bug#19491: /etc/mtab give wrong entries for early-mounted items.

2015-01-04 Thread 宋文武 
Ludovic Courtès  writes:

> 宋文武  skribis:
>
>> Notably, I have:
>>   /dev/sda1 /root ext4 rw,relatime,data=ordered 0 0
>>   none /root//dev devtmpfs rw 0 0
>>
>> Which should be:
>>   /dev/sda1 / ...
>>   none /dev ...
>>
>> Why not make /etc/mtab a symlink to /proc/self/mounts?
>
> This sounds appealing, but it seems that libmount in util-linux will
> insist to update /etc/mtab.
>
> Do you have evidence that it could cope with it?
Not sure, but I think libmount should just work fine.

util-linux support this since 2.19:
  https://www.kernel.org/pub/linux/utils/util-linux/v2.19/v2.19-ReleaseNotes

Debian make this default at 2011:
  https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=494001
>
> Thanks,
> Ludo’.

bug#19491: /etc/mtab give wrong entries for early-mounted items.

2015-01-04 Thread Ludovic Courtès 
宋文武  skribis:

> Ludovic Courtès  writes:
>
>> 宋文武  skribis:
>>
>>> Notably, I have:
>>>   /dev/sda1 /root ext4 rw,relatime,data=ordered 0 0
>>>   none /root//dev devtmpfs rw 0 0
>>>
>>> Which should be:
>>>   /dev/sda1 / ...
>>>   none /dev ...
>>>
>>> Why not make /etc/mtab a symlink to /proc/self/mounts?
>>
>> This sounds appealing, but it seems that libmount in util-linux will
>> insist to update /etc/mtab.
>>
>> Do you have evidence that it could cope with it?
> Not sure, but I think libmount should just work fine.
>
> util-linux support this since 2.19:
>   https://www.kernel.org/pub/linux/utils/util-linux/v2.19/v2.19-ReleaseNotes
>
> Debian make this default at 2011:
>   https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=494001

OK, it sounds like it’s safe, then.  :-)

Would you like to propose a patch?  There are changes to make around
linux-boot.scm, and also in (guix build syscalls) the #:update-mtab?
argument should default to #f.

Thanks,
Ludo’.

bug#18926: cmake-build-system should set CMAKE_INSTALL_LIBDIR and CMAKE_PREFIX_PATH

2015-01-04 Thread Ludovic Courtès 
宋文武  skribis:

> Ludovic Courtès  writes:
>
>> 宋文武  skribis:
>>
>>> Ludovic Courtès  writes:
>>>
 宋文武  skribis:

> if CMAKE_INSTALL_LIBDIR not set to `lib`, GNUInstallDirs.cmake will 
> install
> libraries files to $out/lib64.

 I found several CMake-built libraries on x86_64 (graphite2, openjpeg,
 qjson) that all use lib/, not lib64/.

 Then I found one counterexample, libftdi; however, setting
 CMAKE_INSTALL_LIBDIR=lib doesn’t make any different: it still installs
 libraries in $out/lib64.  Any idea?
>>> libftdi (not using GNUInstallDirs.cmake) handle this itself by
>>> set LIB_SUFFIX, look like we have to set it specifically.
>>
>> Ah, OK.
>>
>> Still, graphite2, openjpeg, and qjson all install to $prefix/lib, even
>> when not passing CMAKE_INSTALL_LIBDIR=lib.  Is it really needed?
> Not sure, it's just for packages using GNUInstallDirs.cmake.
>>
> if CMAKE_PREFIX_PATH not set to PATH of `inputs`, cmake will unable to
> find cmake modules of inputs.

 You’re talking about .cmake files, right?
>>> yes, but now I find out that they can under both lib/cmake and share/cmake.
>>
>> In that case...
>>
 +  ;; Search path used by 'FIND_PACKAGE' and 'INCLUDE'.
 +  (search-path-specification
 +   (variable "CMAKE_MODULE_PATH")
 +   (directories '("lib/cmake")
>> ^
>> ... just add "share/cmake" here. ---’
>>
>> Could you try that?
> Well, it not work, here is the exact output by liblxqt:

[...]

>>> after add CMAKE_PREFIX_PATH with:
>>>
>>>   (search-path-specification
>>> (variable "CMAKE_PREFIX_PATH")
>>> (directories '("")))
>>
>> I thought about it, but that "" is inelegant and not as clear.
> Yes, it not feeling good, but I have no better idea now.

