svn commit: r212061 - head/sys/dev/bge
Pyun YongHyeon
pyunyh at gmail.com
Sat Sep 4 06:27:16 UTC 2010
On Fri, Sep 3, 2010 at 9:29 PM, Alexander Kabaev <kabaev at gmail.com> wrote:
> On Tue, 31 Aug 2010 17:33:48 +0000 (UTC)
> Pyun YongHyeon <yongari at FreeBSD.org> wrote:
>
>> Author: yongari
>> Date: Tue Aug 31 17:33:48 2010
>> New Revision: 212061
>> URL: http://svn.freebsd.org/changeset/base/212061
>>
>> Log:
>> Split common parent DMA tag into ring DMA tag and TX/RX mbuf DMA
>> tag. All controllers that are not BCM5755 or higher have 4GB
>> boundary DMA bug. Previously bge(4) used 32bit DMA address to
>> workaround the bug(r199670). However this caused the use of bounce
>> buffers such that it resulted in poor performance for systems which
>> have more than 4GB memory. Because bus_dma(9) honors boundary
>> restriction requirement of DMA tag for dynamic buffers, having a
>> separate TX/RX mbuf DMA tag will greatly reduce the possibility of
>> using bounce buffers. For DMA buffers allocated with
>> bus_dmamem_alloc(9), now bge(4) explicitly checks whether the
>> requested memory region crossed the boundary or not.
>> With this change, only the DMA buffer that crossed the boundary
>> will use 32bit DMA address. Other DMA buffers are not affected as
>> separate DMA tag is created for each DMA buffer.
>> Even if 32bit DMA address space is used for a buffer, the chance to
>> use bounce buffer is still very low as the size of buffer is small.
>> This change should eliminate most usage of bounce buffers on
>> systems that have more than 4GB memory.
>>
>> More correct fix would be teaching bus_dma(9) to honor boundary
>> restriction for buffers created with bus_dmamem_alloc(9) but it
>> seems that is not easy.
>>
>> While I'm here cleanup bge_dma_map_addr() and remove unnecessary
>> member variables in bge_dmamap_arg structure.
>>
>> Tested by: marcel
>>
>> Modified:
>> head/sys/dev/bge/if_bge.c
>> head/sys/dev/bge/if_bgereg.h
>>
>
> Hi,
>
> with this commit my AMD64 machine reliably reboots after several
> minutes of uptime. Attempting cvs update of ports from repository
> located on NFS server seems to make it happen sooner.
>
>
> bge0: <Broadcom NetXtreme Gigabit Ethernet Controller, ASIC rev.
> 0x001002> mem 0xfd8f0000-0xfd8fffff irq 24 at device 9.0 on pci2
> miibus0: <MII bus> on bge0
> brgphy0: <BCM5703 10/100/1000baseTX PHY> PHY 1 on miibus0
> brgphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT,
> 1000baseT-FDX, auto
> bge0: Ethernet address: 00:e0:81:28:24:e7
> bge0: [ITHREAD]
>
> Booting kernel tree from just before this commit makes everything work
> again.
Sorry, it seems I forgot allocating statistics block. This may cause
issues for controllers(e.g. 570x) that support hardware MAC statistics
counters. This could be the reason why I couldn't see the issue on 5761.
I'm on vacation so I'll fix it Tuesday. If it's urgent please back out this
change.
> --
> Alexander Kabaev
>
More information about the svn-src-all
mailing list