svn commit: r304518 - stable/9/contrib/binutils/bfd

[Pedro F. Giffuni]

Author: pfg
Date: Sat Aug 20 02:21:45 2016
New Revision: 304518
URL: https://svnweb.freebsd.org/changeset/base/304518

Log:
  MFC r303147
  binutils: fix "Bad value" error in bfd for MIPS when using -Bsymbolic.
  
  From OpenBSD's log:
  
  Inspired by https://sourceware.org/ml/binutils/2010-08/msg00333.html,
  but expressed differently so there are no GPLv3 issues.
  
  Obtained from:OpenBSD (CVS rev. 1.7)

Modified:
  stable/9/contrib/binutils/bfd/elfxx-mips.c
Directory Properties:
  stable/9/contrib/binutils/   (props changed)

Modified: stable/9/contrib/binutils/bfd/elfxx-mips.c
==
--- stable/9/contrib/binutils/bfd/elfxx-mips.c  Sat Aug 20 02:18:44 2016
(r304517)
+++ stable/9/contrib/binutils/bfd/elfxx-mips.c  Sat Aug 20 02:21:45 2016
(r304518)
@@ -4797,7 +4797,7 @@ mips_elf_create_dynamic_relocation (bfd 
   /* We must now calculate the dynamic symbol table index to use
  in the relocation.  */
   if (h != NULL
-  && (!h->root.def_regular
+  && (sec == NULL || !h->root.def_regular
  || (info->shared && !info->symbolic && !h->root.forced_local)))
 {
   indx = h->root.dynindx;
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304517 - stable/10/contrib/binutils/bfd

[Pedro F. Giffuni]

Author: pfg
Date: Sat Aug 20 02:18:44 2016
New Revision: 304517
URL: https://svnweb.freebsd.org/changeset/base/304517

Log:
  MFC r303147
  binutils: fix "Bad value" error in bfd for MIPS when using -Bsymbolic.
  
  From OpenBSD's log:
  
  Inspired by https://sourceware.org/ml/binutils/2010-08/msg00333.html,
  but expressed differently so there are no GPLv3 issues.
  
  Obtained from:OpenBSD (CVS rev. 1.7)

Modified:
  stable/10/contrib/binutils/bfd/elfxx-mips.c
Directory Properties:
  stable/10/   (props changed)

Modified: stable/10/contrib/binutils/bfd/elfxx-mips.c
==
--- stable/10/contrib/binutils/bfd/elfxx-mips.c Sat Aug 20 02:14:14 2016
(r304516)
+++ stable/10/contrib/binutils/bfd/elfxx-mips.c Sat Aug 20 02:18:44 2016
(r304517)
@@ -4801,7 +4801,7 @@ mips_elf_create_dynamic_relocation (bfd 
   /* We must now calculate the dynamic symbol table index to use
  in the relocation.  */
   if (h != NULL
-  && (!h->root.def_regular
+  && (sec == NULL || !h->root.def_regular
  || (info->shared && !info->symbolic && !h->root.forced_local)))
 {
   indx = h->root.dynindx;
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304516 - stable/11/contrib/binutils/bfd

[Pedro F. Giffuni]

Author: pfg
Date: Sat Aug 20 02:14:14 2016
New Revision: 304516
URL: https://svnweb.freebsd.org/changeset/base/304516

Log:
  MFC r303147
  binutils: fix "Bad value" error in bfd for MIPS when using -Bsymbolic.
  
  From OpenBSD's log:
  
  Inspired by https://sourceware.org/ml/binutils/2010-08/msg00333.html,
  but expressed differently so there are no GPLv3 issues.
  
  Obtained from:OpenBSD (CVS rev. 1.7)

Modified:
  stable/11/contrib/binutils/bfd/elfxx-mips.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/contrib/binutils/bfd/elfxx-mips.c
==
--- stable/11/contrib/binutils/bfd/elfxx-mips.c Sat Aug 20 02:06:59 2016
(r304515)
+++ stable/11/contrib/binutils/bfd/elfxx-mips.c Sat Aug 20 02:14:14 2016
(r304516)
@@ -4801,7 +4801,7 @@ mips_elf_create_dynamic_relocation (bfd 
   /* We must now calculate the dynamic symbol table index to use
  in the relocation.  */
   if (h != NULL
-  && (!h->root.def_regular
+  && (sec == NULL || !h->root.def_regular
  || (info->shared && !info->symbolic && !h->root.forced_local)))
 {
   indx = h->root.dynindx;
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304515 - in stable/11: contrib/openresolv sbin/resolvconf

[Pedro F. Giffuni]

Author: pfg
Date: Sat Aug 20 02:06:59 2016
New Revision: 304515
URL: https://svnweb.freebsd.org/changeset/base/304515

Log:
  MFC r303062, r303567, r303593:
  MFV r298167, r300962, r303048:
  openresolv(8): update to version 3.8.1.
  
  Among the new features it attempts to support alternative init systems.

Modified:
  stable/11/contrib/openresolv/Makefile
  stable/11/contrib/openresolv/configure
  stable/11/contrib/openresolv/dnsmasq.in
  stable/11/contrib/openresolv/libc.in
  stable/11/contrib/openresolv/named.in
  stable/11/contrib/openresolv/resolvconf.8.in
  stable/11/contrib/openresolv/resolvconf.conf.5.in
  stable/11/contrib/openresolv/resolvconf.in
  stable/11/contrib/openresolv/unbound.in
  stable/11/sbin/resolvconf/Makefile
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/contrib/openresolv/Makefile
==
--- stable/11/contrib/openresolv/Makefile   Sat Aug 20 00:55:58 2016
(r304514)
+++ stable/11/contrib/openresolv/Makefile   Sat Aug 20 02:06:59 2016
(r304515)
@@ -1,5 +1,4 @@
 PKG=   openresolv
-VERSION=   3.7.3
 
 # Nasty hack so that make clean works without configure being run
 _CONFIG_MK!=   test -e config.mk && echo config.mk || echo config-null.mk
@@ -10,14 +9,12 @@ SBINDIR?=/sbin
 SYSCONFDIR?=   /etc
 LIBEXECDIR?=   /libexec/resolvconf
 VARDIR?=   /var/run/resolvconf
-RCDIR?=/etc/rc.d
-RESTARTCMD?=   if ${RCDIR}/\1 status >/dev/null 2>\&1; then \
-   ${RCDIR}/\1 restart; \
-   fi
 
 INSTALL?=  install
 SED?=  sed
 
+VERSION!=  ${SED} -n 's/OPENRESOLV_VERSION="\(.*\)".*/\1/p' resolvconf.in
+
 BINMODE?=  0755
 DOCMODE?=  0644
 MANMODE?=  0444
@@ -33,7 +30,9 @@ SED_SYSCONFDIR=   -e 's:@SYSCONFDIR@:${SY
 SED_LIBEXECDIR=-e 's:@LIBEXECDIR@:${LIBEXECDIR}:g'
 SED_VARDIR=-e 's:@VARDIR@:${VARDIR}:g'
 SED_RCDIR= -e 's:@RCDIR@:${RCDIR}:g'
-SED_RESTARTCMD=-e 's:@RESTARTCMD \(.*\)@:${RESTARTCMD}:g'
+SED_RESTARTCMD=-e 's:@RESTARTCMD@:${RESTARTCMD}:g'
+SED_RCDIR= -e 's:@RCDIR@:${RCDIR}:g'
+SED_STATUSARG= -e 's:@STATUSARG@:${STATUSARG}:g'
 
 DISTPREFIX?=   ${PKG}-${VERSION}
 DISTFILEGZ?=   ${DISTPREFIX}.tar.gz
@@ -44,9 +43,10 @@ FOSSILID?=   current
 
 all: ${TARGET}
 
-.in:
+.in: Makefile ${CONFIG_MK}
${SED}  ${SED_SBINDIR} ${SED_SYSCONFDIR} ${SED_LIBEXECDIR} \
-   ${SED_VARDIR} ${SED_RCDIR} ${SED_RESTARTCMD} \
+   ${SED_VARDIR} \
+   ${SED_RCDIR} ${SED_RESTARTCMD} ${SED_RCDIR} ${SED_STATUSARG} \
$< > $@
 
 clean:

Modified: stable/11/contrib/openresolv/configure
==
--- stable/11/contrib/openresolv/configure  Sat Aug 20 00:55:58 2016
(r304514)
+++ stable/11/contrib/openresolv/configure  Sat Aug 20 02:06:59 2016
(r304515)
@@ -8,6 +8,7 @@ HOST=
 TARGET=
 RESTARTCMD=
 RCDIR=
+STATUSARG=
 
 for x do
opt=${x%%=*}
@@ -33,6 +34,8 @@ for x do
--target) TARGET=$var;;
--libdir) LIBDIR=$var;;
--restartcmd) RESTARTCMD=$var;;
+   --rcdir) RCDIR=$var;;
+   --statusarg) STATUSARG=$var;;
--includedir) eval INCLUDEDIR="$INCLUDEDIR${INCLUDEDIR:+ }$var";;
--datadir|--infodir) ;; # ignore autotools
--disable-maintainer-mode|--disable-dependency-tracking) ;;
@@ -117,7 +120,17 @@ echo "Configuring openresolv for ... $OS
 rm -rf $CONFIG_MK
 echo "# $OS" >$CONFIG_MK
 
-for x in SYSCONFDIR SBINDIR LIBEXECDIR VARDIR MANDIR; do
+# On FreeBSD, /etc/init.d/foo status returns 0 if foo is not enabled
+# regardless of if it's not running.
+# So we force onestatus to work around this silly bug.
+if [ -z "$STATUSARG" ]; then
+   case "$OS" in
+   freebsd*)   STATUSARG="onestatus";;
+   esac
+fi
+
+for x in SYSCONFDIR SBINDIR LIBEXECDIR VARDIR MANDIR RESTARTCMD RCDIR STATUSARG
+do
eval v=\$$x
# Make files look nice for import
l=$((10 - ${#x}))
@@ -126,96 +139,6 @@ for x in SYSCONFDIR SBINDIR LIBEXECDIR V
echo "$x=$t $v" >>$CONFIG_MK
 done
 
-if [ -z "$RESTARTCMD" ]; then
-   printf "Checking for systemd ... "
-   if [ -x /bin/systemctl ]; then
-   RESTARTCMD="/bin/systemctl try-restart \1"
-   echo "yes"
-   elif [ -x /usr/bin/systemctl ]; then
-   RESTARTCMD="/usr/bin/systemctl try-restart \1"
-   echo "yes"
-   else
-   echo "no"
-   fi
-fi
-
-# Arch upgraded to systemd, so this check has to be just after systemd
-# but higher than the others
-if [ -z "$RESTARTCMD" ]; then
-   printf "Checking for Arch ... "
-   if [ -e /etc/arch-release -a -d /etc/rc.d ]; then
-   RCDIR=/etc/rc.d
-   RESTARTCMD="[ -e /var/run/daemons/\1 ] \&\& 

svn commit: r304514 - head/sys/powerpc/booke

[Justin Hibbits]

Author: jhibbits
Date: Sat Aug 20 00:55:58 2016
New Revision: 304514
URL: https://svnweb.freebsd.org/changeset/base/304514

Log:
  Skip HID1 initialization on e6500 cores, it doesn't exist.
  
  With this, and some drivers removed, a T2080 dev board boots to mountroot.
  
  Submitted by: Ivan Krivonos 

Modified:
  head/sys/powerpc/booke/locore.S

Modified: head/sys/powerpc/booke/locore.S
==
--- head/sys/powerpc/booke/locore.S Sat Aug 20 00:49:29 2016
(r304513)
+++ head/sys/powerpc/booke/locore.S Sat Aug 20 00:55:58 2016
(r304514)
@@ -140,6 +140,8 @@ __start:
beq 1f
cmpli   0, 0, %r3, FSL_E5500
beq 1f
+   cmpli   0, 0, %r3, FSL_E6500
+   beq 1f
 
lis %r3, HID1_E500_DEFAULT_SET@h
ori %r3, %r3, HID1_E500_DEFAULT_SET@l
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304513 - in head: . share/man/man4/man4.i386 sys/conf sys/dev/ie sys/i386/conf sys/modules sys/modules/ie sys/pc98/conf

[John Baldwin]

Author: jhb
Date: Sat Aug 20 00:49:29 2016
New Revision: 304513
URL: https://svnweb.freebsd.org/changeset/base/304513

Log:
  Remove the ie(4) driver for Intel 82586 ISA Ethernet adapters.
  
  This driver only supports 10Mb Ethernet using PIO (the hardware supports
  DMA, but the driver only does PIO).  There are not any PCCard adapters
  supported by this driver, only ISA cards.  In addition, it does not use
  bus_space but instead uses bcopy with volatile pointers triggering a
  host of warnings.  (if_ie.c is one of 3 files always built with
  -Wno-error)
  
  Relnotes: yes

Deleted:
  head/share/man/man4/man4.i386/ie.4
  head/sys/dev/ie/
  head/sys/modules/ie/
Modified:
  head/ObsoleteFiles.inc
  head/share/man/man4/man4.i386/Makefile
  head/sys/conf/files
  head/sys/i386/conf/NOTES
  head/sys/modules/Makefile
  head/sys/pc98/conf/NOTES

Modified: head/ObsoleteFiles.inc
==
--- head/ObsoleteFiles.inc  Sat Aug 20 00:34:19 2016(r304512)
+++ head/ObsoleteFiles.inc  Sat Aug 20 00:49:29 2016(r304513)
@@ -38,6 +38,8 @@
 #   xargs -n1 | sort | uniq -d;
 # done
 
+# 20160819: Remove ie(4)
+OLD_FILES+=usr/share/man/man4/i386/ie.4.gz
 # 20160819: Remove spic(4)
 OLD_FILES+=usr/share/man/man4/spic.4.gz
 # 20160819: Remove wl(4) and wlconfig(8)

Modified: head/share/man/man4/man4.i386/Makefile
==
--- head/share/man/man4/man4.i386/Makefile  Sat Aug 20 00:34:19 2016
(r304512)
+++ head/share/man/man4/man4.i386/Makefile  Sat Aug 20 00:49:29 2016
(r304513)
@@ -16,7 +16,6 @@ MAN=  aic.4 \
fe.4 \
glxiic.4 \
glxsb.4 \
-   ie.4 \
longrun.4 \
mse.4 \
npx.4 \

Modified: head/sys/conf/files
==
--- head/sys/conf/files Sat Aug 20 00:34:19 2016(r304512)
+++ head/sys/conf/files Sat Aug 20 00:49:29 2016(r304513)
@@ -1579,8 +1579,6 @@ dev/ida/ida.c optional ida
 dev/ida/ida_disk.c optional ida
 dev/ida/ida_eisa.c optional ida eisa
 dev/ida/ida_pci.c  optional ida pci
-dev/ie/if_ie.c optional ie isa nowerror
-dev/ie/if_ie_isa.c optional ie isa
 dev/iicbus/ad7418.coptional ad7418
 dev/iicbus/ds1307.coptional ds1307
 dev/iicbus/ds133x.coptional ds133x

Modified: head/sys/i386/conf/NOTES
==
--- head/sys/i386/conf/NOTESSat Aug 20 00:34:19 2016(r304512)
+++ head/sys/i386/conf/NOTESSat Aug 20 00:49:29 2016(r304513)
@@ -624,12 +624,6 @@ hint.ed.0.at="isa"
 hint.ed.0.port="0x280"
 hint.ed.0.irq="5"
 hint.ed.0.maddr="0xd8000"
-device ie  # EtherExpress 8/16, 3C507, StarLAN 10 etc.
-# Hints only required for Starlan
-hint.ie.2.at="isa"
-hint.ie.2.port="0x300"
-hint.ie.2.irq="5"
-hint.ie.2.maddr="0xd"
 device ipw # Intel 2100 wireless NICs.
 device iwi # Intel 2200BG/2225BG/2915ABG wireless NICs.
 device iwn # Intel 4965/1000/5000/6000 wireless NICs.

Modified: head/sys/modules/Makefile
==
--- head/sys/modules/Makefile   Sat Aug 20 00:34:19 2016(r304512)
+++ head/sys/modules/Makefile   Sat Aug 20 00:49:29 2016(r304513)
@@ -141,7 +141,6 @@ SUBDIR= \
${_ibcs2} \
${_ichwd} \
${_ida} \
-   ${_ie} \
if_bridge \
if_disc \
if_edsc \
@@ -705,7 +704,6 @@ _elink= elink
 _glxiic=   glxiic
 _glxsb=glxsb
 #_ibcs2=   ibcs2
-_ie=   ie
 _mse=  mse
 _ncr=  ncr
 _ncv=  ncv

Modified: head/sys/pc98/conf/NOTES
==
--- head/sys/pc98/conf/NOTESSat Aug 20 00:34:19 2016(r304512)
+++ head/sys/pc98/conf/NOTESSat Aug 20 00:49:29 2016(r304513)
@@ -390,11 +390,6 @@ hint.ed.0.at="isa"
 hint.ed.0.port="0x280"
 hint.ed.0.irq="5"
 hint.ed.0.maddr="0xd8000"
-device ie  # Hints only required for Starlan
-hint.ie.2.at="isa"
-hint.ie.2.port="0x300"
-hint.ie.2.irq="5"
-hint.ie.2.maddr="0xd"
 #devicele
 # Hint for the PC98-only C-NET(98)S C-bus front-end of le(4).
 hint.le.0.at="isa"
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304506 - in head: . release/doc/en_US.ISO8859-1/hardware share/man/man4/man4.i386 sys/conf sys/dev/wl sys/i386/conf sys/modules sys/modules/wl targets/pseudo/userland tools/build/mk t

[Warner Losh]

On Fri, Aug 19, 2016 at 4:27 PM, John Baldwin <j...@freebsd.org> wrote:
> Author: jhb
> Date: Fri Aug 19 22:27:14 2016
> New Revision: 304506
> URL: https://svnweb.freebsd.org/changeset/base/304506
>
> Log:
>   Remove the wl(4) driver and wlconfig(8) utility.
>
>   The wl(4) driver supports pre-802.11 PCCard wireless adapters that
>   are slower than 802.11b.  They do not work with any of the 802.11
>   framework and the driver hasn't been reported to actually work in a
>   long time.

s/PC Card/ISA/g

There never were any PC Card wl cards. These cards were ISA only
and ran at up to 1Mbps.

Warner

>   Relnotes: yes
>
> Deleted:
>   head/share/man/man4/man4.i386/wl.4
>   head/sys/dev/wl/
>   head/sys/modules/wl/
>   head/tools/kerneldoc/subsys/Doxyfile-dev_wl
>   head/usr.sbin/wlconfig/
> Modified:
>   head/ObsoleteFiles.inc
>   head/release/doc/en_US.ISO8859-1/hardware/article.xml
>   head/share/man/man4/man4.i386/Makefile
>   head/sys/conf/files
>   head/sys/conf/options
>   head/sys/i386/conf/NOTES
>   head/sys/modules/Makefile
>   head/targets/pseudo/userland/Makefile.depend
>   head/tools/build/mk/OptionalObsoleteFiles.inc
>   head/usr.sbin/Makefile.i386
>
> Modified: head/ObsoleteFiles.inc
> ==
> --- head/ObsoleteFiles.inc  Fri Aug 19 22:13:01 2016(r304505)
> +++ head/ObsoleteFiles.inc  Fri Aug 19 22:27:14 2016(r304506)
> @@ -38,6 +38,10 @@
>  #   xargs -n1 | sort | uniq -d;
>  # done
>
> +# 20160819: Remove wl(4) and wlconfig(8)
> +OLD_FILES+=usr/share/man/man4/i386/wl.4.gz
> +OLD_FILES+=usr/sbin/wlconfig
> +OLD_FILES+=usr/share/man/man8/i386/wlconfig.8.gz
>  # 20160819: Remove si(4) and sicontrol(8)
>  OLD_FILES+=usr/share/man/man4/si.4.gz
>  OLD_FILES+=usr/sbin/sicontrol
>
> Modified: head/release/doc/en_US.ISO8859-1/hardware/article.xml
> ==
> --- head/release/doc/en_US.ISO8859-1/hardware/article.xml   Fri Aug 19 
> 22:13:01 2016(r304505)
> +++ head/release/doc/en_US.ISO8859-1/hardware/article.xml   Fri Aug 19 
> 22:27:14 2016(r304506)
> @@ -1023,9 +1023,6 @@
> Intersil PRISM-2.5, Intersil Prism-3, and Symbol Spectrum24
> chipsets ( driver)
>
> -  [] NCR / ATT / Lucent Technologies WaveLan
> -   T1-speed ISA/radio LAN cards ( driver)
> -
>[, ] Intel PRO/Wireless 3945ABG
> MiniPCI network adapters ( driver)
>
>
> Modified: head/share/man/man4/man4.i386/Makefile
> ==
> --- head/share/man/man4/man4.i386/Makefile  Fri Aug 19 22:13:01 2016  
>   (r304505)
> +++ head/share/man/man4/man4.i386/Makefile  Fri Aug 19 22:27:14 2016  
>   (r304506)
> @@ -31,8 +31,7 @@ MAN=  aic.4 \
> streams.4 \
> svr4.4 \
> vpd.4 \
> -   vx.4 \
> -   wl.4
> +   vx.4
>
>  MLINKS=CPU_ELAN.4 CPU_SOEKRIS.4
>  MLINKS+=pae.4 PAE.4
>
> Modified: head/sys/conf/files
> ==
> --- head/sys/conf/files Fri Aug 19 22:13:01 2016(r304505)
> +++ head/sys/conf/files Fri Aug 19 22:27:14 2016(r304506)
> @@ -2933,7 +2933,6 @@ dev/wb/if_wb.coptional wb pci
>  dev/wi/if_wi.c optional wi
>  dev/wi/if_wi_pccard.c  optional wi pccard
>  dev/wi/if_wi_pci.c optional wi pci
> -dev/wl/if_wl.c optional wl isa
>  dev/wpi/if_wpi.c   optional wpi pci
>  wpifw.coptional wpifw
>   \
> compile-with"${AWK} -f $S/tools/fw_stub.awk wpi.fw:wpifw:153229 
> -mwpi -c${.TARGET}" \
>
> Modified: head/sys/conf/options
> ==
> --- head/sys/conf/options   Fri Aug 19 22:13:01 2016(r304505)
> +++ head/sys/conf/options   Fri Aug 19 22:27:14 2016(r304506)
> @@ -218,8 +218,6 @@ SW_WATCHDOG opt_watchdog.h
>  TURNSTILE_PROFILING
>  UMTX_PROFILING
>  VERBOSE_SYSINIT
> -WLCACHEopt_wavelan.h
> -WLDEBUGopt_wavelan.h
>
>  # POSIX kernel options
>  P1003_1B_MQUEUEopt_posix.h
>
> Modified: head/sys/i386/conf/NOTES
> ==
> --- head/sys/i386/conf/NOTESFri Aug 19 22:13:01 2016(r304505)
> +++ head/sys/i386/conf/NOTESFri Aug 19 22:27:14 2016(r304506)
> @@ -648,11 +64

svn commit: r304512 - in stable/11: lib/libc/sys sys/kern

[Bryan Drewery]

Author: bdrewery
Date: Sat Aug 20 00:34:19 2016
New Revision: 304512
URL: https://svnweb.freebsd.org/changeset/base/304512

Log:
  MFC r304288:
  
Garbage collect _umtx_lock(2)/_umtx_unlock(2) references removed in r263318.

Modified:
  stable/11/lib/libc/sys/Symbol.map
  stable/11/sys/kern/capabilities.conf
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/lib/libc/sys/Symbol.map
==
--- stable/11/lib/libc/sys/Symbol.map   Sat Aug 20 00:22:39 2016
(r304511)
+++ stable/11/lib/libc/sys/Symbol.map   Sat Aug 20 00:34:19 2016
(r304512)
@@ -34,9 +34,7 @@ FBSD_1.0 {
__setugid;
__syscall;
__sysctl;
-   _umtx_lock;
_umtx_op;
-   _umtx_unlock;
abort2;
accept;
access;
@@ -455,12 +453,8 @@ FBSDprivate_1.0 {
__sys___syscall;
___sysctl;
__sys___sysctl;
-   __umtx_lock;
-   __sys__umtx_lock;
__umtx_op;
__sys__umtx_op;
-   __umtx_unlock;
-   __sys__umtx_unlock;
_abort2;
__sys_abort2;
_accept;

Modified: stable/11/sys/kern/capabilities.conf
==
--- stable/11/sys/kern/capabilities.confSat Aug 20 00:22:39 2016
(r304511)
+++ stable/11/sys/kern/capabilities.confSat Aug 20 00:34:19 2016
(r304512)
@@ -64,9 +64,7 @@ __sysctl
 ##
 ## XXRW: Need to check this very carefully.
 ##
-_umtx_lock
 _umtx_op
-_umtx_unlock
 
 ##
 ## Allow process termination using abort2(2).
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304511 - in stable: 10/sys/dev/pci 8/sys/dev/pci 9/sys/dev/pci

[John Baldwin]

Author: jhb
Date: Sat Aug 20 00:22:39 2016
New Revision: 304511
URL: https://svnweb.freebsd.org/changeset/base/304511

Log:
  MFC 298950: Fix an off by one error when remapping MSI-X vectors.
  
  pci_remap_msix() can be used to alter the mapping of allocated
  MSI-X vectors to the MSI-X table.  The code had an off by one error
  when adding the IRQ resources after performing a remap.  This was
  fatal for any vectors in the table that used the "last" valid IRQ as
  those vectors were assigned a garbage IRQ value.

Modified:
  stable/8/sys/dev/pci/pci.c
Directory Properties:
  stable/8/sys/   (props changed)
  stable/8/sys/dev/   (props changed)
  stable/8/sys/dev/pci/   (props changed)

Changes in other areas also in this revision:
Modified:
  stable/10/sys/dev/pci/pci.c
  stable/9/sys/dev/pci/pci.c
Directory Properties:
  stable/10/   (props changed)
  stable/9/sys/   (props changed)
  stable/9/sys/dev/   (props changed)

Modified: stable/8/sys/dev/pci/pci.c
==
--- stable/8/sys/dev/pci/pci.c  Sat Aug 20 00:08:10 2016(r304510)
+++ stable/8/sys/dev/pci/pci.c  Sat Aug 20 00:22:39 2016(r304511)
@@ -1617,7 +1617,7 @@ pci_remap_msix_method(device_t dev, devi
for (i = 0; i < count; i++) {
if (vectors[i] == 0)
continue;
-   irq = msix->msix_vectors[vectors[i]].mv_irq;
+   irq = msix->msix_vectors[vectors[i] - 1].mv_irq;
resource_list_add(>resources, SYS_RES_IRQ, i + 1, irq,
irq, 1);
}
@@ -1631,7 +1631,7 @@ pci_remap_msix_method(device_t dev, devi
printf("---");
else
printf("%d",
-   msix->msix_vectors[vectors[i]].mv_irq);
+   msix->msix_vectors[vectors[i] - 1].mv_irq);
}
printf("\n");
}
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304511 - in stable: 10/sys/dev/pci 8/sys/dev/pci 9/sys/dev/pci

[John Baldwin]

Author: jhb
Date: Sat Aug 20 00:22:39 2016
New Revision: 304511
URL: https://svnweb.freebsd.org/changeset/base/304511

Log:
  MFC 298950: Fix an off by one error when remapping MSI-X vectors.
  
  pci_remap_msix() can be used to alter the mapping of allocated
  MSI-X vectors to the MSI-X table.  The code had an off by one error
  when adding the IRQ resources after performing a remap.  This was
  fatal for any vectors in the table that used the "last" valid IRQ as
  those vectors were assigned a garbage IRQ value.

Modified:
  stable/9/sys/dev/pci/pci.c
Directory Properties:
  stable/9/sys/   (props changed)
  stable/9/sys/dev/   (props changed)

Changes in other areas also in this revision:
Modified:
  stable/10/sys/dev/pci/pci.c
  stable/8/sys/dev/pci/pci.c
Directory Properties:
  stable/10/   (props changed)
  stable/8/sys/   (props changed)
  stable/8/sys/dev/   (props changed)
  stable/8/sys/dev/pci/   (props changed)

Modified: stable/9/sys/dev/pci/pci.c
==
--- stable/9/sys/dev/pci/pci.c  Sat Aug 20 00:08:10 2016(r304510)
+++ stable/9/sys/dev/pci/pci.c  Sat Aug 20 00:22:39 2016(r304511)
@@ -1660,7 +1660,7 @@ pci_remap_msix_method(device_t dev, devi
for (i = 0; i < count; i++) {
if (vectors[i] == 0)
continue;
-   irq = msix->msix_vectors[vectors[i]].mv_irq;
+   irq = msix->msix_vectors[vectors[i] - 1].mv_irq;
resource_list_add(>resources, SYS_RES_IRQ, i + 1, irq,
irq, 1);
}
@@ -1674,7 +1674,7 @@ pci_remap_msix_method(device_t dev, devi
printf("---");
else
printf("%d",
-   msix->msix_vectors[vectors[i]].mv_irq);
+   msix->msix_vectors[vectors[i] - 1].mv_irq);
}
printf("\n");
}
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r298950 - head/sys/dev/pci

[John Baldwin]

On Friday, August 19, 2016 08:21:03 AM Harry Schmalzbauer wrote:
>  Bezüglich John Baldwin's Nachricht vom 03.05.2016 02:35 (localtime):
> > Author: jhb
> > Date: Tue May  3 00:35:11 2016
> > New Revision: 298950
> > URL: https://svnweb.freebsd.org/changeset/base/298950
> >
> > Log:
> >   Fix an off by one error when remapping MSI-X vectors.
> >   
> >   pci_remap_msix() can be used to alter the mapping of allocated
> >   MSI-X vectors to the MSI-X table.  The code had an off by one error
> >   when adding the IRQ resources after performing a remap.  This was
> >   fatal for any vectors in the table that used the "last" valid IRQ as
> >   those vectors were assigned a garbage IRQ value.
> >   
> >   MFC after:3 days
> 
> Was this superseded by any other merge?
> As far as I saw, it's missing in stable/10?

Sorry, the reminder somehow got marked as done when it wasn't done yet.
I've merged this back to 8.x and later.  Thanks for the reminder!

-- 
John Baldwin
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304511 - in stable: 10/sys/dev/pci 8/sys/dev/pci 9/sys/dev/pci

[John Baldwin]

Author: jhb
Date: Sat Aug 20 00:22:39 2016
New Revision: 304511
URL: https://svnweb.freebsd.org/changeset/base/304511

Log:
  MFC 298950: Fix an off by one error when remapping MSI-X vectors.
  
  pci_remap_msix() can be used to alter the mapping of allocated
  MSI-X vectors to the MSI-X table.  The code had an off by one error
  when adding the IRQ resources after performing a remap.  This was
  fatal for any vectors in the table that used the "last" valid IRQ as
  those vectors were assigned a garbage IRQ value.

Modified:
  stable/10/sys/dev/pci/pci.c
Directory Properties:
  stable/10/   (props changed)

Changes in other areas also in this revision:
Modified:
  stable/8/sys/dev/pci/pci.c
  stable/9/sys/dev/pci/pci.c
Directory Properties:
  stable/8/sys/   (props changed)
  stable/8/sys/dev/   (props changed)
  stable/8/sys/dev/pci/   (props changed)
  stable/9/sys/   (props changed)
  stable/9/sys/dev/   (props changed)

Modified: stable/10/sys/dev/pci/pci.c
==
--- stable/10/sys/dev/pci/pci.c Sat Aug 20 00:08:10 2016(r304510)
+++ stable/10/sys/dev/pci/pci.c Sat Aug 20 00:22:39 2016(r304511)
@@ -1744,7 +1744,7 @@ pci_remap_msix_method(device_t dev, devi
for (i = 0; i < count; i++) {
if (vectors[i] == 0)
continue;
-   irq = msix->msix_vectors[vectors[i]].mv_irq;
+   irq = msix->msix_vectors[vectors[i] - 1].mv_irq;
resource_list_add(>resources, SYS_RES_IRQ, i + 1, irq,
irq, 1);
}
@@ -1758,7 +1758,7 @@ pci_remap_msix_method(device_t dev, devi
printf("---");
else
printf("%d",
-   msix->msix_vectors[vectors[i]].mv_irq);
+   msix->msix_vectors[vectors[i] - 1].mv_irq);
}
printf("\n");
}
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304510 - head/sys/dev/ixl

[Eric Joyner]

Author: erj
Date: Sat Aug 20 00:08:10 2016
New Revision: 304510
URL: https://svnweb.freebsd.org/changeset/base/304510

Log:
  ixlv(4): Fix ixlv(4) not loading when loaded as a kernel module and netmap is 
enabled.
  
  Define (unused) netmap variables; ixlv(4) doesn't support netmap yet.
  
  Reported by:  sergey.koz...@intel.com
  Sponsored by: Intel Corporation

Modified:
  head/sys/dev/ixl/if_ixlv.c

Modified: head/sys/dev/ixl/if_ixlv.c
==
--- head/sys/dev/ixl/if_ixlv.c  Fri Aug 19 23:44:07 2016(r304509)
+++ head/sys/dev/ixl/if_ixlv.c  Sat Aug 20 00:08:10 2016(r304510)
@@ -217,6 +217,10 @@ TUNABLE_INT("hw.ixlv.tx_itr", _tx_i
 SYSCTL_INT(_hw_ixlv, OID_AUTO, tx_itr, CTLFLAG_RDTUN,
 _tx_itr, 0, "TX Interrupt Rate");
 
+/* Fix when building as a standalone module when netmap is enabled */
+#if defined(DEV_NETMAP) && !defined(NETMAP_IXL_MAIN)
+int ixl_rx_miss, ixl_rx_miss_bufs, ixl_crcstrip;
+#endif
 
 /*
  *  Device identification routine
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304509 - head/sys/arm/allwinner

[Emmanuel Vadot]

Author: manu
Date: Fri Aug 19 23:44:07 2016
New Revision: 304509
URL: https://svnweb.freebsd.org/changeset/base/304509

Log:
  if_emac: Before generating a random MAC address, try using the SID rootkey
  to generate one. This is was U-Boot does to generate a random MAC so we end
  up with the same MAC address as if U-Boot did generate it.
  
  MFC after:1 week

Modified:
  head/sys/arm/allwinner/if_emac.c

Modified: head/sys/arm/allwinner/if_emac.c
==
--- head/sys/arm/allwinner/if_emac.cFri Aug 19 23:39:08 2016
(r304508)
+++ head/sys/arm/allwinner/if_emac.cFri Aug 19 23:44:07 2016
(r304509)
@@ -77,6 +77,7 @@ __FBSDID("$FreeBSD$");
 #include 
 
 #include 
+#include 
 
 #include 
 
@@ -167,12 +168,17 @@ static void
 emac_get_hwaddr(struct emac_softc *sc, uint8_t *hwaddr)
 {
uint32_t val0, val1, rnd;
+   u_char rootkey[16];
 
/*
 * Try to get MAC address from running hardware.
 * If there is something non-zero there just use it.
 *
 * Otherwise set the address to a convenient locally assigned address,
+* using the SID rootkey.
+* This is was uboot does so we end up with the same mac as if uboot
+* did set it.
+* If we can't get the root key, generate a random one,
 * 'bsd' + random 24 low-order bits. 'b' is 0x62, which has the locally
 * assigned bit set, and the broadcast/multicast bit clear.
 */
@@ -186,13 +192,23 @@ emac_get_hwaddr(struct emac_softc *sc, u
hwaddr[4] = (val0 >> 8) & 0xff;
hwaddr[5] = (val0 >> 0) & 0xff;
} else {
-   rnd = arc4random() & 0x00ff;
-   hwaddr[0] = 'b';
-   hwaddr[1] = 's';
-   hwaddr[2] = 'd';
-   hwaddr[3] = (rnd >> 16) & 0xff;
-   hwaddr[4] = (rnd >> 8) & 0xff;
-   hwaddr[5] = (rnd >> 0) & 0xff;
+   if (aw_sid_get_rootkey(rootkey) == 0) {
+   hwaddr[0] = 0x2;
+   hwaddr[1] = rootkey[3];
+   hwaddr[2] = rootkey[12];
+   hwaddr[3] = rootkey[13];
+   hwaddr[4] = rootkey[14];
+   hwaddr[5] = rootkey[15];
+   }
+   else {
+   rnd = arc4random() & 0x00ff;
+   hwaddr[0] = 'b';
+   hwaddr[1] = 's';
+   hwaddr[2] = 'd';
+   hwaddr[3] = (rnd >> 16) & 0xff;
+   hwaddr[4] = (rnd >> 8) & 0xff;
+   hwaddr[5] = (rnd >> 0) & 0xff;
+   }
}
if (bootverbose)
printf("MAC address: %s\n", ether_sprintf(hwaddr));
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304508 - in head: . share/man/man4 sys/conf sys/i386/conf sys/i386/isa

[John Baldwin]

Author: jhb
Date: Fri Aug 19 23:39:08 2016
New Revision: 304508
URL: https://svnweb.freebsd.org/changeset/base/304508

Log:
  Remove the spic(4) driver for the Sony Vaoi Jogdial.
  
