headers
Bastian Blank | 8 Aug 2008 09:47

[PATCH] DocBook: don't compress mandocs

The mandocs target uses gzip to compress the resulting manpages. If this
target runs more than once (using make -jX) at the same time, gzip may
run more then once with the same set of files. It just bails out if it
is not able to find a file because a concurrent run already removed it.

Fix this by removing the gzip call and only work with uncompressed
manpages.

Signed-off-by: Bastian Blank <waldi <at> debian.org>

diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile
index 1d1b345..3fb6ef2 100644
--- a/Documentation/DocBook/Makefile
+++ b/Documentation/DocBook/Makefile
 <at>  <at>  -51,7 +51,7  <at>  <at>  mandocs: $(MAN)

 installmandocs: mandocs
 	mkdir -p /usr/local/man/man9/
-	install Documentation/DocBook/man/*.9.gz /usr/local/man/man9/
+	install Documentation/DocBook/man/*.9 /usr/local/man/man9/
 
 ###
 #External programs used
 <at>  <at>  -156,7 +156,7  <at>  <at>  quiet_cmd_db2html = HTML   $ <at> 
             cp $(PNG-$(basename $(notdir $ <at> ))) $(patsubst %.html,%,$ <at> ); fi

 quiet_cmd_db2man = MAN     $ <at> 
-      cmd_db2man = if grep -q refentry $<; then xmlto man $(XMLTOFLAGS) -o $(obj)/man $< ; gzip -f
$(obj)/man/*.9; fi
+      cmd_db2man = if grep -q refentry $<; then xmlto man $(XMLTOFLAGS) -o $(obj)/man $<; fi
(Continue reading)

Permalink | Reply | 4 comments |
headers
Sam Ravnborg | 8 Aug 2008 23:45
Gravatar

Re: [PATCH] DocBook: don't compress mandocs

On Fri, Aug 08, 2008 at 09:47:24AM +0200, Bastian Blank wrote:
> The mandocs target uses gzip to compress the resulting manpages. If this
> target runs more than once (using make -jX) at the same time, gzip may
> run more then once with the same set of files. It just bails out if it
> is not able to find a file because a concurrent run already removed it.
> 
> Fix this by removing the gzip call and only work with uncompressed
> manpages.

If it failded with gzip being run more than once before this
patch does not solve the bug - it simply makes it harder to trigger.

Anything that does not work with a make -j100 build is broken.
And please understand that we have people that uses "make -j100" when building
the kernel.

These people most likely does not do "make mandocs" but we should still fix
the bug and not try to paper it over in this way.

So try to analyse why we in the first place will run gzip in parallele and solve that.

	Sam
--
To unsubscribe from this list: send the line "unsubscribe linux-doc" in
the body of a message to majordomo <at> vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Permalink | Reply | 2 comments |
headers
Bastian Blank | 9 Aug 2008 08:42

Re: [PATCH] DocBook: don't compress mandocs

On Fri, Aug 08, 2008 at 11:45:04PM +0200, Sam Ravnborg wrote:
> On Fri, Aug 08, 2008 at 09:47:24AM +0200, Bastian Blank wrote:
> > The mandocs target uses gzip to compress the resulting manpages. If this
> > target runs more than once (using make -jX) at the same time, gzip may
> > run more then once with the same set of files. It just bails out if it
> > is not able to find a file because a concurrent run already removed it.
> > 
> > Fix this by removing the gzip call and only work with uncompressed
> > manpages.
> 
> If it failded with gzip being run more than once before this
> patch does not solve the bug - it simply makes it harder to trigger.

No, the problem does not longer exists then. Or do you see a "*" to be
expanded on runtime?

> Anything that does not work with a make -j100 build is broken.
> And please understand that we have people that uses "make -j100" when building
> the kernel.

I only use -j64 ...

> These people most likely does not do "make mandocs" but we should still fix
> the bug and not try to paper it over in this way.
> 
> So try to analyse why we in the first place will run gzip in parallele and solve that.

make mandocs
-> make a.9 b.9 c.9 d.9 ... z.9
   -> xmlto a.xml; gzip *
(Continue reading)

Permalink | Reply | 0 comments |
headers
Rafał Miłecki | 16 Aug 2008 11:17
Picon

Add info about git bisect skip in "quick git", please

Current version of quick git:
http://www.kernel.org/doc/local/git-quick.html
suggest using git-bisect log and git-bisect replay when hitting broken version.

As I can read in man of git-bisect there is much easier sollution
(without logging, replaying, guessing) simply called "skip".

Could someone update current documentation to suggest users using this
easier-skipiing way? Using log & replay caused many problems for me
(it was my first contact with git).

--

-- 
Rafał Miłecki
Permalink | Reply | 1 comment |
headers
frans | 17 Aug 2008 11:07
Picon

[TRIVIAL] [PATCH] Documentation/filesystems/ramfs-rootfs-initramfs.txt

A trivial patch on ramfs-rootfs-initramfs.txt 
First a file hello.c is created, then the file hello2.c is compiled.
Changed this to hello.c

Mailed this before to trivial <at> kernel.org only but did not get a reply and
haven't seen it applied yet, so resubmitting to a wider audience...

