headers
Sonic Zhang | 1 Feb 2011 04:52
Picon

[PATCH] buildroot:package: Remove configure target parameter in Makefile.autotools.in

From: Sonic Zhang <sonic.zhang@...>

configure parameter --target has different meaning from buildroot environment
parameter GNU_TARGET_NAME. GNU_TARGET_NAME is actually equal to --host.

  --build=BUILD     configure for building on BUILD [guessed]
  --host=HOST       cross-compile to build programs to run on HOST [BUILD]
  --target=TARGET   configure for building compilers for TARGET [HOST]

--target is rarely used by most packages and is redundant in configure line.
If some packages do need --target, they can append it to buildroot environment
parameter INETUTILS_CONF_OPT in packages' makefile.

Signed-off-by: Sonic Zhang <sonic.zhang@...>
---
 package/Makefile.autotools.in |    1 -
 1 files changed, 0 insertions(+), 1 deletions(-)

diff --git a/package/Makefile.autotools.in b/package/Makefile.autotools.in
index d445348..151e5cf 100644
--- a/package/Makefile.autotools.in
+++ b/package/Makefile.autotools.in
 <at>  <at>  -87,7 +87,6  <at>  <at>  define $(2)_CONFIGURE_CMDS
 	$$(TARGET_CONFIGURE_ARGS) \
 	$$($$(PKG)_CONF_ENV) \
 	./configure \
-		--target=$$(GNU_TARGET_NAME) \
 		--host=$$(GNU_TARGET_NAME) \
 		--build=$$(GNU_HOST_NAME) \
 		--prefix=/usr \
(Continue reading)

Permalink | Reply | 11 comments |
headers
Mike Frysinger | 1 Feb 2011 04:59
Picon

Re: [PATCH] buildroot: package: add bluez utils

On Tue, Jan 25, 2011 at 04:52, Bob Liu wrote:
> Add bluez-utils.

new packages should be submitted to the buildroot list.  we can cc
this list in the process.

> --- /dev/null
> +++ b/package/bluez_utils/bluez_utils.mk
> +BLUEZ_UTILS_VERSION = 4.86

4.87 was just released

> +BLUEZ_UTILS_AUTORECONF = NO

this is redundant.  the default is "NO" already.

> +BLUEZ_UTILS_INSTALL_STAGING = NO

why not ?  it installs a library that people can & want to link against.

> +BLUEZ_UTILS_CONF_OPT = \
> +

this is useless and should be punted

> +BLUEZ_UTILS_DEPENDENCIES += uclibc

this doesnt make much sense.  bluez-utils needs a C library, but not
uclibc.  the toolchain takes care of this.
-mike
(Continue reading)

Permalink | Reply | 0 comments |
headers
Mike Frysinger | 1 Feb 2011 05:02
Picon

Re: [PATCH] buildroot:package: Remove configure target parameter in Makefile.autotools.in

On Mon, Jan 31, 2011 at 22:52, Sonic Zhang wrote:
> configure parameter --target has different meaning from buildroot environment
> parameter GNU_TARGET_NAME. GNU_TARGET_NAME is actually equal to --host.
>
>  --build=BUILD     configure for building on BUILD [guessed]
>  --host=HOST       cross-compile to build programs to run on HOST [BUILD]
>  --target=TARGET   configure for building compilers for TARGET [HOST]
>
> --target is rarely used by most packages and is redundant in configure line.
> If some packages do need --target, they can append it to buildroot environment
> parameter INETUTILS_CONF_OPT in packages' makefile.

not really ... the usage is harmless.  it might be a little redundant
since target tends to default to host, but that's about it.  what are
you actually trying to address here ?
-mike
_______________________________________________
Uclinux-dist-devel mailing list
Uclinux-dist-devel <at> blackfin.uclinux.org
https://blackfin.uclinux.org/mailman/listinfo/uclinux-dist-devel
Permalink | Reply | 9 comments |
headers
Zhang, Sonic | 1 Feb 2011 08:48
Favicon

Re: [PATCH] buildroot:package: Remove configure target parameter in Makefile.autotools.in