  This hardware is not present on any modern systems.  The driver is quite
  hackish (raw inb/outb instead of bus_space, and raw inb/outb to random
  I/O ports to enable ACPI since it predated proper ACPI support).
  
  Relnotes: yes

Deleted:
  head/share/man/man4/spic.4
  head/sys/i386/isa/spic.c
  head/sys/i386/isa/spicreg.h
Modified:
  head/ObsoleteFiles.inc
  head/share/man/man4/Makefile
  head/sys/conf/files.i386
  head/sys/i386/conf/NOTES

Modified: head/ObsoleteFiles.inc
==
--- head/ObsoleteFiles.inc  Fri Aug 19 23:31:56 2016(r304507)
+++ head/ObsoleteFiles.inc  Fri Aug 19 23:39:08 2016(r304508)
@@ -38,6 +38,8 @@
 #   xargs -n1 | sort | uniq -d;
 # done
 
+# 20160819: Remove spic(4)
+OLD_FILES+=usr/share/man/man4/spic.4.gz
 # 20160819: Remove wl(4) and wlconfig(8)
 OLD_FILES+=usr/share/man/man4/i386/wl.4.gz
 OLD_FILES+=usr/sbin/wlconfig

Modified: head/share/man/man4/Makefile
==
--- head/share/man/man4/MakefileFri Aug 19 23:31:56 2016
(r304507)
+++ head/share/man/man4/MakefileFri Aug 19 23:39:08 2016
(r304508)
@@ -496,7 +496,6 @@ MAN=aac.4 \
snd_via82c686.4 \
snd_vibes.4 \
snp.4 \
-   spic.4 \
${_spkr.4} \
splash.4 \
sppp.4 \

Modified: head/sys/conf/files.i386
==
--- head/sys/conf/files.i386Fri Aug 19 23:31:56 2016(r304507)
+++ head/sys/conf/files.i386Fri Aug 19 23:39:08 2016(r304508)
@@ -516,7 +516,6 @@ i386/isa/elink.coptional ep | ie
 i386/isa/npx.c optional npx
 i386/isa/pmtimer.c optional pmtimer
 i386/isa/prof_machdep.coptional profiling-routine
-i386/isa/spic.coptional spic
 i386/linux/imgact_linux.c  optional compat_linux
 i386/linux/linux_dummy.c   optional compat_linux
 i386/linux/linux_machdep.c optional compat_linux

Modified: head/sys/i386/conf/NOTES
==
--- head/sys/i386/conf/NOTESFri Aug 19 23:31:56 2016(r304507)
+++ head/sys/i386/conf/NOTESFri Aug 19 23:39:08 2016(r304508)
@@ -834,7 +834,6 @@ device  hyperv  # HyperV drivers
 # vpd: Vital Product Data kernel interface
 # pmtimer: Adjust system timer at wakeup time
 # pbio: Parallel (8255 PPI) basic I/O (mode 0) port (e.g. Advantech PCL-724)
-# spic: Sony Programmable I/O controller (VAIO notebooks)
 # asmc: Apple System Management Controller
 # si: Specialix International SI/XIO or SX intelligent serial card driver
 # tpm: Trusted Platform Module
@@ -871,9 +870,6 @@ device  pmtimer
 device pbio
 hint.pbio.0.at="isa"
 hint.pbio.0.port="0x360"
-device spic
-hint.spic.0.at="isa"
-hint.spic.0.port="0x10a0"
 device asmc
 device tpm
 device padlock_rng # VIA Padlock RNG
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304507 - in stable: 10/sys/boot/efi/loader 11/sys/boot/efi/loader

[John Baldwin]

Author: jhb
Date: Fri Aug 19 23:31:56 2016
New Revision: 304507
URL: https://svnweb.freebsd.org/changeset/base/304507

Log:
  MFC 304018: Add defines needed to export SMBIOS serial numbers
  
  Some defines needed for exporting serial numbers from the SMBIOS were
  missed during integration of SMBIOS support in the EFI boot loader (r281138).
  This is needed for getting the hostid set from the system hardware UUID.
  
  PR:   206031

Modified:
  stable/11/sys/boot/efi/loader/Makefile
Directory Properties:
  stable/11/   (props changed)

Changes in other areas also in this revision:
Modified:
  stable/10/sys/boot/efi/loader/Makefile
Directory Properties:
  stable/10/   (props changed)

Modified: stable/11/sys/boot/efi/loader/Makefile
==
--- stable/11/sys/boot/efi/loader/Makefile  Fri Aug 19 22:27:14 2016
(r304506)
+++ stable/11/sys/boot/efi/loader/Makefile  Fri Aug 19 23:31:56 2016
(r304507)
@@ -63,6 +63,18 @@ CFLAGS+= -DNO_PCI -DEFI
 LIBSTAND=  ${.OBJDIR}/../../../../lib/libstand/libstand.a
 .endif
 
+.if !defined(BOOT_HIDE_SERIAL_NUMBERS)
+# Export serial numbers, UUID, and asset tag from loader.
+CFLAGS+= -DSMBIOS_SERIAL_NUMBERS
+.if defined(BOOT_LITTLE_ENDIAN_UUID)
+# Use little-endian UUID format as defined in SMBIOS 2.6.
+CFLAGS+= -DSMBIOS_LITTLE_ENDIAN_UUID
+.elif defined(BOOT_NETWORK_ENDIAN_UUID)
+# Use network-endian UUID format for backward compatibility.
+CFLAGS+= -DSMBIOS_NETWORK_ENDIAN_UUID
+.endif
+.endif
+
 .if ${MK_FORTH} != "no"
 BOOT_FORTH=yes
 CFLAGS+=   -DBOOT_FORTH
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304507 - in stable: 10/sys/boot/efi/loader 11/sys/boot/efi/loader

[John Baldwin]

Author: jhb
Date: Fri Aug 19 23:31:56 2016
New Revision: 304507
URL: https://svnweb.freebsd.org/changeset/base/304507

Log:
  MFC 304018: Add defines needed to export SMBIOS serial numbers
  
  Some defines needed for exporting serial numbers from the SMBIOS were
  missed during integration of SMBIOS support in the EFI boot loader (r281138).
  This is needed for getting the hostid set from the system hardware UUID.
  
  PR:   206031

Modified:
  stable/10/sys/boot/efi/loader/Makefile
Directory Properties:
  stable/10/   (props changed)

Changes in other areas also in this revision:
Modified:
  stable/11/sys/boot/efi/loader/Makefile
Directory Properties:
  stable/11/   (props changed)

Modified: stable/10/sys/boot/efi/loader/Makefile
==
--- stable/10/sys/boot/efi/loader/Makefile  Fri Aug 19 22:27:14 2016
(r304506)
+++ stable/10/sys/boot/efi/loader/Makefile  Fri Aug 19 23:31:56 2016
(r304507)
@@ -50,6 +50,18 @@ CFLAGS+= -DEFI_ZFS_BOOT
 .endif
 CFLAGS+=   -DNO_PCI -DEFI
 
+.if !defined(BOOT_HIDE_SERIAL_NUMBERS)
+# Export serial numbers, UUID, and asset tag from loader.
+CFLAGS+= -DSMBIOS_SERIAL_NUMBERS
+.if defined(BOOT_LITTLE_ENDIAN_UUID)
+# Use little-endian UUID format as defined in SMBIOS 2.6.
+CFLAGS+= -DSMBIOS_LITTLE_ENDIAN_UUID
+.elif defined(BOOT_NETWORK_ENDIAN_UUID)
+# Use network-endian UUID format for backward compatibility.
+CFLAGS+= -DSMBIOS_NETWORK_ENDIAN_UUID
+.endif
+.endif
+
 .if ${MK_FORTH} != "no"
 BOOT_FORTH=yes
 CFLAGS+=   -DBOOT_FORTH
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304506 - in head: . release/doc/en_US.ISO8859-1/hardware share/man/man4/man4.i386 sys/conf sys/dev/wl sys/i386/conf sys/modules sys/modules/wl targets/pseudo/userland tools/build/mk to...

[John Baldwin]

Author: jhb
Date: Fri Aug 19 22:27:14 2016
New Revision: 304506
URL: https://svnweb.freebsd.org/changeset/base/304506

Log:
  Remove the wl(4) driver and wlconfig(8) utility.
  
  The wl(4) driver supports pre-802.11 PCCard wireless adapters that
  are slower than 802.11b.  They do not work with any of the 802.11
  framework and the driver hasn't been reported to actually work in a
  long time.
  
  Relnotes: yes

Deleted:
  head/share/man/man4/man4.i386/wl.4
  head/sys/dev/wl/
  head/sys/modules/wl/
  head/tools/kerneldoc/subsys/Doxyfile-dev_wl
  head/usr.sbin/wlconfig/
Modified:
  head/ObsoleteFiles.inc
  head/release/doc/en_US.ISO8859-1/hardware/article.xml
  head/share/man/man4/man4.i386/Makefile
  head/sys/conf/files
  head/sys/conf/options
  head/sys/i386/conf/NOTES
  head/sys/modules/Makefile
  head/targets/pseudo/userland/Makefile.depend
  head/tools/build/mk/OptionalObsoleteFiles.inc
  head/usr.sbin/Makefile.i386

Modified: head/ObsoleteFiles.inc
==
--- head/ObsoleteFiles.inc  Fri Aug 19 22:13:01 2016(r304505)
+++ head/ObsoleteFiles.inc  Fri Aug 19 22:27:14 2016(r304506)
@@ -38,6 +38,10 @@
 #   xargs -n1 | sort | uniq -d;
 # done
 
+# 20160819: Remove wl(4) and wlconfig(8)
+OLD_FILES+=usr/share/man/man4/i386/wl.4.gz
+OLD_FILES+=usr/sbin/wlconfig
+OLD_FILES+=usr/share/man/man8/i386/wlconfig.8.gz
 # 20160819: Remove si(4) and sicontrol(8)
 OLD_FILES+=usr/share/man/man4/si.4.gz
 OLD_FILES+=usr/sbin/sicontrol

Modified: head/release/doc/en_US.ISO8859-1/hardware/article.xml
==
--- head/release/doc/en_US.ISO8859-1/hardware/article.xml   Fri Aug 19 
22:13:01 2016(r304505)
+++ head/release/doc/en_US.ISO8859-1/hardware/article.xml   Fri Aug 19 
22:27:14 2016(r304506)
@@ -1023,9 +1023,6 @@
Intersil PRISM-2.5, Intersil Prism-3, and Symbol Spectrum24
chipsets ( driver)
 
-  [] NCR / ATT / Lucent Technologies WaveLan
-   T1-speed ISA/radio LAN cards ( driver)
-
   [, ] Intel PRO/Wireless 3945ABG
MiniPCI network adapters ( driver)
 

Modified: head/share/man/man4/man4.i386/Makefile
==
--- head/share/man/man4/man4.i386/Makefile  Fri Aug 19 22:13:01 2016
(r304505)
+++ head/share/man/man4/man4.i386/Makefile  Fri Aug 19 22:27:14 2016
(r304506)
@@ -31,8 +31,7 @@ MAN=  aic.4 \
streams.4 \
svr4.4 \
vpd.4 \
-   vx.4 \
-   wl.4
+   vx.4
 
 MLINKS=CPU_ELAN.4 CPU_SOEKRIS.4
 MLINKS+=pae.4 PAE.4

Modified: head/sys/conf/files
==
--- head/sys/conf/files Fri Aug 19 22:13:01 2016(r304505)
+++ head/sys/conf/files Fri Aug 19 22:27:14 2016(r304506)
@@ -2933,7 +2933,6 @@ dev/wb/if_wb.coptional wb pci
 dev/wi/if_wi.c optional wi
 dev/wi/if_wi_pccard.c  optional wi pccard
 dev/wi/if_wi_pci.c optional wi pci
-dev/wl/if_wl.c optional wl isa
 dev/wpi/if_wpi.c   optional wpi pci
 wpifw.coptional wpifw  
\
compile-with"${AWK} -f $S/tools/fw_stub.awk wpi.fw:wpifw:153229 
-mwpi -c${.TARGET}" \

Modified: head/sys/conf/options
==
--- head/sys/conf/options   Fri Aug 19 22:13:01 2016(r304505)
+++ head/sys/conf/options   Fri Aug 19 22:27:14 2016(r304506)
@@ -218,8 +218,6 @@ SW_WATCHDOG opt_watchdog.h
 TURNSTILE_PROFILING
 UMTX_PROFILING
 VERBOSE_SYSINIT
-WLCACHEopt_wavelan.h
-WLDEBUGopt_wavelan.h
 
 # POSIX kernel options
 P1003_1B_MQUEUEopt_posix.h

Modified: head/sys/i386/conf/NOTES
==
--- head/sys/i386/conf/NOTESFri Aug 19 22:13:01 2016(r304505)
+++ head/sys/i386/conf/NOTESFri Aug 19 22:27:14 2016(r304506)
@@ -648,11 +648,6 @@ hint.sbni.0.port="0x210"
 hint.sbni.0.irq="0xefdead"
 hint.sbni.0.flags="0"
 device vmx # VMware VMXNET3 Ethernet
-device wl
-hint.wl.0.at="isa"
-hint.wl.0.port="0x300"
-optionsWLCACHE # enables the signal-strength cache
-optionsWLDEBUG # enables verbose debugging output
 device wpi # Intel 3945ABG wireless NICs.
 
 # IEEE 802.11 adapter firmware modules

Modified: head/sys/modules/Makefile
==
--- head/sys/modules/Makefile   Fri Aug 19 22:13:01 2016(r304505)
+++ head/sys/modules/Makefile   Fri Aug 19 22:27:14 2016  

svn commit: r304505 - head/release/doc/en_US.ISO8859-1/hardware

[John Baldwin]

Author: jhb
Date: Fri Aug 19 22:13:01 2016
New Revision: 304505
URL: https://svnweb.freebsd.org/changeset/base/304505

Log:
  Remove mentions of the mcd(4), scd(4), and si(4) drivers.
  
  Relnotes: yes

Modified:
  head/release/doc/en_US.ISO8859-1/hardware/article.xml

Modified: head/release/doc/en_US.ISO8859-1/hardware/article.xml
==
--- head/release/doc/en_US.ISO8859-1/hardware/article.xml   Fri Aug 19 
22:05:22 2016(r304504)
+++ head/release/doc/en_US.ISO8859-1/hardware/article.xml   Fri Aug 19 
22:13:01 2016(r304505)
@@ -772,23 +772,9 @@

 

- [] Sony proprietary interface (all models)
-   ()
-   
-
-   
  ATAPI IDE interface ()

   
-
-  [] The following device is unmaintained:
-
-  
-   
- Mitsumi proprietary CD-ROM interface (all models)
-   ()
-   
-  
 
 
 
@@ -1235,13 +1221,6 @@
 
   
 
-  [, ] Specialix SI/XIO/SX multiport
-   serial cards, with both the older SIHOST2.x and the
-   enhanced (transputer based, aka JET) host cards
-   (ISA, EISA and PCI) are supported.  Note that the newer SX+
-   PCI cards are not currently supported.  (
-   driver)
-
   [] Internel serial interfaces (
driver)
 
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304504 - head/etc

[John Baldwin]

Author: jhb
Date: Fri Aug 19 22:05:22 2016
New Revision: 304504
URL: https://svnweb.freebsd.org/changeset/base/304504

Log:
  Remove stale drivers (amd(4) and asr(4)) from the SCSI controller regex.

Modified:
  head/etc/devd.conf

Modified: head/etc/devd.conf
==
--- head/etc/devd.conf  Fri Aug 19 21:51:42 2016(r304503)
+++ head/etc/devd.conf  Fri Aug 19 22:05:22 2016(r304504)
@@ -19,7 +19,7 @@ options {
# Setup some shorthand for regex that we use later in the file.
#XXX Yes, these are gross -- imp
set scsi-controller-regex
-   "(aac|adv|adw|aha|ahb|ahc|ahd|aic|amd|amr|asr|bt|ciss|ct|dpt|\
+   "(aac|adv|adw|aha|ahb|ahc|ahd|aic|amr|bt|ciss|ct|dpt|\
esp|ida|iir|ips|isp|mlx|mly|mpt|ncr|ncv|nsp|stg|sym|trm)\
[0-9]+";
set wifi-driver-regex
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304503 - in head: etc release/doc/en_US.ISO8859-1/hardware sys/conf sys/dev/wds sys/modules sys/modules/wds sys/pc98/conf sys/sparc64/conf tools/kerneldoc/subsys

[John Baldwin]

Author: jhb
Date: Fri Aug 19 21:51:42 2016
New Revision: 304503
URL: https://svnweb.freebsd.org/changeset/base/304503

Log:
  Remove the wds(4) driver for the WD700 ISA SCSI HBA.
  
  While this driver does do DMA, it bounce buffers all transactions through
  a single 64k buffer.  It also does not have a manpage.
  
  Relnotes: yes

Deleted:
  head/sys/dev/wds/
  head/sys/modules/wds/
  head/tools/kerneldoc/subsys/Doxyfile-dev_wds
Modified:
  head/etc/devd.conf
  head/release/doc/en_US.ISO8859-1/hardware/article.xml
  head/sys/conf/NOTES
  head/sys/conf/files
  head/sys/modules/Makefile
  head/sys/pc98/conf/NOTES
  head/sys/sparc64/conf/NOTES

Modified: head/etc/devd.conf
==
--- head/etc/devd.conf  Fri Aug 19 21:14:27 2016(r304502)
+++ head/etc/devd.conf  Fri Aug 19 21:51:42 2016(r304503)
@@ -20,7 +20,7 @@ options {
#XXX Yes, these are gross -- imp
set scsi-controller-regex
"(aac|adv|adw|aha|ahb|ahc|ahd|aic|amd|amr|asr|bt|ciss|ct|dpt|\
-   esp|ida|iir|ips|isp|mlx|mly|mpt|ncr|ncv|nsp|stg|sym|trm|wds)\
+   esp|ida|iir|ips|isp|mlx|mly|mpt|ncr|ncv|nsp|stg|sym|trm)\
[0-9]+";
set wifi-driver-regex
"(ath|bwi|bwn|ipw|iwi|iwm|iwn|malo|mwl|ral|rsu|rum|run|uath|\

Modified: head/release/doc/en_US.ISO8859-1/hardware/article.xml
==
--- head/release/doc/en_US.ISO8859-1/hardware/article.xml   Fri Aug 19 
21:14:27 2016(r304502)
+++ head/release/doc/en_US.ISO8859-1/hardware/article.xml   Fri Aug 19 
21:51:42 2016(r304503)
@@ -751,9 +751,6 @@
 
   
 
-  [] The wds(4) driver supports the WD7000 SCSI
-   controller.
-
   With all supported SCSI controllers, full support is
provided for SCSI-I, SCSI-II, and SCSI-III peripherals,
including hard disks, optical disks, tape drives (including

Modified: head/sys/conf/NOTES
==
--- head/sys/conf/NOTES Fri Aug 19 21:14:27 2016(r304502)
+++ head/sys/conf/NOTES Fri Aug 19 21:51:42 2016(r304503)
@@ -1560,7 +1560,6 @@ options   TERMINAL_KERN_ATTR=(FG_LIGHTRED
 #  53C876, 53C885,  53C895, 53C895A, 53C896,  53C897, 53C1510D,
 #  53C1010-33, 53C1010-66.
 # trm: Tekram DC395U/UW/F DC315U adapters.
-# wds: WD7000
 
 #
 # Note that the order is important in order for Buslogic ISA/EISA cards to be
@@ -1602,11 +1601,6 @@ device   mpt
 device ncr
 device sym
 device trm
-device wds
-hint.wds.0.at="isa"
-hint.wds.0.port="0x350"
-hint.wds.0.irq="11"
-hint.wds.0.drq="6"
 
 # The aic7xxx driver will attempt to use memory mapped I/O for all PCI
 # controllers that have it configured only if this option is set. 
Unfortunately,

Modified: head/sys/conf/files
==
--- head/sys/conf/files Fri Aug 19 21:14:27 2016(r304502)
+++ head/sys/conf/files Fri Aug 19 21:51:42 2016(r304503)
@@ -2930,7 +2930,6 @@ dev/vxge/vxgehal/vxgehal-channel.coptio
 dev/vxge/vxgehal/vxgehal-fifo.coptional vxge
 dev/watchdog/watchdog.cstandard
 dev/wb/if_wb.c optional wb pci
-dev/wds/wd7000.c   optional wds isa
 dev/wi/if_wi.c optional wi
 dev/wi/if_wi_pccard.c  optional wi pccard
 dev/wi/if_wi_pci.c optional wi pci

Modified: head/sys/modules/Makefile
==
--- head/sys/modules/Makefile   Fri Aug 19 21:14:27 2016(r304502)
+++ head/sys/modules/Makefile   Fri Aug 19 21:51:42 2016(r304503)
@@ -390,7 +390,6 @@ SUBDIR= \
${_vxge} \
wb \
${_wbwd} \
-   ${_wds} \
${_wi} \
${_wl} \
wlan \
@@ -718,7 +717,6 @@ _sbni=  sbni
 _streams=  streams
 _stg=  stg
 _svr4= svr4
-_wds=  wds
 .if ${MACHINE} == "i386"
 .if ${MK_EISA} != "no"
 _ahb=  ahb

Modified: head/sys/pc98/conf/NOTES
==
--- head/sys/pc98/conf/NOTESFri Aug 19 21:14:27 2016(r304502)
+++ head/sys/pc98/conf/NOTESFri Aug 19 21:51:42 2016(r304503)
@@ -595,7 +595,6 @@ nodeviceahb
 nodevice   ahd
 nodevice   mpt
 nodevice   trm
-nodevice   wds
 nodevice   dpt
 nodevice   ciss
 nodevice   iir

Modified: head/sys/sparc64/conf/NOTES
==
--- head/sys/sparc64/conf/NOTES Fri Aug 19 21:14:27 2016(r304502)
+++ head/sys/sparc64/conf/NOTES Fri Aug 19 21:51:42 2016(r304503)
@@ -121,7 +121,6 @@ nodevicesnd_vibes
 nodevice   

svn commit: r304502 - in head: . share/man/man4 sys/amd64/conf sys/conf sys/dev/si sys/i386/conf sys/modules sys/modules/si targets/pseudo/userland tools/kerneldoc/subsys usr.sbin usr.sbin/sicontrol

[John Baldwin]

Author: jhb
Date: Fri Aug 19 21:14:27 2016
New Revision: 304502
URL: https://svnweb.freebsd.org/changeset/base/304502

Log:
  Remove the si(4) driver and sicontrol(8) for Specialix serial cards.
  
  The si(4) driver supported multiport serial adapters for ISA, EISA, and
  PCI buses.  This driver does not use bus_space, instead it depends on
  direct use of the pointer returned by rman_get_virtual().  It is also
  still locked by Giant and calls for patch testing to convert it to use
  bus_space were unanswered.
  
  Relnotes: yes

Deleted:
  head/share/man/man4/si.4
  head/sys/dev/si/
  head/sys/modules/si/
  head/tools/kerneldoc/subsys/Doxyfile-dev_si
  head/usr.sbin/sicontrol/
Modified:
  head/ObsoleteFiles.inc
  head/share/man/man4/Makefile
  head/sys/amd64/conf/NOTES
  head/sys/conf/files
  head/sys/i386/conf/NOTES
  head/sys/modules/Makefile
  head/targets/pseudo/userland/Makefile.depend
  head/usr.sbin/Makefile.amd64
  head/usr.sbin/Makefile.i386

Modified: head/ObsoleteFiles.inc
==
--- head/ObsoleteFiles.inc  Fri Aug 19 20:53:55 2016(r304501)
+++ head/ObsoleteFiles.inc  Fri Aug 19 21:14:27 2016(r304502)
@@ -38,6 +38,10 @@
 #   xargs -n1 | sort | uniq -d;
 # done
 
+# 20160819: Remove si(4) and sicontrol(8)
+OLD_FILES+=usr/share/man/man4/si.4.gz
+OLD_FILES+=usr/sbin/sicontrol
+OLD_FILES+=usr/share/man/man8/sicontrol.8.gz
 # 20160819: Remove scd(4)
 OLD_FILES+=usr/share/man/man4/scd.4.gz
 # 20160815: Remove mcd(4)

Modified: head/share/man/man4/Makefile
==
--- head/share/man/man4/MakefileFri Aug 19 20:53:55 2016
(r304501)
+++ head/share/man/man4/MakefileFri Aug 19 21:14:27 2016
(r304502)
@@ -453,7 +453,6 @@ MAN=aac.4 \
sf.4 \
${_sfxge.4} \
sge.4 \
-   si.4 \
siba.4 \
siftr.4 \
siis.4 \

Modified: head/sys/amd64/conf/NOTES
==
--- head/sys/amd64/conf/NOTES   Fri Aug 19 20:53:55 2016(r304501)
+++ head/sys/amd64/conf/NOTES   Fri Aug 19 21:14:27 2016(r304502)
@@ -542,7 +542,6 @@ hint.pbio.0.port="0x360"
 device smbios
 device vpd
 device asmc
-device si
 device tpm
 device padlock_rng # VIA Padlock RNG
 device rdrand_rng  # Intel Bull Mountain RNG

Modified: head/sys/conf/files
==
--- head/sys/conf/files Fri Aug 19 20:53:55 2016(r304501)
+++ head/sys/conf/files Fri Aug 19 21:14:27 2016(r304502)
@@ -2511,12 +2511,6 @@ dev/sdhci/sdhci_if.m optional sdhci
 dev/sdhci/sdhci_pci.c  optional sdhci pci
 dev/sf/if_sf.c optional sf pci
 dev/sge/if_sge.c   optional sge pci
-dev/si/si.coptional si
-dev/si/si2_z280.c  optional si
-dev/si/si3_t225.c  optional si
-dev/si/si_eisa.c   optional si eisa
-dev/si/si_isa.coptional si isa
-dev/si/si_pci.coptional si pci
 dev/siba/siba_bwn.coptional siba_bwn pci
 dev/siba/siba_core.c   optional siba_bwn pci
 dev/siis/siis.coptional siis pci

Modified: head/sys/i386/conf/NOTES
==
--- head/sys/i386/conf/NOTESFri Aug 19 20:53:55 2016(r304501)
+++ head/sys/i386/conf/NOTESFri Aug 19 21:14:27 2016(r304502)
@@ -880,7 +880,6 @@ device  spic
 hint.spic.0.at="isa"
 hint.spic.0.port="0x10a0"
 device asmc
-device si
 device tpm
 device padlock_rng # VIA Padlock RNG
 device rdrand_rng  # Intel Bull Mountain RNG

Modified: head/sys/modules/Makefile
==
--- head/sys/modules/Makefile   Fri Aug 19 20:53:55 2016(r304501)
+++ head/sys/modules/Makefile   Fri Aug 19 21:14:27 2016(r304502)
@@ -330,7 +330,6 @@ SUBDIR= \
${_sf} \
${_sfxge} \
sge \
-   ${_si} \
siba_bwn \
siftr \
siis \
@@ -584,7 +583,6 @@ _rdma=  rdma
 .endif
 _safe= safe
 _scsi_low= scsi_low
-_si=   si
 _speaker=  speaker
 _splash=   splash
 _sppp= sppp

Modified: head/targets/pseudo/userland/Makefile.depend
==
--- head/targets/pseudo/userland/Makefile.dependFri Aug 19 20:53:55 
2016(r304501)
+++ head/targets/pseudo/userland/Makefile.dependFri Aug 19 21:14:27 
2016(r304502)
@@ -834,7 +834,6 @@ DIRDEPS.amd64= \
usr.sbin/lptcont

svn commit: r304501 - head/tools/kerneldoc/subsys

[John Baldwin]

Author: jhb
Date: Fri Aug 19 20:53:55 2016
New Revision: 304501
URL: https://svnweb.freebsd.org/changeset/base/304501

Log:
  Remove doxygen files for mcd(4) and scd(4) drivers.
  
  Submitted by: ak

Deleted:
  head/tools/kerneldoc/subsys/Doxyfile-dev_mcd
  head/tools/kerneldoc/subsys/Doxyfile-dev_scd
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304500 - releng/11.0/share/timedef

[Baptiste Daroussin]

Author: bapt
Date: Fri Aug 19 20:37:32 2016
New Revision: 304500
URL: https://svnweb.freebsd.org/changeset/base/304500

Log:
  MFC r303354,303373,303404,304045,304192
  
  Set date and time formats back to what they were before CLDR
  
  While CLDR brings us a good and up to date source data to generate locales for
  all databses we are using for locales, it is not the case of LC_TIME. Where it
  does not defines the informations we need.
  
  Put back all the date and time formats from the old locales.
  Make it statically for now (in order to be able to merge it now into
  11.0-RELEASE). The generation tools will be updated soon.
  
  That gives us time to properly work on LC_TIME during the 12 timeframe.
  
  While here fix abbreviated month for af_ZA (which are already fixed in CLDR
  data upstream)
  
  In locales where AP/PM was not defined before CLDR data, remove again the 
AP/PM
  informations
  
  For locales where AP/PM was defined before CLDR data, keep the CLDR 
information
  which was properly translated.
  
  r303354,303373 by kevlo
  r303404,304192 by jkim
  
  Approved by:  re (gjb)

Modified:
  releng/11.0/share/timedef/af_ZA.UTF-8.src
  releng/11.0/share/timedef/am_ET.UTF-8.src
  releng/11.0/share/timedef/ar_JO.UTF-8.src
  releng/11.0/share/timedef/ar_MA.UTF-8.src
  releng/11.0/share/timedef/ar_SA.UTF-8.src
  releng/11.0/share/timedef/be_BY.CP1131.src
  releng/11.0/share/timedef/be_BY.CP1251.src
  releng/11.0/share/timedef/be_BY.ISO8859-5.src
  releng/11.0/share/timedef/be_BY.UTF-8.src
  releng/11.0/share/timedef/bg_BG.CP1251.src
  releng/11.0/share/timedef/bg_BG.UTF-8.src
  releng/11.0/share/timedef/ca_IT.ISO8859-15.src
  releng/11.0/share/timedef/ca_IT.UTF-8.src
  releng/11.0/share/timedef/cs_CZ.ISO8859-2.src
  releng/11.0/share/timedef/cs_CZ.UTF-8.src
  releng/11.0/share/timedef/da_DK.ISO8859-15.src
  releng/11.0/share/timedef/da_DK.UTF-8.src
  releng/11.0/share/timedef/de_AT.ISO8859-15.src
  releng/11.0/share/timedef/de_AT.UTF-8.src
  releng/11.0/share/timedef/de_DE.ISO8859-15.src
  releng/11.0/share/timedef/de_DE.UTF-8.src
  releng/11.0/share/timedef/el_GR.ISO8859-7.src
  releng/11.0/share/timedef/el_GR.UTF-8.src
  releng/11.0/share/timedef/en_CA.UTF-8.src
  releng/11.0/share/timedef/en_GB.UTF-8.src
  releng/11.0/share/timedef/en_IE.UTF-8.src
  releng/11.0/share/timedef/en_PH.UTF-8.src
  releng/11.0/share/timedef/en_SG.UTF-8.src
  releng/11.0/share/timedef/en_US.UTF-8.src
  releng/11.0/share/timedef/en_ZA.UTF-8.src
  releng/11.0/share/timedef/es_AR.ISO8859-1.src
  releng/11.0/share/timedef/es_CR.UTF-8.src
  releng/11.0/share/timedef/es_ES.ISO8859-15.src
  releng/11.0/share/timedef/es_ES.UTF-8.src
  releng/11.0/share/timedef/es_MX.ISO8859-1.src
  releng/11.0/share/timedef/es_MX.UTF-8.src
  releng/11.0/share/timedef/et_EE.ISO8859-15.src
  releng/11.0/share/timedef/eu_ES.UTF-8.src
  releng/11.0/share/timedef/fi_FI.ISO8859-15.src
  releng/11.0/share/timedef/fi_FI.UTF-8.src
  releng/11.0/share/timedef/fr_BE.ISO8859-15.src
  releng/11.0/share/timedef/fr_BE.UTF-8.src
  releng/11.0/share/timedef/fr_CA.ISO8859-15.src
  releng/11.0/share/timedef/fr_CA.UTF-8.src
  releng/11.0/share/timedef/fr_CH.ISO8859-15.src
  releng/11.0/share/timedef/fr_CH.UTF-8.src
  releng/11.0/share/timedef/fr_FR.ISO8859-15.src
  releng/11.0/share/timedef/fr_FR.UTF-8.src
  releng/11.0/share/timedef/he_IL.UTF-8.src
  releng/11.0/share/timedef/hi_IN.ISCII-DEV.src
  releng/11.0/share/timedef/hi_IN.UTF-8.src
  releng/11.0/share/timedef/hr_HR.ISO8859-2.src
  releng/11.0/share/timedef/hr_HR.UTF-8.src
  releng/11.0/share/timedef/hu_HU.ISO8859-2.src
  releng/11.0/share/timedef/hu_HU.UTF-8.src
  releng/11.0/share/timedef/hy_AM.ARMSCII-8.src
  releng/11.0/share/timedef/hy_AM.UTF-8.src
  releng/11.0/share/timedef/is_IS.ISO8859-15.src
  releng/11.0/share/timedef/is_IS.UTF-8.src
  releng/11.0/share/timedef/it_CH.ISO8859-15.src
  releng/11.0/share/timedef/it_CH.UTF-8.src
  releng/11.0/share/timedef/it_IT.ISO8859-15.src
  releng/11.0/share/timedef/it_IT.UTF-8.src
  releng/11.0/share/timedef/ja_JP.SJIS.src
  releng/11.0/share/timedef/ja_JP.UTF-8.src
  releng/11.0/share/timedef/ja_JP.eucJP.src
  releng/11.0/share/timedef/kk_KZ.UTF-8.src
  releng/11.0/share/timedef/ko_KR.UTF-8.src
  releng/11.0/share/timedef/ko_KR.eucKR.src   (contents, props changed)
  releng/11.0/share/timedef/lt_LT.ISO8859-13.src
  releng/11.0/share/timedef/lt_LT.UTF-8.src
  releng/11.0/share/timedef/lv_LV.ISO8859-13.src
  releng/11.0/share/timedef/lv_LV.UTF-8.src
  releng/11.0/share/timedef/mn_MN.UTF-8.src
  releng/11.0/share/timedef/nb_NO.ISO8859-15.src
  releng/11.0/share/timedef/nb_NO.UTF-8.src
  releng/11.0/share/timedef/nl_BE.UTF-8.src
  releng/11.0/share/timedef/nl_NL.UTF-8.src
  releng/11.0/share/timedef/nn_NO.ISO8859-15.src
  releng/11.0/share/timedef/nn_NO.UTF-8.src
  releng/11.0/share/timedef/pl_PL.ISO8859-2.src
  releng/11.0/share/timedef/pl_PL.UTF-8.src
  releng/11.0/share/timedef/pt_BR.ISO8859-1.src
  releng/11.0/share/timedef/pt_BR.UTF-8.src
  

svn commit: r304499 - in stable: 10/lib/libc/sys 10/sys/kern 10/sys/sys 10/tests/sys/kern 11/lib/libc/sys 11/sys/kern 11/sys/sys 11/tests/sys/kern

[John Baldwin]

Author: jhb
Date: Fri Aug 19 20:17:57 2016
New Revision: 304499
URL: https://svnweb.freebsd.org/changeset/base/304499

Log:
  MFC 303001: Add PTRACE_VFORK to trace vfork events.
  
