[request-sponsor] Requesting sponsor for CR #6799167 ("real gcc build fails in libshell") ...
"Roger A. Faulkner" wrote:
> > "Roger A. Faulkner" wrote:
> > > > Roland Mainz wrote:
> > > > > This is a sponsor request to "fix" ([1]) CR #6799167
> > > > > (http://bugs.opensolaris.org/view_bug.do?bug_id=6799167 - "real gcc
> > > > > build fails in libshell").
> > > > >
> > > > > [1]=The "fix" is more a workaround since I don't know exactly how the
> > > > > ctf*-tools can be fixed.
> > > > >
> > > > > My contributor ID is "OS0025".
> > > > >
> > > > > Webrev follows later when my "hg clone" is complete...
> > > >
> > > > Webrev is now available at
> > > > http://cr.opensolaris.org/~gisburn/ksh93_integration_cr_6799167_001/ -
> > > > can anyone give me a short code review for the change, please (we'll
> > > > file a new bug in bugster to make sure the original issues doesn't get
> > > > forgotten) ?
> > >
> > > The code change looks good to me (and compiles for me).
> > > (I've been concerned about the not-buildable-with-gcc
> > > issue for some time now. I thought I was the only one.)
> >
> > It was AFAIK the only one (there are other VLA users in the tree but
> > they all use the arithmetric expressions and therefore avoid this
> > xxx@@@!!!-trap).
> >
> > > Nit:
> > > Could you break the comment into more than one line,
> > > to make it fit in 80-columns, something like:
> > >
> > > /*
> > > * Allocate one extra array entry as workaround for:
> > > * CR 6799167 real gcc build fails in libshell
> > > */
> >
> > Fixed. I've uploaded the new webrev as
> > http://cr.opensolaris.org/~gisburn/ksh93_integration_cr_6799167_002/
>
> Beautiful. Thanks.
For the log:
I've filed CR #6811985 ('"ctfconvert" fails when gcc x86 executable uses
"VLA" (="Variable Length Array")') to get the bug in "ctfconvert" fixed
(see attached email, too).
Bye,
Roland
--
__ . . __
(o.\ \/ /.o) roland.mainz at nrubsig.org
\__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer
/O /==\ O\ TEL +49 641 3992797
(;O/ \/ \O;)
-- next part --
An embedded message was scrubbed...
From: bugmail-sen...@sun.com
Subject: CR 6811985 Created, P2 opensolaris/triage-queue "ctfconvert" fails
when gcc x86 executable uses "VLA" (="Variable Length Array")
Date: Mon, 2 Mar 2009 11:51:47 -0700 (MST)
Size: 6017
URL:
[request-sponsor] Requesting sponsor for CR #6799167 ("real gcc build fails in libshell") ...
John Sonnenschein wrote: > On 2-Mar-09, at 10:05 AM, Roger A. Faulkner wrote: > >> Roland Mainz wrote: [snip] > > Nit: > > Could you break the comment into more than one line, > > to make it fit in 80-columns, something like: > > > >/* > > * Allocate one extra array entry as workaround for: > > * CR 6799167 real gcc build fails in libshell > > */ > > Sorry to butt in here, but I was under the impression that code should > never make reference to the bug it fixes. Am I mistaken here? No, you're right... but AFAIK there is an exception when you add a workaround which is expected to be removed (at least there is sufficient precedent in usr/src/uts/ for this kind of comments). IMO no _permanent_ code should contain such comments but the workarounds need such stuff (to avoid excessively large comment blocks) that there is a clear sign why it was done this way... Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 3992797 (;O/ \/ \O;)
[request-sponsor] Requesting sponsor for CR #6799167 ("real gcc build fails in libshell") ...
"Roger A. Faulkner" wrote: > > Roland Mainz wrote: > > > This is a sponsor request to "fix" ([1]) CR #6799167 > > > (http://bugs.opensolaris.org/view_bug.do?bug_id=6799167 - "real gcc > > > build fails in libshell"). > > > > > > [1]=The "fix" is more a workaround since I don't know exactly how the > > > ctf*-tools can be fixed. > > > > > > My contributor ID is "OS0025". > > > > > > Webrev follows later when my "hg clone" is complete... > > > > Webrev is now available at > > http://cr.opensolaris.org/~gisburn/ksh93_integration_cr_6799167_001/ - > > can anyone give me a short code review for the change, please (we'll > > file a new bug in bugster to make sure the original issues doesn't get > > forgotten) ? > > The code change looks good to me (and compiles for me). > (I've been concerned about the not-buildable-with-gcc > issue for some time now. I thought I was the only one.) It was AFAIK the only one (there are other VLA users in the tree but they all use the arithmetric expressions and therefore avoid this xxx@@@!!!-trap). > Nit: > Could you break the comment into more than one line, > to make it fit in 80-columns, something like: > > /* > * Allocate one extra array entry as workaround for: > * CR 6799167 real gcc build fails in libshell > */ Fixed. I've uploaded the new webrev as http://cr.opensolaris.org/~gisburn/ksh93_integration_cr_6799167_002/ Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 3992797 (;O/ \/ \O;)
[request-sponsor] Requesting sponsor for CR #6799167 ("real gcc build fails in libshell") ...
