[Patches] [ python-Patches-803998 ] [_ssl.c] SSL_write() called
with -1 as size
SourceForge.net
noreply at sourceforge.net
Mon Oct 27 09:25:30 EST 2003
Patches item #803998, was opened at 2003-09-10 22:51
Message generated for change (Comment added) made by loewis
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=305470&aid=803998&group_id=5470
Category: Modules
Group: None
>Status: Closed
>Resolution: Accepted
Priority: 5
Submitted By: Ilario Nardinocchi (illo)
Assigned to: Martin v. Löwis (loewis)
Summary: [_ssl.c] SSL_write() called with -1 as size
Initial Comment:
In the inner do/while loop of PySSL_SSLwrite(), whenever a
SSL_write() needs to be called again after a
SSL_ERROR_WANT_{READ,WRITE} error, its "len" parameter
is set to -1, leading to a segfault.
I can't believe nobody ever noticed that :)
----------------------------------------------------------------------
>Comment By: Martin v. Löwis (loewis)
Date: 2003-10-27 15:25
Message:
Logged In: YES
user_id=21627
Thanks for the patch. Committed (with that change) as _ssl.c
1.13.6.2 and 1.15; NEWS 1.831.4.69.
----------------------------------------------------------------------
Comment By: Ilario Nardinocchi (illo)
Date: 2003-10-26 18:17
Message:
Logged In: YES
user_id=626684
since there is no '&count' in the _ssl.c code I'm using, I assume it's
a typo I fixed just after uploading the patch, but I cannot recall it,
it happened a long ago. Sorry.
I don't upload the patch because from what I'm seeing it's just a
matter of deleting the '&'.
----------------------------------------------------------------------
Comment By: Martin v. Löwis (loewis)
Date: 2003-10-26 17:51
Message:
Logged In: YES
user_id=21627
Why are you passing the address of count, and not the value
of count, to SSL_write?
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=305470&aid=803998&group_id=5470
More information about the Patches
mailing list