* Kbuild: Ignore GREP_OPTIONS env variable
@ 2013-10-30 16:06 Borislav Petkov
2013-11-06 21:42 ` Michal Marek
0 siblings, 1 reply; 6+ messages in thread
From: Borislav Petkov @ 2013-10-30 16:06 UTC (permalink / raw)
To: linux-kbuild, lkml; +Cc: x86-ml
Hi,
this is just a general poll of what people think about the issue here,
just so that at least we've talked about it.
So I had defined GREP_OPTIONS=--color=always on one of my boxes and had
forgotten about it and the kernel build started failing because we use
grep quite a while in the tree and it started issuing shell color markup
which generated garbage files, like the syscall headers on x86, for
example.
I have a fix below which seems to take care of it but what is the
general opinion: Do we want to be more robust against the environment we
find on a machine before building the kernel or let the user figure it
out himself that he should be using
GREP_OPTIONS=--color=auto
in the first place and it is his own moronic fault if he does 'always'?
Opinions, comments?
Thanks.
--
From: Borislav Petkov <bp@suse.de>
Subject: [PATCH] Kbuild: Ignore GREP_OPTIONS env variable
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
When building the kernel in a shell which defines GREP_OPTIONS so that
grep behavior is modified, we can break the generation of the syscalls
table like so:
__SYSCALL_COMMON(^[[01;31m^[[K0^[[m^[[K, sys_read, sys_read)
__SYSCALL_COMMON(^[[01;31m^[[K1^[[m^[[K, sys_write, sys_write)
__SYSCALL_COMMON(^[[01;31m^[[K1^[[m^[[K0, sys_mprotect, sys_mprotect) ...
This is just the initial breakage, later we barf when generating
modules.
In this case, GREP_OPTIONS contains "--color=always" which adds the shell
colors markup and completely fudges the headers under ...generated/asm/.
Fix that by ignoring the GREP_OPTIONS variable for the whole kernel
build as we tend to use grep at a bunch of places.
Signed-off-by: Borislav Petkov <bp@suse.de>
---
Makefile | 3 +++
1 file changed, 3 insertions(+)
diff --git a/Makefile b/Makefile
index 868c0eb67b08..7a2acd9a5d1c 100644
--- a/Makefile
+++ b/Makefile
@@ -16,6 +16,9 @@ NAME = One Giant Leap for Frogkind
# o print "Entering directory ...";
MAKEFLAGS += -rR --no-print-directory
+GREP_OPTIONS=
+export GREP_OPTIONS
+
# Avoid funny character set dependencies
unexport LC_ALL
LC_COLLATE=C
--
1.8.4
--
Regards/Gruss,
Boris.
Sent from a fat crate under my desk. Formatting is fine.
--
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: Kbuild: Ignore GREP_OPTIONS env variable
2013-10-30 16:06 Kbuild: Ignore GREP_OPTIONS env variable Borislav Petkov
@ 2013-11-06 21:42 ` Michal Marek
2013-11-11 14:27 ` [PATCH -v1.1] " Borislav Petkov
0 siblings, 1 reply; 6+ messages in thread
From: Michal Marek @ 2013-11-06 21:42 UTC (permalink / raw)
To: Borislav Petkov, linux-kbuild, lkml; +Cc: x86-ml
Dne 30.10.2013 17:06, Borislav Petkov napsal(a):
> So I had defined GREP_OPTIONS=--color=always on one of my boxes and had
> forgotten about it and the kernel build started failing because we use
> grep quite a while in the tree and it started issuing shell color markup
> which generated garbage files, like the syscall headers on x86, for
> example.
>
> I have a fix below which seems to take care of it but what is the
> general opinion: Do we want to be more robust against the environment we
> find on a machine before building the kernel or let the user figure it
> out himself that he should be using
>
> GREP_OPTIONS=--color=auto
>
> in the first place and it is his own moronic fault if he does 'always'?
I think that on a scale from one to ten, building with
GREP_OPTIONS=--color=always and building with LC_COLLATE=<language with
non-ASCII sorting rules> score about the same. And are already
sanitizing the LC_* variables.
> +GREP_OPTIONS=
> +export GREP_OPTIONS
Just remove it from the enviromnet completely, like the next statement does:
> +
> # Avoid funny character set dependencies
> unexport LC_ALL
> LC_COLLATE=C
Michal
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH -v1.1] Kbuild: Ignore GREP_OPTIONS env variable
2013-11-06 21:42 ` Michal Marek
@ 2013-11-11 14:27 ` Borislav Petkov
2013-11-11 17:00 ` Michal Marek
0 siblings, 1 reply; 6+ messages in thread
From: Borislav Petkov @ 2013-11-11 14:27 UTC (permalink / raw)
To: Michal Marek; +Cc: linux-kbuild, lkml, x86-ml
From: Borislav Petkov <bp@suse.de>
Subject: [PATCH] Kbuild: Ignore GREP_OPTIONS env variable
When building the kernel in a shell which defines GREP_OPTIONS so that
grep behavior is modified, we can break the generation of the syscalls
table like so:
__SYSCALL_COMMON(^[[01;31m^[[K0^[[m^[[K, sys_read, sys_read)
__SYSCALL_COMMON(^[[01;31m^[[K1^[[m^[[K, sys_write, sys_write)
__SYSCALL_COMMON(^[[01;31m^[[K1^[[m^[[K0, sys_mprotect, sys_mprotect) ...
This is just the initial breakage, later we barf when generating
modules.
In this case, GREP_OPTIONS contains "--color=always" which adds the shell
colors markup and completely fudges the headers under ...generated/asm/.
Fix that by unexporting the GREP_OPTIONS variable for the whole kernel
build as we tend to use grep at a bunch of places.
Signed-off-by: Borislav Petkov <bp@suse.de>
---
Makefile | 3 +++
1 file changed, 3 insertions(+)
diff --git a/Makefile b/Makefile
index 868c0eb67b08..73503a7755fd 100644
--- a/Makefile
+++ b/Makefile
@@ -22,6 +22,9 @@ LC_COLLATE=C
LC_NUMERIC=C
export LC_COLLATE LC_NUMERIC
+# Avoid interference with shell env settings
+unexport GREP_OPTIONS
+
# We are using a recursive build, so we need to do a little thinking
# to get the ordering right.
#
--
1.8.4
--
Regards/Gruss,
Boris.
Sent from a fat crate under my desk. Formatting is fine.
--
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH -v1.1] Kbuild: Ignore GREP_OPTIONS env variable
2013-11-11 14:27 ` [PATCH -v1.1] " Borislav Petkov
@ 2013-11-11 17:00 ` Michal Marek
2013-11-11 20:43 ` Ingo Molnar
0 siblings, 1 reply; 6+ messages in thread
From: Michal Marek @ 2013-11-11 17:00 UTC (permalink / raw)
To: Borislav Petkov; +Cc: linux-kbuild, lkml, x86-ml
On Mon, Nov 11, 2013 at 03:27:43PM +0100, Borislav Petkov wrote:
> From: Borislav Petkov <bp@suse.de>
> Subject: [PATCH] Kbuild: Ignore GREP_OPTIONS env variable
>
> When building the kernel in a shell which defines GREP_OPTIONS so that
> grep behavior is modified, we can break the generation of the syscalls
> table like so:
>
> __SYSCALL_COMMON(^[[01;31m^[[K0^[[m^[[K, sys_read, sys_read)
> __SYSCALL_COMMON(^[[01;31m^[[K1^[[m^[[K, sys_write, sys_write)
> __SYSCALL_COMMON(^[[01;31m^[[K1^[[m^[[K0, sys_mprotect, sys_mprotect) ...
The email had literal escape sequences and git am preserved them, so git
show displayed red zeros and onces. That was quite scary, so I replaced
the escape character with ^[.
Anyway, applied to kbuild.git#kbuild now.
