* [Buildroot] [PATCH v8 1/1] mosh: new package
@ 2015-07-26 16:34 Christian Stewart
2015-07-26 17:01 ` Yann E. MORIN
` (2 more replies)
0 siblings, 3 replies; 7+ messages in thread
From: Christian Stewart @ 2015-07-26 16:34 UTC (permalink / raw)
To: buildroot
Adding mosh, the mobile shell. Mosh uses ssh or dropbear as an initial transport to
start mosh-server which uses UDP to communicate with the client.
Supports a predictive model to enhance performance on weak connections
and compensate for general lag. Also supports disconnections and
reconnections seamlessly.
Signed-off-by: Christian Stewart <christian@paral.in>
---
package/Config.in | 1 +
package/mosh/0001-remove-system-locale-calls.patch | 46 ++++++++++++++++++++++
package/mosh/Config.in | 19 +++++++++
package/mosh/mosh.hash | 2 +
package/mosh/mosh.mk | 13 ++++++
5 files changed, 81 insertions(+)
create mode 100644 package/mosh/0001-remove-system-locale-calls.patch
create mode 100644 package/mosh/Config.in
create mode 100644 package/mosh/mosh.hash
create mode 100644 package/mosh/mosh.mk
diff --git a/package/Config.in b/package/Config.in
index cfa7512..d1e6640 100644
--- a/package/Config.in
+++ b/package/Config.in
@@ -1250,6 +1250,7 @@ endif
source "package/mongoose/Config.in"
source "package/mongrel2/Config.in"
source "package/monkey/Config.in"
+ source "package/mosh/Config.in"
source "package/mosquitto/Config.in"
source "package/mrouted/Config.in"
source "package/mtr/Config.in"
diff --git a/package/mosh/0001-remove-system-locale-calls.patch b/package/mosh/0001-remove-system-locale-calls.patch
new file mode 100644
index 0000000..114e6ed
--- /dev/null
+++ b/package/mosh/0001-remove-system-locale-calls.patch
@@ -0,0 +1,46 @@
+The locale command is not available on many systems. As this variable
+is unused and appears to have been written with the intent of
+displaying the locale settings to the user, it's not really necessary.
+As this breaks Mosh on a lot of systems, it's best to remove the calls.
+
+Upstream status: refused, see: https://github.com/keithw/mosh/issues/650
+
+Signed-off-by: Christian Stewart <christian@paral.in>
+---
+ src/frontend/mosh-server.cc | 4 +++-
+ src/frontend/stmclient.cc | 4 +++-
+ 2 files changed, 6 insertions(+), 2 deletions(-)
+
+diff --git a/src/frontend/mosh-server.cc b/src/frontend/mosh-server.cc
+index 76ed2ed..0d2f222 100644
+--- a/src/frontend/mosh-server.cc
++++ b/src/frontend/mosh-server.cc
+@@ -313,7 +313,9 @@ int main( int argc, char *argv[] )
+ fprintf( stderr, "mosh-server needs a UTF-8 native locale to run.\n\n" );
+ fprintf( stderr, "Unfortunately, the local environment (%s) specifies\nthe character set \"%s\",\n\n", native_ctype.str().c_str(), native_charset.c_str() );
+ fprintf( stderr, "The client-supplied environment (%s) specifies\nthe character set \"%s\".\n\n", client_ctype.str().c_str(), client_charset.c_str() );
+- int unused __attribute((unused)) = system( "locale" );
++
++ fprintf( stderr, "This is a buildroot system, 'locale' debug output has been removed." );
++
+ exit( 1 );
+ }
+ }
+diff --git a/src/frontend/stmclient.cc b/src/frontend/stmclient.cc
+index 9e4d916..06fac8b 100644
+--- a/src/frontend/stmclient.cc
++++ b/src/frontend/stmclient.cc
+@@ -85,7 +85,9 @@ void STMClient::init( void )
+
+ fprintf( stderr, "mosh-client needs a UTF-8 native locale to run.\n\n" );
