Re: [PATCH v7 07/19] xen: implement hook to fetch e820 memory map

2013-12-21 Thread Konstantin Belousov
This also looks fine, with the style nit noted in the patch.

On Thu, Dec 19, 2013 at 07:54:44PM +0100, Roger Pau Monne wrote:
 ---
  sys/amd64/amd64/machdep.c   |   50 ++
  sys/amd64/include/pc/bios.h |2 +
  sys/amd64/include/sysarch.h |1 +
  sys/x86/xen/pv.c|   26 ++
  4 files changed, 60 insertions(+), 19 deletions(-)
 
 diff --git a/sys/amd64/amd64/machdep.c b/sys/amd64/amd64/machdep.c
 index a2dcb90..6bbfe5a 100644
 --- a/sys/amd64/amd64/machdep.c
 +++ b/sys/amd64/amd64/machdep.c
 @@ -177,11 +177,15 @@ SYSINIT(cpu, SI_SUB_CPU, SI_ORDER_FIRST, cpu_startup, 
 NULL);
  /* Preload data parse function */
  static caddr_t native_parse_preload_data(u_int64_t);
  
 +/* Native function to fetch and parse the e820 map */
 +static void native_parse_memmap(caddr_t, vm_paddr_t *, int *);
 +
  /* Default init_ops implementation. */
  struct init_ops init_ops = {
   .parse_preload_data =   native_parse_preload_data,
   .early_delay_init = i8254_init,
   .early_delay =  i8254_delay,
 + .parse_memmap = native_parse_memmap,
  };
  
  /*
 @@ -1418,21 +1422,12 @@ add_physmap_entry(uint64_t base, uint64_t length, 
 vm_paddr_t *physmap,
   return (1);
  }
  
 -static void
 -add_smap_entries(struct bios_smap *smapbase, vm_paddr_t *physmap,
 -int *physmap_idx)
 +void
 +bios_add_smap_entries(struct bios_smap *smapbase, u_int32_t smapsize,
 +  vm_paddr_t *physmap, int *physmap_idx)
  {
   struct bios_smap *smap, *smapend;
 - u_int32_t smapsize;
  
 - /*
 -  * Memory map from INT 15:E820.
 -  *
 -  * subr_module.c says:
 -  * Consumer may safely assume that size value precedes data.
 -  * ie: an int32_t immediately precedes smap.
 -  */
 - smapsize = *((u_int32_t *)smapbase - 1);
   smapend = (struct bios_smap *)((uintptr_t)smapbase + smapsize);
  
   for (smap = smapbase; smap  smapend; smap++) {
 @@ -1449,6 +1444,29 @@ add_smap_entries(struct bios_smap *smapbase, 
 vm_paddr_t *physmap,
   }
  }
  
 +static void
 +native_parse_memmap(caddr_t kmdp, vm_paddr_t *physmap, int *physmap_idx)
 +{
 + struct bios_smap *smap;
 + u_int32_t size;
 +
 + /*
 +  * Memory map from INT 15:E820.
 +  *
 +  * subr_module.c says:
 +  * Consumer may safely assume that size value precedes data.
 +  * ie: an int32_t immediately precedes smap.
 +  */
 +
 + smap = (struct bios_smap *)preload_search_info(kmdp,
 + MODINFO_METADATA | MODINFOMD_SMAP);
 + if (smap == NULL)
 + panic(No BIOS smap info from loader!);
 + size = *((u_int32_t *)smap - 1);
 +
 + bios_add_smap_entries(smap, size, physmap, physmap_idx);
 +}
 +
  /*
   * Populate the (physmap) array with base/bound pairs describing the
   * available physical memory in the system, then test this memory and
 @@ -1466,19 +1484,13 @@ getmemsize(caddr_t kmdp, u_int64_t first)
   vm_paddr_t pa, physmap[PHYSMAP_SIZE];
   u_long physmem_start, physmem_tunable, memtest;
   pt_entry_t *pte;
 - struct bios_smap *smapbase;
   quad_t dcons_addr, dcons_size;
  
   bzero(physmap, sizeof(physmap));
   basemem = 0;
   physmap_idx = 0;
  
 - smapbase = (struct bios_smap *)preload_search_info(kmdp,
 - MODINFO_METADATA | MODINFOMD_SMAP);
 - if (smapbase == NULL)
 - panic(No BIOS smap info from loader!);
 -
 - add_smap_entries(smapbase, physmap, physmap_idx);
 + init_ops.parse_memmap(kmdp, physmap, physmap_idx);
  