>-----Original Message-----
>From: uclinux-dist-devel-bounces@...
>[mailto:uclinux-dist-devel-bounces@...] On
>Behalf Of Mike Frysinger
>Sent: Tuesday, February 01, 2011 3:00 PM
>To: Sonic Zhang
>Cc: buildroot@...; uclinux-dist-devel@...
>Subject: Re: [uclinux-dist-devel] [PATCH] buildroot:package:
>Remove configure target parameter in Makefile.autotools.in
>
>On Mon, Jan 31, 2011 at 23:43, Sonic Zhang wrote:
>
>please dont top post
>
>> Some autotools based software packages generate prefix name to the
>> binary after running install if you set --target parameter.
>>
>> For example:
>>
>> --target=arm
>>
>> rcp is rename to "linux-arm-rcp" after running install
>
>this is due to program_transform_name which is easily disabled:
>    program_transform_name=
>

Not only inetutils, but many other packages need "program_transform_name=" in this case. Because rare
packages require --target to be set, why not do something to ease many?
(Continue reading)

Permalink | Reply | 1 comment |
headers
Mike Frysinger | 1 Feb 2011 18:15
Picon

Re: musb host hibernate issue (SR#: 1124547)

On Mon, Jan 31, 2011 at 17:17, Manguane, Jorge wrote:
> When you come out of hibernate do you send the RESUME signal on the bus (by
> setting the RESUME bit in the USB_POWER register) and make sure the SUSPEND
> bit is cleared?  In the case where the “slave device” sends a remote wake-up
> signal to the host (Blackfin) it might expect the RESUME signal to be
> extended by the host for 20 ms.
>
> So, perhaps, instead of the added delay you should, right after waking up
> from hibernate, try to set the RESUME bit in the MUSB_POWER register to
> transmit the Resume signal on the bus and turn it off after at least 20 ms.
>
> Incidentally, the above procedure should also be followed when waking up
> from Sleep mode upon receiving a Resume interrupt.

Sonic/Bob will obviously know more than i, but it seems that we are
doing this ...

musb_core.c:musb_stage0_irq() does set RESUME in the MUSB_POWER register, and
clears the SUSPENDM bit.  and it delays the timer for 20 msecs.  so it would
see we're doing what they're asking for already.

http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=drivers/usb/musb/musb_core.c
...
    /* in host mode, the peripheral may issue remote wakeup.
     * in peripheral mode, the host may resume the link.
     * spurious RESUME irqs happen too, paired with SUSPEND.
     */
    if (int_usb & MUSB_INTR_RESUME) {
            switch (musb->xceiv->state) {
            case OTG_STATE_A_SUSPEND:
(Continue reading)

Permalink | Reply | 6 comments |
headers
Grant Likely | 2 Feb 2011 05:05
Picon

Re: [PATCH] spi:add support for slave function in current spi framework

On Mon, Jan 31, 2011 at 04:46:51PM +0800, Aaron.Wu wrote:
> This is for the current git trunk and it's basically from  the patch once post by Ken Mills.  It's a draft
version showing how we plan to implement the spi slave funtion by introducing some changes on the current
spi framework, if this style of implementation is going to be accepted then we will go on to do more work on it.
> 
> Signed-off-by: Aaron.Wu <Aaron.Wu06@...>
> ---
>  drivers/spi/spi.c       |  401 ++++++++++++++++++++++++++++++++++++++++-------
>  include/linux/spi/spi.h |   97 ++++++++++--
>  2 files changed, 432 insertions(+), 66 deletions(-)

Hi Aaron,

I made a few comments about the patch itself, but there is something
more fundamental that needs to be addressed first, so I'll move it to
the top...

> diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h
> index a3059c2..8663947 100644
> --- a/include/linux/spi/spi.h
> +++ b/include/linux/spi/spi.h
>  <at>  <at>  -24,15 +24,18  <at>  <at> 
>  #include <linux/slab.h>
>  
>  /*
> - * INTERFACES between SPI master-side drivers and SPI infrastructure.
> - * (There's no SPI slave support for Linux yet...)
> - */
> + * INTERFACES between SPI Master/Slave side drivers and
> + * SPI infrastructure.
(Continue reading)

Permalink | Reply | 2 comments |
headers
Mike Frysinger | 3 Feb 2011 07:32
Picon
Favicon
Gravatar

[PATCH] net/irda: convert bfin_sir to common Blackfin UART header

No need to duplicate these defines now that the common Blackfin code has
unified these for all UART devices.

