Re: dsdt: error while compiling with ACPI_MEMDEBUG

On Fri, 14 Apr 2017, Anton Lindqvist wrote: > While compiling the GENERIC.MP kernel on amd64 with ACPI_MEMDEBUG > defined, I get the following error: > > dsdt.c:411: warning: format '%x' expects type 'unsigned int', but > argument 4 has type 'long int' > > Adding the 'l' modifier fixes the erro

Re: sparc64 ld fix

On Fri, Apr 14, 2017 at 6:22 AM, Mark Kettenis wrote: > Turns out that clang generates R_SPARC_WPLT30 relocations against > local symbols, which our binutils doesn't handle. Since the Solaris > native assembler produces such relocations (and expects them to be > handled as R_SPARC_WDISP30) and ne

iked remove SA bundles

Hi, iked does not use SA bundles. Does anyone plan to implement this? ok to remove the dead code? bluhm Index: sbin/iked/iked.h === RCS file: /data/mirror/openbsd/cvs/src/sbin/iked/iked.h,v retrieving revision 1.114 diff -u -p -r1.

ipsecctl rename SA bundles

Hi, Rename all SA groups to bundles consistently. The first kernel commit in 2000 that introduced the features already called them SA bundles. The word group is taken for Diffie-Hellman. ok? bluhm Index: sbin/ipsecctl/ipsecctl.c

Re: vmd: VMs with auto-configured L3 interfaces (call for testing)

On Wed, Apr 12, 2017 at 01:44:48PM +0200, Reyk Floeter wrote: > Hi, > > we want to make it easier to run NAT'ed and auto-configured VMs that > don't need switches, L2, manual scripts, or any additional servers on > the host. I wrote a new mode that just needs vmd, pf, and forwarding. Tested on a

Re: vmd: VMs with auto-configured L3 interfaces (call for testing)

On Fri, Apr 14, 2017 at 12:29:53PM -0400, Josh Grosse wrote: > Tested on an amd64 host... Following up on my own post to state this was tested both with "local" in vm.conf as well as with vmctl -L.

sparc64 ld fix

Turns out that clang generates R_SPARC_WPLT30 relocations against local symbols, which our binutils doesn't handle. Since the Solaris native assembler produces such relocations (and expects them to be handled as R_SPARC_WDISP30) and newer binutils handles them as well, I think we should as well.

Re: arm64 pmap tlb flushing diff

On Fri, Apr 14, 2017 at 11:51:52AM +0200, Mark Kettenis wrote: > While reading the arm64 code the ranges covered by the tlb flushes in > the code that sets the l1/l2/l3 tables made no sense to me. I came to > the conclusion that these flushes aren't necessary. When we enter the > final page table

arm64 pmap tlb flushing diff

While reading the arm64 code the ranges covered by the tlb flushes in the code that sets the l1/l2/l3 tables made no sense to me. I came to the conclusion that these flushes aren't necessary. When we enter the final page table entry, we flush the tlb again which flushes any cached information for

vmm/amd64 __func__

Hi, This patch is a continuation of https://marc.info/?l=openbsd-tech&m=149214401819485&w=2 * Make __func__ usage more consistent in vmm/amd64 * Typo cant->can't in the following error messages... vcpu_run_vmx: cant read exit reason vmm_get_exit_qualification: cant extract exit qual * O

nsd 4.1.16

rc1 is out, running fine here with minimal-responses: yes for some time tests/OKs? diff --git configlexer.lex configlexer.lex index 42dbd152f16..547e7db3e02 100644 --- configlexer.lex +++ configlexer.lex @@ -271,6 +271,7 @@ zonefiles-check{COLON} { LEXOUT(("v(%s) ", yytext)); return VAR_ZON

dsdt: error while compiling with ACPI_MEMDEBUG

Hi, While compiling the GENERIC.MP kernel on amd64 with ACPI_MEMDEBUG defined, I get the following error: dsdt.c:411: warning: format '%x' expects type 'unsigned int', but argument 4 has type 'long int' Adding the 'l' modifier fixes the error. Index: dsdt.c ==