   /*
* Find the 'base memory' segment for SMP
 diff --git a/sys/amd64/include/pc/bios.h b/sys/amd64/include/pc/bios.h
 index e7d568e..92d4265 100644
 --- a/sys/amd64/include/pc/bios.h
 +++ b/sys/amd64/include/pc/bios.h
 @@ -106,6 +106,8 @@ struct bios_oem {
  int  bios_oem_strings(struct bios_oem *oem, u_char *buffer, size_t maxlen);
  uint32_t bios_sigsearch(uint32_t start, u_char *sig, int siglen, int paralen,
   int sigofs);
 +void bios_add_smap_entries(struct bios_smap *smapbase, u_int32_t smapsize,
 +   vm_paddr_t *physmap, int *physmap_idx);
This uses the GNU formatting conventions, please indent the continuation
line same as the previous line.


pgpT69W0qSlA3.pgp
Description: PGP signature


Re: [PATCH v7 10/19] xen: add hook for AP bootstrap memory reservation

2013-12-21 Thread Konstantin Belousov
On Thu, Dec 19, 2013 at 07:54:47PM +0100, Roger Pau Monne wrote:
 This hook will only be implemented for bare metal, Xen doesn't require
 any bootstrap code since APs are started in long mode with paging
 enabled.
 ---
  sys/amd64/amd64/machdep.c   |6 +-
  sys/amd64/include/sysarch.h |3 +++
  2 files changed, 8 insertions(+), 1 deletions(-)
 
 diff --git a/sys/amd64/amd64/machdep.c b/sys/amd64/amd64/machdep.c
 index 6bbfe5a..a811a9b 100644
 --- a/sys/amd64/amd64/machdep.c
 +++ b/sys/amd64/amd64/machdep.c
 @@ -186,6 +186,9 @@ struct init_ops init_ops = {
   .early_delay_init = i8254_init,
   .early_delay =  i8254_delay,
   .parse_memmap = native_parse_memmap,
 +#ifdef SMP
 + .mp_bootaddress =   mp_bootaddress,
 +#endif
  };
  
  /*
 @@ -1507,7 +1510,8 @@ getmemsize(caddr_t kmdp, u_int64_t first)
  
  #ifdef SMP
   /* make hole for AP bootstrap code */
 - physmap[1] = mp_bootaddress(physmap[1] / 1024);
 + if (init_ops.mp_bootaddress)
 + physmap[1] = init_ops.mp_bootaddress(physmap[1] / 1024);
  #endif
  
   /*
 diff --git a/sys/amd64/include/sysarch.h b/sys/amd64/include/sysarch.h
 index 084223e..77f4b29 100644
 --- a/sys/amd64/include/sysarch.h
 +++ b/sys/amd64/include/sysarch.h
 @@ -16,6 +16,9 @@ struct init_ops {
   void(*early_delay_init)(void);
   void(*early_delay)(int);
   void(*parse_memmap)(caddr_t, vm_paddr_t *, int *);
 +#ifdef SMP
 + u_int   (*mp_bootaddress)(u_int);
 +#endif
  };
  
  extern struct init_ops init_ops;

I suggest to make the init_ops contain the mp_bootaddress unconditionally,
instead of making it depended on SMP.  The #ifdef makes the kernel binary
interface fragile and depended on the config, which would cause problems
in future if any module needs any interation with init_ops.


pgpvtMJQoKY9a.pgp
Description: PGP signature


PACKAGESITE spam

2013-12-21 Thread dt71

I've just installed a very recent -CURRENT, and now I'm performing a big 
portupgrade procedure. I get the following message spammed a lot:

pkg: PACKAGESITE in pkg.conf is deprecated. Please create a repository 
configuration file
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Steve Kargl
On Sat, Dec 21, 2013 at 07:35:56PM +0100, d...@gmx.com wrote:
 I've just installed a very recent -CURRENT, and now I'm performing a big 
 portupgrade procedure. I get the following message spammed a lot:
 
 pkg: PACKAGESITE in pkg.conf is deprecated. Please create a repository 
 configuration file

Yeah, I noticed that spam along with the spam that is being
spewed to /var/log/messaage, e.g., 

Dec 21 10:27:28 laptop-kargl pkg-static: libwpg-0.2.2 installed
Dec 21 10:31:15 laptop-kargl pkg-static: libcdr-0.0.14 installed
Dec 21 10:32:35 laptop-kargl pkg-static: openjpeg-1.5.0_2 installed
Dec 21 10:38:33 laptop-kargl pkg-static: poppler-data-0.4.6 installed
Dec 21 10:39:48 laptop-kargl pkg-static: poppler-0.22.2 installed
Dec 21 10:40:32 laptop-kargl pkg-static: ilmbase-2.1.0 installed
Dec 21 10:44:28 laptop-kargl pkg-static: OpenEXR-2.1.0_1 installed
Dec 21 10:47:36 laptop-kargl pkg-static: vigra-1.9.0_4 installed
Dec 21 10:51:00 laptop-kargl pkg-static: lp_solve-5.5.2.0 installed

Can you (portmgr) please mute these messages?

