From mboxrd@z Thu Jan 1 00:00:00 1970 Path: news.gmane.io!.POSTED.blaine.gmane.org!not-for-mail From: Andy Moreton Newsgroups: gmane.emacs.bugs Subject: bug#32605: [w64] (random) never returns negative Date: Fri, 13 Aug 2021 22:12:29 +0100 Message-ID: <861r6xoxqa.fsf@gmail.com> References: <855zzpf86u.fsf@gmail.com> <87zhx1ktp0.fsf@gmx.net> <87zhwwhp9i.fsf@gmail.com> <87mtpmls3p.fsf_-_@gnus.org> <83o8a2dbjo.fsf@gnu.org> <86bl62s8qm.fsf@gmail.com> <83czqhdfhm.fsf@gnu.org> Mime-Version: 1.0 Content-Type: text/plain Injection-Info: ciao.gmane.io; posting-host="blaine.gmane.org:116.202.254.214"; logging-data="38127"; mail-complaints-to="usenet@ciao.gmane.io" User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/28.0.50 (windows-nt) To: 32605@debbugs.gnu.org Cancel-Lock: sha1:zL2IYZM7/vzv4HBTkMHnscxxq0k= Original-X-From: bug-gnu-emacs-bounces+geb-bug-gnu-emacs=m.gmane-mx.org@gnu.org Fri Aug 13 23:13:10 2021 Return-path: Envelope-to: geb-bug-gnu-emacs@m.gmane-mx.org Original-Received: from lists.gnu.org ([209.51.188.17]) by ciao.gmane.io with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1mEeUI-0009pl-3J for geb-bug-gnu-emacs@m.gmane-mx.org; Fri, 13 Aug 2021 23:13:10 +0200 Original-Received: from localhost ([::1]:45110 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mEeUG-0000Fk-TH for geb-bug-gnu-emacs@m.gmane-mx.org; Fri, 13 Aug 2021 17:13:08 -0400 Original-Received: from eggs.gnu.org ([2001:470:142:3::10]:37706) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mEeUA-0000FZ-Kq for bug-gnu-emacs@gnu.org; Fri, 13 Aug 2021 17:13:02 -0400 Original-Received: from debbugs.gnu.org ([209.51.188.43]:59824) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1mEeUA-0001vN-EG for bug-gnu-emacs@gnu.org; Fri, 13 Aug 2021 17:13:02 -0400 Original-Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1mEeUA-0005qL-75 for bug-gnu-emacs@gnu.org; Fri, 13 Aug 2021 17:13:02 -0400 X-Loop: help-debbugs@gnu.org In-Reply-To: <855zzpf86u.fsf@gmail.com> Resent-From: Andy Moreton Original-Sender: "Debbugs-submit" Resent-CC: bug-gnu-emacs@gnu.org Resent-Date: Fri, 13 Aug 2021 21:13:02 +0000 Resent-Message-ID: Resent-Sender: help-debbugs@gnu.org X-GNU-PR-Message: followup 32605 X-GNU-PR-Package: emacs X-GNU-PR-Keywords: confirmed X-Debbugs-Original-To: bug-gnu-emacs@gnu.org Original-Received: via spool by submit@debbugs.gnu.org id=B.162888916822441 (code B ref -1); Fri, 13 Aug 2021 21:13:02 +0000 Original-Received: (at submit) by debbugs.gnu.org; 13 Aug 2021 21:12:48 +0000 Original-Received: from localhost ([127.0.0.1]:43137 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1mEeTv-0005ps-OR for submit@debbugs.gnu.org; Fri, 13 Aug 2021 17:12:47 -0400 Original-Received: from lists.gnu.org ([209.51.188.17]:60844) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1mEeTq-0005pg-MR for submit@debbugs.gnu.org; Fri, 13 Aug 2021 17:12:45 -0400 Original-Received: from eggs.gnu.org ([2001:470:142:3::10]:37696) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mEeTq-0000FL-CW for bug-gnu-emacs@gnu.org; Fri, 13 Aug 2021 17:12:42 -0400 Original-Received: from ciao.gmane.io ([116.202.254.214]:57540) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mEeTp-0001br-2o for bug-gnu-emacs@gnu.org; Fri, 13 Aug 2021 17:12:42 -0400 Original-Received: from list by ciao.gmane.io with local (Exim 4.92) (envelope-from ) id 1mEeTl-0009AX-2g for bug-gnu-emacs@gnu.org; Fri, 13 Aug 2021 23:12:37 +0200 X-Injected-Via-Gmane: http://gmane.org/ Received-SPF: pass client-ip=116.202.254.214; envelope-from=geb-bug-gnu-emacs@m.gmane-mx.org; helo=ciao.gmane.io X-Spam_score_int: 5 X-Spam_score: 0.5 X-Spam_bar: / X-Spam_report: (0.5 / 5.0 requ) BAYES_00=-1.9, DKIM_ADSP_CUSTOM_MED=0.001, FORGED_GMAIL_RCVD=1, FREEMAIL_FORGED_FROMDOMAIN=0.249, FREEMAIL_FROM=0.001, HEADER_FROM_DIFFERENT_DOMAINS=0.25, NML_ADSP_CUSTOM_MED=0.9, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=no autolearn_force=no X-Spam_action: no action X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list X-BeenThere: bug-gnu-emacs@gnu.org List-Id: "Bug reports for GNU Emacs, the Swiss army knife of text editors" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: bug-gnu-emacs-bounces+geb-bug-gnu-emacs=m.gmane-mx.org@gnu.org Original-Sender: "bug-gnu-emacs" Xref: news.gmane.io gmane.emacs.bugs:211789 Archived-At: On Fri 13 Aug 2021, Eli Zaretskii wrote: >> From: Andy Moreton >> Date: Thu, 12 Aug 2021 21:34:09 +0100 >> >> > int val = ((rand_as183 () << 15) | rand_as183 ()); >> > #ifdef __x86_64__ >> > return 2 * val - 0x3FFFFFFF; >> > #else >> > return val; >> > #endif >> > >> > Andy, can you test this, please? >> >> That does not produce any negative random numbers within a reasonable >> number of attempts (a few dozen calls). > > Thanks for testing. You elided the detail of my previous message: Surely real the problem is that RAND_BITS is 31, but the random() in w32.c does not provide 31 random bits (and thus fails to meet the API contract). In 32bit builds this problem is hidden because 30 bits are sufficient for a fixnum, so the value of bit30 in the result is ignored. On 64bit builds, 62 bits are needed for a fixnum, and trying to assemble a random number from multiple components does not work if RAND_BITS says 31 bits are usable, but the highest bit in that value is always zero. Please answer that. This function appears to not work properly at all. >> Instead, calling rand_as183 again (as below) does produce positive and >> negative random numbers on 32bit and 64bit builds with a similar number >> of attempts: >> >> return ((rand_as183 () << 30) | (rand_as183 () << 15) | rand_as183 ()); >> >> While this may be less efficient, it at least meets the contract of >> providing 31 random bits. > > What about the variant below, does it produce better results? > > int val = ((rand_as183 () << 15) | rand_as183 ()); > #ifdef __x86_64__ > return 2 * val - 0x7FFFFFFF; > #else > return val; > #endif Why is this any better ? On 32bit builds it does not return 31 random bits (only a 30bit value) and on 64bit builds the lowest bit is not random. I'm not sure I see the point of this bit manipulation as it does not spread randomness throughout the bits in the returned value. AndyM