kmemleak issue on ARM target
in [Kernel]
|
Prev: [Patch 1/2] cpufreq: unexport (un)lock_policy_rwsem* functions
Next: evdev keyboard driver stopped working after upgrading from Kernel 2.6.33-rc8 to 2.6.33 (official release)
From: naveen yadav on 4 Mar 2010 04:20 Hi Catalin Marinas , W am facing one issue on ARM target, we have 512 MB ram on our target, we port your patch of kmemleak(http://linux.derkeiler.com/Mailing-Lists/Kernel/2009-04/msg11830.html) We are facing problem in DEBUG_KMEMLEAK_EARLY_LOG_SIZE we cannot increase its size above 1000 because of our kernel Image size for embedded board has some limit that if it increase we cannot execute it. so is there any implementaion possible using vmalloc and not statically allocating the log of array or else any suggestion. kind regards Naveen -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Catalin Marinas on 4 Mar 2010 05:20
On Thu, 2010-03-04 at 09:13 +0000, naveen yadav wrote: > W am facing one issue on ARM target, we have 512 MB ram on our target, > we port your patch of > kmemleak(http://linux.derkeiler.com/Mailing-Lists/Kernel/2009-04/msg11830.html) > > We are facing problem in DEBUG_KMEMLEAK_EARLY_LOG_SIZE we cannot > increase its size above 1000 because of our kernel Image size for > embedded board > has some limit that if it increase we cannot execute it. so is there > any implementaion possible using vmalloc and not statically allocating > the log of array or else any suggestion. Not really. This buffer needs to be static because it is used before kmemleak is fully initialised. It's also tracking bootmem allocations. But this size should not increase the Image file size as it should go in the BSS section. An additional question - why do you need to increase this size? I found 400 to be enough usually. Do you get any errors? Since you backported kmemleak, please make sure that you check the latest code in mm/kmemleak.c as there are some bug-fixes. -- Catalin -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |