[pull request] Pull request for branch yem-kconfig-for-next

[pull request] Pull request for branch yem-kconfig-for-next

[Yann E. MORIN]

From: "Yann E. MORIN" <yann.morin.1998@free.fr>

Michal,

Please pull this to revert:
    422c809: fix randomising choice entries in presence of KCONFIG_ALLCONFIG

It causes more harm that it fixes issues, and I couldn't find a proper
fix for now. To avoid breakage so close to the opening of the merge
window, it makes more sense to revert it than to wait for a fix.

As I already said earlier, I'll be on vacation the next two weeks, so
I should find some time to look back at this again, hopefully before
-rc1 is out (but any sane person that value his/her life should not hold
his/her breath...)


The following changes since commit e43956e607692f9b1c710311e4a6591ffba1edf0:

  kconfig: implement KCONFIG_PROBABILITY for randconfig (2013-04-25 00:16:30 +0200)

are available in the git repository at:

  git://gitorious.org/linux-kconfig/linux-kconfig.git yem-kconfig-for-next

for you to fetch changes up to 23a5dfdad22a574d19d7cc19b391f9ce0d3c2f21:

  Revert "kconfig: fix randomising choice entries in presence of KCONFIG_ALLCONFIG" (2013-04-26 23:21:59 +0200)

----------------------------------------------------------------
Yann E. MORIN (1):
      Revert "kconfig: fix randomising choice entries in presence of KCONFIG_ALLCONFIG"

 scripts/kconfig/confdata.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

Regards,
Yann E. MORIN

.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

Re: [pull request] Pull request for branch yem-kconfig-for-next

[Michal Marek]

On 27.4.2013 00:17, Yann E. MORIN wrote:
> From: "Yann E. MORIN" <yann.morin.1998@free.fr>
> 
> Michal,
> 
> Please pull this to revert:
>     422c809: fix randomising choice entries in presence of KCONFIG_ALLCONFIG

Pulled, thanks!

Michal

[pull request] Pull request for branch yem-kconfig-for-next

[Yann E. MORIN]

From: "Yann E. MORIN" <yann.morin.1998@free.fr>

Michal,

Please pull this fix to restore compilation of the qconf frontend.

Regards,
Yann E. MORIN.


The following changes since commit 23a5dfdad22a574d19d7cc19b391f9ce0d3c2f21:

  Revert "kconfig: fix randomising choice entries in presence of KCONFIG_ALLCONFIG" (2013-04-26 23:21:59 +0200)

are available in the git repository at:

  git://gitorious.org/linux-kconfig/linux-kconfig.git yem-kconfig-for-next

for you to fetch changes up to 21ca352b71ca252e1933b1538fe89da8a04395c3:

  kconfig: fix lists definition for C++ (2013-04-29 19:55:56 +0200)

----------------------------------------------------------------
Yann E. MORIN (1):
      kconfig: fix lists definition for C++

 scripts/kconfig/list.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

Re: [pull request] Pull request for branch yem-kconfig-for-next

[Michal Marek]

On 29.4.2013 19:59, Yann E. MORIN wrote:
> The following changes since commit 23a5dfdad22a574d19d7cc19b391f9ce0d3c2f21:
> 
>   Revert "kconfig: fix randomising choice entries in presence of KCONFIG_ALLCONFIG" (2013-04-26 23:21:59 +0200)
> 
> are available in the git repository at:
> 
>   git://gitorious.org/linux-kconfig/linux-kconfig.git yem-kconfig-for-next

Pulled, thanks.

Michal

[pull request] Pull request for branch yem-kconfig-for-next

[Yann E. MORIN]

From: "Yann E. MORIN" <yann.morin.1998@free.fr>

Hello Michal, All!

These are the patches against kconfig that I have accumulated for 3.11,
and for which I plan to later send a proper pull-request.

