Thomas Pfaff | 1 Dec 16:46 2008
Picon

Re: azalia: differentiate between line-in and line-out

On Sun, 30 Nov 2008 22:29:22 +0000
Jacob Meuser <jakemsr <at> sdf.lonestar.org> wrote:

> this has long bugged me.  currently we call both line-in and line-out
> just "line".  this make line-in, "line-in" and line-out "line".

Seems to work fine for me.

$ diff -u mixerctl-old mixerctl-new
--- mixerctl-old	Mon Dec  1 16:21:26 2008
+++ mixerctl-new	Mon Dec  1 16:42:16 2008
 <at>  <at>  -4,13 +4,13  <at>  <at> 
 inputs.dac5=126,126 
 inputs.dac6=126,126 
 inputs.sel_source=adc2  [ adc2 adc3 adc4 ]
-record.sel2_source=sel13  [ sel13 cd line5 line6 sel18 mix3 ]
+record.sel2_source=sel13  [ sel13 cd line3 line4 sel18 mix3 ]
 outputs.sel2_mute=off  [ off on ]
 outputs.sel2=124,124 
-record.sel3_source=sel13  [ sel13 cd line5 line6 sel18 mix3 ]
+record.sel3_source=sel13  [ sel13 cd line3 line4 sel18 mix3 ]
 outputs.sel3_mute=off  [ off on ]
 outputs.sel3=124,124 
-inputs.sel4_source=sel13  [ sel13 cd line5 line6 sel18 mix3 ]
+inputs.sel4_source=sel13  [ sel13 cd line3 line4 sel18 mix3 ]
 outputs.sel4_mute=off  [ off on ]
 outputs.sel4=124,124 
 inputs.beep_mute=off  [ off on ]
 <at>  <at>  -24,10 +24,10  <at>  <at> 
 outputs.mic_mute=off  [ off on ]
(Continue reading)

David Gwynne | 2 Dec 03:45 2008
Picon

em diff for testing

this is the diff for em that i committed last week during h2k8 with
some extra changes that hopefully fix the issues people experienced
with it.

the goal of this diff is to reduce the up front mbuf allocation on
the rx ring, thereby saving a lot of kva on systems that dont need
a full rx ring. the other advantage is on livelocked systems, em
can be slowed down to allow the system to continue running.

thanks to jsing for pointing out some issues with the original diff.

Index: if_em.c
===================================================================
RCS file: /cvs/src/sys/dev/pci/if_em.c,v
retrieving revision 1.199
diff -u -p -r1.199 if_em.c
--- if_em.c	29 Nov 2008 10:23:29 -0000	1.199
+++ if_em.c	2 Dec 2008 02:36:08 -0000
 <at>  <at>  -31,7 +31,7  <at>  <at>  POSSIBILITY OF SUCH DAMAGE.

 ***************************************************************************/

-/* $OpenBSD: if_em.c,v 1.199 2008/11/29 10:23:29 sthen Exp $ */
+/* $OpenBSD: if_em.c,v 1.197 2008/11/26 00:14:48 dlg Exp $ */
 /* $FreeBSD: if_em.c,v 1.46 2004/09/29 18:28:28 mlaier Exp $ */

 #include <dev/pci/if_em.h>
 <at>  <at>  -164,6 +164,12  <at>  <at>  void em_update_stats_counters(struct em_
 void em_txeof(struct em_softc *);
 int  em_allocate_receive_structures(struct em_softc *);
(Continue reading)

Pierre Riteau | 2 Dec 23:18 2008
Picon

Re: azalia: differentiate between line-in and line-out

On Sun, Nov 30, 2008 at 10:29:22PM +0000, Jacob Meuser wrote:
> this has long bugged me.  currently we call both line-in and line-out
> just "line".  this make line-in, "line-in" and line-out "line".
> 
> I'm pretty sure the reason only "line" was used was to avoid name
> truncation, but I don't think that is a problem anymore.  but I
> can't know without some tests ...

