This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH v3 4/7] manual: Refactor errno @comments.


On 05/16/2017 04:06 AM, Joseph Myers wrote:
> On Tue, 16 May 2017, Rical Jasan wrote:
>> errno.texi contains a large number of standards annotations which also
>> document the error string on the same line.  The surrounding
>> commentary indicates this information should be retained, so it is
>> moved to the subsequent comment documenting the error code value.  In
>> many cases, there was a redundant "@c DO NOT REMOVE" comment, which
>> was relocated to the front.  Additionally, a number of "Linux???"
>> standards appeared under a section of the manual stating the following
>> errors came from Linux, so the question marks were removed.
> 
> Have you made sure that the generated sysdeps/gnu/errlist.c is unchanged 
> by this patch?  And sysdeps/mach/hurd/bits/errno.h (you might need to 
> figure out what commands to run by hand for that one, in the absence of a 
> way to build a cross compiler and glibc for Hurd).  (NaCl also generates 
> bits/errno.h - not checked in for NaCl - from errno.texi, but that port 
> should be going away soon.)

I see on the Hurd homepage: "Although it is possible to bootstrap the
GNU/Hurd system from the sources by cross-compiling and installing the
system software and the basic applications, this is a difficult process.
It is not recommended that you do this." [1]

Setting up a VM looks promising, though. [2]  IIUC, assuming there's a
compiler and other necessary tools in that image, all I should need to
do is `make' glibc with and without the patch, and compare the resultant
sysdeps/mach/hurd/bits/errno.h, correct?

Or, do you have any idea what might be an easier solution?  As far as
manually generating the file goes, sysdeps/mach/hurd/Makefile uses, in
addition to errno.texi, $(mach-errnos-deps) and
$(common-objpfx)errnos.d.  I don't see where mach-errnos-deps even comes
from, and if I don't have the input, and don't know what the expected
output is, I don't know that I'd trust it much.

Perhaps someone using the Hurd could check the patch.

Rical

[1] https://www.gnu.org/software/hurd/hurd.html
[2] https://www.gnu.org/software/hurd/hurd/running/debian.html


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]