-- 
Steve
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Glen Barber
On Sat, Dec 21, 2013 at 12:05:38PM -0800, Steve Kargl wrote:
 On Sat, Dec 21, 2013 at 07:35:56PM +0100, d...@gmx.com wrote:
  I've just installed a very recent -CURRENT, and now I'm performing a big 
  portupgrade procedure. I get the following message spammed a lot:
  
  pkg: PACKAGESITE in pkg.conf is deprecated. Please create a repository 
  configuration file
 
 Yeah, I noticed that spam along with the spam that is being
 spewed to /var/log/messaage, e.g., 
 
 Dec 21 10:27:28 laptop-kargl pkg-static: libwpg-0.2.2 installed
 Dec 21 10:31:15 laptop-kargl pkg-static: libcdr-0.0.14 installed
 Dec 21 10:32:35 laptop-kargl pkg-static: openjpeg-1.5.0_2 installed
 Dec 21 10:38:33 laptop-kargl pkg-static: poppler-data-0.4.6 installed
 Dec 21 10:39:48 laptop-kargl pkg-static: poppler-0.22.2 installed
 Dec 21 10:40:32 laptop-kargl pkg-static: ilmbase-2.1.0 installed
 Dec 21 10:44:28 laptop-kargl pkg-static: OpenEXR-2.1.0_1 installed
 Dec 21 10:47:36 laptop-kargl pkg-static: vigra-1.9.0_4 installed
 Dec 21 10:51:00 laptop-kargl pkg-static: lp_solve-5.5.2.0 installed
 
 Can you (portmgr) please mute these messages?
 

echo 'SYSLOG: no'  /usr/local/etc/pkg.conf

Glen



pgpcL2CgmtLq7.pgp
Description: PGP signature


Re: PACKAGESITE spam

2013-12-21 Thread Steve Kargl
On Sat, Dec 21, 2013 at 03:10:26PM -0500, Glen Barber wrote:
 On Sat, Dec 21, 2013 at 12:05:38PM -0800, Steve Kargl wrote:
  On Sat, Dec 21, 2013 at 07:35:56PM +0100, d...@gmx.com wrote:
   I've just installed a very recent -CURRENT, and now I'm performing a big 
   portupgrade procedure. I get the following message spammed a lot:
   
   pkg: PACKAGESITE in pkg.conf is deprecated. Please create a repository 
   configuration file
  
  Yeah, I noticed that spam along with the spam that is being
  spewed to /var/log/messaage, e.g., 
  
  Dec 21 10:27:28 laptop-kargl pkg-static: libwpg-0.2.2 installed
  Dec 21 10:44:28 laptop-kargl pkg-static: OpenEXR-2.1.0_1 installed
  Dec 21 10:47:36 laptop-kargl pkg-static: vigra-1.9.0_4 installed
  Dec 21 10:51:00 laptop-kargl pkg-static: lp_solve-5.5.2.0 installed
  
  Can you (portmgr) please mute these messages?
  
 
 echo 'SYSLOG: no'  /usr/local/etc/pkg.conf
 

It did not ask how to stop this stupidity.  I asked to have
this stupidity stopped by default.  The spewing of this 
information in /var/log/messages provides NOTHING.  Please
turn it off by default.

-- 
Steve
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Julien Laffaye

On 12/21/2013 9:14 PM, Steve Kargl wrote:

On Sat, Dec 21, 2013 at 03:10:26PM -0500, Glen Barber wrote:

On Sat, Dec 21, 2013 at 12:05:38PM -0800, Steve Kargl wrote:

On Sat, Dec 21, 2013 at 07:35:56PM +0100, d...@gmx.com wrote:

I've just installed a very recent -CURRENT, and now I'm performing a big 
portupgrade procedure. I get the following message spammed a lot:

pkg: PACKAGESITE in pkg.conf is deprecated. Please create a repository 
configuration file

Yeah, I noticed that spam along with the spam that is being
spewed to /var/log/messaage, e.g.,

Dec 21 10:27:28 laptop-kargl pkg-static: libwpg-0.2.2 installed
Dec 21 10:44:28 laptop-kargl pkg-static: OpenEXR-2.1.0_1 installed
Dec 21 10:47:36 laptop-kargl pkg-static: vigra-1.9.0_4 installed
Dec 21 10:51:00 laptop-kargl pkg-static: lp_solve-5.5.2.0 installed

Can you (portmgr) please mute these messages?


echo 'SYSLOG: no'  /usr/local/etc/pkg.conf


It did not ask how to stop this stupidity.  I asked to have
this stupidity stopped by default.  The spewing of this
information in /var/log/messages provides NOTHING.  Please
turn it off by default.