In ‘core-updates’ it’s now possible to specify file patterns, so we
could look for *.cmake files under lib/cmake or share/cmake like this:

  (search-path-specification
(variable "CMAKE_MODULE_PATH")
(files '("lib/cmake" "share/cmake"))
(file-type 'regular)
(file-pattern "\\.cmake$"))

Would it help?  What else is needed?

Ludo’.

bug#18895: 'patch-shebangs' phase vs. cross-compilation

2015-01-04 Thread Ludovic Courtès 
Fixed in commit ac70048.

Ludo’.

bug#19315: per-user profile directory not created/mentioned

2015-01-04 Thread Ludovic Courtès 
l...@gnu.org (Ludovic Courtès) skribis:

> rekado  skribis:
>
>> the manual does not mention that a per-user profile directory owned by
>> the user account has to be created after installation of guix.
>
> It actually does, under “Invoking guix package” (see
> ):
>
>  In a multi-user setup, user profiles must be stored in a place
>   registered as a "garbage-collector root", which ‘$HOME/.guix-profile’
>   points to (*note Invoking guix gc::).  That directory is normally
>   ‘LOCALSTATEDIR/profiles/per-user/USER’, where LOCALSTATEDIR is the value
>   passed to ‘configure’ as ‘--localstatedir’, and USER is the user name.
>   It must be created by ‘root’, with USER as the owner.  When it does not
>   exist, or is not owned by USER, ‘guix package’ emits an error about it.
>
> Would you suggest making it more prominent somehow?

Commit 15ddeff merges code from nix-daemon.  Among other things, it adds
code to create this ‘profiles/per-user’ directory with appropriate
permissions (see LocalStore::LocalStore(bool reserveSpace) in
local-store.cc.)

Commit 4379c35 updates the manual accordingly.

Thanks,
Ludo’.

bug#19512: Failed substitution yields incomplete store item

2015-01-04 Thread David Thompson 
While working on 'guix publish', I managed to mess up the store.  :(

Here's the first substitution that failed due to an invalid eof marker
for some reason:

  dave@labrys ~$ sudo _NIX_OPTIONS="substitute-urls=http://192.168.1.157:8080"; 
guix substitute-binary --substitute 
/gnu/store/iw3jn6a1avv78pp5v2cv42vyh0d8zi0g-guile-toxcore-0.1-6a9fbe0 
/tmp/guile-toxcore
  
  found valid signature for 
'/gnu/store/iw3jn6a1avv78pp5v2cv42vyh0d8zi0g-guile-toxcore-0.1-6a9fbe0', from 
'http://192.168.1.157:8080/nar/iw3jn6a1avv78pp5v2cv42vyh0d8zi0g-guile-toxcore-0.1-6a9fbe0'
  sha256:xl35awxl6kgl5b2eum7kfgqlnap7a7itczpyxptf4kq25pasafka
  downloading 
`/gnu/store/iw3jn6a1avv78pp5v2cv42vyh0d8zi0g-guile-toxcore-0.1-6a9fbe0' (0.1 
MiB installed)...
  
http://192.168.1.157/.../iw3jn6a1avv78pp5v2cv42vyh0d8zi0g-guile-toxcore-0.1-6a9fbe0
 100.0% of 127.5 KiBguix substitute-binary: error: invalid nar end-of-file 
marker


When the same command is run again, there's a new error:

  dave@labrys ~$ sudo _NIX_OPTIONS="substitute-urls=http://192.168.1.157:8080"; 
guix substitute-binary --substitute 
/gnu/store/iw3jn6a1avv78pp5v2cv42vyh0d8zi0g-guile-toxcore-0.1-6a9fbe0 
/tmp/guile-toxcore
  
  found valid signature for 
'/gnu/store/iw3jn6a1avv78pp5v2cv42vyh0d8zi0g-guile-toxcore-0.1-6a9fbe0', from 
'http://192.168.1.157:8080/nar/iw3jn6a1avv78pp5v2cv42vyh0d8zi0g-guile-toxcore-0.1-6a9fbe0'
  sha256:xl35awxl6kgl5b2eum7kfgqlnap7a7itczpyxptf4kq25pasafka
  downloading 
`/gnu/store/iw3jn6a1avv78pp5v2cv42vyh0d8zi0g-guile-toxcore-0.1-6a9fbe0' (0.1 
MiB installed)...
  
http://192.168.1.157/.../iw3jn6a1avv78pp5v2cv42vyh0d8zi0g-guile-toxcore-0.1-6a9fbe0
   6.3% of 127.5 KiBguix substitute-binary: error: mkdir: File exists

Any idea what could be going wrong?

-- 
David Thompson
Web Developer - Free Software Foundation - http://fsf.org
GPG Key: 0FF1D807
Support the FSF: https://fsf.org/donate