Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Steven Levine

Pages: [1] 2 3 ... 8
1
Utilities / Re: SYSLOGD - stopped working...SYSxxxx, but why?
« on: March 16, 2026, 08:59:22 pm »
The SYS1808 is cmd.exe's way of telling you to go look for a popuplog entry.

The 0005 is the low order 16-bits of the c0000005 exception code.  The truncation is probably a side effect of the fact that cmd.exe is antique 16-bit code.

There are several easy ways to look at the code at the CS:EIP=005b:00023cd2 exception address. I used ida to do a quick disassembly.  The benefit is that I got symbolic names for many of the called functions. Another option is crank up the OpenWatcom debugger and view the code near the exception address.  Set a breakpoint at the call instruction and run the code.

Steven


2
Utilities / Re: SYSLOGD - stopped working...SYSxxxx, but why?
« on: March 16, 2026, 08:21:59 am »
Hi,

If appears you did something to your services file so that it no longer contains a valid syslog entry.

FWIW, Mike built his syslogd with Watcom and provided full source code.  You might want to consider modifying the code so that that it does not trap when GETSERVBYNAME returns NULL.

Steven

3
Internet / Re: Problems with OpenSSL library after update.
« on: February 08, 2026, 04:51:55 am »
Oops - wrong repo and wrong branch.  In the current netlabs sources, the code has moved to crypto\cpuid.c.

Some of the interesting code points are at

crypto\cpuid.c:18
unsigned int OPENSSL_ia32cap_P[4];

crypto\cpuid.c:110
    if ((env = ossl_getenv("OPENSSL_ia32cap")) != NULL) {
        int off = (env[0] == '~') ? 1 : 0;

poly1305-x86.pl:119
      if ($avx>1) {
   &mov   ("ecx",&DWP(8,"edi"));
   &test   ("ecx",1<<5);         # AVX2?
   &jz   (&label("no_sse2"));

Since vmovdqa is an AVX2 instruction, we want to turn off

  bit #64+5 denoting availability of AVX2 instructions;

to disable the problematic code path.

The brute force

  set OPENSSL_ia32cap=:0

should do this.  The less brutish

  set OPENSSL_ia32cap=:~0x20

should also work, but might leave other problem code paths enabled.

4
Internet / Re: Problems with OpenSSL library after update.
« on: February 07, 2026, 06:55:23 am »
It occurred to me that we might not be the only platform that might bump up against this kind of compatibility problem, so I went hunting.

It appears it should be possible to configure OPENSSL at runtime to avoid attempting to use the problematic code via the OPENSSL_ia32cap environment variable.

See

  https://docs.openssl.org/1.1.1/man3/OPENSSL_ia32cap/#description

and

  https://stackoverflow.com/questions/40091888/how-to-disable-avx-instructions-in-openssl

For the curious, the relevant code is at

crypto\cryptlib.c:694
    if ((env = getenv("OPENSSL_ia32cap"))) {

in the netlabs openssl sources.


5
Has DDNSAPC.CMD ever worked for you in the past?

Looking at the script, I wonder if an argument is required.  Typically IBM would create a Desktop object to run the script and provide the site specific argument in the object parameters.

6
Internet / Re: Problems with OpenSSL library after update.
« on: January 29, 2026, 04:21:18 am »
The updated exceptq report helps a lot.  I'm going to chalk this trap up to bad luck.  Here's what I suspect is
happening.  We are trying to execute

 vmovdqa ymm7, [rbx+40h]

and exceptq tells us that

 EBX : read/exec  memory at 0001:00148930 in CRYPTO11

which confirms that ebx+40 is not properly aligned.  It also tells us that ebx is probably pointing at constant data stored in the code segment - possibly OPENSSL_ia32cap_P.  A copy of poly1305-x86.s would make this easier
to figure this out.

IAC, whatever the data is, it's now in an unaligned location in 1.1.1w.

To avoid this, we would need to figure out how to ensure the data is aligned and change to code to use vmovdqu rather than vmovdqa.

7
Internet / Re: Problems with OpenSSL library after update.
« on: January 25, 2026, 09:02:16 pm »
I recommend you install the debuginfo packages for the modules listed in the "Labels on the Stack" section.  This will make analyzing the exceptq report more efficient.

Also, you might want to consider installing http:/www.warpcave.com/betas/exceptq-7.11.6.2-shl-2023-12-31.zip.  It can help with understanding the call chain for binaries that do not use traditional stack frames.

8
General Discussion / Re: Spam Bots
« on: January 25, 2026, 08:50:08 pm »
Not unexpectedly, the Lynx UA works for firefox too.  Thanks.

Turns out changing "Mozilla" to "OS2zilla" in the default OS/2 UA works too.  So we don't have to give up on our OS/2 oriented UAs.

9
Applications / Re: XTide
« on: January 21, 2026, 06:39:51 am »
Without looking at the code, I have to guess that -DHAVE_ISSETUGID=0 is part of the problem.  This is going to trigger the compatibility code to build.  However, since we have a issetugid(), this is going to cause a problem.

10
Networking / Re: uacme
« on: January 21, 2026, 03:31:12 am »
You want,

  https://smedley.id.au/tmp/uacme-1.2.4-os2-20240817.zip

A version of the scripts I have used can be found at

  http://www.warpcave.com/betas/uacme-scripts-for-peter-2024-12-06-20240818.zip

The documentation is a bit rough.  The scripts should complain if you neglect to create one or more of the required directories or if you don't have the apache confs setup quite right.

11
General Discussion / Re: Spam Bots
« on: January 20, 2026, 03:16:07 am »
Both dooble5 and dooble6 work here too, but they are slower than firefox and still less stable.  Thanks.

12
General Discussion / Re: Spam Bots
« on: January 19, 2026, 06:01:31 pm »
I just noticed that anubis seems to be preventing me from accessing edm2 with firefox and seamonkdy.  Did I miss a setting somewhere?  If not, can the same fixes that are used for os2world be applied to edm2?

13
Applications / Re: General Software Testing
« on: January 19, 2026, 09:36:22 am »
Turns out I misread the popuplog.  it's not a stack problem.  I cranked up the debugger to see what the code was really doing that triggered the trap.  The trap is at

035B                 les     di, pntr_2FC8   ; MartinI trap here

so it's a pointer problem.  The pointer is set at:

0293                 call    far ptr WinLoadPointer
0298                 mov     word ptr pntr_2FC8, ax; return pntr not handle ??
029B                 mov     word ptr pntr_2FC8+2, dx

What is odd about this is that based on the code, WinLoadPointer is returning a pointer, not a handle like current versions of WinLoadPointer.  I have to assume that this app was built to run against a really old version of PM.  Possibly for OS/2 2.1

The pointer returned by WinLoadPointer looks OK, but the selector (1b30) is a ring0 selector which is not usable by a ring3 app.

It would be interesting to see if someone with an OS/2 2.x install can run this app.

14
Applications / Re: General Software Testing
« on: January 17, 2026, 03:48:25 am »
This is stack overflow.  You might try enlarging the stack.  Since this is a 16-bit app and the stack is in the same segment as the dgroup, this might be effective.

15
Applications / Re: FM2
« on: January 04, 2026, 06:33:01 am »
Hi,

I pointed Gregg at this thread.

To recover you trac userid/password is probably going to require a message to the netlabs community mailing list.  The member.xml link is no longer functional.


Pages: [1] 2 3 ... 8