* Re: [Patch v4 08/12] memory-hotplug: remove memmap of sparse-vmemmap
From: Wen Congyang @ 2012-11-30 1:45 UTC (permalink / raw)
To: Jianguo Wu
Cc: linux-s390, linux-ia64, Len Brown, linux-acpi, linux-sh, x86,
linux-kernel, cmetcalf, linux-mm, Yasuaki Ishimatsu, paulus,
Minchan Kim, KOSAKI Motohiro, David Rientjes, sparclinux,
Christoph Lameter, linuxppc-dev, Andrew Morton, Jiang Liu
In-Reply-To: <50B5DC00.20103@huawei.com>
At 11/28/2012 05:40 PM, Jianguo Wu Wrote:
> Hi Congyang,
>
> I think vmemmap's pgtable pages should be freed after all entries are cleared, I have a patch to do this.
> The code logic is the same as [Patch v4 09/12] memory-hotplug: remove page table of x86_64 architecture.
>
> How do you think about this?
>
> Signed-off-by: Jianguo Wu <wujianguo@huawei.com>
> Signed-off-by: Jiang Liu <jiang.liu@huawei.com>
> ---
> include/linux/mm.h | 1 +
> mm/sparse-vmemmap.c | 214 +++++++++++++++++++++++++++++++++++++++++++++++++++
> mm/sparse.c | 5 +-
> 3 files changed, 218 insertions(+), 2 deletions(-)
>
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index 5657670..1f26af5 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -1642,6 +1642,7 @@ int vmemmap_populate(struct page *start_page, unsigned long pages, int node);
> void vmemmap_populate_print_last(void);
> void register_page_bootmem_memmap(unsigned long section_nr, struct page *map,
> unsigned long size);
> +void vmemmap_free(struct page *memmap, unsigned long nr_pages);
>
> enum mf_flags {
> MF_COUNT_INCREASED = 1 << 0,
> diff --git a/mm/sparse-vmemmap.c b/mm/sparse-vmemmap.c
> index 1b7e22a..242cb28 100644
> --- a/mm/sparse-vmemmap.c
> +++ b/mm/sparse-vmemmap.c
> @@ -29,6 +29,10 @@
> #include <asm/pgalloc.h>
> #include <asm/pgtable.h>
>
> +#ifdef CONFIG_MEMORY_HOTREMOVE
> +#include <asm/tlbflush.h>
> +#endif
> +
> /*
> * Allocate a block of memory to be used to back the virtual memory map
> * or to back the page tables that are used to create the mapping.
> @@ -224,3 +228,213 @@ void __init sparse_mem_maps_populate_node(struct page **map_map,
> vmemmap_buf_end = NULL;
> }
> }
> +
> +#ifdef CONFIG_MEMORY_HOTREMOVE
> +static void vmemmap_free_pages(struct page *page, int order)
> +{
> + struct zone *zone;
> + unsigned long magic;
> +
> + magic = (unsigned long) page->lru.next;
> + if (magic == SECTION_INFO || magic == MIX_SECTION_INFO) {
> + put_page_bootmem(page);
> +
> + zone = page_zone(page);
> + zone_span_writelock(zone);
> + zone->present_pages++;
> + zone_span_writeunlock(zone);
> + totalram_pages++;
> + } else {
> + if (is_vmalloc_addr(page_address(page)))
> + vfree(page_address(page));
Hmm, vmemmap doesn't use vmalloc() to allocate memory.
> + else
> + free_pages((unsigned long)page_address(page), order);
> + }
> +}
> +
> +static void free_pte_table(pmd_t *pmd)
> +{
> + pte_t *pte, *pte_start;
> + int i;
> +
> + pte_start = (pte_t *)pmd_page_vaddr(*pmd);
> + for (i = 0; i < PTRS_PER_PTE; i++) {
> + pte = pte_start + i;
> + if (pte_val(*pte))
> + return;
> + }
> +
> + /* free a pte talbe */
> + vmemmap_free_pages(pmd_page(*pmd), 0);
> + spin_lock(&init_mm.page_table_lock);
> + pmd_clear(pmd);
> + spin_unlock(&init_mm.page_table_lock);
> +}
> +
> +static void free_pmd_table(pud_t *pud)
> +{
> + pmd_t *pmd, *pmd_start;
> + int i;
> +
> + pmd_start = (pmd_t *)pud_page_vaddr(*pud);
> + for (i = 0; i < PTRS_PER_PMD; i++) {
> + pmd = pmd_start + i;
> + if (pmd_val(*pmd))
> + return;
> + }
> +
> + /* free a pmd talbe */
> + vmemmap_free_pages(pud_page(*pud), 0);
> + spin_lock(&init_mm.page_table_lock);
> + pud_clear(pud);
> + spin_unlock(&init_mm.page_table_lock);
> +}
> +
> +static void free_pud_table(pgd_t *pgd)
> +{
> + pud_t *pud, *pud_start;
> + int i;
> +
> + pud_start = (pud_t *)pgd_page_vaddr(*pgd);
> + for (i = 0; i < PTRS_PER_PUD; i++) {
> + pud = pud_start + i;
> + if (pud_val(*pud))
> + return;
> + }
> +
> + /* free a pud table */
> + vmemmap_free_pages(pgd_page(*pgd), 0);
> + spin_lock(&init_mm.page_table_lock);
> + pgd_clear(pgd);
> + spin_unlock(&init_mm.page_table_lock);
> +}
> +
> +static int split_large_page(pte_t *kpte, unsigned long address, pte_t *pbase)
> +{
> + struct page *page = pmd_page(*(pmd_t *)kpte);
> + int i = 0;
> + unsigned long magic;
> + unsigned long section_nr;
> +
> + __split_large_page(kpte, address, pbase);
> + __flush_tlb_all();
> +
> + magic = (unsigned long) page->lru.next;
> + if (magic == SECTION_INFO) {
> + section_nr = pfn_to_section_nr(page_to_pfn(page));
> + while (i < PTRS_PER_PMD) {
> + page++;
> + i++;
> + get_page_bootmem(section_nr, page, SECTION_INFO);
> + }
> + }
> +
> + return 0;
> +}
> +
> +static void vmemmap_pte_remove(pmd_t *pmd, unsigned long addr, unsigned long end)
> +{
> + pte_t *pte;
> + unsigned long next;
> +
> + pte = pte_offset_kernel(pmd, addr);
> + for (; addr < end; pte++, addr += PAGE_SIZE) {
> + next = (addr + PAGE_SIZE) & PAGE_MASK;
> + if (next > end)
> + next = end;
> +
> + if (pte_none(*pte))
> + continue;
> + if (IS_ALIGNED(addr, PAGE_SIZE) &&
> + IS_ALIGNED(end, PAGE_SIZE)) {
> + vmemmap_free_pages(pte_page(*pte), 0);
> + spin_lock(&init_mm.page_table_lock);
> + pte_clear(&init_mm, addr, pte);
> + spin_unlock(&init_mm.page_table_lock);
If addr or end is not alianed with PAGE_SIZE, you may leak some
memory.
> + }
> + }
> +
> + free_pte_table(pmd);
> + __flush_tlb_all();
> +}
> +
> +static void vmemmap_pmd_remove(pud_t *pud, unsigned long addr, unsigned long end)
> +{
> + unsigned long next;
> + pmd_t *pmd;
> +
> + pmd = pmd_offset(pud, addr);
> + for (; addr < end; addr = next, pmd++) {
> + next = pmd_addr_end(addr, end);
> + if (pmd_none(*pmd))
> + continue;
> +
> + if (cpu_has_pse) {
> + unsigned long pte_base;
> +
> + if (IS_ALIGNED(addr, PMD_SIZE) &&
> + IS_ALIGNED(next, PMD_SIZE)) {
> + vmemmap_free_pages(pmd_page(*pmd),
> + get_order(PMD_SIZE));
> + spin_lock(&init_mm.page_table_lock);
> + pmd_clear(pmd);
> + spin_unlock(&init_mm.page_table_lock);
> + continue;
> + }
> +
> + /*
> + * We use 2M page, but we need to remove part of them,
> + * so split 2M page to 4K page.
> + */
> + pte_base = get_zeroed_page(GFP_ATOMIC | __GFP_NOTRACK);
get_zeored_page() may fail. You should handle this error.
> + split_large_page((pte_t *)pmd, addr, (pte_t *)pte_base);
> + __flush_tlb_all();
> +
> + spin_lock(&init_mm.page_table_lock);
> + pmd_populate_kernel(&init_mm, pmd, (pte_t *)pte_base);
> + spin_unlock(&init_mm.page_table_lock);
> + }
> +
> + vmemmap_pte_remove(pmd, addr, next);
> + }
> +
> + free_pmd_table(pud);
> + __flush_tlb_all();
> +}
> +
> +static void vmemmap_pud_remove(pgd_t *pgd, unsigned long addr, unsigned long end)
> +{
> + unsigned long next;
> + pud_t *pud;
> +
> + pud = pud_offset(pgd, addr);
> + for (; addr < end; addr = next, pud++) {
> + next = pud_addr_end(addr, end);
> + if (pud_none(*pud))
> + continue;
> +
> + vmemmap_pmd_remove(pud, addr, next);
> + }
> +
> + free_pud_table(pgd);
> + __flush_tlb_all();
> +}
> +
> +void vmemmap_free(struct page *memmap, unsigned long nr_pages)
> +{
> + unsigned long addr = (unsigned long)memmap;
> + unsigned long end = (unsigned long)(memmap + nr_pages);
> + unsigned long next;
> +
> + for (; addr < end; addr = next) {
> + pgd_t *pgd = pgd_offset_k(addr);
> +
> + next = pgd_addr_end(addr, end);
> + if (!pgd_present(*pgd))
> + continue;
> +
> + vmemmap_pud_remove(pgd, addr, next);
> + sync_global_pgds(addr, next);
The parameter for sync_global_pgds() is [start, end], not
[start, end)
> + }
> +}
> +#endif
> diff --git a/mm/sparse.c b/mm/sparse.c
> index fac95f2..3a16d68 100644
> --- a/mm/sparse.c
> +++ b/mm/sparse.c
> @@ -613,12 +613,13 @@ static inline struct page *kmalloc_section_memmap(unsigned long pnum, int nid,
> /* This will make the necessary allocations eventually. */
> return sparse_mem_map_populate(pnum, nid);
> }
> -static void __kfree_section_memmap(struct page *memmap, unsigned long nr_pages)
> +static void __kfree_section_memmap(struct page *page, unsigned long nr_pages)
Why do you change this line?
> {
> - return; /* XXX: Not implemented yet */
> + vmemmap_free(page, nr_pages);
> }
> static void free_map_bootmem(struct page *page, unsigned long nr_pages)
> {
> + vmemmap_free(page, nr_pages);
> }
> #else
> static struct page *__kmalloc_section_memmap(unsigned long nr_pages)
^ permalink raw reply
* Re: [PATCH] dts/t4240: Added SEC-5.0 device tree.
From: Kim Phillips @ 2012-11-29 21:36 UTC (permalink / raw)
To: Vakul Garg; +Cc: linuxppc-dev
In-Reply-To: <1354163290-20431-1-git-send-email-vakul@freescale.com>
On Thu, 29 Nov 2012 09:58:10 +0530
Vakul Garg <vakul@freescale.com> wrote:
> Signed-off-by: Vakul Garg <vakul@freescale.com>
> ---
> This patch has a dependency on the patch titled:
> "Corrected PAMU device address range to encompass 6 PAMU partitions."
>
> arch/powerpc/boot/dts/fsl/qoriq-sec5.0-0.dtsi | 109 +++++++++++++++++++++++++
> arch/powerpc/boot/dts/fsl/t4240si-post.dtsi | 2 +
> 2 files changed, 111 insertions(+), 0 deletions(-)
> create mode 100644 arch/powerpc/boot/dts/fsl/qoriq-sec5.0-0.dtsi
um, t4240 support isn't upstream yet.
Kim
^ permalink raw reply
* [PATCH] powerpc/oprofile: Fix error in oprofile power7_marked_instr_event() function
From: Carl E. Love @ 2012-11-29 16:42 UTC (permalink / raw)
To: bherren, linuxppc-dev
Ben:
Please review the following patch. If it is acceptable, will you please
commit it to the mainline tree. Thanks.
Carl Love
P.S. Looks like I sent it to the wrong mailing list the first time to
get it into the patch queue.
--------------------------------------------------------------------------
powerpc/oprofile: Fix error in oprofile power7_marked_instr_event() function
The calculation for the left shift of the mask OPROFILE_PM_PMCSEL_MSK has an
error. The calculation is should be to shift left by (max_cntrs - cntr) times
the width of the pmsel field width. However, the #define OPROFILE_MAX_PMC_NUM
was used instead of OPROFILE_PMSEL_FIELD_WIDTH. This patch fixes the
calculation.
Signed-off-by: Carl Love <cel@us.ibm.com>
---
arch/powerpc/oprofile/op_model_power4.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/powerpc/oprofile/op_model_power4.c b/arch/powerpc/oprofile/op_model_power4.c
index 315f949..f444b94 100644
--- a/arch/powerpc/oprofile/op_model_power4.c
+++ b/arch/powerpc/oprofile/op_model_power4.c
@@ -52,7 +52,7 @@ static int power7_marked_instr_event(u64 mmcr1)
for (pmc = 0; pmc < 4; pmc++) {
psel = mmcr1 & (OPROFILE_PM_PMCSEL_MSK
<< (OPROFILE_MAX_PMC_NUM - pmc)
- * OPROFILE_MAX_PMC_NUM);
+ * OPROFILE_PMSEL_FIELD_WIDTH);
psel = (psel >> ((OPROFILE_MAX_PMC_NUM - pmc)
* OPROFILE_PMSEL_FIELD_WIDTH)) & ~1ULL;
unit = mmcr1 & (OPROFILE_PM_UNIT_MSK
--
1.7.12.rc1.22.gbfbf4d4
^ permalink raw reply related
* Re: [PATCH 1/4 v2] iommu/fsl: Store iommu domain information pointer in archdata.
From: Kumar Gala @ 2012-11-28 17:35 UTC (permalink / raw)
To: Sethi Varun-B16395
Cc: Wood Scott-B07421, joerg.roedel@amd.com, Tabi Timur-B04825,
linux-kernel@vger.kernel.org, iommu@lists.linux-foundation.org,
linuxppc-dev@lists.ozlabs.org
In-Reply-To: <C5ECD7A89D1DC44195F34B25E172658D29AE62@039-SN2MPN1-013.039d.mgd.msft.net>
On Nov 25, 2012, at 11:33 PM, Sethi Varun-B16395 wrote:
> Hi Kumar,
> Can you please apply this patch.
>=20
> Regards
> Varun
Was waiting on the others to apply this all together. (ie getting an =
Ack from Joerg, and follow comments from Timur to be resolved)
- k
>=20
>> -----Original Message-----
>> From: Sethi Varun-B16395
>> Sent: Tuesday, November 20, 2012 7:25 PM
>> To: joerg.roedel@amd.com; iommu@lists.linux-foundation.org; linuxppc-
>> dev@lists.ozlabs.org; linux-kernel@vger.kernel.org; Wood =
Scott-B07421;
>> Tabi Timur-B04825
>> Cc: Sethi Varun-B16395
>> Subject: [PATCH 1/4 v2] iommu/fsl: Store iommu domain information =
pointer
>> in archdata.
>>=20
>> Add a new field in the device (powerpc) archdata structure for =
storing
>> iommu domain information pointer. This pointer is stored when the =
device
>> is attached to a particular domain.
>>=20
>> Signed-off-by: Varun Sethi <Varun.Sethi@freescale.com>
>> ---
>> arch/powerpc/include/asm/device.h | 4 ++++
>> 1 files changed, 4 insertions(+), 0 deletions(-)
>>=20
>> diff --git a/arch/powerpc/include/asm/device.h
>> b/arch/powerpc/include/asm/device.h
>> index 77e97dd..6dc79fe 100644
>> --- a/arch/powerpc/include/asm/device.h
>> +++ b/arch/powerpc/include/asm/device.h
>> @@ -28,6 +28,10 @@ struct dev_archdata {
>> void *iommu_table_base;
>> } dma_data;
>>=20
>> + /* IOMMU domain information pointer. This would be set
>> + * when this device is attached to an iommu_domain.
>> + */
>> + void *iommu_domain;
>> #ifdef CONFIG_SWIOTLB
>> dma_addr_t max_direct_dma_addr;
>> #endif
>> --
>> 1.7.4.1
>=20
^ permalink raw reply
* Re: [PATCH] powerpc/pci-hotplug: fix the rescanned pci device's dma_set_mask issue
From: Kumar Gala @ 2012-11-28 17:34 UTC (permalink / raw)
To: Chen Yuanquan-B41889; +Cc: linuxppc-dev@lists.ozlabs.org list, Zang Roy-R61911
In-Reply-To: <50B5782A.4050308@freescale.com>
On Nov 27, 2012, at 8:34 PM, Chen Yuanquan-B41889 wrote:
> On 11/25/2012 08:41 PM, Kumar Gala wrote:
>> On Nov 22, 2012, at 10:29 PM, Yuanquan Chen wrote:
>>=20
>>> On powerpc arch, dma_ops of rescanned pci device after system's =
booting up won't be
>>> initialized by system, so it will fail to execute the dma_set_mask =
in the device's
>>> driver. Initialize it to solve this issue.
>>>=20
>>> Signed-off-by: Yuanquan Chen <B41889@freescale.com>
>>> ---
>>> arch/powerpc/include/asm/dma-mapping.h | 7 +++++--
>>> 1 file changed, 5 insertions(+), 2 deletions(-)
>> This is not the right way to get the dma_ops setup. You need to find =
some other point for the hotplug scenario to get the dma_ops setup.
>>=20
>> - k
>=20
> Hi Kumar,
>=20
> I read the code about pci bus scan and rescan. Only the =
pcibios_fixup_bus in pci_scan_child_bus and
> pcibios_enable_device in pci_rescan_bus are arch related code. The =
pcibios_fixup_bus won't be called
> for the rescanned PCI devices due to the bus->is_added has been set =
for the first scanning at boot time.
> So I think it's more reasonable to do the same work as =
pcibios_fixup_bus for rescanned PCI device in
> pcibios_enable_device. The patch code is a copy of =
pcibios_setup_bus_devices called by pcibios_fixup_bus,
> It can solve the dma_set_mask and irq related issues of rescanned PCI =
device on powerpc arch. What's
> your opinion?
>=20
> Thanks,
> yuanquan
>=20
> diff --git a/arch/powerpc/kernel/pci-common.c =
b/arch/powerpc/kernel/pci-common.c
> index 7f94f76..30f7d61 100644
> --- a/arch/powerpc/kernel/pci-common.c
> +++ b/arch/powerpc/kernel/pci-common.c
> @@ -1496,6 +1496,23 @@ int pcibios_enable_device(struct pci_dev *dev, =
int mask)
> if (ppc_md.pcibios_enable_device_hook(dev))
> return -EINVAL;
>=20
> + if (!dev->is_added) {
> + set_dev_node(&dev->dev, pcibus_to_node(dev->bus));
> +
> + /* Hook up default DMA ops */
> + set_dma_ops(&dev->dev, pci_dma_ops);
> + set_dma_offset(&dev->dev, PCI_DRAM_OFFSET);
> +
> + /* Additional platform DMA/iommu setup */
> + if (ppc_md.pci_dma_dev_setup)
> + ppc_md.pci_dma_dev_setup(dev);
> +
> + /* Read default IRQs and fixup if necessary */
> + pci_read_irq_line(dev);
> + if (ppc_md.pci_irq_fixup)
> + ppc_md.pci_irq_fixup(dev);
> + }
> +
> return pci_enable_resources(dev, mask);
> }
>=20
This looks better than the first patch.. I'm surprised that the IBM =
guys don't run into a similar issue with their hotplug support.
Maybe Ben or someone else will comment.
- k
>>> diff --git a/arch/powerpc/include/asm/dma-mapping.h =
b/arch/powerpc/include/asm/dma-mapping.h
>>> index 7816087..22eae53 100644
>>> --- a/arch/powerpc/include/asm/dma-mapping.h
>>> +++ b/arch/powerpc/include/asm/dma-mapping.h
>>> @@ -126,8 +126,11 @@ static inline int dma_supported(struct device =
*dev, u64 mask)
>>> {
>>> struct dma_map_ops *dma_ops =3D get_dma_ops(dev);
>>>=20
>>> - if (unlikely(dma_ops =3D=3D NULL))
>>> - return 0;
>>> + if (unlikely(dma_ops =3D=3D NULL)) {
>>> + set_dma_ops(dev, &dma_direct_ops);
>>> + set_dma_offset(dev, PCI_DRAM_OFFSET);
>>> + dma_ops =3D &dma_direct_ops;
>>> + }
>>> if (dma_ops->dma_supported =3D=3D NULL)
>>> return 1;
>>> return dma_ops->dma_supported(dev, mask);
>>> --=20
>>> 1.7.9.5
>>>=20
>>>=20
>>> _______________________________________________
>>> Linuxppc-dev mailing list
>>> Linuxppc-dev@lists.ozlabs.org
>>> https://lists.ozlabs.org/listinfo/linuxppc-dev
>>=20
>>=20
>>=20
>=20
^ permalink raw reply
* Re: [PATCH] export of_reconfig_notifier_[register,unregister]
From: Benjamin Herrenschmidt @ 2012-11-29 6:07 UTC (permalink / raw)
To: Nathan Fontenot
Cc: Stephen Rothwell, linux-kernel, Rob Herring, linux-next,
Paul Mackerras, linuxppc-dev
In-Reply-To: <50B66922.9090209@linux.vnet.ibm.com>
On Wed, 2012-11-28 at 13:42 -0600, Nathan Fontenot wrote:
> The of reconfiguration notification chains should be exported for use
> by modules.
>
> Signed-off-by:Nathan Fontenot <nfont@linux.vnet.ibm.com>
Grant, I'm applying that into the powerpc tree (with a slightly modified
subject line). You don't need to add it to your tree at this stage I
suspect.
Cheers,
Ben.
> ---
> Index: linux-next/drivers/of/base.c
> ===================================================================
> --- linux-next.orig/drivers/of/base.c 2012-11-28 09:18:02.000000000 -0600
> +++ linux-next/drivers/of/base.c 2012-11-28 11:05:00.000000000 -0600
> @@ -1282,11 +1282,13 @@
> {
> return blocking_notifier_chain_register(&of_reconfig_chain, nb);
> }
> +EXPORT_SYMBOL_GPL(of_reconfig_notifier_register);
>
> int of_reconfig_notifier_unregister(struct notifier_block *nb)
> {
> return blocking_notifier_chain_unregister(&of_reconfig_chain, nb);
> }
> +EXPORT_SYMBOL_GPL(of_reconfig_notifier_unregister);
>
> int of_reconfig_notify(unsigned long action, void *p)
> {
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply
* [PATCH] dts/t4240: Added SEC-5.0 device tree.
