Anders Broman wrote:
I don't think we can/should turn off canaries in se_ allocations.
Instead we should create a new canary-less allocator. (Not sure what
such a thing should
be named, of course...)
Well as I see it EP memory is not a problem we only use one chunk (10M)
During the life
-Ursprungligt meddelande-
Från: wireshark-dev-boun...@wireshark.org
[mailto:wireshark-dev-boun...@wireshark.org] För Jeff Morriss
Skickat: den 9 oktober 2009 22:06
Till: Developer support list for Wireshark
Ämne: Re: [Wireshark-dev] New packet list - out of memory?
Anders Broman wrote
Le mercredi 07 octobre 2009 à 17:01 +0200, Anders Broman a écrit :
Hi,
If some one's interested here's the emem statistics for the file:
#define EMEM_ALLOCS_PER_CHUNK (EMEM_PACKET_CHUNK_SIZE / 64)
PWS Commit
43 49264 512
332 052 471 116
oktober 2009 09:34
To: Developer support list for Wireshark
Subject: Re: [Wireshark-dev] New packet list - out of memory?
Le mercredi 07 octobre 2009 à 17:01 +0200, Anders Broman a écrit :
Hi,
If some one's interested here's the emem statistics for the file:
#define EMEM_ALLOCS_PER_CHUNK
didier wrote:
But are canaries used at all? In my understanding without
DEBUG_INTENSE_CANARY_CHECKS they are never checked and it's unset by
default.
Erm, emem_free_all() checks that the canaries haven't been corrupted:
if (memcmp(npc-canary_info-canary[i], canary,
-Original Message-
From: wireshark-dev-boun...@wireshark.org
[mailto:wireshark-dev-boun...@wireshark.org] On Behalf Of Jeff Morriss
Sent: den 8 oktober 2009 15:58
To: Developer support list for Wireshark
Subject: Re: [Wireshark-dev] New packet list - out of memory?
didier wrote
Anders Broman wrote:
Well as I see it EP memory is not a problem we only use one chunk (10M)
During the life time of a packet so memory efficency isn't a big issue.
But when dealing with large files waisting +30% of the memory is not an
option I think.
A way to still test se_alloc() could
Hi,
Le jeudi 08 octobre 2009 à 09:58 -0400, Jeff Morriss a écrit :
didier wrote:
But are canaries used at all? In my understanding without
DEBUG_INTENSE_CANARY_CHECKS they are never checked and it's unset by
default.
Erm, emem_free_all() checks that the canaries haven't been corrupted:
Gerald Combs wrote:
We could add a preference item which lets the user disable SE
canaries.
Maybe a slider with fast on one end and safe on the other, just to
let them know what they're getting into. :)
Not knowing the extent of the :) above, it is worth pointing out that
users are not
Hi,
Le mercredi 07 octobre 2009 à 13:59 +0200, Anders Broman a écrit :
Didier
More or less vanilla Wireshark, no colors. Enclosed proto Hierarchy. I can't
share this file unfortunatly. The reason why I chose it is just that it
caused 'Out of memmory' with the new packet list.
Thanks, Are you
-Ursprungligt meddelande-
Från: wireshark-dev-boun...@wireshark.org
[mailto:wireshark-dev-boun...@wireshark.org] För didier
Skickat: den 9 oktober 2009 03:45
Till: Developer support list for Wireshark
Ämne: Re: [Wireshark-dev] New packet list - out of memory?
Hi,
Le mercredi 07 octobre
-Original Message-
From: wireshark-dev-boun...@wireshark.org
[mailto:wireshark-dev-boun...@wireshark.org] On Behalf Of Bill Meier
Sent: den 6 oktober 2009 00:49
To: Developer support list for Wireshark
Subject: Re: [Wireshark-dev] New packet list - out of memory?
Bill Meier wrote
Anders Broman wrote:
Should we have a new allocator without the canaries?
I think either the current se_ allocations implementation need to be changed
Or the function removed. The result where WIRESHARK_DEBUG_SE_NO_CHUNKS=T
And loading is faster is puzzling.
Trying that on Linux (64-bit) with
Le mardi 06 octobre 2009 à 12:05 +0200, Anders Broman a écrit :
146 MB 661417 pkts (TCP reassembly)
Version 1.2.2 (SVN Rev 29910)
PWS Commit
42 01661 876
704 656 1 639 256
Loading time ~57s
Old packet list SVN 30353
PWS
Le mardi 06 octobre 2009 à 12:21 -0400, Jeff Morriss a écrit :
Another question: do we really need to use se_ allocations here? Do we
need the canaries or are we just using se_ allocations because they get
cleaned up for us automatically?
PacketListRecord size is 20 bytes (could be 16) and
Hi,
I got a 146 Mbyte trace file, loading it with the new
packet list WS blows up Out of memory i in emem.c (~line 353)
Looking in the task manager 279 216K s allcated. Using my second
build without the new packet list The file loads fine using 492 304K.
Changing se_alloc to g_slice_alloc0
From: wireshark-dev-boun...@wireshark.org
[mailto:wireshark-dev-boun...@wireshark.org] On Behalf Of Anders Broman
Sent: den 5 oktober 2009 12:32
To: Developer support list for Wireshark
Subject: [Wireshark-dev] New packet list - out of memory?
Hi,
I got a 146 Mbyte trace file, loading
Anders Broman wrote:
Hi,
Experimenting loading the file there is 163 calls to VirtualAlloc()
~1.630 GB memory used
Changing
#define EMEM_ALLOCS_PER_CHUNK (EMEM_PACKET_CHUNK_SIZE / 512)
to:
#define EMEM_ALLOCS_PER_CHUNK (EMEM_PACKET_CHUNK_SIZE / 64)
Gives only 26 VirtualAlloc() so it
Bill Meier wrote:
A data point: A few days ago while working on bug #2375 I noticed that
on Windows that if I defined the env logicals to not use se_alloc etc
that *much* less memory was required to load a capture. (20% of that
used with se_alloc etc ? I don't quite remember).
Details:
19 matches
Mail list logo