Note-whorthy this time:
  - fix values of tristates that are selected by boolean choices (Arve)
  - fix choice randomisation in presence of KCONFIG_ALLCONFIG    (me)
  - fix choice randomisation selecting more than one value in
    a choice (but only if it is conditional)                     (me)
  - fix choice-in-a-choice randomisation not selecting any value
    for the inner-most choice                                    (me)

Also, some code-cleanups and eye-candy:
  - mconf and nconf code cleanups                                (Dirk, Sedat)
  - mconf and nconf eye-candy                                    (Dirk)
  - scripts/config script-name in help text                      (Clément)
  - heuristic to sort found symbols by relevance                 (me)
  - more randconfig debugging help                               (me)

Note: this series is for review before I send an official pull-request.
I want to avoid the debacle [0] [1] of the last kconfig breakage I caused
with my previous kconfig "fix" to randomising choices in presence of
KCONFIG_ALLCONFIG.

I would like to send an official pull-request early next week, so all of
these can land early in -next, so they get some real-life t4esting before
the next merge windows open. So please, please, test this carefully. ;-)

[0] http://marc.info/?t=136684274600002&r=1&w=2
[1] http://marc.info/?t=136696367200003&r=1&w=2

Regards,
Yann E. MORIN.


The following changes since commit f722406faae2d073cc1d01063d1123c35425939e:

  Linux 3.10-rc1 (2013-05-11 17:14:08 -0700)

are available in the git repository at:

  git://gitorious.org/linux-kconfig/linux-kconfig.git yem-kconfig-for-next

for you to fetch changes up to b4e0ac8385fecde0363a2499323f81b4a91e8430:

  kconfig: fix randomising choice entries in presence of KCONFIG_ALLCONFIG (2013-06-18 23:59:02 +0200)

----------------------------------------------------------------
Arve Hjønnevåg (1):
      kconfig: Fix defconfig when one choice menu selects options that another choice menu depends on

Clement Chauplannaz (1):
      scripts/config: replace hard-coded script name by a dynamic value

Dirk Gouders (4):
      kconfig/lxdialog: handle newline characters in print_autowrap()
      mconf: use function calls instead of ncurses' variables LINES and COLS
      nconf: use function calls instead of ncurses' variables LINES and COLS
      mconf/nconf: mark empty menus/menuconfigs different from non-empty ones

Sedat Dilek (2):
      kconfig/lxdialog: Add definitions for mininimum (re)size values
      kconfig/lxdialog: Use new mininimum resize definitions in conf_choice()

Yann E. MORIN (6):
      kconfig/conf: fix randconfig setting multiple symbols in a choice
      kconfig/conf: accept a base-16 seed for randconfig
      kconfig/conf: print the seed used to initialise the RNG for randconfig
      kconfig: sort found symbols by relevance
      kconfig: loop as long as we changed some symbols in randconfig
      kconfig: fix randomising choice entries in presence of KCONFIG_ALLCONFIG

 scripts/config                       |  12 ++--
 scripts/kconfig/conf.c               |   6 +-
 scripts/kconfig/confdata.c           |  39 +++++++++----
 scripts/kconfig/expr.h               |   3 +
 scripts/kconfig/lkc.h                |   3 +-
 scripts/kconfig/lkc_proto.h          |   1 +
 scripts/kconfig/lxdialog/checklist.c |   8 +--
 scripts/kconfig/lxdialog/dialog.h    |  14 +++++
 scripts/kconfig/lxdialog/inputbox.c  |   8 +--
 scripts/kconfig/lxdialog/menubox.c   |   6 +-
 scripts/kconfig/lxdialog/textbox.c   |   6 +-
 scripts/kconfig/lxdialog/util.c      |  46 ++++++++-------
 scripts/kconfig/lxdialog/yesno.c     |   8 +--
 scripts/kconfig/mconf.c              |  19 ++++---
 scripts/kconfig/menu.c               |  16 ++++++
 scripts/kconfig/nconf.c              |  37 +++++++-----
 scripts/kconfig/nconf.gui.c          |  20 ++++---
 scripts/kconfig/symbol.c             | 107 ++++++++++++++++++++++++++++++++---
 18 files changed, 263 insertions(+), 96 deletions(-)
-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

