x86/iommu: PVH Dom0 workarounds for missing RMRR/IRSV entries
cover x86/iommu: PVH Dom0 workarounds for missing RMRR/IRSV entries
Hello,
The following series implement a workaround for missing RMRR/IRSV entries for a PVH Dom0. It's based on the iommu_inclusive_mapping VTd option, which is generalized to be used by both VTd and AMD-Vi hardware.
The PVH workaround identity maps all regions marked as reserved in the memory map.
The series can be found at:
git://xenbits.xen.org/people/royger/xen.git iommu_inclusive_v1
Thanks, Roger.
Roger Pau Monne (4):
iommu: remove unneeded return from iommu_hwdom_init
iommu: generalize iommu_inclusive_mapping
x86/iommu: reorder conditions used in the inclusive iommu mappings x86/iommu: add PVH support to the inclusive options
docs/misc/xen-command-line.markdown | 20 +++++ xen/drivers/passthrough/arm/iommu.c | 4 + xen/drivers/passthrough/iommu.c | 8 +- xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 -- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +-------------- xen/drivers/passthrough/x86/iommu.c | 117 ++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 151 insertions(+), 74 deletions(-)
Signature · 7 lines
2.18.0 _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
PATCH 1/4 iommu: remove unneeded return from iommu_hwdom_init
---
Cc: Jan Beulich <[email protected]>
--- xen/drivers/passthrough/iommu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
xen/drivers/passthrough/iommu.c +1 -1
@@ -207,7 +207,7 @@ void __hwdom_init iommu_hwdom_init(struct domain *d) d->domain_id, rc); } -return hd->platform_ops->hwdom_init (d); + hd->platform_ops->hwdom_init (d); }void iommu_teardown (struct domain *d)
Signature · 7 lines
2.18.0 _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 1/4] iommu: remove unneeded return from iommu_hwdom_init
Quoted 11 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]] On Behalf Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Jan Beulich <[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 1/4] iommu: remove unneeded return fromiommu_hwdom_initSigned-off-by Roger Pau Monné <[email protected]>
Yeah, I noticed this too...
Quoted 25 lines
---
Cc: Jan Beulich <[email protected]>--- xen/drivers/passthrough/iommu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)xen/drivers/passthrough/iommu.c +1
-1@@ -207,7 +207,7 @@ void __hwdom_init iommu_hwdom_init(struct domain *d) d->domain_id, rc); } -return hd->platform_ops->hwdom_init (d); + hd->platform_ops->hwdom_init (d); }void
iommu_teardown(struct domain *d)
--
2.18.0
Quoted 4 lines
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
PATCH 2/4 iommu: generalize iommu_inclusive_mapping
In reply to Roger Pau MonneIntroduce a new iommu=inclusive generic option that supersedes iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.
Note that is a noop for ARM hardware.
---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Kevin Tian <[email protected]>
--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------ xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)
docs/misc/xen-command-line.markdown +14 -0
@@ -1198,6 +1198,17 @@ detection of systems known to misbehave upon accesses to that port. >> Enable IOMMU debugging code (implies `verbose`). +> `inclusive` + +> Default: `true` + +>> Use this to work around firmware issues providing incorrect RMRR or IVMD +>> entries. Rather than only mapping RAM pages for IOMMU accesses for Dom0, +>> with this option all pages up to 4GB, not marked as unusable in the E820 +>> table, will get a mapping established. Note that this option is only +>> applicable to a PV dom0. Also note that if `dom0-strict` mode is enabled +>> then conventional RAM pages not assigned to dom0 will not be mapped. + ### iommu\_dev\_iotlb\_timeout > `= <integer>`