+ fprintf( stderr, "Unfortunately, the client's environment (%s) specifies\nthe character set \"%s\".\n\n", native_ctype.str().c_str(), native_charset.c_str() );
+- int unused __attribute((unused)) = system( "locale" );
++
++ fprintf( stderr, "This is a buildroot system, 'locale' debug output has been removed." );
++
+ exit( 1 );
+ }
+
+--
+2.1.4
+
diff --git a/package/mosh/Config.in b/package/mosh/Config.in
new file mode 100644
index 0000000..9841b2e
--- /dev/null
+++ b/package/mosh/Config.in
@@ -0,0 +1,19 @@
+comment "mosh needs a toolchain w/ threads"
+ depends on !BR2_TOOLCHAIN_HAS_THREADS
+
+config BR2_PACKAGE_MOSH
+ bool "mosh"
+ depends on BR2_arm || BR2_i386 || BR2_mipsel || BR2_x86_64 # protobuf
+ depends on BR2_TOOLCHAIN_HAS_THREADS #protobuf
+ depends on BR2_HOSTARCH = "x86_64" || BR2_HOSTARCH = "x86" # protobuf
+ select BR2_PACKAGE_PROTOBUF
+ select BR2_PACKAGE_NCURSES
+ select BR2_PACKAGE_OPENSSL
+ select BR2_PACKAGE_OPENSSH if !BR2_PACKAGE_DROPBEAR_CLIENT
+ select BR2_PACKAGE_ZLIB
+ help
+ Remote terminal application that supports intermittent connectivity,
+ allows roaming, and provides speculative local echo and line editing
+ of user keystrokes.
+
+ https://mosh.mit.edu
diff --git a/package/mosh/mosh.hash b/package/mosh/mosh.hash
new file mode 100644
index 0000000..ab925b5
--- /dev/null
+++ b/package/mosh/mosh.hash
@@ -0,0 +1,2 @@
+# From: http://mailman.mit.edu/pipermail/mosh-users/2015-July/000283.html
+sha256 1af809e5d747c333a852fbf7acdbf4d354dc4bbc2839e3afe5cf798190074be3 mosh-1.2.5.tar.gz
diff --git a/package/mosh/mosh.mk b/package/mosh/mosh.mk
new file mode 100644
index 0000000..dd1ef93
--- /dev/null
+++ b/package/mosh/mosh.mk
@@ -0,0 +1,13 @@
+################################################################################
+#
+# mosh
+#
+################################################################################
+
+MOSH_VERSION = 1.2.5
+MOSH_SITE = https://mosh.mit.edu/
+MOSH_DEPENDENCIES = zlib ncurses protobuf openssl
+MOSH_LICENSE = GPLv3+ with exception
+MOSH_LICENSE_FILES = COPYING COPYING.iOS
+
+$(eval $(autotools-package))
--
2.1.4
^ permalink raw reply related [flat|nested] 7+ messages in thread
* [Buildroot] [PATCH v8 1/1] mosh: new package
2015-07-26 16:34 [Buildroot] [PATCH v8 1/1] mosh: new package Christian Stewart
@ 2015-07-26 17:01 ` Yann E. MORIN
2015-07-26 17:35 ` Thomas Petazzoni
2015-07-26 22:20 ` Baruch Siach
2 siblings, 0 replies; 7+ messages in thread
From: Yann E. MORIN @ 2015-07-26 17:01 UTC (permalink / raw)
To: buildroot
Christian, All,
On 2015-07-26 09:34 -0700, Christian Stewart spake thusly:
> Adding mosh, the mobile shell. Mosh uses ssh or dropbear as an initial transport to
> start mosh-server which uses UDP to communicate with the client.
> Supports a predictive model to enhance performance on weak connections
> and compensate for general lag. Also supports disconnections and
> reconnections seamlessly.
>
> Signed-off-by: Christian Stewart <christian@paral.in>
mosh needs host-pkgconf:
checking pkg-config is at least version 0.9.0... ./configure: line
7378: /home/ymorin/dev/buildroot/O/host/usr/bin/pkg-config: No such
file or directory
With that fixed:
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[yann.morin.1998 at free.fr: build-tested only]
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
(I think Thomas will fix that when applying...)
