X-Authentication-Warning: delorie.com: mail set sender to djgpp-workers-bounces using -f
X-Recipient: djgpp-workers AT delorie DOT com
Subject: Re: gcc-8.2.0
To: djgpp-workers AT delorie DOT com
References: <248415bb-553f-6dfa-6d72-7e736cdfa127 AT iki DOT fi>
 <CAA-ihx-C61K1_M9eLUdmrYDVWwsi_g4ZxSWq29UVqJX_bkCwaA AT mail DOT gmail DOT com>
 <3513a9b6-08b7-9c7d-399b-1180a9a67ced AT iki DOT fi>
 <CAA-ihx-i=xR5KkitDb60w1OJPpL4hXLvnLRSQuCmnE_w0M7MZQ AT mail DOT gmail DOT com>
From: "Andris Pavenis (andris DOT pavenis AT iki DOT fi) [via djgpp-workers AT delorie DOT com]" <djgpp-workers AT delorie DOT com>
Message-ID: <b6c3548b-2be6-938c-169b-c5412f988b70@iki.fi>
Date: Mon, 6 Aug 2018 07:06:34 +0300
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.9.1
MIME-Version: 1.0
In-Reply-To: <CAA-ihx-i=xR5KkitDb60w1OJPpL4hXLvnLRSQuCmnE_w0M7MZQ@mail.gmail.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US-large
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from Quoted-Printable to 8bit by delorie.com id w7646j76021813
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

On 08/06/2018 12:46 AM, Rugxulo (rugxulo AT gmail DOT com) [via djgpp-workers AT delorie DOT com] wrote:
> "Unfortunately Microsoft has broken DPMI in Windows 10 32 bit since March 2018"
And it is going from bad to worse:

Initially only nested DPMI applications crashed.

Now (last 2 fast ring insider builds) any DJGPP application that reads from console crashes with 
ntvdm error (The /supplied user buffer/is /not valid/for the /requested operation)

/Earlier problem is still there: tried 'sh -c sh' and it crashed as earlier

I can still build GCC on my old ThinkPad notebook (Windows Vista Business SP3 32 bit) .
Tried FreeDOS 1.2 in VirtualBox VM but:
    - even unzipping gcc820s.zip do not work (writing part of files fails on FAT32 partition when 
LFN is enabled). Additionally unzipping last for hours...
    - pipes (like 'ls -l | grep something') do not work
- ssso/not wo//


/
> Depressing ... but then again, I don't use NTVDM anymore, stuck with
> old Win7 64-bit (for no good reason).
>
> I did (again) test rebuilding p7zip 16.02 (C++), under old DOSEMU, and
> it still seems to mostly work okay. So that's good. (Haven't yet again
> tried rebuilding in pure FreeDOS with SFNs, but I know it works, I've
> done it before, with minor patches.)
>
>>> GCC 8.2.0's gcc.info (erroneously) says:
>>>
>>>     The default, if no C language dialect options are given, is
>>>     `-std=gnu11'.
>>>
>>> But later says this (correctly):
>>>
>>>     `gnu17'
>>>     `gnu18'
>>>     GNU dialect of ISO C17.  This is the default for C code.
>>>
>>> DJGPP's GCC 8.2.0 (default) = gnu17 = 201710L
>>>
>>> Of course, a quick grep does show mention of "201710L" in cpp.info
>>>
>> Source for INFO files is not modified for DJGPP port in any way.
> I know, but it was still surprising to me, so I thought I'd mention it
> anyways. No big deal.
>
Nobody has noticed it earlier or at least not reported it to upstream. How often are You reading 
GCC own documentation. Usually one probably just search for part one is interested about.

Andris