On Mon, Mar 02, 2009 at 11:01:54AM -0800, Garrett D'Amore wrote: > John Sonnenschein wrote: > >Sorry to butt in here, but I was under the impression that code should > >never make reference to the bug it fixes. Am I mistaken here? > > Normally yes. However in this case, I think since the "workaround" may > need some explanation, its not a bad idea to reference it. The correct CR is: 6379193 ctfconvert needs to support VLA for gcc's DWARF (which I just duped the new one to) regards john
[request-sponsor] Requesting sponsor for CR #6799167 ("real gcc build fails in libshell") ...
Roland Mainz wrote: > This is a sponsor request to "fix" ([1]) CR #6799167 > (http://bugs.opensolaris.org/view_bug.do?bug_id=6799167 - "real gcc > build fails in libshell"). > > [1]=The "fix" is more a workaround since I don't know exactly how the > ctf*-tools can be fixed. > > My contributor ID is "OS0025". > > Webrev follows later when my "hg clone" is complete... Webrev is now available at http://cr.opensolaris.org/~gisburn/ksh93_integration_cr_6799167_001/ - can anyone give me a short code review for the change, please (we'll file a new bug in bugster to make sure the original issues doesn't get forgotten) ? Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 3992797 (;O/ \/ \O;)
[request-sponsor] Requesting sponsor for CR #6799167("real gcc build fails in libshell") ...
Milan Jurik wrote:
> Roland Mainz pe v po 02. 03. 2009 v 05:38 +0100:
> > This is a sponsor request to "fix" ([1]) CR #6799167
> > (http://bugs.opensolaris.org/view_bug.do?bug_id=6799167 - "real gcc
> > build fails in libshell").
> >
> > [1]=The "fix" is more a workaround since I don't know exactly how the
> > ctf*-tools can be fixed.
> >
> > My contributor ID is "OS0025".
> >
> > Webrev follows later when my "hg clone" is complete...
>
> I will sponsor it (if it's ksh93 issue).
It's an issue in the ctf*-tools but we failed to find a fix for these
tools for more than a month and IMO breaking the gcc build for such a
long time IMO not acceptable. Since I don't know how to fix the ctf
tools the patch in
http://cr.opensolaris.org/~gisburn/ksh93_integration_cr_6799167_001/
works around the problem by using an arithmetric expression in the VLA
declaration which seems to work in this case.
To explain the workaround: The following testcase causes "ctfconvert" to
fail with the error "die 8002: failed to get unsigned (form 0x13)":
-- snip --
#include
#include
#include
int main(int ac, char *av[])
{
int len=strlen(av[0])+1;
char name[len];
strcpy(name, av[0]);
printf("name='%s'\n", name);
return EXIT_SUCCESS;
}
-- snip --
The problem goes away if I replace the first two lines of the |main()|
function like this...
-- snip --
int len=strlen(av[0]);
char name[len+1];
-- snip --
... e.g. VLA declarations seem to work if you use an arithemtric
expression instead of a plain variable... but I can't explain _why_ this
works... ;-(
Bye,
Roland
--
__ . . __
(o.\ \/ /.o) roland.mainz at nrubsig.org
\__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer
/O /==\ O\ TEL +49 641 3992797
(;O/ \/ \O;)
[request-sponsor] Requesting sponsor for CR #6799167 ("real gcc build fails in libshell") ...
John Sonnenschein wrote:
> Sorry to butt in here, but I was under the impression that code should
> never make reference to the bug it fixes. Am I mistaken here?
Normally yes. However in this case, I think since the "workaround" may
need some explanation, its not a bad idea to reference it.
- Garrett
>
> -JohnS
> On 2-Mar-09, at 10:05 AM, Roger A. Faulkner wrote:
>
>>
>>> Date: Mon, 02 Mar 2009 16:20:44 +0100
>>> From: Roland Mainz
>>> Subject: Re: [request-sponsor] Requesting sponsor for CR #6799167
>>> ("real gcc
>> build fails in libshell") ...