Regards,
Yann E. MORIN.
> ---
> package/Config.in | 1 +
> package/mosh/0001-remove-system-locale-calls.patch | 46 ++++++++++++++++++++++
> package/mosh/Config.in | 19 +++++++++
> package/mosh/mosh.hash | 2 +
> package/mosh/mosh.mk | 13 ++++++
> 5 files changed, 81 insertions(+)
> create mode 100644 package/mosh/0001-remove-system-locale-calls.patch
> create mode 100644 package/mosh/Config.in
> create mode 100644 package/mosh/mosh.hash
> create mode 100644 package/mosh/mosh.mk
>
> diff --git a/package/Config.in b/package/Config.in
> index cfa7512..d1e6640 100644
> --- a/package/Config.in
> +++ b/package/Config.in
> @@ -1250,6 +1250,7 @@ endif
> source "package/mongoose/Config.in"
> source "package/mongrel2/Config.in"
> source "package/monkey/Config.in"
> + source "package/mosh/Config.in"
> source "package/mosquitto/Config.in"
> source "package/mrouted/Config.in"
> source "package/mtr/Config.in"
> diff --git a/package/mosh/0001-remove-system-locale-calls.patch b/package/mosh/0001-remove-system-locale-calls.patch
> new file mode 100644
> index 0000000..114e6ed
> --- /dev/null
> +++ b/package/mosh/0001-remove-system-locale-calls.patch
> @@ -0,0 +1,46 @@
> +The locale command is not available on many systems. As this variable
> +is unused and appears to have been written with the intent of
> +displaying the locale settings to the user, it's not really necessary.
> +As this breaks Mosh on a lot of systems, it's best to remove the calls.
> +
> +Upstream status: refused, see: https://github.com/keithw/mosh/issues/650
> +
> +Signed-off-by: Christian Stewart <christian@paral.in>
> +---
> + src/frontend/mosh-server.cc | 4 +++-
> + src/frontend/stmclient.cc | 4 +++-
> + 2 files changed, 6 insertions(+), 2 deletions(-)
> +
> +diff --git a/src/frontend/mosh-server.cc b/src/frontend/mosh-server.cc
> +index 76ed2ed..0d2f222 100644
> +--- a/src/frontend/mosh-server.cc
> ++++ b/src/frontend/mosh-server.cc
> +@@ -313,7 +313,9 @@ int main( int argc, char *argv[] )
> + fprintf( stderr, "mosh-server needs a UTF-8 native locale to run.\n\n" );
> + fprintf( stderr, "Unfortunately, the local environment (%s) specifies\nthe character set \"%s\",\n\n", native_ctype.str().c_str(), native_charset.c_str() );
> + fprintf( stderr, "The client-supplied environment (%s) specifies\nthe character set \"%s\".\n\n", client_ctype.str().c_str(), client_charset.c_str() );
> +- int unused __attribute((unused)) = system( "locale" );
> ++
> ++ fprintf( stderr, "This is a buildroot system, 'locale' debug output has been removed." );
> ++
> + exit( 1 );
> + }
> + }
> +diff --git a/src/frontend/stmclient.cc b/src/frontend/stmclient.cc
> +index 9e4d916..06fac8b 100644
> +--- a/src/frontend/stmclient.cc
> ++++ b/src/frontend/stmclient.cc
> +@@ -85,7 +85,9 @@ void STMClient::init( void )
> +
> + fprintf( stderr, "mosh-client needs a UTF-8 native locale to run.\n\n" );
> + fprintf( stderr, "Unfortunately, the client's environment (%s) specifies\nthe character set \"%s\".\n\n", native_ctype.str().c_str(), native_charset.c_str() );
> +- int unused __attribute((unused)) = system( "locale" );
> ++
> ++ fprintf( stderr, "This is a buildroot system, 'locale' debug output has been removed." );
> ++
> + exit( 1 );
> + }
> +
> +--
> +2.1.4
> +
> diff --git a/package/mosh/Config.in b/package/mosh/Config.in
> new file mode 100644
> index 0000000..9841b2e
> --- /dev/null
> +++ b/package/mosh/Config.in
> @@ -0,0 +1,19 @@
> +comment "mosh needs a toolchain w/ threads"
> + depends on !BR2_TOOLCHAIN_HAS_THREADS
> +
> +config BR2_PACKAGE_MOSH
> + bool "mosh"
> + depends on BR2_arm || BR2_i386 || BR2_mipsel || BR2_x86_64 # protobuf
> + depends on BR2_TOOLCHAIN_HAS_THREADS #protobuf
> + depends on BR2_HOSTARCH = "x86_64" || BR2_HOSTARCH = "x86" # protobuf
> + select BR2_PACKAGE_PROTOBUF
> + select BR2_PACKAGE_NCURSES
> + select BR2_PACKAGE_OPENSSL
> + select BR2_PACKAGE_OPENSSH if !BR2_PACKAGE_DROPBEAR_CLIENT
> + select BR2_PACKAGE_ZLIB
> + help
> + Remote terminal application that supports intermittent connectivity,
> + allows roaming, and provides speculative local echo and line editing
> + of user keystrokes.
> +
> + https://mosh.mit.edu
> diff --git a/package/mosh/mosh.hash b/package/mosh/mosh.hash
> new file mode 100644
> index 0000000..ab925b5
> --- /dev/null
> +++ b/package/mosh/mosh.hash
> @@ -0,0 +1,2 @@
> +# From: http://mailman.mit.edu/pipermail/mosh-users/2015-July/000283.html
> +sha256 1af809e5d747c333a852fbf7acdbf4d354dc4bbc2839e3afe5cf798190074be3 mosh-1.2.5.tar.gz
> diff --git a/package/mosh/mosh.mk b/package/mosh/mosh.mk
> new file mode 100644
> index 0000000..dd1ef93
> --- /dev/null
> +++ b/package/mosh/mosh.mk
> @@ -0,0 +1,13 @@
> +################################################################################
> +#
> +# mosh
> +#
> +################################################################################
> +
> +MOSH_VERSION = 1.2.5
> +MOSH_SITE = https://mosh.mit.edu/
> +MOSH_DEPENDENCIES = zlib ncurses protobuf openssl
> +MOSH_LICENSE = GPLv3+ with exception
> +MOSH_LICENSE_FILES = COPYING COPYING.iOS
> +
> +$(eval $(autotools-package))
> --
> 2.1.4
>
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot
--
.-----------------.--------------------.------------------.--------------------.
| 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. |
'------------------------------^-------^------------------^--------------------'
^ permalink raw reply [flat|nested] 7+ messages in thread
* [Buildroot] [PATCH v8 1/1] mosh: new package
2015-07-26 16:34 [Buildroot] [PATCH v8 1/1] mosh: new package Christian Stewart
2015-07-26 17:01 ` Yann E. MORIN
@ 2015-07-26 17:35 ` Thomas Petazzoni
2015-07-26 18:02 ` Christian Stewart
2015-07-26 22:20 ` Baruch Siach
2 siblings, 1 reply; 7+ messages in thread
From: Thomas Petazzoni @ 2015-07-26 17:35 UTC (permalink / raw)
To: buildroot
Christian,
On Sun, 26 Jul 2015 09:34:07 -0700, Christian Stewart wrote:
> Adding mosh, the mobile shell. Mosh uses ssh or dropbear as an initial transport to
> start mosh-server which uses UDP to communicate with the client.
> Supports a predictive model to enhance performance on weak connections
> and compensate for general lag. Also supports disconnections and
> reconnections seamlessly.
>
> Signed-off-by: Christian Stewart <christian@paral.in>
I applied, with some more changes:
[Thomas:
- Add missing dependency on host-pkgconf, as noticed by Yann
E. Morin.
- Indicate that openssh/dropbear is a runtime dependency.
- Pass some variables in the configure environment to tell that SSP
support is not available when it isn't. Otherwise, it misdetects
the SSP support as being available, causing a build failure.]
The last item clearly meant that the build was never tested with
uClibc, which is our default C library. Please do at least a test build
with uClibc before submitting new packages.
