Anthony Fok [Fri, 28 Dec 2018 13:18:00 +0000 (06:18 -0700)]
[PATCH] unix: fix Fstatat by using fillStat_t on linux/mips64x
The stat structure on linux/mips64x differ between C library and the kernel,
as described in the stat(2) man page.
Fstat, Lstat and Stat on linux/mips64x already converts the stat structure
using a fillStat_t function, very similar to __xstat_conv in GLIBC.
Doing the same for Fstatat before calling SYS_NEWFSTATAT fixes the
"Fstatat: returned stat does not match Stat/Lstat" error in TestFstatat.
Fixes golang/go#29401
Change-Id: I0b2a7b274acc3c7c9fc7ae2afe722dd6225da383
Reviewed-on: https://go-review.googlesource.com/c/155747 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Tobias Klauser <tobias.klauser@gmail.com>
Gbp-Pq: Name 0004-fix-Fstatat-by-using-fillStat_t-on-linux-mips64x.patch
Make builds reproducible by honoring BUILD_PATH_PREFIX_MAP
Upstream has rejected the patch in this form and promised to implement an
alternative they are happy with instead. That hasn't happened yet though.
[ Dr. Tobias Quathamer ]
* cryptobyte: fix panic due to malformed ASN.1 inputs on 32-bit archs
https://github.com/golang/go/issues/36837
CVE-2020-7919
* net/http: Expect 100-continue panics in httputil.ReverseProxy
https://github.com/golang/go/issues/34902
CVE-2020-15586
* encoding/binary: ReadUvarint and ReadVarint can read an unlimited
number of bytes from invalid inputs
https://github.com/golang/go/issues/40618
CVE-2020-16845
[ Shengjing Zhu ]
* crypto/elliptic: incorrect operations on the P-224 curve
https://github.com/golang/go/issues/43786
CVE-2021-3114
Anthony Fok [Fri, 28 Dec 2018 13:18:00 +0000 (06:18 -0700)]
unix: fix Fstatat by using fillStat_t on linux/mips64x
The stat structure on linux/mips64x differ between C library and the kernel,
as described in the stat(2) man page.
Fstat, Lstat and Stat on linux/mips64x already converts the stat structure
using a fillStat_t function, very similar to __xstat_conv in GLIBC.
Doing the same for Fstatat before calling SYS_NEWFSTATAT fixes the
"Fstatat: returned stat does not match Stat/Lstat" error in TestFstatat.
Fixes golang/go#29401
Change-Id: I0b2a7b274acc3c7c9fc7ae2afe722dd6225da383
Reviewed-on: https://go-review.googlesource.com/c/155747 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Tobias Klauser <tobias.klauser@gmail.com>
Gbp-Pq: Name 0004-fix-Fstatat-by-using-fillStat_t-on-linux-mips64x.patch
Make builds reproducible by honoring BUILD_PATH_PREFIX_MAP
Upstream has rejected the patch in this form and promised to implement an
alternative they are happy with instead. That hasn't happened yet though.
Anthony Fok [Fri, 28 Dec 2018 13:18:00 +0000 (06:18 -0700)]
unix: fix Fstatat by using fillStat_t on linux/mips64x
The stat structure on linux/mips64x differ between C library and the kernel,
as described in the stat(2) man page.
Fstat, Lstat and Stat on linux/mips64x already converts the stat structure
using a fillStat_t function, very similar to __xstat_conv in GLIBC.
Doing the same for Fstatat before calling SYS_NEWFSTATAT fixes the
"Fstatat: returned stat does not match Stat/Lstat" error in TestFstatat.
Fixes golang/go#29401
Change-Id: I0b2a7b274acc3c7c9fc7ae2afe722dd6225da383
Reviewed-on: https://go-review.googlesource.com/c/155747 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Tobias Klauser <tobias.klauser@gmail.com>
Gbp-Pq: Name 0004-fix-Fstatat-by-using-fillStat_t-on-linux-mips64x.patch
Make builds reproducible by honoring BUILD_PATH_PREFIX_MAP
Upstream has rejected the patch in this form and promised to implement an
alternative they are happy with instead. That hasn't happened yet though.