@@ -1212,6 +1223,9 @@ wait descriptor timed out', try increasing this value. > Default: `true` +**WARNING: This command line option is deprecated, and superseded by +_iommu=inclusive_ - using both options in combination is undefined.** + Use this to work around firmware issues providing incorrect RMRR entries. Rather than only mapping RAM pages for IOMMU accesses for Dom0, with this option all pages up to 4GB, not marked as unusable in the E820 table, will
xen/drivers/passthrough/arm/iommu.c +4 -0
@@ -73,3 +73,7 @@ int arch_iommu_populate_page_table(struct domain *d)/* The IOMMU shares the p2m with the CPU */ return -ENOSYS ; } + +void __hwdom_init arch_iommu_hwdom_init(struct domain *d) +{ +}
xen/drivers/passthrough/iommu.c +6 -0
@@ -47,6 +47,9 @@ integer_param("iommu_dev_iotlb_timeout", iommu_dev_iotlb_timeout);* no-igfx Disable VT-d for IGD devices (insecure) * no-amd-iommu-perdev-intremap Don't use per-device interrupt remapping * tables (insecure) +* inclusive Map additional regions into the IOMMU page +* tables in order to workaround bugs in ACPI +* tables. */ custom_param ("iommu" , parse_iommu_param);bool_t __initdata iommu_enable =1 ;
@@ -60,6 +63,7 @@ bool_t __read_mostly iommu_passthrough;bool_t __read_mostly iommu_snoop =1 ;bool_t __read_mostly iommu_qinval =1 ;bool_t __read_mostly iommu_intremap =1 ; +bool __hwdom_initdata iommu_inclusive = true;/* * In the current implementation of VT-d posted interrupts, in some extreme
@@ -208,6 +212,8 @@ void __hwdom_init iommu_hwdom_init(struct domain *d) } hd->platform_ops->hwdom_init (d); + +arch_iommu_hwdom_init (d); }void iommu_teardown (struct domain *d)
xen/drivers/passthrough/vtd/extern.h +0 -2
@@ -99,6 +99,4 @@ void pci_vtd_quirk(const struct pci_dev *);bool_t platform_supports_intremap (void );bool_t platform_supports_x2apic (void ); -void vtd_set_hwdom_mapping (struct domain *d); -#endif // _VTD_EXTERN_H_
xen/drivers/passthrough/vtd/iommu.c +0 -6
@@ -1304,12 +1304,6 @@ static void __hwdom_init intel_iommu_hwdom_init(struct domain *d) {struct acpi_drhd_unit *drhd; -if ( !iommu_passthrough &&is_pv_domain (d) ) - { -/* Set up 1:1 page table for hardware domain. */ -vtd_set_hwdom_mapping (d); - } -setup_hwdom_pci_devices (d, setup_hwdom_device);setup_hwdom_rmrr (d);
xen/drivers/passthrough/vtd/x86/vtd.c +1 -65
75 lines not on this page · raw message
xen/drivers/passthrough/x86/iommu.c +70 -0
82 lines not on this page · raw message
xen/include/xen/iommu.h +2 -0
@@ -35,6 +35,7 @@ extern bool_t iommu_snoop, iommu_qinval, iommu_intremap, iommu_intpost;extern bool_t iommu_hap_pt_share;extern bool_t iommu_debug;extern bool_t amd_iommu_perdev_intremap; +extern bool iommu_inclusive;extern unsigned int iommu_dev_iotlb_timeout;
@@ -49,6 +50,7 @@ void arch_iommu_domain_destroy(struct domain *d);int arch_iommu_domain_init (struct domain *d);int arch_iommu_populate_page_table (struct domain *d);void arch_iommu_check_autotranslated_hwdom (struct domain *d); +void arch_iommu_hwdom_init (struct domain *d);int iommu_construct (struct domain *d);
Signature · 7 lines
2.18.0 _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 53 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]] On Behalf Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------ xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)docs/misc/xen- renamed from docs/misc/xen-command-line.markdown +0
-0command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known to misbehave uponaccesses to that port.
Quoted 3 lines
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies
verbose).+>
inclusive
This is a dom0 (or hwdom) specific setting so perhaps dom0-inclusive?
Actually the dom0 iommu options are starting to get unwieldy as they are conflated with the general host iommu options so I think it may be worthwhile splitting things out into a separate 'dom0-iommu=' top level parameter at this stage. (My reasons are slightly selfish as I intend to add another dom0 iommu option to give it just reserved regions, to avoid unnecessary set-up if we know it will be using PV-IOMMU).
