[PATCH] check for 'flex' before starting tools build

[PATCH] check for 'flex' before starting tools build

[Olaf Hering]

# HG changeset patch
# User Olaf Hering <olaf@aepfle.de>
# Date 1318335031 -7200
# Node ID 5a5181b8c31983ff2f465f85a41df784a5397ed5
# Parent  4b0907c6a08c348962bd976c2976257b412408be
check for 'flex' before starting tools build.

Signed-off-by: Olaf Hering <olaf@aepfle.de>

diff -r 4b0907c6a08c -r 5a5181b8c319 tools/check/check_flex
--- /dev/null
+++ b/tools/check/check_flex
@@ -0,0 +1,6 @@
+#!/bin/sh
+# CHECK-BUILD
+
+. ./funcs.sh
+
+has_or_fail flex

Re: [PATCH] check for 'flex' before starting tools build

[Olaf Hering]

On Tue, Oct 11, Olaf Hering wrote:

> # HG changeset patch
> # User Olaf Hering <olaf@aepfle.de>
> # Date 1318335031 -7200
> # Node ID 5a5181b8c31983ff2f465f85a41df784a5397ed5
> # Parent  4b0907c6a08c348962bd976c2976257b412408be
> check for 'flex' before starting tools build.

A similar check should be added for bison.

After todays 'hg pull -u' and 'make tools xen', two files were modified:

 # hg st
M tools/libxl/libxlu_disk_l.c
M tools/libxl/libxlu_disk_l.h

Shouldnt they be generated files in the first place,
and not part of the SCM?

Olaf

Re: [PATCH] check for 'flex' before starting tools build

[Jan Beulich]

>>> On 11.10.11 at 14:19, Olaf Hering <olaf@aepfle.de> wrote:
> On Tue, Oct 11, Olaf Hering wrote:
> 
>> # HG changeset patch
>> # User Olaf Hering <olaf@aepfle.de>
>> # Date 1318335031 -7200
>> # Node ID 5a5181b8c31983ff2f465f85a41df784a5397ed5
>> # Parent  4b0907c6a08c348962bd976c2976257b412408be
>> check for 'flex' before starting tools build.
> 
> A similar check should be added for bison.
> 
> After todays 'hg pull -u' and 'make tools xen', two files were modified:
> 
>  # hg st
> M tools/libxl/libxlu_disk_l.c
> M tools/libxl/libxlu_disk_l.h
> 
> Shouldnt they be generated files in the first place,
> and not part of the SCM?

Correct - this had happened once already. And neither flex nor
bison should be required (or if they were, likely a minimum version
would also need to be enforced, and the overridable $(FLEX) and
$(BISON) values would also need to be taken into consideration -
when this happened first, I had to use privately built variants on
SLE10 for things to work out properly).

Jan