Signed-off-by: Frans Meulenbroeks <fransmeulenbroeks <at> gmail.com>

diff --git a/Documentation/filesystems/ramfs-rootfs-initramfs.txt b/Documentation/filesystems/ramfs-rootfs-initramfs.txt
index 7be232b..62fe9b1 100644
--- a/Documentation/filesystems/ramfs-rootfs-initramfs.txt
+++ b/Documentation/filesystems/ramfs-rootfs-initramfs.txt
 <at>  <at>  -263,7 +263,7  <at>  <at>  User Mode Linux, like so:
     sleep(999999999);
   }
   EOF
-  gcc -static hello2.c -o init
+  gcc -static hello.c -o init
   echo init | cpio -o -H newc | gzip > test.cpio.gz
   # Testing external initramfs using the initrd loading mechanism.
   qemu -kernel /boot/vmlinuz -initrd test.cpio.gz /dev/zero
--
To unsubscribe from this list: send the line "unsubscribe linux-doc" in
the body of a message to majordomo <at> vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Permalink | Reply | 2 comments |
headers
Rob Landley | 20 Aug 2008 03:55

Re: Add info about git bisect skip in "quick git", please

On Saturday 16 August 2008 04:17:52 Rafał Miłecki wrote:
> Current version of quick git:
> http://www.kernel.org/doc/local/git-quick.html
> suggest using git-bisect log and git-bisect replay when hitting broken
> version.
>
> As I can read in man of git-bisect there is much easier sollution
> (without logging, replaying, guessing) simply called "skip".

That command didn't exist when I wrote the git-quick howto.

> Could someone update current documentation to suggest users using this
> easier-skipiing way? Using log & replay caused many problems for me
> (it was my first contact with git).

The man page says this about "skip":

> But computing the commit to test may be slower afterwards and git may
> eventually not be able to tell the first bad among a bad and one or more
> "skip"ped commits.

But sure, I'll add a note to git-quick about it...

Thanks,

Rob
--

-- 
"One of my most productive days was throwing away 1000 lines of code."
  - Ken Thompson.
--
(Continue reading)

Permalink | Reply | 0 comments |
headers
Rob Landley | 20 Aug 2008 07:56

Re: [TRIVIAL] [PATCH] Documentation/filesystems/ramfs-rootfs-initramfs.txt

On Sunday 17 August 2008 04:07:04 frans wrote:
> A trivial patch on ramfs-rootfs-initramfs.txt
> First a file hello.c is created, then the file hello2.c is compiled.
> Changed this to hello.c
>
> Mailed this before to trivial <at> kernel.org only but did not get a reply and
> haven't seen it applied yet, so resubmitting to a wider audience...
>
> Signed-off-by: Frans Meulenbroeks <fransmeulenbroeks <at> gmail.com>

Signed-off-by: Rob Landley <rob <at> landley.net>

Rob
--

-- 
"One of my most productive days was throwing away 1000 lines of code."
  - Ken Thompson.
--
To unsubscribe from this list: send the line "unsubscribe linux-doc" in
the body of a message to majordomo <at> vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Permalink | Reply | 0 comments |
headers
Jari Aalto | 30 Aug 2008 09:27
Favicon
Gravatar

www.kernel.org/doc/Documentation


FYI, following pages are not accessible:

Jari

$ webcheck http://www.kernel.org/doc/Documentation/
(See http://ch.tudelft.nl/~arthur/webcheck/ )

          + Documentation
               o bad link: http://www.kernel.org/doc/Documentation/i386/: 404: Not Found
               o bad link: http://www.kernel.org/doc/Documentation/time_interpolators.txt: 404:
                 Not Found
               o bad link: http://www.kernel.org/doc/Documentation/x86_64/: 404: Not Found

          + Documentation/filesystems
               o bad link: http://www.kernel.org/doc/Documentation/filesystems/reiser4.txt: 404:
                 Not Found
              o bad link: http://www.kernel.org/doc/Documentation/filesystems/reiser4.txt: 404:
                 Not Found
               o bad link: http://www.kernel.org/doc/Documentation/networking/ethertap.txt: 404:
                 Not Found
          + Documentation/s390
               o bad link: http://www.kernel.org/doc/Documentation/s390/zfcpdump: 404: Not Found
               o bad link: http://www.kernel.org/doc/Documentation/scheduler/sched-design.txt:
          + Documentation/scheduler
               o bad link: http://www.kernel.org/doc/Documentation/i386/: 404: Not Found
               o bad link: http://www.kernel.org/doc/Documentation/time_interpolators.txt: 404:
                 Not Found
               o bad link: http://www.kernel.org/doc/Documentation/x86_64/: 404: Not Found
                 404: Not Found
(Continue reading)

Permalink | Reply | 0 comments |

Gmane