Discussion:
code review 6499087: go103 backport (issue 6499087)
(too old to reply)
adg-iFWiy5xATs8dnm+
2012-09-11 07:33:00 UTC
Permalink
Reviewers: golang-dev_googlegroups.com,

Message:
Hello golang-dev-/***@public.gmane.org,

I'd like you to review this change to
https://go.googlecode.com/hg


Description:
go103 backport

Please review this at http://codereview.appspot.com/6499087/

Affected files:
A misc/backport/backport
A misc/backport/go103
A misc/backport/go103-excluded
A misc/backport/go103-manual
A misc/backport/graftedit
A misc/backport/patches/0-text-template-parse-lex-test.diff
A
misc/backport/patches/1-d5754b3d9f44-encoding-gob-fix-check-for-short-input.diff
A misc/backport/patches/2-2307a931664e-net-http-dont-allow-zero-byte.diff
A misc/backport/patches/3-c552fb2b6a6c-net-http-send-explicit-zero.diff
A
misc/backport/patches/4-bb2316abeea5-cmd-ld-handle-scattered-relocation.diff
A
misc/backport/patches/5-5c4859bc123f-runtime-increase-stack-during-cgo-win-amd64.diff
Andrew Gerrand
2012-09-11 07:41:29 UTC
Permalink
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=google.com; s=20120113;
h=x-beenthere:received-spf:mime-version:sender:in-reply-to:references
:from:date:message-id:subject:to:cc:x-system-of-record
:x-gm-message-state:x-original-sender
:x-original-authentication-results:precedence:mailing-list:list-id
:x-google-group-id:list-post:list-help:list-archive:list-subscribe
:list-unsubscribe:content-type;
bh=y4RoDxxZE9j/7sgxvw+rTvCBm+kx8pQX6vAzLQq5irQ=;
b=aQytuGQni2ZRE3UkRwasQTT5Zl8kSkrF+WZbT9RY3PN+m+HtyZ+GZ2FfN0xUvI89yl
kpZLw16fkgFtJGkiC+kcvrLoCoiYCxjr1f/P7h9/oMoBHF+QuAlms+sPkqz7J6eT30Pu
KeCnFwuTxGfVSS3qcsYhl0l3AsFaJdzZOHpeiDVlqnzbjtw2jle76EZeUi02EJB6Mm7T
BaEWpIxvrqeQCx5JYALylCyrvKEAeteUVzCNSyMbE65igTgsC2H1ImnNj4rzzVM4es5C
mCpb6YDvtlz3s6GIt6v1qMLMwzCHpA4YTuNUlYAmi0JY
Received: by 10.224.76.202 with SMTP id d10mr2891936qak.0.1347349322353;
Tue, 11 Sep 2012 00:42:02 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.224.101.68 with SMTP id b4ls290246qao.0.gmail; Tue, 11 Sep
2012 00:42:01 -0700 (PDT)
Received: by 10.224.223.84 with SMTP id ij20mr7173038qab.5.1347349321779;
Tue, 11 Sep 2012 00:42:01 -0700 (PDT)
Received: by 10.224.223.84 with SMTP id ij20mr7173037qab.5.1347349321763;
Tue, 11 Sep 2012 00:42:01 -0700 (PDT)
Received: from mail-qc0-f175.google.com (mail-qc0-f175.google.com [209.85.216.175])
by gmr-mx.google.com with ESMTPS id t29si1123630qcz.1.2012.09.11.00.42.00
(version=TLSv1/SSLv3 cipher=OTHER);
Tue, 11 Sep 2012 00:42:00 -0700 (PDT)
Received-SPF: pass (google.com: domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.175 as permitted sender) client-ip=209.85.216.175;
Received: by qcad10 with SMTP id d10so107865qca.6
for <golang-dev-/***@public.gmane.org>; Tue, 11 Sep 2012 00:42:00 -0700 (PDT)
Received: by 10.224.31.202 with SMTP id z10mr17753462qac.86.1347349320656;
Tue, 11 Sep 2012 00:42:00 -0700 (PDT)
Received: by 10.224.31.202 with SMTP id z10mr17753457qac.86.1347349320563;
Tue, 11 Sep 2012 00:42:00 -0700 (PDT)
Sender: golang-dev-/***@public.gmane.org
Received: by 10.224.33.202 with HTTP; Tue, 11 Sep 2012 00:41:29 -0700 (PDT)
In-Reply-To: <047d7bd6ac9274f59f04c9681528-hpIqsD4AKlfQT0dZR+***@public.gmane.org>
X-System-Of-Record: true
X-Gm-Message-State: ALoCoQm0LXYhrMXdTIEcvxgr8645MFiy23K+Br/xOTcUXhEy7NOESB5NDdVV1PjxesJ7hngpWntLKk2JK63AJquQpX6duzXC0HPJZnnsIkOfDPyCcOY4UXuFZpkDuOzV1uwdVwKNjZpyaNGrWtJwBPgcasO4CRdvdfdVbT9S8wkzCUX8maNG6qpEZ3iKHzOXNbtw9ZSn1drYdIKlMuonwWmn/5qmKtRmPQ==
X-Original-Sender: adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.175 as permitted sender)
smtp.mail=adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org; dkim=pass header.i=@google.com
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45196>

This is a work-in-progress of the Go 1.0.3 release.

go103 lists the included changes
go103-manual lists the changes manually applied by the patches in the
patches directory (to avoid merge conflicts)
go103-excluded lists the changes that are not included for a variety of reasons

In each of those files, there are four tab-separated columns:

62f087306b18 10 2: +7/-2 text/template/parse: fix bug handling /*/

The hash, the priority, files: lines changed, and the change summary.
Priority >0 means included. 0 means unsure. <0 means excluded.
Changes with priority <-1 were originally supposed to be included, but
were taken out because they caused merge conflicts and didn't seem
worth the pain of doing them manually.

You can import these into Google Docs as a spreadsheet for easier viewing.

This CL in its present state produces a tree that builds and passes
tests on darwin/amd64. I have pushed a clone of the repo (as produced
by the 'backport' script) to this repo:
https://code.google.com/p/go103-test

I would like feedback from golang-dev about the choice of changes
(particularly those at priority 0), and whether you encounter any
issues building or running your programs under the go103-test tree.

Andrew
Andrew Gerrand
2012-09-11 10:03:35 UTC
Permalink
Looks like I neglected to do an "hg add" of the new cgo test files, so
that's why misc/cgo/test fails. Will fix.
d***@public.gmane.org
2012-09-11 10:42:54 UTC
Permalink
From a quick look over go103-excluded, I'd like to throw in my 2 cents
as to which changes I'd like included.


http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded
File misc/backport/go103-excluded (right):

http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode74
misc/backport/go103-excluded:74: 98788e8973f2 -1 7: +500/-500 cmd/gc:
point "no new variables" error at right line number.
This change is actually fairly minor and fixes an issue where the
compiler would confuse the user.

The huge diffstat is because of regenerating the bison files.

http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode98
misc/backport/go103-excluded:98: 272e1dd72156 -1 2: +29/-0 cmd/gc: fix
PkgPath of byte, rune types
This change fixes a regression introduced in go1.0.1 that might break
compatibility for users of reflect.

http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode118
misc/backport/go103-excluded:118: 6429097c1c0f -1 2: +5/-31 cmd/gc: mark
broken type declarations as broken.
This removes a error where the compiler would print out a wrong error
and confuse the user.

Big diffstat on this is because it accidentally clobbered another test
file during submit. If this patch is included, include 391791277b74 as
well

http://codereview.appspot.com/6499087/
dsymonds-iFWiy5xATs8dnm+
2012-09-11 11:45:05 UTC
Permalink
https://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded
File misc/backport/go103-excluded (right):

https://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode98
misc/backport/go103-excluded:98: 272e1dd72156 -1 2: +29/-0 cmd/gc: fix
PkgPath of byte, rune types
Post by d***@public.gmane.org
This change fixes a regression introduced in go1.0.1 that might break
compatibility for users of reflect.
I believe it was actually a regression introduced in 1.0.2, but I agree
it would be good to include this.

https://codereview.appspot.com/6499087/
iant-iFWiy5xATs8dnm+
2012-09-11 13:17:41 UTC
Permalink
https://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded
File misc/backport/go103-excluded (right):

https://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode100
misc/backport/go103-excluded:100: 33a62d86ca2f -1 2: +19/-7 cmd/gc:
Suggest *T in error for x.(T) if it would work.
This change (33a62d86ca2f) is self-contained and I think it can be
helpful for new Go programmers. Not essential but might be nice.

https://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode101
misc/backport/go103-excluded:101: 3532ab6bf5be -1 4: +36/-21 runtime:
use sched_getaffinity for runtime.NumCPU() on Linux
I am mildly in favor of this change (3532ab6bf5be). It's more accurate
and more efficient.

https://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode115
misc/backport/go103-excluded:115: 56ea40aac72b -1 4: +330/-2 runtime:
add vdso support for linux/amd64. Fixes issue 1933.
This change (56ea40aac72b) should be included in 1.0.3.

https://codereview.appspot.com/6499087/
r***@public.gmane.org
2012-09-11 20:59:58 UTC
Permalink
I am additionnally thinking we should include the following changes that
predate Go 1.0.2, plus the compiler bug below which is excluded for some
reason.

* Issue 3180: annoyances with GOROOT_FINAL (make debugging
uncomfortable)
a608a03df703 5a, 6a, 8a: take GOROOT_FINAL into consideration
016c4fefed77 5c, 6c, 8c: take GOROOT_FINAL into consideration
ba91b34fe212 cmd/gc: if $GOROOT_FINAL is set, rewrite file names in
object files

* Issue 3918: serious compiler corruptions on 64 bits
72678c75b2d3 cmd/6c: fix probable code gen bug
a9e09f1fe373 6c, 6g, 6l: add MOVQL to make truncation explicit

* Issue 3824: code semantics violated by compiler
80c98738fdde cmd/gc: fix initialization order involving method calls.


http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded
File misc/backport/go103-excluded (right):

http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode230
misc/backport/go103-excluded:230: 238a5fd52836 -10 3: +35/-6 cmd/gc: fix
escape analysis bug with variable capture in loops.
This change is a one-line fix, is it really not mergeable?