Cheers,
Paul
Quoted 307 lines
docs/misc/xen-command-line.markdown +4
-0+ +> Default: `true` + +>> Use this to work around firmware issues providing incorrect RMRR orIVMD
+>> entries. Rather than only mapping RAM pages for IOMMU accesses for Dom0,
+>> with this option all pages up to 4GB, not marked as unusable in the E820 +>> table, will get a mapping established. Note that this option is only +>> applicable to a PV dom0. Also note that ifdom0-strictmode is enabled +>> then conventional RAM pages not assigned to dom0 will not be mapped.docs/misc/xen-command-line.markdown +4
-0+ ### iommu\_dev\_iotlb\_timeout > `= <integer>`@@ -1212,6 +1223,9 @@ wait descriptor timed out', try increasing this value. > Default: `true` +**WARNING: This command line option is deprecated, and superseded by +_iommu=inclusive_ - using both options in combination is undefined.** + Use this to work around firmware issues providing incorrect RMRR entries. Rather than only mapping RAM pages for IOMMU accesses for Dom0, withthis
option all pages up to 4GB, not marked as unusable in the E820 table, willxen/drivers/passthrough/arm/iommu.c +4
-0@@ -73,3 +73,7 @@ int arch_iommu_populate_page_table(struct domain *d)/* The IOMMU shares the p2m with the CPU */ return -ENOSYS ; } + +void __hwdom_init arch_iommu_hwdom_init(struct domain *d) +{ +}xen/drivers/passthrough/iommu.c +0
-0@@ -47,6 +47,9 @@ integer_param("iommu_dev_iotlb_timeout", iommu_dev_iotlb_timeout); * no-igfx Disable VT-d for IGD devices (insecure) * no-amd-iommu-perdev-intremap Don't use per-device interruptremapping
* tables (insecure) + * inclusive Map additional regions into the IOMMU page + * tables in order to workaround bugs in ACPI + * tables.*/
custom_param("iommu", parse_iommu_param);
bool_t__initdataiommu_enable= 1;xen/drivers/passthrough/iommu.c +1
-0@@ -60,6 +63,7 @@ bool_t __read_mostly iommu_passthrough;bool_t __read_mostly iommu_snoop =1 ;bool_t __read_mostly iommu_qinval =1 ;bool_t __read_mostly iommu_intremap =1 ; +bool __hwdom_initdata iommu_inclusive = true;/* * In the current implementation of VT-d posted interrupts, in some extreme
xen/drivers/passthrough/iommu.c +2
-0@@ -208,6 +212,8 @@ void __hwdom_init iommu_hwdom_init(struct domain *d) } hd->platform_ops->hwdom_init (d); + +arch_iommu_hwdom_init (d); }void
iommu_teardown(struct domain *d)xen/drivers/passthrough/vtd/extern.h +0
-2@@ -99,6 +99,4 @@ void pci_vtd_quirk(const struct pci_dev *);bool_t platform_supports_intremap (void );bool_t platform_supports_x2apic (void ); -void vtd_set_hwdom_mapping (struct domain *d); -#endif // _VTD_EXTERN_H_ xen/drivers/passthrough/vtd/iommu.c +0
-6@@ -1304,12 +1304,6 @@ static void __hwdom_init intel_iommu_hwdom_init(struct domain *d) {struct acpi_drhd_unit *drhd; -if ( !iommu_passthrough &&is_pv_domain (d) ) - { -/* Set up 1:1 page table for hardware domain. */ -vtd_set_hwdom_mapping (d); - } -setup_hwdom_pci_devices (d, setup_hwdom_device);setup_hwdom_rmrr (d);xen/drivers/passthrough/vtd/x86/vtd.c +0
-0@@ -35,8 +35,7 @@* iommu_inclusive_mapping: when set, all memory below 4GB is included in dom0
* 1:1 iommu mappings except xen and unusable regions. */xen/drivers/passthrough/vtd/x86/vtd.c +1