Obviously, if it is on by default, others have different opinions on 
this subject...

___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Greg Rivers

On Sat, 21 Dec 2013, Steve Kargl wrote:


On Sat, Dec 21, 2013 at 03:10:26PM -0500, Glen Barber wrote:

On Sat, Dec 21, 2013 at 12:05:38PM -0800, Steve Kargl wrote:

On Sat, Dec 21, 2013 at 07:35:56PM +0100, d...@gmx.com wrote:
I've just installed a very recent -CURRENT, and now I'm performing a 
big portupgrade procedure. I get the following message spammed a lot:


pkg: PACKAGESITE in pkg.conf is deprecated. Please create a 
repository configuration file


Yeah, I noticed that spam along with the spam that is being spewed to 
/var/log/messaage, e.g.,


Dec 21 10:27:28 laptop-kargl pkg-static: libwpg-0.2.2 installed
Dec 21 10:44:28 laptop-kargl pkg-static: OpenEXR-2.1.0_1 installed
Dec 21 10:47:36 laptop-kargl pkg-static: vigra-1.9.0_4 installed
Dec 21 10:51:00 laptop-kargl pkg-static: lp_solve-5.5.2.0 installed

Can you (portmgr) please mute these messages?



echo 'SYSLOG: no'  /usr/local/etc/pkg.conf



It did not ask how to stop this stupidity.  I asked to have this 
stupidity stopped by default.  The spewing of this information in 
/var/log/messages provides NOTHING.  Please turn it off by default.




Do you really feel that strongly about it?  Having a record of changes to 
the system has always seemed like a feature to me...


--
Greg
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Steve Kargl
On Sat, Dec 21, 2013 at 02:54:39PM -0600, Greg Rivers wrote:
 On Sat, 21 Dec 2013, Steve Kargl wrote:
 
  It did not ask how to stop this stupidity.  I asked to have this 
  stupidity stopped by default.  The spewing of this information in 
  /var/log/messages provides NOTHING.  Please turn it off by default.
 
 
 Do you really feel that strongly about it?  Having a record of changes to 
 the system has always seemed like a feature to me...
 

Yes, I do feel strongly about it.  It is completely unnecesary noise.
It should be off by default.  If someone wants to fill /var up with
useless information, then that someone can turn on the noise.