From: Vakul Garg @ 2012-11-29 4:28 UTC (permalink / raw)
To: linuxppc-dev; +Cc: Vakul Garg
Signed-off-by: Vakul Garg <vakul@freescale.com>
---
This patch has a dependency on the patch titled:
"Corrected PAMU device address range to encompass 6 PAMU partitions."
arch/powerpc/boot/dts/fsl/qoriq-sec5.0-0.dtsi | 109 +++++++++++++++++++++++++
arch/powerpc/boot/dts/fsl/t4240si-post.dtsi | 2 +
2 files changed, 111 insertions(+), 0 deletions(-)
create mode 100644 arch/powerpc/boot/dts/fsl/qoriq-sec5.0-0.dtsi
diff --git a/arch/powerpc/boot/dts/fsl/qoriq-sec5.0-0.dtsi b/arch/powerpc/boot/dts/fsl/qoriq-sec5.0-0.dtsi
new file mode 100644
index 0000000..ffd458f
--- /dev/null
+++ b/arch/powerpc/boot/dts/fsl/qoriq-sec5.0-0.dtsi
@@ -0,0 +1,109 @@
+/*
+ * QorIQ Sec/Crypto 5.0 device tree stub [ controller @ offset 0x300000 ]
+ *
+ * Copyright 2012 Freescale Semiconductor Inc.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of Freescale Semiconductor nor the
+ * names of its contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
+ * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+ * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+ * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+ * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+crypto: crypto@300000 {
+ compatible = "fsl,sec-v5.0", "fsl,sec-v4.0";
+ #address-cells = <1>;
+ #size-cells = <1>;
+ reg = <0x300000 0x10000>;
+ ranges = <0 0x300000 0x10000>;
+ interrupts = <92 2 0 0>;
+
+ sec_jr0: jr@1000 {
+ compatible = "fsl,sec-v5.0-job-ring",
+ "fsl,sec-v4.0-job-ring";
+ reg = <0x1000 0x1000>;
+ interrupts = <88 2 0 0>;
+ };
+
+ sec_jr1: jr@2000 {
+ compatible = "fsl,sec-v5.0-job-ring",
+ "fsl,sec-v4.0-job-ring";
+ reg = <0x2000 0x1000>;
+ interrupts = <89 2 0 0>;
+ };
+
+ sec_jr2: jr@3000 {
+ compatible = "fsl,sec-v5.0-job-ring",
+ "fsl,sec-v4.0-job-ring";
+ reg = <0x3000 0x1000>;
+ interrupts = <90 2 0 0>;
+ };
+
+ sec_jr3: jr@4000 {
+ compatible = "fsl,sec-v5.0-job-ring",
+ "fsl,sec-v4.0-job-ring";
+ reg = <0x4000 0x1000>;
+ interrupts = <91 2 0 0>;
+ };
+
+ rtic@6000 {
+ compatible = "fsl,sec-v5.0-rtic",
+ "fsl,sec-v4.0-rtic";
+ #address-cells = <1>;
+ #size-cells = <1>;
+ reg = <0x6000 0x100>;
+ ranges = <0x0 0x6100 0xe00>;
+
+ rtic_a: rtic-a@0 {
+ compatible = "fsl,sec-v5.0-rtic-memory",
+ "fsl,sec-v4.0-rtic-memory";
+ reg = <0x00 0x20 0x100 0x80>;
+ };
+
+ rtic_b: rtic-b@20 {
+ compatible = "fsl,sec-v5.0-rtic-memory",
+ "fsl,sec-v4.0-rtic-memory";
+ reg = <0x20 0x20 0x200 0x80>;
+ };
+
+ rtic_c: rtic-c@40 {
+ compatible = "fsl,sec-v5.0-rtic-memory",
+ "fsl,sec-v4.0-rtic-memory";
+ reg = <0x40 0x20 0x300 0x80>;
+ };
+
+ rtic_d: rtic-d@60 {
+ compatible = "fsl,sec-v5.0-rtic-memory",
+ "fsl,sec-v4.0-rtic-memory";
+ reg = <0x60 0x20 0x500 0x80>;
+ };
+ };
+};
+
+sec_mon: sec_mon@314000 {
+ compatible = "fsl,sec-v5.0-mon", "fsl,sec-v4.0-mon";
+ reg = <0x314000 0x1000>;
+ interrupts = <93 2 0 0>;
+};
diff --git a/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi b/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi
index 82c9590..39f57c6 100644
--- a/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi
+++ b/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi
@@ -275,6 +275,8 @@
/include/ "qoriq-i2c-1.dtsi"
/include/ "qoriq-duart-0.dtsi"
/include/ "qoriq-duart-1.dtsi"
+/include/ "qoriq-sec5.0-0.dtsi"
+
/*
* Temporarily define cluster 1/2/3's L2 cache nodes in order to pass
--
1.7.7.6
^ permalink raw reply related
* Re: [PATCH] vfio powerpc: enabled on powernv platform
From: Alex Williamson @ 2012-11-29 4:20 UTC (permalink / raw)
To: Alexey Kardashevskiy
Cc: David Gibson, Paul Mackerras, linuxppc-dev, linux-kernel
In-Reply-To: <1354161225-23118-1-git-send-email-aik@ozlabs.ru>
On Thu, 2012-11-29 at 14:53 +1100, Alexey Kardashevskiy wrote:
> This patch initializes IOMMU groups based on the IOMMU
> configuration discovered during the PCI scan on POWERNV
> (POWER non virtualized) platform. The IOMMU groups are
> to be used later by VFIO driver (PCI pass through).
>
> It also implements an API for mapping/unmapping pages for
> guest PCI drivers and providing DMA window properties.
> This API is going to be used later by QEMU-VFIO to handle
> h_put_tce hypercalls from the KVM guest.
>
> Although this driver has been tested only on the POWERNV
> platform, it should work on any platform which supports
> TCE tables.
>
> To enable VFIO on POWER, enable SPAPR_TCE_IOMMU config
> option and configure VFIO as required.
>
> Cc: David Gibson <david@gibson.dropbear.id.au>
> Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
> ---
> arch/powerpc/include/asm/iommu.h | 9 ++
> arch/powerpc/kernel/iommu.c | 159 ++++++++++++++++++++++++++++++++++
> arch/powerpc/platforms/powernv/pci.c | 135 +++++++++++++++++++++++++++++
> drivers/iommu/Kconfig | 8 ++
> 4 files changed, 311 insertions(+)
>
> diff --git a/arch/powerpc/include/asm/iommu.h b/arch/powerpc/include/asm/iommu.h
> index cbfe678..5c7087a 100644
> --- a/arch/powerpc/include/asm/iommu.h
> +++ b/arch/powerpc/include/asm/iommu.h
> @@ -76,6 +76,9 @@ struct iommu_table {
> struct iommu_pool large_pool;
> struct iommu_pool pools[IOMMU_NR_POOLS];
> unsigned long *it_map; /* A simple allocation bitmap for now */
> +#ifdef CONFIG_IOMMU_API
> + struct iommu_group *it_group;
> +#endif
> };
>
> struct scatterlist;
> @@ -147,5 +150,11 @@ static inline void iommu_restore(void)
> }
> #endif
>
> +extern long iommu_clear_tces(struct iommu_table *tbl, unsigned long entry,
> + unsigned long pages);
> +extern long iommu_put_tces(struct iommu_table *tbl, unsigned long entry,
> + uint64_t tce, enum dma_data_direction direction,
> + unsigned long pages);
> +
> #endif /* __KERNEL__ */
> #endif /* _ASM_IOMMU_H */
> diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c
> index ff5a6ce..1225fbb 100644
> --- a/arch/powerpc/kernel/iommu.c
> +++ b/arch/powerpc/kernel/iommu.c
> @@ -44,6 +44,7 @@
> #include <asm/kdump.h>
> #include <asm/fadump.h>
> #include <asm/vio.h>
> +#include <asm/tce.h>
>
> #define DBG(...)
>
> @@ -856,3 +857,161 @@ void iommu_free_coherent(struct iommu_table *tbl, size_t size,
> free_pages((unsigned long)vaddr, get_order(size));
> }
> }
> +
> +#ifdef CONFIG_IOMMU_API
> +/*
> + * SPAPR TCE API
> + */
> +static void tce_flush(struct iommu_table *tbl)
> +{
> + /* Flush/invalidate TLB caches if necessary */
> + if (ppc_md.tce_flush)
> + ppc_md.tce_flush(tbl);
> +
> + /* Make sure updates are seen by hardware */
> + mb();
> +}
> +
> +/*
> + * iommu_clear_tces clears tces and returned the number of pages
> + * which it called put_page() on.
> + */
> +static long clear_tces_nolock(struct iommu_table *tbl, unsigned long entry,
> + unsigned long pages)
> +{
> + int i, retpages = 0;
> + unsigned long oldtce;
> + struct page *page;
> +
> + for (i = 0; i < pages; ++i) {
> + oldtce = ppc_md.tce_get(tbl, entry + i);
> + ppc_md.tce_free(tbl, entry + i, 1);
> +
> + if (!(oldtce & (TCE_PCI_WRITE | TCE_PCI_READ)))
> + continue;
> +
> + page = pfn_to_page(oldtce >> PAGE_SHIFT);
> +
> + WARN_ON(!page);
> + if (!page)
> + continue;
> +
> + if (oldtce & TCE_PCI_WRITE)
> + SetPageDirty(page);
> +
> + if (!(oldtce & ~PAGE_MASK))
> + ++retpages;
I'm confused, it looks like you're trying to only increment the counter
for tce pages aligned at the start of a page, but don't we need to mask
out the read/write and valid bits? Trickiness like this demands a
comment.
> +
> + put_page(page);
> + }
> +
> + return retpages;
> +}
> +
> +/*
> + * iommu_clear_tces clears tces and returned the number of released pages
> + */
> +long iommu_clear_tces(struct iommu_table *tbl, unsigned long entry,
> + unsigned long pages)
> +{
> + int ret;
> + struct iommu_pool *pool = get_pool(tbl, entry);
> +
> + spin_lock(&(pool->lock));
> + ret = clear_tces_nolock(tbl, entry, pages);
> + tce_flush(tbl);
> + spin_unlock(&(pool->lock));
> +
> + return ret;
> +}
> +EXPORT_SYMBOL_GPL(iommu_clear_tces);
> +
> +static int put_tce(struct iommu_table *tbl, unsigned long entry,
> + uint64_t tce, enum dma_data_direction direction)
> +{
> + int ret;
> + struct page *page = NULL;
> + unsigned long kva, offset;
> +
> + /* Map new TCE */
> + offset = (tce & IOMMU_PAGE_MASK) - (tce & PAGE_MASK);
> +
> + ret = get_user_pages_fast(tce & PAGE_MASK, 1,
> + direction != DMA_TO_DEVICE, &page);
> + if (ret < 1) {
> + printk(KERN_ERR "tce_vfio: get_user_pages_fast failed tce=%llx ioba=%lx ret=%d\n",
> + tce, entry << IOMMU_PAGE_SHIFT, ret);
> + if (!ret)
> + ret = -EFAULT;
> + return ret;
> + }
> +
> + kva = (unsigned long) page_address(page);
> + kva += offset;
> +
> + /* tce_build receives a virtual address */
> + entry += tbl->it_offset; /* Offset into real TCE table */
> + ret = ppc_md.tce_build(tbl, entry, 1, kva, direction, NULL);
> +
> + /* tce_build() only returns non-zero for transient errors */
> + if (unlikely(ret)) {
> + printk(KERN_ERR "tce_vfio: tce_put failed on tce=%llx ioba=%lx kva=%lx ret=%d\n",
> + tce, entry << IOMMU_PAGE_SHIFT, kva, ret);
> + put_page(page);
> + return -EIO;
> + }
> +
> + return 0;
> +}
> +
> +/*
> + * iommu_put_tces builds tces and returned the number of actually
> + * locked system pages
> + */
> +long iommu_put_tces(struct iommu_table *tbl, unsigned long entry,
> + uint64_t tce, enum dma_data_direction direction,
> + unsigned long pages)
> +{
> + int i, ret = 0, retpages = 0;
> + struct iommu_pool *pool = get_pool(tbl, entry);
> +
> + BUILD_BUG_ON(PAGE_SIZE < IOMMU_PAGE_SIZE);
> + BUG_ON(direction == DMA_NONE);
> +
> + spin_lock(&(pool->lock));
> +
> + /* Check if any is in use */
> + for (i = 0; i < pages; ++i) {
> + unsigned long oldtce = ppc_md.tce_get(tbl, entry + i);
> + if (oldtce & (TCE_PCI_WRITE | TCE_PCI_READ)) {
> + spin_unlock(&(pool->lock));
> + return -EBUSY;
> + }
> + }
> +
> + /* Put tces to the table */
> + for (i = 0; (i < pages) && !ret; ++i, tce += IOMMU_PAGE_SIZE) {
> + ret = put_tce(tbl, entry + i, tce, direction);
> + /*
> + * As IOMMU page size is always 4K, the system page size
> + * can be 64K and there is no special tracking for IOMMU pages,
> + * we only do rlimit check/update for the very first
> + * 4K IOMMUpage within 64K system page.
> + */
> + if (!(tce & ~PAGE_MASK))
> + ++retpages;
Ah, here's the comment I was looking for, though I'm still not sure
about the read/write bits.
Isn't there an exploit here that a user can lock pages beyond their
limits if they just skip mapping the first 4k of each page? Thanks,
Alex
> + }
> +
> + /* If failed, release locked pages, otherwise return the number of pages */
> + if (ret)
> + clear_tces_nolock(tbl, entry, i);
> + else
> + ret = retpages;
> +
> + tce_flush(tbl);
> + spin_unlock(&(pool->lock));
> +
> + return ret;
> +}
> +EXPORT_SYMBOL_GPL(iommu_put_tces);
> +#endif /* CONFIG_IOMMU_API */
> diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c
> index 05205cf..21250ef 100644
> --- a/arch/powerpc/platforms/powernv/pci.c
> +++ b/arch/powerpc/platforms/powernv/pci.c
> @@ -20,6 +20,7 @@
> #include <linux/irq.h>
> #include <linux/io.h>
> #include <linux/msi.h>
> +#include <linux/iommu.h>
>
> #include <asm/sections.h>
> #include <asm/io.h>
> @@ -613,3 +614,137 @@ void __init pnv_pci_init(void)
> ppc_md.teardown_msi_irqs = pnv_teardown_msi_irqs;
> #endif
> }
> +
> +#ifdef CONFIG_IOMMU_API
> +/*
> + * IOMMU groups support required by VFIO
> + */
> +static int add_device(struct device *dev)
> +{
> + struct iommu_table *tbl;
> + int ret = 0;
> +
> + if (WARN_ON(dev->iommu_group)) {
> + printk(KERN_WARNING "tce_vfio: device %s is already in iommu group %d, skipping\n",
> + dev_name(dev),
> + iommu_group_id(dev->iommu_group));
> + return -EBUSY;
> + }
> +
> + tbl = get_iommu_table_base(dev);
> + if (!tbl) {
> + pr_debug("tce_vfio: skipping device %s with no tbl\n",
> + dev_name(dev));
> + return 0;
> + }
> +
> + pr_debug("tce_vfio: adding %s to iommu group %d\n",
> + dev_name(dev), iommu_group_id(tbl->it_group));
> +
> + ret = iommu_group_add_device(tbl->it_group, dev);
> + if (ret < 0)
> + printk(KERN_ERR "tce_vfio: %s has not been added, ret=%d\n",
> + dev_name(dev), ret);
> +
> + return ret;
> +}
> +
> +static void del_device(struct device *dev)
> +{
> + iommu_group_remove_device(dev);
> +}
> +
> +static int iommu_bus_notifier(struct notifier_block *nb,
> + unsigned long action, void *data)
> +{
> + struct device *dev = data;
> +
> + switch (action) {
> + case BUS_NOTIFY_ADD_DEVICE:
> + return add_device(dev);
> + case BUS_NOTIFY_DEL_DEVICE:
> + del_device(dev);
> + return 0;
> + default:
> + return 0;
> + }
> +}
> +
> +static struct notifier_block tce_iommu_bus_nb = {
> + .notifier_call = iommu_bus_notifier,
> +};
> +
> +static void group_release(void *iommu_data)
> +{
> + struct iommu_table *tbl = iommu_data;
> + tbl->it_group = NULL;
> +}
> +
> +static int __init tce_iommu_init(void)
> +{
> + struct pci_dev *pdev = NULL;
> + struct iommu_table *tbl;
> + struct iommu_group *grp;
> +
> + /* Allocate and initialize IOMMU groups */
> + for_each_pci_dev(pdev) {
> + tbl = get_iommu_table_base(&pdev->dev);
> + if (!tbl)
> + continue;
> +
> + /* Skip already initialized */
> + if (tbl->it_group)
> + continue;
> +
> + grp = iommu_group_alloc();
> + if (IS_ERR(grp)) {
> + printk(KERN_INFO "tce_vfio: cannot create "
> + "new IOMMU group, ret=%ld\n",
> + PTR_ERR(grp));
> + return PTR_ERR(grp);
> + }
> + tbl->it_group = grp;
> + iommu_group_set_iommudata(grp, tbl, group_release);
> + }
> +
> + bus_register_notifier(&pci_bus_type, &tce_iommu_bus_nb);
> +
> + /* Add PCI devices to VFIO groups */
> + for_each_pci_dev(pdev)
> + add_device(&pdev->dev);
> +
> + return 0;
> +}
> +
> +static void __exit tce_iommu_cleanup(void)
> +{
> + struct pci_dev *pdev = NULL;
> + struct iommu_table *tbl;
> + struct iommu_group *grp = NULL;
> +
> + bus_unregister_notifier(&pci_bus_type, &tce_iommu_bus_nb);
> +
> + /* Delete PCI devices from VFIO groups */
> + for_each_pci_dev(pdev)
> + del_device(&pdev->dev);
> +
> + /* Release VFIO groups */
> + for_each_pci_dev(pdev) {
> + tbl = get_iommu_table_base(&pdev->dev);
> + if (!tbl)
> + continue;
> + grp = tbl->it_group;
> +
> + /* Skip (already) uninitialized */
> + if (!grp)
> + continue;
> +
> + /* Do actual release, group_release() is expected to work */
> + iommu_group_put(grp);
> + BUG_ON(tbl->it_group);
> + }
> +}
> +
> +module_init(tce_iommu_init);
> +module_exit(tce_iommu_cleanup);
> +#endif /* CONFIG_IOMMU_API */
> diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
> index 9f69b56..29d11dc 100644
> --- a/drivers/iommu/Kconfig
> +++ b/drivers/iommu/Kconfig
> @@ -187,4 +187,12 @@ config EXYNOS_IOMMU_DEBUG
>
> Say N unless you need kernel log message for IOMMU debugging
>
> +config SPAPR_TCE_IOMMU
> + bool "sPAPR TCE IOMMU Support"
> + depends on PPC_POWERNV
> + select IOMMU_API
> + help
> + Enables bits of IOMMU API required by VFIO. The iommu_ops is
> + still not implemented.
> +
> endif # IOMMU_SUPPORT
^ permalink raw reply
* [PATCH] dts/t4240: Corrected PAMU device address range to encompass 6 PAMU partitions.
From: Vakul Garg @ 2012-11-29 4:23 UTC (permalink / raw)
To: linuxppc-dev; +Cc: Vakul Garg
Signed-off-by: Vakul Garg <vakul@freescale.com>
---
arch/powerpc/boot/dts/fsl/t4240si-post.dtsi | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi b/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi
index ea6e2f4..e8809cd 100644
--- a/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi
+++ b/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi
@@ -231,7 +231,7 @@
iommu@20000 {
compatible = "fsl,pamu-v1.0", "fsl,pamu";
- reg = <0x20000 0x4000>;
+ reg = <0x20000 0x6000>;
interrupts = <
24 2 0 0
16 2 1 30>;
--
1.7.7
^ permalink raw reply related
* [PATCH] dts/t4240: Corrected PAMU device address range to encompass 6 PAMU partitions.
From: Vakul Garg @ 2012-11-29 4:11 UTC (permalink / raw)
To: linuxppc-dev; +Cc: Vakul Garg
Signed-off-by: Vakul Garg <vakul@freescale.com>
---
arch/powerpc/boot/dts/fsl/t4240si-post.dtsi | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi b/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi
index ea6e2f4..e8809cd 100644
--- a/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi
+++ b/arch/powerpc/boot/dts/fsl/t4240si-post.dtsi
@@ -231,7 +231,7 @@
iommu@20000 {
compatible = "fsl,pamu-v1.0", "fsl,pamu";
- reg = <0x20000 0x4000>;
+ reg = <0x20000 0x6000>;
interrupts = <
24 2 0 0
16 2 1 30>;
--
1.7.7
^ permalink raw reply related
* [PATCH] vfio powerpc: enabled on powernv platform
From: Alexey Kardashevskiy @ 2012-11-29 3:53 UTC (permalink / raw)
To: Alex Williamson
Cc: Alexey Kardashevskiy, linux-kernel, Paul Mackerras, linuxppc-dev,
David Gibson
In-Reply-To: <1354138245.1809.226.camel@bling.home>
This patch initializes IOMMU groups based on the IOMMU
configuration discovered during the PCI scan on POWERNV
(POWER non virtualized) platform. The IOMMU groups are
to be used later by VFIO driver (PCI pass through).
It also implements an API for mapping/unmapping pages for
guest PCI drivers and providing DMA window properties.
This API is going to be used later by QEMU-VFIO to handle
h_put_tce hypercalls from the KVM guest.
Although this driver has been tested only on the POWERNV
platform, it should work on any platform which supports
TCE tables.
To enable VFIO on POWER, enable SPAPR_TCE_IOMMU config
option and configure VFIO as required.
Cc: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
---
arch/powerpc/include/asm/iommu.h | 9 ++
arch/powerpc/kernel/iommu.c | 159 ++++++++++++++++++++++++++++++++++
arch/powerpc/platforms/powernv/pci.c | 135 +++++++++++++++++++++++++++++
drivers/iommu/Kconfig | 8 ++
4 files changed, 311 insertions(+)
diff --git a/arch/powerpc/include/asm/iommu.h b/arch/powerpc/include/asm/iommu.h
index cbfe678..5c7087a 100644
--- a/arch/powerpc/include/asm/iommu.h
+++ b/arch/powerpc/include/asm/iommu.h
@@ -76,6 +76,9 @@ struct iommu_table {
struct iommu_pool large_pool;
struct iommu_pool pools[IOMMU_NR_POOLS];
unsigned long *it_map; /* A simple allocation bitmap for now */
+#ifdef CONFIG_IOMMU_API
+ struct iommu_group *it_group;
+#endif
};
struct scatterlist;
@@ -147,5 +150,11 @@ static inline void iommu_restore(void)
}
#endif
+extern long iommu_clear_tces(struct iommu_table *tbl, unsigned long entry,
+ unsigned long pages);
+extern long iommu_put_tces(struct iommu_table *tbl, unsigned long entry,
+ uint64_t tce, enum dma_data_direction direction,
+ unsigned long pages);
+
#endif /* __KERNEL__ */
#endif /* _ASM_IOMMU_H */
diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c
index ff5a6ce..1225fbb 100644
--- a/arch/powerpc/kernel/iommu.c
+++ b/arch/powerpc/kernel/iommu.c
@@ -44,6 +44,7 @@
#include <asm/kdump.h>
#include <asm/fadump.h>
#include <asm/vio.h>
+#include <asm/tce.h>
#define DBG(...)
@@ -856,3 +857,161 @@ void iommu_free_coherent(struct iommu_table *tbl, size_t size,
free_pages((unsigned long)vaddr, get_order(size));
}
}
+
+#ifdef CONFIG_IOMMU_API
+/*
+ * SPAPR TCE API
+ */
+static void tce_flush(struct iommu_table *tbl)
+{
+ /* Flush/invalidate TLB caches if necessary */
+ if (ppc_md.tce_flush)
+ ppc_md.tce_flush(tbl);
+
+ /* Make sure updates are seen by hardware */
+ mb();
+}
+
+/*
+ * iommu_clear_tces clears tces and returned the number of pages
+ * which it called put_page() on.
+ */
+static long clear_tces_nolock(struct iommu_table *tbl, unsigned long entry,
+ unsigned long pages)
+{
+ int i, retpages = 0;
+ unsigned long oldtce;
+ struct page *page;
+
+ for (i = 0; i < pages; ++i) {
+ oldtce = ppc_md.tce_get(tbl, entry + i);
+ ppc_md.tce_free(tbl, entry + i, 1);
+
+ if (!(oldtce & (TCE_PCI_WRITE | TCE_PCI_READ)))
+ continue;
+
+ page = pfn_to_page(oldtce >> PAGE_SHIFT);
+
+ WARN_ON(!page);
+ if (!page)
+ continue;
+
+ if (oldtce & TCE_PCI_WRITE)
+ SetPageDirty(page);
+
+ if (!(oldtce & ~PAGE_MASK))
+ ++retpages;
+
+ put_page(page);
+ }
+
+ return retpages;
+}
+
+/*
+ * iommu_clear_tces clears tces and returned the number of released pages
+ */
+long iommu_clear_tces(struct iommu_table *tbl, unsigned long entry,
+ unsigned long pages)
+{
+ int ret;
+ struct iommu_pool *pool = get_pool(tbl, entry);
+
+ spin_lock(&(pool->lock));
+ ret = clear_tces_nolock(tbl, entry, pages);
+ tce_flush(tbl);
+ spin_unlock(&(pool->lock));
+
+ return ret;
+}
+EXPORT_SYMBOL_GPL(iommu_clear_tces);
+
+static int put_tce(struct iommu_table *tbl, unsigned long entry,
+ uint64_t tce, enum dma_data_direction direction)
+{
+ int ret;
+ struct page *page = NULL;
+ unsigned long kva, offset;
+
+ /* Map new TCE */
+ offset = (tce & IOMMU_PAGE_MASK) - (tce & PAGE_MASK);
+
+ ret = get_user_pages_fast(tce & PAGE_MASK, 1,
+ direction != DMA_TO_DEVICE, &page);
+ if (ret < 1) {
+ printk(KERN_ERR "tce_vfio: get_user_pages_fast failed tce=%llx ioba=%lx ret=%d\n",
+ tce, entry << IOMMU_PAGE_SHIFT, ret);
+ if (!ret)
+ ret = -EFAULT;
+ return ret;
+ }
+
+ kva = (unsigned long) page_address(page);
+ kva += offset;
+
+ /* tce_build receives a virtual address */
+ entry += tbl->it_offset; /* Offset into real TCE table */
+ ret = ppc_md.tce_build(tbl, entry, 1, kva, direction, NULL);
+
+ /* tce_build() only returns non-zero for transient errors */
+ if (unlikely(ret)) {
+ printk(KERN_ERR "tce_vfio: tce_put failed on tce=%llx ioba=%lx kva=%lx ret=%d\n",
+ tce, entry << IOMMU_PAGE_SHIFT, kva, ret);
+ put_page(page);
+ return -EIO;
+ }
+
+ return 0;
+}
+
+/*
+ * iommu_put_tces builds tces and returned the number of actually
+ * locked system pages
+ */
+long iommu_put_tces(struct iommu_table *tbl, unsigned long entry,
+ uint64_t tce, enum dma_data_direction direction,
+ unsigned long pages)
+{
+ int i, ret = 0, retpages = 0;
+ struct iommu_pool *pool = get_pool(tbl, entry);
+
+ BUILD_BUG_ON(PAGE_SIZE < IOMMU_PAGE_SIZE);
+ BUG_ON(direction == DMA_NONE);
+
+ spin_lock(&(pool->lock));
+
+ /* Check if any is in use */
+ for (i = 0; i < pages; ++i) {
+ unsigned long oldtce = ppc_md.tce_get(tbl, entry + i);
+ if (oldtce & (TCE_PCI_WRITE | TCE_PCI_READ)) {
+ spin_unlock(&(pool->lock));
+ return -EBUSY;
+ }
+ }
+
+ /* Put tces to the table */
+ for (i = 0; (i < pages) && !ret; ++i, tce += IOMMU_PAGE_SIZE) {
+ ret = put_tce(tbl, entry + i, tce, direction);
+ /*
+ * As IOMMU page size is always 4K, the system page size
+ * can be 64K and there is no special tracking for IOMMU pages,
+ * we only do rlimit check/update for the very first
+ * 4K IOMMUpage within 64K system page.