  First, PL_FLAG_FORKED events now also set a PL_FLAG_VFORKED flag when
  the new child was created via vfork() rather than fork().  Second, a
  new PL_FLAG_VFORK_DONE event can now be enabled via the PTRACE_VFORK
  event mask.  This new stop is reported after the vfork parent resumes
  due to the child calling exit or exec.  Debuggers can use this stop to
  reinsert breakpoints in the vfork parent process before it resumes.

Modified:
  stable/11/lib/libc/sys/ptrace.2
  stable/11/sys/kern/kern_fork.c
  stable/11/sys/kern/subr_syscall.c
  stable/11/sys/kern/sys_process.c
  stable/11/sys/sys/proc.h
  stable/11/sys/sys/ptrace.h
  stable/11/tests/sys/kern/ptrace_test.c
Directory Properties:
  stable/11/   (props changed)

Changes in other areas also in this revision:
Modified:
  stable/10/lib/libc/sys/ptrace.2
  stable/10/sys/kern/kern_fork.c
  stable/10/sys/kern/subr_syscall.c
  stable/10/sys/kern/sys_process.c
  stable/10/sys/sys/proc.h
  stable/10/sys/sys/ptrace.h
  stable/10/tests/sys/kern/ptrace_test.c
Directory Properties:
  stable/10/   (props changed)

Modified: stable/11/lib/libc/sys/ptrace.2
==
--- stable/11/lib/libc/sys/ptrace.2 Fri Aug 19 19:31:55 2016
(r304498)
+++ stable/11/lib/libc/sys/ptrace.2 Fri Aug 19 20:17:57 2016
(r304499)
@@ -151,6 +151,11 @@ The process ID of the new child process 
 .Va pl_child_pid
 member of
 .Vt "struct ptrace_lwpinfo" .
+If the new child process was created via
+.Xr vfork 2 ,
+the traced process's stop will also include the
+.Dv PL_FLAG_VFORKED
+flag.
 Note that new child processes will be attached with the default
 tracing event mask;
 they do not inherit the event mask of the traced process.
@@ -173,6 +178,33 @@ Note that new processes do not report an
 initial thread,
 and exiting processes do not report an event for the termination of the
 last thread.
+.It Dv PTRACE_VFORK
+Report a stop event when a parent process resumes after a
+.Xr vfork 2 .
+.Pp
+When a thread in the traced process creates a new child process via
+.Xr vfork 2 ,
+the stop that reports
+.Dv PL_FLAG_FORKED
+and
+.Dv PL_FLAG_SCX
+occurs just after the child process is created,
+but before the thread waits for the child process to stop sharing process
+memory.
+If a debugger is not tracing the new child process,
+it must ensure that no breakpoints are enabled in the shared process
+memory before detaching from the new child process.
+This means that no breakpoints are enabled in the parent process either.
+.Pp
+The
+.Dv PTRACE_VFORK
+flag enables a new stop that indicates when the new child process stops
+sharing the process memory of the parent process.
+A debugger can reinsert breakpoints in the parent process and resume it
+in response to this event.
+This event is indicated by setting the
+.Dv PL_FLAG_VFORK_DONE
+flag.
 .El
 .Pp
 The default tracing event mask when attaching to a process via
@@ -501,6 +533,16 @@ is enabled.
 Note that this event is not reported when the last LWP in a process exits.
 The termination of the last thread is reported via a normal process exit
 event.
+.It PL_FLAG_VFORKED
+Indicates that the thread is returning from a call to
+.Xr vfork 2
+that created a new child process.
+This flag is set in addition to
+.Dv PL_FLAG_FORKED .
+.It PL_FLAG_VFORK_DONE
+Indicates that the thread has resumed after a child process created via
+.Xr vfork 2
+has stopped sharing its address space with the traced process.
 .El
 .It pl_sigmask
 The current signal mask of the LWP

Modified: stable/11/sys/kern/kern_fork.c
==
--- stable/11/sys/kern/kern_fork.c  Fri Aug 19 19:31:55 2016
(r304498)
+++ stable/11/sys/kern/kern_fork.c  Fri Aug 19 20:17:57 2016
(r304499)
@@ -736,6 +736,7 @@ do_fork(struct thread *td, struct fork_r
if (fr->fr_flags & RFPPWAIT) {
td->td_pflags |= TDP_RFPPWAIT;
td->td_rfppwait_p = p2;
+   td->td_dbgflags |= TDB_VFORK;
}
PROC_UNLOCK(p2);
 

Modified: stable/11/sys/kern/subr_syscall.c
==
--- stable/11/sys/kern/subr_syscall.c   Fri Aug 19 19:31:55 2016
(r304498)
+++ stable/11/sys/kern/subr_syscall.c   Fri Aug 19 20:17:57 2016
(r304499)
@@ -242,5 +242,13 @@ again:
cv_timedwait(>p_pwait, >p_mtx, hz);
}
PROC_UNLOCK(p2);
+
+   if (td->td_dbgflags & TDB_VFORK) {
+   PROC_LOCK(p);
+   if (p->p_ptevents & PTRACE_VFORK)
+   ptracestop(td, SIGTRAP);
+   td->td_dbgflags &= ~TDB_VFORK;
+   

svn commit: r304499 - in stable: 10/lib/libc/sys 10/sys/kern 10/sys/sys 10/tests/sys/kern 11/lib/libc/sys 11/sys/kern 11/sys/sys 11/tests/sys/kern

[John Baldwin]

Author: jhb
Date: Fri Aug 19 20:17:57 2016
New Revision: 304499
URL: https://svnweb.freebsd.org/changeset/base/304499

Log:
  MFC 303001: Add PTRACE_VFORK to trace vfork events.
  
  First, PL_FLAG_FORKED events now also set a PL_FLAG_VFORKED flag when
  the new child was created via vfork() rather than fork().  Second, a
  new PL_FLAG_VFORK_DONE event can now be enabled via the PTRACE_VFORK
  event mask.  This new stop is reported after the vfork parent resumes
  due to the child calling exit or exec.  Debuggers can use this stop to
  reinsert breakpoints in the vfork parent process before it resumes.

Modified:
  stable/10/lib/libc/sys/ptrace.2
  stable/10/sys/kern/kern_fork.c
  stable/10/sys/kern/subr_syscall.c
  stable/10/sys/kern/sys_process.c
  stable/10/sys/sys/proc.h
  stable/10/sys/sys/ptrace.h
  stable/10/tests/sys/kern/ptrace_test.c
Directory Properties:
  stable/10/   (props changed)

Changes in other areas also in this revision:
Modified:
  stable/11/lib/libc/sys/ptrace.2
  stable/11/sys/kern/kern_fork.c
  stable/11/sys/kern/subr_syscall.c
  stable/11/sys/kern/sys_process.c
  stable/11/sys/sys/proc.h
  stable/11/sys/sys/ptrace.h
  stable/11/tests/sys/kern/ptrace_test.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/10/lib/libc/sys/ptrace.2
==
--- stable/10/lib/libc/sys/ptrace.2 Fri Aug 19 19:31:55 2016
(r304498)
+++ stable/10/lib/libc/sys/ptrace.2 Fri Aug 19 20:17:57 2016
(r304499)
@@ -151,6 +151,11 @@ The process ID of the new child process 
 .Va pl_child_pid
 member of
 .Vt "struct ptrace_lwpinfo" .
+If the new child process was created via
+.Xr vfork 2 ,
+the traced process's stop will also include the
+.Dv PL_FLAG_VFORKED
+flag.
 Note that new child processes will be attached with the default
 tracing event mask;
 they do not inherit the event mask of the traced process.
@@ -173,6 +178,33 @@ Note that new processes do not report an
 initial thread,
 and exiting processes do not report an event for the termination of the
 last thread.
+.It Dv PTRACE_VFORK
+Report a stop event when a parent process resumes after a
+.Xr vfork 2 .
+.Pp
+When a thread in the traced process creates a new child process via
+.Xr vfork 2 ,
+the stop that reports
+.Dv PL_FLAG_FORKED
+and
+.Dv PL_FLAG_SCX
+occurs just after the child process is created,
+but before the thread waits for the child process to stop sharing process
+memory.
+If a debugger is not tracing the new child process,
+it must ensure that no breakpoints are enabled in the shared process
+memory before detaching from the new child process.
+This means that no breakpoints are enabled in the parent process either.
+.Pp
+The
+.Dv PTRACE_VFORK
+flag enables a new stop that indicates when the new child process stops
+sharing the process memory of the parent process.
+A debugger can reinsert breakpoints in the parent process and resume it
+in response to this event.
+This event is indicated by setting the
+.Dv PL_FLAG_VFORK_DONE
+flag.
 .El
 .Pp
 The default tracing event mask when attaching to a process via
@@ -501,6 +533,16 @@ is enabled.
 Note that this event is not reported when the last LWP in a process exits.
 The termination of the last thread is reported via a normal process exit
 event.
+.It PL_FLAG_VFORKED
+Indicates that the thread is returning from a call to
+.Xr vfork 2
+that created a new child process.
+This flag is set in addition to
+.Dv PL_FLAG_FORKED .
+.It PL_FLAG_VFORK_DONE
+Indicates that the thread has resumed after a child process created via
+.Xr vfork 2
+has stopped sharing its address space with the traced process.
 .El
 .It pl_sigmask
 The current signal mask of the LWP

Modified: stable/10/sys/kern/kern_fork.c
==
--- stable/10/sys/kern/kern_fork.c  Fri Aug 19 19:31:55 2016
(r304498)
+++ stable/10/sys/kern/kern_fork.c  Fri Aug 19 20:17:57 2016
(r304499)
@@ -728,6 +728,7 @@ do_fork(struct thread *td, int flags, st
if (flags & RFPPWAIT) {
td->td_pflags |= TDP_RFPPWAIT;
td->td_rfppwait_p = p2;
+   td->td_dbgflags |= TDB_VFORK;
}
PROC_UNLOCK(p2);
if ((flags & RFSTOPPED) == 0) {

Modified: stable/10/sys/kern/subr_syscall.c
==
--- stable/10/sys/kern/subr_syscall.c   Fri Aug 19 19:31:55 2016
(r304498)
+++ stable/10/sys/kern/subr_syscall.c   Fri Aug 19 20:17:57 2016
(r304499)
@@ -249,5 +249,13 @@ again:
cv_timedwait(>p_pwait, >p_mtx, hz);
}
PROC_UNLOCK(p2);
+
+   if (td->td_dbgflags & TDB_VFORK) {
+   PROC_LOCK(p);
+   if (p->p_ptevents & PTRACE_VFORK)
+   ptracestop(td, SIGTRAP);
+   

svn commit: r304498 - in head: . share/man/man4 sys/conf sys/dev/scd sys/modules sys/modules/scd

[John Baldwin]

Author: jhb
Date: Fri Aug 19 19:31:55 2016
New Revision: 304498
URL: https://svnweb.freebsd.org/changeset/base/304498

Log:
  Remove the scd(4) driver for Sony CDU31/33 CD-ROM drives.
  
  This is a driver for a pre-ATAPI ISA CD-ROM adapter.  The driver only
  uses PIO.

Deleted:
  head/share/man/man4/scd.4
  head/sys/dev/scd/
  head/sys/modules/scd/
Modified:
  head/ObsoleteFiles.inc
  head/share/man/man4/Makefile
  head/sys/conf/NOTES
  head/sys/conf/files
  head/sys/modules/Makefile

Modified: head/ObsoleteFiles.inc
==
--- head/ObsoleteFiles.inc  Fri Aug 19 19:30:00 2016(r304497)
+++ head/ObsoleteFiles.inc  Fri Aug 19 19:31:55 2016(r304498)
@@ -38,6 +38,8 @@
 #   xargs -n1 | sort | uniq -d;
 # done
 
+# 20160819: Remove scd(4)
+OLD_FILES+=usr/share/man/man4/scd.4.gz
 # 20160815: Remove mcd(4)
 OLD_FILES+=usr/share/man/man4/mcd.4.gz
 # 20160703: POSIXify locales with variants

Modified: head/share/man/man4/Makefile
==
--- head/share/man/man4/MakefileFri Aug 19 19:30:00 2016
(r304497)
+++ head/share/man/man4/MakefileFri Aug 19 19:31:55 2016
(r304498)
@@ -441,7 +441,6 @@ MAN=aac.4 \
sbp.4 \
sbp_targ.4 \
scc.4 \
-   scd.4 \
sched_4bsd.4 \
sched_ule.4 \
screen.4 \

Modified: head/sys/conf/NOTES
==
--- head/sys/conf/NOTES Fri Aug 19 19:30:00 2016(r304497)
+++ head/sys/conf/NOTES Fri Aug 19 19:31:55 2016(r304498)
@@ -2411,15 +2411,10 @@ options SND_OLDSTEREO
 #
 # Miscellaneous hardware:
 #
-# scd: Sony CD-ROM using proprietary (non-ATAPI) interface
 # bktr: Brooktree bt848/848a/849a/878/879 video capture and TV Tuner board
 # joy: joystick (including IO DATA PCJOY PC Card joystick)
 # cmx: OmniKey CardMan 4040 pccard smartcard reader
 
-# for the Sony CDU31/33A CDROM
-device scd
-hint.scd.0.at="isa"
-hint.scd.0.port="0x230"
 device joy # PnP aware, hints for non-PnP only
 hint.joy.0.at="isa"
 hint.joy.0.port="0x201"

Modified: head/sys/conf/files
==
--- head/sys/conf/files Fri Aug 19 19:30:00 2016(r304497)
+++ head/sys/conf/files Fri Aug 19 19:31:55 2016(r304498)
@@ -2506,8 +2506,6 @@ dev/scc/scc_core.coptional scc
 dev/scc/scc_dev_quicc.coptional scc quicc
 dev/scc/scc_dev_sab82532.c optional scc
 dev/scc/scc_dev_z8530.coptional scc
-dev/scd/scd.c  optional scd isa
-dev/scd/scd_isa.c  optional scd isa
 dev/sdhci/sdhci.c  optional sdhci
 dev/sdhci/sdhci_if.m   optional sdhci
 dev/sdhci/sdhci_pci.c  optional sdhci pci

Modified: head/sys/modules/Makefile
==
--- head/sys/modules/Makefile   Fri Aug 19 19:30:00 2016(r304497)
+++ head/sys/modules/Makefile   Fri Aug 19 19:31:55 2016(r304498)
@@ -322,7 +322,6 @@ SUBDIR= \
${_safe} \
${_sbni} \
scc \
-   scd \
${_scsi_low} \
sdhci \
sdhci_pci \
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304497 - releng/11.0

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 19:30:00 2016
New Revision: 304497
URL: https://svnweb.freebsd.org/changeset/base/304497

Log:
  MFS r304475:
  
MFC r304217:
  
  Trim unneeded bootstrap after r301470 made 9.1 the minimum supported
  release.
  
  Approved by:  re (gjb)

Modified:
  releng/11.0/Makefile.inc1
Directory Properties:
  releng/11.0/   (props changed)

Modified: releng/11.0/Makefile.inc1
==
--- releng/11.0/Makefile.inc1   Fri Aug 19 19:26:16 2016(r304496)
+++ releng/11.0/Makefile.inc1   Fri Aug 19 19:30:00 2016(r304497)
@@ -1579,10 +1579,6 @@ _groff=  gnu/usr.bin/groff \
 _vtfontcvt=usr.bin/vtfontcvt
 .endif
 
-.if ${BOOTSTRAPPING} < 92
-_sed=  usr.bin/sed
-.endif
-
 .if ${BOOTSTRAPPING} < 133
 _libopenbsd=   lib/libopenbsd
 _m4=   usr.bin/m4
@@ -1621,10 +1617,6 @@ _crunchide=  usr.sbin/crunch/crunchide
 _crunchgen=usr.sbin/crunch/crunchgen
 .endif
 
-.if ${BOOTSTRAPPING} >= 900040 && ${BOOTSTRAPPING} < 900041
-_awk=  usr.bin/awk
-.endif
-
 # r296926 -P keymap search path, MFC to stable/10 in r298297
 .if ${BOOTSTRAPPING} < 1003501 || \
(${BOOTSTRAPPING} >= 110 && ${BOOTSTRAPPING} < 1100103)
@@ -1692,7 +1684,6 @@ bootstrap-tools: .PHONY
 ${_gperf} \
 ${_groff} \
 ${_dtc} \
-${_awk} \
 ${_cat} \
 ${_dd} \
 ${_kbdcontrol} \
@@ -1700,7 +1691,6 @@ bootstrap-tools: .PHONY
 ${_libopenbsd} \
 ${_makewhatis} \
 usr.bin/rpcgen \
-${_sed} \
 ${_yacc} \
 ${_m4} \
 ${_lex} \
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304496 - stable/11/share/timedef

[Baptiste Daroussin]

Author: bapt
Date: Fri Aug 19 19:26:16 2016
New Revision: 304496
URL: https://svnweb.freebsd.org/changeset/base/304496

Log:
  MFC r303354,303373,303404,304045,304192
  
  Set date and time formats back to what they were before CLDR
  
  While CLDR brings us a good and up to date source data to generate locales for
  all databses we are using for locales, it is not the case of LC_TIME. Where it
  does not defines the informations we need.
  
  Put back all the date and time formats from the old locales.
  Make it statically for now (in order to be able to merge it now into
  11.0-RELEASE). The generation tools will be updated soon.
  
  That gives us time to properly work on LC_TIME during the 12 timeframe.
  
  While here fix abbreviated month for af_ZA (which are already fixed in CLDR
  data upstream)
  
  In locales where AP/PM was not defined before CLDR data, remove again the 
AP/PM
  informations
  
  For locales where AP/PM was defined before CLDR data, keep the CLDR 
information
  which was properly translated.
  
  r303354,303373 by kevlo
  r303404,304192 by jkim

Modified:
  stable/11/share/timedef/af_ZA.UTF-8.src
  stable/11/share/timedef/am_ET.UTF-8.src
  stable/11/share/timedef/ar_JO.UTF-8.src
  stable/11/share/timedef/ar_MA.UTF-8.src
  stable/11/share/timedef/ar_SA.UTF-8.src
  stable/11/share/timedef/be_BY.CP1131.src
  stable/11/share/timedef/be_BY.CP1251.src
  stable/11/share/timedef/be_BY.ISO8859-5.src
  stable/11/share/timedef/be_BY.UTF-8.src
  stable/11/share/timedef/bg_BG.CP1251.src
  stable/11/share/timedef/bg_BG.UTF-8.src
  stable/11/share/timedef/ca_IT.ISO8859-15.src
  stable/11/share/timedef/ca_IT.UTF-8.src
  stable/11/share/timedef/cs_CZ.ISO8859-2.src
  stable/11/share/timedef/cs_CZ.UTF-8.src
  stable/11/share/timedef/da_DK.ISO8859-15.src
  stable/11/share/timedef/da_DK.UTF-8.src
  stable/11/share/timedef/de_AT.ISO8859-15.src
  stable/11/share/timedef/de_AT.UTF-8.src
  stable/11/share/timedef/de_DE.ISO8859-15.src
  stable/11/share/timedef/de_DE.UTF-8.src
  stable/11/share/timedef/el_GR.ISO8859-7.src
  stable/11/share/timedef/el_GR.UTF-8.src
  stable/11/share/timedef/en_CA.UTF-8.src
  stable/11/share/timedef/en_GB.UTF-8.src
  stable/11/share/timedef/en_IE.UTF-8.src
  stable/11/share/timedef/en_PH.UTF-8.src
  stable/11/share/timedef/en_SG.UTF-8.src
  stable/11/share/timedef/en_US.UTF-8.src
  stable/11/share/timedef/en_ZA.UTF-8.src
  stable/11/share/timedef/es_AR.ISO8859-1.src
  stable/11/share/timedef/es_CR.UTF-8.src
  stable/11/share/timedef/es_ES.ISO8859-15.src
  stable/11/share/timedef/es_ES.UTF-8.src
  stable/11/share/timedef/es_MX.ISO8859-1.src
  stable/11/share/timedef/es_MX.UTF-8.src
  stable/11/share/timedef/et_EE.ISO8859-15.src
  stable/11/share/timedef/eu_ES.UTF-8.src
  stable/11/share/timedef/fi_FI.ISO8859-15.src
  stable/11/share/timedef/fi_FI.UTF-8.src
  stable/11/share/timedef/fr_BE.ISO8859-15.src
  stable/11/share/timedef/fr_BE.UTF-8.src
  stable/11/share/timedef/fr_CA.ISO8859-15.src
  stable/11/share/timedef/fr_CA.UTF-8.src
  stable/11/share/timedef/fr_CH.ISO8859-15.src
  stable/11/share/timedef/fr_CH.UTF-8.src
  stable/11/share/timedef/fr_FR.ISO8859-15.src
  stable/11/share/timedef/fr_FR.UTF-8.src
  stable/11/share/timedef/he_IL.UTF-8.src
  stable/11/share/timedef/hi_IN.ISCII-DEV.src
  stable/11/share/timedef/hi_IN.UTF-8.src
  stable/11/share/timedef/hr_HR.ISO8859-2.src
  stable/11/share/timedef/hr_HR.UTF-8.src
  stable/11/share/timedef/hu_HU.ISO8859-2.src
  stable/11/share/timedef/hu_HU.UTF-8.src
  stable/11/share/timedef/hy_AM.ARMSCII-8.src
  stable/11/share/timedef/hy_AM.UTF-8.src
  stable/11/share/timedef/is_IS.ISO8859-15.src
  stable/11/share/timedef/is_IS.UTF-8.src
  stable/11/share/timedef/it_CH.ISO8859-15.src
  stable/11/share/timedef/it_CH.UTF-8.src
  stable/11/share/timedef/it_IT.ISO8859-15.src
  stable/11/share/timedef/it_IT.UTF-8.src
  stable/11/share/timedef/ja_JP.SJIS.src
  stable/11/share/timedef/ja_JP.UTF-8.src
  stable/11/share/timedef/ja_JP.eucJP.src
  stable/11/share/timedef/kk_KZ.UTF-8.src
  stable/11/share/timedef/ko_KR.UTF-8.src
  stable/11/share/timedef/ko_KR.eucKR.src   (contents, props changed)
  stable/11/share/timedef/lt_LT.ISO8859-13.src
  stable/11/share/timedef/lt_LT.UTF-8.src
  stable/11/share/timedef/lv_LV.ISO8859-13.src
  stable/11/share/timedef/lv_LV.UTF-8.src
  stable/11/share/timedef/mn_MN.UTF-8.src
  stable/11/share/timedef/nb_NO.ISO8859-15.src
  stable/11/share/timedef/nb_NO.UTF-8.src
  stable/11/share/timedef/nl_BE.UTF-8.src
  stable/11/share/timedef/nl_NL.UTF-8.src
  stable/11/share/timedef/nn_NO.ISO8859-15.src
  stable/11/share/timedef/nn_NO.UTF-8.src
  stable/11/share/timedef/pl_PL.ISO8859-2.src
  stable/11/share/timedef/pl_PL.UTF-8.src
  stable/11/share/timedef/pt_BR.ISO8859-1.src
  stable/11/share/timedef/pt_BR.UTF-8.src
  stable/11/share/timedef/pt_PT.ISO8859-15.src
  stable/11/share/timedef/pt_PT.UTF-8.src
  stable/11/share/timedef/ro_RO.ISO8859-2.src
  stable/11/share/timedef/ro_RO.UTF-8.src
  

Re: svn commit: r304495 - releng/11.0/sys/dev/filemon

[Bryan Drewery]

On 8/19/16 12:19 PM, Bryan Drewery wrote:
> Author: bdrewery
> Date: Fri Aug 19 19:19:17 2016
> New Revision: 304495
> URL: https://svnweb.freebsd.org/changeset/base/304495
> 
> Log:
>   MFS r304474:
>   
> MFC r304008:
>   
>   Avoid taking PROC_LOCK in syscalls if not being traced.
>   
>   Approved by:re (XXX)

Woops, Approved by gjb.

-- 
Regards,
Bryan Drewery



signature.asc
Description: OpenPGP digital signature

svn commit: r304495 - releng/11.0/sys/dev/filemon

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 19:19:17 2016
New Revision: 304495
URL: https://svnweb.freebsd.org/changeset/base/304495

Log:
  MFS r304474:
  
MFC r304008:
  
  Avoid taking PROC_LOCK in syscalls if not being traced.
  
  Approved by:  re (XXX)

Modified:
  releng/11.0/sys/dev/filemon/filemon.c
Directory Properties:
  releng/11.0/   (props changed)

Modified: releng/11.0/sys/dev/filemon/filemon.c
==
--- releng/11.0/sys/dev/filemon/filemon.c   Fri Aug 19 19:16:10 2016
(r304494)
+++ releng/11.0/sys/dev/filemon/filemon.c   Fri Aug 19 19:19:17 2016
(r304495)
@@ -137,6 +137,8 @@ filemon_proc_get(struct proc *p)
 {
struct filemon *filemon;
 
+   if (p->p_filemon == NULL)
+   return (NULL);
PROC_LOCK(p);
filemon = filemon_acquire(p->p_filemon);
PROC_UNLOCK(p);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304494 - releng/11.0

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 19:16:10 2016
New Revision: 304494
URL: https://svnweb.freebsd.org/changeset/base/304494

Log:
  MFS r304472:
  
MFC r304006:
  
  Avoid showing the bootstrap make command for check-old, etc.
  
  Approved by:  re (gjb)

Modified:
  releng/11.0/Makefile
  releng/11.0/Makefile.inc1
Directory Properties:
  releng/11.0/   (props changed)

Modified: releng/11.0/Makefile
==
--- releng/11.0/MakefileFri Aug 19 19:02:18 2016(r304493)
+++ releng/11.0/MakefileFri Aug 19 19:16:10 2016(r304494)
@@ -209,7 +209,8 @@ SUB_MAKE= `test -x ${MYMAKE} && echo ${M
 SUB_MAKE= ${MAKE} -m ${.CURDIR}/share/mk
 .endif
 
-_MAKE= PATH=${PATH} ${SUB_MAKE} -f Makefile.inc1 TARGET=${_TARGET} 
TARGET_ARCH=${_TARGET_ARCH}
+_MAKE= PATH=${PATH} MAKE_CMD=${MAKE} ${SUB_MAKE} -f Makefile.inc1 \
+   TARGET=${_TARGET} TARGET_ARCH=${_TARGET_ARCH}
 
 # Only allow meta mode for the whitelisted targets.  See META_TGT_WHITELIST
 # above.

Modified: releng/11.0/Makefile.inc1
==
--- releng/11.0/Makefile.inc1   Fri Aug 19 19:02:18 2016(r304493)
+++ releng/11.0/Makefile.inc1   Fri Aug 19 19:16:10 2016(r304494)
@@ -2372,11 +2372,11 @@ check-old-dirs: .PHONY
done
 
 delete-old: delete-old-files delete-old-dirs .PHONY
-   @echo "To remove old libraries run '${MAKE} delete-old-libs'."
+   @echo "To remove old libraries run '${MAKE_CMD} delete-old-libs'."
 
 check-old: check-old-files check-old-libs check-old-dirs .PHONY
-   @echo "To remove old files and directories run '${MAKE} delete-old'."
-   @echo "To remove old libraries run '${MAKE} delete-old-libs'."
+   @echo "To remove old files and directories run '${MAKE_CMD} 
delete-old'."
+   @echo "To remove old libraries run '${MAKE_CMD} delete-old-libs'."
 
 .endif
 
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304493 - releng/11.0/sys/dev/e1000

[Sean Bruno]

Author: sbruno
Date: Fri Aug 19 19:02:18 2016
New Revision: 304493
URL: https://svnweb.freebsd.org/changeset/base/304493

Log:
  MF stable11 r304337
  e1000:  Add support for Kaby Lake IDs
  Fixup some errors when transitioning to/from low power states.
  
  Approved by:  re (kib)

Modified:
  releng/11.0/sys/dev/e1000/e1000_api.c
  releng/11.0/sys/dev/e1000/e1000_hw.h
  releng/11.0/sys/dev/e1000/e1000_ich8lan.c
  releng/11.0/sys/dev/e1000/e1000_ich8lan.h
  releng/11.0/sys/dev/e1000/e1000_phy.c
  releng/11.0/sys/dev/e1000/if_em.c
Directory Properties:
  releng/11.0/   (props changed)

Modified: releng/11.0/sys/dev/e1000/e1000_api.c
==
--- releng/11.0/sys/dev/e1000/e1000_api.c   Fri Aug 19 18:45:42 2016
(r304492)
+++ releng/11.0/sys/dev/e1000/e1000_api.c   Fri Aug 19 19:02:18 2016
(r304493)
@@ -304,6 +304,10 @@ s32 e1000_set_mac_type(struct e1000_hw *
case E1000_DEV_ID_PCH_SPT_I219_LM2:
case E1000_DEV_ID_PCH_SPT_I219_V2:
case E1000_DEV_ID_PCH_LBG_I219_LM3:
+   case E1000_DEV_ID_PCH_SPT_I219_LM4:
+   case E1000_DEV_ID_PCH_SPT_I219_V4:
+   case E1000_DEV_ID_PCH_SPT_I219_LM5:
+   case E1000_DEV_ID_PCH_SPT_I219_V5:
mac->type = e1000_pch_spt;
break;
case E1000_DEV_ID_82575EB_COPPER:

Modified: releng/11.0/sys/dev/e1000/e1000_hw.h
==
--- releng/11.0/sys/dev/e1000/e1000_hw.hFri Aug 19 18:45:42 2016
(r304492)
+++ releng/11.0/sys/dev/e1000/e1000_hw.hFri Aug 19 19:02:18 2016
(r304493)
@@ -142,6 +142,10 @@ struct e1000_hw;
 #define E1000_DEV_ID_PCH_SPT_I219_LM2  0x15B7 /* Sunrise Point-H PCH */
 #define E1000_DEV_ID_PCH_SPT_I219_V2   0x15B8 /* Sunrise Point-H PCH */
 #define E1000_DEV_ID_PCH_LBG_I219_LM3  0x15B9 /* LEWISBURG PCH */
+#define E1000_DEV_ID_PCH_SPT_I219_LM4  0x15D7
+#define E1000_DEV_ID_PCH_SPT_I219_V4   0x15D8
+#define E1000_DEV_ID_PCH_SPT_I219_LM5  0x15E3
+#define E1000_DEV_ID_PCH_SPT_I219_V5   0x15D6
 #define E1000_DEV_ID_82576 0x10C9
 #define E1000_DEV_ID_82576_FIBER   0x10E6
 #define E1000_DEV_ID_82576_SERDES  0x10E7
@@ -957,9 +961,13 @@ struct e1000_dev_spec_ich8lan {
E1000_MUTEX nvm_mutex;
E1000_MUTEX swflag_mutex;
bool nvm_k1_enabled;
+   bool disable_k1_off;
bool eee_disable;
u16 eee_lp_ability;
enum e1000_ulp_state ulp_state;
+   bool ulp_capability_disabled;
+   bool during_suspend_flow;
+   bool during_dpg_exit;
 };
 
 struct e1000_dev_spec_82575 {

Modified: releng/11.0/sys/dev/e1000/e1000_ich8lan.c
==
--- releng/11.0/sys/dev/e1000/e1000_ich8lan.c   Fri Aug 19 18:45:42 2016
(r304492)
+++ releng/11.0/sys/dev/e1000/e1000_ich8lan.c   Fri Aug 19 19:02:18 2016
(r304493)
@@ -288,7 +288,7 @@ static void e1000_toggle_lanphypc_pch_lp
mac_reg &= ~E1000_CTRL_LANPHYPC_VALUE;
E1000_WRITE_REG(hw, E1000_CTRL, mac_reg);
E1000_WRITE_FLUSH(hw);
-   usec_delay(10);
+   msec_delay(1);
mac_reg &= ~E1000_CTRL_LANPHYPC_OVERRIDE;
E1000_WRITE_REG(hw, E1000_CTRL, mac_reg);
E1000_WRITE_FLUSH(hw);
@@ -1625,7 +1625,17 @@ static s32 e1000_check_for_copper_link_i
hw->phy.ops.write_reg_locked(hw,
 I217_PLL_CLOCK_GATE_REG,
 phy_reg);
-   }
+
+   if (speed == SPEED_1000) {
+   hw->phy.ops.read_reg_locked(hw, HV_PM_CTRL,
+   _reg);
+
+   phy_reg |= HV_PM_CTRL_K1_CLK_REQ;
+
+   hw->phy.ops.write_reg_locked(hw, HV_PM_CTRL,
+phy_reg);
+   }
+}
hw->phy.ops.release(hw);
 
if (ret_val)
@@ -1718,7 +1728,8 @@ static s32 e1000_check_for_copper_link_i
u32 pcieanacfg = E1000_READ_REG(hw, E1000_PCIEANACFG);
u32 fextnvm6 = E1000_READ_REG(hw, E1000_FEXTNVM6);
 
-   if (pcieanacfg & E1000_FEXTNVM6_K1_OFF_ENABLE)
+   if ((pcieanacfg & E1000_FEXTNVM6_K1_OFF_ENABLE) &&
+   (hw->dev_spec.ich8lan.disable_k1_off == FALSE))
fextnvm6 |= E1000_FEXTNVM6_K1_OFF_ENABLE;
else
fextnvm6 &= ~E1000_FEXTNVM6_K1_OFF_ENABLE;

Modified: releng/11.0/sys/dev/e1000/e1000_ich8lan.h
==
--- releng/11.0/sys/dev/e1000/e1000_ich8lan.h 

svn commit: r304492 - head/sys/conf

[John Baldwin]

Author: jhb
Date: Fri Aug 19 18:45:42 2016
New Revision: 304492
URL: https://svnweb.freebsd.org/changeset/base/304492

Log:
  Move cxgb and cxgbe down to the non-mii PCI NIC section.

Modified:
  head/sys/conf/NOTES

Modified: head/sys/conf/NOTES
==
--- head/sys/conf/NOTES Fri Aug 19 18:40:47 2016(r304491)
+++ head/sys/conf/NOTES Fri Aug 19 18:45:42 2016(r304492)
@@ -2116,9 +2116,6 @@ devicebce # Broadcom 
BCM5706/BCM5708 
 device bfe # Broadcom BCM440x 10/100 Ethernet
 device bge # Broadcom BCM570xx Gigabit Ethernet
 device cas # Sun Cassini/Cassini+ and NS DP83065 Saturn
-device cxgb# Chelsio T3 10 Gigabit Ethernet
-device cxgb_t3fw   # Chelsio T3 10 Gigabit Ethernet firmware
-device cxgbe   # Chelsio T4 and T5 1GbE/10GbE/40GbE
 device dc  # DEC/Intel 21143 and various workalikes
 device et  # Agere ET1310 10/100/Gigabit Ethernet
 device fxp # Intel EtherExpress PRO/100B (82557, 82558)
@@ -2149,6 +2146,9 @@ devicewb  # Winbond W89C840F
 device xl  # 3Com 3c90x (``Boomerang'', ``Cyclone'')
 
 # PCI Ethernet NICs.
+device cxgb# Chelsio T3 10 Gigabit Ethernet
+device cxgb_t3fw   # Chelsio T3 10 Gigabit Ethernet firmware
+device cxgbe   # Chelsio T4 and T5 1GbE/10GbE/40GbE
 device de  # DEC/Intel DC21x4x (``Tulip'')
 device em  # Intel Pro/1000 Gigabit Ethernet
 device igb # Intel Pro/1000 PCIE Gigabit Ethernet
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304491 - releng/11.0/sys/conf

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 18:40:47 2016
New Revision: 304491
URL: https://svnweb.freebsd.org/changeset/base/304491

Log:
  MFS r304470:
  
MFC r304005:
  
  PORTS_MODULES: Don't leak in CC/CXX/CPP.
  
  Approved by:  re (gjb)

Modified:
  releng/11.0/sys/conf/kern.post.mk
Directory Properties:
  releng/11.0/   (props changed)

Modified: releng/11.0/sys/conf/kern.post.mk
==
--- releng/11.0/sys/conf/kern.post.mk   Fri Aug 19 18:14:33 2016
(r304490)
+++ releng/11.0/sys/conf/kern.post.mk   Fri Aug 19 18:40:47 2016
(r304491)
@@ -65,6 +65,10 @@ OSRELDATE!=  awk '/^\#define[[:space:]]*_
 # Keep the related ports builds in the obj directory so that they are only 
rebuilt once per kernel build
 WRKDIRPREFIX?= ${MAKEOBJDIRPREFIX}${SRC_BASE}/sys/${KERNCONF}
 PORTSMODULESENV=\
+   env \
+   -u CC \
+   -u CXX \
+   -u CPP \
PATH=${PATH}:${LOCALBASE}/bin:${LOCALBASE}/sbin \
SRC_BASE=${SRC_BASE} \
OSVERSION=${OSRELDATE} \
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304490 - releng/11.0/usr.bin/xinstall

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 18:14:33 2016
New Revision: 304490
URL: https://svnweb.freebsd.org/changeset/base/304490

Log:
  MFS r304469:
  
MFC r303929,r303930,r303931,r303932,r303933:
  
  r303929:
Fix -S with -b not atomically updating the destination file.
  r303930:
Support -v for -l.
  r303931:
Fix -S with -l not being atomic.
  r303932:
Fix -b failure not restoring flags on the destination file.
  r303933:
Squelch a false-positive Clang static analyzer warning.
  
  Approved by:  re (gjb)

Modified:
  releng/11.0/usr.bin/xinstall/xinstall.c
Directory Properties:
  releng/11.0/   (props changed)

Modified: releng/11.0/usr.bin/xinstall/xinstall.c
==
--- releng/11.0/usr.bin/xinstall/xinstall.c Fri Aug 19 18:08:59 2016
(r304489)
+++ releng/11.0/usr.bin/xinstall/xinstall.c Fri Aug 19 18:14:33 2016
(r304490)
@@ -149,6 +149,7 @@ main(int argc, char *argv[])
char *p;
const char *to_name;
 
+   fset = 0;
iflags = 0;
group = owner = NULL;
while ((ch = getopt(argc, argv, "B:bCcD:df:g:h:l:M:m:N:o:pSsT:Uv")) !=
@@ -533,7 +534,9 @@ do_link(const char *from_name, const cha
if (target_sb->st_flags & NOCHANGEBITS)
(void)chflags(to_name, target_sb->st_flags &
 ~NOCHANGEBITS);
-   unlink(to_name);
+   if (verbose)
+   printf("install: link %s -> %s\n",
+   from_name, to_name);
ret = rename(tmpl, to_name);
/*
 * If rename has posix semantics, then the temporary
@@ -543,8 +546,12 @@ do_link(const char *from_name, const cha
(void)unlink(tmpl);
}
return (ret);
-   } else
+   } else {
+   if (verbose)
+   printf("install: link %s -> %s\n",
+   from_name, to_name);
return (link(from_name, to_name));
+   }
 }
 
 /*
@@ -573,14 +580,18 @@ do_symlink(const char *from_name, const 
if (target_sb->st_flags & NOCHANGEBITS)
(void)chflags(to_name, target_sb->st_flags &
 ~NOCHANGEBITS);
-   unlink(to_name);
-
+   if (verbose)
+   printf("install: symlink %s -> %s\n",
+   from_name, to_name);
if (rename(tmpl, to_name) == -1) {
/* Remove temporary link before exiting. */
(void)unlink(tmpl);
err(EX_OSERR, "%s: rename", to_name);
}
} else {
+   if (verbose)
+   printf("install: symlink %s -> %s\n",
+   from_name, to_name);
if (symlink(from_name, to_name) == -1)
err(EX_OSERR, "symlink %s -> %s", from_name, to_name);
}
@@ -886,11 +897,21 @@ install(const char *from_name, const cha
}
if (verbose)
(void)printf("install: %s -> %s\n", to_name, 
backup);
-   if (rename(to_name, backup) < 0) {
+   if (unlink(backup) < 0 && errno != ENOENT) {
+   serrno = errno;
+   if (to_sb.st_flags & NOCHANGEBITS)
+   (void)chflags(to_name, to_sb.st_flags);
+   unlink(tempfile);
+   errno = serrno;
+   err(EX_OSERR, "unlink: %s", backup);
+   }
+   if (link(to_name, backup) < 0) {
serrno = errno;
unlink(tempfile);
+   if (to_sb.st_flags & NOCHANGEBITS)
+   (void)chflags(to_name, to_sb.st_flags);
errno = serrno;
-   err(EX_OSERR, "rename: %s to %s", to_name,
+   err(EX_OSERR, "link: %s to %s", to_name,
 backup);
}
}
@@ -,16 +1132,26 @@ create_newfile(const char *path, int tar
 
if (dobackup) {
if ((size_t)snprintf(backup, MAXPATHLEN, "%s%s",
-   path, suffix) != strlen(path) + strlen(suffix))
+   path, suffix) != strlen(path) + strlen(suffix)) {
+   saved_errno = errno;
+   if (sbp->st_flags & NOCHANGEBITS)
+   

svn commit: r304489 - releng/11.0/share/mk

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 18:08:59 2016
New Revision: 304489
URL: https://svnweb.freebsd.org/changeset/base/304489

Log:
  MFS r304144:
  
MFC r303964:
  
  PROGS: Support INTERNALPROG.prog=yes to not install it.
  
  Approved by:  re (gjb)

Modified:
  releng/11.0/share/mk/bsd.README
  releng/11.0/share/mk/bsd.progs.mk
Directory Properties:
  releng/11.0/   (props changed)

Modified: releng/11.0/share/mk/bsd.README
==
--- releng/11.0/share/mk/bsd.README Fri Aug 19 18:02:36 2016
(r304488)
+++ releng/11.0/share/mk/bsd.README Fri Aug 19 18:08:59 2016
(r304489)
@@ -331,6 +331,7 @@ PROGS_CXX   PROG and PROGS_CXX in one Make
- DEBUG_FLAGS
- DPADD
- DPSRCS
+   - INTERNALPROG (no installation)
- LDADD
- LDFLAGS
- LIBADD

Modified: releng/11.0/share/mk/bsd.progs.mk
==
--- releng/11.0/share/mk/bsd.progs.mk   Fri Aug 19 18:02:36 2016
(r304488)
+++ releng/11.0/share/mk/bsd.progs.mk   Fri Aug 19 18:08:59 2016
(r304489)
@@ -24,8 +24,8 @@ PROGS += ${PROGS_CXX}
 # just one of many
 PROG_OVERRIDE_VARS +=  BINDIR BINGRP BINOWN BINMODE DPSRCS MAN NO_WERROR \
PROGNAME SRCS STRIP WARNS
-PROG_VARS +=   CFLAGS CXXFLAGS DEBUG_FLAGS DPADD LDADD LIBADD LINKS \
-   LDFLAGS MLINKS ${PROG_OVERRIDE_VARS}
+PROG_VARS +=   CFLAGS CXXFLAGS DEBUG_FLAGS DPADD INTERNALPROG LDADD LIBADD \
+   LINKS LDFLAGS MLINKS ${PROG_OVERRIDE_VARS}
 .for v in ${PROG_VARS:O:u}
 .if empty(${PROG_OVERRIDE_VARS:M$v})
 .if defined(${v}.${PROG})
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304488 - head/sys/arm/arm

[Emmanuel Vadot]

Author: manu
Date: Fri Aug 19 18:02:36 2016
New Revision: 304488
URL: https://svnweb.freebsd.org/changeset/base/304488

Log:
  Keep boot parameters in ARM trampoline code
  
  Currently boot parameters (r0 - r3) are forgotten in ARM trampoline code.
  This patch save them at startup and restore them before jumping into kernel
  _start() routine.
  This is usefull when booting with Linux ABI and/or custom bootloader.
  
  Submitted by: Grégory Soutadé 
  Reviewed by:  imp
  Differential Revision:https://reviews.freebsd.org/D7395

Modified:
  head/sys/arm/arm/elf_trampoline.c

Modified: head/sys/arm/arm/elf_trampoline.c
==
--- head/sys/arm/arm/elf_trampoline.c   Fri Aug 19 17:57:14 2016
(r304487)
+++ head/sys/arm/arm/elf_trampoline.c   Fri Aug 19 18:02:36 2016
(r304488)
@@ -39,6 +39,7 @@ __FBSDID("$FreeBSD$");
 #include 
 #include 
 #include 
+#include 
 
 extern char kernel_start[];
 extern char kernel_end[];
@@ -47,7 +48,7 @@ extern void *_end;
 
 void _start(void);
 void __start(void);
-void __startC(void);
+void __startC(unsigned r0, unsigned r1, unsigned r2, unsigned r3);
 
 extern unsigned int cpu_ident(void);
 extern void armv6_idcache_wbinv_all(void);
@@ -124,6 +125,10 @@ static int  arm_dcache_l2_nsets;
 static int  arm_dcache_l2_assoc;
 static int  arm_dcache_l2_linesize;
 
+/*
+ * Boot parameters
+ */
+static struct arm_boot_params s_boot_params;
 
 extern int arm9_dcache_sets_inc;
 extern int arm9_dcache_sets_max;
@@ -172,12 +177,17 @@ bzero(void *addr, int count)
 static void arm9_setup(void);
 
 void
-_startC(void)
+_startC(unsigned r0, unsigned r1, unsigned r2, unsigned r3)
 {
int tmp1;
unsigned int sp = ((unsigned int)&_end & ~3) + 4;
unsigned int pc, kernphysaddr;
 
+   s_boot_params.abp_r0 = r0;
+   s_boot_params.abp_r1 = r1;
+   s_boot_params.abp_r2 = r2;
+   s_boot_params.abp_r3 = r3;
+
/*
 * Figure out the physical address the kernel was loaded at.  This
 * assumes the entry point (this code right here) is in the first page,
@@ -211,8 +221,15 @@ _startC(void)
/* Temporary set the sp and jump to the new location. */
__asm __volatile(
"mov sp, %1\n"
+   "mov r0, %2\n"
+   "mov r1, %3\n"
+   "mov r2, %4\n"
+   "mov r3, %5\n"
"mov pc, %0\n"
-   : : "r" (target_addr), "r" (tmp_sp));
+   : : "r" (target_addr), "r" (tmp_sp),
+   "r" (s_boot_params.abp_r0), "r" (s_boot_params.abp_r1),
+   "r" (s_boot_params.abp_r2), "r" (s_boot_params.abp_r3),
+   : "r0", "r1", "r2", "r3");
 
}
 #endif
@@ -487,6 +504,7 @@ load_kernel(unsigned int kstart, unsigne
vm_offset_t lastaddr = 0;
Elf_Addr ssym = 0;
Elf_Dyn *dp;
+   struct arm_boot_params local_boot_params;
 
eh = (Elf32_Ehdr *)kstart;
ssym = 0;
@@ -555,6 +573,12 @@ load_kernel(unsigned int kstart, unsigne
if (!d)
return ((void *)lastaddr);
 
+   /*
+* Now the stack is fixed, copy boot params
+* before it's overrided
+*/
+   memcpy(_boot_params, _boot_params, sizeof(local_boot_params));
+
j = eh->e_phnum;
for (i = 0; i < j; i++) {
volatile char c;
@@ -604,7 +628,10 @@ load_kernel(unsigned int kstart, unsigne
"mcr p15, 0, %0, c1, c0, 0\n" /* CP15_SCTLR(%0)*/
: "=r" (ssym));
/* Jump to the entry point. */
-   ((void(*)(void))(entry_point - KERNVIRTADDR + curaddr))();
+   ((void(*)(unsigned, unsigned, unsigned, unsigned))
+   (entry_point - KERNVIRTADDR + curaddr))
+   (local_boot_params.abp_r0, local_boot_params.abp_r1,
+   local_boot_params.abp_r2, local_boot_params.abp_r3);
__asm __volatile(".globl func_end\n"
"func_end:");
 
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304487 - in head/sys: kern sys

[Mark Johnston]

Author: markj
Date: Fri Aug 19 17:57:14 2016
New Revision: 304487
URL: https://svnweb.freebsd.org/changeset/base/304487

Log:
  Don't set P2_PTRACE_FSTP in a process that invokes ptrace(PT_TRACE_ME).
  
  Such processes are stopped synchronously by a direct call to
  ptracestop(SIGTRAP) upon exec. P2_PTRACE_FSTP causes the exec()ing thread
  to suspend itself while waiting for a SIGSTOP that never arrives.
  
  Reviewed by:  kib
  MFC after:3 days
  Differential Revision:https://reviews.freebsd.org/D7576

Modified:
  head/sys/kern/kern_fork.c
  head/sys/kern/sys_process.c
  head/sys/sys/proc.h

Modified: head/sys/kern/kern_fork.c
==
--- head/sys/kern/kern_fork.c   Fri Aug 19 17:55:34 2016(r304486)
+++ head/sys/kern/kern_fork.c   Fri Aug 19 17:57:14 2016(r304487)
@@ -1074,7 +1074,7 @@ fork_return(struct thread *td, struct tr
 * parent's children, do it now.
 */
dbg = p->p_pptr->p_pptr;
-   proc_set_traced(p);
+   proc_set_traced(p, true);
CTR2(KTR_PTRACE,
"fork_return: attaching to new child pid %d: oppid %d",
p->p_pid, p->p_oppid);

Modified: head/sys/kern/sys_process.c
==
--- head/sys/kern/sys_process.c Fri Aug 19 17:55:34 2016(r304486)
+++ head/sys/kern/sys_process.c Fri Aug 19 17:57:14 2016(r304487)
@@ -693,12 +693,13 @@ sys_ptrace(struct thread *td, struct ptr
 #endif
 
 void
-proc_set_traced(struct proc *p)
+proc_set_traced(struct proc *p, bool stop)
 {
 
PROC_LOCK_ASSERT(p, MA_OWNED);
p->p_flag |= P_TRACED;
-   p->p_flag2 |= P2_PTRACE_FSTP;
+   if (stop)
+   p->p_flag2 |= P2_PTRACE_FSTP;
p->p_ptevents = PTRACE_DEFAULT;
p->p_oppid = p->p_pptr->p_pid;
 }
@@ -910,7 +911,7 @@ kern_ptrace(struct thread *td, int req, 
switch (req) {
case PT_TRACE_ME:
/* set my trace flag and "owner" so it can read/write me */
-   proc_set_traced(p);
+   proc_set_traced(p, false);
if (p->p_flag & P_PPWAIT)
p->p_flag |= P_PPTRACE;
CTR1(KTR_PTRACE, "PT_TRACE_ME: pid %d", p->p_pid);
@@ -927,7 +928,7 @@ kern_ptrace(struct thread *td, int req, 
 * The old parent is remembered so we can put things back
 * on a "detach".
 */
-   proc_set_traced(p);
+   proc_set_traced(p, true);
if (p->p_pptr != td->td_proc) {
proc_reparent(p, td->td_proc);
}

Modified: head/sys/sys/proc.h
==
--- head/sys/sys/proc.h Fri Aug 19 17:55:34 2016(r304486)
+++ head/sys/sys/proc.h Fri Aug 19 17:57:14 2016(r304487)
@@ -1010,7 +1010,7 @@ void  proc_linkup(struct proc *p, struct 
 struct proc *proc_realparent(struct proc *child);
 void   proc_reap(struct thread *td, struct proc *p, int *status, int options);
 void   proc_reparent(struct proc *child, struct proc *newparent);
-void   proc_set_traced(struct proc *p);
+void   proc_set_traced(struct proc *p, bool stop);
 struct pstats *pstats_alloc(void);
 void   pstats_fork(struct pstats *src, struct pstats *dst);
 void   pstats_free(struct pstats *ps);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304485 - head/sys/dev/cxgbe

[John Baldwin]

Author: jhb
Date: Fri Aug 19 17:54:51 2016
New Revision: 304485
URL: https://svnweb.freebsd.org/changeset/base/304485

Log:
  Reorder sysctls so that nodes shared with the VF driver are added first.
  
  This permits a single early return for VF devices in the routines that
  add sysctl nodes.
  
  Reviewed by:  np
  Sponsored by: Chelsio Communications
  Differential Revision:https://reviews.freebsd.org/D7512

Modified:
  head/sys/dev/cxgbe/t4_main.c

Modified: head/sys/dev/cxgbe/t4_main.c
==
--- head/sys/dev/cxgbe/t4_main.cFri Aug 19 17:54:06 2016
(r304484)
+++ head/sys/dev/cxgbe/t4_main.cFri Aug 19 17:54:51 2016
(r304485)
@@ -4592,6 +4592,32 @@ t4_sysctls(struct adapter *sc)
SYSCTL_ADD_INT(ctx, children, OID_AUTO, "nports", CTLFLAG_RD, NULL,
sc->params.nports, "# of ports");
 
+   SYSCTL_ADD_PROC(ctx, children, OID_AUTO, "doorbells",
+   CTLTYPE_STRING | CTLFLAG_RD, doorbells, sc->doorbells,
+   sysctl_bitfield, "A", "available doorbells");
+
+   SYSCTL_ADD_INT(ctx, children, OID_AUTO, "core_clock", CTLFLAG_RD, NULL,
+   sc->params.vpd.cclk, "core clock frequency (in KHz)");
+
+   SYSCTL_ADD_PROC(ctx, children, OID_AUTO, "holdoff_timers",
+   CTLTYPE_STRING | CTLFLAG_RD, sc->params.sge.timer_val,
+   sizeof(sc->params.sge.timer_val), sysctl_int_array, "A",
+   "interrupt holdoff timer values (us)");
+
+   SYSCTL_ADD_PROC(ctx, children, OID_AUTO, "holdoff_pkt_counts",
+   CTLTYPE_STRING | CTLFLAG_RD, sc->params.sge.counter_val,
+   sizeof(sc->params.sge.counter_val), sysctl_int_array, "A",
+   "interrupt holdoff packet counter values");
+
+   t4_sge_sysctls(sc, ctx, children);
+
+   sc->lro_timeout = 100;
+   SYSCTL_ADD_INT(ctx, children, OID_AUTO, "lro_timeout", CTLFLAG_RW,
+   >lro_timeout, 0, "lro inactive-flush timeout (in us)");
+
+   SYSCTL_ADD_INT(ctx, children, OID_AUTO, "debug_flags", CTLFLAG_RW,
+   >debug_flags, 0, "flags to enable runtime debugging");
+
SYSCTL_ADD_INT(ctx, children, OID_AUTO, "hw_revision", CTLFLAG_RD,
NULL, chip_rev(sc), "chip hardware revision");
 
@@ -4612,10 +4638,6 @@ t4_sysctls(struct adapter *sc)
SYSCTL_ADD_UINT(ctx, children, OID_AUTO, "cfcsum", CTLFLAG_RD, NULL,
sc->cfcsum, "config file checksum");
 
-   SYSCTL_ADD_PROC(ctx, children, OID_AUTO, "doorbells",
-   CTLTYPE_STRING | CTLFLAG_RD, doorbells, sc->doorbells,
-   sysctl_bitfield, "A", "available doorbells");
-
 #define SYSCTL_CAP(name, n, text) \
SYSCTL_ADD_PROC(ctx, children, OID_AUTO, #name, \
CTLTYPE_STRING | CTLFLAG_RD, caps_decoder[n], sc->name, \
@@ -4632,19 +4654,6 @@ t4_sysctls(struct adapter *sc)
SYSCTL_CAP(fcoecaps, 8, "FCoE");
 #undef SYSCTL_CAP
 
-   SYSCTL_ADD_INT(ctx, children, OID_AUTO, "core_clock", CTLFLAG_RD, NULL,
-   sc->params.vpd.cclk, "core clock frequency (in KHz)");
-
-   SYSCTL_ADD_PROC(ctx, children, OID_AUTO, "holdoff_timers",
-   CTLTYPE_STRING | CTLFLAG_RD, sc->params.sge.timer_val,
-   sizeof(sc->params.sge.timer_val), sysctl_int_array, "A",
-   "interrupt holdoff timer values (us)");
-
-   SYSCTL_ADD_PROC(ctx, children, OID_AUTO, "holdoff_pkt_counts",
-   CTLTYPE_STRING | CTLFLAG_RD, sc->params.sge.counter_val,
-   sizeof(sc->params.sge.counter_val), sysctl_int_array, "A",
-   "interrupt holdoff packet counter values");
-
SYSCTL_ADD_INT(ctx, children, OID_AUTO, "nfilters", CTLFLAG_RD,
NULL, sc->tids.nftids, "number of filters");
 
@@ -4652,15 +4661,6 @@ t4_sysctls(struct adapter *sc)
CTLFLAG_RD, sc, 0, sysctl_temperature, "I",
"chip temperature (in Celsius)");
 
-   t4_sge_sysctls(sc, ctx, children);
-
-   sc->lro_timeout = 100;
-   SYSCTL_ADD_INT(ctx, children, OID_AUTO, "lro_timeout", CTLFLAG_RW,
-   >lro_timeout, 0, "lro inactive-flush timeout (in us)");
-
-   SYSCTL_ADD_INT(ctx, children, OID_AUTO, "debug_flags", CTLFLAG_RW,
-   >debug_flags, 0, "flags to enable runtime debugging");
-
 #ifdef SBUF_DRAIN
/*
 * dev.t4nex.X.misc.  Marked CTLFLAG_SKIP to avoid information overload.
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304484 - head/sys/compat/cloudabi64

[Ed Schouten]

Author: ed
Date: Fri Aug 19 17:54:06 2016
New Revision: 304484
URL: https://svnweb.freebsd.org/changeset/base/304484

Log:
  Regenerate system call table after r304483.

Modified:
  head/sys/compat/cloudabi64/cloudabi64_proto.h
  head/sys/compat/cloudabi64/cloudabi64_syscall.h
  head/sys/compat/cloudabi64/cloudabi64_syscalls.c
  head/sys/compat/cloudabi64/cloudabi64_sysent.c

Modified: head/sys/compat/cloudabi64/cloudabi64_proto.h
==
--- head/sys/compat/cloudabi64/cloudabi64_proto.h   Fri Aug 19 17:53:37 
2016(r304483)
+++ head/sys/compat/cloudabi64/cloudabi64_proto.h   Fri Aug 19 17:54:06 
2016(r304484)
@@ -3,7 +3,7 @@
  *
  * DO NOT EDIT-- this file is automatically generated.
  * $FreeBSD$
- * created from FreeBSD
+ * created from FreeBSD: head/sys/contrib/cloudabi/syscalls64.master 304483 
2016-08-19 17:53:37Z ed 
  */
 
 #ifndef _CLOUDABI64_SYSPROTO_H_

Modified: head/sys/compat/cloudabi64/cloudabi64_syscall.h
==
--- head/sys/compat/cloudabi64/cloudabi64_syscall.h Fri Aug 19 17:53:37 
2016(r304483)
+++ head/sys/compat/cloudabi64/cloudabi64_syscall.h Fri Aug 19 17:54:06 
2016(r304484)
@@ -3,7 +3,7 @@
  *
  * DO NOT EDIT-- this file is automatically generated.
  * $FreeBSD$
- * created from FreeBSD
+ * created from FreeBSD: head/sys/contrib/cloudabi/syscalls64.master 304483 
2016-08-19 17:53:37Z ed 
  */
 
 #defineCLOUDABI64_SYS_cloudabi_sys_clock_res_get   0

Modified: head/sys/compat/cloudabi64/cloudabi64_syscalls.c
==
--- head/sys/compat/cloudabi64/cloudabi64_syscalls.cFri Aug 19 17:53:37 
2016(r304483)
+++ head/sys/compat/cloudabi64/cloudabi64_syscalls.cFri Aug 19 17:54:06 
2016(r304484)
@@ -3,7 +3,7 @@
  *
  * DO NOT EDIT-- this file is automatically generated.
  * $FreeBSD$
- * created from FreeBSD
+ * created from FreeBSD: head/sys/contrib/cloudabi/syscalls64.master 304483 
2016-08-19 17:53:37Z ed 
  */
 
 const char *cloudabi64_syscallnames[] = {

Modified: head/sys/compat/cloudabi64/cloudabi64_sysent.c
==
--- head/sys/compat/cloudabi64/cloudabi64_sysent.c  Fri Aug 19 17:53:37 
2016(r304483)
+++ head/sys/compat/cloudabi64/cloudabi64_sysent.c  Fri Aug 19 17:54:06 
2016(r304484)
@@ -3,7 +3,7 @@
  *
  * DO NOT EDIT-- this file is automatically generated.
  * $FreeBSD$
- * created from FreeBSD
+ * created from FreeBSD: head/sys/contrib/cloudabi/syscalls64.master 304483 
2016-08-19 17:53:37Z ed 
  */
 
 #include 
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304483 - head/sys/contrib/cloudabi

[Ed Schouten]

Author: ed
Date: Fri Aug 19 17:53:37 2016
New Revision: 304483
URL: https://svnweb.freebsd.org/changeset/base/304483

Log:
  Use the proper value for svn:keywords.
  
  Pointy hat to:me

Modified:
Directory Properties:
  head/sys/contrib/cloudabi/syscalls64.master   (props changed)
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304482 - head/sys/dev/cxgbe/common

[John Baldwin]

Author: jhb
Date: Fri Aug 19 17:52:48 2016
New Revision: 304482
URL: https://svnweb.freebsd.org/changeset/base/304482

Log:
  Adjust t4_port_init() to work with VF devices.
  
  Specifically, the FW_PORT_CMD may or may not work for a VF (the PF
  driver can choose whether or not to permit access to this command),
  so don't attempt to fetch port information on a VF if permission is
  denied by the PF.
  
  Reviewed by:  np
  Sponsored by: Chelsio Communications
  Differential Revision:https://reviews.freebsd.org/D7511

Modified:
  head/sys/dev/cxgbe/common/t4_hw.c

Modified: head/sys/dev/cxgbe/common/t4_hw.c
==
--- head/sys/dev/cxgbe/common/t4_hw.c   Fri Aug 19 17:51:52 2016
(r304481)
+++ head/sys/dev/cxgbe/common/t4_hw.c   Fri Aug 19 17:52:48 2016
(r304482)
@@ -7938,15 +7938,26 @@ int t4_port_init(struct adapter *adap, i
} while ((adap->params.portvec & (1 << j)) == 0);
}
 
-   c.op_to_portid = htonl(V_FW_CMD_OP(FW_PORT_CMD) |
-  F_FW_CMD_REQUEST | F_FW_CMD_READ |
-  V_FW_PORT_CMD_PORTID(j));
-   c.action_to_len16 = htonl(
-   V_FW_PORT_CMD_ACTION(FW_PORT_ACTION_GET_PORT_INFO) |
-   FW_LEN16(c));
-   ret = t4_wr_mbox(adap, mbox, , sizeof(c), );
-   if (ret)
-   return ret;
+   if (!(adap->flags & IS_VF) ||
+   adap->params.vfres.r_caps & FW_CMD_CAP_PORT) {
+   c.op_to_portid = htonl(V_FW_CMD_OP(FW_PORT_CMD) |
+  F_FW_CMD_REQUEST | F_FW_CMD_READ |
+  V_FW_PORT_CMD_PORTID(j));
+   c.action_to_len16 = htonl(
+   V_FW_PORT_CMD_ACTION(FW_PORT_ACTION_GET_PORT_INFO) |
+   FW_LEN16(c));
+   ret = t4_wr_mbox(adap, mbox, , sizeof(c), );
+   if (ret)
+   return ret;
+
+   ret = be32_to_cpu(c.u.info.lstatus_to_modtype);
+   p->mdio_addr = (ret & F_FW_PORT_CMD_MDIOCAP) ?
+   G_FW_PORT_CMD_MDIOADDR(ret) : -1;
+   p->port_type = G_FW_PORT_CMD_PTYPE(ret);
+   p->mod_type = G_FW_PORT_CMD_MODTYPE(ret);
+
+   init_link_config(>link_cfg, be16_to_cpu(c.u.info.pcap));
+   }
 
ret = t4_alloc_vi(adap, mbox, j, pf, vf, 1, addr, _size);
if (ret < 0)
@@ -7959,14 +7970,6 @@ int t4_port_init(struct adapter *adap, i
p->vi[0].rss_size = rss_size;
t4_os_set_hw_addr(adap, p->port_id, addr);
 
-   ret = be32_to_cpu(c.u.info.lstatus_to_modtype);
-   p->mdio_addr = (ret & F_FW_PORT_CMD_MDIOCAP) ?
-   G_FW_PORT_CMD_MDIOADDR(ret) : -1;
-   p->port_type = G_FW_PORT_CMD_PTYPE(ret);
-   p->mod_type = G_FW_PORT_CMD_MODTYPE(ret);
-
-   init_link_config(>link_cfg, be16_to_cpu(c.u.info.pcap));
-
param = V_FW_PARAMS_MNEM(FW_PARAMS_MNEM_DEV) |
V_FW_PARAMS_PARAM_X(FW_PARAMS_PARAM_DEV_RSSINFO) |
V_FW_PARAMS_PARAM_YZ(p->vi[0].viid);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304481 - head/sys/contrib/cloudabi

[Ed Schouten]

Author: ed
Date: Fri Aug 19 17:51:52 2016
New Revision: 304481
URL: https://svnweb.freebsd.org/changeset/base/304481

Log:
  Add missing SVN keywords keyword.

Modified:
Directory Properties:
  head/sys/contrib/cloudabi/syscalls64.master   (props changed)
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304479 - head/sys/dev/cxgbe/common

[John Baldwin]

Author: jhb
Date: Fri Aug 19 17:49:49 2016
New Revision: 304479
URL: https://svnweb.freebsd.org/changeset/base/304479

Log:
  Add structures for VF-specific adapter parameters.
  
  While here, mark which parameters are PF-specific and which are
  VF-specific.
  
  Sponsored by: Chelsio Communications
  Differential Revision:https://reviews.freebsd.org/D7508

Modified:
  head/sys/dev/cxgbe/common/common.h

Modified: head/sys/dev/cxgbe/common/common.h
==
--- head/sys/dev/cxgbe/common/common.h  Fri Aug 19 17:49:35 2016
(r304478)
+++ head/sys/dev/cxgbe/common/common.h  Fri Aug 19 17:49:49 2016
(r304479)
@@ -280,12 +280,52 @@ struct chip_params {
u16 mps_tcam_size;
 };
 
+/* VF-only parameters. */
+
+/*
+ * Global Receive Side Scaling (RSS) parameters in host-native format.
+ */
+struct rss_params {
+   unsigned int mode;  /* RSS mode */
+   union {
+   struct {
+   u_int synmapen:1;   /* SYN Map Enable */
+   u_int syn4tupenipv6:1;  /* enable hashing 4-tuple IPv6 SYNs */
+   u_int syn2tupenipv6:1;  /* enable hashing 2-tuple IPv6 SYNs */
+   u_int syn4tupenipv4:1;  /* enable hashing 4-tuple IPv4 SYNs */
+   u_int syn2tupenipv4:1;  /* enable hashing 2-tuple IPv4 SYNs */
+   u_int ofdmapen:1;   /* Offload Map Enable */
+   u_int tnlmapen:1;   /* Tunnel Map Enable */
+   u_int tnlalllookup:1;   /* Tunnel All Lookup */
+   u_int hashtoeplitz:1;   /* use Toeplitz hash */
+   } basicvirtual;
+   } u;
+};
+
+/*
+ * Maximum resources provisioned for a PCI VF.
+ */
+struct vf_resources {
+   unsigned int nvi;   /* N virtual interfaces */
+   unsigned int neq;   /* N egress Qs */
+   unsigned int nethctrl;  /* N egress ETH or CTRL Qs */
+   unsigned int niqflint;  /* N ingress Qs/w free list(s) & intr */
+   unsigned int niq;   /* N ingress Qs */
+   unsigned int tc;/* PCI-E traffic class */
+   unsigned int pmask; /* port access rights mask */
+   unsigned int nexactf;   /* N exact MPS filters */
+   unsigned int r_caps;/* read capabilities */
+   unsigned int wx_caps;   /* write/execute capabilities */
+};
+
 struct adapter_params {
struct sge_params sge;
-   struct tp_params  tp;
+   struct tp_params  tp;   /* PF-only */
struct vpd_params vpd;
struct pci_params pci;
-   struct devlog_params devlog;
+   struct devlog_params devlog;/* PF-only */
+   struct rss_params rss;  /* VF-only */
+   struct vf_resources vfres;  /* VF-only */
 
unsigned int sf_size; /* serial flash size in bytes */
unsigned int sf_nsec; /* # of flash sectors */
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304478 - in head: sys/compat/cloudabi sys/compat/cloudabi64 sys/contrib/cloudabi usr.bin/truss

[Ed Schouten]

Author: ed
Date: Fri Aug 19 17:49:35 2016
New Revision: 304478
URL: https://svnweb.freebsd.org/changeset/base/304478

Log:
  Import the new automatically generated system call table for CloudABI.
  
  Now that we've switched over to using the vDSO on CloudABI, it becomes a
  lot easier for us to phase out old features. System call numbering is no
  longer something that's part of the ABI. It's fully based on names. As
  long as the numbering used by the kernel and the vDSO is consistent
  (which it always is), it's all right.
  
  Let's put this to the test by removing a system call (thread_tcb_set())
  that's already unused for quite some time now, but was only left intact
  to serve as a placeholder. Sync in the new system call table that uses
  alphabetic sorting of system calls.
  
