Message ID | 1551430616-42014-1-git-send-email-wen.yang99@zte.com.cn (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | [01/15] ARM: actions: fix a leaked reference by addingmissing of_node_put | expand |
Hi Wen, On Fri, Mar 01, 2019 at 04:56:42PM +0800, Wen Yang wrote: > The call to of_get_next_child returns a node pointer with refcount > incremented thus it must be explicitly decremented after the last > usage. > > Detected by coccinelle with the following warnings: > ./arch/arm/mach-actions/platsmp.c:112:2-8: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 103, but without a corresponding object release within this function. > ./arch/arm/mach-actions/platsmp.c:124:2-8: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 115, but without a corresponding object release within this function. > ./arch/arm/mach-actions/platsmp.c:137:3-9: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 128, but without a corresponding object release within this function. > We have a floating patch for this: https://www.spinics.net/lists/arm-kernel/msg694544.html Andreas: Can you please take a second look at the patchset submitted by Linus Walleij and Russel for simplifying the Actions startup code? Thanks, Mani > Signed-off-by: Wen Yang <wen.yang99@zte.com.cn> > Cc: "Andreas Färber" <afaerber@suse.de> > Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> > Cc: Russell King <linux@armlinux.org.uk> > Cc: linux-arm-kernel@lists.infradead.org > Cc: linux-kernel@vger.kernel.org > --- > arch/arm/mach-actions/platsmp.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/arch/arm/mach-actions/platsmp.c b/arch/arm/mach-actions/platsmp.c > index 4fd479c..1a8e078 100644 > --- a/arch/arm/mach-actions/platsmp.c > +++ b/arch/arm/mach-actions/platsmp.c > @@ -107,6 +107,7 @@ static void __init s500_smp_prepare_cpus(unsigned int max_cpus) > } > > timer_base_addr = of_iomap(node, 0); > + of_node_put(node); > if (!timer_base_addr) { > pr_err("%s: could not map timer registers\n", __func__); > return; > @@ -119,6 +120,7 @@ static void __init s500_smp_prepare_cpus(unsigned int max_cpus) > } > > sps_base_addr = of_iomap(node, 0); > + of_node_put(node); > if (!sps_base_addr) { > pr_err("%s: could not map sps registers\n", __func__); > return; > @@ -132,6 +134,7 @@ static void __init s500_smp_prepare_cpus(unsigned int max_cpus) > } > > scu_base_addr = of_iomap(node, 0); > + of_node_put(node); > if (!scu_base_addr) { > pr_err("%s: could not map scu registers\n", __func__); > return; > -- > 2.9.5 >
On Fri, Mar 1, 2019 at 4:55 PM Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> wrote: > Hi Wen, > > On Fri, Mar 01, 2019 at 04:56:42PM +0800, Wen Yang wrote: > > The call to of_get_next_child returns a node pointer with refcount > > incremented thus it must be explicitly decremented after the last > > usage. > > > > Detected by coccinelle with the following warnings: > > ./arch/arm/mach-actions/platsmp.c:112:2-8: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 103, but without a corresponding object release within this function. > > ./arch/arm/mach-actions/platsmp.c:124:2-8: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 115, but without a corresponding object release within this function. > > ./arch/arm/mach-actions/platsmp.c:137:3-9: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 128, but without a corresponding object release within this function. > > > > We have a floating patch for this: > https://www.spinics.net/lists/arm-kernel/msg694544.html > > Andreas: Can you please take a second look at the patchset submitted by Linus > Walleij and Russel for simplifying the Actions startup code? Andreas wrote a version of simplifying secondary startup in the same spirit as Russell's patches, and it's merged and all is fine I think. If this patch applied on top of the current upstream code I'd say just forget about my patch and merge Wen's patch instead. Yours, Linus Walleij
