Date: Thu, 28 Jan 2010 13:46:17 -0500
From: Michael Breuer <mbreuer@majjas.com>
Subject: Re: Hang: 2.6.32.4 sky2/DMAR (was [PATCH] sky2: Fix WARNING: at
 lib/dma-debug.c:902 check_sync)
In-reply-to: <20100128090835.0d93e53a@nehalam>
To: Stephen Hemminger <shemminger@linux-foundation.org>
Cc: Jarek Poplawski <jarkao2@gmail.com>, David Miller <davem@davemloft.net>,
       akpm@linux-foundation.org, flyboy@gmail.com,
       linux-kernel@vger.kernel.org, netdev@vger.kernel.org,
       Michael Chan <mchan@broadcom.com>, Don Fry <pcnet32@verizon.net>,
       Francois Romieu <romieu@fr.zoreil.com>,
       Matt Carlson <mcarlson@broadcom.com>
Message-id: <4B61DB79.4080703@majjas.com>
MIME-version: 1.0
Content-type: text/plain; charset=ISO-8859-1; format=flowed
Content-transfer-encoding: 7BIT
References: <20100120094103.GA6225@ff.dom.local> <4B58B217.8030001@majjas.com>
 <20100121204133.GB3085@del.dom.local> <4B59E7EB.3050605@majjas.com>
 <20100122215304.GA3105@del.dom.local> <4B5A2362.6000306@majjas.com>
 <20100122230605.GB3105@del.dom.local> <4B5A33D8.90501@majjas.com>
 <20100122234656.GC3105@del.dom.local> <4B5A39BD.8020305@majjas.com>
 <20100123232133.GA3487@del.dom.local> <4B605D1B.60402@majjas.com>
 <20100127085049.5b5048e9@nehalam> <4B60707F.1000608@majjas.com>
 <20100127095614.14313677@nehalam> <4B608128.7090607@majjas.com>
 <4B6089C7.4010803@majjas.com> <4B61ADF1.7060705@majjas.com>
 <4B61BEA4.1030905@majjas.com> <20100128090835.0d93e53a@nehalam>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.7)
 Gecko/20100111 Lightning/1.0b2pre Thunderbird/3.0.1
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2937
Lines: 70

On 1/28/2010 12:08 PM, Stephen Hemminger wrote:
> On Thu, 28 Jan 2010 11:43:16 -0500
> Michael Breuer<mbreuer@majjas.com>  wrote:
>
>    
>> Update: I played with dma-debug. Was being disabled due to lack of
>> memory. I forced it back on while pumping traffic through and got this:
>> Jan 28 11:39:30 mail kernel: ------------[ cut here ]------------
>> Jan 28 11:39:30 mail kernel: WARNING: at lib/dma-debug.c:902
>> check_sync+0xc1/0x43f()
>> Jan 28 11:39:30 mail kernel: Hardware name: System Product Name
>> Jan 28 11:39:30 mail kernel: sky2 0000:06:00.0: DMA-API: device driver
>> tries to sync DMA memory it has not allocated [device
>>      
> This test in dma-debug is bogus. Because the debug code matches
> dma based on address and size; and is perfectly valid to sync a value
> less than size. This is the patch I sent earlier, it isn't 100%
> correct but it will let you keep testing
> ....................................
>
>
> This should fix the dma-debug API code (and documentation), to
> avoid false positives when sync is done on a partial map.
>
> Signed-off-by: Stephen Hemminger<shemminger@vyatta.com>
>
> --- a/Documentation/DMA-API.txt	2010-01-20 15:17:01.390143729 -0800
> +++ b/Documentation/DMA-API.txt	2010-01-20 15:18:48.967875255 -0800
> @@ -377,9 +377,10 @@ void
>   pci_dma_sync_sg(struct pci_dev *hwdev, struct scatterlist *sg,
>   		       int nelems, int direction)
>
> -Synchronise a single contiguous or scatter/gather mapping.  All the
> -parameters must be the same as those passed into the single mapping
> -API.
> +Synchronise a single contiguous or scatter/gather mapping.  The
> +device and handle must be the same as those passed into the single mapping
> +API. The size can be less than the original mapping if only part
> +of the mapping needs to be accessed.
>
>   Notes:  You must do this:
>
> --- a/lib/dma-debug.c	2010-01-20 15:22:55.919519883 -0800
> +++ b/lib/dma-debug.c	2010-01-20 15:26:31.648895638 -0800
> @@ -285,11 +285,9 @@ static struct dma_debug_entry *hash_buck
>   	}
>
>   	/*
> -	 * If we have multiple matches but no perfect-fit, just return
> -	 * NULL.
> +	 * If we have multiple matches but no perfect-fit
> +	 * return best value and let caller deal with it.
>   	 */
> -	ret = (matches == 1) ? ret : NULL;
> -
>   	return ret;
>   }
>
>    
Ok - applied. Noise gone... however I'm not sure whether I'll be able to 
keep dma-debug going long enough to catch anything. num_free_entries 
keeps dropping... looks like entries are not freed. I'm running with a 
huge number for now & sky2 as the driver filter. Is there a reason that 
entries wouldn't be unmapped, or is dma-debug.c just not processing the 
unmap correctly?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/