It causes potential corruption in programs and seems very important to
me.

http://codereview.appspot.com/6499087/
Andrew Gerrand
2012-09-12 00:46:48 UTC
Permalink
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=google.com; s=20120113;
h=x-beenthere:received-spf:mime-version:sender:in-reply-to:references
:from:date:message-id:subject:to:cc:x-system-of-record
:x-gm-message-state:x-original-sender
:x-original-authentication-results:precedence:mailing-list:list-id
:x-google-group-id:list-post:list-help:list-archive:list-subscribe
:list-unsubscribe:content-type;
bh=fyVm678Lh3Li1hy23PjzKLLyWKWGoIdobJUvnC4J0Q8=;
b=fPtJXvZq/AR5IKmspEQR7147L80ooCDSmYgXvrWOu3THowDqYyyIo4WG9iJfCdPeEK
xiDIpCrItzBR0oLaGJeH8gR0X+aPT/i7QkvalnLLLZR97llK7aif09PN6d64/UrO2Ee4
ZK+18rM6aGJF2/3a4y49obNqmTUvzn485xCAqggjeR+Xr/cRxezMvaEMlvBz88PbOkyZ
l8F27xOl3lX8/nWYFNA1j6AVbBcHeGstWFEJUuxJ2cBR/3250Z08WnYQ6t+OC4/S3abS
IzgMr7mgJNGk8iI0yfhYpQV5fgHLQH6hiYa1qLk8irgl
Received: by 10.224.76.202 with SMTP id d10mr3669541qak.0.1347410841111;
Tue, 11 Sep 2012 17:47:21 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.224.209.135 with SMTP id gg7ls2143431qab.6.gmail; Tue, 11 Sep
2012 17:47:20 -0700 (PDT)
Received: by 10.224.189.75 with SMTP id dd11mr8437284qab.6.1347410840181;
Tue, 11 Sep 2012 17:47:20 -0700 (PDT)
Received: by 10.224.189.75 with SMTP id dd11mr8437283qab.6.1347410840168;
Tue, 11 Sep 2012 17:47:20 -0700 (PDT)
Received: from mail-qc0-f177.google.com (mail-qc0-f177.google.com [209.85.216.177])
by gmr-mx.google.com with ESMTPS id fe27si1434021qcb.0.2012.09.11.17.47.19
(version=TLSv1/SSLv3 cipher=OTHER);
Tue, 11 Sep 2012 17:47:19 -0700 (PDT)
Received-SPF: pass (google.com: domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.177 as permitted sender) client-ip=209.85.216.177;
Received: by qcsu28 with SMTP id u28so735922qcs.36
for <golang-dev-/***@public.gmane.org>; Tue, 11 Sep 2012 17:47:19 -0700 (PDT)
Received: by 10.229.135.133 with SMTP id n5mr5373103qct.124.1347410839036;
Tue, 11 Sep 2012 17:47:19 -0700 (PDT)
Received: by 10.229.135.133 with SMTP id n5mr5373098qct.124.1347410838868;
Tue, 11 Sep 2012 17:47:18 -0700 (PDT)
Sender: golang-dev-/***@public.gmane.org
Received: by 10.224.33.202 with HTTP; Tue, 11 Sep 2012 17:46:48 -0700 (PDT)
In-Reply-To: <047d7bd6ac9274f59f04c9681528-hpIqsD4AKlfQT0dZR+***@public.gmane.org>
X-System-Of-Record: true
X-Gm-Message-State: ALoCoQmaxS7mzDANEtabLTXNAVGJGhbEQJZvglgOxfAOoipUUweF+dLXfqXrHcyq4cajsW3r4BwFKL/mMGb1Y3lhoivtOPIPnhW2XTrI1mvSZ7G35SSTEc0Aeq2ZJ3dH15bz41V6uz9t5zVYXA66PACqvagMlciZsaLxpIFDysVDocnymNSmAt6V9YWPcOnfl4imPohfKNbwvIyyafw8URzK1nbzkueEvg==
X-Original-Sender: adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.177 as permitted sender)
smtp.mail=adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org; dkim=pass header.i=@google.com
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45258>

There's one controversial patch here:

86c7b6d67466 10 39: +2811/-476 syscall: return EINVAL when string
arguments have NUL characters

It introduces an API change in syscall (ByteSliceFromString & co). I
am inclined to say this API expansion is permissible given it fixes a
security hole, and will most certainly be in Go 1.1. But this goes
against our policy of not expanding the API in minor point releases.

Thoughts?
Russ Cox
2012-09-12 00:48:31 UTC
Permalink
Received: by 10.52.65.74 with SMTP id v10mr3466325vds.18.1347410913064;
Tue, 11 Sep 2012 17:48:33 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.220.148.143 with SMTP id p15ls1513663vcv.9.gmail; Tue, 11 Sep
2012 17:48:32 -0700 (PDT)
Received: by 10.58.207.43 with SMTP id lt11mr7152747vec.17.1347410912401;
Tue, 11 Sep 2012 17:48:32 -0700 (PDT)
Received: by 10.58.207.43 with SMTP id lt11mr7152746vec.17.1347410912390;
Tue, 11 Sep 2012 17:48:32 -0700 (PDT)
Received: from mail-vc0-f178.google.com (mail-vc0-f178.google.com [209.85.220.178])
by gmr-mx.google.com with ESMTPS id z16si2830901vdf.0.2012.09.11.17.48.31
(version=TLSv1/SSLv3 cipher=OTHER);
Tue, 11 Sep 2012 17:48:31 -0700 (PDT)
Received-SPF: pass (google.com: domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.220.178 as permitted sender) client-ip=209.85.220.178;
Received: by mail-vc0-f178.google.com with SMTP id gb30so1714580vcb.37
for <golang-dev-/***@public.gmane.org>; Tue, 11 Sep 2012 17:48:31 -0700 (PDT)
Received: by 10.52.31.102 with SMTP id z6mr22471316vdh.119.1347410911197; Tue,
11 Sep 2012 17:48:31 -0700 (PDT)
Received: by 10.58.155.137 with HTTP; Tue, 11 Sep 2012 17:48:31 -0700 (PDT)
In-Reply-To: <CAHxS-otLnTLk3UQaNiJ0vbkyLyn3+SVSrY+4MV5MHfaPzVmrAw-JsoAwUIsXosN+***@public.gmane.org>
X-Gm-Message-State: ALoCoQmHvufz53ad90+6UonZcyawGCtVuVHV7RXAIpnBm3QDnWPkzjZ/oONLnq+cJWvCeLZFjDb4
X-Original-Sender: rsc-iFWiy5xATs8dnm+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.220.178 as permitted sender) smtp.mail=rsc-iFWiy5xATs8dnm+***@public.gmane.org
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
Sender: golang-dev-/***@public.gmane.org
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45259>
86c7b6d67466 10 39: +2811/-476 syscall: return EINVAL when string
arguments have NUL characters
It introduces an API change in syscall (ByteSliceFromString & co). I
am inclined to say this API expansion is permissible given it fixes a
security hole, and will most certainly be in Go 1.1. But this goes
against our policy of not expanding the API in minor point releases.
Thoughts?
Unexport the new functions but keep the changes to the existing ones.
Rob Pike
2012-09-12 04:10:40 UTC
Permalink
Received: by 10.50.196.131 with SMTP id im3mr7624273igc.6.1347423041440;
Tue, 11 Sep 2012 21:10:41 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.50.46.194 with SMTP id x2ls8393105igm.1.canary; Tue, 11 Sep
2012 21:10:40 -0700 (PDT)
Received: by 10.43.131.7 with SMTP id ho7mr9343004icc.5.1347423040628;
Tue, 11 Sep 2012 21:10:40 -0700 (PDT)
Received: by 10.43.131.7 with SMTP id ho7mr9343002icc.5.1347423040619;
Tue, 11 Sep 2012 21:10:40 -0700 (PDT)
Received: from mail-ie0-f179.google.com (mail-ie0-f179.google.com [209.85.223.179])
by gmr-mx.google.com with ESMTPS id dd6si1325719igc.0.2012.09.11.21.10.40
(version=TLSv1/SSLv3 cipher=OTHER);
Tue, 11 Sep 2012 21:10:40 -0700 (PDT)
Received-SPF: pass (google.com: domain of ***@golang.org designates 209.85.223.179 as permitted sender) client-ip=209.85.223.179;
Received: by ieje13 with SMTP id e13so2385893iej.38
for <golang-dev-/***@public.gmane.org>; Tue, 11 Sep 2012 21:10:40 -0700 (PDT)
Received: by 10.43.46.135 with SMTP id uo7mr13412448icb.45.1347423040537; Tue,
11 Sep 2012 21:10:40 -0700 (PDT)
Received: by 10.50.191.134 with HTTP; Tue, 11 Sep 2012 21:10:40 -0700 (PDT)
In-Reply-To: <CAHxS-otLnTLk3UQaNiJ0vbkyLyn3+SVSrY+4MV5MHfaPzVmrAw-JsoAwUIsXosN+***@public.gmane.org>
X-Gm-Message-State: ALoCoQkd1yGNItVYPj70VYMQI5d9TSgM5Dhyh+PDbAAqrXg/dlFd9vLqhCCzKbJEtfBid3JnuITC
X-Original-Sender: ***@golang.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of ***@golang.org designates 209.85.223.179 as permitted sender) smtp.mail=***@golang.org
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
Sender: golang-dev-/***@public.gmane.org
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45268>

The change fixes a security problem and should be included.

-rob
adg-iFWiy5xATs8dnm+
2012-09-12 08:11:42 UTC
Permalink
The files now include changes up to and including ffe134f40269.

I included a bunch more net/http patches to get a code base close to tip
(this is what we did for 1.0.1, too), except now the tree builds but
fails with this error:

--- FAIL: TestIssue3595 (0.03 seconds)
transport_test.go:848: Post: write tcp 127.0.0.1:51605: broken pipe
2012/09/12 18:01:42 Unsolicited response received on idle HTTP channel
starting with "H"; err=<nil>
FAIL
FAIL net/http 3.909s

Sigh. Tomorrow I'll try including more of the net/http changes. I didn't
think I missed any that related to this change, but I guess I must have.


http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded
File misc/backport/go103-excluded (right):

http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode98
misc/backport/go103-excluded:98: 272e1dd72156 -1 2: +29/-0 cmd/gc: fix
PkgPath of byte, rune types
Post by d***@public.gmane.org
This change fixes a regression introduced in go1.0.1 that might break
compatibility for users of reflect.
Ok, included.

http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode98
misc/backport/go103-excluded:98: 272e1dd72156 -1 2: +29/-0 cmd/gc: fix
PkgPath of byte, rune types
Post by d***@public.gmane.org
This change fixes a regression introduced in go1.0.1 that might break
compatibility for users of reflect.
Done.

http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode100
misc/backport/go103-excluded:100: 33a62d86ca2f -1 2: +19/-7 cmd/gc:
Suggest *T in error for x.(T) if it would work.
Post by d***@public.gmane.org
This change (33a62d86ca2f) is self-contained and I think it can be
helpful for
Post by d***@public.gmane.org
new Go programmers. Not essential but might be nice.
Added.

http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode101
misc/backport/go103-excluded:101: 3532ab6bf5be -1 4: +36/-21 runtime:
use sched_getaffinity for runtime.NumCPU() on Linux
Post by d***@public.gmane.org
I am mildly in favor of this change (3532ab6bf5be). It's more
accurate and more
Post by d***@public.gmane.org
efficient.
Added.

http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode115
misc/backport/go103-excluded:115: 56ea40aac72b -1 4: +330/-2 runtime:
add vdso support for linux/amd64. Fixes issue 1933.
Post by d***@public.gmane.org
This change (56ea40aac72b) should be included in 1.0.3.
Done.

http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode230
misc/backport/go103-excluded:230: 238a5fd52836 -10 3: +35/-6 cmd/gc: fix
escape analysis bug with variable capture in loops.
Post by d***@public.gmane.org
This change is a one-line fix, is it really not mergeable?
It causes potential corruption in programs and seems very important to
me.

