2005-03-10 00:03:26

by Bagalkote, Sreenivas

[permalink] [raw]
Subject: RE: [ANNOUNCE][PATCH 2.6.11 2/3] megaraid_sas: Announcing new mod ule for LSI Logic's SAS based MegaRAID controllers

>> >
>> >Even for kernels with a 64bit dma_addr_t you can get 32bit dma
>> >addresses
>> >only. As a start check whether the pci_set_dma_mask for
>the 64bit mask
>> >failed - in that case you can always use 32bit SGLs.
>> >
>>
>> Please help me understand: If dma_addr_t is 64 bit, I will get 64bit
>> addresses in scatterlist regardless the outcome of pci_set_dma_mask,
>> won't I? These addresses may have valid or null high
>addresses. My idea
>> was to have 32(64) bit SGLs for 32(64) bit dma_addr_t.
>
>if pci_set_dma_mask for the 64bit mask fails the upper 32bit bit of
>dma_addr_t will guaranteed to be zero, so you don't need to take them
>into account for your hardware SGL.
>

Okay. I understood. Thanks. But please consider this also:

Arjan pointed out that by basing my 32/64 SGL decision solely on
#define IS_DMA64 sizeof(dma_addr_t)==8, I can rely on gcc to optimize the
unused portion of the code away. If I take pci_set_dma_mask also into
account, I can't do that, can I?

Thanks,
Sreenivas