Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
280,626 Commits 606 Branches 0 Tags 3.1 GiB
fc26bdf7d3
Commit Graph

280626 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
mrg
8484d34059 bump ramdisk image size to match build size. 2020-09-07 09:14:47 +00:00
mrg
e9ae3901e6 asan_interceptors.cc needs -O1 for mipsel/mipseb. 2020-09-07 09:09:40 +00:00
mrg
43051be004 bump image size for pmax64 builds. 2020-09-07 09:05:10 +00:00
mrg
0e30b54082 define __HAVE___LWP_GETPRIVATE_FAST 2020-09-07 07:48:23 +00:00
mrg
d4edddf34a i can't confirm or deny that GCC 9 is wrong about parity_valid[] 
having uninitialised members, but since setting up reconstruct
isn't a hot path, just zero init the whole thing to be sure.
 2020-09-07 07:44:53 +00:00
mrg
4dc66dd97f various updates: 
- most builds are working
- arm64, amd64, sparc, evbmips64-eb, and armv7hf-el are known to be
  stable and work
- mips64-el needs new binutils, in testing
- vax, sh3*, riscv64 and mipse[lb] builds are broken in various ways
 2020-09-07 07:43:36 +00:00
rillig
c7c417c4f1 make(1): migrate SuffExpandChildren to Var_ParsePP 2020-09-07 07:15:26 +00:00
rillig
11dcb9bdaf make(1): migrate to Var_ParsePP in Var_Parse and Var_Subst 2020-09-07 07:10:56 +00:00
mrg
3aea87db6a avoid returning stack garbage on platforms that don't have either 
__lwp_gettcb_fast() or __lwp_getprivate_fast().
 2020-09-07 07:10:43 +00:00
mrg
3a14946041 -march=armv3m option is obsolete. remove it from commented kernel 
configurations (we haven't supported those machines lately anyway),
and, for GCC >= 9, from the one caller.
 2020-09-07 07:08:34 +00:00
mrg
a69c7590ce don't error array-bounds and packed-member issues with GCC 9. 2020-09-07 07:07:32 +00:00
rillig
11af043635 make(1): migrate ApplyModifiers to Var_ParsePP 2020-09-07 07:04:30 +00:00
rillig
193b2ca5f4 make(1): migrate Var_Parse to Var_ParsePP in ParseDoDependency 2020-09-07 06:58:02 +00:00
rillig
163aaa9b04 make(1): migrate Var_Parse to Var_ParsePP in archive handling 2020-09-07 06:51:05 +00:00
rillig
a41c30d038 make(1): remove unnecessary cp-- from Arch_ParseArchive 2020-09-07 06:44:53 +00:00
ryoon
702e22bbae Regenerate 2020-09-07 06:33:05 +00:00
ryoon
ce81c731e8 Add Buffalo BWC-35H01 or BWC-30L01 2020-09-07 06:32:13 +00:00
rillig
40b40038da make(1): save a strlen call in ForIterate 2020-09-07 06:28:22 +00:00
rillig
83915ab45f make(1): constify local variables in ForIterate 2020-09-07 06:27:29 +00:00
rillig
ae6eb8a7cc make(1): split local variable in ForIterate 2020-09-07 06:26:18 +00:00
rillig
953112797d make(1): remove redundant includes 2020-09-07 06:20:07 +00:00
rillig
28d4e7745c make(1): convert ForEscapes from #define to enum 2020-09-07 06:01:11 +00:00
rillig
ca751b452a make(1): add local variable for escapes to for_substitute 
Just for debugging.  No change in code size.
 2020-09-07 05:58:08 +00:00
msaitoh
640882309f - Remove extra callout_stop() in ixgbe_detach(). Found by knakahara@. 
- Rename ix{gbe,v}_free_workqueue() to ix{gbe,v}_free_deferred_handlers().
- Add KASSERT() to functions who are called from ixgbe_handle_admin().
 2020-09-07 05:50:58 +00:00
rillig
4186926f7b make(1): clean up comments in job.c 2020-09-07 05:32:12 +00:00
rillig
3f25d2876e make(1): fix archive test when .CURDIR != .PARSEDIR 2020-09-07 05:16:32 +00:00
msaitoh
a8ef69bb97 __predict_false() should be __predict_true() for adapter->osdep.detaching 
== false check.
 2020-09-07 04:15:12 +00:00
msaitoh
ae292bc5eb Don't schedule admin work while shutdown like the recovery mode timer. 
The admin workqueue also runs while !IFF_UP like the recovery mode timer.
Apply the same change of ixgbe.c 1.254 to ixgbe_schedule_admin_tasklet()
to prevent panic. Found by ozaki-r@.
 2020-09-07 03:57:27 +00:00
thorpej
99c7e1fde7 Add the ability to set an alternate cnmagic in the kernel config 
file, e.g.:

options    CNMAGIC="\"+++++\""
 2020-09-07 03:50:41 +00:00
isaki
0c9974fdba Fix misspellings in comment. 2020-09-07 03:36:11 +00:00
mrg
95c5ec5682 remove GCC_NO_ADDR_OF_PACKED_MEMBER now lfs_accessors.h works. 2020-09-07 03:19:04 +00:00
mrg
b8b349010a remove __packed from struct netbsd32_lsdisable. the structure 
is already properly aligned (the embedded int64 already has
align(4) marker applied where needed.)

