wireguard-openbsd/libexec/ld.so/boot.c, branch jd/histogram WireGuard implementation for the OpenBSD kernel https://git.zx2c4.com/wireguard-openbsd/atom/libexec/ld.so/boot.c?h=jd%2Fhistogram 2019-10-20T03:44:49Z For more archs, ld.so itself only needs/uses the arch's "just add load offset" 2019-10-20T03:44:49Z guenther guenther@openbsd.org 2019-10-20T03:44:49Z urn:sha1:17fa88719dad59793b58356460b07659ff5d59db 'relative' relocation. Take advantage of that to simplify ld.so's self-reloc code: * give the exceptional archs (hppa and mips64) copies of the current boot.c as boot_md.c * teach the Makefile to use boot_md.c when present * reduce boot.c down to the minimum necessary to handle just relative reloc * teach the Makefile to fail if the built ld.so has other types of relocs ok visa@ kettenis@ ld.so boot cleanup support: 2019-05-10T13:29:21Z guenther guenther@openbsd.org 2019-05-10T13:29:21Z urn:sha1:c0197e40dd4094880fe205f9c12d3e12e14e7ff3 - put functions and data which are only used before calling the executable's start function into their own page-aligned segments for unmapping (only done on amd64, arm64, armv7, powerpc, and sparc64 so far) - pass .init_array and .preinit_array functions an addition argument which is a callback to get a structure which includes a function that frees the boot text and data - sometimes delay doing RELRO processing: for a shared-object marked DF_1_INITFIRST do it after the object's .init_array, for the executable do it after the .preinit_array - improve test-ld.so to link against libpthread and trigger its initialization late libc changes to use this will come later ok kettenis@ Delay processing of ld.so's own PT_GNU_RELRO section until after environment 2018-10-23T04:01:45Z guenther guenther@openbsd.org 2018-10-23T04:01:45Z urn:sha1:d1b0fb8fb3a964323473438f43f47e69e73ece83 processing and malloc initialization, and then move variables set by those steps, including malloc's mopts structure, into the .data.rel.ro segment. This protects more data and eliminates the extra padding of the mopts. ok kettenis@ Remove dependency on __got_{start,end} symbols and look at PT_GNU_RELRO 2017-01-09T22:51:04Z kettenis kettenis@openbsd.org 2017-01-09T22:51:04Z urn:sha1:41acbfd8967a34b0843c570a9cbae8efcded4ac4 instead. Result in a few more pages that aren't writable on some platforms (such as hppa). Based on an initial diff from guenther@. Thanks to deraadt@ for testing. ok guenther@ Passing in &_DYNAMIC appears to have stuck, so eliminate the extra 2016-08-13T20:57:04Z guenther guenther@openbsd.org 2016-08-13T20:57:04Z urn:sha1:edc0f98f02fdbbee07f8fc79779d22ede8ba999b variable Teach arm to pass &_DYNAMIC to _dl_boot_bind(); can optimize later in tree 2016-08-09T03:58:35Z guenther guenther@openbsd.org 2016-08-09T03:58:35Z urn:sha1:45f3b7e734fc155ce84bb13c5260e10022f5958f As with csu, alpha passes &_DYNAMIC to _reloc_alpha_got(), so just 2016-08-07T03:05:23Z guenther guenther@openbsd.org 2016-08-07T03:05:23Z urn:sha1:fba5f8d8c167217de8c8420e85acfbcbed38e7cc save that and pass it to _dl_boot_bind() too Teach i386 to pass &_DYNAMIC to _dl_boot_bind() 2016-08-07T03:03:44Z guenther guenther@openbsd.org 2016-08-07T03:03:44Z urn:sha1:94bdeb90245ced851e5d0654af0e1e907a9b347f Teach mips64 to pass &_DYNAMIC to _dl_boot_bind() 2016-08-07T03:01:53Z guenther guenther@openbsd.org 2016-08-07T03:01:53Z urn:sha1:f26ff3c14ed07d77734e7801666e90b95d0ebdd5 mips64be testing by deraadt@ hahahah: amd64 has been passing &_DYNAMIC to _dl_boot_bind() for 2016-08-07T02:59:27Z guenther guenther@openbsd.org 2016-08-07T02:59:27Z urn:sha1:3714ee24c60779a59832c599d4f35df148d4b7c6 *149 months*