linux-kernel - [RFC][PATCH] Setting /proc/irq/[n]/smp

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-next>] [day] [month] [year] [list]

Date:	Wed, 2 Dec 2009 10:19:35 -0600
From:	John Blackwood <john.blackwood@...r.com>
To:	Ingo Molnar <mingo@...e.hu>, Rusty Russell <rusty@...tcorp.com.au>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: [RFC][PATCH] Setting /proc/irq/[n]/smp_affinity failure on 8 cpu
 systems

Hi,

I would like to mention an irq set affinity issue occurs on an
older Dell PowerEdge 6650 system:
    - 8 cpus (4 are hyper-threaded),
    - uses the struct apic 'apic_default' support
    - 32 bit (x86) kernel

This issue is that once you change the irq smp_affinity of an irq
to be less than all cpus in the system, you can never change really the
irq smp_affinity back to be all cpus in the system (0xff) again,
even though no error status is returned on the "/bin/echo ff >
/proc/irq/[n]/smp_affinity" operation.

This is due to that fact that BAD_APICID has the same value as
all cpus (0xff) on 32bit kernels, and thus the value returned from
set_desc_affinity() via the cpu_mask_to_apicid_and() function is treated
as a failure in set_ioapic_affinity_irq_desc(), and no affinity changes
are made.

It seems like changing the BAD_APICID value from 0xFF to 0xFFFF for
CONFIG_X86_32 might be one possible solution, and it seems to function
properly for this Dell PowerEdge 6650 system.

Thank you for your time.

Signed-off-by: John Blackwood <john.blackwood@...r.com>


diff -rup linux-2.6.32-rc8/arch/x86/include/asm/apicdef.h new/arch/x86/include/asm/apicdef.h
--- linux-2.6.32-rc8/arch/x86/include/asm/apicdef.h     2009-11-19 16:32:38.000000000 -0600
+++ new/arch/x86/include/asm/apicdef.h  2009-12-01 15:15:34.000000000 -0600
@@ -413,9 +413,6 @@ struct local_apic {

    #undef u32

-#ifdef CONFIG_X86_32
- #define BAD_APICID 0xFFu
-#else
- #define BAD_APICID 0xFFFFu
-#endif
+#define BAD_APICID 0xFFFFu
+
    #endif /* _ASM_X86_APICDEF_H */
diff -rup linux-2.6.32-rc8/arch/x86/include/asm/apic.h new/arch/x86/include/asm/apic.h
--- linux-2.6.32-rc8/arch/x86/include/asm/apic.h        2009-11-19 16:32:38.000000000 -0600
+++ new/arch/x86/include/asm/apic.h     2009-12-01 15:15:34.000000000 -0600
@@ -591,7 +591,7 @@ static inline physid_mask_t default_apic
    #endif /* CONFIG_X86_LOCAL_APIC */

    #ifdef CONFIG_X86_32
-extern u8 cpu_2_logical_apicid[NR_CPUS];
+extern u16 cpu_2_logical_apicid[NR_CPUS];
    #endif

    #endif /* _ASM_X86_APIC_H */
diff -rup linux-2.6.32-rc8/arch/x86/kernel/smpboot.c new/arch/x86/kernel/smpboot.c
--- linux-2.6.32-rc8/arch/x86/kernel/smpboot.c  2009-11-19 16:32:38.000000000 -0600
+++ new/arch/x86/kernel/smpboot.c       2009-12-01 15:15:34.000000000 -0600
@@ -146,7 +146,7 @@ static void unmap_cpu_to_node(int cpu)
    #ifdef CONFIG_X86_32
    static int boot_cpu_logical_apicid;

-u8 cpu_2_logical_apicid[NR_CPUS] __read_mostly =
+u16 cpu_2_logical_apicid[NR_CPUS] __read_mostly =
                                          { [0 ... NR_CPUS-1] = BAD_APICID };

    static void map_cpu_to_logical_apicid(void)



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/