The relevant escape analysis stuff isn't present in go1.0.2 and so the
fix isn't relevant as far as I can tell. Am I wrong?

http://codereview.appspot.com/6499087/diff/15001/misc/backport/go103-excluded
File misc/backport/go103-excluded (right):

http://codereview.appspot.com/6499087/diff/15001/misc/backport/go103-excluded#newcode1
misc/backport/go103-excluded:1: 80c98738fdde 0 3: +48/-1 cmd/gc: fix
initialization order involving method calls.
tried including this but it didn't merge cleanly. want to sort out the
other gc issues first.

http://codereview.appspot.com/6499087/
Russ Cox
2012-09-12 16:27:39 UTC
Permalink
Received: by 10.52.30.133 with SMTP id s5mr3921384vdh.10.1347467261824;
Wed, 12 Sep 2012 09:27:41 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.220.214.66 with SMTP id gz2ls2111564vcb.2.gmail; Wed, 12 Sep
2012 09:27:41 -0700 (PDT)
Received: by 10.58.212.198 with SMTP id nm6mr7637159vec.35.1347467261232;
Wed, 12 Sep 2012 09:27:41 -0700 (PDT)
Received: by 10.58.212.198 with SMTP id nm6mr7637158vec.35.1347467261224;
Wed, 12 Sep 2012 09:27:41 -0700 (PDT)
Received: from mail-vc0-f169.google.com (mail-vc0-f169.google.com [209.85.220.169])
by gmr-mx.google.com with ESMTPS id ef10si4397495vdb.3.2012.09.12.09.27.40
(version=TLSv1/SSLv3 cipher=OTHER);
Wed, 12 Sep 2012 09:27:40 -0700 (PDT)
Received-SPF: pass (google.com: domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.220.169 as permitted sender) client-ip=209.85.220.169;
Received: by mail-vc0-f169.google.com with SMTP id fl13so2873624vcb.0
for <golang-dev-/***@public.gmane.org>; Wed, 12 Sep 2012 09:27:40 -0700 (PDT)
Received: by 10.52.94.171 with SMTP id dd11mr1944483vdb.109.1347467260051;
Wed, 12 Sep 2012 09:27:40 -0700 (PDT)
Received: by 10.58.155.137 with HTTP; Wed, 12 Sep 2012 09:27:39 -0700 (PDT)
In-Reply-To: <047d7b5d42aeb7f2bd04c97cbdbc-hpIqsD4AKlfQT0dZR+***@public.gmane.org>
X-Gm-Message-State: ALoCoQkMrO1N7EkFebBNqXAZSzyAHW5gYz+2nHOoUklw3cG3hauf2yIKtJ9/WdkpgV9gBbXfJm8Z
X-Original-Sender: rsc-iFWiy5xATs8dnm+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.220.169 as permitted sender) smtp.mail=rsc-iFWiy5xATs8dnm+***@public.gmane.org
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
Sender: golang-dev-/***@public.gmane.org
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45300>

issue 3180 asks that we include ba91b34fe212 in 1.0.3. Does it apply cleanly?
Andrew Gerrand
2012-09-12 23:20:42 UTC
Permalink
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=google.com; s=20120113;
h=x-beenthere:received-spf:mime-version:sender:in-reply-to:references
:from:date:message-id:subject:to:cc:x-system-of-record
:x-gm-message-state:x-original-sender
:x-original-authentication-results:precedence:mailing-list:list-id
:x-google-group-id:list-post:list-help:list-archive:list-subscribe
:list-unsubscribe:content-type;
bh=g3jvC2aSpiznQ9F5W/Dn3zW6mork+cULtQ1790ehlLo=;
b=Eg3/n0xG5UXKzNstO4gYAmnKvLV0lxVbHqIuiLM43NMA1Yybkj3sEz6/w1HJK89FKi
O3CE0L0frGw7jwhkZ6s6OHYY2IN7wvkUvxkRPggz7dm76smAJH2ydg/algoV6AVYmgJR
9mNt/wf7qy9m0bEg4989tOmZsS5/vB8YBnylNlLyhrSnk/s80rtCmCCcew7RZFDSO14B
LuqupMPpemgtTzTlzTlRegHSbncJBaEfvewT2C54Swp8cqf46pk+2aSsJuiXbYl7+266
IH12AA73i4pemcNKeY5mB1HgSERSYzQ0U6FcghHV8KtW
Received: by 10.229.105.81 with SMTP id s17mr9697qco.14.1347492075021;
Wed, 12 Sep 2012 16:21:15 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.224.214.132 with SMTP id ha4ls4180846qab.1.gmail; Wed, 12 Sep
2012 16:21:14 -0700 (PDT)
Received: by 10.224.42.68 with SMTP id r4mr595259qae.4.1347492074360;
Wed, 12 Sep 2012 16:21:14 -0700 (PDT)
Received: by 10.224.42.68 with SMTP id r4mr595255qae.4.1347492074255;
Wed, 12 Sep 2012 16:21:14 -0700 (PDT)
Received: from mail-qa0-f45.google.com (mail-qa0-f45.google.com [209.85.216.45])
by gmr-mx.google.com with ESMTPS id fe27si2294514qcb.0.2012.09.12.16.21.13
(version=TLSv1/SSLv3 cipher=OTHER);
Wed, 12 Sep 2012 16:21:13 -0700 (PDT)
Received-SPF: pass (google.com: domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.45 as permitted sender) client-ip=209.85.216.45;
Received: by mail-qa0-f45.google.com with SMTP id c10so3531101qad.18
for <golang-dev-/***@public.gmane.org>; Wed, 12 Sep 2012 16:21:13 -0700 (PDT)
Received: by 10.224.31.202 with SMTP id z10mr1554410qac.86.1347492073166;
Wed, 12 Sep 2012 16:21:13 -0700 (PDT)
Received: by 10.224.31.202 with SMTP id z10mr1554384qac.86.1347492073004; Wed,
12 Sep 2012 16:21:13 -0700 (PDT)
Sender: golang-dev-/***@public.gmane.org
Received: by 10.224.33.202 with HTTP; Wed, 12 Sep 2012 16:20:42 -0700 (PDT)
In-Reply-To: <CAA8EjDRJWxg-Vg5yVwzKX6eQJfY6CrpL2ED8oti1YaNOoAtSKg-JsoAwUIsXosN+***@public.gmane.org>
X-System-Of-Record: true
X-Gm-Message-State: ALoCoQmq1mxLP3eQhgrXDvfxzbqBg6lz6QukwYCSQebvZBJ6JVwR4BYcjnKR4qz8PzaB34O6E7rMydLrzgapk5hwIYokxI7JRKJ/jrB/kqpcS+K6aBXtuxcbReO1SlX+gonJ/D3ATDk7orj6a+O8L+ZGL1rJWnebTdvW52zlgIaux/2oVmB1vi0gT3eV5j+aF8WqA5SdksZcVfagIPrMFxfqVUMpW8uuBg==
X-Original-Sender: adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.45 as permitted sender)
smtp.mail=adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org; dkim=pass header.i=@google.com
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45330>
Post by Russ Cox
issue 3180 asks that we include ba91b34fe212 in 1.0.3. Does it apply cleanly?
The change predates Go 1.0, so it's already in release-branch.go1.

https://code.google.com/p/go/source/detail?r=ba91b34fe212
r***@public.gmane.org
2012-09-12 19:52:45 UTC
Permalink
Post by r***@public.gmane.org
* Issue 3180: annoyances with GOROOT_FINAL (make debugging
uncomfortable)
Post by r***@public.gmane.org
a608a03df703 5a, 6a, 8a: take GOROOT_FINAL into consideration
016c4fefed77 5c, 6c, 8c: take GOROOT_FINAL into consideration
ba91b34fe212 cmd/gc: if $GOROOT_FINAL is set, rewrite file names in
object files

What is the position on this? Issue 3180 mentions the fix for the Go
compiler, but for consistency the fixes for assembler and C should be
included as well.
Post by r***@public.gmane.org
* Issue 3918: serious compiler corruptions on 64 bits
72678c75b2d3 cmd/6c: fix probable code gen bug
a9e09f1fe373 6c, 6g, 6l: add MOVQL to make truncation explicit
adg: what is your opinion on these? We use Go in production and having
potential mysterious bugs whenever integers are converted between 32-bit
and 64-bit is really a serious issue.

http://codereview.appspot.com/6499087/
Andrew Gerrand
2012-09-12 23:31:08 UTC
Permalink
Post by r***@public.gmane.org
Post by r***@public.gmane.org
* Issue 3180: annoyances with GOROOT_FINAL (make debugging
uncomfortable)
Post by r***@public.gmane.org
a608a03df703 5a, 6a, 8a: take GOROOT_FINAL into consideration
016c4fefed77 5c, 6c, 8c: take GOROOT_FINAL into consideration
ba91b34fe212 cmd/gc: if $GOROOT_FINAL is set, rewrite file names in
object files
What is the position on this? Issue 3180 mentions the fix for the Go
compiler, but for consistency the fixes for assembler and C should be
included as well.
Happy to take them if it's not too painful to merge them.
Post by r***@public.gmane.org
Post by r***@public.gmane.org
* Issue 3918: serious compiler corruptions on 64 bits
72678c75b2d3 cmd/6c: fix probable code gen bug
a9e09f1fe373 6c, 6g, 6l: add MOVQL to make truncation explicit
adg: what is your opinion on these? We use Go in production and having
potential mysterious bugs whenever integers are converted between 32-bit
and 64-bit is really a serious issue.
I don't have an opinion. I rely on guidance from people who actually
know the compilers. It's all about risk; will the code be more
reliable with or without the patch?
r***@public.gmane.org
2012-09-12 19:52:57 UTC
Permalink
http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded
File misc/backport/go103-excluded (right):

http://codereview.appspot.com/6499087/diff/2002/misc/backport/go103-excluded#newcode230
misc/backport/go103-excluded:230: 238a5fd52836 -10 3: +35/-6 cmd/gc: fix
escape analysis bug with variable capture in loops.
The bug is present in Go 1.0.2 as mentioned in issue 3975. It breaks
memory safety and should be treated (at least) like security issues.

The relevant line is esc.c:145 in go1.0.2 because of some code move.

http://codereview.appspot.com/6499087/
adg-iFWiy5xATs8dnm+
2012-09-13 00:19:06 UTC
Permalink
Post by Andrew Gerrand
Post by r***@public.gmane.org
Post by r***@public.gmane.org
* Issue 3180: annoyances with GOROOT_FINAL (make debugging
uncomfortable)
Post by r***@public.gmane.org
a608a03df703 5a, 6a, 8a: take GOROOT_FINAL into consideration
016c4fefed77 5c, 6c, 8c: take GOROOT_FINAL into consideration
ba91b34fe212 cmd/gc: if $GOROOT_FINAL is set, rewrite file names in
object files
What is the position on this? Issue 3180 mentions the fix for the Go
compiler, but for consistency the fixes for assembler and C should
be
Post by Andrew Gerrand
Post by r***@public.gmane.org
included as well.
Happy to take them if it's not too painful to merge them.
These merged fine.

http://codereview.appspot.com/6499087/
adg-iFWiy5xATs8dnm+
2012-09-13 00:42:23 UTC
Permalink
fix escape
Post by r***@public.gmane.org
analysis bug with variable capture in loops.
The bug is present in Go 1.0.2 as mentioned in issue 3975. It breaks
memory
Post by r***@public.gmane.org
safety and should be treated (at least) like security issues.
The relevant line is esc.c:145 in go1.0.2 because of some code move.
Can you please take a look at 238a5fd52836.diff ?

I modified the diff to esc.c as I assumed was correct, but now
test/escape.go fails:

# ../test
run escape.go : incorrect output
wrong value want 105 got 106 after for_escapes3
aliased pointers -15 -16 after for_escapes3
panic: BUG: no escape

goroutine 1 [running]:
main.main()
/tmp/go1-test/test/escape.go:220 +0x6b6

goroutine 2 [syscall]:
created by runtime.main
/tmp/go1-test/src/pkg/runtime/proc.c:221
exit status 2

exit status 1


http://codereview.appspot.com/6499087/
r***@public.gmane.org
2012-09-13 06:53:30 UTC
Permalink
fix
Post by r***@public.gmane.org
escape
Post by r***@public.gmane.org
analysis bug with variable capture in loops.
The bug is present in Go 1.0.2 as mentioned in issue 3975. It breaks
memory
Post by r***@public.gmane.org
Post by r***@public.gmane.org
safety and should be treated (at least) like security issues.
The relevant line is esc.c:145 in go1.0.2 because of some code move.
Can you please take a look at 238a5fd52836.diff ?
I modified the diff to esc.c as I assumed was correct, but now
test/escape.go

Sorry it seems things have a little bit more, so you will need this
patch. rsc, do you approve?