Signed-off-by: Mike Frysinger <vapier@...>
---
 drivers/net/irda/bfin_sir.c |   59 +++++++++++++++++++--------------------
 drivers/net/irda/bfin_sir.h |   64 +++----------------------------------------
 2 files changed, 33 insertions(+), 90 deletions(-)

diff --git a/drivers/net/irda/bfin_sir.c b/drivers/net/irda/bfin_sir.c
index f940dfa..9d4ce1a 100644
--- a/drivers/net/irda/bfin_sir.c
+++ b/drivers/net/irda/bfin_sir.c
 <at>  <at>  -67,27 +67,27  <at>  <at>  static void bfin_sir_stop_tx(struct bfin_sir_port *port)
 	disable_dma(port->tx_dma_channel);
 #endif

-	while (!(SIR_UART_GET_LSR(port) & THRE)) {
+	while (!(UART_GET_LSR(port) & THRE)) {
 		cpu_relax();
 		continue;
 	}

-	SIR_UART_STOP_TX(port);
+	UART_CLEAR_IER(port, ETBEI);
 }

 static void bfin_sir_enable_tx(struct bfin_sir_port *port)
 {
-	SIR_UART_ENABLE_TX(port);
(Continue reading)

Permalink | Reply | 2 comments |
headers
Greg KH | 3 Feb 2011 23:20
Gravatar

Re: [PATCH] drivers: char: hvc: add Blackfin JTAG console support

On Mon, Jan 10, 2011 at 09:43:57PM -0500, Mike Frysinger wrote:
> This converts the existing bfin_jtag_comm TTY driver to the HVC layer so
> that the common HVC code can worry about all of the TTY/polling crap and
> leave the Blackfin code to worry about the Blackfin bits.

This patch no longer applies to the linux-next tree.  Can you redo it
and resend it?

thanks,

greg k-h
Permalink | Reply | 1 comment |
headers
Tobias Klauser | 4 Feb 2011 17:45
Picon
Favicon

[PATCH uClinux-dist] boa: Add patch to fix CVE-2009-4496

The patch is taken from the Debian bug tracker.

vulnerability:
http://web.nvd.nist.gov/view/vuln/detail?vulnId=CVE-2009-4496

bug report:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=578035

patch:
http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=5;filename=escape-errorlog.patch;att=1;bug=578035

Signed-off-by: Tobias Klauser <klto@...>
---
 user/boa/boa-0.94.14rc21/src/log.c                 |   30 ++++++++-
 .../patches/boa-0.94.14rc21-escape-errorlog.patch  |   73 ++++++++++++++++++++
 2 files changed, 101 insertions(+), 2 deletions(-)
 create mode 100644 user/boa/patches/boa-0.94.14rc21-escape-errorlog.patch

diff --git a/user/boa/boa-0.94.14rc21/src/log.c b/user/boa/boa-0.94.14rc21/src/log.c
index 9119cd7..8b8aaab 100644
--- a/user/boa/boa-0.94.14rc21/src/log.c
+++ b/user/boa/boa-0.94.14rc21/src/log.c
 <at>  <at>  -156,6 +156,29  <at>  <at>  void log_access(request * req)
            (req->header_user_agent ? req->header_user_agent : "-"));
 }

+static char *escape_pathname(const char *inp)
+{
+    char *escaped, *c;
+
(Continue reading)

Permalink | Reply | 1 comment |
headers
Mike Frysinger | 4 Feb 2011 23:50
Picon

Re: [PATCH uClinux-dist] boa: Add patch to fix CVE-2009-4496

On Fri, Feb 4, 2011 at 11:45, Tobias Klauser wrote:
> The patch is taken from the Debian bug tracker.
>
> vulnerability:
> http://web.nvd.nist.gov/view/vuln/detail?vulnId=CVE-2009-4496
>
> bug report:
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=578035
>
> patch:
> http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=5;filename=escape-errorlog.patch;att=1;bug=578035

thanks, added to svn trunk
-mike
Permalink | Reply | 0 comments |

Gmane