Re: [pull request] Pull request for branch yem-kconfig-for-next

[Michal Marek]

Dne 19.6.2013 00:44, Yann E. MORIN napsal(a):
> From: "Yann E. MORIN" <yann.morin.1998@free.fr>
> 
> Hello Michal, All!
> 
> These are the patches against kconfig that I have accumulated for 3.11,
> and for which I plan to later send a proper pull-request.
> 
> Note-whorthy this time:
>   - fix values of tristates that are selected by boolean choices (Arve)
>   - fix choice randomisation in presence of KCONFIG_ALLCONFIG    (me)
>   - fix choice randomisation selecting more than one value in
>     a choice (but only if it is conditional)                     (me)
>   - fix choice-in-a-choice randomisation not selecting any value
>     for the inner-most choice                                    (me)
> 
> Also, some code-cleanups and eye-candy:
>   - mconf and nconf code cleanups                                (Dirk, Sedat)
>   - mconf and nconf eye-candy                                    (Dirk)
>   - scripts/config script-name in help text                      (Clément)
>   - heuristic to sort found symbols by relevance                 (me)
>   - more randconfig debugging help                               (me)
> 
> Note: this series is for review before I send an official pull-request.
> I want to avoid the debacle [0] [1] of the last kconfig breakage I caused
> with my previous kconfig "fix" to randomising choices in presence of
> KCONFIG_ALLCONFIG.
> 
> I would like to send an official pull-request early next week, so all of
> these can land early in -next, so they get some real-life t4esting before
> the next merge windows open. So please, please, test this carefully. ;-)

I tested this a bit, it works fine. Unless somebody else finds issues,
I'll merge it.

Thanks,
Michal

Re: [pull request] Pull request for branch yem-kconfig-for-next

[Yann E. MORIN]

Michal, All,

On 2013-06-19 22:40 +0200, Michal Marek spake thusly:
> Dne 19.6.2013 00:44, Yann E. MORIN napsal(a):
> > From: "Yann E. MORIN" <yann.morin.1998@free.fr>
> > 
> > Hello Michal, All!
> > 
> > These are the patches against kconfig that I have accumulated for 3.11,
> > and for which I plan to later send a proper pull-request.
> > 
> > Note-whorthy this time:
> >   - fix values of tristates that are selected by boolean choices (Arve)
> >   - fix choice randomisation in presence of KCONFIG_ALLCONFIG    (me)
> >   - fix choice randomisation selecting more than one value in
> >     a choice (but only if it is conditional)                     (me)
> >   - fix choice-in-a-choice randomisation not selecting any value
> >     for the inner-most choice                                    (me)
> > 
> > Also, some code-cleanups and eye-candy:
> >   - mconf and nconf code cleanups                                (Dirk, Sedat)
> >   - mconf and nconf eye-candy                                    (Dirk)
> >   - scripts/config script-name in help text                      (Clément)
> >   - heuristic to sort found symbols by relevance                 (me)
> >   - more randconfig debugging help                               (me)
> > 
> > Note: this series is for review before I send an official pull-request.
> > I want to avoid the debacle [0] [1] of the last kconfig breakage I caused
> > with my previous kconfig "fix" to randomising choices in presence of
> > KCONFIG_ALLCONFIG.
> > 
> > I would like to send an official pull-request early next week, so all of
> > these can land early in -next, so they get some real-life t4esting before
> > the next merge windows open. So please, please, test this carefully. ;-)
> 
> I tested this a bit, it works fine. Unless somebody else finds issues,
> I'll merge it.

Could you please wait a few days (say friday, or monday), I know someone
whose supposed to do some testing on his use-case (Hello Thomas! ;-) ).