-- 
Steve
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


[head tinderbox] failure on i386/i386

2013-12-21 Thread FreeBSD Tinderbox
TB --- 2013-12-21 20:50:20 - tinderbox 2.20 running on freebsd-current.sentex.ca
TB --- 2013-12-21 20:50:20 - FreeBSD freebsd-current.sentex.ca 8.3-PRERELEASE 
FreeBSD 8.3-PRERELEASE #0: Mon Mar 26 13:54:12 EDT 2012 
d...@freebsd-current.sentex.ca:/usr/obj/usr/src/sys/GENERIC  amd64
TB --- 2013-12-21 20:50:20 - starting HEAD tinderbox run for i386/i386
TB --- 2013-12-21 20:50:20 - cleaning the object tree
TB --- 2013-12-21 20:50:20 - /usr/local/bin/svn stat /src
TB --- 2013-12-21 20:50:25 - At svn revision 259696
TB --- 2013-12-21 20:50:26 - building world
TB --- 2013-12-21 20:50:26 - CROSS_BUILD_TESTING=YES
TB --- 2013-12-21 20:50:26 - MAKEOBJDIRPREFIX=/obj
TB --- 2013-12-21 20:50:26 - PATH=/usr/bin:/usr/sbin:/bin:/sbin
TB --- 2013-12-21 20:50:26 - SRCCONF=/dev/null
TB --- 2013-12-21 20:50:26 - TARGET=i386
TB --- 2013-12-21 20:50:26 - TARGET_ARCH=i386
TB --- 2013-12-21 20:50:26 - TZ=UTC
TB --- 2013-12-21 20:50:26 - __MAKE_CONF=/dev/null
TB --- 2013-12-21 20:50:26 - cd /src
TB --- 2013-12-21 20:50:26 - /usr/bin/make -B buildworld
 Building an up-to-date make(1)
 World build started on Sat Dec 21 20:50:34 UTC 2013
 Rebuilding the temporary build tree
 stage 1.1: legacy release compatibility shims
 stage 1.2: bootstrap tools
 stage 2.1: cleaning up the object tree
 stage 2.2: rebuilding the object tree
 stage 2.3: build tools
 stage 3: cross tools
[...]
c++  -O2 -pipe -I/src/lib/clang/libclangsema/../../../contrib/llvm/include 
-I/src/lib/clang/libclangsema/../../../contrib/llvm/tools/clang/include 
-I/src/lib/clang/libclangsema/../../../contrib/llvm/tools/clang/lib/Sema -I. 
-I/src/lib/clang/libclangsema/../../../contrib/llvm/../../lib/clang/include 
-DLLVM_ON_UNIX -DLLVM_ON_FREEBSD -D__STDC_LIMIT_MACROS -D__STDC_CONSTANT_MACROS 
-fno-strict-aliasing -DLLVM_DEFAULT_TARGET_TRIPLE=\i386-unknown-freebsd11.0\ 
-DLLVM_HOST_TRIPLE=\x86_64-unknown-freebsd11.0\ 
-DDEFAULT_SYSROOT=\/obj/i386.i386/src/tmp\ 
-I/obj/i386.i386/src/tmp/legacy/usr/include -fno-exceptions -fno-rtti -c 
/src/lib/clang/libclangsema/../../../contrib/llvm/tools/clang/lib/Sema/SemaConsumer.cpp
 -o SemaConsumer.o
c++  -O2 -pipe -I/src/lib/clang/libclangsema/../../../contrib/llvm/include 
-I/src/lib/clang/libclangsema/../../../contrib/llvm/tools/clang/include 
-I/src/lib/clang/libclangsema/../../../contrib/llvm/tools/clang/lib/Sema -I. 
-I/src/lib/clang/libclangsema/../../../contrib/llvm/../../lib/clang/include 
-DLLVM_ON_UNIX -DLLVM_ON_FREEBSD -D__STDC_LIMIT_MACROS -D__STDC_CONSTANT_MACROS 
-fno-strict-aliasing -DLLVM_DEFAULT_TARGET_TRIPLE=\i386-unknown-freebsd11.0\ 
-DLLVM_HOST_TRIPLE=\x86_64-unknown-freebsd11.0\ 
-DDEFAULT_SYSROOT=\/obj/i386.i386/src/tmp\ 
-I/obj/i386.i386/src/tmp/legacy/usr/include -fno-exceptions -fno-rtti -c 
/src/lib/clang/libclangsema/../../../contrib/llvm/tools/clang/lib/Sema/SemaDecl.cpp
 -o SemaDecl.o
c++  -O2 -pipe -I/src/lib/clang/libclangsema/../../../contrib/llvm/include 
-I/src/lib/clang/libclangsema/../../../contrib/llvm/tools/clang/include 
-I/src/lib/clang/libclangsema/../../../contrib/llvm/tools/clang/lib/Sema -I. 
-I/src/lib/clang/libclangsema/../../../contrib/llvm/../../lib/clang/include 
-DLLVM_ON_UNIX -DLLVM_ON_FREEBSD -D__STDC_LIMIT_MACROS -D__STDC_CONSTANT_MACROS 
-fno-strict-aliasing -DLLVM_DEFAULT_TARGET_TRIPLE=\i386-unknown-freebsd11.0\ 
-DLLVM_HOST_TRIPLE=\x86_64-unknown-freebsd11.0\ 
-DDEFAULT_SYSROOT=\/obj/i386.i386/src/tmp\ 
-I/obj/i386.i386/src/tmp/legacy/usr/include -fno-exceptions -fno-rtti -c 
/src/lib/clang/libclangsema/../../../contrib/llvm/tools/clang/lib/Sema/SemaDeclAttr.cpp
 -o SemaDeclAttr.o
/src/lib/clang/libclangsema/../../../contrib/llvm/tools/clang/lib/Sema/SemaDeclAttr.cpp:
 In function 'bool checkAvailabilityAttr(clang::Sema, clang::SourceRange, 
clang::IdentifierInfo*, clang::VersionTuple, clang::VersionTuple, 
clang::VersionTuple)':
/src/lib/clang/libclangsema/../../../contrib/llvm/tools/clang/lib/Sema/SemaDeclAttr.cpp:2084:
 internal compiler error: in var_ann, at tree-flow-inline.h:128
Please submit a full bug report,
with preprocessed source if appropriate.
See URL:http://gcc.gnu.org/bugs.html for instructions.
*** Error code 1

Stop.
bmake[3]: stopped in /src/lib/clang/libclangsema
*** Error code 1

Stop.
bmake[2]: stopped in /src/lib/clang
*** Error code 1

Stop.
bmake[1]: stopped in /src
*** Error code 1

Stop.
bmake: stopped in /src
*** Error code 1

Stop in /src.
TB --- 2013-12-21 21:19:22 - WARNING: /usr/bin/make returned exit code  1 
TB --- 2013-12-21 21:19:22 - ERROR: failed to build world
TB --- 2013-12-21 21:19:22 - 1526.43 user 170.29 system 1741.63 real


http://tinderbox.freebsd.org/tinderbox-head-build-HEAD-i386-i386.full
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Darren Pilgrim

On 12/21/2013 1:05 PM, Steve Kargl wrote:

On Sat, Dec 21, 2013 at 02:54:39PM -0600, Greg Rivers wrote:

On Sat, 21 Dec 2013, Steve Kargl wrote:


It did not ask how to stop this stupidity.  I asked to have this
stupidity stopped by default.  The spewing of this information in
/var/log/messages provides NOTHING.  Please turn it off by default.