+ */
+ if (!(tce & ~PAGE_MASK))
+ ++retpages;
+ }
+
+ /* If failed, release locked pages, otherwise return the number of pages */
+ if (ret)
+ clear_tces_nolock(tbl, entry, i);
+ else
+ ret = retpages;
+
+ tce_flush(tbl);
+ spin_unlock(&(pool->lock));
+
+ return ret;
+}
+EXPORT_SYMBOL_GPL(iommu_put_tces);
+#endif /* CONFIG_IOMMU_API */
diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c
index 05205cf..21250ef 100644
--- a/arch/powerpc/platforms/powernv/pci.c
+++ b/arch/powerpc/platforms/powernv/pci.c
@@ -20,6 +20,7 @@
#include <linux/irq.h>
#include <linux/io.h>
#include <linux/msi.h>
+#include <linux/iommu.h>
#include <asm/sections.h>
#include <asm/io.h>
@@ -613,3 +614,137 @@ void __init pnv_pci_init(void)
ppc_md.teardown_msi_irqs = pnv_teardown_msi_irqs;
#endif
}
+
+#ifdef CONFIG_IOMMU_API
+/*
+ * IOMMU groups support required by VFIO
+ */
+static int add_device(struct device *dev)
+{
+ struct iommu_table *tbl;
+ int ret = 0;
+
+ if (WARN_ON(dev->iommu_group)) {
+ printk(KERN_WARNING "tce_vfio: device %s is already in iommu group %d, skipping\n",
+ dev_name(dev),
+ iommu_group_id(dev->iommu_group));
+ return -EBUSY;
+ }
+
+ tbl = get_iommu_table_base(dev);
+ if (!tbl) {
+ pr_debug("tce_vfio: skipping device %s with no tbl\n",
+ dev_name(dev));
+ return 0;
+ }
+
+ pr_debug("tce_vfio: adding %s to iommu group %d\n",
+ dev_name(dev), iommu_group_id(tbl->it_group));
+
+ ret = iommu_group_add_device(tbl->it_group, dev);
+ if (ret < 0)
+ printk(KERN_ERR "tce_vfio: %s has not been added, ret=%d\n",
+ dev_name(dev), ret);
+
+ return ret;
+}
+
+static void del_device(struct device *dev)
+{
+ iommu_group_remove_device(dev);
+}
+
+static int iommu_bus_notifier(struct notifier_block *nb,
+ unsigned long action, void *data)
+{
+ struct device *dev = data;
+
+ switch (action) {
+ case BUS_NOTIFY_ADD_DEVICE:
+ return add_device(dev);
+ case BUS_NOTIFY_DEL_DEVICE:
+ del_device(dev);
+ return 0;
+ default:
+ return 0;
+ }
+}
+
+static struct notifier_block tce_iommu_bus_nb = {
+ .notifier_call = iommu_bus_notifier,
+};
+
+static void group_release(void *iommu_data)
+{
+ struct iommu_table *tbl = iommu_data;
+ tbl->it_group = NULL;
+}
+
+static int __init tce_iommu_init(void)
+{
+ struct pci_dev *pdev = NULL;
+ struct iommu_table *tbl;
+ struct iommu_group *grp;
+
+ /* Allocate and initialize IOMMU groups */
+ for_each_pci_dev(pdev) {
+ tbl = get_iommu_table_base(&pdev->dev);
+ if (!tbl)
+ continue;
+
+ /* Skip already initialized */
+ if (tbl->it_group)
+ continue;
+
+ grp = iommu_group_alloc();
+ if (IS_ERR(grp)) {
+ printk(KERN_INFO "tce_vfio: cannot create "
+ "new IOMMU group, ret=%ld\n",
+ PTR_ERR(grp));
+ return PTR_ERR(grp);
+ }
+ tbl->it_group = grp;
+ iommu_group_set_iommudata(grp, tbl, group_release);
+ }
+
+ bus_register_notifier(&pci_bus_type, &tce_iommu_bus_nb);
+
+ /* Add PCI devices to VFIO groups */
+ for_each_pci_dev(pdev)
+ add_device(&pdev->dev);
+
+ return 0;
+}
+
+static void __exit tce_iommu_cleanup(void)
+{
+ struct pci_dev *pdev = NULL;
+ struct iommu_table *tbl;
+ struct iommu_group *grp = NULL;
+
+ bus_unregister_notifier(&pci_bus_type, &tce_iommu_bus_nb);
+
+ /* Delete PCI devices from VFIO groups */
+ for_each_pci_dev(pdev)
+ del_device(&pdev->dev);
+
+ /* Release VFIO groups */
+ for_each_pci_dev(pdev) {
+ tbl = get_iommu_table_base(&pdev->dev);
+ if (!tbl)
+ continue;
+ grp = tbl->it_group;
+
+ /* Skip (already) uninitialized */
+ if (!grp)
+ continue;
+
+ /* Do actual release, group_release() is expected to work */
+ iommu_group_put(grp);
+ BUG_ON(tbl->it_group);
+ }
+}
+
+module_init(tce_iommu_init);
+module_exit(tce_iommu_cleanup);
+#endif /* CONFIG_IOMMU_API */
diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
index 9f69b56..29d11dc 100644
--- a/drivers/iommu/Kconfig
+++ b/drivers/iommu/Kconfig
@@ -187,4 +187,12 @@ config EXYNOS_IOMMU_DEBUG
Say N unless you need kernel log message for IOMMU debugging
+config SPAPR_TCE_IOMMU
+ bool "sPAPR TCE IOMMU Support"
+ depends on PPC_POWERNV
+ select IOMMU_API
+ help
+ Enables bits of IOMMU API required by VFIO. The iommu_ops is
+ still not implemented.
+
endif # IOMMU_SUPPORT
--
1.7.10.4
^ permalink raw reply related
* Re: [PATCH] vfio powerpc: implemented IOMMU driver for VFIO
From: Alexey Kardashevskiy @ 2012-11-29 3:51 UTC (permalink / raw)
To: Alex Williamson
Cc: kvm, linux-kernel, Paul Mackerras, linuxppc-dev, David Gibson
In-Reply-To: <1354136507.1809.215.camel@bling.home>
On 29/11/12 08:01, Alex Williamson wrote:
> On Wed, 2012-11-28 at 18:21 +1100, Alexey Kardashevskiy wrote:
>> VFIO implements platform independent stuff such as
>> a PCI driver, BAR access (via read/write on a file descriptor
>> or direct mapping when possible) and IRQ signaling.
>>
>> The platform dependent part includes IOMMU initialization
>> and handling. This patch implements an IOMMU driver for VFIO
>> which does mapping/unmapping pages for the guest IO and
>> provides information about DMA window (required by a POWERPC
>> guest).
>>
>> The counterpart in QEMU is required to support this functionality.
>>
>> Cc: David Gibson <david@gibson.dropbear.id.au>
>> Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
>> ---
>> drivers/vfio/Kconfig | 6 +
>> drivers/vfio/Makefile | 1 +
>> drivers/vfio/vfio_iommu_spapr_tce.c | 332 +++++++++++++++++++++++++++++++++++
>> include/linux/vfio.h | 33 ++++
>> 4 files changed, 372 insertions(+)
>> create mode 100644 drivers/vfio/vfio_iommu_spapr_tce.c
>>
>> diff --git a/drivers/vfio/Kconfig b/drivers/vfio/Kconfig
>> index 7cd5dec..b464687 100644
>> --- a/drivers/vfio/Kconfig
>> +++ b/drivers/vfio/Kconfig
>> @@ -3,10 +3,16 @@ config VFIO_IOMMU_TYPE1
>> depends on VFIO
>> default n
>>
>> +config VFIO_IOMMU_SPAPR_TCE
>> + tristate
>> + depends on VFIO && SPAPR_TCE_IOMMU
>> + default n
>> +
>> menuconfig VFIO
>> tristate "VFIO Non-Privileged userspace driver framework"
>> depends on IOMMU_API
>> select VFIO_IOMMU_TYPE1 if X86
>> + select VFIO_IOMMU_SPAPR_TCE if PPC_POWERNV
>> help
>> VFIO provides a framework for secure userspace device drivers.
>> See Documentation/vfio.txt for more details.
>> diff --git a/drivers/vfio/Makefile b/drivers/vfio/Makefile
>> index 2398d4a..72bfabc 100644
>> --- a/drivers/vfio/Makefile
>> +++ b/drivers/vfio/Makefile
>> @@ -1,3 +1,4 @@
>> obj-$(CONFIG_VFIO) += vfio.o
>> obj-$(CONFIG_VFIO_IOMMU_TYPE1) += vfio_iommu_type1.o
>> +obj-$(CONFIG_VFIO_IOMMU_SPAPR_TCE) += vfio_iommu_spapr_tce.o
>> obj-$(CONFIG_VFIO_PCI) += pci/
>> diff --git a/drivers/vfio/vfio_iommu_spapr_tce.c b/drivers/vfio/vfio_iommu_spapr_tce.c
>> new file mode 100644
>> index 0000000..b98770e
>> --- /dev/null
>> +++ b/drivers/vfio/vfio_iommu_spapr_tce.c
>> @@ -0,0 +1,332 @@
>> +/*
>> + * VFIO: IOMMU DMA mapping support for TCE on POWER
>> + *
>> + * Copyright (C) 2012 IBM Corp. All rights reserved.
>> + * Author: Alexey Kardashevskiy <aik@ozlabs.ru>
>> + *
>> + * This program is free software; you can redistribute it and/or modify
>> + * it under the terms of the GNU General Public License version 2 as
>> + * published by the Free Software Foundation.
>> + *
>> + * Derived from original vfio_iommu_type1.c:
>> + * Copyright (C) 2012 Red Hat, Inc. All rights reserved.
>> + * Author: Alex Williamson <alex.williamson@redhat.com>
>> + */
>> +
>> +#include <linux/module.h>
>> +#include <linux/pci.h>
>> +#include <linux/slab.h>
>> +#include <linux/uaccess.h>
>> +#include <linux/err.h>
>> +#include <linux/vfio.h>
>> +#include <asm/iommu.h>
>> +
>> +#define DRIVER_VERSION "0.1"
>> +#define DRIVER_AUTHOR "aik@ozlabs.ru"
>> +#define DRIVER_DESC "VFIO IOMMU SPAPR TCE"
>> +
>> +static void tce_iommu_detach_group(void *iommu_data,
>> + struct iommu_group *iommu_group);
>> +
>> +/*
>> + * VFIO IOMMU fd for SPAPR_TCE IOMMU implementation
>> + */
>> +
>> +/*
>> + * This code handles mapping and unmapping of user data buffers
>> + * into DMA'ble space using the IOMMU
>> + */
>> +
>> +#define NPAGE_TO_SIZE(npage) ((size_t)(npage) << PAGE_SHIFT)
>> +
>> +struct vwork {
>> + struct mm_struct *mm;
>> + long npage;
>> + struct work_struct work;
>> +};
>> +
>> +/* delayed decrement/increment for locked_vm */
>> +static void lock_acct_bg(struct work_struct *work)
>> +{
>> + struct vwork *vwork = container_of(work, struct vwork, work);
>> + struct mm_struct *mm;
>> +
>> + mm = vwork->mm;
>> + down_write(&mm->mmap_sem);
>> + mm->locked_vm += vwork->npage;
>> + up_write(&mm->mmap_sem);
>> + mmput(mm);
>> + kfree(vwork);
>> +}
>> +
>> +static void lock_acct(long npage)
>> +{
>> + struct vwork *vwork;
>> + struct mm_struct *mm;
>> +
>> + if (!current->mm)
>> + return; /* process exited */
>> +
>> + if (down_write_trylock(¤t->mm->mmap_sem)) {
>> + current->mm->locked_vm += npage;
>> + up_write(¤t->mm->mmap_sem);
>> + return;
>> + }
>> +
>> + /*
>> + * Couldn't get mmap_sem lock, so must setup to update
>> + * mm->locked_vm later. If locked_vm were atomic, we
>> + * wouldn't need this silliness
>> + */
>> + vwork = kmalloc(sizeof(struct vwork), GFP_KERNEL);
>> + if (!vwork)
>> + return;
>> + mm = get_task_mm(current);
>> + if (!mm) {
>> + kfree(vwork);
>> + return;
>> + }
>> + INIT_WORK(&vwork->work, lock_acct_bg);
>> + vwork->mm = mm;
>> + vwork->npage = npage;
>> + schedule_work(&vwork->work);
>> +}
>
> This looks familiar, should we split it out to a common file instead of
> duplicating it?
It is simple cut-n-paste from type1 driver :)
Moving it to a separate file is up to you but it is quite small piece of
code to move it somewhere, and I have not fixed rlimit handling yet, so
wait a bit.
>> +
>> +/*
>> + * The container descriptor supports only a single group per container.
>> + * Required by the API as the container is not supplied with the IOMMU group
>> + * at the moment of initialization.
>> + */
>> +struct tce_container {
>> + struct mutex lock;
>> + struct iommu_table *tbl;
>> +};
>> +
>> +static void *tce_iommu_open(unsigned long arg)
>> +{
>> + struct tce_container *container;
>> +
>> + if (arg != VFIO_SPAPR_TCE_IOMMU) {
>> + pr_err("tce_vfio: Wrong IOMMU type\n");
>> + return ERR_PTR(-EINVAL);
>> + }
>> +
>> + container = kzalloc(sizeof(*container), GFP_KERNEL);
>> + if (!container)
>> + return ERR_PTR(-ENOMEM);
>> +
>> + mutex_init(&container->lock);
>> +
>> + return container;
>> +}
>> +
>> +static void tce_iommu_release(void *iommu_data)
>> +{
>> + struct tce_container *container = iommu_data;
>> +
>> + WARN_ON(container->tbl && !container->tbl->it_group);
>> + if (container->tbl && container->tbl->it_group)
>> + tce_iommu_detach_group(iommu_data, container->tbl->it_group);
>> +
>> + mutex_destroy(&container->lock);
>> +
>> + kfree(container);
>> +}
>> +
>> +static long tce_iommu_ioctl(void *iommu_data,
>> + unsigned int cmd, unsigned long arg)
>> +{
>> + struct tce_container *container = iommu_data;
>> + unsigned long minsz;
>> + long ret;
>> +
>> + switch (cmd) {
>> + case VFIO_CHECK_EXTENSION: {
>> + return (arg == VFIO_SPAPR_TCE_IOMMU) ? 1 : 0;
>> + }
>> + case VFIO_IOMMU_SPAPR_TCE_GET_INFO: {
>> + struct vfio_iommu_spapr_tce_info info;
>> + struct iommu_table *tbl = container->tbl;
>> +
>> + if (WARN_ON(!tbl))
>> + return -ENXIO;
>> +
>> + minsz = offsetofend(struct vfio_iommu_spapr_tce_info,
>> + dma64_window_size);
>> +
>> + if (copy_from_user(&info, (void __user *)arg, minsz))
>> + return -EFAULT;
>> +
>> + if (info.argsz < minsz)
>> + return -EINVAL;
>> +
>> + info.dma32_window_start = tbl->it_offset << IOMMU_PAGE_SHIFT;
>> + info.dma32_window_size = tbl->it_size << IOMMU_PAGE_SHIFT;
>> + info.dma64_window_start = 0;
>> + info.dma64_window_size = 0;
>> + info.flags = 0;
>> +
>> + if (copy_to_user((void __user *)arg, &info, minsz))
>> + return -EFAULT;
>> +
>> + return 0;
>> + }
>> + case VFIO_IOMMU_MAP_DMA: {
>> + vfio_iommu_spapr_tce_dma_map param;
>> + struct iommu_table *tbl = container->tbl;
>> + enum dma_data_direction direction;
>> + unsigned long locked, lock_limit;
>> +
>> + if (WARN_ON(!tbl))
>> + return -ENXIO;
>> +
>> + minsz = offsetofend(vfio_iommu_spapr_tce_dma_map, size);
>> +
>> + if (copy_from_user(¶m, (void __user *)arg, minsz))
>> + return -EFAULT;
>> +
>> + if (param.argsz < minsz)
>> + return -EINVAL;
>> +
>> + if ((param.flags & VFIO_DMA_MAP_FLAG_READ) &&
>> + (param.flags & VFIO_DMA_MAP_FLAG_WRITE))
>> + direction = DMA_BIDIRECTIONAL;
>> + else if (param.flags & VFIO_DMA_MAP_FLAG_READ)
>> + direction = DMA_TO_DEVICE;
>> + else if (param.flags & VFIO_DMA_MAP_FLAG_WRITE)
>> + direction = DMA_FROM_DEVICE;
>> + else
>> + return -EINVAL;
>> +
>> + if ((param.size & ~IOMMU_PAGE_MASK) ||
>> + (param.iova & ~IOMMU_PAGE_MASK) ||
>> + (param.vaddr & ~IOMMU_PAGE_MASK))
>> + return -EINVAL;
>> +
>> + /* Account for locked pages */
>> + locked = current->mm->locked_vm +
>> + (param.size >> IOMMU_PAGE_SHIFT);
>> + lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
>
> This page accounting doesn't look right. PAGE_SIZE is several orders
> bigger than IOMMU_PAGE_SIZE (right?), but we mix them here, which seems
> like it will over penalize the user. For example, if a user maps 4x4k
> (assume aligned and contiguous) IOMMU pages, isn't that only a single
> pinned system page (assuming >=16k pages).
Oops. My bad. IOMMU_PAGE_SHIFT should be PAGE_SHIFT and should return the
number of system pages.
But we do not track 4K pages so I do not see any easy solution here. Except
fixing iommu_put_tces/iommu_clear_tces (*) to return the number of the very
first 4K IOMMU pages within system 64K pages.
This won't be too accurate but should work, no?
I'll post it as a patch in reply to "vfio powerpc: enabled on powernv
platform".
>> + if (locked > lock_limit && !capable(CAP_IPC_LOCK)) {
>> + pr_warn("RLIMIT_MEMLOCK (%ld) exceeded\n",
>> + rlimit(RLIMIT_MEMLOCK));
>> + return -ENOMEM;
>> + }
>> +
>> + ret = iommu_put_tces(tbl, param.iova >> IOMMU_PAGE_SHIFT,
>> + param.vaddr, direction,
>> + param.size >> IOMMU_PAGE_SHIFT);
>> + if (ret > 0)
>> + lock_acct(ret);
>> +
>> + return ret;
>> + }
>> + case VFIO_IOMMU_UNMAP_DMA: {
>> + vfio_iommu_spapr_tce_dma_unmap param;
>> + struct iommu_table *tbl = container->tbl;
>> +
>> + if (WARN_ON(!tbl))
>> + return -ENXIO;
>> +
>> + minsz = offsetofend(vfio_iommu_spapr_tce_dma_unmap, size);
>> +
>> + if (copy_from_user(¶m, (void __user *)arg, minsz))
>> + return -EFAULT;
>> +
>> + if (param.argsz < minsz)
>> + return -EINVAL;
>> +
>> + if ((param.size & ~IOMMU_PAGE_MASK) ||
>> + (param.iova & ~IOMMU_PAGE_MASK))
>> + return -EINVAL;
>> +
>> + ret = iommu_clear_tces(tbl, param.iova >> IOMMU_PAGE_SHIFT,
>> + param.size >> IOMMU_PAGE_SHIFT);
>> + if (ret > 0)
>> + lock_acct(-ret);
>> +
>> + return ret;
>> + }
>> + default:
>> + pr_warn("tce_vfio: unexpected cmd %x\n", cmd);
>> + }
>> +
>> + return -ENOTTY;
>> +}
>> +
>> +static int tce_iommu_attach_group(void *iommu_data,
>> + struct iommu_group *iommu_group)
>> +{
>> + struct tce_container *container = iommu_data;
>> + struct iommu_table *tbl = iommu_group_get_iommudata(iommu_group);
>> +
>> + BUG_ON(!tbl);
>> + mutex_lock(&container->lock);
>> + pr_debug("tce_vfio: Attaching group #%u to iommu %p\n",
>> + iommu_group_id(iommu_group), iommu_group);
>> + if (container->tbl) {
>> + pr_warn("tce_vfio: Only one group per IOMMU container is allowed, existing id=%d, attaching id=%d\n",
>> + iommu_group_id(container->tbl->it_group),
>> + iommu_group_id(iommu_group));
>> + mutex_unlock(&container->lock);
>> + return -EBUSY;
>> + }
>> +
>> + container->tbl = tbl;
>> + iommu_clear_tces(tbl, tbl->it_offset, tbl->it_size);
>> + mutex_unlock(&container->lock);
>> +
>> + return 0;
>> +}
>> +
>> +static void tce_iommu_detach_group(void *iommu_data,
>> + struct iommu_group *iommu_group)
>> +{
>> + struct tce_container *container = iommu_data;
>> + struct iommu_table *tbl = iommu_group_get_iommudata(iommu_group);
>> +
>> + BUG_ON(!tbl);
>> + mutex_lock(&container->lock);
>> + if (tbl != container->tbl) {
>> + pr_warn("tce_vfio: detaching group #%u, expected group is #%u\n",
>> + iommu_group_id(iommu_group),
>> + iommu_group_id(tbl->it_group));
>> + } else {
>> +
>> + pr_debug("tce_vfio: detaching group #%u from iommu %p\n",
>> + iommu_group_id(iommu_group), iommu_group);
>> +
>> + iommu_clear_tces(tbl, tbl->it_offset, tbl->it_size);
>> + container->tbl = NULL;
>> + }
>> + mutex_unlock(&container->lock);
>> +}
>> +
>> +const struct vfio_iommu_driver_ops tce_iommu_driver_ops = {
>> + .name = "iommu-vfio-powerpc",
>> + .owner = THIS_MODULE,
>> + .open = tce_iommu_open,
>> + .release = tce_iommu_release,
>> + .ioctl = tce_iommu_ioctl,
>> + .attach_group = tce_iommu_attach_group,
>> + .detach_group = tce_iommu_detach_group,
>> +};
>> +
>> +static int __init tce_iommu_init(void)
>> +{
>> + return vfio_register_iommu_driver(&tce_iommu_driver_ops);
>> +}
>> +
>> +static void __exit tce_iommu_cleanup(void)
>> +{
>> + vfio_unregister_iommu_driver(&tce_iommu_driver_ops);
>> +}
>> +
>> +module_init(tce_iommu_init);
>> +module_exit(tce_iommu_cleanup);
>> +
>> +MODULE_VERSION(DRIVER_VERSION);
>> +MODULE_LICENSE("GPL v2");
>> +MODULE_AUTHOR(DRIVER_AUTHOR);
>> +MODULE_DESCRIPTION(DRIVER_DESC);
>> +
>> diff --git a/include/linux/vfio.h b/include/linux/vfio.h
>> index 0a4f180..820af1e 100644
>> --- a/include/linux/vfio.h
>> +++ b/include/linux/vfio.h
>> @@ -99,6 +99,7 @@ extern void vfio_unregister_iommu_driver(
>> /* Extensions */
>>
>> #define VFIO_TYPE1_IOMMU 1
>> +#define VFIO_SPAPR_TCE_IOMMU 2
>>
>> /*
>> * The IOCTL interface is designed for extensibility by embedding the
>> @@ -442,4 +443,36 @@ struct vfio_iommu_type1_dma_unmap {
>>
>> #define VFIO_IOMMU_UNMAP_DMA _IO(VFIO_TYPE, VFIO_BASE + 14)
>>
>> +/* -------- Additional API for SPAPR TCE (Server POWERPC) IOMMU -------- */
>> +
>> +/*
>> + * The SPAPR TCE info struct provides the information about the PCI bus
>> + * address ranges available for DMA, these values are programmed into
>> + * the hardware so the guest has to know that information.