* Fix normalization of headers with spaces before the colon.
https://github.com/golang/go/issues/34541
CVE-2019-16276
* Refresh patches
* Adapt gbp.conf for security branch
Anthony Fok [Fri, 28 Dec 2018 13:18:00 +0000 (06:18 -0700)]
unix: fix Fstatat by using fillStat_t on linux/mips64x
The stat structure on linux/mips64x differ between C library and the kernel,
as described in the stat(2) man page.
Fstat, Lstat and Stat on linux/mips64x already converts the stat structure
using a fillStat_t function, very similar to __xstat_conv in GLIBC.
Doing the same for Fstatat before calling SYS_NEWFSTATAT fixes the
"Fstatat: returned stat does not match Stat/Lstat" error in TestFstatat.
Fixes golang/go#29401
Change-Id: I0b2a7b274acc3c7c9fc7ae2afe722dd6225da383
Reviewed-on: https://go-review.googlesource.com/c/155747 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Tobias Klauser <tobias.klauser@gmail.com>
Gbp-Pq: Name 0004-fix-Fstatat-by-using-fillStat_t-on-linux-mips64x.patch
Make builds reproducible by honoring BUILD_PATH_PREFIX_MAP
Upstream has rejected the patch in this form and promised to implement an
alternative they are happy with instead. That hasn't happened yet though.
* Fix Denial of Service vulnerabilities in the HTTP/2 implementation.
https://github.com/golang/go/issues/33631
CVE-2019-9512, CVE-2019-9514
* Fix multiple Parsing Issues in URL.Parse
https://github.com/golang/go/issues/29098
CVE-2019-14809
Anthony Fok [Fri, 28 Dec 2018 13:18:00 +0000 (06:18 -0700)]
unix: fix Fstatat by using fillStat_t on linux/mips64x
The stat structure on linux/mips64x differ between C library and the kernel,
as described in the stat(2) man page.
Fstat, Lstat and Stat on linux/mips64x already converts the stat structure
using a fillStat_t function, very similar to __xstat_conv in GLIBC.
Doing the same for Fstatat before calling SYS_NEWFSTATAT fixes the
"Fstatat: returned stat does not match Stat/Lstat" error in TestFstatat.
Fixes golang/go#29401
Change-Id: I0b2a7b274acc3c7c9fc7ae2afe722dd6225da383
Reviewed-on: https://go-review.googlesource.com/c/155747 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Tobias Klauser <tobias.klauser@gmail.com>
Gbp-Pq: Name 0004-fix-Fstatat-by-using-fillStat_t-on-linux-mips64x.patch
Make builds reproducible by honoring BUILD_PATH_PREFIX_MAP
Upstream has rejected the patch in this form and promised to implement an
alternative they are happy with instead. That hasn't happened yet though.
This rule is problematic. 1<<(32-uint32(d)) <= int32(c) meant to
say that it is true if c is greater than the largest possible
value of the right shift. But when d==1, 1<<(32-1) is negative
and results in the wrong comparison.
Anthony Fok [Fri, 28 Dec 2018 13:18:00 +0000 (06:18 -0700)]
unix: fix Fstatat by using fillStat_t on linux/mips64x
The stat structure on linux/mips64x differ between C library and the kernel,
as described in the stat(2) man page.
Fstat, Lstat and Stat on linux/mips64x already converts the stat structure
using a fillStat_t function, very similar to __xstat_conv in GLIBC.
Doing the same for Fstatat before calling SYS_NEWFSTATAT fixes the
"Fstatat: returned stat does not match Stat/Lstat" error in TestFstatat.
Fixes golang/go#29401
Change-Id: I0b2a7b274acc3c7c9fc7ae2afe722dd6225da383
Reviewed-on: https://go-review.googlesource.com/c/155747 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Tobias Klauser <tobias.klauser@gmail.com>
Gbp-Pq: Name 0004-fix-Fstatat-by-using-fillStat_t-on-linux-mips64x.patch
Make builds reproducible by honoring BUILD_PATH_PREFIX_MAP
Upstream has rejected the patch in this form and promised to implement an
alternative they are happy with instead. That hasn't happened yet though.
This rule is problematic. 1<<(32-uint32(d)) <= int32(c) meant to
say that it is true if c is greater than the largest possible
value of the right shift. But when d==1, 1<<(32-1) is negative
and results in the wrong comparison.
