linux-kernel - Re: [PATCH v2] x86/entry_32: Move CLEAR_CPU

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: Windows password security audit tool. GUI, reports in PDF.

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

Date: Sat, 29 Jun 2024 11:41:07 -0400
From: Brian Gerst <brgerst@...il.com>
To: Pawan Gupta <pawan.kumar.gupta@...ux.intel.com>
Cc: Borislav Petkov <bp@...en8.de>, Dave Hansen <dave.hansen@...ux.intel.com>, 
	linux-kernel@...r.kernel.org, x86@...nel.org, 
	Robert Gill <rtgill82@...il.com>, Jari Ruusu <jariruusu@...tonmail.com>, 
	"Linux regression tracking (Thorsten Leemhuis)" <regressions@...mhuis.info>, antonio.gomez.iglesias@...ux.intel.com, 
	daniel.sneddon@...ux.intel.com
Subject: Re: [PATCH v2] x86/entry_32: Move CLEAR_CPU_BUFFERS before restoring segments

On Thu, Jun 27, 2024 at 10:57 PM Pawan Gupta
<pawan.kumar.gupta@...ux.intel.com> wrote:
>
> Robert Gill reported below #GP when dosemu software was executing vm86()
> system call:
>
>   general protection fault: 0000 [#1] PREEMPT SMP
>   CPU: 4 PID: 4610 Comm: dosemu.bin Not tainted 6.6.21-gentoo-x86 #1
>   Hardware name: Dell Inc. PowerEdge 1950/0H723K, BIOS 2.7.0 10/30/2010
>   EIP: restore_all_switch_stack+0xbe/0xcf
>   EAX: 00000000 EBX: 00000000 ECX: 00000000 EDX: 00000000
>   ESI: 00000000 EDI: 00000000 EBP: 00000000 ESP: ff8affdc
>   DS: 0000 ES: 0000 FS: 0000 GS: 0033 SS: 0068 EFLAGS: 00010046
>   CR0: 80050033 CR2: 00c2101c CR3: 04b6d000 CR4: 000406d0
>   Call Trace:
>    show_regs+0x70/0x78
>    die_addr+0x29/0x70
>    exc_general_protection+0x13c/0x348
>    exc_bounds+0x98/0x98
>    handle_exception+0x14d/0x14d
>    exc_bounds+0x98/0x98
>    restore_all_switch_stack+0xbe/0xcf
>    exc_bounds+0x98/0x98
>    restore_all_switch_stack+0xbe/0xcf
>
> This only happens when VERW based mitigations like MDS, RFDS are enabled.
> This is because segment registers can have funky values with vm86() that
> can result in #GP when executing VERW. Intel SDM vol. 2C documents the
> following behavior for VERW instruction:
>
>   #GP(0) - If a memory operand effective address is outside the CS, DS, ES,
>            FS, or GS segment limit.

This isn't limited to just VM86 mode, since any user DS that isn't a
flat segment can also cause problems.

> CLEAR_CPU_BUFFERS macro executes VERW instruction before returning to
> user space. Add CLEAR_CPU_BUFFERS to the macro RESTORE_REGS before it
> restores segment registers. In vm86 mode kernel does not support SYSCALL
> and SYSENTER instructions, so the problem is only limited to int80 path
> in 32-bit mode. In the opportunistic SYSEXIT path use
> CLEAR_CPU_BUFFERS_SAFE that ensures a sane %ds value.

The simpler fix is to use an SS segment override (verw
%ss:mds_verw_sel), since the stack segment is still valid right up to
the IRET/SYSEXIT.

Brian Gerst