>> + *
>> + * Pages within 32 bit window should be explicitely mapped/unmapped via ioctls.
> ^^^^^^^^^^^
> explicitly
>
>> + * 64 bit window (not supported at the moment for the guest) is supposed to
>> + * be mapped completely to the guest memory so the devices capable of 64bit
>> + * DMA will not have to use map/unmap ioctls.
>> + *
>> + * The IOMMU page size is always 4K.
>> + */
>
> Thanks,
>
> Alex
>
>> +
>> +struct vfio_iommu_spapr_tce_info {
>> + __u32 argsz;
>> + __u32 flags; /* reserved for future use */
>> + __u32 dma32_window_start; /* 32 bit window start (bytes) */
>> + __u32 dma32_window_size; /* 32 bit window size (bytes) */
>> + __u64 dma64_window_start; /* 64 bit window start (bytes) */
>> + __u64 dma64_window_size; /* 64 bit window size (bytes) */
>> +};
>> +
>> +#define VFIO_IOMMU_SPAPR_TCE_GET_INFO _IO(VFIO_TYPE, VFIO_BASE + 12)
>> +
>> +/* Reuse type1 map/unmap structs as they are the same at the moment */
>> +typedef struct vfio_iommu_type1_dma_map vfio_iommu_spapr_tce_dma_map;
>> +typedef struct vfio_iommu_type1_dma_unmap vfio_iommu_spapr_tce_dma_unmap;
>> +
>> +/* ***************************************************************** */
>> +
>> #endif /* VFIO_H */
>
>
>
--
Alexey
^ permalink raw reply
* [PATCH 2/2] powerpc: collapse some equivalent asm macros
From: Stephen Rothwell @ 2012-11-28 23:57 UTC (permalink / raw)
To: Benjamin Herrenschmidt; +Cc: Alan Modra, ppc-dev
In-Reply-To: <20121129105525.0e47b3b3e51aebe7ded3b7b6@canb.auug.org.au>
[-- Attachment #1: Type: text/plain, Size: 3744 bytes --]
Since we removed the section changes from them, some of the entry point
macros are now the same, so just use the obvious ones.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
arch/powerpc/include/asm/ppc_asm.h | 41 ------------------------------------
arch/powerpc/kernel/head_64.S | 6 +++---
arch/powerpc/kernel/misc_32.S | 2 +-
arch/powerpc/kernel/misc_64.S | 2 +-
4 files changed, 5 insertions(+), 46 deletions(-)
diff --git a/arch/powerpc/include/asm/ppc_asm.h b/arch/powerpc/include/asm/ppc_asm.h
index 5c9af12..896204c 100644
--- a/arch/powerpc/include/asm/ppc_asm.h
+++ b/arch/powerpc/include/asm/ppc_asm.h
@@ -212,32 +212,6 @@ name: \
.type GLUE(.,name),@function; \
GLUE(.,name):
-#define _INIT_GLOBAL(name) \
- .align 2 ; \
- .globl name; \
- .globl GLUE(.,name); \
- .section ".opd","aw"; \
-name: \
- .quad GLUE(.,name); \
- .quad .TOC.@tocbase; \
- .quad 0; \
- .previous; \
- .type GLUE(.,name),@function; \
-GLUE(.,name):
-
-#define _KPROBE(name) \
- .align 2 ; \
- .globl name; \
- .globl GLUE(.,name); \
- .section ".opd","aw"; \
-name: \
- .quad GLUE(.,name); \
- .quad .TOC.@tocbase; \
- .quad 0; \
- .previous; \
- .type GLUE(.,name),@function; \
-GLUE(.,name):
-
#define _STATIC(name) \
.align 2 ; \
.section ".opd","aw"; \
@@ -249,17 +223,6 @@ name: \
.type GLUE(.,name),@function; \
GLUE(.,name):
-#define _INIT_STATIC(name) \
- .align 2 ; \
- .section ".opd","aw"; \
-name: \
- .quad GLUE(.,name); \
- .quad .TOC.@tocbase; \
- .quad 0; \
- .previous; \
- .type GLUE(.,name),@function; \
-GLUE(.,name):
-
#else /* 32-bit */
#define _ENTRY(n) \
@@ -271,10 +234,6 @@ n:
.globl n; \
n:
-#define _KPROBE(n) \
- .globl n; \
-n:
-
#endif
/*
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index 8d3edb9..721d17b 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -355,7 +355,7 @@ _GLOBAL(__start_initialization_multiplatform)
__REF
-_INIT_STATIC(__boot_from_prom)
+_STATIC(__boot_from_prom)
#ifdef CONFIG_PPC_OF_BOOT_TRAMPOLINE
/* Save parameters */
mr r31,r3
@@ -687,7 +687,7 @@ p_toc: .llong __toc_start + 0x8000 - 0b
/*
* This is where the main kernel code starts.
*/
-_INIT_STATIC(start_here_multiplatform)
+_STATIC(start_here_multiplatform)
/* set up the TOC (real address) */
bl .relative_toc
@@ -757,7 +757,7 @@ _INIT_STATIC(start_here_multiplatform)
b . /* prevent speculative execution */
/* This is where all platforms converge execution */
-_INIT_GLOBAL(start_here_common)
+_GLOBAL(start_here_common)
/* relocation is on at this point */
std r1,PACAKSAVE(r13)
diff --git a/arch/powerpc/kernel/misc_32.S b/arch/powerpc/kernel/misc_32.S
index 342c4d9..595af40 100644
--- a/arch/powerpc/kernel/misc_32.S
+++ b/arch/powerpc/kernel/misc_32.S
@@ -328,7 +328,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_UNIFIED_ID_CACHE)
* flush_icache_range(unsigned long start, unsigned long stop)
*/
.section ".kprobes.text","a"
-_KPROBE(__flush_icache_range)
+_ENTRY(__flush_icache_range)
BEGIN_FTR_SECTION
blr /* for 601, do nothing */
END_FTR_SECTION_IFSET(CPU_FTR_COHERENT_ICACHE)
diff --git a/arch/powerpc/kernel/misc_64.S b/arch/powerpc/kernel/misc_64.S
index 0fbe125..bc29512 100644
--- a/arch/powerpc/kernel/misc_64.S
+++ b/arch/powerpc/kernel/misc_64.S
@@ -68,7 +68,7 @@ PPC64_CACHES:
*/
.section ".kprobes.text","a"
-_KPROBE(__flush_icache_range)
+_GLOBAL(__flush_icache_range)
/*
* Flush the data cache to memory
--
1.7.10.280.gaa39
--
Cheers,
Stephen Rothwell sfr@canb.auug.org.au
[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]
^ permalink raw reply related
* [PATCH 1/2] powerpc: remove section changes from _GLOBAL() and friends
From: Stephen Rothwell @ 2012-11-28 23:55 UTC (permalink / raw)
To: Benjamin Herrenschmidt; +Cc: Alan Modra, ppc-dev
[-- Attachment #1: Type: text/plain, Size: 23929 bytes --]
These sometimes produce unexpected results and make it hard to put the
start up code (for 64 bit) into the .head.text section.
This also adds some explicit .texts at the start of asm files that did
not have them.
Suggested by Alan Modra.
Cc: Alan Modra <amodra@au1.ibm.com>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
arch/powerpc/include/asm/ppc_asm.h | 7 -------
arch/powerpc/kernel/cpu_setup_44x.S | 1 +
arch/powerpc/kernel/cpu_setup_6xx.S | 1 +
arch/powerpc/kernel/cpu_setup_a2.S | 1 +
arch/powerpc/kernel/cpu_setup_fsl_booke.S | 1 +
arch/powerpc/kernel/cpu_setup_pa6t.S | 1 +
arch/powerpc/kernel/cpu_setup_power.S | 1 +
arch/powerpc/kernel/cpu_setup_ppc970.S | 1 +
arch/powerpc/kernel/entry_32.S | 2 ++
arch/powerpc/kernel/fpu.S | 1 +
arch/powerpc/kernel/head_64.S | 7 +++++++
arch/powerpc/kernel/head_fsl_booke.S | 2 ++
arch/powerpc/kernel/idle_6xx.S | 5 +++++
arch/powerpc/kernel/idle_book3e.S | 2 ++
arch/powerpc/kernel/l2cr_6xx.S | 1 +
arch/powerpc/kernel/misc_32.S | 4 ++++
arch/powerpc/kernel/misc_64.S | 1 +
arch/powerpc/kernel/ppc_save_regs.S | 2 ++
arch/powerpc/kernel/reloc_32.S | 2 ++
arch/powerpc/kernel/reloc_64.S | 2 ++
arch/powerpc/kernel/swsusp_32.S | 5 +++++
arch/powerpc/kernel/swsusp_booke.S | 5 +++++
arch/powerpc/kernel/systbl.S | 2 ++
arch/powerpc/kernel/vector.S | 2 ++
arch/powerpc/kvm/book3s_hv_interrupts.S | 2 ++
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 2 ++
arch/powerpc/kvm/book3s_interrupts.S | 2 ++
arch/powerpc/kvm/book3s_rmhandlers.S | 2 ++
arch/powerpc/kvm/booke_interrupts.S | 2 ++
arch/powerpc/kvm/bookehv_interrupts.S | 2 ++
arch/powerpc/kvm/fpu.S | 2 ++
arch/powerpc/lib/checksum_64.S | 2 ++
arch/powerpc/lib/copypage_power7.S | 2 ++
arch/powerpc/lib/copyuser_64.S | 1 +
arch/powerpc/lib/copyuser_power7.S | 2 ++
arch/powerpc/lib/div64.S | 2 ++
arch/powerpc/lib/hweight_64.S | 2 ++
arch/powerpc/lib/ldstfp.S | 2 ++
arch/powerpc/lib/mem_64.S | 2 ++
arch/powerpc/lib/memcpy_64.S | 1 +
arch/powerpc/lib/memcpy_power7.S | 2 ++
arch/powerpc/mm/slb_low.S | 2 ++
arch/powerpc/mm/tlb_nohash_low.S | 2 ++
arch/powerpc/platforms/pasemi/powersave.S | 2 ++
arch/powerpc/platforms/powermac/cache.S | 2 ++
arch/powerpc/platforms/powernv/opal-wrappers.S | 2 ++
arch/powerpc/sysdev/6xx-suspend.S | 2 ++
arch/powerpc/sysdev/dcr-low.S | 2 ++
48 files changed, 98 insertions(+), 7 deletions(-)
diff --git a/arch/powerpc/include/asm/ppc_asm.h b/arch/powerpc/include/asm/ppc_asm.h
index ea2a86e..5c9af12 100644
--- a/arch/powerpc/include/asm/ppc_asm.h
+++ b/arch/powerpc/include/asm/ppc_asm.h
@@ -200,7 +200,6 @@ END_FW_FTR_SECTION_IFSET(FW_FEATURE_SPLPAR)
#define GLUE(a,b) XGLUE(a,b)
#define _GLOBAL(name) \
- .section ".text"; \
.align 2 ; \
.globl name; \
.globl GLUE(.,name); \
@@ -214,7 +213,6 @@ name: \
GLUE(.,name):
#define _INIT_GLOBAL(name) \
- __REF; \
.align 2 ; \
.globl name; \
.globl GLUE(.,name); \
@@ -228,7 +226,6 @@ name: \
GLUE(.,name):
#define _KPROBE(name) \
- .section ".kprobes.text","a"; \
.align 2 ; \
.globl name; \
.globl GLUE(.,name); \
@@ -242,7 +239,6 @@ name: \
GLUE(.,name):
#define _STATIC(name) \
- .section ".text"; \
.align 2 ; \
.section ".opd","aw"; \
name: \
@@ -254,7 +250,6 @@ name: \
GLUE(.,name):
#define _INIT_STATIC(name) \
- __REF; \
.align 2 ; \
.section ".opd","aw"; \
name: \
@@ -272,13 +267,11 @@ GLUE(.,name):
n:
#define _GLOBAL(n) \
- .text; \
.stabs __stringify(n:F-1),N_FUN,0,0,n;\
.globl n; \
n:
#define _KPROBE(n) \
- .section ".kprobes.text","a"; \
.globl n; \
n:
diff --git a/arch/powerpc/kernel/cpu_setup_44x.S b/arch/powerpc/kernel/cpu_setup_44x.S
index e32b4a9..3cac909 100644
--- a/arch/powerpc/kernel/cpu_setup_44x.S
+++ b/arch/powerpc/kernel/cpu_setup_44x.S
@@ -17,6 +17,7 @@
#include <asm/cputable.h>
#include <asm/ppc_asm.h>
+ .text
_GLOBAL(__setup_cpu_440ep)
b __init_fpu_44x
_GLOBAL(__setup_cpu_440epx)
diff --git a/arch/powerpc/kernel/cpu_setup_6xx.S b/arch/powerpc/kernel/cpu_setup_6xx.S
index f8cd9fb..4b65da9 100644
--- a/arch/powerpc/kernel/cpu_setup_6xx.S
+++ b/arch/powerpc/kernel/cpu_setup_6xx.S
@@ -17,6 +17,7 @@
#include <asm/cache.h>
#include <asm/mmu.h>
+ .text
_GLOBAL(__setup_cpu_603)
mflr r5
BEGIN_MMU_FTR_SECTION
diff --git a/arch/powerpc/kernel/cpu_setup_a2.S b/arch/powerpc/kernel/cpu_setup_a2.S
index 61f079e..3b52baa 100644
--- a/arch/powerpc/kernel/cpu_setup_a2.S
+++ b/arch/powerpc/kernel/cpu_setup_a2.S
@@ -36,6 +36,7 @@
(MMUCR1_EXTEND_PID | MMUCR1_CSINV_NEVER | MMUCR1_IRRE | \
MMUCR1_DRRE | MMUCR1_TLBWE_BINV)
+ .text
_GLOBAL(__setup_cpu_a2)
/* Some of these are actually thread local and some are
* core local but doing it always won't hurt
diff --git a/arch/powerpc/kernel/cpu_setup_fsl_booke.S b/arch/powerpc/kernel/cpu_setup_fsl_booke.S
index dcd8819..fb5204f 100644
--- a/arch/powerpc/kernel/cpu_setup_fsl_booke.S
+++ b/arch/powerpc/kernel/cpu_setup_fsl_booke.S
@@ -19,6 +19,7 @@
#include <asm/mmu-book3e.h>
#include <asm/asm-offsets.h>
+ .text
_GLOBAL(__e500_icache_setup)
mfspr r0, SPRN_L1CSR1
andi. r3, r0, L1CSR1_ICE
diff --git a/arch/powerpc/kernel/cpu_setup_pa6t.S b/arch/powerpc/kernel/cpu_setup_pa6t.S
index d62cb9c..81fd7ac 100644
--- a/arch/powerpc/kernel/cpu_setup_pa6t.S
+++ b/arch/powerpc/kernel/cpu_setup_pa6t.S
@@ -25,6 +25,7 @@
#include <asm/asm-offsets.h>
#include <asm/cache.h>
+ .text
/* Right now, restore and setup are the same thing */
_GLOBAL(__restore_cpu_pa6t)
_GLOBAL(__setup_cpu_pa6t)
diff --git a/arch/powerpc/kernel/cpu_setup_power.S b/arch/powerpc/kernel/cpu_setup_power.S
index 57cf140..ced4a39 100644
--- a/arch/powerpc/kernel/cpu_setup_power.S
+++ b/arch/powerpc/kernel/cpu_setup_power.S
@@ -16,6 +16,7 @@
#include <asm/asm-offsets.h>
#include <asm/cache.h>
+ .text
/* Entry: r3 = crap, r4 = ptr to cputable entry
*
* Note that we can be called twice for pseudo-PVRs
diff --git a/arch/powerpc/kernel/cpu_setup_ppc970.S b/arch/powerpc/kernel/cpu_setup_ppc970.S
index 12fac8d..220b479 100644
--- a/arch/powerpc/kernel/cpu_setup_ppc970.S
+++ b/arch/powerpc/kernel/cpu_setup_ppc970.S
@@ -16,6 +16,7 @@
#include <asm/asm-offsets.h>
#include <asm/cache.h>
+ .text
_GLOBAL(__cpu_preinit_ppc970)
/* Do nothing if not running in HV mode */
mfmsr r0
diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S
index 9499385..b2a4838 100644
--- a/arch/powerpc/kernel/entry_32.S
+++ b/arch/powerpc/kernel/entry_32.S
@@ -45,6 +45,8 @@
#define LOAD_MSR_KERNEL(r, x) li r,(x)
#endif
+ .text
+
#ifdef CONFIG_BOOKE
.globl mcheck_transfer_to_handler
mcheck_transfer_to_handler:
diff --git a/arch/powerpc/kernel/fpu.S b/arch/powerpc/kernel/fpu.S
index e0ada05..728f700 100644
--- a/arch/powerpc/kernel/fpu.S
+++ b/arch/powerpc/kernel/fpu.S
@@ -50,6 +50,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX); \
#define REST_32FPVSRS(n,c,base) __REST_32FPVSRS(n,__REG_##c,__REG_##base)
#define SAVE_32FPVSRS(n,c,base) __SAVE_32FPVSRS(n,__REG_##c,__REG_##base)
+ .text
/*
* This task wants to use the FPU now.
* On UP, disable FP for the task which had the FPU previously,
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index 116f086..8d3edb9 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -23,6 +23,7 @@
*/
#include <linux/threads.h>
+#include <linux/init.h>
#include <asm/reg.h>
#include <asm/page.h>
#include <asm/mmu.h>
@@ -352,6 +353,8 @@ _GLOBAL(__start_initialization_multiplatform)
b .__after_prom_start
#endif /* CONFIG_PPC_BOOK3E */
+ __REF
+
_INIT_STATIC(__boot_from_prom)
#ifdef CONFIG_PPC_OF_BOOT_TRAMPOLINE
/* Save parameters */
@@ -390,6 +393,8 @@ _INIT_STATIC(__boot_from_prom)
* from OF while CONFIG_PPC_OF_BOOT_TRAMPOLINE isn't selected */
trap
+ .text
+
_STATIC(__after_prom_start)
#ifdef CONFIG_RELOCATABLE
/* process relocations for the final address of the kernel */
@@ -677,6 +682,8 @@ _GLOBAL(relative_toc)
p_toc: .llong __toc_start + 0x8000 - 0b
+ __REF
+
/*
* This is where the main kernel code starts.
*/
diff --git a/arch/powerpc/kernel/head_fsl_booke.S b/arch/powerpc/kernel/head_fsl_booke.S
index 6f62a73..4ec5625 100644
--- a/arch/powerpc/kernel/head_fsl_booke.S
+++ b/arch/powerpc/kernel/head_fsl_booke.S
@@ -776,6 +776,8 @@ tlb_write_entry:
mfspr r10, SPRN_SPRG_RSCRATCH0
rfi /* Force context change */
+ ,text
+
#ifdef CONFIG_SPE
/* Note that the SPE support is closely modeled after the AltiVec
* support. Changes to one are likely to be applicable to the
diff --git a/arch/powerpc/kernel/idle_6xx.S b/arch/powerpc/kernel/idle_6xx.S
index 1686916..4c414e9 100644
--- a/arch/powerpc/kernel/idle_6xx.S
+++ b/arch/powerpc/kernel/idle_6xx.S
@@ -187,6 +187,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_DUAL_PLL_750FX)
.data
+/*
+ * hmm, should this ".text" be here? The following _GLOBAL()s
+ * used to switch to the text section
+ */
+ .text
_GLOBAL(nap_save_msscr0)
.space 4*NR_CPUS
diff --git a/arch/powerpc/kernel/idle_book3e.S b/arch/powerpc/kernel/idle_book3e.S
index 4c7cb400..0d90d6a 100644
--- a/arch/powerpc/kernel/idle_book3e.S
+++ b/arch/powerpc/kernel/idle_book3e.S
@@ -17,6 +17,8 @@
#include <asm/processor.h>
#include <asm/thread_info.h>
+ .text
+
/* 64-bit version only for now */
#ifdef CONFIG_PPC64
diff --git a/arch/powerpc/kernel/l2cr_6xx.S b/arch/powerpc/kernel/l2cr_6xx.S
index 97ec855..000b337 100644
--- a/arch/powerpc/kernel/l2cr_6xx.S
+++ b/arch/powerpc/kernel/l2cr_6xx.S
@@ -96,6 +96,7 @@ IMPORTANT:
* passed in.
* -- paulus.
*/
+ .text
_GLOBAL(_set_L2CR)
/* Make sure this is a 750 or 7400 chip */
BEGIN_FTR_SECTION
diff --git a/arch/powerpc/kernel/misc_32.S b/arch/powerpc/kernel/misc_32.S
index 19e096b..342c4d9 100644
--- a/arch/powerpc/kernel/misc_32.S
+++ b/arch/powerpc/kernel/misc_32.S
@@ -327,6 +327,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_UNIFIED_ID_CACHE)
*
* flush_icache_range(unsigned long start, unsigned long stop)
*/
+ .section ".kprobes.text","a"
_KPROBE(__flush_icache_range)
BEGIN_FTR_SECTION
blr /* for 601, do nothing */
@@ -357,6 +358,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_COHERENT_ICACHE)
sync /* additional sync needed on g4 */
isync
blr
+
+ .text
+
/*
* Write any modified data cache blocks out to memory.
* Does not invalidate the corresponding cache lines (especially for
diff --git a/arch/powerpc/kernel/misc_64.S b/arch/powerpc/kernel/misc_64.S
index 5cfa800..0fbe125 100644
--- a/arch/powerpc/kernel/misc_64.S
+++ b/arch/powerpc/kernel/misc_64.S
@@ -67,6 +67,7 @@ PPC64_CACHES:
* flush all bytes from start through stop-1 inclusive
*/
+ .section ".kprobes.text","a"
_KPROBE(__flush_icache_range)
/*
diff --git a/arch/powerpc/kernel/ppc_save_regs.S b/arch/powerpc/kernel/ppc_save_regs.S
index 1b1787d..48b41a0 100644
--- a/arch/powerpc/kernel/ppc_save_regs.S
+++ b/arch/powerpc/kernel/ppc_save_regs.S
@@ -13,6 +13,8 @@
#include <asm/asm-offsets.h>
#include <asm/ptrace.h>
+ .text
+
/*
* Grab the register values as they are now.
* This won't do a particularly good job because we really
diff --git a/arch/powerpc/kernel/reloc_32.S b/arch/powerpc/kernel/reloc_32.S
index ef46ba6..180a356a 100644
--- a/arch/powerpc/kernel/reloc_32.S
+++ b/arch/powerpc/kernel/reloc_32.S
@@ -31,6 +31,8 @@ R_PPC_RELATIVE = 22
* r3 = desired final address
*/
+ .text
+
_GLOBAL(relocate)
mflr r0 /* Save our LR */
diff --git a/arch/powerpc/kernel/reloc_64.S b/arch/powerpc/kernel/reloc_64.S
index b47a0e1..3ced962 100644
--- a/arch/powerpc/kernel/reloc_64.S
+++ b/arch/powerpc/kernel/reloc_64.S
@@ -15,6 +15,8 @@ RELA = 7
RELACOUNT = 0x6ffffff9
R_PPC64_RELATIVE = 22
+ .text
+
/*
* r3 = desired final address of kernel
*/
diff --git a/arch/powerpc/kernel/swsusp_32.S b/arch/powerpc/kernel/swsusp_32.S
index ba4dee3..7cd3243 100644
--- a/arch/powerpc/kernel/swsusp_32.S
+++ b/arch/powerpc/kernel/swsusp_32.S
@@ -33,6 +33,11 @@
.section .data
.align 5
+/* should this ".text" really be here? The _GLOBAL() used to have
+ * the ".text" embedded in it.
+ */
+ .text
+
_GLOBAL(swsusp_save_area)
.space SL_SIZE
diff --git a/arch/powerpc/kernel/swsusp_booke.S b/arch/powerpc/kernel/swsusp_booke.S
index 11a3930..f30f5d3 100644
--- a/arch/powerpc/kernel/swsusp_booke.S
+++ b/arch/powerpc/kernel/swsusp_booke.S
@@ -39,6 +39,11 @@
.section .data
.align 5
+/*
+ * Should this ".text" be here? The _GLOBAL() used to have
+ * the ".text" embedded in it.