Thank you!

Regards,
Yann E. MORIN.

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

[pull request] Pull request for branch yem-kconfig-for-next

[Yann E. MORIN]

From: "Yann E. MORIN" <yann.morin.1998@free.fr>

Hello Michal, All,

Please pull this kconfig fix from Kees that enables 64-bit-wide
(ie. signed long) [int,hex] ranges.

Regards,
Yann E. MORIN.


The following changes since commit 490f16171119a16e05d670306c105f3b45c38837:

  Revert "kconfig: fix randomising choice entries in presence of KCONFIG_ALLCONFIG" (2013-06-26 15:49:00 +0200)

are available in the git repository at:

  git://gitorious.org/linux-kconfig/linux-kconfig.git yem-kconfig-for-next

for you to fetch changes up to b57caaaed2bd127fe656e6c145970ed6a05c0125:

  kconfig: allow "hex" and "range" to support longs (2013-06-29 15:30:17 +0200)

----------------------------------------------------------------
Kees Cook (1):
      kconfig: allow "hex" and "range" to support longs

 scripts/kconfig/symbol.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

[PATCH 1/1] kconfig: allow "hex" and "range" to support longs

[Yann E. MORIN]

From: Kees Cook <keescook@chromium.org>

The parsing routines for Kconfig files use strtol(), but store and
render values as int. Switch types and formating to long to support a
wider range of values. For example, 0x80000000 wasn't representable.

Signed-off-by: Kees Cook <keescook@chromium.org>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
---
 scripts/kconfig/symbol.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/scripts/kconfig/symbol.c b/scripts/kconfig/symbol.c
index 387d554..d550300 100644
--- a/scripts/kconfig/symbol.c
+++ b/scripts/kconfig/symbol.c
@@ -136,7 +136,7 @@ static struct property *sym_get_range_prop(struct symbol *sym)
 	return NULL;
 }
 
-static int sym_get_range_val(struct symbol *sym, int base)
+static long sym_get_range_val(struct symbol *sym, int base)
 {
 	sym_calc_value(sym);
 	switch (sym->type) {
@@ -155,7 +155,7 @@ static int sym_get_range_val(struct symbol *sym, int base)
 static void sym_validate_range(struct symbol *sym)
 {
 	struct property *prop;
-	int base, val, val2;
+	long base, val, val2;
 	char str[64];
 
 	switch (sym->type) {
@@ -179,9 +179,9 @@ static void sym_validate_range(struct symbol *sym)
 			return;
 	}
 	if (sym->type == S_INT)
-		sprintf(str, "%d", val2);
+		sprintf(str, "%ld", val2);
 	else
-		sprintf(str, "0x%x", val2);
+		sprintf(str, "0x%lx", val2);
 	sym->curr.val = strdup(str);
 }
 
@@ -594,7 +594,7 @@ bool sym_string_valid(struct symbol *sym, const char *str)
 bool sym_string_within_range(struct symbol *sym, const char *str)
 {
 	struct property *prop;
-	int val;
+	long val;
 
 	switch (sym->type) {
 	case S_STRING:
-- 
1.8.1.2

Re: [PATCH 1/1] kconfig: allow "hex" and "range" to support longs

[Geert Uytterhoeven]

On Sat, Jun 29, 2013 at 4:21 PM, Yann E. MORIN <yann.morin.1998@free.fr> wrote:
> From: Kees Cook <keescook@chromium.org>
>
> The parsing routines for Kconfig files use strtol(), but store and
> render values as int. Switch types and formating to long to support a
> wider range of values. For example, 0x80000000 wasn't representable.

0x80000000 does fit in an int. If it's printed as hex, it's treated as unsigned.
Is there a "0" missing, or am I missing something?

