Date: Sat, 11 Jul 2015 10:26:37 -0600 From: Warner Losh <imp@bsdimp.com> To: Craig Rodrigues <rodrigc@crodrigues.org> Cc: John-Mark Gurney <jmg@funkthat.com>, "src-committers@freebsd.org" <src-committers@freebsd.org>, "svn-src-all@freebsd.org" <svn-src-all@freebsd.org>, "svn-src-head@freebsd.org" <svn-src-head@freebsd.org> Subject: Re: svn commit: r285366 - in head: lib/libmd sys/crypto/sha2 Message-ID: <34F75D0D-33F1-4D4F-81E5-A1DA7664868C@bsdimp.com> In-Reply-To: <CAG=rPVcva-GOLsSbbVOPWssP1SB=hAPwyib73ARUwED0skDX5g@mail.gmail.com> References: <201507110312.t6B3CZBv081280@repo.freebsd.org> <20150711080439.GD8523@funkthat.com> <CAG=rPVcCSzHKwqyOzAZOG=N9pBMpwFbY%2BCMOM655BanjnNVQ-w@mail.gmail.com> <20150711142920.GE8523@funkthat.com> <CAG=rPVcva-GOLsSbbVOPWssP1SB=hAPwyib73ARUwED0skDX5g@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
[-- Attachment #1 --] > On Jul 11, 2015, at 8:33 AM, Craig Rodrigues <rodrigc@crodrigues.org> wrote: > > > > On Sat, Jul 11, 2015 at 10:29 AM, John-Mark Gurney <jmg@funkthat.com> wrote: > Craig Rodrigues wrote this message on Sat, Jul 11, 2015 at 10:16 -0400: > > > So it looks like _libmd_SHA256_File is not defined in the legacy > > library used during the bootstrap. > > The closest I can see is that sha256hl.c (from mdXhl.c) is including > "sha256.h" with double quotes, and getting the wrong header file.. > > > Yes, I think you are right. > > If I go to src/lib/libmd, and type: > > make -V CFLAGS > > I see: > > -O2 -pipe -I/builds/FreeBSD_HEAD/lib/libmd -DWEAK_REFS -std=gnu99 -Qunused-arguments -fstack-protector -Wno-pointer-sign -Wno-empty-body -Wno-string-plus-int -Wno-unused-const-variable -Wno-tautological-compare -Wno-unused-value -Wno-parentheses-equality -Wno-unused-function -Wno-enum-conversion -Wno-switch -Wno-switch-enum -Wno-knr-promoted-parameter -Wno-parentheses > > After your change, there is no sha256.h file in the src/lib/libmd directory, so the bootstrap stage > is picking up sha256.h from /usr/include , which causes problems. You’d be better off retaining sha256.h as an empty file, or adding it to the legacy includes. But for some reason those aren’t appearing on this command line (suggesting this is built after we’re done with -legacy). If so, an empty file would likely cover many of the sins here. I think you can add a CFLAGS.sha256hi.c=-E to get the intermediate output of the pre-processor to see exactly where this file is coming from. Or you can use the .depend file from the build to see where it was picked up from (which might be easier). Unless you can resolve this quickly, I’d go with the blackout. Warner [-- Attachment #2 --] -----BEGIN PGP SIGNATURE----- Comment: GPGTools - https://gpgtools.org iQIcBAEBCgAGBQJVoUO9AAoJEGwc0Sh9sBEAlz8QAKiRcCiMe6yGhb/2ozXEqnJo zYV3LdrxYWAmxkO/pCjobWUzcaWQRSR9snqZTTIDbO/ACpgnXqKhOpS3OZzfScyZ 6HwQATjoGU+3zAej6ZlOO4ULkdKcM90oFWVyvEfk5mmpUbIn3O6eOEYUJk4LfvsP N8r+3mGykhZ5HRa/M2+l5ixsXWmDWGXWZxU5NHNuPSayho3B0ySvMZDfw9xblSku NL1UbBmjF0/PzreWdQjEf1u8BclO+YwUZx8mc2veHHnvqkwV9F9iu2+ndQeeqKDe JUWMSiTAO9iWNy4UI7iu/KK2dLNsa/e+Pz1Gs+dWTNWbj8EHedi3Cu46WEFbwbcc hvUiE4nmsCW3PHp0/8asUCSg4o64IFhuPAYbglY7gwdVTj7ztMFraACRY6AN1KgP Al8YFcNdOYWBOojbeFp02kPqQZGCQy5krzurqLhojU8IMETWNnwWZv6ynVg38Mkk YPgImCnoUcbJ3WlNCq6cO0dyC+SPHgU2J8JHrQVunTU+pfI2KBHBceSSJpL3lEf8 XTYRKL15m9lAPXkoDNMia7NLFlymsm6orWF+c3+MDPxEOYBLrLTgRUiJLjGiVhiz hvoeFmPKD3Q4y2xG3JgLc5UkMm71045Qi8Jlb4VtAdKJgvkBZsK9CnodLFMjhoAA YZE/JL7GO/AIsNxMAPNc =jgZg -----END PGP SIGNATURE-----
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?34F75D0D-33F1-4D4F-81E5-A1DA7664868C>