Re: iwx(4) 40MHz channel support

On Tue, Oct 12, 2021 at 02:47:54PM +0200, Stefan Sperling wrote: > > This patch adds support for 40MHz channels to iwx(4). > > Please sync your source tree before attempting to apply this patch. > I have committed some changes to this driver today which this patch > is based on. > > Works for

table-procexec for opensmtpd (another try)

Hi all, I've made a refactored version of table-procexec, hopefully with a lot less redundancy in code. This patch adds the table-procexec backend which is configured with a timeout of 500 milliseconds. Currently this is hardcoded, but that is easy enough to change and shouldnt be the holdback.

lrint(3) and llrint(3) implementation

Currently the lib/libm/msun/run-lrint_test regress fails on powerpc64 and other platforms. Our implementation came from NetBSD, but NetBSD switched to the implementation from FreeBSD some time ago. That is the same implementation that we already use for lrintl(3) and llrintl(3). Diff below

crypto dispatch error

Hi, The function crypto-dispatch() never returns an error. Make it void to avoid error handling in the callers. ok? bluhm Index: crypto/crypto.c === RCS file: /data/mirror/openbsd/cvs/src/sys/crypto/crypto.c,v retrieving revision

Re: acme-client: don't reach into X509

Theo Buehler(t...@theobuehler.org) on 2021.10.13 13:55:14 +0200: > In an upcoming libcrypto bump, we will make a few structs in libcrypto > opaque. This needs a small change in acme-client. Fetch the extension > stack using X509_get0_extensions() and iterate using the stack API. > Note that

Re: Switch to kqueue based select(2)

On 13/10/21(Wed) 11:41, Alexander Bluhm wrote: > On Sat, Oct 02, 2021 at 09:10:13AM +0200, Martin Pieuchot wrote: > > ok? > > OK bluhm@ > > > + /* Maxium number of events per iteration */ > > Maximum > > > +int > > +pselcollect(struct proc *p, struct kevent *kevp, fd_set *pobits[3],

patch: vnode lock: remove vop_generic_{,is,un}lock functions

Hi, The following diff removes vop_generic_{,un,is}lock functions. These functions are only stubs (returning 0). Replace them by using nullop function (same behaviour). There is no intented behaviour changes. While here, I reordered some vop_islocked member in structs to be next others

ipsec redundant null checks

Hi, These NULL checks are never reached. The compiler optimizes away the first one as m->m_pkthdr dereference m before the check. All callers of ipsec_common_input_cb() never call it with NULL as they dereference m before. ok? bluhm Index: netinet/ipsec_input.c

Re: isakmpd: remove ifdefs for prehistoric OPENSSL_VERSIONs

On Wed, Oct 13, 2021 at 01:46:04PM +0200, Theo Buehler wrote: > I don't think anyone will want to compile this against OpenSSL 0.9.7 or > earlier. Calling OpenSSL_add_all_algorithms() has not been necessary > for a few years, so let's remove libcrypto.c. OK bluhm@ > Index: Makefile >

acme-client: don't reach into X509

In an upcoming libcrypto bump, we will make a few structs in libcrypto opaque. This needs a small change in acme-client. Fetch the extension stack using X509_get0_extensions() and iterate using the stack API. Note that sk_*_num() returns -1 on NULL, so we won't enter the for loop and the extsz

isakmpd: remove ifdefs for prehistoric OPENSSL_VERSIONs

I don't think anyone will want to compile this against OpenSSL 0.9.7 or earlier. Calling OpenSSL_add_all_algorithms() has not been necessary for a few years, so let's remove libcrypto.c. Index: Makefile === RCS file:

Re: Switch to kqueue based select(2)

On Sat, Oct 02, 2021 at 09:10:13AM +0200, Martin Pieuchot wrote: > ok? OK bluhm@ > + /* Maxium number of events per iteration */ Maximum > +int > +pselcollect(struct proc *p, struct kevent *kevp, fd_set *pobits[3], > +int *ncollected) > +{ > +#ifdef DIAGNOSTIC > + /* Filter