>>> To: "request-sponsor at opensolaris.org"
>>> , Milan
>> Jurik , ksh93-integration-discuss
>>
>>>
>>> Roland Mainz wrote:
>>>> This is a sponsor request to "fix" ([1]) CR #6799167
>>>> (http://bugs.opensolaris.org/view_bug.do?bug_id=6799167 - "real gcc
>>>> build fails in libshell").
>>>>
>>>> [1]=The "fix" is more a workaround since I don't know exactly how the
>>>> ctf*-tools can be fixed.
>>>>
>>>> My contributor ID is "OS0025".
>>>>
>>>> Webrev follows later when my "hg clone" is complete...
>>>
>>> Webrev is now available at
>>> http://cr.opensolaris.org/~gisburn/ksh93_integration_cr_6799167_001/ -
>>> can anyone give me a short code review for the change, please (we'll
>>> file a new bug in bugster to make sure the original issues doesn't get
>>> forgotten) ?
>>
>> The code change looks good to me (and compiles for me).
>> (I've been concerned about the not-buildable-with-gcc
>> issue for some time now. I thought I was the only one.)
>>
>> Nit:
>> Could you break the comment into more than one line,
>> to make it fit in 80-columns, something like:
>>
>>/*
>> * Allocate one extra array entry as workaround for:
>> * CR 6799167 real gcc build fails in libshell
>> */
>>
>> Thanks,
>> Roger
>>
>> ___
>> request-sponsor mailing list
>> request-sponsor at opensolaris.org
>
> ___
> request-sponsor mailing list
> request-sponsor at opensolaris.org
[request-sponsor] Requesting sponsor for CR #6799167 ("real gcc build fails in libshell") ...
JohnS> ... I was under the impression that code should JohnS> never make reference to the bug it fixes. Am I mistaken here? No, you are quite correct. Referring to an open bug e.g.: /* * Work around silly misfeature until 699 is fixed. */ is OK, but the comment proposed in this case is not. Something like: /* * Allocate one extra array entry to keep gcc happy. */ should suffice. -- John http://blogs.sun.com/jbeck
[request-sponsor] Requesting sponsor for CR #6799167 ("real gcc build fails in libshell") ...
Sorry to butt in here, but I was under the impression that code should
never make reference to the bug it fixes. Am I mistaken here?
-JohnS
On 2-Mar-09, at 10:05 AM, Roger A. Faulkner wrote:
>
>> Date: Mon, 02 Mar 2009 16:20:44 +0100
>> From: Roland Mainz
>> Subject: Re: [request-sponsor] Requesting sponsor for CR #6799167
>> ("real gcc
> build fails in libshell") ...
>> To: "request-sponsor at opensolaris.org" > >, Milan
> Jurik , ksh93-integration-discuss
>
>>
>> Roland Mainz wrote:
>>> This is a sponsor request to "fix" ([1]) CR #6799167
>>> (http://bugs.opensolaris.org/view_bug.do?bug_id=6799167 - "real gcc
>>> build fails in libshell").
>>>
>>> [1]=The "fix" is more a workaround since I don't know exactly how
>>> the
>>> ctf*-tools can be fixed.
>>>
>>> My contributor ID is "OS0025".
>>>
>>> Webrev follows later when my "hg clone" is complete...
>>
>> Webrev is now available at
>> http://cr.opensolaris.org/~gisburn/
>> ksh93_integration_cr_6799167_001/ -
>> can anyone give me a short code review for the change, please (we'll
>> file a new bug in bugster to make sure the original issues doesn't
>> get
>> forgotten) ?
>
> The code change looks good to me (and compiles for me).
> (I've been concerned about the not-buildable-with-gcc
> issue for some time now. I thought I was the only one.)
>
> Nit:
> Could you break the comment into more than one line,
> to make it fit in 80-columns, something like:
>
>/*
> * Allocate one extra array entry as workaround for:
> * CR 6799167 real gcc build fails in libshell
> */
>
> Thanks,
> Roger
>
> ___
> request-sponsor mailing list
> request-sponsor at opensolaris.org
[request-sponsor] Requesting sponsor for CR #6799167 ("real gcc build fails in libshell") ...
> Date: Mon, 02 Mar 2009 19:17:31 +0100
> From: Roland Mainz
> Subject: Re: [request-sponsor] Requesting sponsor for CR #6799167 ("real gcc
build fails in libshell") ...