+ */
+ .text
_GLOBAL(swsusp_save_area)
.space SL_SIZE
diff --git a/arch/powerpc/kernel/systbl.S b/arch/powerpc/kernel/systbl.S
index 93219c3..af3d3be 100644
--- a/arch/powerpc/kernel/systbl.S
+++ b/arch/powerpc/kernel/systbl.S
@@ -36,6 +36,8 @@
#define PPC_SYS_SPU(func) PPC_SYS(func)
#define SYSX_SPU(f, f3264, f32) SYSX(f, f3264, f32)
+ .text
+
#ifdef CONFIG_PPC64
#define sys_sigpending sys_ni_syscall
#define sys_old_getrlimit sys_ni_syscall
diff --git a/arch/powerpc/kernel/vector.S b/arch/powerpc/kernel/vector.S
index e830289..8560e88 100644
--- a/arch/powerpc/kernel/vector.S
+++ b/arch/powerpc/kernel/vector.S
@@ -7,6 +7,8 @@
#include <asm/page.h>
#include <asm/ptrace.h>
+ .text
+
/*
* load_up_altivec(unused, unused, tsk)
* Disable VMX for the task which had it previously,
diff --git a/arch/powerpc/kvm/book3s_hv_interrupts.S b/arch/powerpc/kvm/book3s_hv_interrupts.S
index 84035a5..f319e2f 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupts.S
+++ b/arch/powerpc/kvm/book3s_hv_interrupts.S
@@ -28,6 +28,8 @@
#include <asm/exception-64s.h>
#include <asm/ppc-opcode.h>
+ .text
+
/*****************************************************************************
* *
* Guest entry / exit code that is in kernel module memory (vmalloc) *
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 74a24bb..f8667e3 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -28,6 +28,8 @@
#include <asm/exception-64s.h>
#include <asm/kvm_book3s_asm.h>
+ .text
+
/*****************************************************************************
* *
* Real Mode handlers that need to be in the linear mapping *
diff --git a/arch/powerpc/kvm/book3s_interrupts.S b/arch/powerpc/kvm/book3s_interrupts.S
index 48cbbf8..672c21a 100644
--- a/arch/powerpc/kvm/book3s_interrupts.S
+++ b/arch/powerpc/kvm/book3s_interrupts.S
@@ -50,6 +50,8 @@
PPC_LL r30, VCPU_GPR(R30)(vcpu); \
PPC_LL r31, VCPU_GPR(R31)(vcpu); \
+ .text
+
/*****************************************************************************
* *
* Guest entry / exit code that is in kernel module memory (highmem) *
diff --git a/arch/powerpc/kvm/book3s_rmhandlers.S b/arch/powerpc/kvm/book3s_rmhandlers.S
index 9ecf6e3..4883e92 100644
--- a/arch/powerpc/kvm/book3s_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_rmhandlers.S
@@ -28,6 +28,8 @@
#include <asm/exception-64s.h>
#endif
+ .text
+
/*****************************************************************************
* *
* Real Mode handlers that need to be in low physical memory *
diff --git a/arch/powerpc/kvm/booke_interrupts.S b/arch/powerpc/kvm/booke_interrupts.S
index bb46b32..f379cda 100644
--- a/arch/powerpc/kvm/booke_interrupts.S
+++ b/arch/powerpc/kvm/booke_interrupts.S
@@ -73,6 +73,8 @@ _GLOBAL(kvmppc_handler_\ivor_nr)
bctr
.endm
+ .text
+
_GLOBAL(kvmppc_handlers_start)
KVM_HANDLER BOOKE_INTERRUPT_CRITICAL SPRN_SPRG_RSCRATCH_CRIT SPRN_CSRR0
KVM_HANDLER BOOKE_INTERRUPT_MACHINE_CHECK SPRN_SPRG_RSCRATCH_MC SPRN_MCSRR0
diff --git a/arch/powerpc/kvm/bookehv_interrupts.S b/arch/powerpc/kvm/bookehv_interrupts.S
index 099fe82..4ec294f 100644
--- a/arch/powerpc/kvm/bookehv_interrupts.S
+++ b/arch/powerpc/kvm/bookehv_interrupts.S
@@ -60,6 +60,8 @@
#define NEED_DEAR 0x00000002 /* save faulting DEAR */
#define NEED_ESR 0x00000004 /* save faulting ESR */
+ .text
+
/*
* On entry:
* r4 = vcpu, r5 = srr0, r6 = srr1
diff --git a/arch/powerpc/kvm/fpu.S b/arch/powerpc/kvm/fpu.S
index bf68d59..dfe7d48 100644
--- a/arch/powerpc/kvm/fpu.S
+++ b/arch/powerpc/kvm/fpu.S
@@ -22,6 +22,8 @@
/* Instructions operating on single parameters */
+ .text
+
/*
* Single operation with one input operand
*
diff --git a/arch/powerpc/lib/checksum_64.S b/arch/powerpc/lib/checksum_64.S
index 167f725..539d9ac 100644
--- a/arch/powerpc/lib/checksum_64.S
+++ b/arch/powerpc/lib/checksum_64.S
@@ -17,6 +17,8 @@
#include <asm/errno.h>
#include <asm/ppc_asm.h>
+ .text
+
/*
* ip_fast_csum(r3=buf, r4=len) -- Optimized for IP header
* len is in words and is always >= 5.
diff --git a/arch/powerpc/lib/copypage_power7.S b/arch/powerpc/lib/copypage_power7.S
index 0ef75bf..df72190 100644
--- a/arch/powerpc/lib/copypage_power7.S
+++ b/arch/powerpc/lib/copypage_power7.S
@@ -20,6 +20,8 @@
#include <asm/page.h>
#include <asm/ppc_asm.h>
+ .text
+
_GLOBAL(copypage_power7)
/*
* We prefetch both the source and destination using enhanced touch
diff --git a/arch/powerpc/lib/copyuser_64.S b/arch/powerpc/lib/copyuser_64.S
index d73a590..9e57cd5 100644
--- a/arch/powerpc/lib/copyuser_64.S
+++ b/arch/powerpc/lib/copyuser_64.S
@@ -9,6 +9,7 @@
#include <asm/processor.h>
#include <asm/ppc_asm.h>
+ .text
.align 7
_GLOBAL(__copy_tofrom_user)
BEGIN_FTR_SECTION
diff --git a/arch/powerpc/lib/copyuser_power7.S b/arch/powerpc/lib/copyuser_power7.S
index 0d24ff1..c2b2269 100644
--- a/arch/powerpc/lib/copyuser_power7.S
+++ b/arch/powerpc/lib/copyuser_power7.S
@@ -19,6 +19,8 @@
*/
#include <asm/ppc_asm.h>
+ .text
+
.macro err1
100:
.section __ex_table,"a"
diff --git a/arch/powerpc/lib/div64.S b/arch/powerpc/lib/div64.S
index 83d9832..439c5f5 100644
--- a/arch/powerpc/lib/div64.S
+++ b/arch/powerpc/lib/div64.S
@@ -16,6 +16,8 @@
#include <asm/ppc_asm.h>
#include <asm/processor.h>
+ .text
+
_GLOBAL(__div64_32)
lwz r5,0(r3) # get the dividend into r5/r6
lwz r6,4(r3)
diff --git a/arch/powerpc/lib/hweight_64.S b/arch/powerpc/lib/hweight_64.S
index 9b96ff2..6db0f35 100644
--- a/arch/powerpc/lib/hweight_64.S
+++ b/arch/powerpc/lib/hweight_64.S
@@ -22,6 +22,8 @@
/* Note: This code relies on -mminimal-toc */
+ .text
+
_GLOBAL(__arch_hweight8)
BEGIN_FTR_SECTION
b .__sw_hweight8
diff --git a/arch/powerpc/lib/ldstfp.S b/arch/powerpc/lib/ldstfp.S
index 85aec08..2707323 100644
--- a/arch/powerpc/lib/ldstfp.S
+++ b/arch/powerpc/lib/ldstfp.S
@@ -37,6 +37,8 @@ reg = reg + 1
.endr
.endm
+ .text
+
/* Get the contents of frN into fr0; N is in r3. */
_GLOBAL(get_fpr)
mflr r0
diff --git a/arch/powerpc/lib/mem_64.S b/arch/powerpc/lib/mem_64.S
index f4fcb0b..6f7e636 100644
--- a/arch/powerpc/lib/mem_64.S
+++ b/arch/powerpc/lib/mem_64.S
@@ -12,6 +12,8 @@
#include <asm/errno.h>
#include <asm/ppc_asm.h>
+ .text
+
_GLOBAL(memset)
neg r0,r3
rlwimi r4,r4,8,16,23
diff --git a/arch/powerpc/lib/memcpy_64.S b/arch/powerpc/lib/memcpy_64.S
index d2bbbc8..d318765 100644
--- a/arch/powerpc/lib/memcpy_64.S
+++ b/arch/powerpc/lib/memcpy_64.S
@@ -9,6 +9,7 @@
#include <asm/processor.h>
#include <asm/ppc_asm.h>
+ .text
.align 7
_GLOBAL(memcpy)
BEGIN_FTR_SECTION
diff --git a/arch/powerpc/lib/memcpy_power7.S b/arch/powerpc/lib/memcpy_power7.S
index 0663630..e804ab8 100644
--- a/arch/powerpc/lib/memcpy_power7.S
+++ b/arch/powerpc/lib/memcpy_power7.S
@@ -19,6 +19,8 @@
*/
#include <asm/ppc_asm.h>
+ .text
+
_GLOBAL(memcpy_power7)
#ifdef CONFIG_ALTIVEC
cmpldi r5,16
diff --git a/arch/powerpc/mm/slb_low.S b/arch/powerpc/mm/slb_low.S
index 1a16ca2..fd2413b 100644
--- a/arch/powerpc/mm/slb_low.S
+++ b/arch/powerpc/mm/slb_low.S
@@ -23,6 +23,8 @@
#include <asm/pgtable.h>
#include <asm/firmware.h>
+ .text
+
/* void slb_allocate_realmode(unsigned long ea);
*
* Create an SLB entry for the given EA (user or kernel).
diff --git a/arch/powerpc/mm/tlb_nohash_low.S b/arch/powerpc/mm/tlb_nohash_low.S
index 626ad08..487f71a 100644
--- a/arch/powerpc/mm/tlb_nohash_low.S
+++ b/arch/powerpc/mm/tlb_nohash_low.S
@@ -35,6 +35,8 @@
#include <asm/processor.h>
#include <asm/bug.h>
+ .text
+
#if defined(CONFIG_40x)
/*
diff --git a/arch/powerpc/platforms/pasemi/powersave.S b/arch/powerpc/platforms/pasemi/powersave.S
index 56f45ad..29a64fa 100644
--- a/arch/powerpc/platforms/pasemi/powersave.S
+++ b/arch/powerpc/platforms/pasemi/powersave.S
@@ -43,6 +43,8 @@
1: cmpd r3,r3; \
bne 1b
+ .text
+
_doze:
PRE_SLEEP_SEQUENCE
DOZE
diff --git a/arch/powerpc/platforms/powermac/cache.S b/arch/powerpc/platforms/powermac/cache.S
index 6be1a4a..af387c9 100644
--- a/arch/powerpc/platforms/powermac/cache.S
+++ b/arch/powerpc/platforms/powermac/cache.S
@@ -18,6 +18,8 @@
#include <asm/ppc_asm.h>
#include <asm/cputable.h>
+ .text
+
/*
* Flush and disable all data caches (dL1, L2, L3). This is used
* when going to sleep, when doing a PMU based cpufreq transition,
diff --git a/arch/powerpc/platforms/powernv/opal-wrappers.S b/arch/powerpc/platforms/powernv/opal-wrappers.S
index 3bb07e5..1a9648a 100644
--- a/arch/powerpc/platforms/powernv/opal-wrappers.S
+++ b/arch/powerpc/platforms/powernv/opal-wrappers.S
@@ -14,6 +14,8 @@
#include <asm/asm-offsets.h>
#include <asm/opal.h>
+ .text
+
/* TODO:
*
* - Trace irqs in/off (needs saving/restoring all args, argh...)
diff --git a/arch/powerpc/sysdev/6xx-suspend.S b/arch/powerpc/sysdev/6xx-suspend.S
index cf48e9c..c87ced0 100644
--- a/arch/powerpc/sysdev/6xx-suspend.S
+++ b/arch/powerpc/sysdev/6xx-suspend.S
@@ -16,6 +16,8 @@
#include <asm/thread_info.h>
#include <asm/asm-offsets.h>
+ .text
+
_GLOBAL(mpc6xx_enter_standby)
mflr r4
diff --git a/arch/powerpc/sysdev/dcr-low.S b/arch/powerpc/sysdev/dcr-low.S
index d3098ef..aadc7ec 100644
--- a/arch/powerpc/sysdev/dcr-low.S
+++ b/arch/powerpc/sysdev/dcr-low.S
@@ -26,6 +26,8 @@
EMIT_BUG_ENTRY 1b,__FILE__,__LINE__,0; \
blr
+ .text
+
_GLOBAL(__mfdcr)
DCR_ACCESS_PROLOG(__mfdcr_table)
--
1.7.10.280.gaa39
--
Cheers,
Stephen Rothwell sfr@canb.auug.org.au
[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]
^ permalink raw reply related
* Re: [PATCH V7 1/7] kbuild: centralize .dts->.dtb rule
From: Benjamin Herrenschmidt @ 2012-11-28 23:28 UTC (permalink / raw)
To: Stephen Warren
Cc: Michal Marek, linux-arch, linux-kernel, Rob Herring,
Paul Mackerras, linuxppc-dev
In-Reply-To: <50B65903.1050400@wwwdotorg.org>
On Wed, 2012-11-28 at 11:33 -0700, Stephen Warren wrote:
> Following this patch (http://lkml.org/lkml/2012/11/27/555), I posted a
> series of patches to convert almost all architectures to using the
> centralized rule. The one architecture I didn't convert was PowerPC.
>
> I didn't convert it because arch/powerpc/boot/Makefile contains a large
> number of rules (to generate *Image.% where % is a board name) that
> depend on %.dtb, which is expected to be in arch/powerpc/boot rather
> than arch/powerpc/boot/dts. Now, I guess it's possible to convert them
> all to expect the .dtb files to be in dts/ and also have
> arch/powerpc/boot/Makefile call make in boot/dts/ to make each required
> .dtb file. However, the patch would be a bit larger than all the other
> architecture patches. Do you want me to do that conversion, or would you
> rather I leave PowerPC alone? Thanks for any feedback.
Kumar, any objection to moving the dtb's to arch/powerpc/boot/dtb/ ?
Other than breaking a script or two out there ...
Cheers,
Ben.
^ permalink raw reply
* Re: [PATCH] vfio powerpc: enabled on powernv platform
From: Alex Williamson @ 2012-11-28 21:30 UTC (permalink / raw)
To: Alexey Kardashevskiy
Cc: David Gibson, Paul Mackerras, linuxppc-dev, linux-kernel
In-Reply-To: <1354087113-26733-1-git-send-email-aik@ozlabs.ru>
On Wed, 2012-11-28 at 18:18 +1100, Alexey Kardashevskiy wrote:
> This patch initializes IOMMU groups based on the IOMMU
> configuration discovered during the PCI scan on POWERNV
> (POWER non virtualized) platform. The IOMMU groups are
> to be used later by VFIO driver (PCI pass through).
>
> It also implements an API for mapping/unmapping pages for
> guest PCI drivers and providing DMA window properties.
> This API is going to be used later by QEMU-VFIO to handle
> h_put_tce hypercalls from the KVM guest.
>
> Although this driver has been tested only on the POWERNV
> platform, it should work on any platform which supports
> TCE tables.
>
> To enable VFIO on POWER, enable SPAPR_TCE_IOMMU config
> option and configure VFIO as required.
>
> Cc: David Gibson <david@gibson.dropbear.id.au>
> Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
> ---
> arch/powerpc/include/asm/iommu.h | 9 +++
> arch/powerpc/kernel/iommu.c | 147 ++++++++++++++++++++++++++++++++++
> arch/powerpc/platforms/powernv/pci.c | 135 +++++++++++++++++++++++++++++++
> drivers/iommu/Kconfig | 8 ++
> 4 files changed, 299 insertions(+)
>
> diff --git a/arch/powerpc/include/asm/iommu.h b/arch/powerpc/include/asm/iommu.h
> index cbfe678..5c7087a 100644
> --- a/arch/powerpc/include/asm/iommu.h
> +++ b/arch/powerpc/include/asm/iommu.h
> @@ -76,6 +76,9 @@ struct iommu_table {
> struct iommu_pool large_pool;
> struct iommu_pool pools[IOMMU_NR_POOLS];
> unsigned long *it_map; /* A simple allocation bitmap for now */
> +#ifdef CONFIG_IOMMU_API
> + struct iommu_group *it_group;
> +#endif
> };
>
> struct scatterlist;
> @@ -147,5 +150,11 @@ static inline void iommu_restore(void)
> }
> #endif
>
> +extern long iommu_clear_tces(struct iommu_table *tbl, unsigned long entry,
> + unsigned long pages);
> +extern long iommu_put_tces(struct iommu_table *tbl, unsigned long entry,
> + uint64_t tce, enum dma_data_direction direction,
> + unsigned long pages);
> +
> #endif /* __KERNEL__ */
> #endif /* _ASM_IOMMU_H */
> diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c
> index ff5a6ce..1456b6e 100644
> --- a/arch/powerpc/kernel/iommu.c
> +++ b/arch/powerpc/kernel/iommu.c
> @@ -44,6 +44,7 @@
> #include <asm/kdump.h>
> #include <asm/fadump.h>
> #include <asm/vio.h>
> +#include <asm/tce.h>
>
> #define DBG(...)
>
> @@ -856,3 +857,149 @@ void iommu_free_coherent(struct iommu_table *tbl, size_t size,
> free_pages((unsigned long)vaddr, get_order(size));
> }
> }
> +
> +#ifdef CONFIG_IOMMU_API
> +/*
> + * SPAPR TCE API
> + */
> +static void tce_flush(struct iommu_table *tbl)
> +{
> + /* Flush/invalidate TLB caches if necessary */
> + if (ppc_md.tce_flush)
> + ppc_md.tce_flush(tbl);
> +
> + /* Make sure updates are seen by hardware */
> + mb();
> +}
> +
> +/*
> + * iommu_clear_tces clears tces and returned the number of pages
> + * which it called put_page() on.
> + */
> +static long clear_tces_nolock(struct iommu_table *tbl, unsigned long entry,
> + unsigned long pages)
> +{
> + int i, pages_put = 0;
> + unsigned long oldtce;
> + struct page *page;
> +
> + for (i = 0; i < pages; ++i) {
> + oldtce = ppc_md.tce_get(tbl, entry + i);
> + ppc_md.tce_free(tbl, entry + i, 1);
> +
> + if (!(oldtce & (TCE_PCI_WRITE | TCE_PCI_READ)))
> + continue;
> +
> + page = pfn_to_page(oldtce >> PAGE_SHIFT);
> +
> + WARN_ON(!page);
> + if (!page)
> + continue;
> +
> + if (oldtce & TCE_PCI_WRITE)
> + SetPageDirty(page);
> +
> + ++pages_put;
> + put_page(page);
> + }
> +
> + return pages_put;
> +}
> +
> +/*
> + * iommu_clear_tces clears tces and returned the number of released pages
> + */
> +long iommu_clear_tces(struct iommu_table *tbl, unsigned long entry,
> + unsigned long pages)
> +{
> + int ret;
> + struct iommu_pool *pool = get_pool(tbl, entry);
> +
> + spin_lock(&(pool->lock));
> + ret = clear_tces_nolock(tbl, entry, pages);
> + tce_flush(tbl);
> + spin_unlock(&(pool->lock));
> +
> + return ret;
> +}
> +EXPORT_SYMBOL_GPL(iommu_clear_tces);
> +
> +static int put_tce(struct iommu_table *tbl, unsigned long entry,
> + uint64_t tce, enum dma_data_direction direction)
> +{
> + int ret;
> + struct page *page = NULL;
> + unsigned long kva, offset;
> +
> + /* Map new TCE */
> + offset = (tce & IOMMU_PAGE_MASK) - (tce & PAGE_MASK);
> +
> + ret = get_user_pages_fast(tce & PAGE_MASK, 1,
> + direction != DMA_TO_DEVICE, &page);
> + if (ret < 1) {
> + printk(KERN_ERR "tce_vfio: get_user_pages_fast failed tce=%llx ioba=%lx ret=%d\n",
> + tce, entry << IOMMU_PAGE_SHIFT, ret);
> + if (!ret)
> + ret = -EFAULT;
> + return ret;
> + }
> +
> + kva = (unsigned long) page_address(page);
> + kva += offset;
> +
> + /* tce_build receives a virtual address */
> + entry += tbl->it_offset; /* Offset into real TCE table */
> + ret = ppc_md.tce_build(tbl, entry, 1, kva, direction, NULL);
> +
> + /* tce_build() only returns non-zero for transient errors */
> + if (unlikely(ret)) {
> + printk(KERN_ERR "tce_vfio: tce_put failed on tce=%llx ioba=%lx kva=%lx ret=%d\n",
> + tce, entry << IOMMU_PAGE_SHIFT, kva, ret);
> + put_page(page);
> + return -EIO;
> + }
> +
> + return 0;
> +}
> +
> +/*
> + * iommu_put_tces builds tces and returned the number of actually locked pages
> + */
> +long iommu_put_tces(struct iommu_table *tbl, unsigned long entry,
> + uint64_t tce, enum dma_data_direction direction,
> + unsigned long pages)
> +{
> + int i, ret = 0;
> + struct iommu_pool *pool = get_pool(tbl, entry);
> +
> + BUILD_BUG_ON(PAGE_SIZE < IOMMU_PAGE_SIZE);
> + BUG_ON(direction == DMA_NONE);
> +
> + spin_lock(&(pool->lock));
> +
> + /* Check if any is in use */
> + for (i = 0; i < pages; ++i) {
> + unsigned long oldtce = ppc_md.tce_get(tbl, entry + i);
> + if (oldtce & (TCE_PCI_WRITE | TCE_PCI_READ)) {
> + spin_unlock(&(pool->lock));
> + return -EBUSY;
> + }
> + }
> +
> + /* Put tces to the table */
> + for (i = 0; (i < pages) && !ret; ++i, tce += IOMMU_PAGE_SIZE)
> + ret = put_tce(tbl, entry + i, tce, direction);
> +
> + /* If failed, release locked pages, otherwise return the number of pages */
> + if (ret)
> + clear_tces_nolock(tbl, entry, i);
> + else
> + ret = pages;
> +
> + tce_flush(tbl);
> + spin_unlock(&(pool->lock));
> +
> + return ret;
> +}
Nice, no more kmalloc! I'm still concerned about the IOMMU page size
mismatch here. If nothing else, the comment is misleading since we're
locking system pages, but returning tce pages. The user would therefore
need to multiply their locked memory limit by (PAGE_SHIFT -
IOMMU_PAGE_SHIFT). Thanks,
Alex
> +EXPORT_SYMBOL_GPL(iommu_put_tces);
> +#endif /* CONFIG_IOMMU_API */
> diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c
> index 05205cf..21250ef 100644
> --- a/arch/powerpc/platforms/powernv/pci.c
> +++ b/arch/powerpc/platforms/powernv/pci.c
> @@ -20,6 +20,7 @@
> #include <linux/irq.h>
> #include <linux/io.h>
> #include <linux/msi.h>
> +#include <linux/iommu.h>
>
> #include <asm/sections.h>
> #include <asm/io.h>
> @@ -613,3 +614,137 @@ void __init pnv_pci_init(void)
> ppc_md.teardown_msi_irqs = pnv_teardown_msi_irqs;
> #endif
> }
> +
> +#ifdef CONFIG_IOMMU_API
> +/*
> + * IOMMU groups support required by VFIO
> + */
> +static int add_device(struct device *dev)
> +{
> + struct iommu_table *tbl;
> + int ret = 0;
> +
> + if (WARN_ON(dev->iommu_group)) {
> + printk(KERN_WARNING "tce_vfio: device %s is already in iommu group %d, skipping\n",
> + dev_name(dev),
> + iommu_group_id(dev->iommu_group));
> + return -EBUSY;
> + }
> +
> + tbl = get_iommu_table_base(dev);
> + if (!tbl) {
> + pr_debug("tce_vfio: skipping device %s with no tbl\n",
> + dev_name(dev));
> + return 0;
> + }
> +
> + pr_debug("tce_vfio: adding %s to iommu group %d\n",
> + dev_name(dev), iommu_group_id(tbl->it_group));
> +
> + ret = iommu_group_add_device(tbl->it_group, dev);
> + if (ret < 0)
> + printk(KERN_ERR "tce_vfio: %s has not been added, ret=%d\n",
> + dev_name(dev), ret);
> +
> + return ret;
> +}
> +
> +static void del_device(struct device *dev)
> +{
> + iommu_group_remove_device(dev);
> +}
> +
> +static int iommu_bus_notifier(struct notifier_block *nb,
> + unsigned long action, void *data)
> +{
> + struct device *dev = data;
> +
> + switch (action) {
> + case BUS_NOTIFY_ADD_DEVICE:
> + return add_device(dev);
> + case BUS_NOTIFY_DEL_DEVICE:
> + del_device(dev);
> + return 0;
> + default:
> + return 0;
> + }
> +}
> +
> +static struct notifier_block tce_iommu_bus_nb = {
> + .notifier_call = iommu_bus_notifier,
> +};
> +
> +static void group_release(void *iommu_data)
> +{
> + struct iommu_table *tbl = iommu_data;
> + tbl->it_group = NULL;
> +}
> +
> +static int __init tce_iommu_init(void)
> +{
> + struct pci_dev *pdev = NULL;
> + struct iommu_table *tbl;
> + struct iommu_group *grp;
> +
> + /* Allocate and initialize IOMMU groups */
> + for_each_pci_dev(pdev) {
> + tbl = get_iommu_table_base(&pdev->dev);
> + if (!tbl)
> + continue;
> +
> + /* Skip already initialized */
> + if (tbl->it_group)
> + continue;
> +
> + grp = iommu_group_alloc();
> + if (IS_ERR(grp)) {
> + printk(KERN_INFO "tce_vfio: cannot create "
> + "new IOMMU group, ret=%ld\n",
> + PTR_ERR(grp));
> + return PTR_ERR(grp);
> + }
> + tbl->it_group = grp;
> + iommu_group_set_iommudata(grp, tbl, group_release);
> + }
> +
> + bus_register_notifier(&pci_bus_type, &tce_iommu_bus_nb);
> +
> + /* Add PCI devices to VFIO groups */
> + for_each_pci_dev(pdev)
> + add_device(&pdev->dev);
> +
> + return 0;
> +}
> +
> +static void __exit tce_iommu_cleanup(void)
> +{
> + struct pci_dev *pdev = NULL;
> + struct iommu_table *tbl;
> + struct iommu_group *grp = NULL;
> +
> + bus_unregister_notifier(&pci_bus_type, &tce_iommu_bus_nb);
> +
> + /* Delete PCI devices from VFIO groups */
> + for_each_pci_dev(pdev)
> + del_device(&pdev->dev);
> +
> + /* Release VFIO groups */
> + for_each_pci_dev(pdev) {
> + tbl = get_iommu_table_base(&pdev->dev);
> + if (!tbl)
> + continue;
> + grp = tbl->it_group;
> +
> + /* Skip (already) uninitialized */
> + if (!grp)
> + continue;
> +
> + /* Do actual release, group_release() is expected to work */
> + iommu_group_put(grp);
> + BUG_ON(tbl->it_group);
> + }
> +}
> +
> +module_init(tce_iommu_init);
> +module_exit(tce_iommu_cleanup);
> +#endif /* CONFIG_IOMMU_API */
> diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
> index 9f69b56..29d11dc 100644
> --- a/drivers/iommu/Kconfig
> +++ b/drivers/iommu/Kconfig
> @@ -187,4 +187,12 @@ config EXYNOS_IOMMU_DEBUG
>
> Say N unless you need kernel log message for IOMMU debugging
>
> +config SPAPR_TCE_IOMMU
> + bool "sPAPR TCE IOMMU Support"
> + depends on PPC_POWERNV
> + select IOMMU_API
> + help
> + Enables bits of IOMMU API required by VFIO. The iommu_ops is
> + still not implemented.