  Obtained from:https://github.com/NuxiNL/cloudabi

Added:
  head/sys/contrib/cloudabi/syscalls64.master
Deleted:
  head/sys/contrib/cloudabi/syscalls.master
Modified:
  head/sys/compat/cloudabi/cloudabi_thread.c
  head/sys/compat/cloudabi64/Makefile
  head/sys/compat/cloudabi64/cloudabi64_proto.h
  head/sys/compat/cloudabi64/cloudabi64_syscall.h
  head/sys/compat/cloudabi64/cloudabi64_syscalls.c
  head/sys/compat/cloudabi64/cloudabi64_sysent.c
  head/sys/compat/cloudabi64/cloudabi64_systrace_args.c
  head/sys/contrib/cloudabi/cloudabi64_types.h
  head/sys/contrib/cloudabi/cloudabi_types.h
  head/sys/contrib/cloudabi/cloudabi_types_common.h
  head/sys/contrib/cloudabi/cloudabi_vdso_aarch64.c
  head/sys/contrib/cloudabi/cloudabi_vdso_x86_64.c
  head/usr.bin/truss/syscalls.c

Modified: head/sys/compat/cloudabi/cloudabi_thread.c
==
--- head/sys/compat/cloudabi/cloudabi_thread.c  Fri Aug 19 17:48:47 2016
(r304477)
+++ head/sys/compat/cloudabi/cloudabi_thread.c  Fri Aug 19 17:49:35 2016
(r304478)
@@ -60,14 +60,6 @@ cloudabi_sys_thread_exit(struct thread *
 }
 
 int
-cloudabi_sys_thread_tcb_set(struct thread *td,
-struct cloudabi_sys_thread_tcb_set_args *uap)
-{
-
-   return (cpu_set_user_tls(td, uap->tcb));
-}
-
-int
 cloudabi_sys_thread_yield(struct thread *td,
 struct cloudabi_sys_thread_yield_args *uap)
 {

Modified: head/sys/compat/cloudabi64/Makefile
==
--- head/sys/compat/cloudabi64/Makefile Fri Aug 19 17:48:47 2016
(r304477)
+++ head/sys/compat/cloudabi64/Makefile Fri Aug 19 17:49:35 2016
(r304478)
@@ -8,7 +8,7 @@ sysent: cloudabi64_sysent.c cloudabi64_s
 
 cloudabi64_sysent.c cloudabi64_syscall.h cloudabi64_proto.h \
 cloudabi64_syscalls.c cloudabi64_systrace_args.c: \
-../../kern/makesyscalls.sh ../../contrib/cloudabi/syscalls.master \
+../../kern/makesyscalls.sh ../../contrib/cloudabi/syscalls64.master \
 syscalls.conf
-   sh ../../kern/makesyscalls.sh ../../contrib/cloudabi/syscalls.master \
+   sh ../../kern/makesyscalls.sh ../../contrib/cloudabi/syscalls64.master \
syscalls.conf

Modified: head/sys/compat/cloudabi64/cloudabi64_proto.h
==
--- head/sys/compat/cloudabi64/cloudabi64_proto.h   Fri Aug 19 17:48:47 
2016(r304477)
+++ head/sys/compat/cloudabi64/cloudabi64_proto.h   Fri Aug 19 17:49:35 
2016(r304478)
@@ -3,7 +3,7 @@
  *
  * DO NOT EDIT-- this file is automatically generated.
  * $FreeBSD$
- * created from FreeBSD: head/sys/contrib/cloudabi/syscalls.master 297468 
2016-03-31 18:50:06Z ed 
+ * created from FreeBSD
  */
 
 #ifndef _CLOUDABI64_SYSPROTO_H_
@@ -236,6 +236,14 @@ struct cloudabi64_sys_poll_args {
char out_l_[PADL_(cloudabi64_event_t *)]; cloudabi64_event_t * out; 
char out_r_[PADR_(cloudabi64_event_t *)];
char nsubscriptions_l_[PADL_(size_t)]; size_t nsubscriptions; char 
nsubscriptions_r_[PADR_(size_t)];
 };
+struct cloudabi64_sys_poll_fd_args {
+   char fd_l_[PADL_(cloudabi_fd_t)]; cloudabi_fd_t fd; char 
fd_r_[PADR_(cloudabi_fd_t)];
+   char in_l_[PADL_(const cloudabi64_subscription_t *)]; const 
cloudabi64_subscription_t * in; char in_r_[PADR_(const 
cloudabi64_subscription_t *)];
+   char nin_l_[PADL_(size_t)]; size_t nin; char nin_r_[PADR_(size_t)];
+   char out_l_[PADL_(cloudabi64_event_t *)]; cloudabi64_event_t * out; 
char out_r_[PADR_(cloudabi64_event_t *)];
+   char nout_l_[PADL_(size_t)]; size_t nout; char nout_r_[PADR_(size_t)];
+   char timeout_l_[PADL_(const cloudabi64_subscription_t *)]; const 
cloudabi64_subscription_t * timeout; char timeout_r_[PADR_(const 
cloudabi64_subscription_t *)];
+};
 struct cloudabi_sys_proc_exec_args {
char fd_l_[PADL_(cloudabi_fd_t)]; cloudabi_fd_t fd; char 
fd_r_[PADR_(cloudabi_fd_t)];
char data_l_[PADL_(const void *)]; const void * data; char 
data_r_[PADR_(const void *)];
@@ -302,20 +310,9 @@ struct cloudabi_sys_thread_exit_args {
char 

svn commit: r304477 - releng/11.0/usr.bin/truss

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 17:48:47 2016
New Revision: 304477
URL: https://svnweb.freebsd.org/changeset/base/304477

Log:
  MFS r304143:
  
MFC r303934,r303937,r303942:
  
  r303934:
Support rmdir(2).
  r303937:
Use proper argument length for rmdir(2) for r303934.
  r303942:
Fix sorting in r303934.
  
  Approved by:  re (gjb)

Modified:
  releng/11.0/usr.bin/truss/syscalls.c
Directory Properties:
  releng/11.0/   (props changed)

Modified: releng/11.0/usr.bin/truss/syscalls.c
==
--- releng/11.0/usr.bin/truss/syscalls.cFri Aug 19 17:37:32 2016
(r304476)
+++ releng/11.0/usr.bin/truss/syscalls.cFri Aug 19 17:48:47 2016
(r304477)
@@ -281,6 +281,8 @@ static struct syscall decoded_syscalls[]
  .args = { { Atfd, 0 }, { Name, 1 }, { Atfd, 2 }, { Name, 3 } } },
{ .name = "rfork", .ret_type = 1, .nargs = 1,
  .args = { { Rforkflags, 0 } } },
+   { .name = "rmdir", .ret_type = 1, .nargs = 1,
+ .args = { { Name, 0 } } },
{ .name = "select", .ret_type = 1, .nargs = 5,
  .args = { { Int, 0 }, { Fd_set, 1 }, { Fd_set, 2 }, { Fd_set, 3 },
{ Timeval, 4 } } },
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304476 - head/lib/libc/sys

[John Baldwin]

Author: jhb
Date: Fri Aug 19 17:37:32 2016
New Revision: 304476
URL: https://svnweb.freebsd.org/changeset/base/304476

Log:
  Fix various nits in the aio operation manpages.
  
  - Avoid double use of "request" in a single sentence.  Instead, describe
aio_sigevent as being used to request notification of the associated
operation's completion.  This matches the language used to describe
aio_sigevent in aio(4).
  - Simplify the prohibition on modifying buffers while requests are in
flight.
  - Fix case mismatch.
  - Drop note about not using stack variables. C programmers should be able
to figure out if a stack variable is safe based on the later warning
about the life cycle requirements of control blocks.
  - Remove prohibition on modifying the I/O buffer for aio_fsync() since
it does not use an I/O buffer.  For aio_mlock(), prohibit modifications
to the mapping (e.g. due to mprotect, munmap, mmap, etc.) but do not
prohibit modifications to the memory backing the buffer (stores into
the pages backing the buffer).
  
  Requested by: wblock (1,2), kib (4)
  Reviewed by:  kib, rpokala, wblock
  MFC after:3 days
  Sponsored by: Chelsio Communications
  Differential Revision:https://reviews.freebsd.org/D7462

Modified:
  head/lib/libc/sys/aio_fsync.2
  head/lib/libc/sys/aio_mlock.2
  head/lib/libc/sys/aio_read.2
  head/lib/libc/sys/aio_write.2

Modified: head/lib/libc/sys/aio_fsync.2
==
--- head/lib/libc/sys/aio_fsync.2   Fri Aug 19 17:03:14 2016
(r304475)
+++ head/lib/libc/sys/aio_fsync.2   Fri Aug 19 17:37:32 2016
(r304476)
@@ -24,7 +24,7 @@
 .\"
 .\" $FreeBSD$
 .\"
-.Dd July 21, 2016
+.Dd August 19, 2016
 .Dt AIO_FSYNC 2
 .Os
 .Sh NAME
@@ -74,16 +74,14 @@ the call returns without having enqueued
 .Pp
 The
 .Fa iocb->aio_sigevent
-structure can be used to request notification of the request's
+structure can be used to request notification of the operation's
 completion as described in
 .Xr aio 4 .
 .Sh RESTRICTIONS
-The asynchronous I/O Control Block structure pointed to by
+The Asynchronous I/O Control Block structure pointed to by
 .Fa iocb
 must remain valid until the
 operation has completed.
-For this reason, use of auto (stack) variables
-for these objects is discouraged.
 .Pp
 The asynchronous I/O control buffer
 .Fa iocb
@@ -91,9 +89,8 @@ should be zeroed before the
 .Fn aio_fsync
 call to avoid passing bogus context information to the kernel.
 .Pp
-Modifications of the Asynchronous I/O Control Block structure or the
-buffer contents after the request has been enqueued, but before the
-request has completed, are not allowed.
+Modification of the Asynchronous I/O Control Block structure is not allowed
+while the request is queued.
 .Sh RETURN VALUES
 .Rv -std aio_fsync
 .Sh ERRORS

Modified: head/lib/libc/sys/aio_mlock.2
==
--- head/lib/libc/sys/aio_mlock.2   Fri Aug 19 17:03:14 2016
(r304475)
+++ head/lib/libc/sys/aio_mlock.2   Fri Aug 19 17:37:32 2016
(r304476)
@@ -24,7 +24,7 @@
 .\"
 .\" $FreeBSD$
 .\"
-.Dd July 21, 2016
+.Dd August 19, 2016
 .Dt AIO_MLOCK 2
 .Os
 .Sh NAME
@@ -67,7 +67,7 @@ then the call returns without having enq
 .Pp
 The
 .Fa iocb->aio_sigevent
-structure can be used to request notification of the request's
+structure can be used to request notification of the operation's
 completion as described in
 .Xr aio 4 .
 .Sh RESTRICTIONS
@@ -77,8 +77,6 @@ and the buffer that the
 .Fa iocb->aio_buf
 member of that structure references must remain valid until the
 operation has completed.
-For this reason, use of auto (stack) variables
-for these objects is discouraged.
 .Pp
 The asynchronous I/O control buffer
 .Fa iocb
@@ -87,8 +85,8 @@ should be zeroed before the
 call to avoid passing bogus context information to the kernel.
 .Pp
 Modifications of the Asynchronous I/O Control Block structure or the
-buffer contents after the request has been enqueued, but before the
-request has completed, are not allowed.
+memory mapping described by the virtual address range are not allowed
+while the request is queued.
 .Sh RETURN VALUES
 .Rv -std aio_mlock
 .Sh ERRORS

Modified: head/lib/libc/sys/aio_read.2
==
--- head/lib/libc/sys/aio_read.2Fri Aug 19 17:03:14 2016
(r304475)
+++ head/lib/libc/sys/aio_read.2Fri Aug 19 17:37:32 2016
(r304476)
@@ -24,7 +24,7 @@
 .\"
 .\" $FreeBSD$
 .\"
-.Dd July 21, 2016
+.Dd August 19, 2016
 .Dt AIO_READ 2
 .Os
 .Sh NAME
@@ -82,7 +82,7 @@ not be referenced after the request is e
 .Pp
 The
 .Fa iocb->aio_sigevent
-structure can be used to request notification of the request's
+structure can be used to request notification of the operation's
 completion as described in
 .Xr aio 4 .
 .Sh RESTRICTIONS
@@ -92,8 +92,6 

svn commit: r304475 - stable/11

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 17:03:14 2016
New Revision: 304475
URL: https://svnweb.freebsd.org/changeset/base/304475

Log:
  MFC r304217:
  
Trim unneeded bootstrap after r301470 made 9.1 the minimum supported 
release.

Modified:
  stable/11/Makefile.inc1
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/Makefile.inc1
==
--- stable/11/Makefile.inc1 Fri Aug 19 17:02:14 2016(r304474)
+++ stable/11/Makefile.inc1 Fri Aug 19 17:03:14 2016(r304475)
@@ -1579,10 +1579,6 @@ _groff=  gnu/usr.bin/groff \
 _vtfontcvt=usr.bin/vtfontcvt
 .endif
 
-.if ${BOOTSTRAPPING} < 92
-_sed=  usr.bin/sed
-.endif
-
 .if ${BOOTSTRAPPING} < 133
 _libopenbsd=   lib/libopenbsd
 _m4=   usr.bin/m4
@@ -1621,10 +1617,6 @@ _crunchide=  usr.sbin/crunch/crunchide
 _crunchgen=usr.sbin/crunch/crunchgen
 .endif
 
-.if ${BOOTSTRAPPING} >= 900040 && ${BOOTSTRAPPING} < 900041
-_awk=  usr.bin/awk
-.endif
-
 # r296926 -P keymap search path, MFC to stable/10 in r298297
 .if ${BOOTSTRAPPING} < 1003501 || \
(${BOOTSTRAPPING} >= 110 && ${BOOTSTRAPPING} < 1100103)
@@ -1692,7 +1684,6 @@ bootstrap-tools: .PHONY
 ${_gperf} \
 ${_groff} \
 ${_dtc} \
-${_awk} \
 ${_cat} \
 ${_dd} \
 ${_kbdcontrol} \
@@ -1700,7 +1691,6 @@ bootstrap-tools: .PHONY
 ${_libopenbsd} \
 ${_makewhatis} \
 usr.bin/rpcgen \
-${_sed} \
 ${_yacc} \
 ${_m4} \
 ${_lex} \
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304474 - stable/11/sys/dev/filemon

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 17:02:14 2016
New Revision: 304474
URL: https://svnweb.freebsd.org/changeset/base/304474

Log:
  MFC r304008:
  
Avoid taking PROC_LOCK in syscalls if not being traced.

Modified:
  stable/11/sys/dev/filemon/filemon.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/sys/dev/filemon/filemon.c
==
--- stable/11/sys/dev/filemon/filemon.c Fri Aug 19 17:02:05 2016
(r304473)
+++ stable/11/sys/dev/filemon/filemon.c Fri Aug 19 17:02:14 2016
(r304474)
@@ -137,6 +137,8 @@ filemon_proc_get(struct proc *p)
 {
struct filemon *filemon;
 
+   if (p->p_filemon == NULL)
+   return (NULL);
PROC_LOCK(p);
filemon = filemon_acquire(p->p_filemon);
PROC_UNLOCK(p);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304473 - stable/10/sys/dev/filemon

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 17:02:05 2016
New Revision: 304473
URL: https://svnweb.freebsd.org/changeset/base/304473

Log:
  MFC r304008:
  
Avoid taking PROC_LOCK in syscalls if not being traced.

Modified:
  stable/10/sys/dev/filemon/filemon.c
Directory Properties:
  stable/10/   (props changed)

Modified: stable/10/sys/dev/filemon/filemon.c
==
--- stable/10/sys/dev/filemon/filemon.c Fri Aug 19 17:01:13 2016
(r304472)
+++ stable/10/sys/dev/filemon/filemon.c Fri Aug 19 17:02:05 2016
(r304473)
@@ -137,6 +137,8 @@ filemon_proc_get(struct proc *p)
 {
struct filemon *filemon;
 
+   if (p->p_filemon == NULL)
+   return (NULL);
PROC_LOCK(p);
filemon = filemon_acquire(p->p_filemon);
PROC_UNLOCK(p);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304472 - stable/11

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 17:01:13 2016
New Revision: 304472
URL: https://svnweb.freebsd.org/changeset/base/304472

Log:
  MFC r304006:
  
Avoid showing the bootstrap make command for check-old, etc.

Modified:
  stable/11/Makefile
  stable/11/Makefile.inc1
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/Makefile
==
--- stable/11/Makefile  Fri Aug 19 16:59:16 2016(r304471)
+++ stable/11/Makefile  Fri Aug 19 17:01:13 2016(r304472)
@@ -209,7 +209,8 @@ SUB_MAKE= `test -x ${MYMAKE} && echo ${M
 SUB_MAKE= ${MAKE} -m ${.CURDIR}/share/mk
 .endif
 
-_MAKE= PATH=${PATH} ${SUB_MAKE} -f Makefile.inc1 TARGET=${_TARGET} 
TARGET_ARCH=${_TARGET_ARCH}
+_MAKE= PATH=${PATH} MAKE_CMD=${MAKE} ${SUB_MAKE} -f Makefile.inc1 \
+   TARGET=${_TARGET} TARGET_ARCH=${_TARGET_ARCH}
 
 # Only allow meta mode for the whitelisted targets.  See META_TGT_WHITELIST
 # above.

Modified: stable/11/Makefile.inc1
==
--- stable/11/Makefile.inc1 Fri Aug 19 16:59:16 2016(r304471)
+++ stable/11/Makefile.inc1 Fri Aug 19 17:01:13 2016(r304472)
@@ -2372,11 +2372,11 @@ check-old-dirs: .PHONY
done
 
 delete-old: delete-old-files delete-old-dirs .PHONY
-   @echo "To remove old libraries run '${MAKE} delete-old-libs'."
+   @echo "To remove old libraries run '${MAKE_CMD} delete-old-libs'."
 
 check-old: check-old-files check-old-libs check-old-dirs .PHONY
-   @echo "To remove old files and directories run '${MAKE} delete-old'."
-   @echo "To remove old libraries run '${MAKE} delete-old-libs'."
+   @echo "To remove old files and directories run '${MAKE_CMD} 
delete-old'."
+   @echo "To remove old libraries run '${MAKE_CMD} delete-old-libs'."
 
 .endif
 
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304470 - stable/11/sys/conf

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 16:59:01 2016
New Revision: 304470
URL: https://svnweb.freebsd.org/changeset/base/304470

Log:
  MFC r304005:
  
PORTS_MODULES: Don't leak in CC/CXX/CPP.

Modified:
  stable/11/sys/conf/kern.post.mk
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/sys/conf/kern.post.mk
==
--- stable/11/sys/conf/kern.post.mk Fri Aug 19 16:56:52 2016
(r304469)
+++ stable/11/sys/conf/kern.post.mk Fri Aug 19 16:59:01 2016
(r304470)
@@ -65,6 +65,10 @@ OSRELDATE!=  awk '/^\#define[[:space:]]*_
 # Keep the related ports builds in the obj directory so that they are only 
rebuilt once per kernel build
 WRKDIRPREFIX?= ${MAKEOBJDIRPREFIX}${SRC_BASE}/sys/${KERNCONF}
 PORTSMODULESENV=\
+   env \
+   -u CC \
+   -u CXX \
+   -u CPP \
PATH=${PATH}:${LOCALBASE}/bin:${LOCALBASE}/sbin \
SRC_BASE=${SRC_BASE} \
OSVERSION=${OSRELDATE} \
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304471 - stable/10/usr.bin/xinstall

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 16:59:16 2016
New Revision: 304471
URL: https://svnweb.freebsd.org/changeset/base/304471

Log:
  MFC r303929,r303930,r303931,r303932,r303933:
  
r303929:
  Fix -S with -b not atomically updating the destination file.
r303930:
  Support -v for -l.
r303931:
  Fix -S with -l not being atomic.
r303932:
  Fix -b failure not restoring flags on the destination file.
r303933:
  Squelch a false-positive Clang static analyzer warning.

Modified:
  stable/10/usr.bin/xinstall/xinstall.c
Directory Properties:
  stable/10/   (props changed)

Modified: stable/10/usr.bin/xinstall/xinstall.c
==
--- stable/10/usr.bin/xinstall/xinstall.c   Fri Aug 19 16:59:01 2016
(r304470)
+++ stable/10/usr.bin/xinstall/xinstall.c   Fri Aug 19 16:59:16 2016
(r304471)
@@ -151,6 +151,7 @@ main(int argc, char *argv[])
char *p;
const char *to_name;
 
+   fset = 0;
iflags = 0;
group = owner = NULL;
while ((ch = getopt(argc, argv, "B:bCcD:df:g:h:l:M:m:N:o:pSsT:Uv")) !=
@@ -535,7 +536,9 @@ do_link(const char *from_name, const cha
if (target_sb->st_flags & NOCHANGEBITS)
(void)chflags(to_name, target_sb->st_flags &
 ~NOCHANGEBITS);
-   unlink(to_name);
+   if (verbose)
+   printf("install: link %s -> %s\n",
+   from_name, to_name);
ret = rename(tmpl, to_name);
/*
 * If rename has posix semantics, then the temporary
@@ -545,8 +548,12 @@ do_link(const char *from_name, const cha
(void)unlink(tmpl);
}
return (ret);
-   } else
+   } else {
+   if (verbose)
+   printf("install: link %s -> %s\n",
+   from_name, to_name);
return (link(from_name, to_name));
+   }
 }
 
 /*
@@ -575,14 +582,18 @@ do_symlink(const char *from_name, const 
if (target_sb->st_flags & NOCHANGEBITS)
(void)chflags(to_name, target_sb->st_flags &
 ~NOCHANGEBITS);
-   unlink(to_name);
-
+   if (verbose)
+   printf("install: symlink %s -> %s\n",
+   from_name, to_name);
if (rename(tmpl, to_name) == -1) {
/* Remove temporary link before exiting. */
(void)unlink(tmpl);
err(EX_OSERR, "%s: rename", to_name);
}
} else {
+   if (verbose)
+   printf("install: symlink %s -> %s\n",
+   from_name, to_name);
if (symlink(from_name, to_name) == -1)
err(EX_OSERR, "symlink %s -> %s", from_name, to_name);
}
@@ -882,11 +893,21 @@ install(const char *from_name, const cha
}
if (verbose)
(void)printf("install: %s -> %s\n", to_name, 
backup);
-   if (rename(to_name, backup) < 0) {
+   if (unlink(backup) < 0 && errno != ENOENT) {
+   serrno = errno;
+   if (to_sb.st_flags & NOCHANGEBITS)
+   (void)chflags(to_name, to_sb.st_flags);
+   unlink(tempfile);
+   errno = serrno;
+   err(EX_OSERR, "unlink: %s", backup);
+   }
+   if (link(to_name, backup) < 0) {
serrno = errno;
unlink(tempfile);
+   if (to_sb.st_flags & NOCHANGEBITS)
+   (void)chflags(to_name, to_sb.st_flags);
errno = serrno;
-   err(EX_OSERR, "rename: %s to %s", to_name,
+   err(EX_OSERR, "link: %s to %s", to_name,
 backup);
}
}
@@ -1109,16 +1130,26 @@ create_newfile(const char *path, int tar
 
if (dobackup) {
if ((size_t)snprintf(backup, MAXPATHLEN, "%s%s",
-   path, suffix) != strlen(path) + strlen(suffix))
+   path, suffix) != strlen(path) + strlen(suffix)) {
+   saved_errno = errno;
+   if (sbp->st_flags & NOCHANGEBITS)
+   (void)chflags(path, sbp->st_flags);
+   errno = 

svn commit: r304469 - stable/11/usr.bin/xinstall

[Bryan Drewery]

Author: bdrewery
Date: Fri Aug 19 16:56:52 2016
New Revision: 304469
URL: https://svnweb.freebsd.org/changeset/base/304469

Log:
  MFC r303929,r303930,r303931,r303932,r303933:
  
r303929:
  Fix -S with -b not atomically updating the destination file.
r303930:
  Support -v for -l.
r303931:
  Fix -S with -l not being atomic.
r303932:
  Fix -b failure not restoring flags on the destination file.
r303933:
  Squelch a false-positive Clang static analyzer warning.

Modified:
  stable/11/usr.bin/xinstall/xinstall.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/usr.bin/xinstall/xinstall.c
==
--- stable/11/usr.bin/xinstall/xinstall.c   Fri Aug 19 16:45:21 2016
(r304468)
+++ stable/11/usr.bin/xinstall/xinstall.c   Fri Aug 19 16:56:52 2016
(r304469)
@@ -149,6 +149,7 @@ main(int argc, char *argv[])
char *p;
const char *to_name;
 
+   fset = 0;
iflags = 0;
group = owner = NULL;
while ((ch = getopt(argc, argv, "B:bCcD:df:g:h:l:M:m:N:o:pSsT:Uv")) !=
@@ -533,7 +534,9 @@ do_link(const char *from_name, const cha
if (target_sb->st_flags & NOCHANGEBITS)
(void)chflags(to_name, target_sb->st_flags &
 ~NOCHANGEBITS);
-   unlink(to_name);
+   if (verbose)
+   printf("install: link %s -> %s\n",
+   from_name, to_name);
ret = rename(tmpl, to_name);
/*
 * If rename has posix semantics, then the temporary
@@ -543,8 +546,12 @@ do_link(const char *from_name, const cha
(void)unlink(tmpl);
}
return (ret);
-   } else
+   } else {
+   if (verbose)
+   printf("install: link %s -> %s\n",
+   from_name, to_name);
return (link(from_name, to_name));
+   }
 }
 
 /*
@@ -573,14 +580,18 @@ do_symlink(const char *from_name, const 
if (target_sb->st_flags & NOCHANGEBITS)
(void)chflags(to_name, target_sb->st_flags &
 ~NOCHANGEBITS);
-   unlink(to_name);
-
+   if (verbose)
+   printf("install: symlink %s -> %s\n",
+   from_name, to_name);
if (rename(tmpl, to_name) == -1) {
/* Remove temporary link before exiting. */
(void)unlink(tmpl);
err(EX_OSERR, "%s: rename", to_name);
}
} else {
+   if (verbose)
+   printf("install: symlink %s -> %s\n",
+   from_name, to_name);
if (symlink(from_name, to_name) == -1)
err(EX_OSERR, "symlink %s -> %s", from_name, to_name);
}
@@ -886,11 +897,21 @@ install(const char *from_name, const cha
}
if (verbose)
(void)printf("install: %s -> %s\n", to_name, 
backup);
-   if (rename(to_name, backup) < 0) {
+   if (unlink(backup) < 0 && errno != ENOENT) {
+   serrno = errno;
+   if (to_sb.st_flags & NOCHANGEBITS)
+   (void)chflags(to_name, to_sb.st_flags);
+   unlink(tempfile);
+   errno = serrno;
+   err(EX_OSERR, "unlink: %s", backup);
+   }
+   if (link(to_name, backup) < 0) {
serrno = errno;
unlink(tempfile);
+   if (to_sb.st_flags & NOCHANGEBITS)
+   (void)chflags(to_name, to_sb.st_flags);
errno = serrno;
-   err(EX_OSERR, "rename: %s to %s", to_name,
+   err(EX_OSERR, "link: %s to %s", to_name,
 backup);
}
}
@@ -,16 +1132,26 @@ create_newfile(const char *path, int tar
 
if (dobackup) {
if ((size_t)snprintf(backup, MAXPATHLEN, "%s%s",
-   path, suffix) != strlen(path) + strlen(suffix))
+   path, suffix) != strlen(path) + strlen(suffix)) {
+   saved_errno = errno;
+   if (sbp->st_flags & NOCHANGEBITS)
+   (void)chflags(path, sbp->st_flags);
+   errno = 

Re: svn commit: r304142 - head/usr.sbin/bsdinstall/partedit

[Nathan Whitehorn]

On 08/19/16 08:57, Allan Jude wrote:

On 2016-08-19 10:13, Warner Losh wrote:

On Fri, Aug 19, 2016 at 12:51 AM, Dag-Erling Smørgrav  wrote:

Warner Losh  writes:

Allan Jude  writes:

Which makes more sense:

A) If stripesize == 0, use some sane value like 4096

I don't like this.


B) Some other combination that uses the reported stripe size, unless it
is 0, in which case it uses 4096 (or some other value controlled by a
different new sysctl)

Don't like this so much.


C) create kern.geom.min_stripe_size with a default of 512, but users can
set 4096 if they use only 4k devices. (doesn't really solve the problem
for the installer)

Default it to 4k, and allow users to set it to 512. If the drive
reports < this value
report this value instead.

I don't like either option.  Option D (which I don't like either, but
which should at least work in most cases) is a sysctl that specifies a
minimum factor, and set the reported stripe size to the least common
multiple of that number and the actual stripe or sector size.  This is
what my bsdinstall patch does.  However, I think that pushing this down
to a layer where it will affect all applications is a terrible idea,
because we have no way of knowing what will break[*], and it can
seriously mislead users and hinder troubleshooting - especially if it is
enabled by default rather than only when necessary.

I took a look into the implications of doing a 4k stripesize 'automatically'
this morning. I found a few places in g_part where it would actively
hurt when coupled with gpart's insistence on aligning things. So I
now think it's a bad idea. This will make it harder for FreeBSD to
generate arbitrary disk layouts. And I'm not too sure about what
things like gstripe would report as a result and if this would actually
interfere if you had a large, but not power of two stripe size.


I don't think it's a good idea to enforce stripe alignment everywhere,
either.  It works for partitions because they are very large relative to
the stripe size, and at worst we will waste a few millionths of the disk
on inter-partition gaps, which should only occur between the partition
table and the boot partition, and possibly, if the stripe size is very
large, between the boot partition and the swap partition.  But forcing
filesystems to respect the stripe size will lead to no end of trouble,
because RAID volumes can have stripe sizes of 16 kB or more.  I think it
is important to align partitions during installation because of the huge
performance impact of misaligned partitions on AF disks, but despite
what Nathan claims, I never advocated applying the same logic
everywhere.

Yea, having poked at it for just a little while, I agree. The installer is the
right place to make sure we don't cross-thread the 4k sectors. Stripe size
means too many other things to have it be useful in that context.

Warner


Maybe instead we just change gpart to default to 4k alignment, but users
can always override with -a 512 or some other value?

Then the installer behaves the same as a user typing 'gpart', but we
don't mess with the entire geom layer?



ZFS also looks at this, so we would need it there, and there might be a 
few others. At the very least, gpart and the installer should have the 
same behavior.


Warner, could you elaborate on what you mean by "cross-threading"? Are 
you worried about nested partition tables in which the outer one is 
misaligned? Having the disk drivers report a 4K stripesize seems to have 
worked so far; I'm not sure how reporting it more often would cause 
problems. You wouldn't want all drivers, or all GEOM layers, to do this, 
of course. If the problem is that "stripe size" has an overloaded 
meaning, we could easily add another GEOM property.

-Nathan
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304142 - head/usr.sbin/bsdinstall/partedit

[Nathan Whitehorn]

On 08/18/16 22:33, Warner Losh wrote:

On Aug 18, 2016, at 11:21 PM, Nathan Whitehorn  wrote:



On 08/18/16 21:15, Warner Losh wrote:

On Thu, Aug 18, 2016 at 6:56 AM, Allan Jude  wrote:

On 08/18/16 05:50 AM, Dag-Erling Smørgrav wrote:

Nathan Whitehorn  writes:

OK. In which configurations? My Dell servers, for instance, don't do
this. How are they set up? What drivers are being used? Is this
something that affects passthrough disks, RAIDs, disk images?

Most LSI MegaRAID controllers don't have real passthrough, only JBOD.
You can query the drive with "camcontrol identify passX", but the
controller does not report a stripe size for the volume (mfidY).


The point is that *if the reported stripe size is wrong*, more things
than partition alignment in the installer will suffer for it.

It's not wrong, it's non-existent, and I'm getting really tired of
repeating myself.


Fixing the installer with a bandaid in the run-up to a release is
fine, but *we need to fix the underlying problem*.

We can't, because hardware sucks, and I'm getting really tired of
repeating myself.

DES


Which makes more sense:

A) If stripesize == 0, use some sane value like 4096

I don't like this.


B) Some other combination that uses the reported stripe size, unless it
is 0, in which case it uses 4096 (or some other value controlled by a
different new sysctl)

Don't like this so much.


C) create kern.geom.min_stripe_size with a default of 512, but users can
set 4096 if they use only 4k devices. (doesn't really solve the problem
for the installer)

Default it to 4k, and allow users to set it to 512. If the drive
reports < this value
report this value instead. You'll need to make this a tunable. Then the upper
layers wouldn't care. There's a small chance that some SD cards might be
reporting values that are too large. But I think it is confined to SD cards and
if I see too many more I'll do something specific in the SD driver.

Warner



That sounds good to me and I think can clean up a lot of code and potential 
foot-shooting. Who is planning to make the patch? I'm happy to do anything that 
would be helpful.

The patch is super-easy, but I need to get the concept validated and make sure 
that it does not have unintended side effects.

Warner

Sounds great. There is no urgency here -- we have a good solution for 
11.0 already -- so taking time to do it right sounds good. I believe 
stripesize is only consumed by disk formatting tools, so unintended side 
effects at least should be minimal.

-Nathan
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304142 - head/usr.sbin/bsdinstall/partedit

[Allan Jude]

On 2016-08-19 10:13, Warner Losh wrote:
> On Fri, Aug 19, 2016 at 12:51 AM, Dag-Erling Smørgrav  wrote:
>> Warner Losh  writes:
>>> Allan Jude  writes:
 Which makes more sense:

 A) If stripesize == 0, use some sane value like 4096
>>>
>>> I don't like this.
>>>
 B) Some other combination that uses the reported stripe size, unless it
 is 0, in which case it uses 4096 (or some other value controlled by a
 different new sysctl)
>>>
>>> Don't like this so much.
>>>
 C) create kern.geom.min_stripe_size with a default of 512, but users can
 set 4096 if they use only 4k devices. (doesn't really solve the problem
 for the installer)
>>>
>>> Default it to 4k, and allow users to set it to 512. If the drive
>>> reports < this value
>>> report this value instead.
>>
>> I don't like either option.  Option D (which I don't like either, but
>> which should at least work in most cases) is a sysctl that specifies a
>> minimum factor, and set the reported stripe size to the least common
>> multiple of that number and the actual stripe or sector size.  This is
>> what my bsdinstall patch does.  However, I think that pushing this down
>> to a layer where it will affect all applications is a terrible idea,
>> because we have no way of knowing what will break[*], and it can
>> seriously mislead users and hinder troubleshooting - especially if it is
>> enabled by default rather than only when necessary.
> 
> I took a look into the implications of doing a 4k stripesize 'automatically'
> this morning. I found a few places in g_part where it would actively
> hurt when coupled with gpart's insistence on aligning things. So I
> now think it's a bad idea. This will make it harder for FreeBSD to
> generate arbitrary disk layouts. And I'm not too sure about what
> things like gstripe would report as a result and if this would actually
> interfere if you had a large, but not power of two stripe size.
> 
>> I don't think it's a good idea to enforce stripe alignment everywhere,
>> either.  It works for partitions because they are very large relative to
>> the stripe size, and at worst we will waste a few millionths of the disk
>> on inter-partition gaps, which should only occur between the partition
>> table and the boot partition, and possibly, if the stripe size is very
>> large, between the boot partition and the swap partition.  But forcing
>> filesystems to respect the stripe size will lead to no end of trouble,
>> because RAID volumes can have stripe sizes of 16 kB or more.  I think it
>> is important to align partitions during installation because of the huge
>> performance impact of misaligned partitions on AF disks, but despite
>> what Nathan claims, I never advocated applying the same logic
>> everywhere.
> 
> Yea, having poked at it for just a little while, I agree. The installer is the
> right place to make sure we don't cross-thread the 4k sectors. Stripe size
> means too many other things to have it be useful in that context.
> 
> Warner
> 

Maybe instead we just change gpart to default to 4k alignment, but users
can always override with -a 512 or some other value?

Then the installer behaves the same as a user typing 'gpart', but we
don't mess with the entire geom layer?

-- 
Allan Jude



signature.asc
Description: OpenPGP digital signature

Re: svn commit: r304142 - head/usr.sbin/bsdinstall/partedit

[Allan Jude]

On 2016-08-19 00:15, Warner Losh wrote:
>> Which makes more sense:
>>
>> A) If stripesize == 0, use some sane value like 4096
> 
> I don't like this.
> 
>> B) Some other combination that uses the reported stripe size, unless it
>> is 0, in which case it uses 4096 (or some other value controlled by a
>> different new sysctl)
> 
> Don't like this so much.
> 
>> C) create kern.geom.min_stripe_size with a default of 512, but users can
>> set 4096 if they use only 4k devices. (doesn't really solve the problem
>> for the installer)
> 
> Default it to 4k, and allow users to set it to 512. If the drive
> reports < this value
> report this value instead. You'll need to make this a tunable. Then the upper
> layers wouldn't care. There's a small chance that some SD cards might be
> reporting values that are too large. But I think it is confined to SD cards 
> and
> if I see too many more I'll do something specific in the SD driver.
> 
> Warner
> 

I think I mentioned this earlier in the thread,, but I do have a USB
thumbstick that reports an 8mb stripe size.