Anthony Fok [Fri, 28 Dec 2018 13:18:00 +0000 (06:18 -0700)]
unix: fix Fstatat by using fillStat_t on linux/mips64x
The stat structure on linux/mips64x differ between C library and the kernel,
as described in the stat(2) man page.
Fstat, Lstat and Stat on linux/mips64x already converts the stat structure
using a fillStat_t function, very similar to __xstat_conv in GLIBC.
Doing the same for Fstatat before calling SYS_NEWFSTATAT fixes the
"Fstatat: returned stat does not match Stat/Lstat" error in TestFstatat.
Fixes golang/go#29401
Change-Id: I0b2a7b274acc3c7c9fc7ae2afe722dd6225da383
Reviewed-on: https://go-review.googlesource.com/c/155747 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Tobias Klauser <tobias.klauser@gmail.com>
Gbp-Pq: Name 0004-fix-Fstatat-by-using-fillStat_t-on-linux-mips64x.patch
Make builds reproducible by honoring BUILD_PATH_PREFIX_MAP
Upstream has rejected the patch in this form and promised to implement an
alternative they are happy with instead. That hasn't happened yet though.
Anthony Fok [Fri, 28 Dec 2018 18:39:40 +0000 (18:39 +0000)]
golang-1.11 (1.11.4-4) unstable; urgency=medium
* Team upload.
* Add patch "unix: fix Fstatat by using fillStat_t on linux/mips64x"
This fixes the "Fstatat: returned stat does not match Stat/Lstat"
errors detected by TestFstatat.
See https://go-review.googlesource.com/c/sys/+/155747
* Re-enable TestFstatat on mips64el
cmd/internal/objfile: only consider executable segments for load address
Reportedly on some new Fedora systems the linker is producing extra
load segments, basically making the dynamic section non-executable.
We were assuming that the first load segment could be used to
determine the program's load offset, but that is no longer true.
Use the first executable load segment instead.
Fixes #26369
Change-Id: I5ee31ddeef2e8caeed3112edc5149065a6448456
Reviewed-on: https://go-review.googlesource.com/127895
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Gbp-Pq: Name 0002-cmd-internal-objfile-only-consider-executable-segmen.patch
Make builds reproducible by honoring BUILD_PATH_PREFIX_MAP
Upstream has rejected the patch in this form and promised to implement an
alternative they are happy with instead. That hasn't happened yet though.
[ Dr. Tobias Quathamer ]
* Switch team address to tracker.d.o
* Do not compress favicon.ico.
Thanks to Dato Simó <dato@debian.org> (Closes: #917132)
* Update patches for git-buildpackage
[ Anthony Fok ]
* Bump Standards-Version to 4.3.0 (no change)
cmd/internal/objfile: only consider executable segments for load address
Reportedly on some new Fedora systems the linker is producing extra
load segments, basically making the dynamic section non-executable.
We were assuming that the first load segment could be used to
determine the program's load offset, but that is no longer true.
Use the first executable load segment instead.
Fixes #26369
Change-Id: I5ee31ddeef2e8caeed3112edc5149065a6448456
Reviewed-on: https://go-review.googlesource.com/127895
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Gbp-Pq: Name only-consider-executable-segments.patch
* New upstream version 1.10.7
* Override two false positive Lintian errors (missing depends
on sensible-utils)
* Add lintian overrides for testdata
* Include /usr/share/dpkg/architecture.mk for DEB_HOST_ARCH
* Fix Lintian warnings about wrong interpreter path
* Make two scripts executable which have been missed by upstream
* Update d/copyright
* Remove unused lintian overrides
cmd/internal/objfile: only consider executable segments for load address
Reportedly on some new Fedora systems the linker is producing extra
load segments, basically making the dynamic section non-executable.
We were assuming that the first load segment could be used to
determine the program's load offset, but that is no longer true.
Use the first executable load segment instead.
Fixes #26369
Change-Id: I5ee31ddeef2e8caeed3112edc5149065a6448456
Reviewed-on: https://go-review.googlesource.com/127895
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Gbp-Pq: Name only-consider-executable-segments.patch