[ams@kemisten.nu: rpctrace --help output]
Ping pong... --- Start of forwarded message --- Date: Tue, 16 Sep 2003 15:06:18 +0200 (MEST) From: "Alfred M. Szmidt" <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Subject: rpctrace --help output This make the output of "rpcrace --help" less annoying. Here is an example of what it look like now (note the `.' before the "Report bugs to ..." string). ~/src/hurd $ rpctrace --help Usage: rpctrace [OPTION...] COMMAND [ARG...] Trace Mach Remote Procedure Calls. -I, --rpc-list=FILERead FILE for assocations of message ID numbers to names. -o, --output=FILE Send trace output to FILE instead of stderr. -?, --help Give this help list --usageGive a short usage message -V, --version Print program version Mandatory or optional arguments to long options are also mandatory or optional for any corresponding short options. . Report bugs to [EMAIL PROTECTED] ~/src/hurd $ 2003-09-16 Alfred M. Szmidt <[EMAIL PROTECTED]> * rpctrace.c (doc): Remove spurious vertical tab. Index: utils/rpctrace.c - --- utils/rpctrace.c +++ utils/rpctrace.c @@ -45,8 +45,7 @@ static const struct argp_option options[ static const char *args_doc = "COMMAND [ARG...]"; static const char *doc = - -"Trace Mach Remote Procedure Calls." - -"\v."; +"Trace Mach Remote Procedure Calls."; /* The msgid_ihash table maps msgh_id values to names. */ ___ Bug-hurd mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/bug-hurd --- End of forwarded message --- ___ Bug-hurd mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/bug-hurd
[ams@kemisten.nu: Re: storeread --help output]
Ping pong...
--- Start of forwarded message ---
Date: Wed, 24 Sep 2003 09:19:10 +0200 (MEST)
From: "Alfred M. Szmidt" <[EMAIL PROTECTED]>
To: Marcus Brinkmann <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Subject: Re: storeread --help output
storeread should have a small description of what it does before
the \v, though.
Will this do my lord?
2003-09-16 Alfred M. Szmidt <[EMAIL PROTECTED]>
* storeread.c (doc): Added description about the program.
Index: utils/storeread.c
- --- utils/storeread.c
+++ utils/storeread.c
@@ -36,7 +36,9 @@
{0, 0}
};
char *arg_doc = "FILE [ADDR [LENGTH]]...";
- -char *doc = "\vADDR is in blocks, and defaults to 0; LENGTH is in bytes, and
defaults to the remainder of FILE.";
+char *doc = "Write portions of the contents of a store to stdout"
+ "\vADDR is in blocks, and defaults to 0; LENGTH is in bytes, and"
+ " defaults to the remainder of FILE.";
int
main (int argc, char **argv)
___
Bug-hurd mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-hurd
--- End of forwarded message ---
___
Bug-hurd mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-hurd
[ams@kemisten.nu: Re: bugs in idvec-verify.c, almost...]
Ping pong...
--- Start of forwarded message ---
Date: Tue, 16 Sep 2003 15:06:18 +0200 (MEST)
From: "Alfred M. Szmidt" <[EMAIL PROTECTED]>
To: "Alfred M. Szmidt" <[EMAIL PROTECTED]>
CC: [EMAIL PROTECTED]
Subject: Re: bugs in idvec-verify.c, almost...
Since there are quite a few bugs of the same nature here, it makes for
quite a long read. So to ease for the reader, here is the list of
bugs that I have found so far (the gdb output is at the end
somewhere). All of them segfault in either ugids-argp.c or
idvec-verify.c, and they do it for the same reason, i.e. the struct
they polute isn't checked if it is NULL.
A patch is applied at the end, with ChangeLog (search for the string
"*** Patch ***"), and it has been tested. If someone could be nice
enough to comment about the ChangeLog entry then that would be nice,
since I don't really like it.
Oh, should we be nice enough and report some kind of a useful error
message in idvec-verify:verify_passwd()? Because right now it would do
this:
$ addauth 123
addauth: Authentication failure: Invalid argument
Note that this is the only the case for UID's, user/group names report
the sane message "Unknown user/group", and if the user has a valid
entry in /etc/passwd (and /etc/shadow depending on its password,
i.e. it is "x").
* `addauth' with a UID that doesn't exist (idvec-verify).
* `addauth' with a GID that doesn't exist (idvec-verify).
* `addauth' with a user name that doesn't exist in /etc/passwd and
/etc/shadow (ugids-argp.c).
* `addauth' with a group name that doesn't exist in /etc/group
(ugids-argp.c)
* `addauth' with a user name that doesn't exist in /etc/shadow, but
exists in /etc/passwd (ugids-argp.c).
Running `addauth' with a UID that doesn't exist:
Starting program: /bin/addauth 123
Program received signal SIGSEGV, Segmentation fault.
0x01025323 in verify_id (id=123, is_group=0, multiple=0, getpass_fn=0,
getpass_hook=0x0, verify_fn=0x1026e40 ,
verify_hook=0x1017aa0)
at /src/hurd-2003-09-06/libshouldbeinlibc/idvec-verify.c:282
Running `addauth' with a GID that doesn't exist:
Starting program: /bin/addauth -g 123
Program received signal SIGSEGV, Segmentation fault.
0x01025162 in verify_id (id=123, is_group=1, multiple=0, getpass_fn=0,
getpass_hook=0x0, verify_fn=0x1026e40 ,
verify_hook=0x1017aa0)
at /src/hurd-2003-09-06/libshouldbeinlibc/idvec-verify.c:270
Running `addauth' with a user name that doesn't exist in /etc/passwd
and /etc/shadow:
Starting program: /bin/addauth toor
Program received signal SIGSEGV, Segmentation fault.
0x01025ab4 in parse_opt (key=0, arg=0x101800d "toor", state=0x1017a6c)
at /src/hurd-2003-09-06/libshouldbeinlibc/ugids-argp.c:88
Running `addauth' with a group name that doesn't exist in /etc/group:
Starting program: /bin/addauth -g toor
Program received signal SIGSEGV, Segmentation fault.
0x01025a0a in parse_opt (key=103, arg=0x1018010 "toor", state=0x1017a6c)
at /src/hurd-2003-09-06/libshouldbeinlibc/ugids-argp.c:126
Running `addauth' with a user name that doesn't exist in /etc/shadow,
but exists in /etc/passwd:
Starting program: /bin/addauth fool
Program received signal SIGSEGV, Segmentation fault.
0x0102537f in verify_id (id=666, is_group=0, multiple=0, getpass_fn=0,
getpass_hook=0x0, verify_fn=0x1026e40 ,
verify_hook=0x1017aa0)
at /src/hurd-2003-09-06/libshouldbeinlibc/idvec-verify.c:294
*** Patch ***
2003-09-15 Alfred M. Szmidt <[EMAIL PROTECTED]>
* idvec-verify.c (verify_passwd,verify_id): Check if the spwd and
passwd structures are NULL, and return an error if so.
(sys_encrypt): Check if NULL, return an error if so.
* ugids-argp.c (parse_opt): Check if the group and passwd
structures are NULL, and return an error if so.
Index: libshouldbeinlibc/idvec-verify.c
- --- libshouldbeinlibc/idvec-verify.c
+++ libshouldbeinlibc/idvec-verify.c
@@ -99,17 +99,22 @@ verify_passwd (const char *password,
{
/* When encrypted password is "x", try shadow passwords. */
struct spwd _sp, *sp;
- - if (getspnam_r (pw->pw_name, &_sp, sp_lookup_buf,
- - sizeof sp_lookup_buf, &sp) == 0)
- - return sp->sp_pwdp;
+ getspnam_r (pw->pw_name, &_sp, sp_lookup_buf,
+ sizeof sp_lookup_buf, &sp);
+ if (sp == NULL)
+ return NULL;
+ return sp->sp_pwdp;
}
return pw->pw_passwd;
}
- - if (getpwuid_r (wheel_uid, &_pw, lookup_buf, sizeof lookup_buf, &pw))
- - return errno ?: EINVAL;
+ getpwuid_r (wheel_uid, &_pw, lookup_buf, sizeof lookup_buf, &pw);
+ if (pw == NULL)
+ return EINVAL;
sys_encrypted = check_shadow (pw);
+ if (sys_encrypted == NULL)
+ return EINVAL;
encrypted = crypt (password, sys_encrypted);
if (! encrypted)
@@ -264,41 +
Re: bugs in idvec-verify.c, almost...
You should not remove the checking of the return value of get*_r. When these return nonzero, that value (also in errno) should be returned. If it returns zero that means "nothing went wrong", in which case the result being null means "not found" and that case is what we need to handle differently. ___ Bug-hurd mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/bug-hurd
[PATCH][BETA 1] ext2fs for large stores
Since a couple of weeks I'm doing many light compilations (ext3fs) with the alpha 3 patch. There wasn't a single failure in which clues lead to the patch so this beta 1 patch is essentialy alpha 3 with some cleanups of the code. As usual, you can download it from http://debian.fmi.uni-sofia.bg/~ogi/hurd/ext3fs/ It's recommended that e2fsprogs is patched too with the patch from the same site. FUTURE -- I'll do some heavy compilations (glibc) and if the patch performs well it'll be announced as Release Candidate 1 which will be uploaded to Savannah Patch system. It will have ChangeLog entries, ready to be included in the Hurd. (I've signed the GNU papers under #32260.) Regards -- Ognyan Kulev <[EMAIL PROTECTED],fsa-bg.org,jabber.net}> 7D9F 66E6 68B7 A62B 0FCF EB04 80BF 3A8C A252 9782 ___ Bug-hurd mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/bug-hurd
Rx: VICODIN is Here - Just Pennies per Tablet...gwyneth
VICODIN Is Here FRE.E-shipping on a 3 month supply (90-count) www.rxanytime.biz/superstore Vicodin ES 30 tablets - 4.63 per tablet ES 90 tablets - 2.32 per tablet - BEST DEAL - Plus FRE.E-SHIPPING www.rxanytime.biz/superstore No more ads - www.rxanytime.biz/a.html pewlj ___ Bug-hurd mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/bug-hurd
Re: help needed implementing a new store class
On Sat, 25 Oct 2003, Roland McGrath wrote: > You should run storeio under gdb and see what is going on when it tries to > load your module. Thanks for your reply, but i don't know how to debug storeio :-( If i should continue this thread in help-hurd please to tell me it i installed the debian package libc0.3-dbg (that installs the libc with debug in /lib/debug) and compiled the hurd, then i've done three test: $ cp crypt.c hurd/libstore ; cd hurd/libstore added -lcrypto in Makefile for OpenSSL $ make libstore_crypt.so.0.3 $ export LD_LIBRARY_PATH=$PWD:/lib/debug $ cd ../storeio $ make $ settrans -Pac ~/test storeio -T crypt ~/pippo.enc crypt: Invalid argument to --store-type well i try to run a typed store to have the time to pause: $ settrans -Pac ~/test storeio -T crypt ~/pippo.enc Translator pid: 1234 Pausing... in another shell (in the same directory with the same LD_LIBRARY_PATH): $ gdb -d ../libstore storeio 1234 Attaching to program: storeio, process 1234 Reading symbols from /lib/libstore.so.0.3...done. Loaded symbols for /lib/libstore.so.0.3 Reading symbols from /libc.so.0.3...done. Loaded symbols for /libc.so.0.3 why my LD_LIBRARY_PATH environment variable is not used for dynamic linking? I think this only appears when using settrans the second test was: $ cd hurd/libstore created a test store class from file store class $ make libstore_test.so.0.3 $ settrans -Pac ~/test ../storeio/storeio -T test ~/pippo.enc test: Invalid argument to --store-type the third: $ cd hurd/storeio $ make storeio.static $ settrans -Pac ~/test storeio.static -T crypt ~/pippo.enc $ cat ~/test Invalid argument (or similar i don't remember) $ settrans -Pac ~/test storeio.static -T file ~/pippo.enc $ cat ~/test Invalid argument (or similar) Should i compile a static storeio in another way? It is normal that my LD_LIBRARY_PATH is not used when setting translators? It is correct what i've done to debug? If it is ok i'll try to reinstall the system with an updated Debian GNU Hurd (i'm using J2). c-ya! 3 http://www.trek.eu.org/ k PGPKey: 7016731A57D4A69B 1A8EE5E90EF2608E (since 1995) ___ Bug-hurd mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/bug-hurd
[home-job:00062] $BL$>5Bz9-9p"(!Z:_Bp%9%?%C%UJg=8![(B
$B!cG[?.l9g$O:o=|4j$$$^$9!#(B (B $B!Z!Z!Z!Z!Z:_Bp%9%?%C%UJg=8![![![![![(B $B%$%s%?!<%M%C%H$G%[!<%`%Z!<%8$r8+$k;v$,$G$-!"(B $B%a!<%k$NAw$N$*;E;v$H3]$1;}$A$7$F$b0B?4$G$9!#(B $BOC=Q!&4+M6!&?ML.!&%;%_%J!\:Y(B $B"-(B (Bhttp://iis.to/together/ (B $B!c%H%%%.%c%6!<4k2h!d(B (B $B"(LBOG%a!<%k$K$J$i$J$$MM!"0lEY8B$j$NAw?.$r?4$,$1$F$*$j$^$9!#(B $B!!=EJ#%A%'%C%/$O$7$F$*$j$^$9$,!"K|$,0lB3$1$FAw?.$5$l$?>l9g$O(B $B!!7h$7$F0-0U$O$4$6$$$^$;$s$N$G$4MFl9g$O!"$*http://bb.yahoo.co.jp/ (B (B (B (B___ (BBug-hurd mailing list (B[EMAIL PROTECTED] (Bhttp://mail.gnu.org/mailman/listinfo/bug-hurd