Do you really feel that strongly about it?  Having a record of changes to
the system has always seemed like a feature to me...



Yes, I do feel strongly about it.  It is completely unnecesary noise.
It should be off by default.  If someone wants to fill /var up with
useless information, then that someone can turn on the noise.


It's about what's safe in the common case.  There are significant 
security risks inherent in pkg's activities, so having a written 
external record is the safe option.


I don't buy the fill up /var argument.  If your /var is so small that 
pkg's logging risks filling it up, why are you not logging to an 
external syslog server?  There are much more voluminous sources of logs 
on a FreeBSD system.

___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Steve Kargl
On Sat, Dec 21, 2013 at 01:24:55PM -0800, Darren Pilgrim wrote:
 On 12/21/2013 1:05 PM, Steve Kargl wrote:
  On Sat, Dec 21, 2013 at 02:54:39PM -0600, Greg Rivers wrote:
  On Sat, 21 Dec 2013, Steve Kargl wrote:
 
  It did not ask how to stop this stupidity.  I asked to have this
  stupidity stopped by default.  The spewing of this information in
  /var/log/messages provides NOTHING.  Please turn it off by default.
 
 
  Do you really feel that strongly about it?  Having a record of changes to
  the system has always seemed like a feature to me...
 
 
  Yes, I do feel strongly about it.  It is completely unnecesary noise.
  It should be off by default.  If someone wants to fill /var up with
  useless information, then that someone can turn on the noise.
 
 It's about what's safe in the common case.  There are significant 
 security risks inherent in pkg's activities, so having a written 
 external record is the safe option.
 
 I don't buy the fill up /var argument.  If your /var is so small that 
 pkg's logging risks filling it up, why are you not logging to an 
 external syslog server?  There are much more voluminous sources of logs 
 on a FreeBSD system.

It has nothing to do with the size of /var, really.  It is completely
useless information.  You want to know what package are installed, use
'pkg info'.  Packages do not spontaneously install themselves.  If
your system is so insecure that you are worried that some unpriveleged
user installed a package, you have bigger problems.

-- 
steve


-- 
Steve
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Baptiste Daroussin
On Sat, Dec 21, 2013 at 01:39:59PM -0800, Steve Kargl wrote:
 On Sat, Dec 21, 2013 at 01:24:55PM -0800, Darren Pilgrim wrote:
  On 12/21/2013 1:05 PM, Steve Kargl wrote:
   On Sat, Dec 21, 2013 at 02:54:39PM -0600, Greg Rivers wrote:
   On Sat, 21 Dec 2013, Steve Kargl wrote:
  
   It did not ask how to stop this stupidity.  I asked to have this
   stupidity stopped by default.  The spewing of this information in
   /var/log/messages provides NOTHING.  Please turn it off by default.
  
  
   Do you really feel that strongly about it?  Having a record of changes to
   the system has always seemed like a feature to me...
  
  
   Yes, I do feel strongly about it.  It is completely unnecesary noise.
   It should be off by default.  If someone wants to fill /var up with
   useless information, then that someone can turn on the noise.
  
  It's about what's safe in the common case.  There are significant 
  security risks inherent in pkg's activities, so having a written 
  external record is the safe option.
  
  I don't buy the fill up /var argument.  If your /var is so small that 
  pkg's logging risks filling it up, why are you not logging to an 
  external syslog server?  There are much more voluminous sources of logs 
  on a FreeBSD system.
 
 It has nothing to do with the size of /var, really.  It is completely
 useless information.  You want to know what package are installed, use
 'pkg info'.  Packages do not spontaneously install themselves.  If
 your system is so insecure that you are worried that some unpriveleged
 user installed a package, you have bigger problems.
 
 -- 
 steve
 
 
 -- 
 Steve
 ___
 freebsd-current@freebsd.org mailing list
 http://lists.freebsd.org/mailman/listinfo/freebsd-current
 To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org

this has been done and activated for reason, first for lot of companies, it is
important (PCI DSS requirement for example), secondly I receive tons of request
to actiavte on by default while you are the first to request it off by default

Bapt


pgpuaUQdFoTIm.pgp
Description: PGP signature


Re: PACKAGESITE spam

2013-12-21 Thread Steve Kargl
On Sat, Dec 21, 2013 at 11:14:39PM +0100, Baptiste Daroussin wrote:
 
 this has been done and activated for reason, first for lot of companies,

companies can turn it on it they want it.

 secondly I receive tons of request to actiavte on by default while
 you are the first to request it off by default

I certainly can't refute 'tons of [private] requests'.  There is
no discussing of such logging in freebsd-current, freebsd-hackers,
or freebsd-ports lists.

Other than the noise in /var/log/message, what does this provide
that 'pkg info' doesn't!  Please turn of this feature by default.

-- 
Steve
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Julien Laffaye

On 12/22/2013 12:04 AM, Steve Kargl wrote:

On Sat, Dec 21, 2013 at 11:14:39PM +0100, Baptiste Daroussin wrote:

this has been done and activated for reason, first for lot of companies,

companies can turn it on it they want it.


And if you want to save the few extra kB, you can turn it off...


secondly I receive tons of request to actiavte on by default while
you are the first to request it off by default

I certainly can't refute 'tons of [private] requests'.  There is
no discussing of such logging in freebsd-current, freebsd-hackers,
or freebsd-ports lists.

Other than the noise in /var/log/message, what does this provide
that 'pkg info' doesn't!  Please turn of this feature by default.



Actually, you are the one making noise...
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Steve Kargl
On Sun, Dec 22, 2013 at 12:16:00AM +0100, Julien Laffaye wrote:
 On 12/22/2013 12:04 AM, Steve Kargl wrote:
  On Sat, Dec 21, 2013 at 11:14:39PM +0100, Baptiste Daroussin wrote:
  this has been done and activated for reason, first for lot of companies,
  companies can turn it on it they want it.
 
 And if you want to save the few extra kB, you can turn it off...

I have turned it off.  

  secondly I receive tons of request to actiavte on by default while
  you are the first to request it off by default
  I certainly can't refute 'tons of [private] requests'.  There is
  no discussing of such logging in freebsd-current, freebsd-hackers,
  or freebsd-ports lists.
 
  Other than the noise in /var/log/message, what does this provide
  that 'pkg info' doesn't!  Please turn of this feature by default.
 
 
 Actually, you are the one making noise...

What does this logging in /var/log/message provide that is
not provided by 'pkg info'?  It is useless noise.

-- 
Steve
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Julien Laffaye

On 12/22/2013 12:25 AM, Steve Kargl wrote:

On Sun, Dec 22, 2013 at 12:16:00AM +0100, Julien Laffaye wrote:

On 12/22/2013 12:04 AM, Steve Kargl wrote:

On Sat, Dec 21, 2013 at 11:14:39PM +0100, Baptiste Daroussin wrote:

this has been done and activated for reason, first for lot of companies,

companies can turn it on it they want it.

And if you want to save the few extra kB, you can turn it off...

I have turned it off.


secondly I receive tons of request to actiavte on by default while
you are the first to request it off by default

I certainly can't refute 'tons of [private] requests'.  There is
no discussing of such logging in freebsd-current, freebsd-hackers,
or freebsd-ports lists.

Other than the noise in /var/log/message, what does this provide
that 'pkg info' doesn't!  Please turn of this feature by default.


Actually, you are the one making noise...

What does this logging in /var/log/message provide that is
not provided by 'pkg info'?  It is useless noise.



Forensics ?
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Florian Smeets
On 22/12/13 00:04, Steve Kargl wrote:
 On Sat, Dec 21, 2013 at 11:14:39PM +0100, Baptiste Daroussin wrote:

 
 Other than the noise in /var/log/message, what does this provide
 that 'pkg info' doesn't!  Please turn of this feature by default.
 

No please don't, it's a very useful feature, you can see when packages
were installed or upgraded. Which can be very useful in certain
environments.

Additionally, looking at other systems (SLES, CentOS, Debian and Ubuntu
from the top of my head) they all have a log that keeps track of
packages. The idea cannot be *that* useless.

Florian



signature.asc
Description: OpenPGP digital signature


Re: PACKAGESITE spam

2013-12-21 Thread Steve Kargl
On Sat, Dec 21, 2013 at 06:29:17PM -0500, Garrett Wollman wrote:
 In article 20131221230448.ga61...@troutmask.apl.washington.edu,
 Steve Kargl writes:
 
 Other than the noise in /var/log/message, what does this provide
 that 'pkg info' doesn't!
 
 A record of when packages were installed and removed.
 