But I agree, I think a sysctl where you set the minimum (default 4096,
but settable to 512), that is a tunable that can be overwritten in
loader.conf

What do we want for the logic as to what values it can be set to?

Any power of 2 greater than or equal to 512?

-- 
Allan Jude



signature.asc
Description: OpenPGP digital signature

Re: svn commit: r304187 - in head: . share/man/man4 sys/conf sys/dev/mcd sys/modules sys/modules/mcd

[Bruce Evans]

On Fri, 19 Aug 2016, Alexey Dokuchaev wrote:


On Fri, Aug 19, 2016 at 09:12:53PM +1000, Bruce Evans wrote:

I normally use my version of biosboot for boot2.  I improved its caching
just a couple of years ago.  It was using 9K buffer optimized for 1440K
floppies.  Now it uses a 32K buffer.  Booting a 5.5MB kernel takes a
fraction of a second.


That looks like change that everyone would benefit from.  Consider posting
a patch or committing it yourself. ;-)


Oops, it is actually loading that takes a fraction of a second.  Not
much different than with -current boot2 or old loader.  -current boot2
also works for me, but I managed to squeeze more of the features that
I want into biosboot (everything except ufs2).  Booting from the boot1
prompt takes 20-25 seconds here.


Of course I don't use modules, so not many seeks are needed.


Why not, they're convenient (apart from "kernel version mismatch" crap
that I keep hitting now and then)?  AFAIR they were slow to load at some
point but that was (not so) recently fixed.


Same reason that I don't use shared libraries if possible - they are
larger, slower and more difficult to debug.  For kernel development,
the version control problem is large.  Modules are good for avoiding
rebooting when developing something in a single module, but I usually
work on either small changes that panic often or system wide-changes
that need recompiling everything.  Both require rebooting a lot to
test, and I can rebuild a kernel and without modules and reboot it
almost faster than I can remember where the modules directories are.

Bruce
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304187 - in head: . share/man/man4 sys/conf sys/dev/mcd sys/modules sys/modules/mcd

[Warner Losh]

On Fri, Aug 19, 2016 at 3:35 AM, Toomas Soome  wrote:
>
>> On 19. aug 2016, at 12:24, Konstantin Belousov  wrote:
>>
>> On Fri, Aug 19, 2016 at 11:50:35AM +0300, Toomas Soome wrote:
>>>
 On 19. aug 2016, at 10:39, Konstantin Belousov  wrote:

 On Thu, Aug 18, 2016 at 09:28:57PM -0600, Warner Losh wrote:
> On Thu, Aug 18, 2016 at 12:50 AM, Julian Elischer  
> wrote:
>> On 16/08/2016 4:54 AM, John Baldwin wrote:
>>>
>>> On Monday, August 15, 2016 08:38:02 PM John Baldwin wrote:

 Author: jhb
 Date: Mon Aug 15 20:38:02 2016
 New Revision: 304187
 URL: https://svnweb.freebsd.org/changeset/base/304187

 Log:
 Remove the mcd(4) driver for Mitsumi CD-ROM players.
This is a driver for a pre-ATAPI ISA CD-ROM adapter.  As noted in
 the manpage, this driver is only useful as a backend to cdcontrol to
 play audio CDs since it doesn't use DMA, so its data performance is
 "abysmal" (and that was true in the mid 90's).
>>>
>>> No one stepped up to test patches for it either when I last posted 
>>> patches
>>> to
>>> convert it from timeout(9) to callout(9).  I have a few more drivers 
>>> that
>>> are
>>> both very old and that people have no business using in 12 (think ISA
>>> adapters that don't do DMA and can't be used with pccard) that I will be
>>> removing over the next little while.  I brought up a list of drivers on
>>> arch@
>>> a couple of years ago and the conversation drifted off into the weeds
>>> about
>>> trimming GENERIC, etc.  No one objected to the specific drivers I listed
>>> though (and I got a few pleas of "please remove").  If someone shows up
>>> desperately clutching an ISA adapter they can always dig up the source
>>> from
>>> svn and deal with forward porting it for whatever API changes have
>>> happened
>>> since it was removed.
>>
>>
>> I would imagine any machine still holding one of these probably has not
>> enough memory to run FreeBSD.
>>
>> would we still run in 2MB?
>
> With insane levels of tuning, we can run in 32MB userland that can do
> things. Even 64MB is tight w/o some tuning. 16MB is almost certainly
> right out except for very specialized situations. 2MB? We can't even
> load the loader in that :(. Oh, and all these memory configs are only
> possible if you tweak the loader's block cache...
>

 32MB is quite usable.  Without any tuning, you get slightly less than 10MB
 for userspace, which is enough to for many things, and plenty if swap is
 added.

 Note that you cannot boot on such configurations since loader was broken,
 but if you do manage to jump to kernel, things were fine several months
 ago.  I tested my relatively recent OOM changes on 32MB qemu config.
> Warner

>>>
>>> If the target is to go as low memory as possible, sure, you can strip all 
>>> off, from boot loader point, you should load kernel from stage2 and not use 
>>> loader at all (you can load and jump kernel even now from stage2, assuming 
>>> it wont need any special configuration from loader config) etc etc. This 
>>> means highly specialized build and has nothing to do with generic all 
>>> purpose system.
>>>
>> Why you describe this as an 'alternative' ? Before that loader changes,
>> I regularly tested on 32MB qemy i386 image and 64MB amd64 image. I do
>> not see anything extreme in these configs. They use normal boot path,
>> which provides kernels with debugging symbols, metadata, loaded modules
>> etc. Why should I use deficient boot2-only loading, which, additionally,
>> cannot work on amd64 ?
>>
>> More, this is the only reasonable way for most developers to ensure that
>> system is still usable on tiny configs found on embedded devices. Right
>> now the min which I have to set up is 128MB, and VM changes are simply not
>> tested on anything smaller. It is guaranteed that small systems will grow
>> regressions fast.  And I will not jump through the hoops to mitigate
>> breakage induced by other people' changes.
>>
>>> Also at some point, there is an question about how reasonable it is to have 
>>> such configuration as part of generic code base for special bits like boot 
>>> loader itself, as the problem is, testing all those variants is becoming 
>>> impossible and even keeping reasonable code base in all of the #if #else 
>>> #endif spaghetti is getting quite hard and error prone.
>>>
 From developers point of view, it is not really encouraging to have 
 possible feedback like ???oh, but you did break my 32MB system boot??? ;) 
 This does bring back some memories however. For first 2 unix systems I was 
 dealing with, one had 8MB and another had 12MB of memory??? it was ~ 
 1992-1993;)
>>>
>> Not 

Re: svn commit: r304187 - in head: . share/man/man4 sys/conf sys/dev/mcd sys/modules sys/modules/mcd

[Warner Losh]

On Fri, Aug 19, 2016 at 5:27 AM, Konstantin Belousov
 wrote:
> On Fri, Aug 19, 2016 at 09:12:53PM +1000, Bruce Evans wrote:
>> On Fri, 19 Aug 2016, Konstantin Belousov wrote:
>>
>> > On Thu, Aug 18, 2016 at 09:28:57PM -0600, Warner Losh wrote:
>> >> On Thu, Aug 18, 2016 at 12:50 AM, Julian Elischer  
>> >> wrote:
>> >>> On 16/08/2016 4:54 AM, John Baldwin wrote:
>> 
>>  On Monday, August 15, 2016 08:38:02 PM John Baldwin wrote:
>> > ,,,
>> > Log:
>> >Remove the mcd(4) driver for Mitsumi CD-ROM players.
>> >   This is a driver for a pre-ATAPI ISA CD-ROM adapter.  As noted in
>> >the manpage, this driver is only useful as a backend to cdcontrol to
>> >play audio CDs since it doesn't use DMA, so its data performance is
>> >"abysmal" (and that was true in the mid 90's).
>> 
>>  No one stepped up to test patches for it either when I last posted 
>>  patches
>>  to
>>  convert it from timeout(9) to callout(9).  I have a few more drivers 
>>  that
>>  are
>>  ...
>> >>>
>> >>> I would imagine any machine still holding one of these probably has not
>> >>> enough memory to run FreeBSD.
>> >>>
>> >>> would we still run in 2MB?
>> >>
>> >> With insane levels of tuning, we can run in 32MB userland that can do
>> >> things. Even 64MB is tight w/o some tuning. 16MB is almost certainly
>> >> right out except for very specialized situations. 2MB? We can't even
>> >> load the loader in that :(. Oh, and all these memory configs are only
>> >> possible if you tweak the loader's block cache...
>> >
>> > 32MB is quite usable.  Without any tuning, you get slightly less than 10MB
>> > for userspace, which is enough to for many things, and plenty if swap is
>> > added.
>>
>> No, 32MB needs lots of needs tuning.  -current seems to need about 16MB
>> more than just a few months ago when I last discussed this with you.
>> My i386 system doesn't have many drivers or a bloated userland (*),
>> but it took the following tuning plus my PAE tuning fixes to boot in
>> 32MB:
>> - disable [l]em1
>> - reduce tx and buffers to 256 for em0.  They default to 4096 for em.  That
>>is something like 8K * 1500 bytes = 12MB for em0 alone.  lem[1] wants
>>another 12MB.  I think this is not all statically allocated, but the
>>drivers hang onto that much.
>> This now longer works.  -current without my PAE tuning fixes hangs mounting
>> root or in usb initialization with this tuning and 40MB.  -current with my
>> PAE tuning fixes hangs similarly with 32MB; with 40MB it boots to the
>> start of multiuser but then hangs (it starts 1 getty, then 2 sendmails
>> and kills them with "out of swap space") and 1 rpcbind (also killed).  I
>> don't use swap, but it was needed 20 years ago on a system that actually
>> had 32MB of memory.
>>
>> (*) /bin/sh doing nothing much in -current i386: size 6532K res 1924K
>>  /bin/shin my ~5.2  i386: size  864K res  592K
>>
>> The kernel size is 5.5MB text 370K data 2.2MB bss (lots of bloat in bss
>> for debugging and vt).  In single user mode, with 40MB to start, there
>> is 22116K wired and 2516K free.  A few programs can be run in 2516K
>> without swap if they have res 592K and not 1924K.
>>
>> > Note that you cannot boot on such configurations since loader was broken,
>> > but if you do manage to jump to kernel, things were fine several months
>> > ago.  I tested my relatively recent OOM changes on 32MB qemu config.
>>
>> I have no problems booting such configuratations since I don't use the
>> current loader and only use old loader to change the environment to
>> set up special configurations like this.  I use my version of biosboot
>> for boot2.  This requires fixing 2 layers of complicated breakage in
>> init386().  vm86 is now used before the TSS and PIC resources that it
>> uses are initialized, but only in paths that are not normally used
>> because they are for memory sizing that is normally done by loader
>> :-(.
>>
>> I normally use my version of biosboot for boot2.  I improved its caching
>> just a couple of years ago.  It was using 9K buffer optimized for 1440K
>> floppies.  Now it uses a 32K buffer.  Booting a 5.5MB kernel takes a
>> fraction of a second.  Of course I don't use modules, so not many seeks
>> are needed.
>
> Of course I use modules and do not use GENERIC.  I just tried: with today
> HEAD, and old loader on 32MB VM, I get 11MB free in single-user mode.
> Active+inactive is ~4MB, and 1M is eaten by buffers, which is about right
> for init+/bin/sh+top idle system.

The situation on x86 must be a lot better than arm. My old Atmel boards with
32MB have < 1MB free when booted to the login prompt (more at single user)
and need special tuning to reduce the 5MB of network buffers allocated to
be anything approaching useful.

Warner
___
svn-src-all@freebsd.org mailing list

Re: svn commit: r304142 - head/usr.sbin/bsdinstall/partedit

[Warner Losh]

On Fri, Aug 19, 2016 at 12:51 AM, Dag-Erling Smørgrav  wrote:
> Warner Losh  writes:
>> Allan Jude  writes:
>> > Which makes more sense:
>> >
>> > A) If stripesize == 0, use some sane value like 4096
>>
>> I don't like this.
>>
>> > B) Some other combination that uses the reported stripe size, unless it
>> > is 0, in which case it uses 4096 (or some other value controlled by a
>> > different new sysctl)
>>
>> Don't like this so much.
>>
>> > C) create kern.geom.min_stripe_size with a default of 512, but users can
>> > set 4096 if they use only 4k devices. (doesn't really solve the problem
>> > for the installer)
>>
>> Default it to 4k, and allow users to set it to 512. If the drive
>> reports < this value
>> report this value instead.
>
> I don't like either option.  Option D (which I don't like either, but
> which should at least work in most cases) is a sysctl that specifies a
> minimum factor, and set the reported stripe size to the least common
> multiple of that number and the actual stripe or sector size.  This is
> what my bsdinstall patch does.  However, I think that pushing this down
> to a layer where it will affect all applications is a terrible idea,
> because we have no way of knowing what will break[*], and it can
> seriously mislead users and hinder troubleshooting - especially if it is
> enabled by default rather than only when necessary.

I took a look into the implications of doing a 4k stripesize 'automatically'
this morning. I found a few places in g_part where it would actively
hurt when coupled with gpart's insistence on aligning things. So I
now think it's a bad idea. This will make it harder for FreeBSD to
generate arbitrary disk layouts. And I'm not too sure about what
things like gstripe would report as a result and if this would actually
interfere if you had a large, but not power of two stripe size.

> I don't think it's a good idea to enforce stripe alignment everywhere,
> either.  It works for partitions because they are very large relative to
> the stripe size, and at worst we will waste a few millionths of the disk
> on inter-partition gaps, which should only occur between the partition
> table and the boot partition, and possibly, if the stripe size is very
> large, between the boot partition and the swap partition.  But forcing
> filesystems to respect the stripe size will lead to no end of trouble,
> because RAID volumes can have stripe sizes of 16 kB or more.  I think it
> is important to align partitions during installation because of the huge
> performance impact of misaligned partitions on AF disks, but despite
> what Nathan claims, I never advocated applying the same logic
> everywhere.

Yea, having poked at it for just a little while, I agree. The installer is the
right place to make sure we don't cross-thread the 4k sectors. Stripe size
means too many other things to have it be useful in that context.

Warner
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304466 - stable/9/sys/contrib/pf/net

[Kristof Provost]

Author: kp
Date: Fri Aug 19 13:39:36 2016
New Revision: 304466
URL: https://svnweb.freebsd.org/changeset/base/304466

Log:
  MFC r304152:
  
  pf: Add missing byte-order swap to pf_match_addr_range
  
  Without this, rules using address ranges (e.g. "10.1.1.1 - 10.1.1.5") did not
  match addresses correctly on little-endian systems.
  
  PR:   211796
  Obtained from:OpenBSD (sthen)

Modified:
  stable/9/sys/contrib/pf/net/pf.c

Modified: stable/9/sys/contrib/pf/net/pf.c
==
--- stable/9/sys/contrib/pf/net/pf.cFri Aug 19 13:12:57 2016
(r304465)
+++ stable/9/sys/contrib/pf/net/pf.cFri Aug 19 13:39:36 2016
(r304466)
@@ -2691,8 +2691,8 @@ pf_match_addr_range(struct pf_addr *b, s
switch (af) {
 #ifdef INET
case AF_INET:
-   if ((a->addr32[0] < b->addr32[0]) ||
-   (a->addr32[0] > e->addr32[0]))
+   if ((ntohl(a->addr32[0]) < ntohl(b->addr32[0])) ||
+   (ntohl(a->addr32[0]) > ntohl(e->addr32[0])))
return (0);
break;
 #endif /* INET */
@@ -2702,15 +2702,15 @@ pf_match_addr_range(struct pf_addr *b, s
 
/* check a >= b */
for (i = 0; i < 4; ++i)
-   if (a->addr32[i] > b->addr32[i])
+   if (ntohl(a->addr32[i]) > ntohl(b->addr32[i]))
break;
-   else if (a->addr32[i] < b->addr32[i])
+   else if (ntohl(a->addr32[i]) < ntohl(b->addr32[i]))
return (0);
/* check a <= e */
for (i = 0; i < 4; ++i)
-   if (a->addr32[i] < e->addr32[i])
+   if (ntohl(a->addr32[i]) < ntohl(e->addr32[i]))
break;
-   else if (a->addr32[i] > e->addr32[i])
+   else if (ntohl(a->addr32[i]) > ntohl(e->addr32[i]))
return (0);
break;
}
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304187 - in head: . share/man/man4 sys/conf sys/dev/mcd sys/modules sys/modules/mcd

[Alexey Dokuchaev]

On Fri, Aug 19, 2016 at 09:12:53PM +1000, Bruce Evans wrote:
> I normally use my version of biosboot for boot2.  I improved its caching
> just a couple of years ago.  It was using 9K buffer optimized for 1440K
> floppies.  Now it uses a 32K buffer.  Booting a 5.5MB kernel takes a
> fraction of a second.

That looks like change that everyone would benefit from.  Consider posting
a patch or committing it yourself. ;-)

> Of course I don't use modules, so not many seeks are needed.

Why not, they're convenient (apart from "kernel version mismatch" crap
that I keep hitting now and then)?  AFAIR they were slow to load at some
point but that was (not so) recently fixed.

./danfe
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304464 - in head/sys/arm/allwinner: . a10

[Emmanuel Vadot]

Author: manu
Date: Fri Aug 19 12:48:32 2016
New Revision: 304464
URL: https://svnweb.freebsd.org/changeset/base/304464

Log:
  Allwinner: Move a10_padconf.c into a10 subdirectory.

Added:
  head/sys/arm/allwinner/a10/a10_padconf.c
 - copied unchanged from r304463, head/sys/arm/allwinner/a10_padconf.c
Deleted:
  head/sys/arm/allwinner/a10_padconf.c
Modified:
  head/sys/arm/allwinner/a10/files.a10

Copied: head/sys/arm/allwinner/a10/a10_padconf.c (from r304463, 
head/sys/arm/allwinner/a10_padconf.c)
==
--- /dev/null   00:00:00 1970   (empty, because file is newly added)
+++ head/sys/arm/allwinner/a10/a10_padconf.cFri Aug 19 12:48:32 2016
(r304464, copy of r304463, head/sys/arm/allwinner/a10_padconf.c)
@@ -0,0 +1,231 @@
+/*-
+ * Copyright (c) 2016 Emmanuel Vadot 
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *notice, this list of conditions and the following disclaimer in the
+ *documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ */
+
+#include 
+__FBSDID("$FreeBSD$");
+
+#include 
+#include 
+#include 
+#include 
+
+#include 
+
+#ifdef SOC_ALLWINNER_A10
+
+const static struct allwinner_pins a10_pins[] = {
+   {"PA0",  0, 0,  {"gpio_in", "gpio_out", "emac", "spi1", "uart2", NULL, 
NULL, NULL}},
+   {"PA1",  0, 1,  {"gpio_in", "gpio_out", "emac", "spi1", "uart2", NULL, 
NULL, NULL}},
+   {"PA2",  0, 2,  {"gpio_in", "gpio_out", "emac", "spi1", "uart2", NULL, 
NULL, NULL}},
+   {"PA3",  0, 3,  {"gpio_in", "gpio_out", "emac", "spi1", "uart2", NULL, 
NULL, NULL}},
+   {"PA4",  0, 4,  {"gpio_in", "gpio_out", "emac", "spi1", NULL, NULL, 
NULL, NULL}},
+   {"PA5",  0, 5,  {"gpio_in", "gpio_out", "emac", "spi3", NULL, NULL, 
NULL, NULL}},
+   {"PA6",  0, 6,  {"gpio_in", "gpio_out", "emac", "spi3", NULL, NULL, 
NULL, NULL}},
+   {"PA7",  0, 7,  {"gpio_in", "gpio_out", "emac", "spi3", NULL, NULL, 
NULL, NULL}},
+   {"PA8",  0, 8,  {"gpio_in", "gpio_out", "emac", "spi3", NULL, NULL, 
NULL, NULL}},
+   {"PA9",  0, 9,  {"gpio_in", "gpio_out", "emac", "spi3", NULL, NULL, 
NULL, NULL}},
+   {"PA10", 0, 10, {"gpio_in", "gpio_out", "emac", NULL, "uart1", NULL, 
NULL, NULL}},
+   {"PA11", 0, 11, {"gpio_in", "gpio_out", "emac", NULL, "uart1", NULL, 
NULL, NULL}},
+   {"PA12", 0, 12, {"gpio_in", "gpio_out", "emac", "uart6", "uart1", NULL, 
NULL, NULL}},
+   {"PA13", 0, 13, {"gpio_in", "gpio_out", "emac", "uart6", "uart1", NULL, 
NULL, NULL}},
+   {"PA14", 0, 14, {"gpio_in", "gpio_out", "emac", "uart7", "uart1", NULL, 
NULL, NULL}},
+   {"PA15", 0, 15, {"gpio_in", "gpio_out", "emac", "uart7", "uart1", NULL, 
NULL, NULL}},
+   {"PA16", 0, 16, {"gpio_in", "gpio_out", NULL, "can", "uart1", NULL, 
NULL, NULL}},
+   {"PA17", 0, 17, {"gpio_in", "gpio_out", NULL, "can", "uart1", NULL, 
NULL, NULL}},
+
+   {"PB0",  1, 0,  {"gpio_in", "gpio_out", "i2c0", NULL, NULL, NULL, NULL, 
NULL}},
+   {"PB1",  1, 1,  {"gpio_in", "gpio_out", "i2c0", NULL, NULL, NULL, NULL, 
NULL}},
+   {"PB2",  1, 2,  {"gpio_in", "gpio_out", "pwm", NULL, NULL, NULL, NULL, 
NULL}},
+   {"PB3",  1, 3,  {"gpio_in", "gpio_out", "ir0", NULL, NULL, NULL, NULL, 
NULL}},
+   {"PB4",  1, 4,  {"gpio_in", "gpio_out", "ir0", NULL, NULL, NULL, NULL, 
NULL}},
+   {"PB5",  1, 5,  {"gpio_in", "gpio_out", "i2s", "ac97", NULL, NULL, 
NULL, NULL}},
+   {"PB6",  1, 6,  {"gpio_in", "gpio_out", "i2s", "ac97", NULL, NULL, 
NULL, NULL}},
+   {"PB7",  1, 7,  {"gpio_in", "gpio_out", "i2s", "ac97", NULL, NULL, 
NULL, NULL}},
+   {"PB8",  1, 8,  {"gpio_in", "gpio_out", "i2s", "ac97", NULL, NULL, 
NULL, NULL}},
+   {"PB9",  1, 9,  {"gpio_in", "gpio_out", "i2s", NULL, NULL, NULL, NULL, 
NULL}},
+   {"PB10", 1, 10, 

svn commit: r304463 - stable/10/sys/netpfil/pf

[Kristof Provost]

Author: kp
Date: Fri Aug 19 11:36:00 2016
New Revision: 304463
URL: https://svnweb.freebsd.org/changeset/base/304463

Log:
  MFC r304152:
  
  pf: Add missing byte-order swap to pf_match_addr_range
  
  Without this, rules using address ranges (e.g. "10.1.1.1 - 10.1.1.5") did not
  match addresses correctly on little-endian systems.
  
  PR:   211796
  Obtained from:OpenBSD (sthen)

Modified:
  stable/10/sys/netpfil/pf/pf.c
Directory Properties:
  stable/10/   (props changed)

Modified: stable/10/sys/netpfil/pf/pf.c
==
--- stable/10/sys/netpfil/pf/pf.c   Fri Aug 19 11:31:30 2016
(r304462)
+++ stable/10/sys/netpfil/pf/pf.c   Fri Aug 19 11:36:00 2016
(r304463)
@@ -2564,8 +2564,8 @@ pf_match_addr_range(struct pf_addr *b, s
switch (af) {
 #ifdef INET
case AF_INET:
-   if ((a->addr32[0] < b->addr32[0]) ||
-   (a->addr32[0] > e->addr32[0]))
+   if ((ntohl(a->addr32[0]) < ntohl(b->addr32[0])) ||
+   (ntohl(a->addr32[0]) > ntohl(e->addr32[0])))
return (0);
break;
 #endif /* INET */
@@ -2575,15 +2575,15 @@ pf_match_addr_range(struct pf_addr *b, s
 
/* check a >= b */
for (i = 0; i < 4; ++i)
-   if (a->addr32[i] > b->addr32[i])
+   if (ntohl(a->addr32[i]) > ntohl(b->addr32[i]))
break;
-   else if (a->addr32[i] < b->addr32[i])
+   else if (ntohl(a->addr32[i]) < ntohl(b->addr32[i]))
return (0);
/* check a <= e */
for (i = 0; i < 4; ++i)
-   if (a->addr32[i] < e->addr32[i])
+   if (ntohl(a->addr32[i]) < ntohl(e->addr32[i]))
break;
-   else if (a->addr32[i] > e->addr32[i])
+   else if (ntohl(a->addr32[i]) > ntohl(e->addr32[i]))
return (0);
break;
}
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304462 - stable/11/sys/netpfil/pf

[Kristof Provost]

Author: kp
Date: Fri Aug 19 11:31:30 2016
New Revision: 304462
URL: https://svnweb.freebsd.org/changeset/base/304462

Log:
  MFC r304152:
  
  pf: Add missing byte-order swap to pf_match_addr_range
  
  Without this, rules using address ranges (e.g. "10.1.1.1 - 10.1.1.5") did not
  match addresses correctly on little-endian systems.
  
  PR:   211796
  Obtained from:OpenBSD (sthen)

Modified:
  stable/11/sys/netpfil/pf/pf.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/sys/netpfil/pf/pf.c
==
--- stable/11/sys/netpfil/pf/pf.c   Fri Aug 19 11:12:59 2016
(r304461)
+++ stable/11/sys/netpfil/pf/pf.c   Fri Aug 19 11:31:30 2016
(r304462)
@@ -2600,8 +2600,8 @@ pf_match_addr_range(struct pf_addr *b, s
switch (af) {
 #ifdef INET
case AF_INET:
-   if ((a->addr32[0] < b->addr32[0]) ||
-   (a->addr32[0] > e->addr32[0]))
+   if ((ntohl(a->addr32[0]) < ntohl(b->addr32[0])) ||
+   (ntohl(a->addr32[0]) > ntohl(e->addr32[0])))
return (0);
break;
 #endif /* INET */
@@ -2611,15 +2611,15 @@ pf_match_addr_range(struct pf_addr *b, s
 
/* check a >= b */
for (i = 0; i < 4; ++i)
-   if (a->addr32[i] > b->addr32[i])
+   if (ntohl(a->addr32[i]) > ntohl(b->addr32[i]))
break;
-   else if (a->addr32[i] < b->addr32[i])
+   else if (ntohl(a->addr32[i]) < ntohl(b->addr32[i]))
return (0);
/* check a <= e */
for (i = 0; i < 4; ++i)
-   if (a->addr32[i] < e->addr32[i])
+   if (ntohl(a->addr32[i]) < ntohl(e->addr32[i]))
break;
-   else if (a->addr32[i] > e->addr32[i])
+   else if (ntohl(a->addr32[i]) > ntohl(e->addr32[i]))
return (0);
break;
}
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304187 - in head: . share/man/man4 sys/conf sys/dev/mcd sys/modules sys/modules/mcd

[Konstantin Belousov]

On Fri, Aug 19, 2016 at 09:12:53PM +1000, Bruce Evans wrote:
> On Fri, 19 Aug 2016, Konstantin Belousov wrote:
> 
> > On Thu, Aug 18, 2016 at 09:28:57PM -0600, Warner Losh wrote:
> >> On Thu, Aug 18, 2016 at 12:50 AM, Julian Elischer  
> >> wrote:
> >>> On 16/08/2016 4:54 AM, John Baldwin wrote:
> 
>  On Monday, August 15, 2016 08:38:02 PM John Baldwin wrote:
> > ,,,
> > Log:
> >Remove the mcd(4) driver for Mitsumi CD-ROM players.
> >   This is a driver for a pre-ATAPI ISA CD-ROM adapter.  As noted in
> >the manpage, this driver is only useful as a backend to cdcontrol to
> >play audio CDs since it doesn't use DMA, so its data performance is
> >"abysmal" (and that was true in the mid 90's).
> 
>  No one stepped up to test patches for it either when I last posted 
>  patches
>  to
>  convert it from timeout(9) to callout(9).  I have a few more drivers that
>  are
>  ...
> >>>
> >>> I would imagine any machine still holding one of these probably has not
> >>> enough memory to run FreeBSD.
> >>>
> >>> would we still run in 2MB?
> >>
> >> With insane levels of tuning, we can run in 32MB userland that can do
> >> things. Even 64MB is tight w/o some tuning. 16MB is almost certainly
> >> right out except for very specialized situations. 2MB? We can't even
> >> load the loader in that :(. Oh, and all these memory configs are only
> >> possible if you tweak the loader's block cache...
> >
> > 32MB is quite usable.  Without any tuning, you get slightly less than 10MB
> > for userspace, which is enough to for many things, and plenty if swap is
> > added.
> 
> No, 32MB needs lots of needs tuning.  -current seems to need about 16MB
> more than just a few months ago when I last discussed this with you.
> My i386 system doesn't have many drivers or a bloated userland (*),
> but it took the following tuning plus my PAE tuning fixes to boot in
> 32MB:
> - disable [l]em1
> - reduce tx and buffers to 256 for em0.  They default to 4096 for em.  That
>is something like 8K * 1500 bytes = 12MB for em0 alone.  lem[1] wants
>another 12MB.  I think this is not all statically allocated, but the
>drivers hang onto that much.
> This now longer works.  -current without my PAE tuning fixes hangs mounting
> root or in usb initialization with this tuning and 40MB.  -current with my
> PAE tuning fixes hangs similarly with 32MB; with 40MB it boots to the
> start of multiuser but then hangs (it starts 1 getty, then 2 sendmails
> and kills them with "out of swap space") and 1 rpcbind (also killed).  I
> don't use swap, but it was needed 20 years ago on a system that actually
> had 32MB of memory.
> 
> (*) /bin/sh doing nothing much in -current i386: size 6532K res 1924K
>  /bin/shin my ~5.2  i386: size  864K res  592K
> 
> The kernel size is 5.5MB text 370K data 2.2MB bss (lots of bloat in bss
> for debugging and vt).  In single user mode, with 40MB to start, there
> is 22116K wired and 2516K free.  A few programs can be run in 2516K
> without swap if they have res 592K and not 1924K.
> 
> > Note that you cannot boot on such configurations since loader was broken,
> > but if you do manage to jump to kernel, things were fine several months
> > ago.  I tested my relatively recent OOM changes on 32MB qemu config.
> 
> I have no problems booting such configuratations since I don't use the
> current loader and only use old loader to change the environment to
> set up special configurations like this.  I use my version of biosboot
> for boot2.  This requires fixing 2 layers of complicated breakage in
> init386().  vm86 is now used before the TSS and PIC resources that it
> uses are initialized, but only in paths that are not normally used
> because they are for memory sizing that is normally done by loader
> :-(.
> 
> I normally use my version of biosboot for boot2.  I improved its caching
> just a couple of years ago.  It was using 9K buffer optimized for 1440K
> floppies.  Now it uses a 32K buffer.  Booting a 5.5MB kernel takes a
> fraction of a second.  Of course I don't use modules, so not many seeks
> are needed.

Of course I use modules and do not use GENERIC.  I just tried: with today
HEAD, and old loader on 32MB VM, I get 11MB free in single-user mode.
Active+inactive is ~4MB, and 1M is eaten by buffers, which is about right
for init+/bin/sh+top idle system.

Anyway, judging from the other responses, this is the lost case.
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304187 - in head: . share/man/man4 sys/conf sys/dev/mcd sys/modules sys/modules/mcd

[Bruce Evans]

On Fri, 19 Aug 2016, Konstantin Belousov wrote:


On Thu, Aug 18, 2016 at 09:28:57PM -0600, Warner Losh wrote:

On Thu, Aug 18, 2016 at 12:50 AM, Julian Elischer  wrote:

On 16/08/2016 4:54 AM, John Baldwin wrote:


On Monday, August 15, 2016 08:38:02 PM John Baldwin wrote:

,,,
Log:
   Remove the mcd(4) driver for Mitsumi CD-ROM players.
  This is a driver for a pre-ATAPI ISA CD-ROM adapter.  As noted in
   the manpage, this driver is only useful as a backend to cdcontrol to
   play audio CDs since it doesn't use DMA, so its data performance is
   "abysmal" (and that was true in the mid 90's).


No one stepped up to test patches for it either when I last posted patches
to
convert it from timeout(9) to callout(9).  I have a few more drivers that
are
...


I would imagine any machine still holding one of these probably has not
enough memory to run FreeBSD.

would we still run in 2MB?


With insane levels of tuning, we can run in 32MB userland that can do
things. Even 64MB is tight w/o some tuning. 16MB is almost certainly
right out except for very specialized situations. 2MB? We can't even
load the loader in that :(. Oh, and all these memory configs are only
possible if you tweak the loader's block cache...


32MB is quite usable.  Without any tuning, you get slightly less than 10MB
for userspace, which is enough to for many things, and plenty if swap is
added.


No, 32MB needs lots of needs tuning.  -current seems to need about 16MB
more than just a few months ago when I last discussed this with you.
My i386 system doesn't have many drivers or a bloated userland (*),
but it took the following tuning plus my PAE tuning fixes to boot in
32MB:
- disable [l]em1
- reduce tx and buffers to 256 for em0.  They default to 4096 for em.  That
  is something like 8K * 1500 bytes = 12MB for em0 alone.  lem[1] wants
  another 12MB.  I think this is not all statically allocated, but the
  drivers hang onto that much.
This now longer works.  -current without my PAE tuning fixes hangs mounting
root or in usb initialization with this tuning and 40MB.  -current with my
PAE tuning fixes hangs similarly with 32MB; with 40MB it boots to the
start of multiuser but then hangs (it starts 1 getty, then 2 sendmails
and kills them with "out of swap space") and 1 rpcbind (also killed).  I
don't use swap, but it was needed 20 years ago on a system that actually
had 32MB of memory.

(*) /bin/sh doing nothing much in -current i386: size 6532K res 1924K
/bin/shin my ~5.2  i386: size  864K res  592K

The kernel size is 5.5MB text 370K data 2.2MB bss (lots of bloat in bss
for debugging and vt).  In single user mode, with 40MB to start, there
is 22116K wired and 2516K free.  A few programs can be run in 2516K
without swap if they have res 592K and not 1924K.


Note that you cannot boot on such configurations since loader was broken,
but if you do manage to jump to kernel, things were fine several months
ago.  I tested my relatively recent OOM changes on 32MB qemu config.


I have no problems booting such configuratations since I don't use the
current loader and only use old loader to change the environment to
set up special configurations like this.  I use my version of biosboot
for boot2.  This requires fixing 2 layers of complicated breakage in
init386().  vm86 is now used before the TSS and PIC resources that it
uses are initialized, but only in paths that are not normally used
because they are for memory sizing that is normally done by loader
:-(.

I normally use my version of biosboot for boot2.  I improved its caching
just a couple of years ago.  It was using 9K buffer optimized for 1440K
floppies.  Now it uses a 32K buffer.  Booting a 5.5MB kernel takes a
fraction of a second.  Of course I don't use modules, so not many seeks
are needed.

Bruce
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304461 - head/sys/arm/nvidia

[Michal Meloun]

Author: mmel
Date: Fri Aug 19 11:12:59 2016
New Revision: 304461
URL: https://svnweb.freebsd.org/changeset/base/304461

Log:
  TEGRA: Remove forgotten debug printf.