-58-static bool_t __hwdom_initdata iommu_inclusive_mapping = 1; -boolean_param ("iommu_inclusive_mapping" , iommu_inclusive_mapping); +boolean_param ("iommu_inclusive_mapping" , iommu_inclusive);void *map_vtd_domain_page (u64 maddr) {@@ -108,66 +107,3 @@ void hvm_dpci_isairq_eoi(struct domain *d, unsigned int isairq) spin_unlock(&d->event_lock); } -void __hwdom_init vtd_set_hwdom_mapping(struct domain *d) -{ - unsigned long i, j, tmp, top, max_pfn; - - BUG_ON(!is_hardware_domain(d)); - - max_pfn = (GB(4) >> PAGE_SHIFT) - 1; - top = max(max_pdx, pfn_to_pdx(max_pfn) + 1); - - for ( i = 0; i < top; i++ ) - { - unsigned long pfn = pdx_to_pfn(i); - bool map; - int rc = 0; - - /* - * Set up 1:1 mapping for dom0. Default to include only - * conventional RAM areas and let RMRRs include needed reserved - * regions. When set, the inclusive mapping additionally maps in - * every pfn up to 4GB except those that fall in unusable ranges. - */ - if ( pfn > max_pfn && !mfn_valid(_mfn(pfn)) ) - continue; - - if ( iommu_inclusive_mapping && pfn <= max_pfn ) - map = !page_is_ram_type(pfn, RAM_TYPE_UNUSABLE); - else - map = page_is_ram_type(pfn, RAM_TYPE_CONVENTIONAL); - - if ( !map ) - continue; - - /* Exclude Xen bits */ - if ( xen_in_range(pfn) ) - continue; - - /* - * If dom0-strict mode is enabled then exclude conventional RAM - * and let the common code map dom0's pages. - */ - if ( iommu_dom0_strict && - page_is_ram_type(pfn, RAM_TYPE_CONVENTIONAL) ) - continue; - - tmp = 1 << (PAGE_SHIFT - PAGE_SHIFT_4K); - for ( j = 0; j < tmp; j++ ) - { - int ret = iommu_map_page(d, pfn * tmp + j, pfn * tmp + j, - IOMMUF_readable|IOMMUF_writable); - - if ( !rc ) - rc = ret; - } - - if ( rc ) - printk(XENLOG_WARNING VTDPREFIX " d%d: IOMMU mapping failed:%d\n",
xen/drivers/passthrough/vtd/x86/vtd.c +0
-7- d->domain_id, rc); - -if (!(i & (0xfffff >> (PAGE_SHIFT -PAGE_SHIFT_4K )))) -process_pending_softirqs (); - } -} -xen/drivers/passthrough/x86/iommu.c +70
-0@@ -20,6 +20,8 @@#include <xen/softirq.h> #include <xsm/xsm.h> +#include <asm/setup.h> +void iommu_update_ire_from_apic (unsigned int apic,unsigned int reg,unsigned int value) {@@ -132,6 +134,74 @@ void arch_iommu_domain_destroy(struct domain *d) { } +void __hwdom_init arch_iommu_hwdom_init(struct domain *d) +{ +unsigned long i, j, tmp, top, max_pfn; + +if ( iommu_passthrough || !is_pv_domain (d) ) +return ; + +BUG_ON (!is_hardware_domain (d)); + + max_pfn = (GB (4 ) >>PAGE_SHIFT ) -1 ; + top =max (max_pdx,pfn_to_pdx (max_pfn) +1 ); + +for ( i =0 ; i < top; i++ ) + { +unsigned long pfn =pdx_to_pfn (i); +bool map; +int rc =0 ; + +/* +* Set up 1:1 mapping for dom0. Default to include only +* conventional RAM areas and let RMRRs include needed reserved +* regions. When set, the inclusive mapping additionally maps in +* every pfn up to 4GB except those that fall in unusable ranges. +*/ +if ( pfn > max_pfn && !mfn_valid (_mfn (pfn)) ) +continue ; + +if ( iommu_inclusive && pfn <= max_pfn ) + map = !page_is_ram_type (pfn,RAM_TYPE_UNUSABLE ); +else + map =page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ); + +if ( !map ) +continue ; + +/* Exclude Xen bits */ +if (xen_in_range (pfn) ) +continue ; + +/* +* If dom0-strict mode is enabled then exclude conventional RAM +* and let the common code map dom0's pages. +*/ +if ( iommu_dom0_strict && +page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ) ) +continue ; + + tmp =1 << (PAGE_SHIFT -PAGE_SHIFT_4K ); +for ( j =0 ; j < tmp; j++ ) + { +int ret =iommu_map_page (d, pfn * tmp + j, pfn * tmp + j, + IOMMUF_readable|IOMMUF_writable); + +if ( !rc ) + rc = ret; + } + +if ( rc ) +printk (XENLOG_WARNING "d%d: IOMMU mapping failed: %d\n" , + d->domain_id, rc); + +if (!(i & (0xfffff >> (PAGE_SHIFT -PAGE_SHIFT_4K )))) +process_pending_softirqs (); + } + + +} +/* * Local variables: * mode: C
xen/include/xen/iommu.h +2
-1@@ -35,6 +35,7 @@ extern bool_t iommu_snoop, iommu_qinval, iommu_intremap, iommu_intpost;extern bool_t iommu_hap_pt_share;extern bool_t iommu_debug;extern bool_t amd_iommu_perdev_intremap; +extern bool iommu_inclusive;extern unsigned int iommu_dev_iotlb_timeout;@@ -49,6 +50,7 @@ void arch_iommu_domain_destroy(struct domain *d);int arch_iommu_domain_init (struct domain *d);int arch_iommu_populate_page_table (struct domain *d);void arch_iommu_check_autotranslated_hwdom (struct domain *d); +void arch_iommu_hwdom_init (struct domain *d);int iommu_construct (struct domain *d); --2.18.0
Quoted 4 lines
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 53 lines · Paul Durrant
Quoted 53 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]] On Behalf Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------ xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)docs/misc/xen- renamed from docs/misc/xen-command-line.markdown +0
-0command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known to misbehave uponaccesses to that port.
Quoted 7 lines
Quoted 3 lines
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies
verbose).+>
inclusiveThis is a dom0 (or hwdom) specific setting so perhaps dom0-inclusive?
Actually the dom0 iommu options are starting to get unwieldy as they are conflated with the general host iommu options so I think it may be worthwhile splitting things out into a separate 'dom0-iommu=' top level parameter at this stage. (My reasons are slightly selfish as I intend to add another dom0 iommu option to give it just reserved regions, to avoid unnecessary set-up if we know it will be using PV-IOMMU).
Mapping just the reserved regions is what I actually do for PVH with iommu=inclusive (patch 4/4), so maybe it would make sense to speak about the naming here in order to use the same naming for PV and PVH.
TBH I don't really like the dom0- prefix, the command line iommu options either apply to all domains or Dom0 only, having domu-inclusive for example makes no sense IMO.
Maybe the current iommu_inclusive_mapping could be mapped to iommu=inclusive-mapping instead of iommu=inclusive and the new option could be added as iommu=reserved-mapping?
That would work for PVH AFAICT.
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 84 lines
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:16
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 2 lines · Paul Durrant
Quoted 2 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]] OnBehalf
Quoted 7 lines
Quoted 7 lines
Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>;Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 35 lines
Quoted 35 lines
<[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------ xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)Quoted 6 lines
docs/misc/xen-command-line.markdown detection of systems known to misbehave
Quoted 6 lines
docs/misc/xen- renamed from docs/misc/xen-command-line.markdown +0
-0command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known to misbehaveupon
Quoted 9 lines
Quoted 5 lines
accesses to that port.
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies
verbose).+>
inclusiveThis is a dom0 (or hwdom) specific setting so perhaps dom0-inclusive?