> +
> endif # IOMMU_SUPPORT
^ permalink raw reply
* Re: [PATCH] vfio powerpc: implemented IOMMU driver for VFIO
From: Alex Williamson @ 2012-11-28 21:01 UTC (permalink / raw)
To: Alexey Kardashevskiy
Cc: kvm, linux-kernel, Paul Mackerras, linuxppc-dev, David Gibson
In-Reply-To: <1354087299-26791-1-git-send-email-aik@ozlabs.ru>
On Wed, 2012-11-28 at 18:21 +1100, Alexey Kardashevskiy wrote:
> VFIO implements platform independent stuff such as
> a PCI driver, BAR access (via read/write on a file descriptor
> or direct mapping when possible) and IRQ signaling.
>
> The platform dependent part includes IOMMU initialization
> and handling. This patch implements an IOMMU driver for VFIO
> which does mapping/unmapping pages for the guest IO and
> provides information about DMA window (required by a POWERPC
> guest).
>
> The counterpart in QEMU is required to support this functionality.
>
> Cc: David Gibson <david@gibson.dropbear.id.au>
> Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
> ---
> drivers/vfio/Kconfig | 6 +
> drivers/vfio/Makefile | 1 +
> drivers/vfio/vfio_iommu_spapr_tce.c | 332 +++++++++++++++++++++++++++++++++++
> include/linux/vfio.h | 33 ++++
> 4 files changed, 372 insertions(+)
> create mode 100644 drivers/vfio/vfio_iommu_spapr_tce.c
>
> diff --git a/drivers/vfio/Kconfig b/drivers/vfio/Kconfig
> index 7cd5dec..b464687 100644
> --- a/drivers/vfio/Kconfig
> +++ b/drivers/vfio/Kconfig
> @@ -3,10 +3,16 @@ config VFIO_IOMMU_TYPE1
> depends on VFIO
> default n
>
> +config VFIO_IOMMU_SPAPR_TCE
> + tristate
> + depends on VFIO && SPAPR_TCE_IOMMU
> + default n
> +
> menuconfig VFIO
> tristate "VFIO Non-Privileged userspace driver framework"
> depends on IOMMU_API
> select VFIO_IOMMU_TYPE1 if X86
> + select VFIO_IOMMU_SPAPR_TCE if PPC_POWERNV
> help
> VFIO provides a framework for secure userspace device drivers.
> See Documentation/vfio.txt for more details.
> diff --git a/drivers/vfio/Makefile b/drivers/vfio/Makefile
> index 2398d4a..72bfabc 100644
> --- a/drivers/vfio/Makefile
> +++ b/drivers/vfio/Makefile
> @@ -1,3 +1,4 @@
> obj-$(CONFIG_VFIO) += vfio.o
> obj-$(CONFIG_VFIO_IOMMU_TYPE1) += vfio_iommu_type1.o
> +obj-$(CONFIG_VFIO_IOMMU_SPAPR_TCE) += vfio_iommu_spapr_tce.o
> obj-$(CONFIG_VFIO_PCI) += pci/
> diff --git a/drivers/vfio/vfio_iommu_spapr_tce.c b/drivers/vfio/vfio_iommu_spapr_tce.c
> new file mode 100644
> index 0000000..b98770e
> --- /dev/null
> +++ b/drivers/vfio/vfio_iommu_spapr_tce.c
> @@ -0,0 +1,332 @@
> +/*
> + * VFIO: IOMMU DMA mapping support for TCE on POWER
> + *
> + * Copyright (C) 2012 IBM Corp. All rights reserved.
> + * Author: Alexey Kardashevskiy <aik@ozlabs.ru>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 as
> + * published by the Free Software Foundation.
> + *
> + * Derived from original vfio_iommu_type1.c:
> + * Copyright (C) 2012 Red Hat, Inc. All rights reserved.
> + * Author: Alex Williamson <alex.williamson@redhat.com>
> + */
> +
> +#include <linux/module.h>
> +#include <linux/pci.h>
> +#include <linux/slab.h>
> +#include <linux/uaccess.h>
> +#include <linux/err.h>
> +#include <linux/vfio.h>
> +#include <asm/iommu.h>
> +
> +#define DRIVER_VERSION "0.1"
> +#define DRIVER_AUTHOR "aik@ozlabs.ru"
> +#define DRIVER_DESC "VFIO IOMMU SPAPR TCE"
> +
> +static void tce_iommu_detach_group(void *iommu_data,
> + struct iommu_group *iommu_group);
> +
> +/*
> + * VFIO IOMMU fd for SPAPR_TCE IOMMU implementation
> + */
> +
> +/*
> + * This code handles mapping and unmapping of user data buffers
> + * into DMA'ble space using the IOMMU
> + */
> +
> +#define NPAGE_TO_SIZE(npage) ((size_t)(npage) << PAGE_SHIFT)
> +
> +struct vwork {
> + struct mm_struct *mm;
> + long npage;
> + struct work_struct work;
> +};
> +
> +/* delayed decrement/increment for locked_vm */
> +static void lock_acct_bg(struct work_struct *work)
> +{
> + struct vwork *vwork = container_of(work, struct vwork, work);
> + struct mm_struct *mm;
> +
> + mm = vwork->mm;
> + down_write(&mm->mmap_sem);
> + mm->locked_vm += vwork->npage;
> + up_write(&mm->mmap_sem);
> + mmput(mm);
> + kfree(vwork);
> +}
> +
> +static void lock_acct(long npage)
> +{
> + struct vwork *vwork;
> + struct mm_struct *mm;
> +
> + if (!current->mm)
> + return; /* process exited */
> +
> + if (down_write_trylock(¤t->mm->mmap_sem)) {
> + current->mm->locked_vm += npage;
> + up_write(¤t->mm->mmap_sem);
> + return;
> + }
> +
> + /*
> + * Couldn't get mmap_sem lock, so must setup to update
> + * mm->locked_vm later. If locked_vm were atomic, we
> + * wouldn't need this silliness
> + */
> + vwork = kmalloc(sizeof(struct vwork), GFP_KERNEL);
> + if (!vwork)
> + return;
> + mm = get_task_mm(current);
> + if (!mm) {
> + kfree(vwork);
> + return;
> + }
> + INIT_WORK(&vwork->work, lock_acct_bg);
> + vwork->mm = mm;
> + vwork->npage = npage;
> + schedule_work(&vwork->work);
> +}
This looks familiar, should we split it out to a common file instead of
duplicating it?
> +
> +/*
> + * The container descriptor supports only a single group per container.
> + * Required by the API as the container is not supplied with the IOMMU group
> + * at the moment of initialization.
> + */
> +struct tce_container {
> + struct mutex lock;
> + struct iommu_table *tbl;
> +};
> +
> +static void *tce_iommu_open(unsigned long arg)
> +{
> + struct tce_container *container;
> +
> + if (arg != VFIO_SPAPR_TCE_IOMMU) {
> + pr_err("tce_vfio: Wrong IOMMU type\n");
> + return ERR_PTR(-EINVAL);
> + }
> +
> + container = kzalloc(sizeof(*container), GFP_KERNEL);
> + if (!container)
> + return ERR_PTR(-ENOMEM);
> +
> + mutex_init(&container->lock);
> +
> + return container;
> +}
> +
> +static void tce_iommu_release(void *iommu_data)
> +{
> + struct tce_container *container = iommu_data;
> +
> + WARN_ON(container->tbl && !container->tbl->it_group);
> + if (container->tbl && container->tbl->it_group)
> + tce_iommu_detach_group(iommu_data, container->tbl->it_group);
> +
> + mutex_destroy(&container->lock);
> +
> + kfree(container);
> +}
> +
> +static long tce_iommu_ioctl(void *iommu_data,
> + unsigned int cmd, unsigned long arg)
> +{
> + struct tce_container *container = iommu_data;
> + unsigned long minsz;
> + long ret;
> +
> + switch (cmd) {
> + case VFIO_CHECK_EXTENSION: {
> + return (arg == VFIO_SPAPR_TCE_IOMMU) ? 1 : 0;
> + }
> + case VFIO_IOMMU_SPAPR_TCE_GET_INFO: {
> + struct vfio_iommu_spapr_tce_info info;
> + struct iommu_table *tbl = container->tbl;
> +
> + if (WARN_ON(!tbl))
> + return -ENXIO;
> +
> + minsz = offsetofend(struct vfio_iommu_spapr_tce_info,
> + dma64_window_size);
> +
> + if (copy_from_user(&info, (void __user *)arg, minsz))
> + return -EFAULT;
> +
> + if (info.argsz < minsz)
> + return -EINVAL;
> +
> + info.dma32_window_start = tbl->it_offset << IOMMU_PAGE_SHIFT;
> + info.dma32_window_size = tbl->it_size << IOMMU_PAGE_SHIFT;
> + info.dma64_window_start = 0;
> + info.dma64_window_size = 0;
> + info.flags = 0;
> +
> + if (copy_to_user((void __user *)arg, &info, minsz))
> + return -EFAULT;
> +
> + return 0;
> + }
> + case VFIO_IOMMU_MAP_DMA: {
> + vfio_iommu_spapr_tce_dma_map param;
> + struct iommu_table *tbl = container->tbl;
> + enum dma_data_direction direction;
> + unsigned long locked, lock_limit;
> +
> + if (WARN_ON(!tbl))
> + return -ENXIO;
> +
> + minsz = offsetofend(vfio_iommu_spapr_tce_dma_map, size);
> +
> + if (copy_from_user(¶m, (void __user *)arg, minsz))
> + return -EFAULT;
> +
> + if (param.argsz < minsz)
> + return -EINVAL;
> +
> + if ((param.flags & VFIO_DMA_MAP_FLAG_READ) &&
> + (param.flags & VFIO_DMA_MAP_FLAG_WRITE))
> + direction = DMA_BIDIRECTIONAL;
> + else if (param.flags & VFIO_DMA_MAP_FLAG_READ)
> + direction = DMA_TO_DEVICE;
> + else if (param.flags & VFIO_DMA_MAP_FLAG_WRITE)
> + direction = DMA_FROM_DEVICE;
> + else
> + return -EINVAL;
> +
> + if ((param.size & ~IOMMU_PAGE_MASK) ||
> + (param.iova & ~IOMMU_PAGE_MASK) ||
> + (param.vaddr & ~IOMMU_PAGE_MASK))
> + return -EINVAL;
> +
> + /* Account for locked pages */
> + locked = current->mm->locked_vm +
> + (param.size >> IOMMU_PAGE_SHIFT);
> + lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
This page accounting doesn't look right. PAGE_SIZE is several orders
bigger than IOMMU_PAGE_SIZE (right?), but we mix them here, which seems
like it will over penalize the user. For example, if a user maps 4x4k
(assume aligned and contiguous) IOMMU pages, isn't that only a single
pinned system page (assuming >=16k pages).
> + if (locked > lock_limit && !capable(CAP_IPC_LOCK)) {
> + pr_warn("RLIMIT_MEMLOCK (%ld) exceeded\n",
> + rlimit(RLIMIT_MEMLOCK));
> + return -ENOMEM;
> + }
> +
> + ret = iommu_put_tces(tbl, param.iova >> IOMMU_PAGE_SHIFT,
> + param.vaddr, direction,
> + param.size >> IOMMU_PAGE_SHIFT);
> + if (ret > 0)
> + lock_acct(ret);
> +
> + return ret;
> + }
> + case VFIO_IOMMU_UNMAP_DMA: {
> + vfio_iommu_spapr_tce_dma_unmap param;
> + struct iommu_table *tbl = container->tbl;
> +
> + if (WARN_ON(!tbl))
> + return -ENXIO;
> +
> + minsz = offsetofend(vfio_iommu_spapr_tce_dma_unmap, size);
> +
> + if (copy_from_user(¶m, (void __user *)arg, minsz))
> + return -EFAULT;
> +
> + if (param.argsz < minsz)
> + return -EINVAL;
> +
> + if ((param.size & ~IOMMU_PAGE_MASK) ||
> + (param.iova & ~IOMMU_PAGE_MASK))
> + return -EINVAL;
> +
> + ret = iommu_clear_tces(tbl, param.iova >> IOMMU_PAGE_SHIFT,
> + param.size >> IOMMU_PAGE_SHIFT);
> + if (ret > 0)
> + lock_acct(-ret);
> +
> + return ret;
> + }
> + default:
> + pr_warn("tce_vfio: unexpected cmd %x\n", cmd);
> + }
> +
> + return -ENOTTY;
> +}
> +
> +static int tce_iommu_attach_group(void *iommu_data,
> + struct iommu_group *iommu_group)
> +{
> + struct tce_container *container = iommu_data;
> + struct iommu_table *tbl = iommu_group_get_iommudata(iommu_group);
> +
> + BUG_ON(!tbl);
> + mutex_lock(&container->lock);
> + pr_debug("tce_vfio: Attaching group #%u to iommu %p\n",
> + iommu_group_id(iommu_group), iommu_group);
> + if (container->tbl) {
> + pr_warn("tce_vfio: Only one group per IOMMU container is allowed, existing id=%d, attaching id=%d\n",
> + iommu_group_id(container->tbl->it_group),
> + iommu_group_id(iommu_group));
> + mutex_unlock(&container->lock);
> + return -EBUSY;
> + }
> +
> + container->tbl = tbl;
> + iommu_clear_tces(tbl, tbl->it_offset, tbl->it_size);
> + mutex_unlock(&container->lock);
> +
> + return 0;
> +}
> +
> +static void tce_iommu_detach_group(void *iommu_data,
> + struct iommu_group *iommu_group)
> +{
> + struct tce_container *container = iommu_data;
> + struct iommu_table *tbl = iommu_group_get_iommudata(iommu_group);
> +
> + BUG_ON(!tbl);
> + mutex_lock(&container->lock);
> + if (tbl != container->tbl) {
> + pr_warn("tce_vfio: detaching group #%u, expected group is #%u\n",
> + iommu_group_id(iommu_group),
> + iommu_group_id(tbl->it_group));
> + } else {
> +
> + pr_debug("tce_vfio: detaching group #%u from iommu %p\n",
> + iommu_group_id(iommu_group), iommu_group);
> +
> + iommu_clear_tces(tbl, tbl->it_offset, tbl->it_size);
> + container->tbl = NULL;
> + }
> + mutex_unlock(&container->lock);
> +}
> +
> +const struct vfio_iommu_driver_ops tce_iommu_driver_ops = {
> + .name = "iommu-vfio-powerpc",
> + .owner = THIS_MODULE,
> + .open = tce_iommu_open,
> + .release = tce_iommu_release,
> + .ioctl = tce_iommu_ioctl,
> + .attach_group = tce_iommu_attach_group,
> + .detach_group = tce_iommu_detach_group,
> +};
> +
> +static int __init tce_iommu_init(void)
> +{
> + return vfio_register_iommu_driver(&tce_iommu_driver_ops);
> +}
> +
> +static void __exit tce_iommu_cleanup(void)
> +{
> + vfio_unregister_iommu_driver(&tce_iommu_driver_ops);
> +}
> +
> +module_init(tce_iommu_init);
> +module_exit(tce_iommu_cleanup);
> +
> +MODULE_VERSION(DRIVER_VERSION);
> +MODULE_LICENSE("GPL v2");
> +MODULE_AUTHOR(DRIVER_AUTHOR);
> +MODULE_DESCRIPTION(DRIVER_DESC);
> +
> diff --git a/include/linux/vfio.h b/include/linux/vfio.h
> index 0a4f180..820af1e 100644
> --- a/include/linux/vfio.h
> +++ b/include/linux/vfio.h
> @@ -99,6 +99,7 @@ extern void vfio_unregister_iommu_driver(
> /* Extensions */
>
> #define VFIO_TYPE1_IOMMU 1
> +#define VFIO_SPAPR_TCE_IOMMU 2
>
> /*
> * The IOCTL interface is designed for extensibility by embedding the
> @@ -442,4 +443,36 @@ struct vfio_iommu_type1_dma_unmap {
>
> #define VFIO_IOMMU_UNMAP_DMA _IO(VFIO_TYPE, VFIO_BASE + 14)
>
> +/* -------- Additional API for SPAPR TCE (Server POWERPC) IOMMU -------- */
> +
> +/*
> + * The SPAPR TCE info struct provides the information about the PCI bus
> + * address ranges available for DMA, these values are programmed into
> + * the hardware so the guest has to know that information.
> + *
> + * Pages within 32 bit window should be explicitely mapped/unmapped via ioctls.
^^^^^^^^^^^
explicitly
> + * 64 bit window (not supported at the moment for the guest) is supposed to
> + * be mapped completely to the guest memory so the devices capable of 64bit
> + * DMA will not have to use map/unmap ioctls.
> + *
> + * The IOMMU page size is always 4K.
> + */
Thanks,
Alex
> +
> +struct vfio_iommu_spapr_tce_info {
> + __u32 argsz;
> + __u32 flags; /* reserved for future use */
> + __u32 dma32_window_start; /* 32 bit window start (bytes) */
> + __u32 dma32_window_size; /* 32 bit window size (bytes) */
> + __u64 dma64_window_start; /* 64 bit window start (bytes) */
> + __u64 dma64_window_size; /* 64 bit window size (bytes) */
> +};
> +
> +#define VFIO_IOMMU_SPAPR_TCE_GET_INFO _IO(VFIO_TYPE, VFIO_BASE + 12)
> +
> +/* Reuse type1 map/unmap structs as they are the same at the moment */
> +typedef struct vfio_iommu_type1_dma_map vfio_iommu_spapr_tce_dma_map;
> +typedef struct vfio_iommu_type1_dma_unmap vfio_iommu_spapr_tce_dma_unmap;
> +
> +/* ***************************************************************** */
> +
> #endif /* VFIO_H */
^ permalink raw reply
* Re: [GIT PULL 0/8] perf/urgent fixes
From: Josh Boyer @ 2012-11-28 20:27 UTC (permalink / raw)
To: Ingo Molnar, Thomas Gleixner, Linus Torvalds,
Arnaldo Carvalho de Melo
Cc: Arnaldo Carvalho de Melo, David Howells, linux-arch,
Peter Zijlstra, Robert Richter, Namhyung Kim, Anton Blanchard,
linux-kernel, Xiao Guangrong, Arnaldo Carvalho de Melo,
linuxppc-dev, Dong Hao, Borislav Petkov, David Ahern,
Runzhen Wang, Paul Mackerras, x86, Sukadev Bhattiprolu
In-Reply-To: <1353716453-9693-1-git-send-email-acme@infradead.org>
On Fri, Nov 23, 2012 at 7:20 PM, Arnaldo Carvalho de Melo
<acme@ghostprotocols.net> wrote:
> Hi Ingo,
>
> Tested using a cross-compiler and directly on a Raspberry pi (ARM) with
> raspbian.
>
> Please consider pulling.
>
> - Arnaldo
>
> The following changes since commit 18423d3562f396206e0928a71177eeb2edfed077:
>
> Merge tag 'perf-urgent-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent (2012-11-13 18:51:51 +0100)
>
> are available in the git repository at:
>
>
> git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux tags/perf-urgent-for-mingo
>
> for you to fetch changes up to 7321090f6751c9987c26a8c81c63680d16a614d7:
>
> perf kvm: Fix building perf kvm on non x86 arches (2012-11-23 20:40:17 -0300)
>
> ----------------------------------------------------------------
> perf/urgent fixes
>
> . Don't build 'perf kvm stat" on non-x86 arches, fix from Xiao Guangrong.
>
> . UAPI fixes to get perf building again in non-x86 arches, from David Howells.
>
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
>
> ----------------------------------------------------------------
> Arnaldo Carvalho de Melo (1):
> Merge tag 'perf-uapi-20121119' of git://git.infradead.org/users/dhowells/linux-headers into perf/urgent
>
> David Howells (6):
> x86: Export asm/{svm.h,vmx.h,perf_regs.h}
> Merge branch 'x86-pre-uapi' into perf-uapi
> tools: Define a Makefile function to do subdir processing
> tools: Honour the O= flag when tool build called from a higher Makefile
> tools: Pass the target in descend
> perf: Make perf build for x86 with UAPI disintegration applied
>
> Sukadev Bhattiprolu (1):
> perf powerpc: Use uapi/unistd.h to fix build error
>
> Xiao Guangrong (2):
> perf kvm: Rename perf_kvm to perf_kvm_stat
> perf kvm: Fix building perf kvm on non x86 arches
I probably sound like a broken record at this point, but I've not seen
this go into any tip branch, nor is it in Linus' tree. Hopefully this
gets pulled before 3.7 is released.
josh
^ permalink raw reply
* [PATCH] export of_reconfig_notifier_[register,unregister]
From: Nathan Fontenot @ 2012-11-28 19:42 UTC (permalink / raw)
To: Stephen Rothwell
Cc: linux-kernel, Rob Herring, linux-next, Paul Mackerras,
linuxppc-dev
In-Reply-To: <20121128140308.abb21ec8f61553a2ed077fe3@canb.auug.org.au>
The of reconfiguration notification chains should be exported for use
by modules.
Signed-off-by:Nathan Fontenot <nfont@linux.vnet.ibm.com>
---
Index: linux-next/drivers/of/base.c
===================================================================
--- linux-next.orig/drivers/of/base.c 2012-11-28 09:18:02.000000000 -0600
+++ linux-next/drivers/of/base.c 2012-11-28 11:05:00.000000000 -0600
@@ -1282,11 +1282,13 @@
{
return blocking_notifier_chain_register(&of_reconfig_chain, nb);
}
+EXPORT_SYMBOL_GPL(of_reconfig_notifier_register);
int of_reconfig_notifier_unregister(struct notifier_block *nb)
{
return blocking_notifier_chain_unregister(&of_reconfig_chain, nb);
}
+EXPORT_SYMBOL_GPL(of_reconfig_notifier_unregister);
int of_reconfig_notify(unsigned long action, void *p)
{
^ permalink raw reply
* Re: 3.7-rc7: BUG: MAX_STACK_TRACE_ENTRIES too low!