Modified:
  head/sys/arm/nvidia/tegra_pcie.c

Modified: head/sys/arm/nvidia/tegra_pcie.c
==
--- head/sys/arm/nvidia/tegra_pcie.cFri Aug 19 10:53:17 2016
(r304460)
+++ head/sys/arm/nvidia/tegra_pcie.cFri Aug 19 11:12:59 2016
(r304461)
@@ -567,7 +567,6 @@ tegra_pcib_msi_intr(void *arg)
/* Handle one vector. */
while (reg != 0) {
bit = ffs(reg) - 1;
-//printf("%s: i: %d, reg: 0x%08X, bit: 0x%08X, addr: 0x%08llX\n", __func__, i, 
reg, bit, rman_get_start(sc->afi_mem_res));
/* Send EOI */
AFI_WR4(sc, AFI_MSI_VEC(i), 1 << bit);
irq = i * AFI_MSI_INTR_IN_REG + bit;
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304439 - head/sys/dev/usb/net

[YongHyeon PYUN]

On Fri, Aug 19, 2016 at 12:39:58PM +0200, Hans Petter Selasky wrote:
> On 08/19/16 11:22, YongHyeon PYUN wrote:
> >On Fri, Aug 19, 2016 at 11:11:56AM +0200, Hans Petter Selasky wrote:
> >>On 08/19/16 10:55, YongHyeon PYUN wrote:
> >>>I think the order is right but it was not tested on big-endian
> >>>systems.
> >>
> >>Hi,
> >>
> >>I'm pretty sure the ifdef is wrong, because you write the fields one at
> >>a time, using htole32():
> >>
> >>txhdr.mss = 0;
> >>txhdr.len = 
> >>htole32(AXGE_TXBYTES(m->m_pkthdr.len));
> >>
> >>Big endian machines don't re-order variables like this.
> >>
> >>You should remove the #else part.
> >
> >Wouldn't USB stack pass txhdr structure without any
> >modification? And controller want to see len (low 32bits address)
> >first and then mss (high 32bits address). On big endian systems I
> >guess this should be reversed in host memory layout.  This is so
> >confusing so I could be wrong.
> 
> The USB stack passes TXHDR as-is and the host controller is byte 
> oriented, not 64-bit word oriented. That's why the layout is the same as 
> long as you assign per 32-bit field.
> 

Ok, fixed in r304458.
Thanks for pointing it out!
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304459 - in head/sys: arm/arm arm/nvidia arm/ti/omap4 arm64/arm64 dev/fdt dev/gpio dev/iicbus dev/ofw dev/pci dev/vnic kern mips/mips sys

[Michal Meloun]

Author: mmel
Date: Fri Aug 19 10:52:39 2016
New Revision: 304459
URL: https://svnweb.freebsd.org/changeset/base/304459

Log:
  INTRNG: Rework handling with resources. Partially revert r301453.
   - Read interrupt properties at bus enumeration time and store
 it into global mapping table.
   - At bus_activate_resource() time, given mapping entry is resolved and
 connected to real interrupt source. A copy of mapping entry is attached
 to given resource.
   - At bus_setup_intr() time, mapping entry stored in resource is used
 for delivery of requested interrupt configuration.
   - For MSI/MSIX interrupts, mapping entry is created within
 pci_alloc_msi()/pci_alloc_msix() call.
   - For legacy PCI interrupts, mapping entry must be created within
 pcib_route_interrupt() by pcib driver itself.
  
  Reviewed by: nwhitehorn, andrew
  Differential Revision: https://reviews.freebsd.org/D7493

Modified:
  head/sys/arm/arm/nexus.c
  head/sys/arm/nvidia/tegra_lic.c
  head/sys/arm/ti/omap4/omap4_wugen.c
  head/sys/arm64/arm64/nexus.c
  head/sys/dev/fdt/simplebus.c
  head/sys/dev/gpio/gpiobus.c
  head/sys/dev/gpio/gpiobusvar.h
  head/sys/dev/gpio/ofw_gpiobus.c
  head/sys/dev/iicbus/ofw_iicbus.c
  head/sys/dev/ofw/ofw_bus_subr.c
  head/sys/dev/ofw/ofw_bus_subr.h
  head/sys/dev/ofw/ofwbus.c
  head/sys/dev/pci/pci_host_generic.c
  head/sys/dev/vnic/mrml_bridge.c
  head/sys/dev/vnic/thunder_mdio_fdt.c
  head/sys/kern/bus_if.m
  head/sys/kern/pic_if.m
  head/sys/kern/subr_bus.c
  head/sys/kern/subr_intr.c
  head/sys/mips/mips/nexus.c
  head/sys/sys/bus.h
  head/sys/sys/intr.h

Modified: head/sys/arm/arm/nexus.c
==
--- head/sys/arm/arm/nexus.cFri Aug 19 10:51:30 2016(r304458)
+++ head/sys/arm/arm/nexus.cFri Aug 19 10:52:39 2016(r304459)
@@ -64,6 +64,7 @@ __FBSDID("$FreeBSD$");
 
 #ifdef FDT
 #include 
+#include 
 #include "ofw_bus_if.h"
 #endif
 
@@ -379,6 +380,11 @@ nexus_activate_resource(device_t bus, de
 #endif
rman_set_virtual(r, (void *)vaddr);
rman_set_bushandle(r, vaddr);
+   return (0);
+   } else if (type == SYS_RES_IRQ) {
+#ifdef INTRNG
+   intr_activate_irq(child, r);
+#endif
}
return (0);
 }
@@ -390,17 +396,23 @@ nexus_deactivate_resource(device_t bus, 
bus_size_t psize;
bus_space_handle_t vaddr;
 
-   psize = (bus_size_t)rman_get_size(r);
-   vaddr = rman_get_bushandle(r);
+   if (type == SYS_RES_MEMORY || type == SYS_RES_IOPORT) {
+   psize = (bus_size_t)rman_get_size(r);
+   vaddr = rman_get_bushandle(r);
 
-   if (vaddr != 0) {
+   if (vaddr != 0) {
 #ifdef FDT
-   bus_space_unmap(fdtbus_bs_tag, vaddr, psize);
+   bus_space_unmap(fdtbus_bs_tag, vaddr, psize);
 #else
-   pmap_unmapdev((vm_offset_t)vaddr, (vm_size_t)psize);
+   pmap_unmapdev((vm_offset_t)vaddr, (vm_size_t)psize);
+#endif
+   rman_set_virtual(r, NULL);
+   rman_set_bushandle(r, 0);
+   }
+   } else if (type == SYS_RES_IRQ) {
+#ifdef INTRNG
+   intr_deactivate_irq(child, r);
 #endif
-   rman_set_virtual(r, NULL);
-   rman_set_bushandle(r, 0);
}
 
return (rman_deactivate_resource(r));
@@ -411,11 +423,22 @@ static int
 nexus_ofw_map_intr(device_t dev, device_t child, phandle_t iparent, int icells,
 pcell_t *intr)
 {
-
-#ifdef INTRNG
-   return (INTR_IRQ_INVALID);
-#else
+#ifndef INTRNG
return (intr_fdt_map_irq(iparent, intr, icells));
-#endif
+#else
+   u_int irq;
+   struct intr_map_data_fdt *fdt_data;
+   size_t len;
+
+   len = sizeof(*fdt_data) + icells * sizeof(pcell_t);
+   fdt_data = (struct intr_map_data_fdt *)intr_alloc_map_data(
+   INTR_MAP_DATA_FDT, len, M_WAITOK | M_ZERO);
+   fdt_data->iparent = iparent;
+   fdt_data->ncells = icells;
+   memcpy(fdt_data->cells, intr, icells * sizeof(pcell_t));
+   irq = intr_map_irq(NULL, iparent, (struct intr_map_data *)fdt_data);
+   return (irq);
+#endif /* INTRNG */
 }
-#endif
+#endif /* FDT */
+

Modified: head/sys/arm/nvidia/tegra_lic.c
==
--- head/sys/arm/nvidia/tegra_lic.c Fri Aug 19 10:51:30 2016
(r304458)
+++ head/sys/arm/nvidia/tegra_lic.c Fri Aug 19 10:52:39 2016
(r304459)
@@ -88,12 +88,12 @@ struct tegra_lic_sc {
 };
 
 static int
-tegra_lic_alloc_intr(device_t dev, struct intr_irqsrc *isrc,
+tegra_lic_activate_intr(device_t dev, struct intr_irqsrc *isrc,
 struct resource *res, struct intr_map_data *data)
 {
struct tegra_lic_sc *sc = device_get_softc(dev);
 
-   return (PIC_ALLOC_INTR(sc->parent, isrc, res, data));
+   return (PIC_ACTIVATE_INTR(sc->parent, isrc, res, data));
 

svn commit: r304460 - head/sys/arm/nvidia

[Michal Meloun]

Author: mmel
Date: Fri Aug 19 10:53:17 2016
New Revision: 304460
URL: https://svnweb.freebsd.org/changeset/base/304460

Log:
  TEGRA: Implement MSI/MSIX interrupts for pcie controller.

Modified:
  head/sys/arm/nvidia/tegra_pcie.c

Modified: head/sys/arm/nvidia/tegra_pcie.c
==
--- head/sys/arm/nvidia/tegra_pcie.cFri Aug 19 10:52:39 2016
(r304459)
+++ head/sys/arm/nvidia/tegra_pcie.cFri Aug 19 10:53:17 2016
(r304460)
@@ -33,20 +33,20 @@ __FBSDID("$FreeBSD$");
 
 #include 
 #include 
+#include 
+#include 
+#include 
 #include 
-#include 
 #include 
 #include 
 #include 
-#include 
-#include 
 #include 
-#include 
-#include 
 
 #include 
 
 #include 
+#include 
+#include 
 #include 
 
 #include 
@@ -68,8 +68,9 @@ __FBSDID("$FreeBSD$");
 #include 
 
 #include "ofw_bus_if.h"
+#include "msi_if.h"
 #include "pcib_if.h"
-
+#include "pic_if.h"
 
 
 #defineAFI_AXI_BAR0_SZ 0x000
@@ -93,17 +94,10 @@ __FBSDID("$FreeBSD$");
 #defineAFI_MSI_BAR_SZ  0x060
 #defineAFI_MSI_FPCI_BAR_ST 0x064
 #defineAFI_MSI_AXI_BAR_ST  0x068
-
-
-#defineAFI_AXI_BAR6_SZ 0x134
-#defineAFI_AXI_BAR7_SZ 0x138
-#defineAFI_AXI_BAR8_SZ 0x13c
-#defineAFI_AXI_BAR6_START  0x140
-#defineAFI_AXI_BAR7_START  0x144
-#defineAFI_AXI_BAR8_START  0x148
-#defineAFI_FPCI_BAR6   0x14c
-#defineAFI_FPCI_BAR7   0x150
-#defineAFI_FPCI_BAR8   0x154
+#define AFI_MSI_VEC(x) (0x06c + 4 * (x))
+#define AFI_MSI_EN_VEC(x)  (0x08c + 4 * (x))
+#define AFI_MSI_INTR_IN_REG32
+#define AFI_MSI_REGS   8
 
 #defineAFI_CONFIGURATION   0x0ac
 #define AFI_CONFIGURATION_EN_FPCI  (1 << 0)
@@ -209,6 +203,8 @@ __FBSDID("$FreeBSD$");
 
 #defineTEGRA_PCIE_LINKUP_TIMEOUT   200
 
+#define TEGRA_PCIB_MSI_ENABLE
+
 #defineDEBUG
 #ifdef DEBUG
 #definedebugf(fmt, args...) do { printf(fmt,##args); } while (0)
@@ -258,6 +254,13 @@ static struct ofw_compat_data compat_dat
{NULL,  0},
 };
 
+#defineTEGRA_FLAG_MSI_USED 0x0001
+struct tegra_pcib_irqsrc {
+   struct intr_irqsrc  isrc;
+   u_int   irq;
+   u_int   flags;
+};
+
 struct tegra_pcib_port {
int enabled;
int port_idx;   /* chip port index */
@@ -271,6 +274,7 @@ struct tegra_pcib_port {
 };
 
 #defineTEGRA_PCIB_MAX_PORTS3
+#defineTEGRA_PCIB_MAX_MSI  AFI_MSI_INTR_IN_REG * AFI_MSI_REGS
 struct tegra_pcib_softc {
struct ofw_pci_softcofw_pci;
device_tdev;
@@ -303,7 +307,7 @@ struct tegra_pcib_softc {
regulator_t supply_vddio_pex_ctl;
regulator_t supply_avdd_pll_erefe;
 
-   uint32_tmsi_bitmap;
+   vm_offset_t msi_page;   /* VA of MSI page */
bus_addr_t  cfg_base_addr;  /* base address of config */
bus_size_t  cfg_cur_offs;   /* currently mapped window */
bus_space_handle_t  cfg_handle; /* handle of config window */
@@ -311,9 +315,9 @@ struct tegra_pcib_softc {
int lanes_cfg;
int num_ports;
struct tegra_pcib_port *ports[TEGRA_PCIB_MAX_PORTS];
+   struct tegra_pcib_irqsrc *isrcs;
 };
 
-
 static int
 tegra_pcib_maxslots(device_t dev)
 {
@@ -324,13 +328,15 @@ static int
 tegra_pcib_route_interrupt(device_t bus, device_t dev, int pin)
 {
struct tegra_pcib_softc *sc;
+   u_int irq;
 
sc = device_get_softc(bus);
-   device_printf(bus, "route pin %d for device %d.%d to %ju\n",
+   irq = intr_map_clone_irq(rman_get_start(sc->irq_res));
+   device_printf(bus, "route pin %d for device %d.%d to %u\n",
  pin, pci_get_slot(dev), pci_get_function(dev),
- rman_get_start(sc->irq_res));
+ irq);
 
-   return (rman_get_start(sc->irq_res));
+   return (irq);
 }
 
 static int
@@ -471,84 +477,320 @@ static int tegra_pci_intr(void *arg)
return (FILTER_HANDLED);
 }
 
-#if defined(TEGRA_PCI_MSI)
+/* ---
+ *
+ * PCI MSI interface
+ */
+static int
+tegra_pcib_alloc_msi(device_t pci, device_t child, int count, int maxcount,
+int *irqs)
+{
+   phandle_t msi_parent;
+
+   /*  ofw_bus_msimap() don't 

svn commit: r304458 - head/sys/dev/usb/net

[Pyun YongHyeon]

Author: yongari
Date: Fri Aug 19 10:51:30 2016
New Revision: 304458
URL: https://svnweb.freebsd.org/changeset/base/304458

Log:
  Host controller is byte oriented.  Fix wrong assumption on big-endian
  systems.
  
  Pointed out by:   hselasky

Modified:
  head/sys/dev/usb/net/if_axgereg.h

Modified: head/sys/dev/usb/net/if_axgereg.h
==
--- head/sys/dev/usb/net/if_axgereg.h   Fri Aug 19 09:11:50 2016
(r304457)
+++ head/sys/dev/usb/net/if_axgereg.h   Fri Aug 19 10:51:30 2016
(r304458)
@@ -154,21 +154,15 @@ enum {
 #defineAXGE_N_FRAMES   16
 
 struct axge_frame_txhdr {
-#if BYTE_ORDER == LITTLE_ENDIAN
uint32_tlen;
-   uint32_tmss;
-#else
-   uint32_tmss;
-   uint32_tlen;
-#endif
-} __packed;
-
 #defineAXGE_TXLEN_MASK 0x0001
 #defineAXGE_VLAN_INSERT0x2000
 #defineAXGE_CSUM_DISABLE   0x8000
+   uint32_tmss;
 #defineAXGE_MSS_MASK   0x3FFF
 #defineAXGE_PADDING0x80008000
 #defineAXGE_VLAN_TAG_MASK  0x
+} __packed;
 
 #defineAXGE_TXBYTES(x) ((x) & AXGE_TXLEN_MASK)
 
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304439 - head/sys/dev/usb/net

[Hans Petter Selasky]

On 08/19/16 11:22, YongHyeon PYUN wrote:

On Fri, Aug 19, 2016 at 11:11:56AM +0200, Hans Petter Selasky wrote:

On 08/19/16 10:55, YongHyeon PYUN wrote:

I think the order is right but it was not tested on big-endian
systems.


Hi,

I'm pretty sure the ifdef is wrong, because you write the fields one at
a time, using htole32():

txhdr.mss = 0;
txhdr.len = htole32(AXGE_TXBYTES(m->m_pkthdr.len));

Big endian machines don't re-order variables like this.

You should remove the #else part.


Wouldn't USB stack pass txhdr structure without any
modification? And controller want to see len (low 32bits address)
first and then mss (high 32bits address). On big endian systems I
guess this should be reversed in host memory layout.  This is so
confusing so I could be wrong.


The USB stack passes TXHDR as-is and the host controller is byte 
oriented, not 64-bit word oriented. That's why the layout is the same as 
long as you assign per 32-bit field.


--HPS
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304187 - in head: . share/man/man4 sys/conf sys/dev/mcd sys/modules sys/modules/mcd

[Toomas Soome]

> On 19. aug 2016, at 12:24, Konstantin Belousov  wrote:
> 
> On Fri, Aug 19, 2016 at 11:50:35AM +0300, Toomas Soome wrote:
>> 
>>> On 19. aug 2016, at 10:39, Konstantin Belousov  wrote:
>>> 
>>> On Thu, Aug 18, 2016 at 09:28:57PM -0600, Warner Losh wrote:
 On Thu, Aug 18, 2016 at 12:50 AM, Julian Elischer  
 wrote:
> On 16/08/2016 4:54 AM, John Baldwin wrote:
>> 
>> On Monday, August 15, 2016 08:38:02 PM John Baldwin wrote:
>>> 
>>> Author: jhb
>>> Date: Mon Aug 15 20:38:02 2016
>>> New Revision: 304187
>>> URL: https://svnweb.freebsd.org/changeset/base/304187
>>> 
>>> Log:
>>> Remove the mcd(4) driver for Mitsumi CD-ROM players.
>>>This is a driver for a pre-ATAPI ISA CD-ROM adapter.  As noted in
>>> the manpage, this driver is only useful as a backend to cdcontrol to
>>> play audio CDs since it doesn't use DMA, so its data performance is
>>> "abysmal" (and that was true in the mid 90's).
>> 
>> No one stepped up to test patches for it either when I last posted 
>> patches
>> to
>> convert it from timeout(9) to callout(9).  I have a few more drivers that
>> are
>> both very old and that people have no business using in 12 (think ISA
>> adapters that don't do DMA and can't be used with pccard) that I will be
>> removing over the next little while.  I brought up a list of drivers on
>> arch@
>> a couple of years ago and the conversation drifted off into the weeds
>> about
>> trimming GENERIC, etc.  No one objected to the specific drivers I listed
>> though (and I got a few pleas of "please remove").  If someone shows up
>> desperately clutching an ISA adapter they can always dig up the source
>> from
>> svn and deal with forward porting it for whatever API changes have
>> happened
>> since it was removed.
> 
> 
> I would imagine any machine still holding one of these probably has not
> enough memory to run FreeBSD.
> 
> would we still run in 2MB?
 
 With insane levels of tuning, we can run in 32MB userland that can do
 things. Even 64MB is tight w/o some tuning. 16MB is almost certainly
 right out except for very specialized situations. 2MB? We can't even
 load the loader in that :(. Oh, and all these memory configs are only
 possible if you tweak the loader's block cache...
 
>>> 
>>> 32MB is quite usable.  Without any tuning, you get slightly less than 10MB
>>> for userspace, which is enough to for many things, and plenty if swap is
>>> added.
>>> 
>>> Note that you cannot boot on such configurations since loader was broken,
>>> but if you do manage to jump to kernel, things were fine several months
>>> ago.  I tested my relatively recent OOM changes on 32MB qemu config.
 Warner
>>> 
>> 
>> If the target is to go as low memory as possible, sure, you can strip all 
>> off, from boot loader point, you should load kernel from stage2 and not use 
>> loader at all (you can load and jump kernel even now from stage2, assuming 
>> it wont need any special configuration from loader config) etc etc. This 
>> means highly specialized build and has nothing to do with generic all 
>> purpose system.
>> 
> Why you describe this as an 'alternative' ? Before that loader changes,
> I regularly tested on 32MB qemy i386 image and 64MB amd64 image. I do
> not see anything extreme in these configs. They use normal boot path,
> which provides kernels with debugging symbols, metadata, loaded modules
> etc. Why should I use deficient boot2-only loading, which, additionally,
> cannot work on amd64 ?
> 
> More, this is the only reasonable way for most developers to ensure that
> system is still usable on tiny configs found on embedded devices. Right
> now the min which I have to set up is 128MB, and VM changes are simply not
> tested on anything smaller. It is guaranteed that small systems will grow
> regressions fast.  And I will not jump through the hoops to mitigate
> breakage induced by other people' changes.
> 
>> Also at some point, there is an question about how reasonable it is to have 
>> such configuration as part of generic code base for special bits like boot 
>> loader itself, as the problem is, testing all those variants is becoming 
>> impossible and even keeping reasonable code base in all of the #if #else 
>> #endif spaghetti is getting quite hard and error prone.
>> 
>>> From developers point of view, it is not really encouraging to have 
>>> possible feedback like ???oh, but you did break my 32MB system boot??? ;) 
>>> This does bring back some memories however. For first 2 unix systems I was 
>>> dealing with, one had 8MB and another had 12MB of memory??? it was ~ 
>>> 1992-1993;)
>> 
> Not mine, but you (?) indirectly broke system for people who do use 32MB
> on other arches, since low memory config on dev systems become 128MB.
> I cared about 

Re: svn commit: r304187 - in head: . share/man/man4 sys/conf sys/dev/mcd sys/modules sys/modules/mcd

[Konstantin Belousov]

On Fri, Aug 19, 2016 at 11:50:35AM +0300, Toomas Soome wrote:
> 
> > On 19. aug 2016, at 10:39, Konstantin Belousov  wrote:
> > 
> > On Thu, Aug 18, 2016 at 09:28:57PM -0600, Warner Losh wrote:
> >> On Thu, Aug 18, 2016 at 12:50 AM, Julian Elischer  
> >> wrote:
> >>> On 16/08/2016 4:54 AM, John Baldwin wrote:
>  
>  On Monday, August 15, 2016 08:38:02 PM John Baldwin wrote:
> > 
> > Author: jhb
> > Date: Mon Aug 15 20:38:02 2016
> > New Revision: 304187
> > URL: https://svnweb.freebsd.org/changeset/base/304187
> > 
> > Log:
> >  Remove the mcd(4) driver for Mitsumi CD-ROM players.
> > This is a driver for a pre-ATAPI ISA CD-ROM adapter.  As noted in
> >  the manpage, this driver is only useful as a backend to cdcontrol to
> >  play audio CDs since it doesn't use DMA, so its data performance is
> >  "abysmal" (and that was true in the mid 90's).
>  
>  No one stepped up to test patches for it either when I last posted 
>  patches
>  to
>  convert it from timeout(9) to callout(9).  I have a few more drivers that
>  are
>  both very old and that people have no business using in 12 (think ISA
>  adapters that don't do DMA and can't be used with pccard) that I will be
>  removing over the next little while.  I brought up a list of drivers on
>  arch@
>  a couple of years ago and the conversation drifted off into the weeds
>  about
>  trimming GENERIC, etc.  No one objected to the specific drivers I listed
>  though (and I got a few pleas of "please remove").  If someone shows up
>  desperately clutching an ISA adapter they can always dig up the source
>  from
>  svn and deal with forward porting it for whatever API changes have
>  happened
>  since it was removed.
> >>> 
> >>> 
> >>> I would imagine any machine still holding one of these probably has not
> >>> enough memory to run FreeBSD.
> >>> 
> >>> would we still run in 2MB?
> >> 
> >> With insane levels of tuning, we can run in 32MB userland that can do
> >> things. Even 64MB is tight w/o some tuning. 16MB is almost certainly
> >> right out except for very specialized situations. 2MB? We can't even
> >> load the loader in that :(. Oh, and all these memory configs are only
> >> possible if you tweak the loader's block cache...
> >> 
> > 
> > 32MB is quite usable.  Without any tuning, you get slightly less than 10MB
> > for userspace, which is enough to for many things, and plenty if swap is
> > added.
> > 
> > Note that you cannot boot on such configurations since loader was broken,
> > but if you do manage to jump to kernel, things were fine several months
> > ago.  I tested my relatively recent OOM changes on 32MB qemu config.
> >> Warner
> > 
> 
> If the target is to go as low memory as possible, sure, you can strip all 
> off, from boot loader point, you should load kernel from stage2 and not use 
> loader at all (you can load and jump kernel even now from stage2, assuming it 
> wont need any special configuration from loader config) etc etc. This means 
> highly specialized build and has nothing to do with generic all purpose 
> system.
> 
Why you describe this as an 'alternative' ? Before that loader changes,
I regularly tested on 32MB qemy i386 image and 64MB amd64 image. I do
not see anything extreme in these configs. They use normal boot path,
which provides kernels with debugging symbols, metadata, loaded modules
etc. Why should I use deficient boot2-only loading, which, additionally,
cannot work on amd64 ?

More, this is the only reasonable way for most developers to ensure that
system is still usable on tiny configs found on embedded devices. Right
now the min which I have to set up is 128MB, and VM changes are simply not
tested on anything smaller. It is guaranteed that small systems will grow
regressions fast.  And I will not jump through the hoops to mitigate
breakage induced by other people' changes.

> Also at some point, there is an question about how reasonable it is to have 
> such configuration as part of generic code base for special bits like boot 
> loader itself, as the problem is, testing all those variants is becoming 
> impossible and even keeping reasonable code base in all of the #if #else 
> #endif spaghetti is getting quite hard and error prone.
> 
> >From developers point of view, it is not really encouraging to have possible 
> >feedback like ???oh, but you did break my 32MB system boot??? ;) This does 
> >bring back some memories however. For first 2 unix systems I was dealing 
> >with, one had 8MB and another had 12MB of memory??? it was ~ 1992-1993;)
> 
Not mine, but you (?) indirectly broke system for people who do use 32MB
on other arches, since low memory config on dev systems become 128MB.
I cared about 32MB before, but not any longer.

> Right now the loader and stage2 are set to use 64MB heap to make it possible 
> to implement zfs feature 

Re: svn commit: r304439 - head/sys/dev/usb/net

[YongHyeon PYUN]

On Fri, Aug 19, 2016 at 11:11:56AM +0200, Hans Petter Selasky wrote:
> On 08/19/16 10:55, YongHyeon PYUN wrote:
> >I think the order is right but it was not tested on big-endian
> >systems.
> 
> Hi,
> 
> I'm pretty sure the ifdef is wrong, because you write the fields one at 
> a time, using htole32():
> 
> txhdr.mss = 0;
> txhdr.len = htole32(AXGE_TXBYTES(m->m_pkthdr.len));
> 
> Big endian machines don't re-order variables like this.
> 
> You should remove the #else part.

Wouldn't USB stack pass txhdr structure without any
modification? And controller want to see len (low 32bits address)
first and then mss (high 32bits address). On big endian systems I
guess this should be reversed in host memory layout.  This is so
confusing so I could be wrong.
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304457 - releng/11.0/usr.sbin/bsdinstall/partedit

[Dag-Erling Smørgrav]

Author: des
Date: Fri Aug 19 09:11:50 2016
New Revision: 304457
URL: https://svnweb.freebsd.org/changeset/base/304457

Log:
  MFH (r304142): ensure stripe size is non-zero multiple of 4096
  
  PR:   211361
  Approved by:  re (gjb)

Modified:
  releng/11.0/usr.sbin/bsdinstall/partedit/gpart_ops.c
Directory Properties:
  releng/11.0/   (props changed)

Modified: releng/11.0/usr.sbin/bsdinstall/partedit/gpart_ops.c
==
--- releng/11.0/usr.sbin/bsdinstall/partedit/gpart_ops.cFri Aug 19 
08:44:37 2016(r304456)
+++ releng/11.0/usr.sbin/bsdinstall/partedit/gpart_ops.cFri Aug 19 
09:11:50 2016(r304457)
@@ -795,6 +795,7 @@ gpart_max_free(struct ggeom *geom, intma
 {
struct gconfig *gc;
struct gprovider *pp, **providers;
+   intmax_t sectorsize, stripesize, offset;
intmax_t lastend;
intmax_t start, end;
intmax_t maxsize, maxstart;
@@ -845,12 +846,25 @@ gpart_max_free(struct ggeom *geom, intma
 
pp = LIST_FIRST(>lg_consumer)->lg_provider;
 
-   /* Compute beginning of new partition and maximum available space */
-   if (pp->lg_stripesize > 0 &&
-   (maxstart*pp->lg_sectorsize % pp->lg_stripesize) != 0) {
-   intmax_t offset = (pp->lg_stripesize -
-   ((maxstart*pp->lg_sectorsize) % pp->lg_stripesize)) /
-   pp->lg_sectorsize;
+   /*
+* Round the start and size of the largest available space up to
+* the nearest multiple of the adjusted stripe size.
+*
+* The adjusted stripe size is the least common multiple of the
+* actual stripe size, or the sector size if no stripe size was
+* reported, and 4096.  The reason for this is that contemporary
+* disks often have 4096-byte physical sectors but report 512
+* bytes instead for compatibility with older / broken operating
+* systems and BIOSes.  For the same reasons, virtualized storage
+* may also report a 512-byte stripe size, or none at all.
+*/
+   sectorsize = pp->lg_sectorsize;
+   if ((stripesize = pp->lg_stripesize) == 0)
+   stripesize = sectorsize;
+   while (stripesize % 4096 != 0)
+   stripesize *= 2;
+   if ((offset = maxstart * sectorsize % stripesize) != 0) {
+   offset = (stripesize - offset) / sectorsize;
maxstart += offset;
maxsize -= offset;
}
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304439 - head/sys/dev/usb/net

[Hans Petter Selasky]

On 08/19/16 10:55, YongHyeon PYUN wrote:

I think the order is right but it was not tested on big-endian
systems.


Hi,

I'm pretty sure the ifdef is wrong, because you write the fields one at 
a time, using htole32():


txhdr.mss = 0;
txhdr.len = htole32(AXGE_TXBYTES(m->m_pkthdr.len));

Big endian machines don't re-order variables like this.

You should remove the #else part.

--HPS
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304439 - head/sys/dev/usb/net

[YongHyeon PYUN]

On Fri, Aug 19, 2016 at 10:50:47AM +0200, Hans Petter Selasky wrote:
> On 08/19/16 02:50, Pyun YongHyeon wrote:
> >Modified: head/sys/dev/usb/net/if_axgereg.h
> >==
> >--- head/sys/dev/usb/net/if_axgereg.hFri Aug 19 00:03:41 2016 
> >(r304438)
> >+++ head/sys/dev/usb/net/if_axgereg.hFri Aug 19 00:50:32 2016 
> >(r304439)
> >@@ -156,19 +156,20 @@ enum {
> > struct axge_frame_txhdr {
> > #if BYTE_ORDER == LITTLE_ENDIAN
> > uint32_tlen;
> >-#define AXGE_TXLEN_MASK 0x0001
> >-#define AXGE_VLAN_INSERT0x2000
> >-#define AXGE_CSUM_DISABLE   0x8000
> > uint32_tmss;
> >-#define AXGE_MSS_MASK   0x3FFF
> >-#define AXGE_PADDING0x80008000
> >-#define AXGE_VLAN_TAG_MASK  0x
> > #else
> > uint32_tmss;
> > uint32_tlen;
> > #endif
> > } __packed;
> >
> 
> Hi,
> 
> Is it correct to switch the order of mss and len variables for 
> bit/little endian? Looks buggy to me.
> 

I think the order is right but it was not tested on big-endian
systems.
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304187 - in head: . share/man/man4 sys/conf sys/dev/mcd sys/modules sys/modules/mcd

[Toomas Soome]

> On 19. aug 2016, at 10:39, Konstantin Belousov  wrote:
> 
> On Thu, Aug 18, 2016 at 09:28:57PM -0600, Warner Losh wrote:
>> On Thu, Aug 18, 2016 at 12:50 AM, Julian Elischer  wrote:
>>> On 16/08/2016 4:54 AM, John Baldwin wrote:
 
 On Monday, August 15, 2016 08:38:02 PM John Baldwin wrote:
> 
> Author: jhb
> Date: Mon Aug 15 20:38:02 2016
> New Revision: 304187
> URL: https://svnweb.freebsd.org/changeset/base/304187
> 
> Log:
>  Remove the mcd(4) driver for Mitsumi CD-ROM players.
> This is a driver for a pre-ATAPI ISA CD-ROM adapter.  As noted in
>  the manpage, this driver is only useful as a backend to cdcontrol to
>  play audio CDs since it doesn't use DMA, so its data performance is
>  "abysmal" (and that was true in the mid 90's).
 
 No one stepped up to test patches for it either when I last posted patches
 to
 convert it from timeout(9) to callout(9).  I have a few more drivers that
 are
 both very old and that people have no business using in 12 (think ISA
 adapters that don't do DMA and can't be used with pccard) that I will be
 removing over the next little while.  I brought up a list of drivers on
 arch@
 a couple of years ago and the conversation drifted off into the weeds
 about
 trimming GENERIC, etc.  No one objected to the specific drivers I listed
 though (and I got a few pleas of "please remove").  If someone shows up
 desperately clutching an ISA adapter they can always dig up the source
 from
 svn and deal with forward porting it for whatever API changes have
 happened
 since it was removed.
>>> 
>>> 
>>> I would imagine any machine still holding one of these probably has not
>>> enough memory to run FreeBSD.
>>> 
>>> would we still run in 2MB?
>> 
>> With insane levels of tuning, we can run in 32MB userland that can do
>> things. Even 64MB is tight w/o some tuning. 16MB is almost certainly
>> right out except for very specialized situations. 2MB? We can't even
>> load the loader in that :(. Oh, and all these memory configs are only
>> possible if you tweak the loader's block cache...
>> 
> 
> 32MB is quite usable.  Without any tuning, you get slightly less than 10MB
> for userspace, which is enough to for many things, and plenty if swap is
> added.
> 
> Note that you cannot boot on such configurations since loader was broken,
> but if you do manage to jump to kernel, things were fine several months
> ago.  I tested my relatively recent OOM changes on 32MB qemu config.
>> Warner
> 

If the target is to go as low memory as possible, sure, you can strip all off, 
from boot loader point, you should load kernel from stage2 and not use loader 
at all (you can load and jump kernel even now from stage2, assuming it wont 
need any special configuration from loader config) etc etc. This means highly 
specialized build and has nothing to do with generic all purpose system.

Also at some point, there is an question about how reasonable it is to have 
such configuration as part of generic code base for special bits like boot 
loader itself, as the problem is, testing all those variants is becoming 
impossible and even keeping reasonable code base in all of the #if #else #endif 
spaghetti is getting quite hard and error prone.

From developers point of view, it is not really encouraging to have possible 
feedback like “oh, but you did break my 32MB system boot” ;) This does bring 
back some memories however. For first 2 unix systems I was dealing with, one 
had 8MB and another had 12MB of memory… it was ~ 1992-1993;)

Right now the loader and stage2 are set to use 64MB heap to make it possible to 
implement zfs feature support and later on, for more features.

Also note that UEFI setups are much harder to deal with regard of memory 
management, because as long as BS are in control, you can not really control 
the memory management there and can end up with fragmented unusable (for kernel 
loading) layout. This is especially nasty as apparently some (buggy) systems 
actually have runtime services using boot services memory areas, so you can end 
up in setup where you can not re-use BS memory and those chunks can be all over 
the low memory address space… 

rgds,
toomas
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304439 - head/sys/dev/usb/net

[Sepherosa Ziehau]

On Fri, Aug 19, 2016 at 4:50 PM, Hans Petter Selasky  wrote:
> On 08/19/16 02:50, Pyun YongHyeon wrote:
>>
>> Modified: head/sys/dev/usb/net/if_axgereg.h
>>
>> ==
>> --- head/sys/dev/usb/net/if_axgereg.h   Fri Aug 19 00:03:41 2016
>> (r304438)
>> +++ head/sys/dev/usb/net/if_axgereg.h   Fri Aug 19 00:50:32 2016
>> (r304439)
>> @@ -156,19 +156,20 @@ enum {
>>  struct axge_frame_txhdr {
>>  #if BYTE_ORDER == LITTLE_ENDIAN
>> uint32_tlen;
>> -#defineAXGE_TXLEN_MASK 0x0001
>> -#defineAXGE_VLAN_INSERT0x2000
>> -#defineAXGE_CSUM_DISABLE   0x8000
>> uint32_tmss;
>> -#defineAXGE_MSS_MASK   0x3FFF
>> -#defineAXGE_PADDING0x80008000
>> -#defineAXGE_VLAN_TAG_MASK  0x
>>  #else
>> uint32_tmss;
>> uint32_tlen;
>>  #endif
>>  } __packed;
>>
>
> Hi,
>
> Is it correct to switch the order of mss and len variables for bit/little
> endian? Looks buggy to me.

