From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pa0-x242.google.com (mail-pa0-x242.google.com [IPv6:2607:f8b0:400e:c03::242]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 3rVF1J5sKVzDqh8 for ; Thu, 16 Jun 2016 04:10:28 +1000 (AEST) Received: by mail-pa0-x242.google.com with SMTP id us13so1975590pab.1 for ; Wed, 15 Jun 2016 11:10:28 -0700 (PDT) Subject: Re: [PATCH] Remove a tiny memory leak in drivers/of To: Mathieu Malaterre , linuxppc-dev@lists.ozlabs.org References: <1466002919-535-1-git-send-email-mathieu.malaterre@gmail.com> Cc: robh+dt@kernel.org, devicetree@vger.kernel.org From: Frank Rowand Message-ID: <57619A0D.5080906@gmail.com> Date: Wed, 15 Jun 2016 11:10:21 -0700 MIME-Version: 1.0 In-Reply-To: <1466002919-535-1-git-send-email-mathieu.malaterre@gmail.com> Content-Type: text/plain; charset=windows-1252 List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Mathieu, On 06/15/16 08:01, Mathieu Malaterre wrote: > On my PowerMac device-tree would generate a duplicate name: > > [ 0.023043] device-tree: Duplicate name in PowerPC,G4@0, renamed to "l2-cache#1" > > in this case a newly allocated name is generated by `safe_name`. However > in this case it is never deallocated. > > The bug was found using kmemleak reported as: > > unreferenced object 0xdf532e60 (size 32): > comm "swapper", pid 1, jiffies 4294892300 (age 1993.532s) > hex dump (first 32 bytes): > 6c 32 2d 63 61 63 68 65 23 31 00 dd e4 dd 1e c2 l2-cache#1...... > ec d4 ba ce 04 ec cc de 8e 85 e9 ca c4 ec cc 9e ................ > backtrace: > [] kvasprintf+0x64/0xc8 > [] kasprintf+0x4c/0x5c > [] safe_name.isra.1+0x80/0xc4 > [] __of_attach_node_sysfs+0x6c/0x11c > [] of_core_init+0x8c/0xf8 > [] kernel_init_freeable+0xd4/0x208 > [] kernel_init+0x24/0x11c > [] ret_from_kernel_thread+0x5c/0x64 > > Link: https://bugzilla.kernel.org/show_bug.cgi?id=120331 > Signed-off-by: Mathieu Malaterre > --- > drivers/of/base.c | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) > > diff --git a/drivers/of/base.c b/drivers/of/base.c > index ebf84e3..3cb11df 100644 > --- a/drivers/of/base.c > +++ b/drivers/of/base.c > @@ -174,11 +174,15 @@ int __of_attach_node_sysfs(struct device_node *np) > rc = kobject_add(&np->kobj, NULL, "%s", > safe_name(&of_kset->kobj, "base")); > } else { > - name = safe_name(&np->parent->kobj, kbasename(np->full_name)); > + const char *orig_name = kbasename(np->full_name); > + > + name = safe_name(&np->parent->kobj, orig_name); > if (!name || !name[0]) > return -EINVAL; > > rc = kobject_add(&np->kobj, &np->parent->kobj, "%s", name); > + if (name != orig_name) > + kfree(name); > } > if (rc) > return rc; > Thanks for reporting the issue and the cause of the issue. You have exposed an issue that affects additional safe_name() call sites that the proposed patch does not fix. I will send you another patch that also addresses the other safe_name() call sites, please test it to see if it fixes the memory leak that you reported. -Frank