Michal
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH -v1.1] Kbuild: Ignore GREP_OPTIONS env variable
2013-11-11 17:00 ` Michal Marek
@ 2013-11-11 20:43 ` Ingo Molnar
2013-11-11 20:58 ` Borislav Petkov
0 siblings, 1 reply; 6+ messages in thread
From: Ingo Molnar @ 2013-11-11 20:43 UTC (permalink / raw)
To: Michal Marek; +Cc: Borislav Petkov, linux-kbuild, lkml, x86-ml
* Michal Marek <mmarek@suse.cz> wrote:
> On Mon, Nov 11, 2013 at 03:27:43PM +0100, Borislav Petkov wrote:
> > From: Borislav Petkov <bp@suse.de>
> > Subject: [PATCH] Kbuild: Ignore GREP_OPTIONS env variable
> >
> > When building the kernel in a shell which defines GREP_OPTIONS so that
> > grep behavior is modified, we can break the generation of the syscalls
> > table like so:
> >
> > __SYSCALL_COMMON(^[[01;31m^[[K0^[[m^[[K, sys_read, sys_read)
> > __SYSCALL_COMMON(^[[01;31m^[[K1^[[m^[[K, sys_write, sys_write)
> > __SYSCALL_COMMON(^[[01;31m^[[K1^[[m^[[K0, sys_mprotect, sys_mprotect) ...
>
> The email had literal escape sequences and git am preserved them, so git
> show displayed red zeros and onces. [..]
Neat trick if it was intentionally!
> [...] That was quite scary, [...]
Scary if it was unintentional ;-)
Thanks,
Ingo
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH -v1.1] Kbuild: Ignore GREP_OPTIONS env variable
2013-11-11 20:43 ` Ingo Molnar
@ 2013-11-11 20:58 ` Borislav Petkov
0 siblings, 0 replies; 6+ messages in thread
From: Borislav Petkov @ 2013-11-11 20:58 UTC (permalink / raw)
To: Ingo Molnar; +Cc: Michal Marek, linux-kbuild, lkml, x86-ml
On Mon, Nov 11, 2013 at 09:43:30PM +0100, Ingo Molnar wrote:
> * Michal Marek <mmarek@suse.cz> wrote:
> > On Mon, Nov 11, 2013 at 03:27:43PM +0100, Borislav Petkov wrote:
> > > From: Borislav Petkov <bp@suse.de>
> > > Subject: [PATCH] Kbuild: Ignore GREP_OPTIONS env variable
> > >
> > > When building the kernel in a shell which defines GREP_OPTIONS so that
> > > grep behavior is modified, we can break the generation of the syscalls
> > > table like so:
> > >
> > > __SYSCALL_COMMON(^[[01;31m^[[K0^[[m^[[K, sys_read, sys_read)
> > > __SYSCALL_COMMON(^[[01;31m^[[K1^[[m^[[K, sys_write, sys_write)
> > > __SYSCALL_COMMON(^[[01;31m^[[K1^[[m^[[K0, sys_mprotect, sys_mprotect) ...
> >
> > The email had literal escape sequences and git am preserved them, so git
> > show displayed red zeros and onces. [..]
>
> Neat trick if it was intentionally!
>
> > [...] That was quite scary, [...]
>
> Scary if it was unintentional ;-)
Yeah, all our computers are bleeding :-) Well, I sent the first version
on the evening before Helloween...
<scary music playing, an old door is creaking...>
--
Regards/Gruss,
Boris.
Sent from a fat crate under my desk. Formatting is fine.
--
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2013-11-11 20:59 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-10-30 16:06 Kbuild: Ignore GREP_OPTIONS env variable Borislav Petkov
2013-11-06 21:42 ` Michal Marek
2013-11-11 14:27 ` [PATCH -v1.1] " Borislav Petkov
2013-11-11 17:00 ` Michal Marek
2013-11-11 20:43 ` Ingo Molnar
2013-11-11 20:58 ` Borislav Petkov
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.