WireGuard implementation for the OpenBSD kernel https://git.zx2c4.com/wireguard-openbsd/atom/lib/libc/arch?h=jd%2Fqueueboosts 2020-03-13T09:31:24Z 2020-03-13T09:31:24Z deraadt deraadt@openbsd.org 2020-03-13T09:31:24Z urn:sha1:f74a2b77455090a00a6234c40c5f02aa52986a5b problems as 64-bit models. To resolve the syscall speculation, as a first step "nop; nop" was added after all occurances of the syscall ("swi 0") instruction. Then the kernel was changed to jump over the 2 extra instructions. In this final step, those pair of nops are converted into the speculation-blocking sequence ("dsb nsh; isb"). Don't try to build through these multiple steps, use a snapshot instead. Packages matching the new ABI will be out in a while... ok kettenis 2020-03-11T12:53:19Z deraadt deraadt@openbsd.org 2020-03-11T12:53:19Z urn:sha1:f377c7672bd38a70b428cd7751ceff0af609e750 problems as 64-bit models. For the syscall instruction issue, add nop;nop after swi 0, in preparation for jumping over a speculation barrier here later. ok kettenis 2020-02-18T12:19:11Z kettenis kettenis@openbsd.org 2020-02-18T12:19:11Z urn:sha1:5ecc9681133f1894e81c385eddd09981f78f1193 a syscall, replace the double nop with a dsb nsh; isb; sequence which stops the CPU from speculating any further. This fix was suggested by Anthony Steinhauser. ok deraadt@ 2020-01-26T02:19:43Z kettenis kettenis@openbsd.org 2020-01-26T02:19:43Z urn:sha1:0c401ffc2a2550c32105ce180746b0a95071123b The will be replaced by a speculation barrier as soon as we teach the kernel to skip over these two instructions when returning from a system call. ok patrick@, deraadt@ 2019-11-10T22:23:28Z guenther guenther@openbsd.org 2019-11-10T22:23:28Z urn:sha1:00b6e344ba691d3d76ced8394494ab6d26b6766c as well as those in arch/arm/gen/divsi3.S. This cleans up the PLTs on the 32bit archs. luna88k testing by aoyama@ "looks good" kettenis@, testing and ok deraadt@ 2019-10-27T21:07:13Z guenther guenther@openbsd.org 2019-10-27T21:07:13Z urn:sha1:c2f6c1bf3d56ccc4b69356e111e3b83c014e2fe7 PLT entries and prevent overriding ok kettenis@ deraadt@ 2019-10-26T18:47:10Z guenther guenther@openbsd.org 2019-10-26T18:47:10Z urn:sha1:589b4acdbba65d6ecc6e77d9aac5e27618bc75fb by the ASM stub, which is also in libc. The compiler only generates invocations of the latter. ok mpi@ deraadt@ kettenis@ 2019-10-23T19:33:41Z kettenis kettenis@openbsd.org 2019-10-23T19:33:41Z urn:sha1:a0cc66ea236f244ba12da80a5560f08fb0318b57 ok deraadt@ 2019-07-14T03:36:37Z guenther guenther@openbsd.org 2019-07-14T03:36:37Z urn:sha1:b21062f52ee8bbb169b189af297cd24a86cef8d8 Tweak the PSEUDO() macro to avoid that. problem noted and tested by kettenis@ 2019-06-21T17:00:58Z jca jca@openbsd.org 2019-06-21T17:00:58Z urn:sha1:dcaa94fa8333d546e96b49f352c2dd028e96fd9d Bug exposed by erratic sqlite3 behavior used in ports/devel/proj, as pointed out by landry@. Richard Hipps (SQLite) pointed at the culprit (_Qp_div), many thanks. Adapted from FreeBSD revision 146673 by Stephen Paskaluk and stefanf@FreeBSD. FreeBSD commit message: """ Fix long (and long long) to long double, unsigned to long double and unsigned long (and unsigned long long) to long double conversions. - Add a parameter that specifies the position of the sign bit to the _QP_TTOQ macro, previously it always looked at bit 31. Pass a negative number to disable sign inspection for unsigned types. This fixes _Qp_xtoq(), _Qp_uitoq() and _Qp_uxtoq(). - In the functions __fpu_itof() and __fpu_xtof(), look at the sign bit to decide whether we're doing a conversion from an unsigned type. If so, don't negate the mantissa if the integer exceeds the biggest signed number. """ ok deraadt@