The date/time a package is installed should be recorded
in the pkg database and be available with 'pkg info -f'.
Of course, it appeat that this information s not properly
recorded.

-- 
Steve
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Glen Barber
On Sat, Dec 21, 2013 at 03:44:11PM -0800, Steve Kargl wrote:
 On Sat, Dec 21, 2013 at 06:29:17PM -0500, Garrett Wollman wrote:
  In article 20131221230448.ga61...@troutmask.apl.washington.edu,
  Steve Kargl writes:
  
  Other than the noise in /var/log/message, what does this provide
  that 'pkg info' doesn't!
  
  A record of when packages were installed and removed.
  
 
 The date/time a package is installed should be recorded
 in the pkg database and be available with 'pkg info -f'.
 Of course, it appeat that this information s not properly
 recorded.
 

Because databases are never corruptible?

Glen



pgplQ34lUTbQB.pgp
Description: PGP signature


Re: PACKAGESITE spam

2013-12-21 Thread Steve Kargl
On Sat, Dec 21, 2013 at 06:47:01PM -0500, Glen Barber wrote:
 On Sat, Dec 21, 2013 at 03:44:11PM -0800, Steve Kargl wrote:
  On Sat, Dec 21, 2013 at 06:29:17PM -0500, Garrett Wollman wrote:
   In article 20131221230448.ga61...@troutmask.apl.washington.edu,
   Steve Kargl writes:
   
   Other than the noise in /var/log/message, what does this provide
   that 'pkg info' doesn't!
   
   A record of when packages were installed and removed.
   
  
  The date/time a package is installed should be recorded
  in the pkg database and be available with 'pkg info -f'.
  Of course, it appeat that this information s not properly
  recorded.
  
 
 Because databases are never corruptible?
 

/var/log/message never gets corrupted?

-- 
Steve
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Baptiste Daroussin
On Sat, Dec 21, 2013 at 03:44:11PM -0800, Steve Kargl wrote:
 On Sat, Dec 21, 2013 at 06:29:17PM -0500, Garrett Wollman wrote:
  In article 20131221230448.ga61...@troutmask.apl.washington.edu,
  Steve Kargl writes:
  
  Other than the noise in /var/log/message, what does this provide
  that 'pkg info' doesn't!
  
  A record of when packages were installed and removed.
  
 
 The date/time a package is installed should be recorded
 in the pkg database and be available with 'pkg info -f'.
 Of course, it appeat that this information s not properly
 recorded.
 
 -- 
 Steve
 ___
 freebsd-current@freebsd.org mailing list
 http://lists.freebsd.org/mailman/listinfo/freebsd-current
 To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org

Works for installation and we do track it, does not work for removal

seriously are installing that many packages everyday that it bother you?

regards,
Bapt


pgpe1334kYS1T.pgp
Description: PGP signature


Re: PACKAGESITE spam

2013-12-21 Thread Glen Barber
On Sat, Dec 21, 2013 at 03:49:26PM -0800, Steve Kargl wrote:
 /var/log/message never gets corrupted?
 

If a machine is compromised, I can review log from remote syslog server.
It is impossible to trust the package information, relying on the local
database alone.

Glen



pgpklyEF3RMKl.pgp
Description: PGP signature


Re: PACKAGESITE spam

2013-12-21 Thread Garrett Wollman
In article 20131221230448.ga61...@troutmask.apl.washington.edu,
Steve Kargl writes:

Other than the noise in /var/log/message, what does this provide
that 'pkg info' doesn't!

A record of when packages were installed and removed.

-GAWollman

___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Jakub Lach
I was surprised initially, but haven't thought about it much since.



--
View this message in context: 
http://freebsd.1045724.n5.nabble.com/PACKAGESITE-spam-tp5870583p5870666.html
Sent from the freebsd-current mailing list archive at Nabble.com.
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: PACKAGESITE spam

2013-12-21 Thread Shane Ambler
On 22/12/2013 09:34, Steve Kargl wrote:
 On Sat, Dec 21, 2013 at 11:14:39PM +0100, Baptiste Daroussin wrote:

 this has been done and activated for reason, first for lot of companies,
 
 companies can turn it on it they want it.

Generally the more secure option is default, you can turn off security
options as you choose.

 secondly I receive tons of request to actiavte on by default while
 you are the first to request it off by default
 
 I certainly can't refute 'tons of [private] requests'.  There is
 no discussing of such logging in freebsd-current, freebsd-hackers,
 or freebsd-ports lists.
 
 Other than the noise in /var/log/message, what does this provide
 that 'pkg info' doesn't!  Please turn of this feature by default.
 

Isn't this noise the same as we had with the old package system?
So really nothing has changed, well now we can turn it off which I
don't think we could before.

___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org