Skip to contentSkip to conversation

powerpc/dma: Always configure ZONE_DMA32 correctly

PATCH In review Benjamin Herrenschmidt 2 participants 4 messages Feb 20, 2015 – Feb 21, 2015 PowerPC

lore
Share & tools
lore
#0 lore raw

PATCH 1/3 powerpc/dma: Always configure ZONE_DMA32 correctly

We do this for consistency and also in order to support the use of a consistent mask smaller than the dma mask in subsequent patches.

---
 arch/powerpc/kernel/dma-swiotlb.c | 3 ---
 arch/powerpc/mm/mem.c             | 6 ++++++
 2 files changed, 6 insertions(+), 3 deletions(-)
arch/powerpc/kernel/dma-swiotlb.c +0 -3
index 7359797..cb92f94 100644
@@ -110,9 +110,6 @@ void __init swiotlb_detect_4g(void)
 {
 	if ((memblock_end_of_DRAM() - 1) > 0xffffffff) {
 		ppc_swiotlb_enable = 1;
-#ifdef CONFIG_ZONE_DMA32
-		limit_zone_pfn(ZONE_DMA32, (1ULL << 32) >> PAGE_SHIFT);
-#endif
 	}
 }
 
arch/powerpc/mm/mem.c +6 -0
index b7285a5..f146ef0 100644
@@ -307,6 +307,12 @@ void __init paging_init(void)
 	printk(KERN_DEBUG "Memory hole size: %ldMB\n",
 	       (long int)((top_of_ram - total_ram) >> 20));
 
+#ifdef CONFIG_ZONE_DMA32
+	/* Default limit for ZONE_DMA32, platform might limit it
+	 * further due to PCI bridge addressing limitations
+	 */
+	limit_zone_pfn(ZONE_DMA32, (1ULL << 32) >> PAGE_SHIFT);
+#endif
 #ifdef CONFIG_HIGHMEM
 	top_zone = ZONE_HIGHMEM;
 	limit_zone_pfn(ZONE_NORMAL, lowmem_end_addr >> PAGE_SHIFT);
#1 lore raw
Quoted 28 lines · Benjamin Herrenschmidt

We do this for consistency and also in order to support the use of a consistent mask smaller than the dma mask in subsequent patches.

---
 arch/powerpc/kernel/dma-swiotlb.c | 3 ---
 arch/powerpc/mm/mem.c             | 6 ++++++
 2 files changed, 6 insertions(+), 3 deletions(-)
arch/powerpc/kernel/dma-swiotlb.c +0 -3
index 7359797..cb92f94 100644
@@ -110,9 +110,6 @@ void __init swiotlb_detect_4g(void)
 {
 	if ((memblock_end_of_DRAM() - 1) > 0xffffffff) {
 		ppc_swiotlb_enable = 1;
-#ifdef CONFIG_ZONE_DMA32
-		limit_zone_pfn(ZONE_DMA32, (1ULL << 32) >> PAGE_SHIFT);
-#endif
 	}
 }
 
arch/powerpc/mm/mem.c +0 -0
index b7285a5..f146ef0 100644

arch/powerpc/mm/mem.c void __init paging_init(void) · line 316

Quoted 10 lines
arch/powerpc/mm/mem.c +6 -0
@@ -307,6 +307,12 @@ void __init paging_init(void)
 	printk(KERN_DEBUG "Memory hole size: %ldMB\n",
 	       (long int)((top_of_ram - total_ram) >> 20));
 
+#ifdef CONFIG_ZONE_DMA32
+	/* Default limit for ZONE_DMA32, platform might limit it
+	 * further due to PCI bridge addressing limitations
+	 */
+	limit_zone_pfn(ZONE_DMA32, (1ULL << 32) >> PAGE_SHIFT);
+#endif
 #ifdef CONFIG_HIGHMEM

Hmm, won't this break the PCI on the fsl SoCs? The ZONE_DMA32 is set to 2GB due to the PCI inbound window limitation on fsl SoCs. This will override that unconditionally. Please see the commit 84f44cc56c09 ("powerpc/fsl-pci:
Limit ZONE_DMA32 to 2GiB on 64-bit platforms") for more detail.

Thanks,
Kevin

#2 lore raw
Quoted 32 lines · Kevin Hao
Quoted 31 lines · Benjamin Herrenschmidt

We do this for consistency and also in order to support the use of a consistent mask smaller than the dma mask in subsequent patches.

---
 arch/powerpc/kernel/dma-swiotlb.c | 3 ---
 arch/powerpc/mm/mem.c             | 6 ++++++
 2 files changed, 6 insertions(+), 3 deletions(-)
arch/powerpc/kernel/dma-swiotlb.c +0 -3
index 7359797..cb92f94 100644
@@ -110,9 +110,6 @@ void __init swiotlb_detect_4g(void)
 {
 	if ((memblock_end_of_DRAM() - 1) > 0xffffffff) {
 		ppc_swiotlb_enable = 1;
-#ifdef CONFIG_ZONE_DMA32
-		limit_zone_pfn(ZONE_DMA32, (1ULL << 32) >> PAGE_SHIFT);
-#endif
 	}
 }
 
arch/powerpc/mm/mem.c +0 -0
index b7285a5..f146ef0 100644
@@ -307,6 +307,12 @@ void __init paging_init(void)
 	printk(KERN_DEBUG "Memory hole size: %ldMB\n",
 	       (long int)((top_of_ram - total_ram) >> 20));
 
Quoted 12 lines

arch/powerpc/mm/mem.c void __init paging_init(void) · line 316

Quoted 7 lines
arch/powerpc/mm/mem.c +6 -0
+#ifdef CONFIG_ZONE_DMA32
+	/* Default limit for ZONE_DMA32, platform might limit it
+	 * further due to PCI bridge addressing limitations
+	 */
+	limit_zone_pfn(ZONE_DMA32, (1ULL << 32) >> PAGE_SHIFT);
+#endif
 #ifdef CONFIG_HIGHMEM

Hmm, won't this break the PCI on the fsl SoCs? The ZONE_DMA32 is set to 2GB due to the PCI inbound window limitation on fsl SoCs. This will override that unconditionally. Please see the commit 84f44cc56c09 ("powerpc/fsl-pci:
Limit ZONE_DMA32 to 2GiB on 64-bit platforms") for more detail.

No it won't, or am I missing something ? Ie, limit_zone_pfn() will only reduce the size of the window, never increase it.

Cheers,
Ben.

Quoted 2 lines

Thanks,
Kevin

#3 lore raw
Quoted 2 lines · Benjamin Herrenschmidt

No it won't, or am I missing something ? Ie, limit_zone_pfn() will only reduce the size of the window, never increase it.

Yes, you are right. Just overlook this. Sorry for the noise.

Thanks,
Kevin