Also, one comment below.
> diff --git a/package/mosh/0001-remove-system-locale-calls.patch b/package/mosh/0001-remove-system-locale-calls.patch
> new file mode 100644
> index 0000000..114e6ed
> --- /dev/null
> +++ b/package/mosh/0001-remove-system-locale-calls.patch
> @@ -0,0 +1,46 @@
> +The locale command is not available on many systems. As this variable
> +is unused and appears to have been written with the intent of
> +displaying the locale settings to the user, it's not really necessary.
> +As this breaks Mosh on a lot of systems, it's best to remove the calls.
> +
> +Upstream status: refused, see: https://github.com/keithw/mosh/issues/650
I think upstream refused it, because the approach of your patch is not
acceptable upstream.
> +Signed-off-by: Christian Stewart <christian@paral.in>
> +---
> + src/frontend/mosh-server.cc | 4 +++-
> + src/frontend/stmclient.cc | 4 +++-
> + 2 files changed, 6 insertions(+), 2 deletions(-)
> +
> +diff --git a/src/frontend/mosh-server.cc b/src/frontend/mosh-server.cc
> +index 76ed2ed..0d2f222 100644
> +--- a/src/frontend/mosh-server.cc
> ++++ b/src/frontend/mosh-server.cc
> +@@ -313,7 +313,9 @@ int main( int argc, char *argv[] )
> + fprintf( stderr, "mosh-server needs a UTF-8 native locale to run.\n\n" );
> + fprintf( stderr, "Unfortunately, the local environment (%s) specifies\nthe character set \"%s\",\n\n", native_ctype.str().c_str(), native_charset.c_str() );
> + fprintf( stderr, "The client-supplied environment (%s) specifies\nthe character set \"%s\".\n\n", client_ctype.str().c_str(), client_charset.c_str() );
> +- int unused __attribute((unused)) = system( "locale" );
> ++
> ++ fprintf( stderr, "This is a buildroot system, 'locale' debug output has been removed." );
Here is what your patch should do instead:
if (locale command exists)
int unused __attribute((unused)) = system( "locale" );
else
fprintf( stderr, "locale command not found, cannot show related debug output." );
I'm pretty sure this is going to make your patch much more acceptable
upstream. Can you retry with something like that?
Thanks!
Thomas
--
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
^ permalink raw reply [flat|nested] 7+ messages in thread
* [Buildroot] [PATCH v8 1/1] mosh: new package
2015-07-26 17:35 ` Thomas Petazzoni
@ 2015-07-26 18:02 ` Christian Stewart
2015-07-26 19:41 ` Thomas Petazzoni
0 siblings, 1 reply; 7+ messages in thread
From: Christian Stewart @ 2015-07-26 18:02 UTC (permalink / raw)
To: buildroot
Thomas,
No, they will reject it regardless of how it is coded. Checking a global
POSIX command exists is not worth it to them as it exists on most systems
anyway. They requested I make a patch for BR that outputs a special debug
line for buildroot users so they know what they're dealing with if someone
submits a log.
Thank you for the fixes!
Christian
On Sun, Jul 26, 2015, 10:35 AM Thomas Petazzoni <
thomas.petazzoni@free-electrons.com> wrote:
> Christian,
>
> On Sun, 26 Jul 2015 09:34:07 -0700, Christian Stewart wrote:
> > Adding mosh, the mobile shell. Mosh uses ssh or dropbear as an initial
> transport to
> > start mosh-server which uses UDP to communicate with the client.
> > Supports a predictive model to enhance performance on weak connections
> > and compensate for general lag. Also supports disconnections and
> > reconnections seamlessly.
> >
> > Signed-off-by: Christian Stewart <christian@paral.in>
>
> I applied, with some more changes:
>
> [Thomas:
> - Add missing dependency on host-pkgconf, as noticed by Yann
> E. Morin.
> - Indicate that openssh/dropbear is a runtime dependency.
> - Pass some variables in the configure environment to tell that SSP
> support is not available when it isn't. Otherwise, it misdetects
> the SSP support as being available, causing a build failure.]