From: Christian Kujau @ 2012-11-28 18:59 UTC (permalink / raw)
To: Li Zhong; +Cc: paulus, linuxppc-dev, LKML
In-Reply-To: <1354092082.3054.19.camel@ThinkPad-T5421.cn.ibm.com>
On Wed, 28 Nov 2012 at 16:41, Li Zhong wrote:
> Would you please help to try the following fix? I don't have a powerpc32
> machine for test...
I've just applied this to 3.7-rc7 and booted the machine. I don't know how
to trigger this bug, so it might take a while until it happens again - or
not, now with your patch applied.
It happened only 2 times so far, after ~8h and after ~20h:
Nov 5 13:28:20 alice kernel: [ 0.000000] Linux version 3.7.0-rc4
Nov 5 21:00:26 alice kernel: [27148.965634] BUG: MAX_STACK_TRACE_ENTRIES too low!
Nov 26 21:53:43 alice kernel: [ 0.000000] Linux version 3.7.0-rc7
Nov 27 17:15:29 alice kernel: [69731.388717] BUG: MAX_STACK_TRACE_ENTRIES too low!
Thanks,
Christian.
--
BOFH excuse #86:
Runt packets
^ permalink raw reply
* Re: [PATCH V7 1/7] kbuild: centralize .dts->.dtb rule
From: Stephen Warren @ 2012-11-28 18:33 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Paul Mackerras
Cc: Michal Marek, linux-arch, linux-kernel, Rob Herring, linuxppc-dev
In-Reply-To: <1354058956-7199-1-git-send-email-swarren@wwwdotorg.org>
On 11/27/2012 04:29 PM, Stephen Warren wrote:
> All architectures that use cmd_dtc do so in almost the same way. Create
> a central build rule to avoid duplication. The one difference is that
> most current uses of dtc build $(obj)/%.dtb from $(src)/dts/%.dts rather
> than building the .dtb in the same directory as the .dts file. This
> difference will be eliminated arch-by-arch in future patches.
>
> MIPS is the exception here; it already uses the exact same rule as the
> new common rule, so the duplicate is removed in this patch to avoid any
> conflict. arch/mips changes courtesy of Ralf Baechle.
>
> Update Documentation/kbuild to remove the explicit call to cmd_dtc from
> the example, now that the rule exists in a centralized location.
Ben, Paul,
Following this patch (http://lkml.org/lkml/2012/11/27/555), I posted a
series of patches to convert almost all architectures to using the
centralized rule. The one architecture I didn't convert was PowerPC.
I didn't convert it because arch/powerpc/boot/Makefile contains a large
number of rules (to generate *Image.% where % is a board name) that
depend on %.dtb, which is expected to be in arch/powerpc/boot rather
than arch/powerpc/boot/dts. Now, I guess it's possible to convert them
all to expect the .dtb files to be in dts/ and also have
arch/powerpc/boot/Makefile call make in boot/dts/ to make each required
.dtb file. However, the patch would be a bit larger than all the other
architecture patches. Do you want me to do that conversion, or would you
rather I leave PowerPC alone? Thanks for any feedback.
^ permalink raw reply
* Re: [Patch v4 08/12] memory-hotplug: remove memmap of sparse-vmemmap
From: Jianguo Wu @ 2012-11-28 9:40 UTC (permalink / raw)
To: Wen Congyang
Cc: linux-s390, linux-ia64, Len Brown, linux-acpi, linux-sh, x86,
linux-kernel, cmetcalf, linux-mm, Yasuaki Ishimatsu, paulus,
Minchan Kim, KOSAKI Motohiro, David Rientjes, sparclinux,
Christoph Lameter, linuxppc-dev, Andrew Morton, Jiang Liu
In-Reply-To: <1354010422-19648-9-git-send-email-wency@cn.fujitsu.com>
Hi Congyang,
I think vmemmap's pgtable pages should be freed after all entries are cleared, I have a patch to do this.
The code logic is the same as [Patch v4 09/12] memory-hotplug: remove page table of x86_64 architecture.
How do you think about this?
Signed-off-by: Jianguo Wu <wujianguo@huawei.com>
Signed-off-by: Jiang Liu <jiang.liu@huawei.com>
---
include/linux/mm.h | 1 +
mm/sparse-vmemmap.c | 214 +++++++++++++++++++++++++++++++++++++++++++++++++++
mm/sparse.c | 5 +-
3 files changed, 218 insertions(+), 2 deletions(-)
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 5657670..1f26af5 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1642,6 +1642,7 @@ int vmemmap_populate(struct page *start_page, unsigned long pages, int node);
void vmemmap_populate_print_last(void);
void register_page_bootmem_memmap(unsigned long section_nr, struct page *map,
unsigned long size);
+void vmemmap_free(struct page *memmap, unsigned long nr_pages);
enum mf_flags {
MF_COUNT_INCREASED = 1 << 0,
diff --git a/mm/sparse-vmemmap.c b/mm/sparse-vmemmap.c
index 1b7e22a..242cb28 100644
--- a/mm/sparse-vmemmap.c
+++ b/mm/sparse-vmemmap.c
@@ -29,6 +29,10 @@
#include <asm/pgalloc.h>
#include <asm/pgtable.h>
+#ifdef CONFIG_MEMORY_HOTREMOVE
+#include <asm/tlbflush.h>
+#endif
+
/*
* Allocate a block of memory to be used to back the virtual memory map
* or to back the page tables that are used to create the mapping.
@@ -224,3 +228,213 @@ void __init sparse_mem_maps_populate_node(struct page **map_map,
vmemmap_buf_end = NULL;
}
}
+
+#ifdef CONFIG_MEMORY_HOTREMOVE
+static void vmemmap_free_pages(struct page *page, int order)
+{
+ struct zone *zone;
+ unsigned long magic;
+
+ magic = (unsigned long) page->lru.next;
+ if (magic == SECTION_INFO || magic == MIX_SECTION_INFO) {
+ put_page_bootmem(page);
+
+ zone = page_zone(page);
+ zone_span_writelock(zone);
+ zone->present_pages++;
+ zone_span_writeunlock(zone);
+ totalram_pages++;
+ } else {
+ if (is_vmalloc_addr(page_address(page)))
+ vfree(page_address(page));
+ else
+ free_pages((unsigned long)page_address(page), order);
+ }
+}
+
+static void free_pte_table(pmd_t *pmd)
+{
+ pte_t *pte, *pte_start;
+ int i;
+
+ pte_start = (pte_t *)pmd_page_vaddr(*pmd);
+ for (i = 0; i < PTRS_PER_PTE; i++) {
+ pte = pte_start + i;
+ if (pte_val(*pte))
+ return;
+ }
+
+ /* free a pte talbe */
+ vmemmap_free_pages(pmd_page(*pmd), 0);
+ spin_lock(&init_mm.page_table_lock);
+ pmd_clear(pmd);
+ spin_unlock(&init_mm.page_table_lock);
+}
+
+static void free_pmd_table(pud_t *pud)
+{
+ pmd_t *pmd, *pmd_start;
+ int i;
+
+ pmd_start = (pmd_t *)pud_page_vaddr(*pud);
+ for (i = 0; i < PTRS_PER_PMD; i++) {
+ pmd = pmd_start + i;
+ if (pmd_val(*pmd))
+ return;
+ }
+
+ /* free a pmd talbe */
+ vmemmap_free_pages(pud_page(*pud), 0);
+ spin_lock(&init_mm.page_table_lock);
+ pud_clear(pud);
+ spin_unlock(&init_mm.page_table_lock);
+}
+
+static void free_pud_table(pgd_t *pgd)
+{
+ pud_t *pud, *pud_start;
+ int i;
+
+ pud_start = (pud_t *)pgd_page_vaddr(*pgd);
+ for (i = 0; i < PTRS_PER_PUD; i++) {
+ pud = pud_start + i;
+ if (pud_val(*pud))
+ return;
+ }
+
+ /* free a pud table */
+ vmemmap_free_pages(pgd_page(*pgd), 0);
+ spin_lock(&init_mm.page_table_lock);
+ pgd_clear(pgd);
+ spin_unlock(&init_mm.page_table_lock);
+}
+
+static int split_large_page(pte_t *kpte, unsigned long address, pte_t *pbase)
+{
+ struct page *page = pmd_page(*(pmd_t *)kpte);
+ int i = 0;
+ unsigned long magic;
+ unsigned long section_nr;
+
+ __split_large_page(kpte, address, pbase);
+ __flush_tlb_all();
+
+ magic = (unsigned long) page->lru.next;
+ if (magic == SECTION_INFO) {
+ section_nr = pfn_to_section_nr(page_to_pfn(page));
+ while (i < PTRS_PER_PMD) {
+ page++;
+ i++;
+ get_page_bootmem(section_nr, page, SECTION_INFO);
+ }
+ }
+
+ return 0;
+}
+
+static void vmemmap_pte_remove(pmd_t *pmd, unsigned long addr, unsigned long end)
+{
+ pte_t *pte;
+ unsigned long next;
+
+ pte = pte_offset_kernel(pmd, addr);
+ for (; addr < end; pte++, addr += PAGE_SIZE) {
+ next = (addr + PAGE_SIZE) & PAGE_MASK;
+ if (next > end)
+ next = end;
+
+ if (pte_none(*pte))
+ continue;
+ if (IS_ALIGNED(addr, PAGE_SIZE) &&
+ IS_ALIGNED(end, PAGE_SIZE)) {
+ vmemmap_free_pages(pte_page(*pte), 0);
+ spin_lock(&init_mm.page_table_lock);
+ pte_clear(&init_mm, addr, pte);
+ spin_unlock(&init_mm.page_table_lock);
+ }
+ }
+
+ free_pte_table(pmd);
+ __flush_tlb_all();
+}
+
+static void vmemmap_pmd_remove(pud_t *pud, unsigned long addr, unsigned long end)
+{
+ unsigned long next;
+ pmd_t *pmd;
+
+ pmd = pmd_offset(pud, addr);
+ for (; addr < end; addr = next, pmd++) {
+ next = pmd_addr_end(addr, end);
+ if (pmd_none(*pmd))
+ continue;
+
+ if (cpu_has_pse) {
+ unsigned long pte_base;
+
+ if (IS_ALIGNED(addr, PMD_SIZE) &&
+ IS_ALIGNED(next, PMD_SIZE)) {
+ vmemmap_free_pages(pmd_page(*pmd),
+ get_order(PMD_SIZE));
+ spin_lock(&init_mm.page_table_lock);
+ pmd_clear(pmd);
+ spin_unlock(&init_mm.page_table_lock);
+ continue;
+ }
+
+ /*
+ * We use 2M page, but we need to remove part of them,
+ * so split 2M page to 4K page.
+ */
+ pte_base = get_zeroed_page(GFP_ATOMIC | __GFP_NOTRACK);
+ split_large_page((pte_t *)pmd, addr, (pte_t *)pte_base);
+ __flush_tlb_all();
+
+ spin_lock(&init_mm.page_table_lock);
+ pmd_populate_kernel(&init_mm, pmd, (pte_t *)pte_base);
+ spin_unlock(&init_mm.page_table_lock);
+ }
+
+ vmemmap_pte_remove(pmd, addr, next);
+ }
+
+ free_pmd_table(pud);
+ __flush_tlb_all();
+}
+
+static void vmemmap_pud_remove(pgd_t *pgd, unsigned long addr, unsigned long end)
+{
+ unsigned long next;
+ pud_t *pud;
+
+ pud = pud_offset(pgd, addr);
+ for (; addr < end; addr = next, pud++) {
+ next = pud_addr_end(addr, end);
+ if (pud_none(*pud))
+ continue;
+
+ vmemmap_pmd_remove(pud, addr, next);
+ }
+
+ free_pud_table(pgd);
+ __flush_tlb_all();
+}
+
+void vmemmap_free(struct page *memmap, unsigned long nr_pages)
+{
+ unsigned long addr = (unsigned long)memmap;
+ unsigned long end = (unsigned long)(memmap + nr_pages);
+ unsigned long next;
+
+ for (; addr < end; addr = next) {
+ pgd_t *pgd = pgd_offset_k(addr);
+
+ next = pgd_addr_end(addr, end);
+ if (!pgd_present(*pgd))
+ continue;
+
+ vmemmap_pud_remove(pgd, addr, next);
+ sync_global_pgds(addr, next);
+ }
+}
+#endif
diff --git a/mm/sparse.c b/mm/sparse.c
index fac95f2..3a16d68 100644
--- a/mm/sparse.c
+++ b/mm/sparse.c
@@ -613,12 +613,13 @@ static inline struct page *kmalloc_section_memmap(unsigned long pnum, int nid,
/* This will make the necessary allocations eventually. */
return sparse_mem_map_populate(pnum, nid);
}
-static void __kfree_section_memmap(struct page *memmap, unsigned long nr_pages)
+static void __kfree_section_memmap(struct page *page, unsigned long nr_pages)
{
- return; /* XXX: Not implemented yet */
+ vmemmap_free(page, nr_pages);
}
static void free_map_bootmem(struct page *page, unsigned long nr_pages)
{
+ vmemmap_free(page, nr_pages);
}
#else
static struct page *__kmalloc_section_memmap(unsigned long nr_pages)
--
1.7.6.1
On 2012/11/27 18:00, Wen Congyang wrote:
> From: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
>
> All pages of virtual mapping in removed memory cannot be freed, since some pages
> used as PGD/PUD includes not only removed memory but also other memory. So the
> patch checks whether page can be freed or not.
>
> How to check whether page can be freed or not?
> 1. When removing memory, the page structs of the revmoved memory are filled
> with 0FD.
> 2. All page structs are filled with 0xFD on PT/PMD, PT/PMD can be cleared.
> In this case, the page used as PT/PMD can be freed.
>
> Applying patch, __remove_section() of CONFIG_SPARSEMEM_VMEMMAP is integrated
> into one. So __remove_section() of CONFIG_SPARSEMEM_VMEMMAP is deleted.
>
> Note: vmemmap_kfree() and vmemmap_free_bootmem() are not implemented for ia64,
> ppc, s390, and sparc.
>
> CC: David Rientjes <rientjes@google.com>
> CC: Jiang Liu <liuj97@gmail.com>
> CC: Len Brown <len.brown@intel.com>
> CC: Christoph Lameter <cl@linux.com>
> Cc: Minchan Kim <minchan.kim@gmail.com>
> CC: Andrew Morton <akpm@linux-foundation.org>
> CC: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
> Signed-off-by: Jianguo Wu <wujianguo@huawei.com>
> Signed-off-by: Wen Congyang <wency@cn.fujitsu.com>
> ---
> arch/ia64/mm/discontig.c | 8 ++++
> arch/powerpc/mm/init_64.c | 8 ++++
> arch/s390/mm/vmem.c | 8 ++++
> arch/sparc/mm/init_64.c | 8 ++++
> arch/x86/mm/init_64.c | 119 ++++++++++++++++++++++++++++++++++++++++++++++
> include/linux/mm.h | 2 +
> mm/memory_hotplug.c | 17 +------
> mm/sparse.c | 19 ++++----
> 8 files changed, 165 insertions(+), 24 deletions(-)
>
> diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
> index 33943db..0d23b69 100644
> --- a/arch/ia64/mm/discontig.c
> +++ b/arch/ia64/mm/discontig.c
> @@ -823,6 +823,14 @@ int __meminit vmemmap_populate(struct page *start_page,
> return vmemmap_populate_basepages(start_page, size, node);
> }
>
> +void vmemmap_kfree(struct page *memmap, unsigned long nr_pages)
> +{
> +}
> +
> +void vmemmap_free_bootmem(struct page *memmap, unsigned long nr_pages)
> +{
> +}
> +
> void register_page_bootmem_memmap(unsigned long section_nr,
> struct page *start_page, unsigned long size)
> {
> diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c
> index 6466440..df7d155 100644
> --- a/arch/powerpc/mm/init_64.c
> +++ b/arch/powerpc/mm/init_64.c
> @@ -298,6 +298,14 @@ int __meminit vmemmap_populate(struct page *start_page,
> return 0;
> }
>
> +void vmemmap_kfree(struct page *memmap, unsigned long nr_pages)
> +{
> +}
> +
> +void vmemmap_free_bootmem(struct page *memmap, unsigned long nr_pages)
> +{
> +}
> +
> void register_page_bootmem_memmap(unsigned long section_nr,
> struct page *start_page, unsigned long size)
> {
> diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c
> index 4f4803a..ab69c34 100644
> --- a/arch/s390/mm/vmem.c
> +++ b/arch/s390/mm/vmem.c
> @@ -236,6 +236,14 @@ out:
> return ret;
> }
>
> +void vmemmap_kfree(struct page *memmap, unsigned long nr_pages)
> +{
> +}
> +
> +void vmemmap_free_bootmem(struct page *memmap, unsigned long nr_pages)
> +{
> +}
> +
> void register_page_bootmem_memmap(unsigned long section_nr,
> struct page *start_page, unsigned long size)
> {
> diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
> index 75a984b..546855d 100644
> --- a/arch/sparc/mm/init_64.c
> +++ b/arch/sparc/mm/init_64.c
> @@ -2232,6 +2232,14 @@ void __meminit vmemmap_populate_print_last(void)
> }
> }
>
> +void vmemmap_kfree(struct page *memmap, unsigned long nr_pages)
> +{
> +}
> +
> +void vmemmap_free_bootmem(struct page *memmap, unsigned long nr_pages)
> +{
> +}
> +
> void register_page_bootmem_memmap(unsigned long section_nr,
> struct page *start_page, unsigned long size)
> {
> diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
> index 795dae3..e85626d 100644
> --- a/arch/x86/mm/init_64.c
> +++ b/arch/x86/mm/init_64.c
> @@ -998,6 +998,125 @@ vmemmap_populate(struct page *start_page, unsigned long size, int node)
> return 0;
> }
>
> +#define PAGE_INUSE 0xFD
> +
> +unsigned long find_and_clear_pte_page(unsigned long addr, unsigned long end,
> + struct page **pp, int *page_size)
> +{
> + pgd_t *pgd;
> + pud_t *pud;
> + pmd_t *pmd;
> + pte_t *pte = NULL;
> + void *page_addr;
> + unsigned long next;
> +
> + *pp = NULL;
> +
> + pgd = pgd_offset_k(addr);
> + if (pgd_none(*pgd))
> + return pgd_addr_end(addr, end);
> +
> + pud = pud_offset(pgd, addr);
> + if (pud_none(*pud))
> + return pud_addr_end(addr, end);
> +
> + if (!cpu_has_pse) {
> + next = (addr + PAGE_SIZE) & PAGE_MASK;
> + pmd = pmd_offset(pud, addr);
> + if (pmd_none(*pmd))
> + return next;
> +
> + pte = pte_offset_kernel(pmd, addr);
> + if (pte_none(*pte))
> + return next;
> +
> + *page_size = PAGE_SIZE;
> + *pp = pte_page(*pte);
> + } else {
> + next = pmd_addr_end(addr, end);
> +
> + pmd = pmd_offset(pud, addr);
> + if (pmd_none(*pmd))
> + return next;
> +
> + *page_size = PMD_SIZE;
> + *pp = pmd_page(*pmd);
> + }
> +
> + /*
> + * Removed page structs are filled with 0xFD.
> + */
> + memset((void *)addr, PAGE_INUSE, next - addr);
> +
> + page_addr = page_address(*pp);
> +
> + /*
> + * Check the page is filled with 0xFD or not.
> + * memchr_inv() returns the address. In this case, we cannot
> + * clear PTE/PUD entry, since the page is used by other.
> + * So we cannot also free the page.
> + *
> + * memchr_inv() returns NULL. In this case, we can clear
> + * PTE/PUD entry, since the page is not used by other.
> + * So we can also free the page.
> + */
> + if (memchr_inv(page_addr, PAGE_INUSE, *page_size)) {
> + *pp = NULL;
> + return next;
> + }
> +
> + if (!cpu_has_pse)
> + pte_clear(&init_mm, addr, pte);
> + else
> + pmd_clear(pmd);
> +
> + return next;
> +}
> +
> +void vmemmap_kfree(struct page *memmap, unsigned long nr_pages)
> +{
> + unsigned long addr = (unsigned long)memmap;
> + unsigned long end = (unsigned long)(memmap + nr_pages);
> + unsigned long next;
> + struct page *page;
> + int page_size;
> +
> + for (; addr < end; addr = next) {
> + page = NULL;
> + page_size = 0;
> + next = find_and_clear_pte_page(addr, end, &page, &page_size);
> + if (!page)
> + continue;
> +
> + free_pages((unsigned long)page_address(page),
> + get_order(page_size));
> + __flush_tlb_one(addr);
> + }
> +}
> +
> +void vmemmap_free_bootmem(struct page *memmap, unsigned long nr_pages)
> +{
> + unsigned long addr = (unsigned long)memmap;
> + unsigned long end = (unsigned long)(memmap + nr_pages);
> + unsigned long next;
> + struct page *page;
> + int page_size;
> + unsigned long magic;
> +
> + for (; addr < end; addr = next) {
> + page = NULL;
> + page_size = 0;
> + next = find_and_clear_pte_page(addr, end, &page, &page_size);
> + if (!page)
> + continue;
> +
> + magic = (unsigned long) page->lru.next;
> + if (magic == SECTION_INFO)
> + put_page_bootmem(page);
> + flush_tlb_kernel_range(addr, end);
> + }
> +}
> +
> void register_page_bootmem_memmap(unsigned long section_nr,
> struct page *start_page, unsigned long size)
> {
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index 5657670..94d5ccd 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -1642,6 +1642,8 @@ int vmemmap_populate(struct page *start_page, unsigned long pages, int node);
> void vmemmap_populate_print_last(void);
> void register_page_bootmem_memmap(unsigned long section_nr, struct page *map,
> unsigned long size);
> +void vmemmap_kfree(struct page *memmpa, unsigned long nr_pages);
> +void vmemmap_free_bootmem(struct page *memmpa, unsigned long nr_pages);
>
> enum mf_flags {
> MF_COUNT_INCREASED = 1 << 0,
> diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
> index ccc11b6..7797e91 100644
> --- a/mm/memory_hotplug.c
> +++ b/mm/memory_hotplug.c
> @@ -301,19 +301,6 @@ static int __meminit __add_section(int nid, struct zone *zone,
> return register_new_memory(nid, __pfn_to_section(phys_start_pfn));
> }
>
> -#ifdef CONFIG_SPARSEMEM_VMEMMAP
> -static int __remove_section(struct zone *zone, struct mem_section *ms)
> -{
> - int ret = -EINVAL;
> -
> - if (!valid_section(ms))
> - return ret;
> -
> - ret = unregister_memory_section(ms);
> -
> - return ret;
> -}
> -#else
> static int __remove_section(struct zone *zone, struct mem_section *ms)
> {
> unsigned long flags;
> @@ -330,9 +317,9 @@ static int __remove_section(struct zone *zone, struct mem_section *ms)
> pgdat_resize_lock(pgdat, &flags);
> sparse_remove_one_section(zone, ms);
> pgdat_resize_unlock(pgdat, &flags);
> - return 0;
> +
> + return ret;
> }
> -#endif
>
> /*
> * Reasonably generic function for adding memory. It is
> diff --git a/mm/sparse.c b/mm/sparse.c
> index fac95f2..c723bc2 100644
> --- a/mm/sparse.c
> +++ b/mm/sparse.c
> @@ -613,12 +613,13 @@ static inline struct page *kmalloc_section_memmap(unsigned long pnum, int nid,
> /* This will make the necessary allocations eventually. */
> return sparse_mem_map_populate(pnum, nid);
> }
> -static void __kfree_section_memmap(struct page *memmap, unsigned long nr_pages)
> +static void __kfree_section_memmap(struct page *page, unsigned long nr_pages)
> {
> - return; /* XXX: Not implemented yet */
> + vmemmap_kfree(page, nr_pages);
> }
> -static void free_map_bootmem(struct page *page, unsigned long nr_pages)
> +static void free_map_bootmem(struct page *page)
> {
> + vmemmap_free_bootmem(page, PAGES_PER_SECTION);
> }
> #else
> static struct page *__kmalloc_section_memmap(unsigned long nr_pages)
> @@ -658,10 +659,14 @@ static void __kfree_section_memmap(struct page *memmap, unsigned long nr_pages)
> get_order(sizeof(struct page) * nr_pages));
> }
>
> -static void free_map_bootmem(struct page *page, unsigned long nr_pages)
> +static void free_map_bootmem(struct page *page)
> {
> unsigned long maps_section_nr, removing_section_nr, i;
> unsigned long magic;
> + unsigned long nr_pages;
> +
> + nr_pages = PAGE_ALIGN(PAGES_PER_SECTION * sizeof(struct page))
> + >> PAGE_SHIFT;
>
> for (i = 0; i < nr_pages; i++, page++) {
> magic = (unsigned long) page->lru.next;
> @@ -688,7 +693,6 @@ static void free_map_bootmem(struct page *page, unsigned long nr_pages)
> static void free_section_usemap(struct page *memmap, unsigned long *usemap)
> {
> struct page *usemap_page;
> - unsigned long nr_pages;
>
> if (!usemap)
> return;
> @@ -713,10 +717,7 @@ static void free_section_usemap(struct page *memmap, unsigned long *usemap)
> struct page *memmap_page;
> memmap_page = virt_to_page(memmap);
>
> - nr_pages = PAGE_ALIGN(PAGES_PER_SECTION * sizeof(struct page))
> - >> PAGE_SHIFT;
> -
> - free_map_bootmem(memmap_page, nr_pages);
> + free_map_bootmem(memmap_page);
> }
> }
>
^ permalink raw reply related
* fsl spi questions & patch
From: Frans Meulenbroeks @ 2012-11-28 9:39 UTC (permalink / raw)
To: linuxppc-dev
[-- Attachment #1.1: Type: text/plain, Size: 1298 bytes --]
Hi,
I've been playing with spi on mpc8313e and have some things on
spi-fsl-spi.c:
Is QE useful on 8313?