same size structure generated.
 2020-09-07 03:12:51 +00:00
mrg
c2082ece61 remove GCC_NO_ADDR_OF_PACKED_MEMBER for several subdir builds 
that are now handled by lfs_accessors.h internally.
 2020-09-07 03:09:53 +00:00
msaitoh
347bdda96a Add CPUID(EAX=07H, ECX=0) ECX bit 16 LA57 from maxv. 2020-09-07 03:03:09 +00:00
riastradh
efc3e3f98c Suppress -Waddress-of-packed-member just for lfs_accessors.h. 
We can remove -Wno-error=address-of-packed-member from various
makefiles now.
 2020-09-07 02:28:12 +00:00
mrg
aebafec3e5 enable -Wno-error=array-bounds for GCC 9 and hppa. machdep.c has a 
bunch of warnings that i can't workaround yet.
 2020-09-07 02:22:27 +00:00
mrg
3b2eb8312b oops, move GCC_NO_ADDR_OF_PACKED_MEMBER into the right header. 2020-09-07 01:54:26 +00:00
mrg
931f53ce3f avoid an uninit warning with GCC 9. 2020-09-07 01:35:25 +00:00
riastradh
e059bdb000 wg: Use threadpool(9) and workqueue(9) for asynchronous tasks. 
- Using threadpool(9) job per interface to receive incoming handshake
  messages gives the same concurrency for active interfaces but
  doesn't waste kthreads for inactive ones.

  => Can't really do this with a global workqueue(9) because there's
     no bound on the amount of time wg_receive_packets() might run
     for; we really need separate threads or threadpool jobs in order
     to avoid having one interface starve all the others.

- Using a global workqueue(9) for asynchronous peer tasks avoids
  creating unnecessary kthreads.

  => Each task does a more or less bounded amount of work, so it's OK
     to share a global workqueue -- there's no advantage to adding
     concurrency for what is almost certainly going to be CPU-bound
     asymmetric crypto.

  => This way we don't need a thread per peer or iteration over a
     list of all peers, so the task mechanism should no longer be a
     bottleneck to scaling to thousands of peers.

XXX This doesn't distribute the load across CPUs -- it keeps it on
the same CPU where the packet came in.  Should consider doing
something to balance the load -- maybe note if the current CPU is
loaded, and if so, sort CPUs by queue length or some other measure of
load and pick the least loaded one or something.
 2020-09-07 01:15:25 +00:00
riastradh
95085a12ec wg: Use a global pktqueue rather than a per-peer pcq. 
- Improves scalability -- won't hit limit on softints no matter how
  many peers there are.
- Improves parallelism -- softint was kernel-locked to serialize
  access to the pcq.
- Requires per-peer queue on handshake init to avoid dropping first
  packet.
  . Per-peer queue is currently a single packet -- should serve well
    enough for pings, dns queries, tcp connections, &c.
 2020-09-07 01:14:42 +00:00
riastradh
340d2bc52e threadpool: Simplify job reference-counting logic. 
Use atomic_load_relaxed while here.
 2020-09-07 01:08:27 +00:00
riastradh
11002ee06b threadpool: Fix man page for threadpool_job_init. 
This is threadpool_job_init(job, fn, interlock, fmt, args...), not
just threadpool_job_init(job, fn, interlock).
 2020-09-07 01:07:38 +00:00
mrg
ebde11d941 make some prototypes match the builtin properly. GCC 9 complains 
with the old version, GCC 8 is happy with this version.

tested on sparc.
 2020-09-07 00:52:19 +00:00
mrg
3c4e3b4cdb avoid calling printf %s with NULL. fix idea from christos. 2020-09-07 00:48:45 +00:00
mrg
577075c0df adjust some fallback code to be slightly less ugly and avoid 
upsetting GCC 9.
 2020-09-07 00:47:27 +00:00
mrg
078f2392b1 avoid a GCC stringop truncate warning. 2020-09-07 00:46:38 +00:00
mrg
79b5664c2b update status for several ports that now mostly build (perhaps with 
some uncommited changes.)
 2020-09-07 00:38:14 +00:00
mrg
ec02653d43 re-gen. 2020-09-07 00:37:23 +00:00
christos
1124ae8b50 Avoid strict aliasing issue by using a separate buffer.. 2020-09-07 00:36:53 +00:00
riastradh
476688215f wg: Fix debug output now that the priority is mixed into it. 2020-09-07 00:33:08 +00:00