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:   Tue, 21 Nov 2017 08:39:00 -0700
From:   Shuah Khan <shuah@...nel.org>
To:     Mathieu Desnoyers <mathieu.desnoyers@...icios.com>,
        Peter Zijlstra <peterz@...radead.org>,
        "Paul E . McKenney" <paulmck@...ux.vnet.ibm.com>,
        Boqun Feng <boqun.feng@...il.com>,
        Andy Lutomirski <luto@...capital.net>,
        Dave Watson <davejwatson@...com>
Cc:     linux-kernel@...r.kernel.org, linux-api@...r.kernel.org,
        Paul Turner <pjt@...gle.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Russell King <linux@....linux.org.uk>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>,
        "H . Peter Anvin" <hpa@...or.com>, Andrew Hunter <ahh@...gle.com>,
        Andi Kleen <andi@...stfloor.org>, Chris Lameter <cl@...ux.com>,
        Ben Maurer <bmaurer@...com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Josh Triplett <josh@...htriplett.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Catalin Marinas <catalin.marinas@....com>,
        Will Deacon <will.deacon@....com>,
        Michael Kerrisk <mtk.manpages@...il.com>,
        Florian Weimer <fweimer@...hat.com>,
        linux-kselftest@...r.kernel.org,
        Shuah Khan <shuahkh@....samsung.com>,
        Shuah Khan <shuah@...nel.org>
Subject: Re: [RFC PATCH for 4.15 16/22] rseq: selftests: arm: workaround gcc
 asm size guess

On 11/21/2017 07:18 AM, Mathieu Desnoyers wrote:
> Fixes assembler errors:
> /tmp/cceKwI9a.s: Assembler messages:
> /tmp/cceKwI9a.s:849: Error: co-processor offset out of range
> 
> with gcc prior to gcc-7. This can trigger if multiple rseq inline asm
> are used within the same function.
> 
> My best guess on the cause of this issue is that gcc has a hard
> time figuring out the actual size of the inline asm, and therefore
> does not compute the offsets at which literal values can be
> placed from the program counter accurately.
> 
> Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
> CC: "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
> CC: Peter Zijlstra <peterz@...radead.org>
> CC: Paul Turner <pjt@...gle.com>
> CC: Thomas Gleixner <tglx@...utronix.de>
> CC: Andrew Hunter <ahh@...gle.com>
> CC: Andy Lutomirski <luto@...capital.net>
> CC: Andi Kleen <andi@...stfloor.org>
> CC: Dave Watson <davejwatson@...com>
> CC: Chris Lameter <cl@...ux.com>
> CC: Ingo Molnar <mingo@...hat.com>
> CC: "H. Peter Anvin" <hpa@...or.com>
> CC: Ben Maurer <bmaurer@...com>
> CC: Steven Rostedt <rostedt@...dmis.org>
> CC: Josh Triplett <josh@...htriplett.org>
> CC: Linus Torvalds <torvalds@...ux-foundation.org>
> CC: Andrew Morton <akpm@...ux-foundation.org>
> CC: Russell King <linux@....linux.org.uk>
> CC: Catalin Marinas <catalin.marinas@....com>
> CC: Will Deacon <will.deacon@....com>
> CC: Michael Kerrisk <mtk.manpages@...il.com>
> CC: Boqun Feng <boqun.feng@...il.com>
> CC: Florian Weimer <fweimer@...hat.com>
> CC: Shuah Khan <shuah@...nel.org>
> CC: linux-kselftest@...r.kernel.org
> CC: linux-api@...r.kernel.org
> ---
>  tools/testing/selftests/rseq/rseq-arm.h | 33 +++++++++++++++++++++++++++++++++
>  1 file changed, 33 insertions(+)
> 
> diff --git a/tools/testing/selftests/rseq/rseq-arm.h b/tools/testing/selftests/rseq/rseq-arm.h
> index 47953c0cef4f..6d3fda276f4d 100644
> --- a/tools/testing/selftests/rseq/rseq-arm.h
> +++ b/tools/testing/selftests/rseq/rseq-arm.h
> @@ -79,12 +79,15 @@ do {									\
>  		teardown						\
>  		"b %l[" __rseq_str(cmpfail_label) "]\n\t"
>  
> +#define rseq_workaround_gcc_asm_size_guess()	__asm__ __volatile__("")
> +
>  static inline __attribute__((always_inline))
>  int rseq_cmpeqv_storev(intptr_t *v, intptr_t expect, intptr_t newv,
>  		int cpu)
>  {
>  	RSEQ_INJECT_C(9)
>  
> +	rseq_workaround_gcc_asm_size_guess();
>  	__asm__ __volatile__ goto (
>  		RSEQ_ASM_DEFINE_TABLE(__rseq_table, 0x0, 0x0, 1f, 2f-1f, 4f)
>  		RSEQ_ASM_STORE_RSEQ_CS(1, 3f, rseq_cs)
> @@ -115,11 +118,14 @@ int rseq_cmpeqv_storev(intptr_t *v, intptr_t expect, intptr_t newv,
>  		  RSEQ_INJECT_CLOBBER
>  		: abort, cmpfail
>  	);
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 0;
>  abort:
> +	rseq_workaround_gcc_asm_size_guess();
>  	RSEQ_INJECT_FAILED
>  	return -1;
>  cmpfail:
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 1;
>  }
>  
> @@ -129,6 +135,7 @@ int rseq_cmpnev_storeoffp_load(intptr_t *v, intptr_t expectnot,
>  {
>  	RSEQ_INJECT_C(9)
>  
> +	rseq_workaround_gcc_asm_size_guess();
>  	__asm__ __volatile__ goto (
>  		RSEQ_ASM_DEFINE_TABLE(__rseq_table, 0x0, 0x0, 1f, 2f-1f, 4f)
>  		RSEQ_ASM_STORE_RSEQ_CS(1, 3f, rseq_cs)
> @@ -164,11 +171,14 @@ int rseq_cmpnev_storeoffp_load(intptr_t *v, intptr_t expectnot,
>  		  RSEQ_INJECT_CLOBBER
>  		: abort, cmpfail
>  	);
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 0;
>  abort:
> +	rseq_workaround_gcc_asm_size_guess();
>  	RSEQ_INJECT_FAILED
>  	return -1;
>  cmpfail:
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 1;
>  }
>  
> @@ -177,6 +187,7 @@ int rseq_addv(intptr_t *v, intptr_t count, int cpu)
>  {
>  	RSEQ_INJECT_C(9)
>  
> +	rseq_workaround_gcc_asm_size_guess();
>  	__asm__ __volatile__ goto (
>  		RSEQ_ASM_DEFINE_TABLE(__rseq_table, 0x0, 0x0, 1f, 2f-1f, 4f)
>  		RSEQ_ASM_STORE_RSEQ_CS(1, 3f, rseq_cs)
