linux-kernel - Re: [PATCHv6 07/16] x86/vsyscall: Reorganize the #PF emulation code

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <262c0fd2-ac66-4ce7-903f-4062f1fe1d6e@citrix.com>
Date: Sat, 21 Jun 2025 00:08:51 +0100
From: Andrew Cooper <andrew.cooper3@...rix.com>
To: dave.hansen@...el.com
Cc: acme@...hat.com, aik@....com, akpm@...ux-foundation.org,
 alexander.shishkin@...ux.intel.com, ardb@...nel.org, ast@...nel.org,
 bp@...en8.de, brijesh.singh@....com, changbin.du@...wei.com,
 christophe.leroy@...roup.eu, corbet@....net, daniel.sneddon@...ux.intel.com,
 dave.hansen@...ux.intel.com, ebiggers@...gle.com, geert+renesas@...der.be,
 houtao1@...wei.com, hpa@...or.com, jgg@...pe.ca, jgross@...e.com,
 jpoimboe@...nel.org, kai.huang@...el.com, kees@...nel.org,
 kirill.shutemov@...ux.intel.com, leitao@...ian.org,
 linux-doc@...r.kernel.org, linux-efi@...r.kernel.org,
 linux-kernel@...r.kernel.org, linux-mm@...ck.org, linux@...musvillemoes.dk,
 luto@...nel.org, mcgrof@...nel.org, mhiramat@...nel.org,
 michael.roth@....com, mingo@...nel.org, mingo@...hat.com,
 namhyung@...nel.org, paulmck@...nel.org, pawan.kumar.gupta@...ux.intel.com,
 peterz@...radead.org, rick.p.edgecombe@...el.com, rppt@...nel.org,
 sandipan.das@....com, shijie@...amperecomputing.com, sohil.mehta@...el.com,
 tglx@...utronix.de, tj@...nel.org, tony.luck@...el.com,
 vegard.nossum@...cle.com, x86@...nel.org, xin3.li@...el.com,
 xiongwei.song@...driver.com, ytcoode@...il.com
Subject: Re: [PATCHv6 07/16] x86/vsyscall: Reorganize the #PF emulation code

> But, the resulting code is wonky. It needs to do something more like this:
>
> 	if ((error_code & (X86_PF_WRITE | X86_PF_USER)) != X86_PF_USER)
> 		return false;
>
> 	if (error_code & X86_PF_INSTR))
> 		return __emulate_vsyscall(regs, address);

To do this, LASS needs a proper interlink against NX || SMEP.

If neither NX nor SMEP are active, the CPU does not report X86_PF_INSTR,
meaning that fetches are reported as plain reads.

This leads to some fun corner cases in SMAP and now LASS too for virt.

~Andrew