Seems OK for me too with:
azalia0 at pci3 dev 1 function 0 "VIA HD Audio" rev 0x10: apic 2 int 17 (irq 5)
azalia0: codecs: Realtek ALC888

--- mixerctl.old	Tue Dec  2 23:08:29 2008
+++ mixerctl.new	Tue Dec  2 23:14:09 2008
 <at>  <at>  -2,9 +2,9  <at>  <at> 
 record.adc=120,120 
 record.adc2_mute=off  [ off on ]
 record.adc2=120,120 
-inputs.mix_source=mic,line5,cd,line,line2,line3,line4  { mic line5 cd line line2 line3 line4 }
+inputs.mix_source=mic,line-in,cd,line,line2,line3,line4  { mic line-in cd line line2 line3 line4 }
 inputs.mix_mic=120,120 
-inputs.mix_line5=120,120 
+inputs.mix_line-in=120,120 
 inputs.mix_cd=120,120 
 inputs.mix_line=120,120 
 inputs.mix_line2=120,120 
 <at>  <at>  -41,13 +41,13  <at>  <at> 
 inputs.mic=85,85 
 outputs.mic_dir=input  [ input output ]
 outputs.mic_boost=off  [ off on ]
(Continue reading)

Anathae Townsend | 2 Dec 23:52 2008

possible spelling error diff for lockd_lock.c

Index: lockd_lock.c
===================================================================
RCS file: /cvs/src/usr.sbin/rpc.lockd/lockd_lock.c,v
retrieving revision 1.5
diff -u -r1.5 lockd_lock.c
--- lockd_lock.c        25 Nov 2008 00:00:52 -0000      1.5
+++ lockd_lock.c        2 Dec 2008 22:51:05 -0000
 <at>  <at>  -503,7 +503,7  <at>  <at> 

 /*
  *
- * try to acquire the lock described by fl. Eventually fock a child to do a
+ * try to acquire the lock described by fl. Eventually fork a child to do a
  * blocking lock if allowed and required.
  */

Brad | 3 Dec 11:14 2008

em(4) diff for ICH9 IGP M needs testing.

If you have a system with the Intel ICH9 IGP M or ICH9 IGP M AMT chipset
please test the following diff. I'm especially interested in having this
tested with the ICH9 IGP M. Please provide me with a dmesg.

Index: if_em.c
===================================================================
RCS file: /cvs/src/sys/dev/pci/if_em.c,v
retrieving revision 1.198
diff -u -p -r1.198 if_em.c
--- if_em.c	28 Nov 2008 02:44:17 -0000	1.198
+++ if_em.c	28 Nov 2008 04:50:25 -0000
 <at>  <at>  -127,7 +127,9  <at>  <at>  const struct pci_matchid em_devices[] = 
 	{ PCI_VENDOR_INTEL, PCI_PRODUCT_INTEL_ICH9_IFE_G },
 	{ PCI_VENDOR_INTEL, PCI_PRODUCT_INTEL_ICH9_IFE_GT },
 	{ PCI_VENDOR_INTEL, PCI_PRODUCT_INTEL_ICH9_IGP_AMT },
-	{ PCI_VENDOR_INTEL, PCI_PRODUCT_INTEL_ICH9_IGP_C }
+	{ PCI_VENDOR_INTEL, PCI_PRODUCT_INTEL_ICH9_IGP_C },
+	{ PCI_VENDOR_INTEL, PCI_PRODUCT_INTEL_ICH9_IGP_M },
+	{ PCI_VENDOR_INTEL, PCI_PRODUCT_INTEL_ICH9_IGP_M_AMT }
 };

 /*********************************************************************
Index: if_em_hw.c
===================================================================
RCS file: /cvs/src/sys/dev/pci/if_em_hw.c,v
retrieving revision 1.30
diff -u -p -r1.30 if_em_hw.c
--- if_em_hw.c	24 Sep 2008 19:12:59 -0000	1.30
+++ if_em_hw.c	28 Nov 2008 05:10:34 -0000
 <at>  <at>  -460,6 +460,8  <at>  <at>  em_set_mac_type(struct em_hw *hw)
(Continue reading)

Dinar Talypov | 3 Dec 11:24 2008
Picon

Re: [PATCH] if_gre: add ALTQ support second version

Hi,

> 
> There are quite a few KNF issues in the code (gre_start is indented in a
> very strange way) plus some errors that I show inline. Maybe you should
> split gre_start() into grep_start() and gre_sendout() or so. So you don't
> need to indent all of the gre_output() code.
>  
> 
Here is the second version of the patch.

 Best regards,
 Dinar Talypov

Index: if_gre.h
===================================================================
RCS file: /cvs/src/sys/net/if_gre.h,v
retrieving revision 1.11
diff -u -r1.11 if_gre.h
--- if_gre.h	26 Jun 2008 05:42:20 -0000	1.11
+++ if_gre.h	25 Nov 2008 08:39:40 -0000
 <at>  <at>  -141,6 +141,7  <at>  <at> 
 int     gre_ioctl(struct ifnet *, u_long, caddr_t);
 int     gre_output(struct ifnet *, struct mbuf *, struct sockaddr *,
 	    struct rtentry *);
+void 	gre_start(struct ifnet *);
 u_int16_t gre_in_cksum(u_int16_t *, u_int);
 #endif /* _KERNEL */
 #endif /* _NET_IF_GRE_H_ */
Index: if_gre.c
(Continue reading)

Picon

Re: em(4) diff for ICH9 IGP M needs testing.

Brad, 

I have been using jcs <at>  hack for a while. 
It's available here:

http://jcs.org/patches/openbsd-em_match_thinkpad_x200.diff

I noticed that is basically equal to this one you have sent to list, but it just have my card.
I didn't have any problem with it (low traffic, normal usage).

Here's my dmesg:

OpenBSD 4.4-current (GENERIC.MP) #2: Wed Nov 26 00:17:53 BRST 2008
    root <at> dub:/usr/src/sys/arch/amd64/compile/GENERIC.MP
real mem = 3694911488 (3523MB)
avail mem = 3585921024 (3419MB)
mainbus0 at root
bios0 at mainbus0: SMBIOS rev. 2.4  <at>  0xf6590 (57 entries)
bios0: vendor Dell Inc. version "A06" date 10/10/2008
bios0: Dell Inc. Latitude E6400
acpi0 at bios0: rev 2
acpi0: tables DSDT FACP HPET DMAR APIC ASF! MCFG SLIC TCPA SSDT
acpi0: wakeup devices PCI0(S4) PCIE(S4) USB1(S0) USB2(S0) USB3(S0) USB4(S0) USB5(S0) USB6(S0)
EHC2(S0) EHCI(S0) AZAL(S3) RP01(S4) RP02(S4) RP03(S4) RP04(S3) RP05(S3) RP06(S5) LID_(S3) PBTN(S4)
acpitimer0 at acpi0: 3579545 Hz, 24 bits
acpihpet0 at acpi0: 14318179 Hz
acpimadt0 at acpi0 addr 0xfee00000: PC-AT compat
cpu0 at mainbus0: apid 0 (boot processor)
cpu0: Intel(R) Core(TM)2 Duo CPU P8400  <at>  2.26GHz, 2261.33 MHz
cpu0: FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,SBF,SSE3,MWAIT,DS-CPL,VMX,SMX,EST,TM2,CX16,xTPR,NXE,LONG
(Continue reading)

Jason McIntyre | 3 Dec 16:17 2008
Picon

Re: possible spelling error diff for lockd_lock.c

On Tue, Dec 02, 2008 at 03:52:25PM -0700, Anathae Townsend wrote:
> Index: lockd_lock.c
> ===================================================================
> RCS file: /cvs/src/usr.sbin/rpc.lockd/lockd_lock.c,v
> retrieving revision 1.5
> diff -u -r1.5 lockd_lock.c
> --- lockd_lock.c        25 Nov 2008 00:00:52 -0000      1.5
> +++ lockd_lock.c        2 Dec 2008 22:51:05 -0000
>  <at>  <at>  -503,7 +503,7  <at>  <at> 
> 
>  /*
>   *
> - * try to acquire the lock described by fl. Eventually fock a child to do a
> + * try to acquire the lock described by fl. Eventually fork a child to do a
>   * blocking lock if allowed and required.
>   */

fixed, thanks.
jmc

Johan Borch | 3 Dec 21:42 2008
Picon

Re: [patch] pf PPTP nat passthrough patch.

Ermal LuC'i <ermal.luci <at> gmail.com> writes:

> 
> Hello,
> 
> after sometime thinking about this i putted together a patch for NAT
> that should workaround PPTP NAT passthrough problems.
> 

Hi!

Any news regarding pptp-proxy for openbsd?

Regards
Johan

hyjial | 3 Dec 21:01 2008
Picon

man(1) support for Xorg's ``t''s

Hi list,
An issue has been raised on misc some months ago about man(1) not
making the manpage
go through tbl(1) when formatting manpages which have their
format on the first line of their
file rather than in the suffix of their file
name.
I added support for it through a new keyword in man.conf(5). The
man.conf(5) patch gives a
hint about how it works. Basically, if the first
line of the file contains single letters---separated
by spaces---the
corresponding command will be inserted in the command line passed to
system(3)
right before nroff. The letters of the first line before the first space are
ignored.
Hope, this will fit. I am still learning the art of programming and
would appreciate feedback
as to what is flawed in this chunck of code.
I don't
speak roff(1) nor mdoc yet. The manpage would certainly need to be updated.
By
the way, there is a man.conf in /cvs/openbsd/src/usr.bin/man which is
out-dated and
not installed any more. Would that be wise to delete it ?
Regards.

Index: man.conf
===================================================================
RCS file:
(Continue reading)


Gmane