> Signed-off-by: Kees Cook <keescook@chromium.org>
> Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
> Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
> Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
> ---
>  scripts/kconfig/symbol.c | 10 +++++-----
>  1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/scripts/kconfig/symbol.c b/scripts/kconfig/symbol.c
> index 387d554..d550300 100644
> --- a/scripts/kconfig/symbol.c
> +++ b/scripts/kconfig/symbol.c
> @@ -136,7 +136,7 @@ static struct property *sym_get_range_prop(struct symbol *sym)
>         return NULL;
>  }
>
> -static int sym_get_range_val(struct symbol *sym, int base)
> +static long sym_get_range_val(struct symbol *sym, int base)
>  {
>         sym_calc_value(sym);
>         switch (sym->type) {

Changing all these "int" to "long" only matters on 64-bit platforms.
As kconfig is built on the host, this will cause different behavior when
cross-compiling 64-bit (I assume that's where you need it most) kernels
on a 32-bit or a 64-bit host.

Probably you wanted to use "long long" instead of "long", and switch
to "strtoll()", to make them always 64-bit?

Still, beware using 64-bit config symbols with arithmetic shell operations from
a Makefile, cfr. https://lkml.org/lkml/2013/6/13/200.

Sorry for only noticing now, after it went into 3.11-rc1.

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH 1/1] kconfig: allow "hex" and "range" to support longs

[Kees Cook]

On Thu, Jul 18, 2013 at 3:08 AM, Geert Uytterhoeven
<geert@linux-m68k.org> wrote:
> On Sat, Jun 29, 2013 at 4:21 PM, Yann E. MORIN <yann.morin.1998@free.fr> wrote:
>> From: Kees Cook <keescook@chromium.org>
>>
>> The parsing routines for Kconfig files use strtol(), but store and
>> render values as int. Switch types and formating to long to support a
>> wider range of values. For example, 0x80000000 wasn't representable.
>
> 0x80000000 does fit in an int. If it's printed as hex, it's treated as unsigned.
> Is there a "0" missing, or am I missing something?

0x80000000 is 1 more than INT_MAX, but you're right about the hex
printing. Regardless, it was still being truncated in kconfig
(probably due to signed comparisons). I had to switch the formatting
since the type changed, though.

>> Signed-off-by: Kees Cook <keescook@chromium.org>
>> Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
>> Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
>> Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
>> ---
>>  scripts/kconfig/symbol.c | 10 +++++-----
>>  1 file changed, 5 insertions(+), 5 deletions(-)
>>
>> diff --git a/scripts/kconfig/symbol.c b/scripts/kconfig/symbol.c
>> index 387d554..d550300 100644
>> --- a/scripts/kconfig/symbol.c
>> +++ b/scripts/kconfig/symbol.c
>> @@ -136,7 +136,7 @@ static struct property *sym_get_range_prop(struct symbol *sym)
>>         return NULL;
>>  }
>>
>> -static int sym_get_range_val(struct symbol *sym, int base)
>> +static long sym_get_range_val(struct symbol *sym, int base)
>>  {
>>         sym_calc_value(sym);
>>         switch (sym->type) {
>
> Changing all these "int" to "long" only matters on 64-bit platforms.
> As kconfig is built on the host, this will cause different behavior when
> cross-compiling 64-bit (I assume that's where you need it most) kernels
> on a 32-bit or a 64-bit host.
>
> Probably you wanted to use "long long" instead of "long", and switch
> to "strtoll()", to make them always 64-bit?

That's an excellent point; I hadn't considered the cross-compiling
from a 32-bit host.

> Still, beware using 64-bit config symbols with arithmetic shell operations from
> a Makefile, cfr. https://lkml.org/lkml/2013/6/13/200.

Yeah, noted. I don't have plans to pass kconfig items through the shell.

> Sorry for only noticing now, after it went into 3.11-rc1.

Thanks for pointing it out! I don't think there is huge urgency since
nothing is (yet) using > INT_MAX values in kconfig. Regardless, I'll
send a new patch that switches to long long.

-Kees

--
Kees Cook
Chrome OS Security