> To: "Roger A. Faulkner"
> Cc: Milan.Jurik at sun.com, ksh93-integration-discuss at opensolaris.org,
request-sponsor at opensolaris.org
>
> "Roger A. Faulkner" wrote:
> > > Roland Mainz wrote:
> > > > This is a sponsor request to "fix" ([1]) CR #6799167
> > > > (http://bugs.opensolaris.org/view_bug.do?bug_id=6799167 - "real gcc
> > > > build fails in libshell").
> > > >
> > > > [1]=The "fix" is more a workaround since I don't know exactly how the
> > > > ctf*-tools can be fixed.
> > > >
> > > > My contributor ID is "OS0025".
> > > >
> > > > Webrev follows later when my "hg clone" is complete...
> > >
> > > Webrev is now available at
> > > http://cr.opensolaris.org/~gisburn/ksh93_integration_cr_6799167_001/ -
> > > can anyone give me a short code review for the change, please (we'll
> > > file a new bug in bugster to make sure the original issues doesn't get
> > > forgotten) ?
> >
> > The code change looks good to me (and compiles for me).
> > (I've been concerned about the not-buildable-with-gcc
> > issue for some time now. I thought I was the only one.)
>
> It was AFAIK the only one (there are other VLA users in the tree but
> they all use the arithmetric expressions and therefore avoid this
> xxx@@@!!!-trap).
>
> > Nit:
> > Could you break the comment into more than one line,
> > to make it fit in 80-columns, something like:
> >
> > /*
> > * Allocate one extra array entry as workaround for:
> > * CR 6799167 real gcc build fails in libshell
> > */
>
> Fixed. I've uploaded the new webrev as
> http://cr.opensolaris.org/~gisburn/ksh93_integration_cr_6799167_002/
Beautiful. Thanks.
Go for it,
Roger
[request-sponsor] Requesting sponsor for CR #6799167 ("real gcc build fails in libshell") ...
> Date: Mon, 02 Mar 2009 16:20:44 +0100
> From: Roland Mainz
> Subject: Re: [request-sponsor] Requesting sponsor for CR #6799167 ("real gcc
build fails in libshell") ...
> To: "request-sponsor at opensolaris.org" opensolaris.org>, Milan
Jurik , ksh93-integration-discuss
>
> Roland Mainz wrote:
> > This is a sponsor request to "fix" ([1]) CR #6799167
> > (http://bugs.opensolaris.org/view_bug.do?bug_id=6799167 - "real gcc
> > build fails in libshell").
> >
> > [1]=The "fix" is more a workaround since I don't know exactly how the
> > ctf*-tools can be fixed.
> >
> > My contributor ID is "OS0025".
> >
> > Webrev follows later when my "hg clone" is complete...
>
> Webrev is now available at
> http://cr.opensolaris.org/~gisburn/ksh93_integration_cr_6799167_001/ -
> can anyone give me a short code review for the change, please (we'll
> file a new bug in bugster to make sure the original issues doesn't get
> forgotten) ?
The code change looks good to me (and compiles for me).
(I've been concerned about the not-buildable-with-gcc
issue for some time now. I thought I was the only one.)
Nit:
Could you break the comment into more than one line,
to make it fit in 80-columns, something like:
/*
* Allocate one extra array entry as workaround for:
* CR 6799167 real gcc build fails in libshell
*/
Thanks,
Roger
[request-sponsor] Requesting sponsor for CR #6799167 ("real gcc build fails in libshell") ...
Hi, I will sponsor it (if it's ksh93 issue). Best regards, Milan Roland Mainz p??e v po 02. 03. 2009 v 05:38 +0100: > Hi! > > > > This is a sponsor request to "fix" ([1]) CR #6799167 > (http://bugs.opensolaris.org/view_bug.do?bug_id=6799167 - "real gcc > build fails in libshell"). > > [1]=The "fix" is more a workaround since I don't know exactly how the > ctf*-tools can be fixed. > > My contributor ID is "OS0025". > > Webrev follows later when my "hg clone" is complete... > > > > Bye, > Roland >
[request-sponsor] Requesting sponsor for CR #6799167 ("real gcc build fails in libshell") ...
Hi! This is a sponsor request to "fix" ([1]) CR #6799167 (http://bugs.opensolaris.org/view_bug.do?bug_id=6799167 - "real gcc build fails in libshell"). [1]=The "fix" is more a workaround since I don't know exactly how the ctf*-tools can be fixed. My contributor ID is "OS0025". Webrev follows later when my "hg clone" is complete... Bye, Roland -- __ . . __ (o.\ \/ /.o) roland.mainz at nrubsig.org \__\/\/__/ MPEG specialist, C&&JAVA&&Sun&&Unix programmer /O /==\ O\ TEL +49 641 3992797 (;O/ \/ \O;)
