From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by yocto-www.yoctoproject.org (Postfix, from userid 118) id 18D92E00C66; Tue, 29 Dec 2015 04:31:43 -0800 (PST) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on yocto-www.yoctoproject.org X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,SPF_HELO_PASS autolearn=ham version=3.3.1 X-Spam-HAM-Report: * -0.0 SPF_HELO_PASS SPF: HELO matches SPF record * -1.9 BAYES_00 BODY: Bayes spam probability is 0 to 1% * [score: 0.0000] Received: from ptmx.org (ptmx.org [178.63.28.110]) by yocto-www.yoctoproject.org (Postfix) with ESMTP id 6B609E00901 for ; Tue, 29 Dec 2015 04:31:38 -0800 (PST) Received: from [192.168.178.14] (chello062178118086.5.14.vie.surfer.at [62.178.118.86]) by ptmx.org (Postfix) with ESMTPSA id 640EF1FF36 for ; Tue, 29 Dec 2015 13:31:36 +0100 (CET) To: meta-freescale@yoctoproject.org References: <564B3FD3.1030607@pseudoterminal.org> <564B5C5E.60309@pseudoterminal.org> <564C78B6.2000700@pseudoterminal.org> <564CAE4C.3040005@pseudoterminal.org> <56784D25.3040608@pseudoterminal.org> <56791C77.1050001@pseudoterminal.org> <567928F2.4070709@pseudoterminal.org> <56793F40.7090407@pseudoterminal.org> From: Carlos Rafael Giani Message-ID: <56827D28.7010702@pseudoterminal.org> Date: Tue, 29 Dec 2015 13:31:36 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.4.0 MIME-Version: 1.0 In-Reply-To: <56793F40.7090407@pseudoterminal.org> Subject: Re: imxvpudec crash with 3.14.28 and gstreamer1.0 X-BeenThere: meta-freescale@yoctoproject.org X-Mailman-Version: 2.1.13 Precedence: list List-Id: Usage and development list for the meta-fsl-* layers List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 29 Dec 2015 12:31:43 -0000 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Writing this to the mailing list for sake of completeness: this turned out to be a memory leak in the GPU driver, imx-gpu-viv 5.0.11.p4.4. Upgrading to 5.0.11.p4.5 fixed the issue. On 2015-12-22 13:17, Carlos Rafael Giani wrote: > This is strange. For me, it looped over 160 times before I stopped it. > I'll compare your config to that of the boundarydevices kernel. (Also > note that this kernel might have patches that yours doesn't.) > > Try to repeatedly print out the contents of /proc/buddyinfo, > especially the DMA row. I did, and the bins stayed relatively stable. > If these bins rapidly decrease for you, then we might be looking at a > kernel issue, or a problem with imx-vpu. > > On 12/22/2015 12:16 PM, Vikas Patil wrote: >> Hi Carlos, >> >> Yes. After reconfiguration saw memory allocation failure after 35th >> loop. >> >> Attached here the complete log with allocation failure after 53rd time >> with master branch of plug-in and libimxvpuapi and CMA configs I >> mentioned. >> >> Regards, >> Vikash >> >> >> >> On Tue, Dec 22, 2015 at 4:11 PM, Carlos Rafael Giani >> wrote: >>> You mean, "before the CMA reconfiguration, the loop-video test >>> failed in the >>> 5th loop, but after reconfiguration, it still runs after the 37th >>> loop"? >>> >>> >>> On 2015-12-22 11:19, Vikas Patil wrote: >>>> Hi Carlos, >>>> >>>> I could run the video and see the output on dislay after commenting >>>> libgstimxaudio.so but only one loop. Attached here the log. >>>> >>>> Also after enabling following kernel configs I could run the videotest >>>> player application which I have till 37th loop before it was failing >>>> in 5th loop itself. >>>> >>>> CONFIG_DMA_CMA=y >>>> # >>>> # Default contiguous memory area size: >>>> # >>>> CONFIG_CMA_SIZE_MBYTES=256 >>>> CONFIG_CMA_SIZE_SEL_MBYTES=y >>>> # CONFIG_CMA_SIZE_SEL_PERCENTAGE is not set >>>> # CONFIG_CMA_SIZE_SEL_MIN is not set >>>> # CONFIG_CMA_SIZE_SEL_MAX is not set >>>> CONFIG_CMA_ALIGNMENT=8 >>>> CONFIG_CMA_AREAS=7 >>>> >>>> Regards, >>>> Vikas >>>> >>>> On Tue, Dec 22, 2015 at 3:18 PM, Carlos Rafael Giani >>>> wrote: >>>>> This is a different error, however. It seems to be caused by an AAC >>>>> decoding >>>>> issue. Try removing the libgstimxaudio.so file from /usr/lib/ on your >>>>> device. Note that you will need another AAC decoder then. >>>>> gst-libav can >>>>> decode AAC. So can the libgstfaad plugin (part of gst-plugins-bad). >>>>> >>>>> Would it also be possible for you to give me a copy of this >>>>> B01_Baseline1.0_1280_720.MP4 file? >>>>> >>>>> >>>>> On 2015-12-22 10:39, Vikas Patil wrote: >>>>>> Hi Carlos, >>>>>> >>>>>> Still I could not play. Attached here the log. >>>>>> >>>>>> Regards, >>>>>> Vikas >>>>>> >>>>>> On Tue, Dec 22, 2015 at 12:34 AM, Carlos Rafael Giani >>>>>> wrote: >>>>>>> Can you try out the example program I attached? Just run it like >>>>>>> this: >>>>>>> >>>>>>> GST_DEBUG=2,*imx*:5 ./loop-videos -i 5000 -v >>>>>>> "imxipuvideotransform ! >>>>>>> imxeglvivsink" /home/root/B01_Baseline1.0_1280_720.MP4 >>>>>>> >>>>>>> This will run the mp4 video for 5 seconds and then start again. >>>>>>> Note >>>>>>> that >>>>>>> you have to build it with the -std=c++11 compiler flag. >> > >