>
> The last item clearly meant that the build was never tested with
> uClibc, which is our default C library. Please do at least a test build
> with uClibc before submitting new packages.
>
> Also, one comment below.
>
>
> > diff --git a/package/mosh/0001-remove-system-locale-calls.patch
> b/package/mosh/0001-remove-system-locale-calls.patch
> > new file mode 100644
> > index 0000000..114e6ed
> > --- /dev/null
> > +++ b/package/mosh/0001-remove-system-locale-calls.patch
> > @@ -0,0 +1,46 @@
> > +The locale command is not available on many systems. As this variable
> > +is unused and appears to have been written with the intent of
> > +displaying the locale settings to the user, it's not really necessary.
> > +As this breaks Mosh on a lot of systems, it's best to remove the calls.
> > +
> > +Upstream status: refused, see:
> https://github.com/keithw/mosh/issues/650
>
> I think upstream refused it, because the approach of your patch is not
> acceptable upstream.
>
> > +Signed-off-by: Christian Stewart <christian@paral.in>
> > +---
> > + src/frontend/mosh-server.cc | 4 +++-
> > + src/frontend/stmclient.cc | 4 +++-
> > + 2 files changed, 6 insertions(+), 2 deletions(-)
> > +
> > +diff --git a/src/frontend/mosh-server.cc b/src/frontend/mosh-server.cc
> > +index 76ed2ed..0d2f222 100644
> > +--- a/src/frontend/mosh-server.cc
> > ++++ b/src/frontend/mosh-server.cc
> > +@@ -313,7 +313,9 @@ int main( int argc, char *argv[] )
> > + fprintf( stderr, "mosh-server needs a UTF-8 native locale to
> run.\n\n" );
> > + fprintf( stderr, "Unfortunately, the local environment (%s)
> specifies\nthe character set \"%s\",\n\n", native_ctype.str().c_str(),
> native_charset.c_str() );
> > + fprintf( stderr, "The client-supplied environment (%s)
> specifies\nthe character set \"%s\".\n\n", client_ctype.str().c_str(),
> client_charset.c_str() );
> > +- int unused __attribute((unused)) = system( "locale" );
> > ++
> > ++ fprintf( stderr, "This is a buildroot system, 'locale' debug
> output has been removed." );
>
> Here is what your patch should do instead:
>
> if (locale command exists)
> int unused __attribute((unused)) = system( "locale" );
> else
> fprintf( stderr, "locale command not found, cannot show
> related debug output." );
>
> I'm pretty sure this is going to make your patch much more acceptable
> upstream. Can you retry with something like that?
>
> Thanks!
>
> Thomas
> --
> Thomas Petazzoni, CTO, Free Electrons
> Embedded Linux, Kernel and Android engineering
> http://free-electrons.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20150726/47cb4e8e/attachment.html>
^ permalink raw reply [flat|nested] 7+ messages in thread
* [Buildroot] [PATCH v8 1/1] mosh: new package
2015-07-26 18:02 ` Christian Stewart
@ 2015-07-26 19:41 ` Thomas Petazzoni
2015-07-27 21:38 ` Christian Stewart
0 siblings, 1 reply; 7+ messages in thread
From: Thomas Petazzoni @ 2015-07-26 19:41 UTC (permalink / raw)
To: buildroot
Dear Christian Stewart,
Please don't top-post, thanks!
On Sun, 26 Jul 2015 18:02:29 +0000, Christian Stewart wrote:
> No, they will reject it regardless of how it is coded. Checking a global
> POSIX command exists is not worth it to them as it exists on most systems
> anyway. They requested I make a patch for BR that outputs a special debug
> line for buildroot users so they know what they're dealing with if someone
> submits a log.
Can you try anyway? Because what I'm proposing is significantly
different from what you proposed in your pull request. My proposal has
the advantage of keeping the locale output when 'locale' is available
*and* warn that locale is not available otherwise. Basically it exactly
matches the requirements of the upstream mosh developers.
We don't like to keep patches in Buildroot forever, so having an
upstreamable solution is important.
Thanks,
Thomas
--
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
^ permalink raw reply [flat|nested] 7+ messages in thread
* [Buildroot] [PATCH v8 1/1] mosh: new package
2015-07-26 16:34 [Buildroot] [PATCH v8 1/1] mosh: new package Christian Stewart
2015-07-26 17:01 ` Yann E. MORIN
2015-07-26 17:35 ` Thomas Petazzoni
@ 2015-07-26 22:20 ` Baruch Siach
2 siblings, 0 replies; 7+ messages in thread
From: Baruch Siach @ 2015-07-26 22:20 UTC (permalink / raw)
To: buildroot
Hi Christian,
On Sun, Jul 26, 2015 at 09:34:07AM -0700, Christian Stewart wrote:
> diff --git a/package/mosh/Config.in b/package/mosh/Config.in
> new file mode 100644
> index 0000000..9841b2e
> --- /dev/null
> +++ b/package/mosh/Config.in
> @@ -0,0 +1,19 @@
> +comment "mosh needs a toolchain w/ threads"
> + depends on !BR2_TOOLCHAIN_HAS_THREADS
The dependencies comment should depend on supported arches. Since this patch
has already been applied, I'll send a follow-up.
> +config BR2_PACKAGE_MOSH
> + bool "mosh"
> + depends on BR2_arm || BR2_i386 || BR2_mipsel || BR2_x86_64 # protobuf
> + depends on BR2_TOOLCHAIN_HAS_THREADS #protobuf
> + depends on BR2_HOSTARCH = "x86_64" || BR2_HOSTARCH = "x86" # protobuf
> + select BR2_PACKAGE_PROTOBUF
> + select BR2_PACKAGE_NCURSES
> + select BR2_PACKAGE_OPENSSL
> + select BR2_PACKAGE_OPENSSH if !BR2_PACKAGE_DROPBEAR_CLIENT
> + select BR2_PACKAGE_ZLIB
> + help
> + Remote terminal application that supports intermittent connectivity,
> + allows roaming, and provides speculative local echo and line editing
> + of user keystrokes.
> +
> + https://mosh.mit.edu
baruch
--
http://baruch.siach.name/blog/ ~. .~ Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
- baruch at tkos.co.il - tel: +972.2.679.5364, http://www.tkos.co.il -
^ permalink raw reply [flat|nested] 7+ messages in thread
* [Buildroot] [PATCH v8 1/1] mosh: new package
2015-07-26 19:41 ` Thomas Petazzoni
@ 2015-07-27 21:38 ` Christian Stewart
0 siblings, 0 replies; 7+ messages in thread
From: Christian Stewart @ 2015-07-27 21:38 UTC (permalink / raw)
To: buildroot
Hi Thomas,
On Sun, Jul 26, 2015 at 12:41 PM Thomas Petazzoni <
thomas.petazzoni@free-electrons.com> wrote:
> Dear Christian Stewart,
>
[snip]
> Can you try anyway? Because what I'm proposing is significantly
> different from what you proposed in your pull request. My proposal has
> the advantage of keeping the locale output when 'locale' is available
> *and* warn that locale is not available otherwise. Basically it exactly
> matches the requirements of the upstream mosh developers.
>
Their opinion is that since Mosh is going to exit immediately after
"locale" is called anyways, they would prefer to just have the default
"command not found" be printed rather than check that the command exists at
all.
> We don't like to keep patches in Buildroot forever, so having an
> upstreamable solution is important.
>
Honestly it's not as bad as I thought originally to have this error, you
could probably safely delete the patch without any upstream changes or
repercussions.
Christian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20150727/9a055602/attachment.html>
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2015-07-27 21:38 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-07-26 16:34 [Buildroot] [PATCH v8 1/1] mosh: new package Christian Stewart
2015-07-26 17:01 ` Yann E. MORIN
2015-07-26 17:35 ` Thomas Petazzoni
2015-07-26 18:02 ` Christian Stewart
2015-07-26 19:41 ` Thomas Petazzoni
2015-07-27 21:38 ` Christian Stewart
2015-07-26 22:20 ` Baruch Siach
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox