On 05/13/2015 03:56 PM, Benjamin Herrenschmidt wrote: > On Wed, 2015-05-13 at 22:14 +0530, Sreekanth Reddy wrote: >> Hi Brain, >> >> We had a restriction from the firmware that the upper 32 bits of the >> RDPQ pool entries must be same. So to limit this restriction we >> initially set the consistent DMA mask to 32 and then after allocating >> the RDPQ pools we changed the consistent DMA mask to 64 before >> allocating remaining pools. > > Brian, maybe we should just bite that bullet and implement that hybrid > DMA ops I mentioned. I'll see if I can spare some time today to look > at it. It would work as long as we never remove the legacy window using > the DDW APIs (removing the legacy window is broken anyway for other > reasons I think we discussed already). > > As long as we have both the legacy window and the DDW available (or the > legacy and bypass on "nv"), we can then route individual DMAs according > to the corresponding applicable mask. > > I'll try to come up with a patch but I'll need you to test it. Sure. I can help with that. Thanks, Brian > > Cheers, > Ben. > >> Regards, >> Sreekanth >> >> On Wed, May 13, 2015 at 7:42 PM, Brian King <brking@xxxxxxxxxxxxxxxxxx> wrote: >>> On 05/13/2015 08:31 AM, Arnd Bergmann wrote: >>>> On Wednesday 13 May 2015 08:23:41 Brian King wrote: >>>>> On 05/13/2015 03:10 AM, Arnd Bergmann wrote: >>>>>> On Tuesday 12 May 2015 18:24:43 Brian King wrote: >>>>>>> >>>>>>> Commit 5fb1bf8aaa832e1e9ca3198de7bbecb8eff7db9c broke 64 bit DMA for mpt2sas on Power. >>>>>>> That commit changed the sequence for setting up the DMA and coherent DMA masks so >>>>>>> that during initialization the driver requests a 64 bit DMA mask and a 32 bit consistent >>>>>>> DMA mask, then later requests a 64 bit consistent DMA mask. The Power architecture does >>>>>>> not currently support this, which results in always falling back to a 32 bit DMA window, >>>>>>> which has a negative impact on performance. Tweak this algorithm slightly so that >>>>>>> if requesting a 32 bit consistent mask fails after we've successfully set a 64 bit >>>>>>> DMA mask, just try to get a 64 bit consistent mask. This should preserve existing >>>>>>> behavior on platforms that support mixed mask setting and restore previous functionality >>>>>>> to those that do not. >>>>>>> >>>>>>> Signed-off-by: Brian King <brking@xxxxxxxxxxxxxxxxxx> >>>>>> >>>>>> I believe the way the API is designed, it should guarantee that after dma_set_mask() >>>>>> succeeds for a device, dma_set_coherent_mask() with the same mask will also succeed. >>>>>> >>>>>> Could you just call dma_set_mask_and_coherent() here to avoid that complex logic? >>>>> >>>>> I don't think that would work. The mpt2sas driver wants to set the dma mask to 64bits >>>>> but set the coherent mask to 32 bits, then my change is to set the coherent mask to >>>>> 64bits if setting it to 32bit fails. I'm not seeing how using dma_set_mask_and_coherent >>>>> would simplify anything here. >>>> >>>> My question was more about why the driver would ask for a 32-bit coherent >>>> mask to start with. Is this a limitation in the mpt2sas device that can >>>> only have descriptors at low addresses, or is it trying to work around some >>>> bug in a particular host system? >>> >>> I'll need help from Sreekanth in answering that. All that is in the git commit log is: >>> >>> 2. Initially set the consistent DMA mask to 32 bit and then change it to 64 bit mask >>> after allocating RDPQ pools by calling the function _base_change_consistent_dma_mask. >>> This is to ensure that all the upper 32 bits of RDPQ entries's base address to be same. >>> >>> So, I'm not sure my patch won't break something with mpt2sas... This commit also changed >>> how the RDPO pools are allocated, so its possible this won't work. However, I did load >>> it on Power box and I'm not seeing any major problems so far. I am seeing the following message >>> get logged on a regular basis, not sure if its related to this change or not: >>> >>> mpt2sas3: log_info(0x31120303): originator(IOP), code(0x03), sub_code(0x3112) >>> >>> -Brian >>> >>> >>> -- >>> Brian King >>> Power Linux I/O >>> IBM Linux Technology Center >>> >>> > > -- Brian King Power Linux I/O IBM Linux Technology Center -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
