Re: CVS commit: src/sys/dev/pci

[Christos Zoulas]

In article <20150217101125.1591...@cvs.netbsd.org>,
SAITOH Masanobu  wrote:
>-=-=-=-=-=-
>
>Module Name:   src
>Committed By:  msaitoh
>Date:  Tue Feb 17 10:11:25 UTC 2015
>
>Modified Files:
>   src/sys/dev/pci: if_bge.c
>
>Log Message:
> Print bit setting of bge_asf_mode if BGE_DEBUG is set.
>
>
>To generate a diff of this commit:
>cvs rdiff -u -r1.278 -r1.279 src/sys/dev/pci/if_bge.c
>
>Please note that diffs are not public domain; they are subject to the
>copyright notices on the relevant files.
>
>
>-=-=-=-=-=-
>
>Modified files:
>
>Index: src/sys/dev/pci/if_bge.c
>diff -u src/sys/dev/pci/if_bge.c:1.278 src/sys/dev/pci/if_bge.c:1.279
>--- src/sys/dev/pci/if_bge.c:1.278 Wed Feb 11 23:07:13 2015
>+++ src/sys/dev/pci/if_bge.c   Tue Feb 17 10:11:24 2015
>@@ -1,4 +1,4 @@
>-/*$NetBSD: if_bge.c,v 1.278 2015/02/11 23:07:13 msaitoh Exp $ */
>+/*$NetBSD: if_bge.c,v 1.279 2015/02/17 10:11:24 msaitoh Exp $ */
> 
> /*
>  * Copyright (c) 2001 Wind River Systems
>@@ -79,7 +79,7 @@
>  */
> 
> #include 
>-__KERNEL_RCSID(0, "$NetBSD: if_bge.c,v 1.278 2015/02/11 23:07:13
>msaitoh Exp $");
>+__KERNEL_RCSID(0, "$NetBSD: if_bge.c,v 1.279 2015/02/17 10:11:24
>msaitoh Exp $");
> 
> #include 
> #include 
>@@ -6053,6 +6053,7 @@ bge_debug_info(struct bge_softc *sc)
>   if (sc->bge_flags & BGEF_TSO)
>   printf(" - TSO\n");
> 
>+  /* PHY related */
>   if (sc->bge_phy_flags & BGEPHYF_NO_3LED)
>   printf(" - No 3 LEDs\n");
>   if (sc->bge_phy_flags & BGEPHYF_CRC_BUG)
>@@ -6069,6 +6070,14 @@ bge_debug_info(struct bge_softc *sc)
>   printf(" - adjust trim\n");
>   if (sc->bge_phy_flags & BGEPHYF_NO_WIRESPEED)
>   printf(" - no wirespeed\n");
>+
>+  /* ASF related */
>+  if (sc->bge_asf_mode & ASF_ENABLE)
>+  printf(" - ASF enable\n");
>+  if (sc->bge_asf_mode & ASF_NEW_HANDSHARE)
>+  printf(" - ASF new handshake\n");
>+  if (sc->bge_asf_mode & ASF_STACKUP)
>+  printf(" - ASF stackup\n");

Candidates for snprintb()?

christos

Re: CVS commit: [netbsd-7] xsrc/external/mit/MesaLib/dist/src/mapi

[David Laight]

On Sat, Feb 14, 2015 at 08:14:01AM +, Soren Jacobsen wrote:
> Module Name:  xsrc
> Committed By: snj
> Date: Sat Feb 14 08:14:01 UTC 2015
> 
> Modified Files:
>   xsrc/external/mit/MesaLib/dist/src/mapi [netbsd-7]: entry.c
> 
> Log Message:
> Pull up following revision(s) (requested by mrg in ticket #514):
>   external/mit/MesaLib/dist/src/mapi/entry.c: revision 1.2
> disable the use of 32 bit x86 asm code here, it (like the 64 bit code),
> does not work on netbsd and probably needs an x86 guru to fix.

Wouldn't it be more sensible to undefine USE_X86_ASM in the maefile?

David

-- 
David Laight: da...@l8s.co.uk

re: CVS commit: [netbsd-7] xsrc/external/mit/MesaLib/dist/src/mapi

[matthew green]

David Laight writes:
> On Sat, Feb 14, 2015 at 08:14:01AM +, Soren Jacobsen wrote:
> > Module Name:xsrc
> > Committed By:   snj
> > Date:   Sat Feb 14 08:14:01 UTC 2015
> > 
> > Modified Files:
> > xsrc/external/mit/MesaLib/dist/src/mapi [netbsd-7]: entry.c
> > 
> > Log Message:
> > Pull up following revision(s) (requested by mrg in ticket #514):
> > external/mit/MesaLib/dist/src/mapi/entry.c: revision 1.2
> > disable the use of 32 bit x86 asm code here, it (like the 64 bit code),
> > does not work on netbsd and probably needs an x86 guru to fix.
> 
> Wouldn't it be more sensible to undefine USE_X86_ASM in the maefile?

there is a whole bunch of other stuff that seems to work fine,
it is only this TLS-related stuff that is problematic (i'm
guessing it's linux-specific code here.)

we're getting all the other benefit, it seemed better to only
disable this one case.


.mrg.