--- /usr/lib/go/src/cmd/gc/esc.c 2012-06-14 05:23:38.000000000 +0200
+++ src/cmd/gc/esc.c 2012-09-13 08:52:16.180558427 +0200
@@ -143,6 +143,9 @@

n = nod(OADDR, ll->n->closure, N);
n->lineno = ll->n->lineno;
+ // n belongs to the outer scope.
+ n->escloopdepth = saveld;
+ n->curfn = savefn;
typecheck(&n, Erv);
escassign(curfn, n);
}
Post by r***@public.gmane.org
# ../test
run escape.go : incorrect output
wrong value want 105 got 106 after for_escapes3
aliased pointers -15 -16 after for_escapes3
panic: BUG: no escape
main.main()
/tmp/go1-test/test/escape.go:220 +0x6b6
created by runtime.main
/tmp/go1-test/src/pkg/runtime/proc.c:221
exit status 2
exit status 1
http://codereview.appspot.com/6499087/
Andrew Gerrand
2012-09-14 03:59:42 UTC
Permalink
Post by r***@public.gmane.org
Post by r***@public.gmane.org
misc/backport/go103-excluded:230: 238a5fd52836 -10 3: +35/-6
fix
Post by r***@public.gmane.org
escape
analysis bug with variable capture in loops.
The bug is present in Go 1.0.2 as mentioned in issue 3975. It breaks
memory
Post by r***@public.gmane.org
safety and should be treated (at least) like security issues.
The relevant line is esc.c:145 in go1.0.2 because of some code move.
Can you please take a look at 238a5fd52836.diff ?
I modified the diff to esc.c as I assumed was correct, but now
test/escape.go
Sorry it seems things have a little bit more, so you will need this
patch. rsc, do you approve?
--- /usr/lib/go/src/cmd/gc/esc.c 2012-06-14 05:23:38.000000000 +0200
+++ src/cmd/gc/esc.c 2012-09-13 08:52:16.180558427 +0200
@@ -143,6 +143,9 @@
n = nod(OADDR, ll->n->closure, N);
n->lineno = ll->n->lineno;
+ // n belongs to the outer scope.
+ n->escloopdepth = saveld;
+ n->curfn = savefn;
typecheck(&n, Erv);
escassign(curfn, n);
}
Thanks Remy. This patch works well.
rsc-iFWiy5xATs8dnm+
2012-09-13 18:53:40 UTC
Permalink
http://codereview.appspot.com/6499087/diff/7046/misc/backport/backport
File misc/backport/backport (right):

http://codereview.appspot.com/6499087/diff/7046/misc/backport/backport#newcode26
misc/backport/backport:26: yes c |
HGEDITOR=$goroot/misc/backport/graftedit REVISION=$r hg graft -e -D $r
I'm a little worried by this. I didn't need 'yes c |' for go 1.0.2. Does
that imply things are not merging perfectly and you are overriding the
conflicts?

http://codereview.appspot.com/6499087/diff/7046/misc/backport/backport#newcode52
misc/backport/backport:52: exit 0
Delete.

http://codereview.appspot.com/6499087/
rsc-iFWiy5xATs8dnm+
2012-09-13 19:10:04 UTC
Permalink
What about
8eae6e390d20 fix handling of CONNECT in http servemux
5e7fd762f356 fix testing memory usage
7a67d277c7e8 fix concurrent logging and failing in testing



http://codereview.appspot.com/6499087/diff/7046/misc/backport/go103
File misc/backport/go103 (right):

http://codereview.appspot.com/6499087/diff/7046/misc/backport/go103#newcode6
misc/backport/go103:6: 67b66dd9ea4a 100 2: +188/-37 net/http: add
If-None-Match and If-Range support to ServeContent
I don't believe this is appropriate in Go 1.0.3. This is new
functionality.

http://codereview.appspot.com/6499087/
rsc-iFWiy5xATs8dnm+
2012-09-13 19:11:36 UTC
Permalink
http://codereview.appspot.com/6499087/diff/7046/misc/backport/patches/22a65f54a36a.diff
File misc/backport/patches/22a65f54a36a.diff (right):

http://codereview.appspot.com/6499087/diff/7046/misc/backport/patches/22a65f54a36a.diff#newcode1
misc/backport/patches/22a65f54a36a.diff:1: diff -r 0e114e4df4ec -r
22a65f54a36a src/pkg/syscall/exec_plan9.go
Please include the CL description at the top of all these diffs. patch
will ignore it.

http://codereview.appspot.com/6499087/
adg-iFWiy5xATs8dnm+
2012-09-14 02:35:35 UTC
Permalink
http://codereview.appspot.com/6499087/diff/7046/misc/backport/backport
File misc/backport/backport (right):

http://codereview.appspot.com/6499087/diff/7046/misc/backport/backport#newcode26
misc/backport/backport:26: yes c |
HGEDITOR=$goroot/misc/backport/graftedit REVISION=$r hg graft -e -D $r
Post by rsc-iFWiy5xATs8dnm+
I'm a little worried by this. I didn't need 'yes c |' for go 1.0.2.
Does that
Post by rsc-iFWiy5xATs8dnm+
imply things are not merging perfectly and you are overriding the
conflicts?

It's for these prompts:

remote changed src/pkg/net/net_posix.go which local deleted
use (c)hanged version or leave (d)eleted? c

which, for some reason, appeared with the rc version of the script and
automatically chose 'c'. Maybe because rc didn't make hg aware of the
controlling terminal? Whatever the case, this should be the same
behavior as with the rc script.

http://codereview.appspot.com/6499087/diff/7046/misc/backport/go103
File misc/backport/go103 (right):

http://codereview.appspot.com/6499087/diff/7046/misc/backport/go103#newcode6
misc/backport/go103:6: 67b66dd9ea4a 100 2: +188/-37 net/http: add
If-None-Match and If-Range support to ServeContent
Post by rsc-iFWiy5xATs8dnm+
I don't believe this is appropriate in Go 1.0.3. This is new
functionality.

Without it I get test failures:

--- FAIL: TestIssue3595 (0.01 seconds)
transport_test.go:848: Post: write tcp 127.0.0.1:51987: broken pipe
FAIL

Given that this merely makes ServeContent more compliant with the RFC,
I'm inclined to treat it as bugfix-esque.

http://codereview.appspot.com/6499087/diff/7046/misc/backport/patches/22a65f54a36a.diff
File misc/backport/patches/22a65f54a36a.diff (right):

http://codereview.appspot.com/6499087/diff/7046/misc/backport/patches/22a65f54a36a.diff#newcode1
misc/backport/patches/22a65f54a36a.diff:1: diff -r 0e114e4df4ec -r
22a65f54a36a src/pkg/syscall/exec_plan9.go
Post by rsc-iFWiy5xATs8dnm+
Please include the CL description at the top of all these diffs. patch
will
Post by rsc-iFWiy5xATs8dnm+
ignore it.
Done.

http://codereview.appspot.com/6499087/
Russ Cox
2012-09-14 02:53:59 UTC
Permalink
Received: by 10.52.33.34 with SMTP id o2mr182195vdi.12.1347591241485;
Thu, 13 Sep 2012 19:54:01 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.52.92.207 with SMTP id co15ls3286016vdb.5.gmail; Thu, 13 Sep
2012 19:54:00 -0700 (PDT)
Received: by 10.58.145.65 with SMTP id ss1mr202068veb.39.1347591240818;
Thu, 13 Sep 2012 19:54:00 -0700 (PDT)
Received: by 10.58.145.65 with SMTP id ss1mr202067veb.39.1347591240473;
Thu, 13 Sep 2012 19:54:00 -0700 (PDT)
Received: from mail-vc0-f179.google.com (mail-vc0-f179.google.com [209.85.220.179])
by gmr-mx.google.com with ESMTPS id ef10si9561vdb.3.2012.09.13.19.53.59
(version=TLSv1/SSLv3 cipher=OTHER);
Thu, 13 Sep 2012 19:53:59 -0700 (PDT)
Received-SPF: pass (google.com: domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.220.179 as permitted sender) client-ip=209.85.220.179;
Received: by vcbfl13 with SMTP id fl13so799087vcb.10
for <golang-dev-/***@public.gmane.org>; Thu, 13 Sep 2012 19:53:59 -0700 (PDT)
Received: by 10.52.94.171 with SMTP id dd11mr623850vdb.109.1347591239355; Thu,
13 Sep 2012 19:53:59 -0700 (PDT)
Received: by 10.58.155.137 with HTTP; Thu, 13 Sep 2012 19:53:59 -0700 (PDT)
In-Reply-To: <047d7b86c5ec588a0004c9a0470a-hpIqsD4AKlfQT0dZR+***@public.gmane.org>
X-Gm-Message-State: ALoCoQlB/gzkdwPW6Oy5YKpt6YcxdMBfvEe8DTlqAF3KOZLqkJhdmb0T7xZWgmWkCWYL6KYUIYd4
X-Original-Sender: rsc-iFWiy5xATs8dnm+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.220.179 as permitted sender) smtp.mail=rsc-iFWiy5xATs8dnm+***@public.gmane.org
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
Sender: golang-dev-/***@public.gmane.org
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45477>
Post by adg-iFWiy5xATs8dnm+
remote changed src/pkg/net/net_posix.go which local deleted
use (c)hanged version or leave (d)eleted? c
which, for some reason, appeared with the rc version of the script and
automatically chose 'c'. Maybe because rc didn't make hg aware of the
controlling terminal? Whatever the case, this should be the same
behavior as with the rc script.
I just didn't have any of those in go102. Thanks for the explanation.
Post by adg-iFWiy5xATs8dnm+
--- FAIL: TestIssue3595 (0.01 seconds)
transport_test.go:848: Post: write tcp 127.0.0.1:51987: broken pipe
FAIL
Given that this merely makes ServeContent more compliant with the RFC,
I'm inclined to treat it as bugfix-esque.
It changes the behavior of the server based on the caller's setting of
the Etag header, which has never been documented to have any special
meaning to the server code. It's an API change, albeit one without a
corresponding type system change. It would be okay to keep the header
stripping but the If-Range and If-None-Match changes cannot go into an
x.y.z release. It is almost certainly easier to drop the CL whole than
just half.

I have no idea why not having it would cause a test failure. I ran hg
undo 67b66dd9ea4a from tip in my local client, and all.bash and 'go
test net/http' are both happy. The failure you are seeing worries me.
Can you push a version with that? It is probably better to push to new
clones than to make new projects, by the way. Scroll to the bottom of
http://code.google.com/p/go/source/clones and click 'Create a clone'.

Russ
Andrew Gerrand
2012-09-14 02:57:09 UTC
Permalink
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=google.com; s=20120113;
h=x-beenthere:received-spf:mime-version:sender:in-reply-to:references
:from:date:message-id:subject:to:cc:x-system-of-record
:x-gm-message-state:x-original-sender
:x-original-authentication-results:precedence:mailing-list:list-id
:x-google-group-id:list-post:list-help:list-archive:list-subscribe
:list-unsubscribe:content-type;
bh=SMWSfRLqGHLISFEcWOGjed3KKbFVz645QBYsUMH8eZQ=;
b=pLYXTrDt5SehV72n+ReDLDsy+PyzX7B7ojEUh9MFpOWP25e1nv81cQr7tI3k2O52UC
D61Z2xiP5a6ZyFZneE+5MkGJ1DUzaC34sx6tF+Qi4wbB+YX7e7vv0QNko2AKuxYmtbcT
cNnlcUFFniGhEgicUf4vxcrllnqT3705BlRvw2HDZIhCUhVd3c3KQW+YAaOCBwVKLamd
tDt38OmnIY63pgTrfeRiEiA07cQ4BVLEEImLTOHOXyrngzLzcCi2lDmS7px2xj3613TB
b43WB3PnMwAEAWImlFP6HV1cGGFbjhOhkPEPQz8l4sie
Received: by 10.224.221.19 with SMTP id ia19mr487428qab.3.1347591462190;
Thu, 13 Sep 2012 19:57:42 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.224.138.79 with SMTP id z15ls6714120qat.7.gmail; Thu, 13 Sep
2012 19:57:41 -0700 (PDT)
Received: by 10.224.220.12 with SMTP id hw12mr1205404qab.8.1347591461554;
Thu, 13 Sep 2012 19:57:41 -0700 (PDT)
Received: by 10.224.220.12 with SMTP id hw12mr1205402qab.8.1347591461546;
Thu, 13 Sep 2012 19:57:41 -0700 (PDT)
Received: from mail-qa0-f52.google.com (mail-qa0-f52.google.com [209.85.216.52])
by gmr-mx.google.com with ESMTPS id fe27si25668qcb.0.2012.09.13.19.57.40
(version=TLSv1/SSLv3 cipher=OTHER);
Thu, 13 Sep 2012 19:57:40 -0700 (PDT)
Received-SPF: pass (google.com: domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.52 as permitted sender) client-ip=209.85.216.52;
Received: by qabg14 with SMTP id g14so3229796qab.18
for <golang-dev-/***@public.gmane.org>; Thu, 13 Sep 2012 19:57:40 -0700 (PDT)
Received: by 10.224.107.2 with SMTP id z2mr3863659qao.71.1347591460418;
Thu, 13 Sep 2012 19:57:40 -0700 (PDT)
Received: by 10.224.107.2 with SMTP id z2mr3863637qao.71.1347591460258; Thu,
13 Sep 2012 19:57:40 -0700 (PDT)
Sender: golang-dev-/***@public.gmane.org
Received: by 10.224.33.202 with HTTP; Thu, 13 Sep 2012 19:57:09 -0700 (PDT)
In-Reply-To: <CAA8EjDQvg4BzWsQtsMvJzWmqQmxaCWCpOvhR1Q_q79WBXLBBfQ-JsoAwUIsXosN+***@public.gmane.org>
X-System-Of-Record: true
X-Gm-Message-State: ALoCoQnAMBh4De9/lEWmFadWmgq7dsSmHnyZ0aSKoYfT3BwGL46XE7WhToeKgNA9eGDvamGm/XWhxO1s+AhQ/ddpfPE3fLHbDeRepz2gvx4OnCFHYXWlf4oBqbbb3f29PicBSeZtmaM5+aSwk2s1gowPW9z/IYYDaCWeyC+K2D4enGGpSQWxbx+LXPWd3nTfw8Wh46r1VrVgG6hsNbJARdK4xG5Q9cooQQ==
X-Original-Sender: adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.52 as permitted sender)
smtp.mail=adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org; dkim=pass header.i=@google.com
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45480>
Post by Russ Cox
Post by adg-iFWiy5xATs8dnm+
--- FAIL: TestIssue3595 (0.01 seconds)
transport_test.go:848: Post: write tcp 127.0.0.1:51987: broken pipe
FAIL
Given that this merely makes ServeContent more compliant with the RFC,
I'm inclined to treat it as bugfix-esque.
It changes the behavior of the server based on the caller's setting of
the Etag header, which has never been documented to have any special
meaning to the server code. It's an API change, albeit one without a
corresponding type system change. It would be okay to keep the header
stripping but the If-Range and If-None-Match changes cannot go into an
x.y.z release. It is almost certainly easier to drop the CL whole than
just half.
OK, I agree it should be dropped.
Post by Russ Cox
I have no idea why not having it would cause a test failure. I ran hg
undo 67b66dd9ea4a from tip in my local client, and all.bash and 'go
test net/http' are both happy. The failure you are seeing worries me.
Can you push a version with that? It is probably better to push to new
clones than to make new projects, by the way. Scroll to the bottom of
http://code.google.com/p/go/source/clones and click 'Create a clone'.
TestIssue3595 continues to cause problems. It has resurfaced, and I'm
not sure why. I am wondering whether it is flaky against the current
patch set and I am just unlucky. I will continue to investigate.
Andrew Gerrand
2012-09-14 05:12:18 UTC
Permalink
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=google.com; s=20120113;
h=x-beenthere:received-spf:mime-version:sender:in-reply-to:references
:from:date:message-id:subject:to:cc:x-system-of-record
:x-gm-message-state:x-original-sender
:x-original-authentication-results:precedence:mailing-list:list-id
:x-google-group-id:list-post:list-help:list-archive:list-subscribe
:list-unsubscribe:content-type;
bh=yNuW4D8IVn6XR1FukL+17n9QE5qJlmIcg5eaiooWU4Q=;
b=a598oMDhaK66rNQ5aex9uVdGQH1OHlUnmqjL2D28og5qpi8zZpwrzqFhkgIWnofyfw
SoJUtbIhBbydlMefc9exjnxbQSn/QAGmSYC/clD8ag8R0GURp6IZs6625XE3M09fW5cb
kMJpXFzYPRvbv6rU+Qa7iQSn+CdLsIuqKFHrmBNVK2bedNRI94xL46H9jzVpPcmEMluR
o1XLqBtKLaozPSPasSMmlyicAyZbfIQur8AY0fOtL54xj7HBk/0yundiF//xJLYLEOt1
99VPXkLYT30h7Fj5YQr0glyTWlRq+GwK+zpr+PewH/Cu
Received: by 10.224.201.196 with SMTP id fb4mr494896qab.11.1347599571639;
Thu, 13 Sep 2012 22:12:51 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.224.216.6 with SMTP id hg6ls6815794qab.5.gmail; Thu, 13 Sep
2012 22:12:51 -0700 (PDT)
Received: by 10.224.31.20 with SMTP id w20mr1346885qac.2.1347599571100;
Thu, 13 Sep 2012 22:12:51 -0700 (PDT)
Received: by 10.224.31.20 with SMTP id w20mr1346884qac.2.1347599571090;
Thu, 13 Sep 2012 22:12:51 -0700 (PDT)
Received: from mail-qc0-f173.google.com (mail-qc0-f173.google.com [209.85.216.173])
by gmr-mx.google.com with ESMTPS id mx3si63705qcb.3.2012.09.13.22.12.50
(version=TLSv1/SSLv3 cipher=OTHER);
Thu, 13 Sep 2012 22:12:50 -0700 (PDT)
Received-SPF: pass (google.com: domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.173 as permitted sender) client-ip=209.85.216.173;
Received: by qcab12 with SMTP id b12so3337073qca.32
for <golang-dev-/***@public.gmane.org>; Thu, 13 Sep 2012 22:12:50 -0700 (PDT)
Received: by 10.224.110.197 with SMTP id o5mr4528898qap.73.1347599569918;
Thu, 13 Sep 2012 22:12:49 -0700 (PDT)
Received: by 10.224.110.197 with SMTP id o5mr4528884qap.73.1347599569797; Thu,
13 Sep 2012 22:12:49 -0700 (PDT)
Sender: golang-dev-/***@public.gmane.org
Received: by 10.224.33.202 with HTTP; Thu, 13 Sep 2012 22:12:18 -0700 (PDT)
In-Reply-To: <CAA8EjDQvg4BzWsQtsMvJzWmqQmxaCWCpOvhR1Q_q79WBXLBBfQ-JsoAwUIsXosN+***@public.gmane.org>
X-System-Of-Record: true
X-Gm-Message-State: ALoCoQn57cc6ucSKqF3H3AtCE0vezQmc5pwMFhHQMlmn9RD9s7CrglHbqUuvcSWeGMTqB9PxaOZ9y4Jv0+34YjLeCbasULZAcZi4bh09TWyPVvx0RH8QSsB51NWxWm+F92GZBkcPDJZ8hJTSCiZEuSe4kJG1aP+Sb/XMLSRwEx+gLQ4HZVMl7DIGAb5FdYnqo2Xpc3my1dKZnE2SFWcL1kqHcN3ge3Ytcg==
X-Original-Sender: adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.173 as permitted sender)
smtp.mail=adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org; dkim=pass header.i=@google.com
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45502>

I've updated this CL and pushed out a new candidate to this repo:
https://code.google.com/r/adg-go103-test-1

It builds and passes all tests on my macbook. Yay, or so I thought.

On my ubuntu Linux machine it fails the same net/http test that was
failing before:

2012/09/14 14:54:17 Unsolicited response received on idle HTTP channel
starting with "H"; err=<nil>
--- FAIL: TestIssue3595 (0.00 seconds)
transport_test.go:848: Post: write tcp 127.0.0.1:42247: connection
reset by peer
2012/09/14 14:54:18 Unsolicited response received on idle HTTP channel
starting with "H"; err=<nil>
FAIL
FAIL net/http 1.960s

Confusing. I am going to stop for today and resume playing with it on Monday.

Russ, if you'd like to have a crack at this on your Friday that'd be great.

Brad, I know you said you didn't want to be debugging a frankenstein
of your net/http, but can you think of why this might pass on my
macbook but fail on my linux box? The linux box has 6 cores, while the
macbook has 2. Could be related?

Andrew
r***@public.gmane.org
2012-09-14 09:55:47 UTC
Permalink
Post by Andrew Gerrand
Post by adg-iFWiy5xATs8dnm+
Post by adg-iFWiy5xATs8dnm+
I modified the diff to esc.c as I assumed was correct, but now
test/escape.go
Sorry it seems things have a little bit more, so you will need this
patch. rsc, do you approve?
Thanks Remy. This patch works well.
Thanks. I still really think issue 3918 should get a fix or a workaround
for Go 1.0.3. It triggers way too easily to be ignored. I agree that the
commits on tip that fixes the issue are too intrusive.

Can you/rsc/anyone take a look at http://codereview.appspot.com/6497136/
?

It disables the part of the optimizer that leads to the bug, and should
apply really cleanly.

Thanks.

http://codereview.appspot.com/6499087/
Russ Cox
2012-09-14 17:31:14 UTC
Permalink
Received: by 10.229.136.130 with SMTP id r2mr205393qct.0.1347643877126;
Fri, 14 Sep 2012 10:31:17 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.229.175.169 with SMTP id ba41ls5737929qcb.5.gmail; Fri, 14 Sep
2012 10:31:15 -0700 (PDT)
Received: by 10.224.182.74 with SMTP id cb10mr2464324qab.0.1347643875922;
Fri, 14 Sep 2012 10:31:15 -0700 (PDT)
Received: by 10.224.182.74 with SMTP id cb10mr2464321qab.0.1347643875897;
Fri, 14 Sep 2012 10:31:15 -0700 (PDT)
Received: from mail-qa0-f52.google.com (mail-qa0-f52.google.com [209.85.216.52])
by gmr-mx.google.com with ESMTPS id t29si541234qcz.1.2012.09.14.10.31.14
(version=TLSv1/SSLv3 cipher=OTHER);
Fri, 14 Sep 2012 10:31:14 -0700 (PDT)
Received-SPF: pass (google.com: domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.216.52 as permitted sender) client-ip=209.85.216.52;
Received: by qabg14 with SMTP id g14so71248qab.18
for <golang-dev-/***@public.gmane.org>; Fri, 14 Sep 2012 10:31:14 -0700 (PDT)
Received: by 10.224.177.203 with SMTP id bj11mr8790096qab.60.1347643874599;
Fri, 14 Sep 2012 10:31:14 -0700 (PDT)
Received: by 10.49.6.66 with HTTP; Fri, 14 Sep 2012 10:31:14 -0700 (PDT)
In-Reply-To: <047d7b6d82b69548cc04c9a66d7e-hpIqsD4AKlfQT0dZR+***@public.gmane.org>
X-Gm-Message-State: ALoCoQkLucAsj1jBq4XPXOyGjHe+xlsccXDLX4sR8aqHSr6mg7xZX2BuQ9OK7OV1W1PHbnmUIzqI
X-Original-Sender: rsc-iFWiy5xATs8dnm+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.216.52 as permitted sender) smtp.mail=rsc-iFWiy5xATs8dnm+***@public.gmane.org
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
Sender: golang-dev-/***@public.gmane.org
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45550>

Remy, do you have any measurements of the performance difference (say,
on test/bench/go1) if we apply 6497136. I'm a little worried.

Russ
Russ Cox
2012-09-16 20:20:43 UTC
Permalink
Received: by 10.52.21.44 with SMTP id s12mr404213vde.5.1347826845232;
Sun, 16 Sep 2012 13:20:45 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.52.156.112 with SMTP id wd16ls675451vdb.6.gmail; Sun, 16 Sep
2012 13:20:44 -0700 (PDT)
Received: by 10.58.135.42 with SMTP id pp10mr1694224veb.18.1347826844681;
Sun, 16 Sep 2012 13:20:44 -0700 (PDT)
Received: by 10.58.135.42 with SMTP id pp10mr1694223veb.18.1347826844673;
Sun, 16 Sep 2012 13:20:44 -0700 (PDT)
Received: from mail-vc0-f173.google.com (mail-vc0-f173.google.com [209.85.220.173])
by gmr-mx.google.com with ESMTPS id s13si315522vde.2.2012.09.16.13.20.43
(version=TLSv1/SSLv3 cipher=OTHER);
Sun, 16 Sep 2012 13:20:43 -0700 (PDT)
Received-SPF: pass (google.com: domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.220.173 as permitted sender) client-ip=209.85.220.173;
Received: by mail-vc0-f173.google.com with SMTP id fl15so8044210vcb.32
for <golang-dev-/***@public.gmane.org>; Sun, 16 Sep 2012 13:20:43 -0700 (PDT)
Received: by 10.52.91.176 with SMTP id cf16mr2532237vdb.50.1347826843561; Sun,
16 Sep 2012 13:20:43 -0700 (PDT)
Received: by 10.58.155.137 with HTTP; Sun, 16 Sep 2012 13:20:43 -0700 (PDT)
In-Reply-To: <CAA8EjDSgGYBmvEhqWFEaP76MKE3o4QcN=MYzRN589Ys9gW4ZMQ-JsoAwUIsXosN+***@public.gmane.org>
X-Gm-Message-State: ALoCoQnxFOJdtzHiNSCXx/XiLld8zU6XfIfcu2/TjSKrKYwEqF46lKu6vgpcu2kI1dMxT6jmLPHS
X-Original-Sender: rsc-iFWiy5xATs8dnm+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.220.173 as permitted sender) smtp.mail=rsc-iFWiy5xATs8dnm+***@public.gmane.org
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
Sender: golang-dev-/***@public.gmane.org
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45637>

I looked into the net/http test failure in the go103-test-1 repo.
I don't believe 305f67dc3f99 is correct. Can you drop it from the go103 list?

Russ
Andrew Gerrand
2012-09-17 19:28:37 UTC
Permalink
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=google.com; s=20120113;
h=x-beenthere:received-spf:mime-version:sender:in-reply-to:references
:from:date:message-id:subject:to:cc:x-system-of-record
:x-gm-message-state:x-original-sender
:x-original-authentication-results:precedence:mailing-list:list-id
:x-google-group-id:list-post:list-help:list-archive:list-subscribe
:list-unsubscribe:content-type;
bh=GFOLWv2JOai6MWRuz+BzVnyGBWutTVScmHMEIUc5DDA=;
b=eXreZFhSk7ar/v4LIf9yUaI5KS8evkZ2wCCO2M7lOy0cnmyV57p0vrs9tspq7wA4j9
Th4gsdBTJUIqMKuNqa8cjXuMg/SYXVZUrVZNdZVY+6KGBU8RGLQLFvgpPRUo7a2XbLX3
xABzf6ueS34AVGHhlrzuvX+Rt6c78iPUtAXfpzjB5055fnq5D0LFeFoRJuh6i4W19QcU
nOIrKaBrxC10WBrB5y6tNim1j9FaBidS+mJcPQ38X/6EuqyFJcqwHBSaw1knzcgX07A0
jmUc+M2bdGUICXoo3y5Q5Y847FrrpLs8GvuFxhj2jg47
Received: by 10.224.220.139 with SMTP id hy11mr2647126qab.10.1347910151144;
Mon, 17 Sep 2012 12:29:11 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.224.210.4 with SMTP id gi4ls12480295qab.2.gmail; Mon, 17 Sep
2012 12:29:10 -0700 (PDT)
Received: by 10.224.42.68 with SMTP id r4mr8165332qae.4.1347910150393;
Mon, 17 Sep 2012 12:29:10 -0700 (PDT)
Received: by 10.224.42.68 with SMTP id r4mr8165331qae.4.1347910150381;
Mon, 17 Sep 2012 12:29:10 -0700 (PDT)
Received: from mail-qa0-f42.google.com (mail-qa0-f42.google.com [209.85.216.42])
by gmr-mx.google.com with ESMTPS id fe27si2804250qcb.0.2012.09.17.12.29.09
(version=TLSv1/SSLv3 cipher=OTHER);
Mon, 17 Sep 2012 12:29:09 -0700 (PDT)
Received-SPF: pass (google.com: domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.42 as permitted sender) client-ip=209.85.216.42;
Received: by qaeb19 with SMTP id b19so2759365qae.1
for <golang-dev-/***@public.gmane.org>; Mon, 17 Sep 2012 12:29:09 -0700 (PDT)
Received: by 10.229.136.84 with SMTP id q20mr8031204qct.80.1347910149211;
Mon, 17 Sep 2012 12:29:09 -0700 (PDT)
Received: by 10.229.136.84 with SMTP id q20mr8031195qct.80.1347910149049; Mon,
17 Sep 2012 12:29:09 -0700 (PDT)
Sender: golang-dev-/***@public.gmane.org
Received: by 10.224.174.212 with HTTP; Mon, 17 Sep 2012 12:28:37 -0700 (PDT)
In-Reply-To: <CAA8EjDT22-2kRv=8V18W+ScOiZ2dT6Z00EP+_yj3dkdPGsxkPA-JsoAwUIsXosN+***@public.gmane.org>
X-System-Of-Record: true
X-Gm-Message-State: ALoCoQn/sqQMM/419ZHdZTD8AVCd31c6C3jzSMaNQr4xBkZP3kfm0iJzyRMVDILX+F5nMw68FkUxtyke80k0WEzB23GlXANoKoYZ+rfF6BhTa+mKq3ycGvqwgkSigxUmOL38mXZbH/oNEFRnk0DkBurztQn7nj9iaBKt5wPrqEgUdm1GUCRmLYKWEWgayB7g2quvpn/JPAjszP5s8dojfuAGvFrKo9vhkA==
X-Original-Sender: adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.42 as permitted sender)
smtp.mail=adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org; dkim=pass header.i=@google.com
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45698>
Post by Russ Cox
I looked into the net/http test failure in the go103-test-1 repo.
I don't believe 305f67dc3f99 is correct. Can you drop it from the go103 list?
With this removed

305f67dc3f99 10 2: +86/-10 net/http: make client await response
concurrently with writing request

I also need to remove these two

26b62d1db69f 10 1: +17/-13 net/http: fix send on close channel error
81f45da9df0d 10 2: +91/-9 net/http: fix Transport race(s) with high GOMAXPROCS

which seems fine, to me. It's no worse than 1.0.2 at least.

Objections from Brad?

Andrew
Russ Cox
2012-09-17 19:31:24 UTC
Permalink
Received: by 10.52.90.69 with SMTP id bu5mr964448vdb.6.1347910285873;
Mon, 17 Sep 2012 12:31:25 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.52.92.207 with SMTP id co15ls1417161vdb.5.gmail; Mon, 17 Sep
2012 12:31:25 -0700 (PDT)
Received: by 10.58.132.238 with SMTP id ox14mr2643563veb.15.1347910285367;
Mon, 17 Sep 2012 12:31:25 -0700 (PDT)
Received: by 10.58.132.238 with SMTP id ox14mr2643562veb.15.1347910285358;
Mon, 17 Sep 2012 12:31:25 -0700 (PDT)
Received: from mail-vb0-f43.google.com (mail-vb0-f43.google.com [209.85.212.43])
by gmr-mx.google.com with ESMTPS id r14si439006vdu.1.2012.09.17.12.31.24
(version=TLSv1/SSLv3 cipher=OTHER);
Mon, 17 Sep 2012 12:31:24 -0700 (PDT)
Received-SPF: pass (google.com: domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.212.43 as permitted sender) client-ip=209.85.212.43;
Received: by vbbfq11 with SMTP id fq11so5463471vbb.30
for <golang-dev-/***@public.gmane.org>; Mon, 17 Sep 2012 12:31:24 -0700 (PDT)
Received: by 10.52.100.5 with SMTP id eu5mr3942533vdb.34.1347910284178; Mon,
17 Sep 2012 12:31:24 -0700 (PDT)
Received: by 10.58.69.105 with HTTP; Mon, 17 Sep 2012 12:31:24 -0700 (PDT)
In-Reply-To: <CAHxS-ovXVQHEsAVQ8H=0Jmon_tWZvkH78GbwK1Ph4L7JL+1atQ-JsoAwUIsXosN+***@public.gmane.org>
X-Gm-Message-State: ALoCoQkoNf4X/Q+l0Zj9Cu58i8W21pNyKaj3ZqJUvAAvJwyHlSsKvF2HbDuB4rf80VazLOSmtgT7
X-Original-Sender: rsc-iFWiy5xATs8dnm+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.212.43 as permitted sender) smtp.mail=rsc-iFWiy5xATs8dnm+***@public.gmane.org
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
Sender: golang-dev-/***@public.gmane.org
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45700>

lgtm
Andrew Gerrand
2012-09-17 19:41:13 UTC
Permalink
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=google.com; s=20120113;
h=x-beenthere:received-spf:mime-version:sender:in-reply-to:references
:from:date:message-id:subject:to:cc:x-system-of-record
:x-gm-message-state:x-original-sender
:x-original-authentication-results:precedence:mailing-list:list-id
:x-google-group-id:list-post:list-help:list-archive:list-subscribe
:list-unsubscribe:content-type;
bh=1KuIENptpKFidn2aXaW/8fXTfE2GtD6bE3EaVhFdSnE=;
b=CFgklMAjWXfsSBbPilJr9BAGi0qrF2yGIosCU2hfO3frJ7PgpIV9SaJjtZFw603Iov
c0O1wv/winDFG7ofTJBPjn4EviXrgLpgVRYEUAWwR3NBrB5RbIXhYyYEK1Iaj6Xrf1z4
dHBX6M6+vGLszM7bi5CBeg9al/ELQ1P6LZemij4vkErHFb9gtrtv+fEL/BO2lY/p8y1L
0TYD7csYwR1H+5EHCIaHiBIH3XjWE9S8G2Jv4xglVVy2v10YT6pj2DqYEZEKN8waYdGu
Y0yokGTKUz5RhIBL5KgtHuKs+iCQ7YOw1v31Bl1kaiBl
Received: by 10.236.185.130 with SMTP id u2mr1946123yhm.0.1347910904565;
Mon, 17 Sep 2012 12:41:44 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.236.154.232 with SMTP id h68ls4351543yhk.2.gmail; Mon, 17 Sep
2012 12:41:43 -0700 (PDT)
Received: by 10.236.115.226 with SMTP id e62mr8386812yhh.34.1347910903631;
Mon, 17 Sep 2012 12:41:43 -0700 (PDT)
Received: by 10.236.115.226 with SMTP id e62mr8386811yhh.34.1347910903621;
Mon, 17 Sep 2012 12:41:43 -0700 (PDT)
Received: from mail-qc0-f174.google.com (mail-qc0-f174.google.com [209.85.216.174])
by gmr-mx.google.com with ESMTPS id fe27si2815893qcb.0.2012.09.17.12.41.43
(version=TLSv1/SSLv3 cipher=OTHER);
Mon, 17 Sep 2012 12:41:43 -0700 (PDT)
Received-SPF: pass (google.com: domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.174 as permitted sender) client-ip=209.85.216.174;
Received: by qcro28 with SMTP id o28so5516506qcr.33
for <golang-dev-/***@public.gmane.org>; Mon, 17 Sep 2012 12:41:43 -0700 (PDT)
Received: by 10.224.176.132 with SMTP id be4mr30104693qab.45.1347910903507;
Mon, 17 Sep 2012 12:41:43 -0700 (PDT)
Received: by 10.224.176.132 with SMTP id be4mr30104679qab.45.1347910903400;
Mon, 17 Sep 2012 12:41:43 -0700 (PDT)
Sender: golang-dev-/***@public.gmane.org
Received: by 10.224.174.212 with HTTP; Mon, 17 Sep 2012 12:41:13 -0700 (PDT)
In-Reply-To: <CAA8EjDSGD=ckEMyDYjTtxPUJqehJMffuQ=eZ_jjp+vYXALn3Jw-JsoAwUIsXosN+***@public.gmane.org>
X-System-Of-Record: true
X-Gm-Message-State: ALoCoQmQ5DdHq7IWlD9fugFv9ltO/2gqO3gECCeEd1qsrLoTxotkqdTRbEUTA4z5Z/gaIu9ETAngu3MxyLHRqK5wBuhvc5A7chE5xSUcrEPkSx1Hm0LtJzgxKOr75A2zcb7kx+ADohLOkGmrCnCyQb3zr5bv5gkfAb2yKU3za4rm6d+mHSEeH4Y7iDfMwUzH9Aqgq7y2YSS6Jtfv5FW/4cDgWtvsvQhDqA==
X-Original-Sender: adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.174 as permitted sender)
smtp.mail=adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org; dkim=pass header.i=@google.com
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45703>

Candidate repo here:

https://code.google.com/r/adg-go103-test-3
Andrew Gerrand
2012-09-17 19:40:26 UTC
Permalink
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=google.com; s=20120113;
h=x-beenthere:received-spf:mime-version:sender:in-reply-to:references
:from:date:message-id:subject:to:cc:x-system-of-record
:x-gm-message-state:x-original-sender
:x-original-authentication-results:precedence:mailing-list:list-id
:x-google-group-id:list-post:list-help:list-archive:list-subscribe
:list-unsubscribe:content-type;
bh=ZRV6fpJFrJwvo4MgbcGdbGvRxtAlfltesLqDtTrO7Ls=;
b=Yxzm5YkhNW10oRAyzlebxzA9KpvrZ28/zrdKVEO5ydBbh1XlF7Z+hz7rx7g4aQV3dg
1yFIo/RYAlTHGBtBmGpRM97z+8+Y4MMdH0fSAWnhR6t5G8Z3XVdxZPtBYjOPUwyL/xqo
p+L2Dgb5yGPqmRSOXDqDeuMG74U9CJ0DDP66v+vA0V8cXmUEpAC49lHjqRzOEKDddQ7O
CmpSG00zp2KSY6CFn1jBZtJJA1RaR4Vtn539/Txjxf81pjEeNaUv7YyTescLQzJFqGej
qXBfGsvLnmEtzcW8J8xETRAm6PVxMhBqMaku+TtILSwn
Received: by 10.52.28.83 with SMTP id z19mr996751vdg.20.1347910859010;
Mon, 17 Sep 2012 12:40:59 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.52.156.112 with SMTP id wd16ls1475174vdb.6.gmail; Mon, 17 Sep
2012 12:40:58 -0700 (PDT)
Received: by 10.224.42.68 with SMTP id r4mr8189961qae.4.1347910858377;
Mon, 17 Sep 2012 12:40:58 -0700 (PDT)
Received: by 10.224.42.68 with SMTP id r4mr8189960qae.4.1347910858369;
Mon, 17 Sep 2012 12:40:58 -0700 (PDT)
Received: from mail-qa0-f45.google.com (mail-qa0-f45.google.com [209.85.216.45])
by gmr-mx.google.com with ESMTPS id mx3si2813466qcb.3.2012.09.17.12.40.57
(version=TLSv1/SSLv3 cipher=OTHER);
Mon, 17 Sep 2012 12:40:57 -0700 (PDT)
Received-SPF: pass (google.com: domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.45 as permitted sender) client-ip=209.85.216.45;
Received: by mail-qa0-f45.google.com with SMTP id c10so1833746qad.4
for <golang-dev-/***@public.gmane.org>; Mon, 17 Sep 2012 12:40:57 -0700 (PDT)
Received: by 10.229.135.12 with SMTP id l12mr8046406qct.111.1347910857148;
Mon, 17 Sep 2012 12:40:57 -0700 (PDT)
Received: by 10.229.135.12 with SMTP id l12mr8046388qct.111.1347910856895;
Mon, 17 Sep 2012 12:40:56 -0700 (PDT)
Sender: golang-dev-/***@public.gmane.org
Received: by 10.224.174.212 with HTTP; Mon, 17 Sep 2012 12:40:26 -0700 (PDT)
In-Reply-To: <CAHxS-ovXVQHEsAVQ8H=0Jmon_tWZvkH78GbwK1Ph4L7JL+1atQ-JsoAwUIsXosN+***@public.gmane.org>
X-System-Of-Record: true
X-Gm-Message-State: ALoCoQkzf2Cql0ysNy3zgqvKc6Yn/zrgmMo1BqpiRQE6jJJDzKCsbc2ImHM22Yd2PuqszS4jmWlpK8e6kOiq3S4P/1cPFRM5xkXXqrHJoIvz/LEfzDb7arkF8Bm3h9l4oSQjHx99UlibQ08ue70XKgLSdIOfPD8KcSjJNH0Y08XHsPBCfSEJZtWYGtpNUHSiBOnPGY4VjdncUOl9YYxSRP3x3DnMmMS2VQ==
X-Original-Sender: adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.216.45 as permitted sender)
smtp.mail=adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org; dkim=pass header.i=@google.com
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45702>
Post by Andrew Gerrand
Objections from Brad?
Brad says ok.
m***@public.gmane.org
2012-09-16 17:03:12 UTC
Permalink
https://codereview.appspot.com/6499087/diff/14049/misc/backport/go103-excluded
File misc/backport/go103-excluded (right):

https://codereview.appspot.com/6499087/diff/14049/misc/backport/go103-excluded#newcode91
misc/backport/go103-excluded:91: aeaab9df5600 -1 1: +6/-1 cmd/cgo: use 1
as last entry for __cgodebug_data
i recommend taking this change.

it trivially fixed a real cgo bug when using newer versions of Xcode
gcc.

https://codereview.appspot.com/6499087/diff/14049/misc/backport/go103-excluded#newcode133
misc/backport/go103-excluded:133: 7c5f4dad8fae -1 2: +19/-11 pprof: make
it work on windows again
i also recommend taking this CL.

quite some people are asking how to use pprof on Windows, and
this CL will surely help them.

https://codereview.appspot.com/6499087/diff/14049/misc/backport/go103-excluded#newcode134
misc/backport/go103-excluded:134: 80be10893a8b -1 4: +24/-40 syscall:
extract an ExampleLoadLibrary from comment
this is a pure docs change, taking it won't do any harm.

https://codereview.appspot.com/6499087/
r***@public.gmane.org
2012-09-16 21:22:32 UTC
Permalink
Post by Russ Cox
Remy, do you have any measurements of the performance difference (say,
on test/bench/go1) if we apply 6497136. I'm a little worried.
The fannkuch benchmark is quite sensible to disabling the movprop in
that case. What do you advise? It feels awkward to have a "Known issues"
section in release notes.

benchmark old ns/op new ns/op delta
BenchmarkBinaryTree17 9310850000 9207829000 -1.11%
BenchmarkFannkuch11 8313733000 9494323000 +14.20%
BenchmarkGobDecode 30859680 31051680 +0.62%
BenchmarkGobEncode 17888400 18581720 +3.88%
BenchmarkGzip 299812 298566 -0.42%
BenchmarkGunzip 17710 17594 -0.65%
BenchmarkJSONEncode 114687650 117321600 +2.30%
BenchmarkJSONDecode 438836200 442206000 +0.77%
BenchmarkRevcomp25M 1796440000 1742657000 -2.99%
BenchmarkTemplate 663305400 652331200 -1.65%

benchmark old MB/s new MB/s speedup
BenchmarkGobDecode 24.87 24.72 0.99x
BenchmarkGobEncode 42.91 41.31 0.96x
BenchmarkJSONEncode 16.92 16.54 0.98x
BenchmarkJSONDecode 4.42 4.39 0.99x
BenchmarkRevcomp25M 141.48 145.85 1.03x
BenchmarkTemplate 2.93 2.97 1.01x


http://codereview.appspot.com/6499087/
Russ Cox
2012-09-17 18:36:20 UTC
Permalink
Received: by 10.236.154.194 with SMTP id h42mr1888619yhk.8.1347906983269;
Mon, 17 Sep 2012 11:36:23 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.236.124.176 with SMTP id x36ls4193346yhh.1.gmail; Mon, 17 Sep
2012 11:36:21 -0700 (PDT)
Received: by 10.58.33.234 with SMTP id u10mr2568371vei.28.1347906981166;
Mon, 17 Sep 2012 11:36:21 -0700 (PDT)
Received: by 10.58.33.234 with SMTP id u10mr2568370vei.28.1347906981157;
Mon, 17 Sep 2012 11:36:21 -0700 (PDT)
Received: from mail-vb0-f47.google.com (mail-vb0-f47.google.com [209.85.212.47])
by gmr-mx.google.com with ESMTPS id r14si431462vdu.1.2012.09.17.11.36.21
(version=TLSv1/SSLv3 cipher=OTHER);
Mon, 17 Sep 2012 11:36:21 -0700 (PDT)
Received-SPF: pass (google.com: domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.212.47 as permitted sender) client-ip=209.85.212.47;
Received: by mail-vb0-f47.google.com with SMTP id ez10so2686726vbb.6
for <golang-dev-/***@public.gmane.org>; Mon, 17 Sep 2012 11:36:21 -0700 (PDT)
Received: by 10.220.214.208 with SMTP id hb16mr7945245vcb.56.1347906980961;
Mon, 17 Sep 2012 11:36:20 -0700 (PDT)
Received: by 10.58.69.105 with HTTP; Mon, 17 Sep 2012 11:36:20 -0700 (PDT)
In-Reply-To: <047d7b5d46804d660904c9d8416a-hpIqsD4AKlfQT0dZR+***@public.gmane.org>
X-Gm-Message-State: ALoCoQkWHL+Xnf1h00yRCuAUr1tb/YTa/pf1/9lttr34mabfHP+MydPOSsaSXvlETQOgIyvhLCod
X-Original-Sender: rsc-iFWiy5xATs8dnm+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of rsc-iFWiy5xATs8dnm+***@public.gmane.org designates 209.85.212.47 as permitted sender) smtp.mail=rsc-iFWiy5xATs8dnm+***@public.gmane.org
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
Sender: golang-dev-/***@public.gmane.org
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/45688>

I am not very worried about this bug. This has been in Go since day 1
and yet no one has ever reported it. I agree that we should fix it but
I don't think we need to rush out a not-very-soaked fix as part of
1.0.3.

Russ
adg-iFWiy5xATs8dnm+
2012-09-18 15:03:23 UTC
Permalink
https://codereview.appspot.com/6499087/diff/14049/misc/backport/go103-excluded
File misc/backport/go103-excluded (right):

https://codereview.appspot.com/6499087/diff/14049/misc/backport/go103-excluded#newcode91
misc/backport/go103-excluded:91: aeaab9df5600 -1 1: +6/-1 cmd/cgo: use 1
as last entry for __cgodebug_data
Post by m***@public.gmane.org
i recommend taking this change.
it trivially fixed a real cgo bug when using newer versions of Xcode
gcc.

Done.

https://codereview.appspot.com/6499087/diff/14049/misc/backport/go103-excluded#newcode133
misc/backport/go103-excluded:133: 7c5f4dad8fae -1 2: +19/-11 pprof: make
it work on windows again
Post by m***@public.gmane.org
i also recommend taking this CL.
quite some people are asking how to use pprof on Windows, and
this CL will surely help them.
Done.

https://codereview.appspot.com/6499087/diff/14049/misc/backport/go103-excluded#newcode134
misc/backport/go103-excluded:134: 80be10893a8b -1 4: +24/-40 syscall:
extract an ExampleLoadLibrary from comment
Post by m***@public.gmane.org
this is a pure docs change, taking it won't do any harm.
Fails to merge automatically and it's not worth the time to merge it
manually. Sorry.

https://codereview.appspot.com/6499087/
Andrew Gerrand
2012-09-21 20:41:16 UTC
Permalink
I've pushed this to the main repo and the builders are reporting that
windows is broken. Something wrong with net. I will investigate but if any
windows people can suggest a fix I'd appreciate it.

Andrew
Andrew Gerrand
2012-09-21 21:19:22 UTC
Permalink
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=google.com; s=20120113;
h=x-beenthere:received-spf:mime-version:sender:in-reply-to:references
:from:date:message-id:subject:to:cc:x-system-of-record
:x-gm-message-state:x-original-sender
:x-original-authentication-results:precedence:mailing-list:list-id
:x-google-group-id:list-post:list-help:list-archive:list-subscribe
:list-unsubscribe:content-type;
bh=xl7U/FcLrhhmRYWn0GUFFptHl95GvciKMP+82AihDxo=;
b=RIfCBqksN5mGV38vhZn7+jnwjMmBMbhG11SNB06LV7MJ2+vxYMvU8l8PjLDz8VkNpk
cFkmchwqmyO8ZwaVTOniwhQcrC5jG1fRb0iOYiHI0YQjcs0XJET3ZvWPVvozFeMaZbW8
Buvxc/oXVxJ9eVvtZVa+xNztJ6/ZhClYdEVLVwZYmmbp5IhRJRvMUzX2RiSFxlvkiu8j
UdnKqAIfISJJ2KtOVZS/i9Np6FojTXGAYLfkzxdmekSaIixipP8BzRj/HzS/Y5Rjt5T2
zGoS+eoM8BnpQiYzfKq+OkwQkfh7jf/ZHuvLIO89bXd2
Received: by 10.52.21.44 with SMTP id s12mr1281546vde.5.1348262394752;
Fri, 21 Sep 2012 14:19:54 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.220.148.143 with SMTP id p15ls3180223vcv.9.gmail; Fri, 21 Sep
2012 14:19:53 -0700 (PDT)
Received: by 10.58.155.100 with SMTP id vv4mr1137426veb.12.1348262393830;
Fri, 21 Sep 2012 14:19:53 -0700 (PDT)
Received: by 10.58.155.100 with SMTP id vv4mr1137425veb.12.1348262393819;
Fri, 21 Sep 2012 14:19:53 -0700 (PDT)
Received: from mail-vc0-f173.google.com (mail-vc0-f173.google.com [209.85.220.173])
by gmr-mx.google.com with ESMTPS id ef10si570147vdb.3.2012.09.21.14.19.52
(version=TLSv1/SSLv3 cipher=OTHER);
Fri, 21 Sep 2012 14:19:52 -0700 (PDT)
Received-SPF: pass (google.com: domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.220.173 as permitted sender) client-ip=209.85.220.173;
Received: by vcbfl15 with SMTP id fl15so5370410vcb.18
for <golang-dev-/***@public.gmane.org>; Fri, 21 Sep 2012 14:19:52 -0700 (PDT)
Received: by 10.220.214.208 with SMTP id hb16mr3686731vcb.56.1348262392692;
Fri, 21 Sep 2012 14:19:52 -0700 (PDT)
Received: by 10.220.214.208 with SMTP id hb16mr3686724vcb.56.1348262392599;
Fri, 21 Sep 2012 14:19:52 -0700 (PDT)
Sender: golang-dev-/***@public.gmane.org
Received: by 10.220.239.135 with HTTP; Fri, 21 Sep 2012 14:19:22 -0700 (PDT)
In-Reply-To: <CAHxS-ovV68spZMt-dDEJTLQcf+SkVrEM0cDJ24O33UAyqup=DQ-JsoAwUIsXosN+***@public.gmane.org>
X-System-Of-Record: true
X-Gm-Message-State: ALoCoQl2X4oxehdKRHsK4nEhg/KUbT/im0tjVPDjY3oUvnlDXZDSkpNuN2s/3Q/KsEJQc5hcSMNHBslS8fJT0gHN1YrYSBB0Um0Lb7xQ4Im8mGcXWa/Ern5e6zyanmMifuukZZBEXsaFt8VcJMB5Rgiys7jd59fEPgzv9LA5aL/wDY7beVG93Svh3ebUsthH9VM0DPNsd+Y3pzn31dWw1o/Hf402wkwCbA==
X-Original-Sender: adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.220.173 as permitted sender)
smtp.mail=adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org; dkim=pass header.i=@google.com
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/46310>
Post by Andrew Gerrand
I've pushed this to the main repo and the builders are reporting that
windows is broken. Something wrong with net. I will investigate but if any
windows people can suggest a fix I'd appreciate it.
I see the problem. I neglected to backport
golang.org/change/a834a940704c which disables the failing test under
windows.

Will add it to the branch now.

There is an unrelated issue in the syscall windows tests which I will
fix in a separate change.
Andrew Gerrand
2012-09-21 22:22:46 UTC
Permalink
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=google.com; s=20120113;
h=x-beenthere:received-spf:mime-version:sender:in-reply-to:references
:from:date:message-id:subject:to:cc:x-system-of-record
:x-gm-message-state:x-original-sender
:x-original-authentication-results:precedence:mailing-list:list-id
:x-google-group-id:list-post:list-help:list-archive:list-subscribe
:list-unsubscribe:content-type;
bh=yjQYnzC3udznNorTJRJWoJu4pl8JrHnyuZSrCgCjVKA=;
b=WFm19EoHJI2MoxiVFLusUOfrWlGo6H4F3hW4ZaXNJjeqBh6QH6cYWfeWV/15UwmxC5
kGmQaxGppis6oVGhNVUs/alDLsAFXkP3nROKBtm6EgX7Y5rZN00ROkLg3cvNtBJ6tri6
K6llXh/tvQVPVmED6hffEV7krz6RfYvirZ96Q28yQyX96+YGB0IS6EpgFUBGU7Ph2ZwQ
gdFeSt5BZQq2N16Eim2ToLlFjJoFAOiCmjIN/K9UyHEsi6sozdYD0htdYq42CVBkhkCw
qxJeVDiizjs8kI43xJZdohUCVGscQCuXRmfo6G89O2sQ
Received: by 10.52.66.235 with SMTP id i11mr1308292vdt.7.1348266199320;
Fri, 21 Sep 2012 15:23:19 -0700 (PDT)
X-BeenThere: golang-dev-/***@public.gmane.org
Received: by 10.220.148.143 with SMTP id p15ls3201572vcv.9.gmail; Fri, 21 Sep
2012 15:23:18 -0700 (PDT)
Received: by 10.58.222.100 with SMTP id ql4mr1144195vec.22.1348266198307;
Fri, 21 Sep 2012 15:23:18 -0700 (PDT)
Received: by 10.58.222.100 with SMTP id ql4mr1144194vec.22.1348266198297;
Fri, 21 Sep 2012 15:23:18 -0700 (PDT)
Received: from mail-vc0-f176.google.com (mail-vc0-f176.google.com [209.85.220.176])
by gmr-mx.google.com with ESMTPS id r14si577095vdu.1.2012.09.21.15.23.17
(version=TLSv1/SSLv3 cipher=OTHER);
Fri, 21 Sep 2012 15:23:17 -0700 (PDT)
Received-SPF: pass (google.com: domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.220.176 as permitted sender) client-ip=209.85.220.176;
Received: by vcbgb22 with SMTP id gb22so5770842vcb.21
for <golang-dev-/***@public.gmane.org>; Fri, 21 Sep 2012 15:23:17 -0700 (PDT)
Received: by 10.52.75.42 with SMTP id z10mr2274227vdv.14.1348266197179;
Fri, 21 Sep 2012 15:23:17 -0700 (PDT)
Received: by 10.52.75.42 with SMTP id z10mr2274221vdv.14.1348266197068; Fri,
21 Sep 2012 15:23:17 -0700 (PDT)
Sender: golang-dev-/***@public.gmane.org
Received: by 10.220.239.135 with HTTP; Fri, 21 Sep 2012 15:22:46 -0700 (PDT)
In-Reply-To: <CAHxS-ovCjbJPTroiGXYJKmGHDXReOEhkKn3QCSf4SwfL=dMtjw-JsoAwUIsXosN+***@public.gmane.org>
X-System-Of-Record: true
X-Gm-Message-State: ALoCoQn1FuXUPC2//PeuzACHHPjYfGSwYqYFd3yJ+5GHZCUpK2METewrKwE6Xlm1bFU7FIt1kGmUTnSwsWJ0D02Hv38jPymaqFmzba9i2B0rOmykNXTb4bL93Nxuwv3uV4ye0GP+qojbkrfFgm9/DabFIwy1cn3PkSabFB63GFFLTs9PFJiWTi040rHhOQ/2bKzdj5h9vi9D+XvEyloxNumLnoVx6r++rg==
X-Original-Sender: adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org
X-Original-Authentication-Results: gmr-mx.google.com; spf=pass (google.com:
domain of adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org designates 209.85.220.176 as permitted sender)
smtp.mail=adg-hpIqsD4AKlfQT0dZR+***@public.gmane.org; dkim=pass header.i=@google.com
Precedence: list
Mailing-list: list golang-dev-/***@public.gmane.org; contact golang-dev+owners-/***@public.gmane.org
List-ID: <golang-dev.googlegroups.com>
X-Google-Group-Id: 1097896213209
List-Post: <http://groups.google.com/group/golang-dev/post?hl=en_US>, <mailto:golang-dev-/***@public.gmane.org>
List-Help: <http://groups.google.com/support/?hl=en_US>, <mailto:golang-dev+help-/***@public.gmane.org>
List-Archive: <http://groups.google.com/group/golang-dev?hl=en_US>
List-Subscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:golang-dev+subscribe-/***@public.gmane.org>
List-Unsubscribe: <http://groups.google.com/group/golang-dev/subscribe?hl=en_US>,
<mailto:googlegroups-manage+1097896213209+unsubscribe-/***@public.gmane.org>
Archived-At: <http://permalink.gmane.org/gmane.comp.lang.go.devel/46321>

Both these windows issues appear to be fixed now.
Post by Andrew Gerrand
Post by Andrew Gerrand
I've pushed this to the main repo and the builders are reporting that
windows is broken. Something wrong with net. I will investigate but if any
windows people can suggest a fix I'd appreciate it.
I see the problem. I neglected to backport
golang.org/change/a834a940704c which disables the failing test under
windows.
Will add it to the branch now.
There is an unrelated issue in the syscall windows tests which I will
fix in a separate change.
Loading...