> From: Miquel van Smoorenburg [mailto:mikevs@xxxxxxxxxx] > >+static int aac_scsi_32_64(struct fib * fib, struct scsi_cmnd * cmd) > >+{ > >+ if ((sizeof(dma_addr_t) > 4) && > >+ (num_physpages > (0xFFFFFFFFULL >> PAGE_SHIFT)) && > >+ (fib->dev->adapter_info.options & AAC_OPT_SGMAP_HOST64)) > >+ return FAILED; > >+ return aac_scsi_32(fib, cmd); > >+} > > I saw the same thing in the 64-bit dpt_i2o driver, but I Yes, scratch and sniff, caught me :-} > wonder, shouldn't this be something like: > > #include <linux/bootmem.h> > > if ((sizeof(dma_addr_t) > 4) && > (max_pfn > (0xFFFFFFFFULL >> PAGE_SHIFT)) && > (fib->dev->adapter_info.options & AAC_OPT_SGMAP_HOST64)) > > e.g. use max_pfn instead of num_physpages ? Yes, it would be more correct for the moment, save for the fact that it is marked for deletion: * This doesn't seem to be used by the Linux memory manager any * more, but is used by ll_rw_block. If we can get rid of it, we * also get rid of some of the stuff above as well. . . . max_pfn = . . . and unfortunately architecture issues abound with all memory limit variable :-( in arch/arm/mm/init.c: * Note: max_low_pfn and max_pfn reflect the number of _pages_ in * the system, not the maximum PFN. Reminder: If this automated solution represented in this patch, however flawed with any memory limit variable, is too hacky and not acceptable for inclusion, the workaround for the existing drivers and PERC users in 64 bit populated architectures continues to be to manually set aacraid.dacmode=0 in the failing configurations. What variable tells us the maximum physical memory address, is correct in all architectures and is going to stick around? Sincerely -- Mark Salyzyn - 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