> @@ -203,8 +214,10 @@ int rseq_addv(intptr_t *v, intptr_t count, int cpu)
>  		  RSEQ_INJECT_CLOBBER
>  		: abort
>  	);
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 0;
>  abort:
> +	rseq_workaround_gcc_asm_size_guess();
>  	RSEQ_INJECT_FAILED
>  	return -1;
>  }
> @@ -216,6 +229,7 @@ int rseq_cmpeqv_trystorev_storev(intptr_t *v, intptr_t expect,
>  {
>  	RSEQ_INJECT_C(9)
>  
> +	rseq_workaround_gcc_asm_size_guess();
>  	__asm__ __volatile__ goto (
>  		RSEQ_ASM_DEFINE_TABLE(__rseq_table, 0x0, 0x0, 1f, 2f-1f, 4f)
>  		RSEQ_ASM_STORE_RSEQ_CS(1, 3f, rseq_cs)
> @@ -253,11 +267,14 @@ int rseq_cmpeqv_trystorev_storev(intptr_t *v, intptr_t expect,
>  		  RSEQ_INJECT_CLOBBER
>  		: abort, cmpfail
>  	);
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 0;
>  abort:
> +	rseq_workaround_gcc_asm_size_guess();
>  	RSEQ_INJECT_FAILED
>  	return -1;
>  cmpfail:
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 1;
>  }
>  
> @@ -268,6 +285,7 @@ int rseq_cmpeqv_trystorev_storev_release(intptr_t *v, intptr_t expect,
>  {
>  	RSEQ_INJECT_C(9)
>  
> +	rseq_workaround_gcc_asm_size_guess();
>  	__asm__ __volatile__ goto (
>  		RSEQ_ASM_DEFINE_TABLE(__rseq_table, 0x0, 0x0, 1f, 2f-1f, 4f)
>  		RSEQ_ASM_STORE_RSEQ_CS(1, 3f, rseq_cs)
> @@ -306,11 +324,14 @@ int rseq_cmpeqv_trystorev_storev_release(intptr_t *v, intptr_t expect,
>  		  RSEQ_INJECT_CLOBBER
>  		: abort, cmpfail
>  	);
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 0;
>  abort:
> +	rseq_workaround_gcc_asm_size_guess();
>  	RSEQ_INJECT_FAILED
>  	return -1;
>  cmpfail:
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 1;
>  }
>  
> @@ -321,6 +342,7 @@ int rseq_cmpeqv_cmpeqv_storev(intptr_t *v, intptr_t expect,
>  {
>  	RSEQ_INJECT_C(9)
>  
> +	rseq_workaround_gcc_asm_size_guess();
>  	__asm__ __volatile__ goto (
>  		RSEQ_ASM_DEFINE_TABLE(__rseq_table, 0x0, 0x0, 1f, 2f-1f, 4f)
>  		RSEQ_ASM_STORE_RSEQ_CS(1, 3f, rseq_cs)
> @@ -359,11 +381,14 @@ int rseq_cmpeqv_cmpeqv_storev(intptr_t *v, intptr_t expect,
>  		  RSEQ_INJECT_CLOBBER
>  		: abort, cmpfail
>  	);
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 0;
>  abort:
> +	rseq_workaround_gcc_asm_size_guess();
>  	RSEQ_INJECT_FAILED
>  	return -1;
>  cmpfail:
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 1;
>  }
>  
> @@ -376,6 +401,7 @@ int rseq_cmpeqv_trymemcpy_storev(intptr_t *v, intptr_t expect,
>  
>  	RSEQ_INJECT_C(9)
>  
> +	rseq_workaround_gcc_asm_size_guess();
>  	__asm__ __volatile__ goto (
>  		RSEQ_ASM_DEFINE_TABLE(__rseq_table, 0x0, 0x0, 1f, 2f-1f, 4f)
>  		"str %[src], %[rseq_scratch0]\n\t"
> @@ -442,11 +468,14 @@ int rseq_cmpeqv_trymemcpy_storev(intptr_t *v, intptr_t expect,
>  		  RSEQ_INJECT_CLOBBER
>  		: abort, cmpfail
>  	);
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 0;
>  abort:
> +	rseq_workaround_gcc_asm_size_guess();
>  	RSEQ_INJECT_FAILED
>  	return -1;
>  cmpfail:
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 1;
>  }
>  
> @@ -459,6 +488,7 @@ int rseq_cmpeqv_trymemcpy_storev_release(intptr_t *v, intptr_t expect,
>  
>  	RSEQ_INJECT_C(9)
>  
> +	rseq_workaround_gcc_asm_size_guess();
>  	__asm__ __volatile__ goto (
>  		RSEQ_ASM_DEFINE_TABLE(__rseq_table, 0x0, 0x0, 1f, 2f-1f, 4f)
>  		"str %[src], %[rseq_scratch0]\n\t"
> @@ -526,10 +556,13 @@ int rseq_cmpeqv_trymemcpy_storev_release(intptr_t *v, intptr_t expect,
>  		  RSEQ_INJECT_CLOBBER
>  		: abort, cmpfail
>  	);
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 0;
>  abort:
> +	rseq_workaround_gcc_asm_size_guess();
>  	RSEQ_INJECT_FAILED
>  	return -1;
>  cmpfail:
> +	rseq_workaround_gcc_asm_size_guess();
>  	return 1;
>  }
> 

Looks fine to me. For this patch.

Acked-by: Shuah Khan <shuahkh@....samsung.com>

I have comments on other patches in this series.

thanks,
-- Shuah

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