfjit starts by (over)allocating a shadow bpfjit_insn array,
then scans (and goes past the end) instructions, builds CFG
and sets the "unreachable" flag in the shadow array. All
instructions after the last RET have unreachable set to true.
- bpfjit skips emitting any code when unreachable is true
[5 lines not shown]
Use proper naming for registering ether filtering hooks
get_etherpfil_head function name doesn't make sense in that context.
use register_etherpfil_hooks (plural) which registers one hook
(by calling the singular form (register_etherpfil_hook)
per interface at the ether layer.
No functional change intended
Thanks Edgar Fub
set lists: remove support for openssl < 3.5
Should fix 'obsolete' state of quite a number of files
ok riastradh
While here, remove some files that were listed in two sets from one.
x86/efi(4): Set up pcb_onfault handler during efirt access to abort.
This way, if buggy firmware crashes, we have a chance of gracefully
recovering instead of crashing the kernel -- same as we already do on
arm.
PR kern/59235: efi(8) panics
linux ww_mutex: Nix another assertion that trips in sketchy path.
Under the premise (explained in the comment above) that it is
legitimate for the upstream Linux code to do ww_mutex_trylock on a
mutex already held by the current thread, we can't assert
mutex->wwm_state != WW_WANTOWN ||
mutex->wwm_u.ctx->wwx_owner != curlwp
any more than we can assert
mutex->wwm_state != WW_CTX ||
mutex->wwm_u.ctx->wwx_owner != curlwp
because if the latter condition is possible, then any concurrent
thread tries to acquire the mutex a moment before we test that
condition will turn it into the former condition, i.e., transition to
mutex->wwm_state = WW_WANTOWN with mutex->wwm_u.ctx->wwx_owner =
curlwp.
[3 lines not shown]