It probably is a 64bits field.  IMHO, using a 64bits field and
mask/shift-op probably will be better here.

Thanks,
sephe

-- 
Tomorrow Will Never Die
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304439 - head/sys/dev/usb/net

[Hans Petter Selasky]

On 08/19/16 02:50, Pyun YongHyeon wrote:

Modified: head/sys/dev/usb/net/if_axgereg.h
==
--- head/sys/dev/usb/net/if_axgereg.h   Fri Aug 19 00:03:41 2016
(r304438)
+++ head/sys/dev/usb/net/if_axgereg.h   Fri Aug 19 00:50:32 2016
(r304439)
@@ -156,19 +156,20 @@ enum {
 struct axge_frame_txhdr {
 #if BYTE_ORDER == LITTLE_ENDIAN
uint32_tlen;
-#defineAXGE_TXLEN_MASK 0x0001
-#defineAXGE_VLAN_INSERT0x2000
-#defineAXGE_CSUM_DISABLE   0x8000
uint32_tmss;
-#defineAXGE_MSS_MASK   0x3FFF
-#defineAXGE_PADDING0x80008000
-#defineAXGE_VLAN_TAG_MASK  0x
 #else
uint32_tmss;
uint32_tlen;
 #endif
 } __packed;



Hi,

Is it correct to switch the order of mss and len variables for 
bit/little endian? Looks buggy to me.


--HPS
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304456 - releng/11.0/libexec/rtld-elf

[Konstantin Belousov]

Author: kib
Date: Fri Aug 19 08:44:37 2016
New Revision: 304456
URL: https://svnweb.freebsd.org/changeset/base/304456

Log:
  MFC r304011:
  Remove all remaining uses of TAILQ_FOREACH_FROM() from rtld-elf.
  
  MFC r304012:
  Fill phdr and phsize for rtld object.
  
  Approved by:  re (gjb)

Modified:
  releng/11.0/libexec/rtld-elf/rtld.c
Directory Properties:
  releng/11.0/   (props changed)

Modified: releng/11.0/libexec/rtld-elf/rtld.c
==
--- releng/11.0/libexec/rtld-elf/rtld.c Fri Aug 19 08:07:04 2016
(r304455)
+++ releng/11.0/libexec/rtld-elf/rtld.c Fri Aug 19 08:44:37 2016
(r304456)
@@ -1916,6 +1916,7 @@ static void
 init_rtld(caddr_t mapbase, Elf_Auxinfo **aux_info)
 {
 Obj_Entry objtmp;  /* Temporary rtld object */
+const Elf_Ehdr *ehdr;
 const Elf_Dyn *dyn_rpath;
 const Elf_Dyn *dyn_soname;
 const Elf_Dyn *dyn_runpath;
@@ -1954,6 +1955,9 @@ init_rtld(caddr_t mapbase, Elf_Auxinfo *
 
relocate_objects(, true, , 0, NULL);
 }
+ehdr = (Elf_Ehdr *)mapbase;
+objtmp.phdr = (Elf_Phdr *)((char *)mapbase + ehdr->e_phoff);
+objtmp.phsize = ehdr->e_phnum * sizeof(objtmp.phdr[0]);
 
 /* Initialize the object list. */
 TAILQ_INIT(_list);
@@ -2164,8 +2168,7 @@ load_needed_objects(Obj_Entry *first, in
 {
 Obj_Entry *obj;
 
-obj = first;
-TAILQ_FOREACH_FROM(obj, _list, next) {
+for (obj = first; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
if (obj->marker)
continue;
if (process_needed(obj, obj->needed, flags) == -1)
@@ -2769,9 +2772,8 @@ relocate_objects(Obj_Entry *first, bool 
Obj_Entry *obj;
int error;
 
-   error = 0;
-   obj = first;
-   TAILQ_FOREACH_FROM(obj, _list, next) {
+   for (error = 0, obj = first;  obj != NULL;
+   obj = TAILQ_NEXT(obj, next)) {
if (obj->marker)
continue;
error = relocate_object(obj, bind_now, rtldobj, flags,
@@ -2811,8 +2813,7 @@ resolve_objects_ifunc(Obj_Entry *first, 
 {
Obj_Entry *obj;
 
-   obj = first;
-   TAILQ_FOREACH_FROM(obj, _list, next) {
+   for (obj = first; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
if (obj->marker)
continue;
if (resolve_object_ifunc(obj, bind_now, flags, lockstate) == -1)
@@ -4316,7 +4317,7 @@ trace_loaded_objects(Obj_Entry *obj)
 
 list_containers = getenv(_LD("TRACE_LOADED_OBJECTS_ALL"));
 
-TAILQ_FOREACH_FROM(obj, _list, next) {
+for (; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
Needed_Entry*needed;
char*name, *path;
boolis_lib;
@@ -4661,8 +4662,7 @@ allocate_tls(Obj_Entry *objs, void *oldt
 */
free_tls(oldtls, 2*sizeof(Elf_Addr), sizeof(Elf_Addr));
 } else {
-   obj = objs;
-   TAILQ_FOREACH_FROM(obj, _list, next) {
+   for (obj = objs; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
if (obj->marker || obj->tlsoffset == 0)
continue;
addr = segbase - obj->tlsoffset;
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304455 - stable/10/libexec/rtld-elf

[Konstantin Belousov]

Author: kib
Date: Fri Aug 19 08:07:04 2016
New Revision: 304455
URL: https://svnweb.freebsd.org/changeset/base/304455

Log:
  MFC r304012:
  Fill phdr and phsize for rtld object.

Modified:
  stable/10/libexec/rtld-elf/rtld.c
Directory Properties:
  stable/10/   (props changed)

Modified: stable/10/libexec/rtld-elf/rtld.c
==
--- stable/10/libexec/rtld-elf/rtld.c   Fri Aug 19 08:05:08 2016
(r304454)
+++ stable/10/libexec/rtld-elf/rtld.c   Fri Aug 19 08:07:04 2016
(r304455)
@@ -1872,6 +1872,7 @@ static void
 init_rtld(caddr_t mapbase, Elf_Auxinfo **aux_info)
 {
 Obj_Entry objtmp;  /* Temporary rtld object */
+const Elf_Ehdr *ehdr;
 const Elf_Dyn *dyn_rpath;
 const Elf_Dyn *dyn_soname;
 const Elf_Dyn *dyn_runpath;
@@ -1910,6 +1911,9 @@ init_rtld(caddr_t mapbase, Elf_Auxinfo *
 
relocate_objects(, true, , 0, NULL);
 }
+ehdr = (Elf_Ehdr *)mapbase;
+objtmp.phdr = (Elf_Phdr *)((char *)mapbase + ehdr->e_phoff);
+objtmp.phsize = ehdr->e_phnum * sizeof(objtmp.phdr[0]);
 
 /* Initialize the object list. */
 TAILQ_INIT(_list);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304454 - stable/10/libexec/rtld-elf

[Konstantin Belousov]

Author: kib
Date: Fri Aug 19 08:05:08 2016
New Revision: 304454
URL: https://svnweb.freebsd.org/changeset/base/304454

Log:
  MFC r304011:
  Remove all remaining uses of TAILQ_FOREACH_FROM() from rtld-elf.

Modified:
  stable/10/libexec/rtld-elf/rtld.c
Directory Properties:
  stable/10/   (props changed)

Modified: stable/10/libexec/rtld-elf/rtld.c
==
--- stable/10/libexec/rtld-elf/rtld.c   Fri Aug 19 08:00:48 2016
(r304453)
+++ stable/10/libexec/rtld-elf/rtld.c   Fri Aug 19 08:05:08 2016
(r304454)
@@ -2120,8 +2120,7 @@ load_needed_objects(Obj_Entry *first, in
 {
 Obj_Entry *obj;
 
-obj = first;
-TAILQ_FOREACH_FROM(obj, _list, next) {
+for (obj = first; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
if (obj->marker)
continue;
if (process_needed(obj, obj->needed, flags) == -1)
@@ -2720,9 +2719,8 @@ relocate_objects(Obj_Entry *first, bool 
Obj_Entry *obj;
int error;
 
-   error = 0;
-   obj = first;
-   TAILQ_FOREACH_FROM(obj, _list, next) {
+   for (error = 0, obj = first;  obj != NULL;
+   obj = TAILQ_NEXT(obj, next)) {
if (obj->marker)
continue;
error = relocate_object(obj, bind_now, rtldobj, flags,
@@ -2762,8 +2760,7 @@ resolve_objects_ifunc(Obj_Entry *first, 
 {
Obj_Entry *obj;
 
-   obj = first;
-   TAILQ_FOREACH_FROM(obj, _list, next) {
+   for (obj = first; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
if (obj->marker)
continue;
if (resolve_object_ifunc(obj, bind_now, flags, lockstate) == -1)
@@ -4213,7 +4210,7 @@ trace_loaded_objects(Obj_Entry *obj)
 
 list_containers = getenv(LD_ "TRACE_LOADED_OBJECTS_ALL");
 
-TAILQ_FOREACH_FROM(obj, _list, next) {
+for (; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
Needed_Entry*needed;
char*name, *path;
boolis_lib;
@@ -4557,8 +4554,7 @@ allocate_tls(Obj_Entry *objs, void *oldt
 */
free_tls(oldtls, 2*sizeof(Elf_Addr), sizeof(Elf_Addr));
 } else {
-   obj = objs;
-   TAILQ_FOREACH_FROM(obj, _list, next) {
+   for (obj = objs; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
if (obj->marker || obj->tlsoffset == 0)
continue;
addr = segbase - obj->tlsoffset;
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304453 - in stable/11: lib/libsysdecode libexec/rtld-elf

[Konstantin Belousov]

Author: kib
Date: Fri Aug 19 08:00:48 2016
New Revision: 304453
URL: https://svnweb.freebsd.org/changeset/base/304453

Log:
  MFC r304016:
  Move defines common between rtld and libsysdecode into the header.

Added:
  stable/11/libexec/rtld-elf/rtld_utrace.h
 - copied unchanged from r304016, head/libexec/rtld-elf/rtld_utrace.h
Modified:
  stable/11/lib/libsysdecode/Makefile
  stable/11/lib/libsysdecode/utrace.c
  stable/11/libexec/rtld-elf/rtld.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/lib/libsysdecode/Makefile
==
--- stable/11/lib/libsysdecode/Makefile Fri Aug 19 07:59:00 2016
(r304452)
+++ stable/11/lib/libsysdecode/Makefile Fri Aug 19 08:00:48 2016
(r304453)
@@ -9,6 +9,7 @@ SRCS=   errno.c ioctl.c syscallnames.c utr
 INCS=  sysdecode.h
 
 CFLAGS+= -I${.CURDIR}/../../sys
+CFLAGS+= -I${.CURDIR}/../../libexec/rtld-elf
 
 MAN+=  sysdecode.3 \
sysdecode_abi_to_freebsd_errno.3 \

Modified: stable/11/lib/libsysdecode/utrace.c
==
--- stable/11/lib/libsysdecode/utrace.c Fri Aug 19 07:59:00 2016
(r304452)
+++ stable/11/lib/libsysdecode/utrace.c Fri Aug 19 08:00:48 2016
(r304453)
@@ -35,33 +35,11 @@ __FBSDID("$FreeBSD$");
 #include 
 #include 
 #include 
-
-#defineUTRACE_DLOPEN_START 1
-#defineUTRACE_DLOPEN_STOP  2
-#defineUTRACE_DLCLOSE_START3
-#defineUTRACE_DLCLOSE_STOP 4
-#defineUTRACE_LOAD_OBJECT  5
-#defineUTRACE_UNLOAD_OBJECT6
-#defineUTRACE_ADD_RUNDEP   7
-#defineUTRACE_PRELOAD_FINISHED 8
-#defineUTRACE_INIT_CALL9
-#defineUTRACE_FINI_CALL10
-#defineUTRACE_DLSYM_START  11
-#defineUTRACE_DLSYM_STOP   12
-
-struct utrace_rtld {
-   char sig[4];/* 'RTLD' */
-   int event;
-   void *handle;
-   void *mapbase;
-   size_t mapsize;
-   int refcnt;
-   char name[MAXPATHLEN];
-};
+#include "rtld_utrace.h"
 
 #ifdef __LP64__
 struct utrace_rtld32 {
-   char sig[4];/* 'RTLD' */
+   char sig[4];
int event;
uint32_t handle;
uint32_t mapbase;
@@ -189,10 +167,11 @@ sysdecode_utrace(FILE *fp, void *p, size
struct utrace_malloc um;
struct utrace_malloc32 *pm;
 #endif
+   static const char rtld_utrace_sig[RTLD_UTRACE_SIG_SZ] = RTLD_UTRACE_SIG;
 
-   if (len == sizeof(struct utrace_rtld) && bcmp(p, "RTLD", 4) == 0) {
+   if (len == sizeof(struct utrace_rtld) && bcmp(p, rtld_utrace_sig,
+   sizeof(rtld_utrace_sig)) == 0)
return (print_utrace_rtld(fp, p));
-   }
 
if (len == sizeof(struct utrace_malloc)) {
print_utrace_malloc(fp, p);
@@ -200,7 +179,8 @@ sysdecode_utrace(FILE *fp, void *p, size
}
 
 #ifdef __LP64__
-   if (len == sizeof(struct utrace_rtld32) && bcmp(p, "RTLD", 4) == 0) {
+   if (len == sizeof(struct utrace_rtld32) && bcmp(p, rtld_utrace_sig,
+   sizeof(rtld_utrace_sig)) == 0) {
pr = p;
memset(, 0, sizeof(ur));
memcpy(ur.sig, pr->sig, sizeof(ur.sig));

Modified: stable/11/libexec/rtld-elf/rtld.c
==
--- stable/11/libexec/rtld-elf/rtld.c   Fri Aug 19 07:59:00 2016
(r304452)
+++ stable/11/libexec/rtld-elf/rtld.c   Fri Aug 19 08:00:48 2016
(r304453)
@@ -59,6 +59,7 @@
 #include "paths.h"
 #include "rtld_tls.h"
 #include "rtld_printf.h"
+#include "rtld_utrace.h"
 #include "notes.h"
 
 /* Types. */
@@ -273,29 +274,6 @@ char *ld_env_prefix = LD_;
 (dlp)->num_alloc = obj_count,  \
 (dlp)->num_used = 0)
 
-#defineUTRACE_DLOPEN_START 1
-#defineUTRACE_DLOPEN_STOP  2
-#defineUTRACE_DLCLOSE_START3
-#defineUTRACE_DLCLOSE_STOP 4
-#defineUTRACE_LOAD_OBJECT  5
-#defineUTRACE_UNLOAD_OBJECT6
-#defineUTRACE_ADD_RUNDEP   7
-#defineUTRACE_PRELOAD_FINISHED 8
-#defineUTRACE_INIT_CALL9
-#defineUTRACE_FINI_CALL10
-#defineUTRACE_DLSYM_START  11
-#defineUTRACE_DLSYM_STOP   12
-
-struct utrace_rtld {
-   char sig[4];/* 'RTLD' */
-   int event;
-   void *handle;
-   void *mapbase;  /* Used for 'parent' and 'init/fini' */
-   size_t mapsize;
-   int refcnt; /* Used for 'mode' */
-   char name[MAXPATHLEN];
-};
-
 #defineLD_UTRACE(e, h, mb, ms, r, n) do {  

svn commit: r304452 - stable/11/libexec/rtld-elf

[Konstantin Belousov]

Author: kib
Date: Fri Aug 19 07:59:00 2016
New Revision: 304452
URL: https://svnweb.freebsd.org/changeset/base/304452

Log:
  MFC r304012:
  Fill phdr and phsize for rtld object.

Modified:
  stable/11/libexec/rtld-elf/rtld.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/libexec/rtld-elf/rtld.c
==
--- stable/11/libexec/rtld-elf/rtld.c   Fri Aug 19 07:57:43 2016
(r304451)
+++ stable/11/libexec/rtld-elf/rtld.c   Fri Aug 19 07:59:00 2016
(r304452)
@@ -1916,6 +1916,7 @@ static void
 init_rtld(caddr_t mapbase, Elf_Auxinfo **aux_info)
 {
 Obj_Entry objtmp;  /* Temporary rtld object */
+const Elf_Ehdr *ehdr;
 const Elf_Dyn *dyn_rpath;
 const Elf_Dyn *dyn_soname;
 const Elf_Dyn *dyn_runpath;
@@ -1954,6 +1955,9 @@ init_rtld(caddr_t mapbase, Elf_Auxinfo *
 
relocate_objects(, true, , 0, NULL);
 }
+ehdr = (Elf_Ehdr *)mapbase;
+objtmp.phdr = (Elf_Phdr *)((char *)mapbase + ehdr->e_phoff);
+objtmp.phsize = ehdr->e_phnum * sizeof(objtmp.phdr[0]);
 
 /* Initialize the object list. */
 TAILQ_INIT(_list);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304451 - stable/11/libexec/rtld-elf

[Konstantin Belousov]

Author: kib
Date: Fri Aug 19 07:57:43 2016
New Revision: 304451
URL: https://svnweb.freebsd.org/changeset/base/304451

Log:
  MFC r304011:
  Remove all remaining uses of TAILQ_FOREACH_FROM() from rtld-elf.

Modified:
  stable/11/libexec/rtld-elf/rtld.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/libexec/rtld-elf/rtld.c
==
--- stable/11/libexec/rtld-elf/rtld.c   Fri Aug 19 07:56:07 2016
(r304450)
+++ stable/11/libexec/rtld-elf/rtld.c   Fri Aug 19 07:57:43 2016
(r304451)
@@ -2164,8 +2164,7 @@ load_needed_objects(Obj_Entry *first, in
 {
 Obj_Entry *obj;
 
-obj = first;
-TAILQ_FOREACH_FROM(obj, _list, next) {
+for (obj = first; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
if (obj->marker)
continue;
if (process_needed(obj, obj->needed, flags) == -1)
@@ -2769,9 +2768,8 @@ relocate_objects(Obj_Entry *first, bool 
Obj_Entry *obj;
int error;
 
-   error = 0;
-   obj = first;
-   TAILQ_FOREACH_FROM(obj, _list, next) {
+   for (error = 0, obj = first;  obj != NULL;
+   obj = TAILQ_NEXT(obj, next)) {
if (obj->marker)
continue;
error = relocate_object(obj, bind_now, rtldobj, flags,
@@ -2811,8 +2809,7 @@ resolve_objects_ifunc(Obj_Entry *first, 
 {
Obj_Entry *obj;
 
-   obj = first;
-   TAILQ_FOREACH_FROM(obj, _list, next) {
+   for (obj = first; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
if (obj->marker)
continue;
if (resolve_object_ifunc(obj, bind_now, flags, lockstate) == -1)
@@ -4316,7 +4313,7 @@ trace_loaded_objects(Obj_Entry *obj)
 
 list_containers = getenv(_LD("TRACE_LOADED_OBJECTS_ALL"));
 
-TAILQ_FOREACH_FROM(obj, _list, next) {
+for (; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
Needed_Entry*needed;
char*name, *path;
boolis_lib;
@@ -4661,8 +4658,7 @@ allocate_tls(Obj_Entry *objs, void *oldt
 */
free_tls(oldtls, 2*sizeof(Elf_Addr), sizeof(Elf_Addr));
 } else {
-   obj = objs;
-   TAILQ_FOREACH_FROM(obj, _list, next) {
+   for (obj = objs; obj != NULL; obj = TAILQ_NEXT(obj, next)) {
if (obj->marker || obj->tlsoffset == 0)
continue;
addr = segbase - obj->tlsoffset;
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304450 - stable/11/lib/libsysdecode

[Konstantin Belousov]

Author: kib
Date: Fri Aug 19 07:56:07 2016
New Revision: 304450
URL: https://svnweb.freebsd.org/changeset/base/304450

Log:
  MFC r303991:
  Decode 32bit utrace records on the 64bit host.

Modified:
  stable/11/lib/libsysdecode/utrace.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/lib/libsysdecode/utrace.c
==
--- stable/11/lib/libsysdecode/utrace.c Fri Aug 19 07:48:32 2016
(r304449)
+++ stable/11/lib/libsysdecode/utrace.c Fri Aug 19 07:56:07 2016
(r304450)
@@ -33,7 +33,7 @@ __FBSDID("$FreeBSD$");
 #include 
 #include 
 #include 
-#include 
+#include 
 #include 
 
 #defineUTRACE_DLOPEN_START 1
@@ -59,6 +59,18 @@ struct utrace_rtld {
char name[MAXPATHLEN];
 };
 
+#ifdef __LP64__
+struct utrace_rtld32 {
+   char sig[4];/* 'RTLD' */
+   int event;
+   uint32_t handle;
+   uint32_t mapbase;
+   uint32_t mapsize;
+   int refcnt;
+   char name[MAXPATHLEN];
+};
+#endif
+
 static int
 print_utrace_rtld(FILE *fp, void *p)
 {
@@ -145,6 +157,14 @@ struct utrace_malloc {
void *r;
 };
 
+#ifdef __LP64__
+struct utrace_malloc32 {
+   uint32_t p;
+   uint32_t s;
+   uint32_t r;
+};
+#endif
+
 static void
 print_utrace_malloc(FILE *fp, void *p)
 {
@@ -163,6 +183,12 @@ print_utrace_malloc(FILE *fp, void *p)
 int
 sysdecode_utrace(FILE *fp, void *p, size_t len)
 {
+#ifdef __LP64__
+   struct utrace_rtld ur;
+   struct utrace_rtld32 *pr;
+   struct utrace_malloc um;
+   struct utrace_malloc32 *pm;
+#endif
 
if (len == sizeof(struct utrace_rtld) && bcmp(p, "RTLD", 4) == 0) {
return (print_utrace_rtld(fp, p));
@@ -172,6 +198,32 @@ sysdecode_utrace(FILE *fp, void *p, size
print_utrace_malloc(fp, p);
return (1);
}
-   
+
+#ifdef __LP64__
+   if (len == sizeof(struct utrace_rtld32) && bcmp(p, "RTLD", 4) == 0) {
+   pr = p;
+   memset(, 0, sizeof(ur));
+   memcpy(ur.sig, pr->sig, sizeof(ur.sig));
+   ur.event = pr->event;
+   ur.handle = (void *)(uintptr_t)pr->handle;
+   ur.mapbase = (void *)(uintptr_t)pr->mapbase;
+   ur.mapsize = pr->mapsize;
+   ur.refcnt = pr->refcnt;
+   memcpy(ur.name, pr->name, sizeof(ur.name));
+   return (print_utrace_rtld(fp, ));
+   }
+
+   if (len == sizeof(struct utrace_malloc32)) {
+   pm = p;
+   memset(, 0, sizeof(um));
+   um.p = pm->p == (uint32_t)-1 ? (void *)(intptr_t)-1 :
+   (void *)(uintptr_t)pm->p;
+   um.s = pm->s;
+   um.r = (void *)(uintptr_t)pm->r;
+   print_utrace_malloc(fp, );
+   return (1);
+   }
+#endif
+
return (0);
 }
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304449 - stable/11/usr.bin/kdump

[Konstantin Belousov]

Author: kib
Date: Fri Aug 19 07:48:32 2016
New Revision: 304449
URL: https://svnweb.freebsd.org/changeset/base/304449

Log:
  MFC r303990:
  Remove unused prototypes.

Modified:
  stable/11/usr.bin/kdump/kdump.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/usr.bin/kdump/kdump.c
==
--- stable/11/usr.bin/kdump/kdump.c Fri Aug 19 07:05:34 2016
(r304448)
+++ stable/11/usr.bin/kdump/kdump.c Fri Aug 19 07:48:32 2016
(r304449)
@@ -105,8 +105,6 @@ void ktrgenio(struct ktr_genio *, int);
 void ktrpsig(struct ktr_psig *);
 void ktrcsw(struct ktr_csw *);
 void ktrcsw_old(struct ktr_csw_old *);
-void ktruser_malloc(void *);
-void ktruser_rtld(int, void *);
 void ktruser(int, void *);
 void ktrcaprights(cap_rights_t *);
 void ktritimerval(struct itimerval *it);
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304187 - in head: . share/man/man4 sys/conf sys/dev/mcd sys/modules sys/modules/mcd

[Konstantin Belousov]

On Thu, Aug 18, 2016 at 09:28:57PM -0600, Warner Losh wrote:
> On Thu, Aug 18, 2016 at 12:50 AM, Julian Elischer  wrote:
> > On 16/08/2016 4:54 AM, John Baldwin wrote:
> >>
> >> On Monday, August 15, 2016 08:38:02 PM John Baldwin wrote:
> >>>
> >>> Author: jhb
> >>> Date: Mon Aug 15 20:38:02 2016
> >>> New Revision: 304187
> >>> URL: https://svnweb.freebsd.org/changeset/base/304187
> >>>
> >>> Log:
> >>>Remove the mcd(4) driver for Mitsumi CD-ROM players.
> >>>   This is a driver for a pre-ATAPI ISA CD-ROM adapter.  As noted in
> >>>the manpage, this driver is only useful as a backend to cdcontrol to
> >>>play audio CDs since it doesn't use DMA, so its data performance is
> >>>"abysmal" (and that was true in the mid 90's).
> >>
> >> No one stepped up to test patches for it either when I last posted patches
> >> to
> >> convert it from timeout(9) to callout(9).  I have a few more drivers that
> >> are
> >> both very old and that people have no business using in 12 (think ISA
> >> adapters that don't do DMA and can't be used with pccard) that I will be
> >> removing over the next little while.  I brought up a list of drivers on
> >> arch@
> >> a couple of years ago and the conversation drifted off into the weeds
> >> about
> >> trimming GENERIC, etc.  No one objected to the specific drivers I listed
> >> though (and I got a few pleas of "please remove").  If someone shows up
> >> desperately clutching an ISA adapter they can always dig up the source
> >> from
> >> svn and deal with forward porting it for whatever API changes have
> >> happened
> >> since it was removed.
> >
> >
> > I would imagine any machine still holding one of these probably has not
> > enough memory to run FreeBSD.
> >
> > would we still run in 2MB?
> 
> With insane levels of tuning, we can run in 32MB userland that can do
> things. Even 64MB is tight w/o some tuning. 16MB is almost certainly
> right out except for very specialized situations. 2MB? We can't even
> load the loader in that :(. Oh, and all these memory configs are only
> possible if you tweak the loader's block cache...
> 

32MB is quite usable.  Without any tuning, you get slightly less than 10MB
for userspace, which is enough to for many things, and plenty if swap is
added.

Note that you cannot boot on such configurations since loader was broken,
but if you do manage to jump to kernel, things were fine several months
ago.  I tested my relatively recent OOM changes on 32MB qemu config.
> Warner
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

svn commit: r304448 - stable/11/usr.sbin/bsdinstall/partedit

[Dag-Erling Smørgrav]

Author: des
Date: Fri Aug 19 07:05:34 2016
New Revision: 304448
URL: https://svnweb.freebsd.org/changeset/base/304448

Log:
  MFH (r304142): ensure stripe size is non-zero multiple of 4096
  
  PR:   211361

Modified:
  stable/11/usr.sbin/bsdinstall/partedit/gpart_ops.c
Directory Properties:
  stable/11/   (props changed)

Modified: stable/11/usr.sbin/bsdinstall/partedit/gpart_ops.c
==
--- stable/11/usr.sbin/bsdinstall/partedit/gpart_ops.c  Fri Aug 19 05:43:28 
2016(r304447)
+++ stable/11/usr.sbin/bsdinstall/partedit/gpart_ops.c  Fri Aug 19 07:05:34 
2016(r304448)
@@ -795,6 +795,7 @@ gpart_max_free(struct ggeom *geom, intma
 {
struct gconfig *gc;
struct gprovider *pp, **providers;
+   intmax_t sectorsize, stripesize, offset;
intmax_t lastend;
intmax_t start, end;
intmax_t maxsize, maxstart;
@@ -845,12 +846,25 @@ gpart_max_free(struct ggeom *geom, intma
 
pp = LIST_FIRST(>lg_consumer)->lg_provider;
 
-   /* Compute beginning of new partition and maximum available space */
-   if (pp->lg_stripesize > 0 &&
-   (maxstart*pp->lg_sectorsize % pp->lg_stripesize) != 0) {
-   intmax_t offset = (pp->lg_stripesize -
-   ((maxstart*pp->lg_sectorsize) % pp->lg_stripesize)) /
-   pp->lg_sectorsize;
+   /*
+* Round the start and size of the largest available space up to
+* the nearest multiple of the adjusted stripe size.
+*
+* The adjusted stripe size is the least common multiple of the
+* actual stripe size, or the sector size if no stripe size was
+* reported, and 4096.  The reason for this is that contemporary
+* disks often have 4096-byte physical sectors but report 512
+* bytes instead for compatibility with older / broken operating
+* systems and BIOSes.  For the same reasons, virtualized storage
+* may also report a 512-byte stripe size, or none at all.
+*/
+   sectorsize = pp->lg_sectorsize;
+   if ((stripesize = pp->lg_stripesize) == 0)
+   stripesize = sectorsize;
+   while (stripesize % 4096 != 0)
+   stripesize *= 2;
+   if ((offset = maxstart * sectorsize % stripesize) != 0) {
+   offset = (stripesize - offset) / sectorsize;
maxstart += offset;
maxsize -= offset;
}
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304142 - head/usr.sbin/bsdinstall/partedit

[Dag-Erling Smørgrav]

Warner Losh  writes:
> Allan Jude  writes:
> > Which makes more sense:
> >
> > A) If stripesize == 0, use some sane value like 4096
>
> I don't like this.
>
> > B) Some other combination that uses the reported stripe size, unless it
> > is 0, in which case it uses 4096 (or some other value controlled by a
> > different new sysctl)
>
> Don't like this so much.
>
> > C) create kern.geom.min_stripe_size with a default of 512, but users can
> > set 4096 if they use only 4k devices. (doesn't really solve the problem
> > for the installer)
>
> Default it to 4k, and allow users to set it to 512. If the drive
> reports < this value
> report this value instead.

I don't like either option.  Option D (which I don't like either, but
which should at least work in most cases) is a sysctl that specifies a
minimum factor, and set the reported stripe size to the least common
multiple of that number and the actual stripe or sector size.  This is
what my bsdinstall patch does.  However, I think that pushing this down
to a layer where it will affect all applications is a terrible idea,
because we have no way of knowing what will break[*], and it can
seriously mislead users and hinder troubleshooting - especially if it is
enabled by default rather than only when necessary.

I don't think it's a good idea to enforce stripe alignment everywhere,
either.  It works for partitions because they are very large relative to
the stripe size, and at worst we will waste a few millionths of the disk
on inter-partition gaps, which should only occur between the partition
table and the boot partition, and possibly, if the stripe size is very
large, between the boot partition and the swap partition.  But forcing
filesystems to respect the stripe size will lead to no end of trouble,
because RAID volumes can have stripe sizes of 16 kB or more.  I think it
is important to align partitions during installation because of the huge
performance impact of misaligned partitions on AF disks, but despite
what Nathan claims, I never advocated applying the same logic
everywhere.

[*] Apart from audio CDs.  We already know that it will break those.

DES
-- 
Dag-Erling Smørgrav - d...@des.no
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r304142 - head/usr.sbin/bsdinstall/partedit

[Glen Barber]

On Fri, Aug 19, 2016 at 08:21:49AM +0200, Dag-Erling Smørgrav wrote:
> Nathan Whitehorn  writes:
> > This is ridiculous. I've asked a series of technical questions about
> > generalizing a patch you made and that I think is a good idea. In
> > response, those questions have been met with a non-stop torrent of
> > insults and abuse instead of answers, with only one eventual nugget of
> > information in response to one of them -- that you ran into problems
> > with mfid -- to redeem it. I'm done with this discussion.
> 
> I suggest you go and read through the discussion again.  I gave you
> plenty of information, mentioning LSI and hypervisors, for instance, and
> even, as I grew increasingly frustrated by your stonewalling, showed you
> how to inspect the code to confirm that most RAID drivers do not even
> try to report a stripe size.  Meanwhile, you steadfastedly refused to
> address my arguments, simply repeating your position again and again,
> and insisting that the fact that you have never encountered any systems
> that get the stripe size wrong (which I don't believe, but I'm willing
> to believe that you've never noticed) invalidates the fact that I have.
> Not to mention your steadfast defense of the non-existent purity of the
> installer.  When Warner finally intervened, you switched to setting up
> strawmen, asking me to pick and defend one of two positions which I've
> never held.  I'm not going to play that game.
> 
> > Hopefully we can have a real conversation about this at this some
> > point after the release.
> 
> Likewise, but I'm not holding my breath.
> 

I'll mediate this then.  Not right now.  I think everyone needs to step
back for a bit.

Glen



signature.asc
Description: PGP signature

Re: svn commit: r304142 - head/usr.sbin/bsdinstall/partedit

[Dag-Erling Smørgrav]

Nathan Whitehorn  writes:
> This is ridiculous. I've asked a series of technical questions about
> generalizing a patch you made and that I think is a good idea. In
> response, those questions have been met with a non-stop torrent of
> insults and abuse instead of answers, with only one eventual nugget of
> information in response to one of them -- that you ran into problems
> with mfid -- to redeem it. I'm done with this discussion.

I suggest you go and read through the discussion again.  I gave you
plenty of information, mentioning LSI and hypervisors, for instance, and
even, as I grew increasingly frustrated by your stonewalling, showed you
how to inspect the code to confirm that most RAID drivers do not even
try to report a stripe size.  Meanwhile, you steadfastedly refused to
address my arguments, simply repeating your position again and again,
and insisting that the fact that you have never encountered any systems
that get the stripe size wrong (which I don't believe, but I'm willing
to believe that you've never noticed) invalidates the fact that I have.
Not to mention your steadfast defense of the non-existent purity of the
installer.  When Warner finally intervened, you switched to setting up
strawmen, asking me to pick and defend one of two positions which I've
never held.  I'm not going to play that game.

> Hopefully we can have a real conversation about this at this some
> point after the release.

Likewise, but I'm not holding my breath.

DES
-- 
Dag-Erling Smørgrav - d...@des.no
___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Re: svn commit: r298950 - head/sys/dev/pci

[Harry Schmalzbauer]

 Bezüglich John Baldwin's Nachricht vom 03.05.2016 02:35 (localtime):
> Author: jhb
> Date: Tue May  3 00:35:11 2016
> New Revision: 298950
> URL: https://svnweb.freebsd.org/changeset/base/298950
>
> Log:
>   Fix an off by one error when remapping MSI-X vectors.
>   
>   pci_remap_msix() can be used to alter the mapping of allocated
>   MSI-X vectors to the MSI-X table.  The code had an off by one error
>   when adding the IRQ resources after performing a remap.  This was
>   fatal for any vectors in the table that used the "last" valid IRQ as
>   those vectors were assigned a garbage IRQ value.
>   
>   MFC after:  3 days

Was this superseded by any other merge?
As far as I saw, it's missing in stable/10?

Thanks,

-Harry

___
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"