svn commit: r353914 - head/lib/libcasper/services/cap_sysctl

2019-10-22 Thread Kyle Evans
Author: kevans Date: Wed Oct 23 03:23:14 2019 New Revision: 353914 URL: https://svnweb.freebsd.org/changeset/base/353914 Log: cap_sysctl: correct typo from r347534-ish operation & ~limit where limit is a bool is clearly not what was intended, given the line prior. Correct it to use the ca

svn commit: r353912 - head/sys/net

2019-10-22 Thread Kyle Evans
Author: kevans Date: Wed Oct 23 02:15:15 2019 New Revision: 353912 URL: https://svnweb.freebsd.org/changeset/base/353912 Log: tuntap(4): Fix NOINET build after r353741 Shuffle headers around to more appropriate #ifdef OPTION blocks (INET vs. INET6) -- double checked LINT-{NOINET,NOINET6,N

svn commit: r353911 - head/lib/libcasper

2019-10-22 Thread Kyle Evans
Author: kevans Date: Wed Oct 23 01:50:41 2019 New Revision: 353911 URL: https://svnweb.freebsd.org/changeset/base/353911 Log: libcasper/services: include to hook tests to build Note that the cap_sysctl tests are currently failing and need some attention. Modified: head/lib/libcasper/M

svn commit: r353909 - head/usr.bin/calendar/calendars

2019-10-22 Thread Greg Lehey
Author: grog Date: Wed Oct 23 01:11:25 2019 New Revision: 353909 URL: https://svnweb.freebsd.org/changeset/base/353909 Log: Correct spelling, apply appropriate respect. Modified: head/usr.bin/calendar/calendars/calendar.holiday Modified: head/usr.bin/calendar/calendars/calendar.holiday =

svn commit: r353908 - head/sys/powerpc/booke

2019-10-22 Thread Justin Hibbits
Author: jhibbits Date: Wed Oct 23 00:31:19 2019 New Revision: 353908 URL: https://svnweb.freebsd.org/changeset/base/353908 Log: powerpc/booke: Fix Book-E boot post-minidump r353489 added minidump support for powerpc64, but it added a dependency on the dump_avail array. Leaving it uniniti

svn commit: r353907 - head

2019-10-22 Thread Jung-uk Kim
Author: jkim Date: Wed Oct 23 00:05:29 2019 New Revision: 353907 URL: https://svnweb.freebsd.org/changeset/base/353907 Log: Belatedly remove stale debug symbols after r339270. Reported by: danfe MFC after:3 days Modified: head/ObsoleteFiles.inc Modified: head/ObsoleteFiles.inc ==

svn commit: r353906 - head/sys/amd64/amd64

2019-10-22 Thread Mateusz Guzik
Author: mjg Date: Tue Oct 22 22:55:46 2019 New Revision: 353906 URL: https://svnweb.freebsd.org/changeset/base/353906 Log: amd64 pmap: conditionalize per-superpage locks on NUMA Instead of superpages use. The current code employs superpage-wide locking regardless and the better locking gr

svn commit: r353905 - head/sys/amd64/amd64

2019-10-22 Thread Mateusz Guzik
Author: mjg Date: Tue Oct 22 22:54:41 2019 New Revision: 353905 URL: https://svnweb.freebsd.org/changeset/base/353905 Log: amd64 pmap: fixup invlgen lookup for fictitious mappings Similarly to r353438, use dummy entry. Reported and tested by: Neel Chauhan Sponsored by: The Free

svn commit: r353904 - head/sys/fs/pseudofs

2019-10-22 Thread Mateusz Guzik
Author: mjg Date: Tue Oct 22 22:52:53 2019 New Revision: 353904 URL: https://svnweb.freebsd.org/changeset/base/353904 Log: pseudofs: hashed vncache Vast majority of uses the cache are just checking if there is an entry present on process exit (and evicting it if so). Both checking and e

svn commit: r353903 - head/sbin/fsck_ffs

2019-10-22 Thread Kirk McKusick
Author: mckusick Date: Tue Oct 22 22:23:59 2019 New Revision: 353903 URL: https://svnweb.freebsd.org/changeset/base/353903 Log: Replace an uninitialized variable with the correct element from the superblock when doing recovery with journalled soft updates. Reported by: Chuck Silvers MF

svn commit: r353902 - head/sys/arm/arm

2019-10-22 Thread Kyle Evans
Author: kevans Date: Tue Oct 22 21:46:03 2019 New Revision: 353902 URL: https://svnweb.freebsd.org/changeset/base/353902 Log: arm: correct kernelstack allocation size This appears to be a copy-pasto from previous lines that propagated to v6 over the years. Indeed, nothing references kerne

svn commit: r353900 - head/sys/dev/cxgbe

2019-10-22 Thread John Baldwin
Author: jhb Date: Tue Oct 22 20:41:54 2019 New Revision: 353900 URL: https://svnweb.freebsd.org/changeset/base/353900 Log: Split Chelsio send tags into a generic base tag and a ratelimit tag. NIC KTLS will add a new TLS send tag type in cxgbe(4) that is a distinct tag from a ratelimit tag

svn commit: r353899 - in head/sys/dev/cxgbe: . tom

2019-10-22 Thread John Baldwin
Author: jhb Date: Tue Oct 22 20:01:47 2019 New Revision: 353899 URL: https://svnweb.freebsd.org/changeset/base/353899 Log: Always allocate the atid table during attach. Previously the table was allocated on first use by TOE and the ratelimit code. The forthcoming NIC KTLS code also uses

svn commit: r353898 - head/contrib/blacklist/bin

2019-10-22 Thread Ed Maste
Author: emaste Date: Tue Oct 22 18:30:51 2019 New Revision: 353898 URL: https://svnweb.freebsd.org/changeset/base/353898 Log: blacklistd: remove reference to NetBSD npfctl Xr pfctl instead. PR: 221069 Submitted by: "fml" MFC after:3 days Modified: head/contrib/blac

svn commit: r353897 - head/sys/powerpc/ofw

2019-10-22 Thread Leandro Lupori
Author: luporl Date: Tue Oct 22 18:28:58 2019 New Revision: 353897 URL: https://svnweb.freebsd.org/changeset/base/353897 Log: [PPC] Avoid underflows in NUMA domains On POWER8 systems with only one memory domain, the "ibm,associativity" number that corresponds to it is 0, unlike POWER9 sys

Re: svn commit: r353876 - head/sys/kern

2019-10-22 Thread Li-Wen Hsu
On Wed, Oct 23, 2019 at 2:02 AM Gleb Smirnoff wrote: > > On Tue, Oct 22, 2019 at 11:40:36AM -0400, Mark Johnston wrote: > M> > I need help from RISC-V experts. > M> > M> I suspect the problem is that init_proc0() on several arches does not > M> initialize td_kstack_pages. I will test this patch a

Re: svn commit: r353640 - head/sys/kern

2019-10-22 Thread Conrad Meyer
Theoretically the information should be here, but it's still locked: https://admbugs.freebsd.org/show_bug.cgi?id=651 . On Tue, Oct 22, 2019 at 8:51 AM Alan Somers wrote: > > On Wed, Oct 16, 2019 at 7:21 AM Andrew Turner wrote: >> >> Author: andrew >> Date: Wed Oct 16 13:21:01 2019 >> New Revisio

svn commit: r353896 - head/sys/kern

2019-10-22 Thread Gleb Smirnoff
Author: glebius Date: Tue Oct 22 18:05:15 2019 New Revision: 353896 URL: https://svnweb.freebsd.org/changeset/base/353896 Log: Allow epoch tracker to use the very last byte of the stack. Not sure this will help to avoid panic in this function, since it will also use some stack, but makes co

Re: svn commit: r353876 - head/sys/kern

2019-10-22 Thread Gleb Smirnoff
On Tue, Oct 22, 2019 at 11:40:36AM -0400, Mark Johnston wrote: M> > I need help from RISC-V experts. M> M> I suspect the problem is that init_proc0() on several arches does not M> initialize td_kstack_pages. I will test this patch and commit if it M> fixes the panic. M> M> Note that on arm64 and

svn commit: r353894 - head/sys/arm/arm

2019-10-22 Thread Mark Johnston
Author: markj Date: Tue Oct 22 17:21:23 2019 New Revision: 353894 URL: https://svnweb.freebsd.org/changeset/base/353894 Log: Initialize thread0.td_kstack_pages on arm. Fix style on the line below. Reported by: Jenkins (hardware CI lab) MFC after:1 week Sponsored by: The FreeBS

svn commit: r353895 - head/sys/arm64/arm64

2019-10-22 Thread Mark Johnston
Author: markj Date: Tue Oct 22 17:22:10 2019 New Revision: 353895 URL: https://svnweb.freebsd.org/changeset/base/353895 Log: Apply r353893 to arm64. Reported by: Jenkins (hardware CI lab) MFC after:1 week Sponsored by: The FreeBSD Foundation Modified: head/sys/arm64/arm64/locore

svn commit: r353893 - head/sys/riscv/riscv

2019-10-22 Thread Mark Johnston
Author: markj Date: Tue Oct 22 16:52:56 2019 New Revision: 353893 URL: https://svnweb.freebsd.org/changeset/base/353893 Log: Fix thread0 kernel stack initialization on riscv. - td_kstack_pages was not being initialized. - td_kstack is supposed to be the base address of the stack region,

Re: svn commit: r353888 - head/sys/vm

2019-10-22 Thread Enji Cooper
> On Oct 22, 2019, at 08:59, Konstantin Belousov wrote: > > Author: kib > Date: Tue Oct 22 15:59:16 2019 > New Revision: 353888 > URL: https://svnweb.freebsd.org/changeset/base/353888 > > Log: > vm_fault(): extract code to lock the vnode into a helper > vn_fault_lock_vnode(). From an ex

Re: svn commit: r353732 - head/sys/net

2019-10-22 Thread Enji Cooper
> On Oct 22, 2019, at 08:26, Gleb Smirnoff wrote: > > On Tue, Oct 22, 2019 at 07:53:31AM -0700, Enji Cooper wrote: > E> > Author: glebius > E> > Date: Fri Oct 18 15:20:24 2019 > E> > New Revision: 353732 > E> > URL: https://svnweb.freebsd.org/changeset/base/353732 > E> > > E> > Log: > E> > M

svn commit: r353892 - in head/sys: cddl/contrib/opensolaris/uts/common/fs/zfs sys vm

2019-10-22 Thread Konstantin Belousov
Author: kib Date: Tue Oct 22 16:21:24 2019 New Revision: 353892 URL: https://svnweb.freebsd.org/changeset/base/353892 Log: Assert that vnode_pager_setsize() is called with the vnode exclusively locked except for filesystems that set the MNTK_VMSETSIZE_BUG, Set the flag for ZFS. Tested

svn commit: r353891 - in head/sys/fs: nfs nfsclient

2019-10-22 Thread Konstantin Belousov
Author: kib Date: Tue Oct 22 16:17:38 2019 New Revision: 353891 URL: https://svnweb.freebsd.org/changeset/base/353891 Log: Fix interface between nfsclient and vnode pager. Make the nfsclient always call vnode_pager_setsize() with the vnode exclusively locked. This ensures that page fault

svn commit: r353890 - in head/sys: sys vm

2019-10-22 Thread Konstantin Belousov
Author: kib Date: Tue Oct 22 16:09:25 2019 New Revision: 353890 URL: https://svnweb.freebsd.org/changeset/base/353890 Log: Add VV_VMSIZEVNLOCK flag. The flag specifies that vm_fault() handler should check the vnode' vm_object size under the vnode lock. It is converted into the object'

svn commit: r353889 - head/sys/netinet6

2019-10-22 Thread Gleb Smirnoff
Author: glebius Date: Tue Oct 22 16:06:33 2019 New Revision: 353889 URL: https://svnweb.freebsd.org/changeset/base/353889 Log: Execute nd6_dad_timer() in the network epoch, since nd6_dad_duplicated() requires it. Make nd6_dad_starttimer() require network epoch. Two calls out of three happ

svn commit: r353888 - head/sys/vm

2019-10-22 Thread Konstantin Belousov
Author: kib Date: Tue Oct 22 15:59:16 2019 New Revision: 353888 URL: https://svnweb.freebsd.org/changeset/base/353888 Log: vm_fault(): extract code to lock the vnode into a helper vn_fault_lock_vnode(). Tested by:pho Reviewed by: markj Sponsored by: The FreeBSD Foundation MFC af

Re: svn commit: r353640 - head/sys/kern

2019-10-22 Thread Alan Somers
On Wed, Oct 16, 2019 at 7:21 AM Andrew Turner wrote: > Author: andrew > Date: Wed Oct 16 13:21:01 2019 > New Revision: 353640 > URL: https://svnweb.freebsd.org/changeset/base/353640 > > Log: > Stop leaking information from the kernel through timespec > > The timespec struct holds a seconds va

Re: svn commit: r353876 - head/sys/kern

2019-10-22 Thread Mark Johnston
On Tue, Oct 22, 2019 at 08:29:13AM -0700, Gleb Smirnoff wrote: > Li-Wen, > > On Tue, Oct 22, 2019 at 04:28:45PM +0800, Li-Wen Hsu wrote: > L> > Modified: head/sys/kern/subr_epoch.c > L> > > == > L> > --- head/sys/kern/s

Re: svn commit: r353876 - head/sys/kern

2019-10-22 Thread Hans Petter Selasky
On 2019-10-22 17:29, Gleb Smirnoff wrote: L> L> panic: Assertion (vm_offset_t)et >= td->td_kstack && (vm_offset_t)et + L> sizeof(struct epoch_tracker) < td->td_kstack + td->td_kstack_pages * L> PAGE_SIZE failed at /workspace/src/sys/kern/subr_epoch.c:373 Probably doesn't matter that much, but t

Re: svn commit: r353876 - head/sys/kern

2019-10-22 Thread Gleb Smirnoff
Li-Wen, On Tue, Oct 22, 2019 at 04:28:45PM +0800, Li-Wen Hsu wrote: L> > Modified: head/sys/kern/subr_epoch.c L> > == L> > --- head/sys/kern/subr_epoch.c Mon Oct 21 21:21:34 2019(r353875) L> > +++ head/sys/kern

Re: svn commit: r353732 - head/sys/net

2019-10-22 Thread Gleb Smirnoff
On Tue, Oct 22, 2019 at 07:53:31AM -0700, Enji Cooper wrote: E> > Author: glebius E> > Date: Fri Oct 18 15:20:24 2019 E> > New Revision: 353732 E> > URL: https://svnweb.freebsd.org/changeset/base/353732 E> > E> > Log: E> > Make rt_getifa_fib() static. E> E> Why should it be made static? The ans

Re: svn commit: r353732 - head/sys/net

2019-10-22 Thread Enji Cooper
> On Oct 18, 2019, at 08:20, Gleb Smirnoff wrote: > > Author: glebius > Date: Fri Oct 18 15:20:24 2019 > New Revision: 353732 > URL: https://svnweb.freebsd.org/changeset/base/353732 > > Log: > Make rt_getifa_fib() static. Why should it be made static? The answer to this question is missing f

svn commit: r353887 - head/sys/dev/nctgpio

2019-10-22 Thread Andriy Gapon
Author: avg Date: Tue Oct 22 14:20:35 2019 New Revision: 353887 URL: https://svnweb.freebsd.org/changeset/base/353887 Log: nctgpio: improve performance (latency) of operation This change consists of two parts. First, nctgpio now supports hardware access via an I/O port window if it's

svn commit: r353886 - head/sys/vm

2019-10-22 Thread Mark Johnston
Author: markj Date: Tue Oct 22 14:20:06 2019 New Revision: 353886 URL: https://svnweb.freebsd.org/changeset/base/353886 Log: Avoid reloading bucket pointers in uma_vm_zone_stats(). The correctness of per-CPU cache accounting in that function is dependent on reading per-CPU pointers exactl

Re: svn commit: r353876 - head/sys/kern

2019-10-22 Thread Li-Wen Hsu
On Tue, Oct 22, 2019 at 7:12 AM Gleb Smirnoff wrote: > > Author: glebius > Date: Mon Oct 21 23:12:14 2019 > New Revision: 353876 > URL: https://svnweb.freebsd.org/changeset/base/353876 > > Log: > Assert that any epoch tracker belongs to the thread stack. > > Reviewed by: kib > > Modified: >