From 5d2efba64b231a1733c4048d1708d77e07f26426 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Mon, 30 Oct 2006 16:15:59 +1100 Subject: [POWERPC] Use 4kB iommu pages even on 64kB-page systems The 10Gigabit ethernet device drivers appear to be able to chew up all 256MB of TCE mappings on pSeries systems, as evidenced by numerous error messages: iommu_alloc failed, tbl c0000000010d5c48 vaddr c0000000d875eff0 npages 1 Some experimentation indicates that this is essentially because one 1500 byte ethernet MTU gets mapped as a 64K DMA region when the large 64K pages are enabled. Thus, it doesn't take much to exhaust all of the available DMA mappings for a high-speed card. This patch changes the iommu allocator to work with its own unique, distinct page size. Although the patch is long, its actually quite simple: it just #defines a distinct IOMMU_PAGE_SIZE and then uses this in all the places that matter. As a side effect, it also dramatically improves network performance on platforms with H-calls on iommu translation inserts/removes (since we no longer call it 16 times for a 1500 bytes packet when the iommu HW is still 4k). In the future, we might want to make the IOMMU_PAGE_SIZE a variable in the iommu_table instance, thus allowing support for different HW page sizes in the iommu itself. Signed-off-by: Linas Vepstas Signed-off-by: Benjamin Herrenschmidt Acked-by: Olof Johansson Acked-by: Stephen Rothwell Signed-off-by: Paul Mackerras --- arch/powerpc/platforms/iseries/iommu.c | 11 ++--------- 1 file changed, 2 insertions(+), 9 deletions(-) (limited to 'arch/powerpc/platforms/iseries/iommu.c') diff --git a/arch/powerpc/platforms/iseries/iommu.c b/arch/powerpc/platforms/iseries/iommu.c index f4cbbcf8773a..218817d13c5c 100644 --- a/arch/powerpc/platforms/iseries/iommu.c +++ b/arch/powerpc/platforms/iseries/iommu.c @@ -43,9 +43,6 @@ static void tce_build_iSeries(struct iommu_table *tbl, long index, long npages, u64 rc; u64 tce, rpn; - index <<= TCE_PAGE_FACTOR; - npages <<= TCE_PAGE_FACTOR; - while (npages--) { rpn = virt_to_abs(uaddr) >> TCE_SHIFT; tce = (rpn & TCE_RPN_MASK) << TCE_RPN_SHIFT; @@ -75,9 +72,6 @@ static void tce_free_iSeries(struct iommu_table *tbl, long index, long npages) { u64 rc; - npages <<= TCE_PAGE_FACTOR; - index <<= TCE_PAGE_FACTOR; - while (npages--) { rc = HvCallXm_setTce((u64)tbl->it_index, (u64)index, 0); if (rc) @@ -136,10 +130,9 @@ void iommu_table_getparms_iSeries(unsigned long busno, panic("PCI_DMA: parms->size is zero, parms is 0x%p", parms); /* itc_size is in pages worth of table, it_size is in # of entries */ - tbl->it_size = ((parms->itc_size * TCE_PAGE_SIZE) / - TCE_ENTRY_SIZE) >> TCE_PAGE_FACTOR; + tbl->it_size = (parms->itc_size * TCE_PAGE_SIZE) / TCE_ENTRY_SIZE; tbl->it_busno = parms->itc_busno; - tbl->it_offset = parms->itc_offset >> TCE_PAGE_FACTOR; + tbl->it_offset = parms->itc_offset; tbl->it_index = parms->itc_index; tbl->it_blocksize = 1; tbl->it_type = virtbus ? TCE_VB : TCE_PCI; -- cgit v1.2.3