I've tried it (using cpu-qe in my dts file) and see in the boot log that it
is used, but I do not really see any effect when it comes to performance or
cpu usage.
Furthermore:
In the fsl_spi_cpu_irq there is a line:
/* Clear the events */
mpc8xxx_spi_write_reg(®_base->event, events);
Is this really useful? The 8313 book says NE is cleared upon reading and NF
is cleared upon writing.
(this might apply to fsl_spi_cpm_irq too, I do not have info on cpm.
Next, I noticed some spacing between two spi words being sent. It seems the
transmit buffer is not filled when possible, but only when a word is
received (and the previous word is transmitted). By modifying the code
somewhat I was able to roughly double the effective transfer rate in my
test setup (8 Mhz spi clock, 32 bit transfers). Attached is my changed
code. Feedback on it is appreciated.
This patch also eliminated the spinning until TX is done. (actually I am
not sure if this will happen, I would expect NE and NF to be raised roughly
at the same time.
Thanks for any feedback!
Frans Meulenbroeks
PS:it would probably be nice if in board setup one could also set the
(default) value of bits-per-word.
[-- Attachment #1.2: Type: text/html, Size: 1396 bytes --]
[-- Attachment #2: 0001-spi-fsl-spi.c-use-NF-interrupt-instead-of-NE.patch --]
[-- Type: application/octet-stream, Size: 2799 bytes --]
From fe783ac0fe3d2ea4814d282f436c9670f892bda2 Mon Sep 17 00:00:00 2001
From: Frans Meulenbroeks <fransmeulenbroeks@gmail.com>
Date: Wed, 28 Nov 2012 09:42:51 +0100
Subject: [PATCH] spi-fsl-spi.c: use NF interrupt instead of NE
When using NE (not empty) interrupt we get the first interrupt
when the first word is received (and the first one send as
sending and receiving happens simultaneously and sychronously).
In the case of multi-word transmits there is a short delay to
reload the transmit buffer (on my test system 1.5 us).
However it is possible to load the next byte before as there
is a one word hold register.
By using NF (not full) to trigger the interrupt we fill the
register as soon as possible, improving the transmission speed.
(about doubling it with an 8Mhz SPI clock and 32 bit transfers
on MPC8313e)
I also moved the disabling of the interrupt into the _irq
handler (before calling complete() ) instead of after
wait_for_completion().
On my system the NF interrupt kept being raised; this fixed that.
Signed-off-by: Frans Meulenbroeks <fransmeulenbroeks@gmail.com>
---
drivers/spi/spi-fsl-spi.c | 27 ++++++++++-----------------
1 file changed, 10 insertions(+), 17 deletions(-)
diff --git a/drivers/spi/spi-fsl-spi.c b/drivers/spi/spi-fsl-spi.c
index 6a62934..1f81ccd 100644
--- a/drivers/spi/spi-fsl-spi.c
+++ b/drivers/spi/spi-fsl-spi.c
@@ -410,7 +410,7 @@ static int fsl_spi_cpu_bufs(struct mpc8xxx_spi *mspi,
mspi->count = len;
/* enable rx ints */
- mpc8xxx_spi_write_reg(®_base->mask, SPIM_NE);
+ mpc8xxx_spi_write_reg(®_base->mask, SPIM_NF);
/* transmit word */
word = mspi->get_tx(mspi);
@@ -460,9 +460,6 @@ static int fsl_spi_bufs(struct spi_device *spi, struct spi_transfer *t,
wait_for_completion(&mpc8xxx_spi->done);
- /* disable rx ints */
- mpc8xxx_spi_write_reg(®_base->mask, 0);
-
if (mpc8xxx_spi->flags & SPI_CPM_MODE)
fsl_spi_cpm_bufs_complete(mpc8xxx_spi);
@@ -604,23 +601,19 @@ static void fsl_spi_cpu_irq(struct mpc8xxx_spi *mspi, u32 events)
mspi->get_rx(rx_data, mspi);
}
- if ((events & SPIE_NF) == 0)
- /* spin until TX is done */
- while (((events =
- mpc8xxx_spi_read_reg(®_base->event)) &
- SPIE_NF) == 0)
- cpu_relax();
-
/* Clear the events */
mpc8xxx_spi_write_reg(®_base->event, events);
- mspi->count -= 1;
- if (mspi->count) {
- u32 word = mspi->get_tx(mspi);
+ if ((events & SPIE_NF) && (mspi->count > 1))
+ { mspi->count -= 1;
+ if (mspi->count) {
+ u32 word = mspi->get_tx(mspi);
- mpc8xxx_spi_write_reg(®_base->transmit, word);
- } else {
- complete(&mspi->done);
+ mpc8xxx_spi_write_reg(®_base->transmit, word);
+ } else {
+ mpc8xxx_spi_write_reg(®_base->mask, 0); /* disable interrupts */
+ complete(&mspi->done);
+ }
}
}
--
1.7.9.5
^ permalink raw reply related
* Re: 3.7-rc7: BUG: MAX_STACK_TRACE_ENTRIES too low!
From: Li Zhong @ 2012-11-28 8:41 UTC (permalink / raw)
To: Christian Kujau; +Cc: paulus, linuxppc-dev, LKML
In-Reply-To: <alpine.DEB.2.01.1211271916160.7378@trent.utfs.org>
On Tue, 2012-11-27 at 19:22 -0800, Christian Kujau wrote:
> On Tue, 27 Nov 2012 at 19:06, Christian Kujau wrote:
> > the same thing[0] happened again in 3.7-rc7, after ~20h uptime:
>
> I found the following on patchwork, but this seems to deal with powerpc64
> only, while this PowerBook G4 of mine is powerpc32:
>
> http://patchwork.ozlabs.org/patch/193414/
>
> It looks related, but then again, I fail to parse assember...
Hi Christian,
Would you please help to try the following fix? I don't have a powerpc32
machine for test...
Thanks, Zhong
==============================================
diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S
index 9499385..cadebfd 100644
--- a/arch/powerpc/kernel/entry_32.S
+++ b/arch/powerpc/kernel/entry_32.S
@@ -439,6 +439,8 @@ ret_from_fork:
ret_from_kernel_thread:
REST_NVGPRS(r1)
bl schedule_tail
+ li r3,0
+ stw r3,0(r1)
mtlr r14
mr r3,r15
PPC440EP_ERR42
==============================================
>
> Christian.
>
> > [40007.339487] [sched_delayed] sched: RT throttling activated
> > [69731.388717] BUG: MAX_STACK_TRACE_ENTRIES too low!
> > [69731.390371] turning off the locking correctness validator.
> > [69731.391942] Call Trace:
> > [69731.393525] [c9a61c10] [c0009064] show_stack+0x70/0x1bc (unreliable)
> > [69731.395152] [c9a61c50] [c0077460] save_trace+0xfc/0x114
> > [69731.396735] [c9a61c60] [c007be20] __lock_acquire+0x1568/0x19b8
> > [69731.398296] [c9a61d00] [c007c2c0] lock_acquire+0x50/0x70
> > [69731.399857] [c9a61d20] [c0550e28] _raw_spin_lock_irq+0x5c/0x78
> > [69731.401419] [c9a61d40] [c054fb58] __schedule+0xd8/0x534
> > [69731.402972] [c9a61da0] [c0550094] _cond_resched+0x50/0x68
> > [69731.404527] [c9a61db0] [c0479908] dst_gc_task+0xbc/0x258
> > [69731.406070] [c9a61e40] [c004eeb8] process_one_work+0x1f4/0x49c
> > [69731.407585] [c9a61e80] [c004f644] worker_thread+0x14c/0x400
> > [69731.409075] [c9a61eb0] [c0057634] kthread+0xbc/0xc0
> > [69731.410521] [c9a61f40] [c0011ad4] ret_from_kernel_thread+0x5c/0x64
> > [...repeated 54 times...]
> >
> > Anyone knows what this is about?
> >
> > Thanks,
> > Christian.
> >
> > [0] http://lkml.indiana.edu/hypermail/linux/kernel/1211.0/03025.html
^ permalink raw reply related
* [PATCH] vfio powerpc: implemented IOMMU driver for VFIO
From: Alexey Kardashevskiy @ 2012-11-28 7:21 UTC (permalink / raw)
To: Alex Williamson
Cc: kvm, Alexey Kardashevskiy, linux-kernel, Paul Mackerras,
linuxppc-dev, David Gibson
In-Reply-To: <1353992877.1809.156.camel@bling.home>
VFIO implements platform independent stuff such as
a PCI driver, BAR access (via read/write on a file descriptor
or direct mapping when possible) and IRQ signaling.
The platform dependent part includes IOMMU initialization
and handling. This patch implements an IOMMU driver for VFIO
which does mapping/unmapping pages for the guest IO and
provides information about DMA window (required by a POWERPC
guest).
The counterpart in QEMU is required to support this functionality.
Cc: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
---
drivers/vfio/Kconfig | 6 +
drivers/vfio/Makefile | 1 +
drivers/vfio/vfio_iommu_spapr_tce.c | 332 +++++++++++++++++++++++++++++++++++
include/linux/vfio.h | 33 ++++
4 files changed, 372 insertions(+)
create mode 100644 drivers/vfio/vfio_iommu_spapr_tce.c
diff --git a/drivers/vfio/Kconfig b/drivers/vfio/Kconfig
index 7cd5dec..b464687 100644
--- a/drivers/vfio/Kconfig
+++ b/drivers/vfio/Kconfig
@@ -3,10 +3,16 @@ config VFIO_IOMMU_TYPE1
depends on VFIO
default n
+config VFIO_IOMMU_SPAPR_TCE
+ tristate
+ depends on VFIO && SPAPR_TCE_IOMMU
+ default n
+
menuconfig VFIO
tristate "VFIO Non-Privileged userspace driver framework"
depends on IOMMU_API
select VFIO_IOMMU_TYPE1 if X86
+ select VFIO_IOMMU_SPAPR_TCE if PPC_POWERNV
help
VFIO provides a framework for secure userspace device drivers.
See Documentation/vfio.txt for more details.
diff --git a/drivers/vfio/Makefile b/drivers/vfio/Makefile
index 2398d4a..72bfabc 100644
--- a/drivers/vfio/Makefile
+++ b/drivers/vfio/Makefile
@@ -1,3 +1,4 @@
obj-$(CONFIG_VFIO) += vfio.o
obj-$(CONFIG_VFIO_IOMMU_TYPE1) += vfio_iommu_type1.o
+obj-$(CONFIG_VFIO_IOMMU_SPAPR_TCE) += vfio_iommu_spapr_tce.o
obj-$(CONFIG_VFIO_PCI) += pci/
diff --git a/drivers/vfio/vfio_iommu_spapr_tce.c b/drivers/vfio/vfio_iommu_spapr_tce.c
new file mode 100644
index 0000000..b98770e
--- /dev/null
+++ b/drivers/vfio/vfio_iommu_spapr_tce.c
@@ -0,0 +1,332 @@
+/*
+ * VFIO: IOMMU DMA mapping support for TCE on POWER
+ *
+ * Copyright (C) 2012 IBM Corp. All rights reserved.
+ * Author: Alexey Kardashevskiy <aik@ozlabs.ru>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Derived from original vfio_iommu_type1.c:
+ * Copyright (C) 2012 Red Hat, Inc. All rights reserved.
+ * Author: Alex Williamson <alex.williamson@redhat.com>
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+#include <linux/err.h>
+#include <linux/vfio.h>
+#include <asm/iommu.h>
+
+#define DRIVER_VERSION "0.1"
+#define DRIVER_AUTHOR "aik@ozlabs.ru"
+#define DRIVER_DESC "VFIO IOMMU SPAPR TCE"
+
+static void tce_iommu_detach_group(void *iommu_data,
+ struct iommu_group *iommu_group);
+
+/*
+ * VFIO IOMMU fd for SPAPR_TCE IOMMU implementation
+ */
+
+/*
+ * This code handles mapping and unmapping of user data buffers
+ * into DMA'ble space using the IOMMU
+ */
+
+#define NPAGE_TO_SIZE(npage) ((size_t)(npage) << PAGE_SHIFT)
+
+struct vwork {
+ struct mm_struct *mm;
+ long npage;
+ struct work_struct work;
+};
+
+/* delayed decrement/increment for locked_vm */
+static void lock_acct_bg(struct work_struct *work)
+{
+ struct vwork *vwork = container_of(work, struct vwork, work);
+ struct mm_struct *mm;
+
+ mm = vwork->mm;
+ down_write(&mm->mmap_sem);
+ mm->locked_vm += vwork->npage;
+ up_write(&mm->mmap_sem);
+ mmput(mm);
+ kfree(vwork);
+}
+
+static void lock_acct(long npage)
+{
+ struct vwork *vwork;
+ struct mm_struct *mm;
+
+ if (!current->mm)
+ return; /* process exited */
+
+ if (down_write_trylock(¤t->mm->mmap_sem)) {
+ current->mm->locked_vm += npage;
+ up_write(¤t->mm->mmap_sem);
+ return;
+ }
+
+ /*
+ * Couldn't get mmap_sem lock, so must setup to update
+ * mm->locked_vm later. If locked_vm were atomic, we
+ * wouldn't need this silliness
+ */
+ vwork = kmalloc(sizeof(struct vwork), GFP_KERNEL);
+ if (!vwork)
+ return;
+ mm = get_task_mm(current);
+ if (!mm) {
+ kfree(vwork);
+ return;
+ }
+ INIT_WORK(&vwork->work, lock_acct_bg);
+ vwork->mm = mm;
+ vwork->npage = npage;
+ schedule_work(&vwork->work);
+}
+
+/*
+ * The container descriptor supports only a single group per container.
+ * Required by the API as the container is not supplied with the IOMMU group
+ * at the moment of initialization.
+ */
+struct tce_container {
+ struct mutex lock;
+ struct iommu_table *tbl;
+};
+
+static void *tce_iommu_open(unsigned long arg)
+{
+ struct tce_container *container;
+
+ if (arg != VFIO_SPAPR_TCE_IOMMU) {
+ pr_err("tce_vfio: Wrong IOMMU type\n");
+ return ERR_PTR(-EINVAL);
+ }
+
+ container = kzalloc(sizeof(*container), GFP_KERNEL);
+ if (!container)
+ return ERR_PTR(-ENOMEM);
+
+ mutex_init(&container->lock);
+
+ return container;
+}
+
+static void tce_iommu_release(void *iommu_data)
+{
+ struct tce_container *container = iommu_data;
+
+ WARN_ON(container->tbl && !container->tbl->it_group);
+ if (container->tbl && container->tbl->it_group)
+ tce_iommu_detach_group(iommu_data, container->tbl->it_group);
+
+ mutex_destroy(&container->lock);
+
+ kfree(container);
+}
+
+static long tce_iommu_ioctl(void *iommu_data,
+ unsigned int cmd, unsigned long arg)
+{
+ struct tce_container *container = iommu_data;
+ unsigned long minsz;
+ long ret;
+
+ switch (cmd) {
+ case VFIO_CHECK_EXTENSION: {
+ return (arg == VFIO_SPAPR_TCE_IOMMU) ? 1 : 0;
+ }
+ case VFIO_IOMMU_SPAPR_TCE_GET_INFO: {
+ struct vfio_iommu_spapr_tce_info info;
+ struct iommu_table *tbl = container->tbl;
+
+ if (WARN_ON(!tbl))
+ return -ENXIO;
+
+ minsz = offsetofend(struct vfio_iommu_spapr_tce_info,
+ dma64_window_size);
+
+ if (copy_from_user(&info, (void __user *)arg, minsz))
+ return -EFAULT;
+
+ if (info.argsz < minsz)
+ return -EINVAL;
+
+ info.dma32_window_start = tbl->it_offset << IOMMU_PAGE_SHIFT;
+ info.dma32_window_size = tbl->it_size << IOMMU_PAGE_SHIFT;
+ info.dma64_window_start = 0;
+ info.dma64_window_size = 0;
+ info.flags = 0;
+
+ if (copy_to_user((void __user *)arg, &info, minsz))
+ return -EFAULT;
+
+ return 0;
+ }
+ case VFIO_IOMMU_MAP_DMA: {
+ vfio_iommu_spapr_tce_dma_map param;
+ struct iommu_table *tbl = container->tbl;
+ enum dma_data_direction direction;
+ unsigned long locked, lock_limit;
+
+ if (WARN_ON(!tbl))
+ return -ENXIO;
+
+ minsz = offsetofend(vfio_iommu_spapr_tce_dma_map, size);
+
+ if (copy_from_user(¶m, (void __user *)arg, minsz))
+ return -EFAULT;
+
+ if (param.argsz < minsz)
+ return -EINVAL;
+
+ if ((param.flags & VFIO_DMA_MAP_FLAG_READ) &&
+ (param.flags & VFIO_DMA_MAP_FLAG_WRITE))
+ direction = DMA_BIDIRECTIONAL;
+ else if (param.flags & VFIO_DMA_MAP_FLAG_READ)
+ direction = DMA_TO_DEVICE;
+ else if (param.flags & VFIO_DMA_MAP_FLAG_WRITE)
+ direction = DMA_FROM_DEVICE;
+ else
+ return -EINVAL;
+
+ if ((param.size & ~IOMMU_PAGE_MASK) ||
+ (param.iova & ~IOMMU_PAGE_MASK) ||
+ (param.vaddr & ~IOMMU_PAGE_MASK))
+ return -EINVAL;
+
+ /* Account for locked pages */
+ locked = current->mm->locked_vm +
+ (param.size >> IOMMU_PAGE_SHIFT);
+ lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
+ if (locked > lock_limit && !capable(CAP_IPC_LOCK)) {
+ pr_warn("RLIMIT_MEMLOCK (%ld) exceeded\n",
+ rlimit(RLIMIT_MEMLOCK));
+ return -ENOMEM;
+ }
+
+ ret = iommu_put_tces(tbl, param.iova >> IOMMU_PAGE_SHIFT,
+ param.vaddr, direction,
+ param.size >> IOMMU_PAGE_SHIFT);
+ if (ret > 0)
+ lock_acct(ret);
+
+ return ret;
+ }
+ case VFIO_IOMMU_UNMAP_DMA: {
+ vfio_iommu_spapr_tce_dma_unmap param;
+ struct iommu_table *tbl = container->tbl;
+
+ if (WARN_ON(!tbl))
+ return -ENXIO;
+
+ minsz = offsetofend(vfio_iommu_spapr_tce_dma_unmap, size);
+
+ if (copy_from_user(¶m, (void __user *)arg, minsz))
+ return -EFAULT;
+
+ if (param.argsz < minsz)
+ return -EINVAL;
+
+ if ((param.size & ~IOMMU_PAGE_MASK) ||
+ (param.iova & ~IOMMU_PAGE_MASK))
+ return -EINVAL;
+
+ ret = iommu_clear_tces(tbl, param.iova >> IOMMU_PAGE_SHIFT,
+ param.size >> IOMMU_PAGE_SHIFT);
+ if (ret > 0)
+ lock_acct(-ret);
+
+ return ret;
+ }
+ default:
+ pr_warn("tce_vfio: unexpected cmd %x\n", cmd);
+ }
+
+ return -ENOTTY;
+}
+
+static int tce_iommu_attach_group(void *iommu_data,
+ struct iommu_group *iommu_group)
+{
+ struct tce_container *container = iommu_data;
+ struct iommu_table *tbl = iommu_group_get_iommudata(iommu_group);
+
+ BUG_ON(!tbl);
+ mutex_lock(&container->lock);
+ pr_debug("tce_vfio: Attaching group #%u to iommu %p\n",
+ iommu_group_id(iommu_group), iommu_group);
+ if (container->tbl) {
+ pr_warn("tce_vfio: Only one group per IOMMU container is allowed, existing id=%d, attaching id=%d\n",
+ iommu_group_id(container->tbl->it_group),
+ iommu_group_id(iommu_group));
+ mutex_unlock(&container->lock);
+ return -EBUSY;
+ }
+
+ container->tbl = tbl;
+ iommu_clear_tces(tbl, tbl->it_offset, tbl->it_size);
+ mutex_unlock(&container->lock);
+
+ return 0;
+}
+
+static void tce_iommu_detach_group(void *iommu_data,
+ struct iommu_group *iommu_group)
+{
+ struct tce_container *container = iommu_data;
+ struct iommu_table *tbl = iommu_group_get_iommudata(iommu_group);
+
+ BUG_ON(!tbl);
+ mutex_lock(&container->lock);
+ if (tbl != container->tbl) {
+ pr_warn("tce_vfio: detaching group #%u, expected group is #%u\n",
+ iommu_group_id(iommu_group),
+ iommu_group_id(tbl->it_group));
+ } else {
+
+ pr_debug("tce_vfio: detaching group #%u from iommu %p\n",
+ iommu_group_id(iommu_group), iommu_group);
+
+ iommu_clear_tces(tbl, tbl->it_offset, tbl->it_size);
+ container->tbl = NULL;
+ }
+ mutex_unlock(&container->lock);
+}
+
+const struct vfio_iommu_driver_ops tce_iommu_driver_ops = {
+ .name = "iommu-vfio-powerpc",
+ .owner = THIS_MODULE,
+ .open = tce_iommu_open,
+ .release = tce_iommu_release,
+ .ioctl = tce_iommu_ioctl,
+ .attach_group = tce_iommu_attach_group,
+ .detach_group = tce_iommu_detach_group,
+};
+
+static int __init tce_iommu_init(void)
+{
+ return vfio_register_iommu_driver(&tce_iommu_driver_ops);
+}
+
+static void __exit tce_iommu_cleanup(void)
+{
+ vfio_unregister_iommu_driver(&tce_iommu_driver_ops);
+}
+
+module_init(tce_iommu_init);
+module_exit(tce_iommu_cleanup);
+
+MODULE_VERSION(DRIVER_VERSION);
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR(DRIVER_AUTHOR);
+MODULE_DESCRIPTION(DRIVER_DESC);
+
diff --git a/include/linux/vfio.h b/include/linux/vfio.h
index 0a4f180..820af1e 100644
--- a/include/linux/vfio.h
+++ b/include/linux/vfio.h
@@ -99,6 +99,7 @@ extern void vfio_unregister_iommu_driver(
/* Extensions */
#define VFIO_TYPE1_IOMMU 1
+#define VFIO_SPAPR_TCE_IOMMU 2
/*
* The IOCTL interface is designed for extensibility by embedding the
@@ -442,4 +443,36 @@ struct vfio_iommu_type1_dma_unmap {
#define VFIO_IOMMU_UNMAP_DMA _IO(VFIO_TYPE, VFIO_BASE + 14)
+/* -------- Additional API for SPAPR TCE (Server POWERPC) IOMMU -------- */
+
+/*
+ * The SPAPR TCE info struct provides the information about the PCI bus
+ * address ranges available for DMA, these values are programmed into
+ * the hardware so the guest has to know that information.
+ *
+ * Pages within 32 bit window should be explicitely mapped/unmapped via ioctls.
+ * 64 bit window (not supported at the moment for the guest) is supposed to
+ * be mapped completely to the guest memory so the devices capable of 64bit
+ * DMA will not have to use map/unmap ioctls.
+ *
+ * The IOMMU page size is always 4K.
+ */
+
+struct vfio_iommu_spapr_tce_info {
+ __u32 argsz;
+ __u32 flags; /* reserved for future use */
+ __u32 dma32_window_start; /* 32 bit window start (bytes) */
+ __u32 dma32_window_size; /* 32 bit window size (bytes) */
+ __u64 dma64_window_start; /* 64 bit window start (bytes) */
+ __u64 dma64_window_size; /* 64 bit window size (bytes) */
+};
+
+#define VFIO_IOMMU_SPAPR_TCE_GET_INFO _IO(VFIO_TYPE, VFIO_BASE + 12)
+
+/* Reuse type1 map/unmap structs as they are the same at the moment */
+typedef struct vfio_iommu_type1_dma_map vfio_iommu_spapr_tce_dma_map;
+typedef struct vfio_iommu_type1_dma_unmap vfio_iommu_spapr_tce_dma_unmap;
+
+/* ***************************************************************** */
+
#endif /* VFIO_H */
--
1.7.10.4
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox