Re: linux-next: Tree for December 1 (kconfig question)

[Randy Dunlap <randy.dunlap@oracle.com>]

On 12/01/10 14:18, Zimny Lech wrote:
> 2010/12/1 Randy Dunlap <randy.dunlap@oracle.com>:
>> On Wed, 1 Dec 2010 17:04:38 +0100 Zimny Lech wrote:
>>
>>> Ave
>>>
>>> 2010/12/1 Stephen Rothwell <sfr@canb.auug.org.au>:
>>>> Hi all,
>>>>
>>>> Changes since 20101130:
>>>>
>>>
>>> I remember, I reported something similar a week or two ago
>>>
>>>   GEN     /home/test/linux-2.6-obj/Makefile
>>> scripts/kconfig/conf --silentoldconfig Kconfig
>>> .config:167:warning: override: SLOB changes choice state
>>> warning: (IA32_EMULATION && X86_64) selects COMPAT_BINFMT_ELF which
>>> has unmet direct dependencies (COMPAT && BINFMT_ELF)
>>> warning: (MTD_NAND && MTD || MTD_DOC2000 && MTD || MTD_DOC2001 && MTD
>>> || MTD_DOC2001PLUS && MTD) selects MTD_NAND_IDS which has unmet direct
>>> dependencies (MTD
>>> warning: (CRYPTO_LRW && CRYPTO && EXPERIMENTAL || CRYPTO_XTS && CRYPTO
>>> && EXPERIMENTAL || CRYPTO_GHASH && CRYPTO) selects CRYPTO_GF128MUL
>>> which has unmet dire
>>> *
>>> * Restart config...
>>> *
>>> *
>>> * Kernel hacking
>>> *
>>> Enable __deprecated logic (ENABLE_WARN_DEPRECATED) [Y/n/?] y
>>> Enable __must_check logic (ENABLE_MUST_CHECK) [N/y/?] n
>>> Warn for stack frames larger than (needs gcc 4.4) (FRAME_WARN) [2048] 2048
>>> Magic SysRq key (MAGIC_SYSRQ) [N/y/?] n
>>> Strip assembler-generated symbols during link (STRIP_ASM_SYMS) [N/y/?] n
>>> Enable unused/obsolete exported symbols (UNUSED_SYMBOLS) [Y/n/?] y
>>> Debug Filesystem (DEBUG_FS) [Y/?] y
>>> Run 'make headers_check' when building vmlinux (HEADERS_CHECK) [N/y/?] n
>>> Kernel debugging (DEBUG_KERNEL) [Y/n/?] y
>>>   Debug shared IRQ handlers (DEBUG_SHIRQ) [N/y/?] n
>>>   Detect Hard and Soft Lockups (LOCKUP_DETECTOR) [Y/n/?] y
>>> Panic (Reboot) On Soft Lockups (BOOTPARAM_SOFTLOCKUP_PANIC) [Y/n/?] y
>>> Detect Hung Tasks (DETECT_HUNG_TASK) [N/y/?] n
>>> Debug object operations (DEBUG_OBJECTS) [N/y/?] n
>>> RT Mutex debugging, deadlock detection (DEBUG_RT_MUTEXES) [Y/n/?] y
>>> Built-in scriptable tester for rt-mutexes (RT_MUTEX_TESTER) [Y/n/?] y
>>> Spinlock and rw-lock debugging: basic checks (DEBUG_SPINLOCK) [Y/?] y
>>> Mutex debugging: basic checks (DEBUG_MUTEXES) [Y/?] y
>>> Lock debugging: detect incorrect freeing of live locks
>>> (DEBUG_LOCK_ALLOC) [Y/?] y
>>> Lock debugging: prove locking correctness (PROVE_LOCKING) [Y/n/?] y
>>>   RCU debugging: prove RCU correctness (PROVE_RCU) [Y/n/?] y
>>>     RCU debugging: don't disable PROVE_RCU on first splat
>>> (PROVE_RCU_REPEATEDLY) [N/y/?] n
>>> RCU debugging: sparse-based checks for pointer usage
>>> (SPARSE_RCU_POINTER) [Y/n/?] y
>>> Lock usage statistics (LOCK_STAT) [N/y/?] n
>>> Lock dependency engine debugging (DEBUG_LOCKDEP) [Y/n/?] y
>>> Spinlock debugging: sleep-inside-spinlock checking
>>> (DEBUG_SPINLOCK_SLEEP) [Y/n/?] y
>>> Locking API boot-time self-tests (DEBUG_LOCKING_API_SELFTESTS) [N/y/?] n
>>> kobject debugging (DEBUG_KOBJECT) [N/y/?] n
>>> Compile the kernel with debug info (DEBUG_INFO) [N/y/?] n
>>> Debug VM (DEBUG_VM) [Y/n/?] y
>>> Debug VM translations (DEBUG_VIRTUAL) [Y/n/?] y
>>> Debug filesystem writers count (DEBUG_WRITECOUNT) [N/y/?] n
>>> Debug memory initialisation (DEBUG_MEMORY_INIT) [Y/n/?] y
>>> Debug linked list manipulation (DEBUG_LIST) [Y/n/?] y
>>> Linked list sorting test (TEST_LIST_SORT) [N/y/?] n
>>> Debug SG table operations (DEBUG_SG) [N/y/?] n
>>> Debug notifier call chains (DEBUG_NOTIFIERS) [N/y/?] n
>>> Debug credential management (DEBUG_CREDENTIALS) [Y/n/?] y
>>> Compile the kernel with frame pointers (FRAME_POINTER) [Y/?] y
>>> torture tests for RCU (RCU_TORTURE_TEST) [N/y/?] n
>>> Check for stalled CPUs delaying RCU grace periods
>>> (RCU_CPU_STALL_DETECTOR) [Y/n/?] y
>>>   RCU CPU stall timeout in seconds (RCU_CPU_STALL_TIMEOUT) [60] 60
>>>   RCU CPU stall checking starts automatically at boot
>>> (RCU_CPU_STALL_DETECTOR_RUNNABLE) [Y/n/?] y
>>> Self test for the backtrace code (BACKTRACE_SELF_TEST) [Y/n/?] y
>>> Force extended block device numbers and spread them
>>> (DEBUG_BLOCK_EXT_DEVT) [N/y/?] n
>>> Force weak per-cpu definitions (DEBUG_FORCE_WEAK_PER_CPU) [N/y/?] n
>>> Linux Kernel Dump Test Tool Module (LKDTM) [N/y/?] n
>>> CPU notifier error injection module (CPU_NOTIFIER_ERROR_INJECT) [Y/n/?] y
>>> Fault-injection framework (FAULT_INJECTION) [Y/n/?] y
>>>   Fault-injection capabilitiy for alloc_pages() (FAIL_PAGE_ALLOC) [N/y/?] n
>>>   Fault-injection capability for disk IO (FAIL_MAKE_REQUEST) [Y/n/?] y
>>>   Fault-injection capability for faking disk interrupts
>>> (FAIL_IO_TIMEOUT) [Y/n/?] y
>>>   Debugfs entries for fault-injection capabilities
>>> (FAULT_INJECTION_DEBUG_FS) [N/y/?] n
>>> Debug page memory allocations (DEBUG_PAGEALLOC) [N/y/?] (NEW) aborted!
>>>
>>> Console input/output is redirected. Run 'make oldconfig' to update
>>> configuration.
>>>
>>> make[3]: *** [silentoldconfig] Error 1
>>> make[2]: *** [silentoldconfig] Error 2
>>> make[1]: *** No rule to make target `include/config/auto.conf', needed
>>> by `include/config/kernel.release'.  Stop.
>>> make: *** [sub-make] Error 2
>>
>>
>> This is a problem.  Any idea how you achieved this?
> 
> Believe if I say that it was generated by make randconfig? Or this is
> so amazing that you would not believe? :)

Yeah, that's what I expected.

kconfig people:  Is there some way that 'make randconfig' could generate
this?

# CONFIG_SLAB is not set
CONFIG_SLUB=y
CONFIG_SLOB=y


> This build was done (as always) by my build script - the same that I
> send you once.
> 
> It's simple:
> make mrproper
> make O=../dir mrproper
> and in loop:
> make randconfig
> make
> 
> This bug appeared on third run through the loop.
> 
>>
>> # CONFIG_SLAB is not set
>> CONFIG_SLUB=y
>> CONFIG_SLOB=y
>>
>> These are part of a choice and should be mutually exclusive.


-- 
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***