Hi Linus, On Mon, Mar 04, 2019 at 01:29:33PM +0100, Linus Walleij wrote: > On Fri, Mar 1, 2019 at 4:55 PM Manivannan Sadhasivam > <manivannan.sadhasivam@linaro.org> wrote: > > Hi Wen, > > > > On Fri, Mar 01, 2019 at 04:56:42PM +0800, Wen Yang wrote: > > > The call to of_get_next_child returns a node pointer with refcount > > > incremented thus it must be explicitly decremented after the last > > > usage. > > > > > > Detected by coccinelle with the following warnings: > > > ./arch/arm/mach-actions/platsmp.c:112:2-8: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 103, but without a corresponding object release within this function. > > > ./arch/arm/mach-actions/platsmp.c:124:2-8: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 115, but without a corresponding object release within this function. > > > ./arch/arm/mach-actions/platsmp.c:137:3-9: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 128, but without a corresponding object release within this function. > > > > > > > We have a floating patch for this: > > https://www.spinics.net/lists/arm-kernel/msg694544.html > > > > Andreas: Can you please take a second look at the patchset submitted by Linus > > Walleij and Russel for simplifying the Actions startup code? > > Andreas wrote a version of simplifying secondary startup in the > same spirit as Russell's patches, and it's merged and all > is fine I think. > Oops. I think I missed that! Can you please point me to that patch? And how it got merged? I did the PR for actions stuff this time and haven't included any mach-actions patches. Thanks, Mani > If this patch applied on top of the current upstream code I'd say > just forget about my patch and merge Wen's patch instead. > > Yours, > Linus Walleij
On Mon, Mar 4, 2019 at 3:48 PM Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> wrote: > On Mon, Mar 04, 2019 at 01:29:33PM +0100, Linus Walleij wrote: > > On Fri, Mar 1, 2019 at 4:55 PM Manivannan Sadhasivam > > <manivannan.sadhasivam@linaro.org> wrote: > > > Hi Wen, > > > > > > On Fri, Mar 01, 2019 at 04:56:42PM +0800, Wen Yang wrote: > > > > The call to of_get_next_child returns a node pointer with refcount > > > > incremented thus it must be explicitly decremented after the last > > > > usage. > > > > > > > > Detected by coccinelle with the following warnings: > > > > ./arch/arm/mach-actions/platsmp.c:112:2-8: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 103, but without a corresponding object release within this function. > > > > ./arch/arm/mach-actions/platsmp.c:124:2-8: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 115, but without a corresponding object release within this function. > > > > ./arch/arm/mach-actions/platsmp.c:137:3-9: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 128, but without a corresponding object release within this function. > > > > > > > > > > We have a floating patch for this: > > > https://www.spinics.net/lists/arm-kernel/msg694544.html > > > > > > Andreas: Can you please take a second look at the patchset submitted by Linus > > > Walleij and Russel for simplifying the Actions startup code? > > > > Andreas wrote a version of simplifying secondary startup in the > > same spirit as Russell's patches, and it's merged and all > > is fine I think. > > > > Oops. I think I missed that! Can you please point me to that patch? And how it > got merged? I did the PR for actions stuff this time and haven't included any > mach-actions patches. I just did git log arch/arm/mach-actions but I think it came in quite some time ago, not last merge window: But you see: commit 6c2eb3e76fb84e2eb46d484f71fab469c0d9532c "ARM: owl: smp: Drop owl_secondary_boot()" commit bad29933fef76fb6ee577f4a0b6d145c1f52f663 "ARM: owl: smp: Use __pa_symbol()" commit 18cfd9429d8a82c49add8f3ca9d366599bfcac45 "ARM: owl: smp: Drop bogus holding pen" platsmp.c looks just fine these days. Except for what Wei's patch is fixing, of_node_put(). Yours, Linus Walleij
On Tue, Mar 05, 2019 at 07:21:19AM +0530, Manivannan Sadhasivam wrote: > On Tue, 5 Mar, 2019, 4:08 AM Linus Walleij <linus.walleij@linaro.org wrote: > > > On Mon, Mar 4, 2019 at 3:48 PM Manivannan Sadhasivam > > <manivannan.sadhasivam@linaro.org> wrote: > > > On Mon, Mar 04, 2019 at 01:29:33PM +0100, Linus Walleij wrote: > > > > On Fri, Mar 1, 2019 at 4:55 PM Manivannan Sadhasivam > > > > <manivannan.sadhasivam@linaro.org> wrote: > > > > > Hi Wen, > > > > > > > > > > On Fri, Mar 01, 2019 at 04:56:42PM +0800, Wen Yang wrote: > > > > > > The call to of_get_next_child returns a node pointer with refcount > > > > > > incremented thus it must be explicitly decremented after the last > > > > > > usage. > > > > > > > > > > > > Detected by coccinelle with the following warnings: > > > > > > ./arch/arm/mach-actions/platsmp.c:112:2-8: ERROR: missing > > of_node_put; acquired a node pointer with refcount incremented on line 103, > > but without a corresponding object release within this function. > > > > > > ./arch/arm/mach-actions/platsmp.c:124:2-8: ERROR: missing > > of_node_put; acquired a node pointer with refcount incremented on line 115, > > but without a corresponding object release within this function. > > > > > > ./arch/arm/mach-actions/platsmp.c:137:3-9: ERROR: missing > > of_node_put; acquired a node pointer with refcount incremented on line 128, > > but without a corresponding object release within this function. > > > > > > > > > > > > > > > > We have a floating patch for this: > > > > > https://www.spinics.net/lists/arm-kernel/msg694544.html > > > > > > > > > > Andreas: Can you please take a second look at the patchset submitted > > by Linus > > > > > Walleij and Russel for simplifying the Actions startup code? > > > > > > > > Andreas wrote a version of simplifying secondary startup in the > > > > same spirit as Russell's patches, and it's merged and all > > > > is fine I think. > > > > > > > > > > Oops. I think I missed that! Can you please point me to that patch? And > > how it > > > got merged? I did the PR for actions stuff this time and haven't > > included any > > > mach-actions patches. > > > > I just did git log arch/arm/mach-actions but I think it came in quite some > > time > > ago, not last merge window: > > > > But you see: > > commit 6c2eb3e76fb84e2eb46d484f71fab469c0d9532c > > "ARM: owl: smp: Drop owl_secondary_boot()" > > commit bad29933fef76fb6ee577f4a0b6d145c1f52f663 > > "ARM: owl: smp: Use __pa_symbol()" > > commit 18cfd9429d8a82c49add8f3ca9d366599bfcac45 > > "ARM: owl: smp: Drop bogus holding pen" > > > > platsmp.c looks just fine these days. Except for what Wei's patch is > > fixing, > > of_node_put(). > > > > Nope. platsmp.c still requires some cleanup like removing the redundant > bootlock and pen_release flag as pointed out by Russel. Andreas just > replied to your cleanup patches but there was no follow up since that. So, > I guess we can just apply Russell's patches and this patch once Andreas is > fine with it (it looks good to me though). No. My patches are in my tree queued for this merge window. They are part of a series that can not be broken up and merged separately because all the per-platform patches need to be merged before the final patch "ARM: smp: remove arch-provided "pen_release"" otherwise the platforms break. I thought that was already explained.
diff --git a/arch/arm/mach-actions/platsmp.c b/arch/arm/mach-actions/platsmp.c index 4fd479c..1a8e078 100644 --- a/arch/arm/mach-actions/platsmp.c +++ b/arch/arm/mach-actions/platsmp.c @@ -107,6 +107,7 @@ static void __init s500_smp_prepare_cpus(unsigned int max_cpus) } timer_base_addr = of_iomap(node, 0); + of_node_put(node); if (!timer_base_addr) { pr_err("%s: could not map timer registers\n", __func__); return; @@ -119,6 +120,7 @@ static void __init s500_smp_prepare_cpus(unsigned int max_cpus) } sps_base_addr = of_iomap(node, 0); + of_node_put(node); if (!sps_base_addr) { pr_err("%s: could not map sps registers\n", __func__); return; @@ -132,6 +134,7 @@ static void __init s500_smp_prepare_cpus(unsigned int max_cpus) } scu_base_addr = of_iomap(node, 0); + of_node_put(node); if (!scu_base_addr) { pr_err("%s: could not map scu registers\n", __func__); return;
The call to of_get_next_child returns a node pointer with refcount incremented thus it must be explicitly decremented after the last usage. Detected by coccinelle with the following warnings: ./arch/arm/mach-actions/platsmp.c:112:2-8: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 103, but without a corresponding object release within this function. ./arch/arm/mach-actions/platsmp.c:124:2-8: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 115, but without a corresponding object release within this function. ./arch/arm/mach-actions/platsmp.c:137:3-9: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 128, but without a corresponding object release within this function. Signed-off-by: Wen Yang <wen.yang99@zte.com.cn> Cc: "Andreas Färber" <afaerber@suse.de> Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Cc: Russell King <linux@armlinux.org.uk> Cc: linux-arm-kernel@lists.infradead.org Cc: linux-kernel@vger.kernel.org --- arch/arm/mach-actions/platsmp.c | 3 +++ 1 file changed, 3 insertions(+)