Actually the dom0 iommu options are starting to get unwieldy as they are
conflated with the general host iommu options so I think it may be worthwhile splitting things out into a separate 'dom0-iommu=' top level parameter at this stage. (My reasons are slightly selfish as I intend to add another dom0 iommu option to give it just reserved regions, to avoid unnecessary set-up if we know it will be using PV-IOMMU).
Quoted 8 lines
Mapping just the reserved regions is what I actually do for PVH with iommu=inclusive (patch 4/4), so maybe it would make sense to speak about the
naming here in order to use the same naming for PV and PVH.TBH I don't really like the dom0- prefix, the command line iommu options either apply to all domains or Dom0 only, having domu-inclusive for example makes no sense IMO.
No, I think there are some options that you may want to apply to dom0 only, but these are more like the dom0_mem or dom0_max_vpus options. Particularly, the inclusive option is probably something that is only desirable for dom0. Clearly dom0-passthrough and dom0-strict are already defined to relate to dom0 only, and options such as 'reserved' should only be specific on the command line in relation to dom0 IMO. For other domains such an option should be specified via xl.cfg.
Paul
Quoted 8 lines
Maybe the current
iommu_inclusive_mappingcould be mapped to iommu=inclusive-mapping instead of iommu=inclusive and the new option could be added as iommu=reserved-mapping?That would work for PVH AFAICT.
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 89 lines · Paul Durrant
Quoted 89 lines
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:16
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 2 lines · Paul Durrant
Quoted 2 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]] OnBehalf
Quoted 7 lines
Quoted 7 lines
Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>;Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 35 lines
Quoted 35 lines
<[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------ xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)Quoted 6 lines
docs/misc/xen-command-line.markdown detection of systems known to misbehave
Quoted 6 lines
docs/misc/xen- renamed from docs/misc/xen-command-line.markdown +0
-0command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known to misbehaveupon
Quoted 9 lines
Quoted 5 lines
accesses to that port.
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies
verbose).+>
inclusiveThis is a dom0 (or hwdom) specific setting so perhaps dom0-inclusive?
Actually the dom0 iommu options are starting to get unwieldy as they are
conflated with the general host iommu options so I think it may be worthwhile splitting things out into a separate 'dom0-iommu=' top level parameter at this stage. (My reasons are slightly selfish as I intend to add another dom0 iommu option to give it just reserved regions, to avoid unnecessary set-up if we know it will be using PV-IOMMU).
Mapping just the reserved regions is what I actually do for PVH with iommu=inclusive (patch 4/4), so maybe it would make sense to speak about the
naming here in order to use the same naming for PV and PVH.
Quoted 5 lines
Quoted 3 lines
TBH I don't really like the dom0- prefix, the command line iommu options either apply to all domains or Dom0 only, having domu-inclusive for example makes no sense IMO.
No, I think there are some options that you may want to apply to dom0 only, but these are more like the
dom0_memordom0_max_vpusoptions. Particularly, the inclusive option is probably something that is only desirable for dom0. Clearly dom0-passthrough and dom0-strict are already defined to relate to dom0 only, and options such as 'reserved' should only be specific on the command line in relation to dom0 IMO. For other domains such an option should be specified via xl.cfg.
Yes, we already have a bunch of those, so then I think dom0-inclusive and dom0-reserved would be appropriate?
dom0-inclusive-mapping or dom0-reserved-mapping seems too long.
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 126 lines
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:34
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 6 lines · Paul Durrant
Quoted 6 lines
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:16
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei LiuQuoted 2 lines
Quoted 2 lines
George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>;
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 7 lines
Quoted 7 lines
<[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 2 lines · Paul Durrant
Quoted 2 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]]On
Quoted 6 lines
Quoted 6 lines
Behalf
Quoted 5 lines
Quoted 5 lines
Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; GeorgeDunlap
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
<[email protected]>; Andrew Cooper
<[email protected]>; Ian JacksonQuoted 9 lines
Quoted 9 lines
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 6 lines
Quoted 6 lines
<[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional changeon
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Intel hardware, while AMD hardware will gain the same functionality
of
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
mapping almost everything below the 4GB boundary.
Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------ xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)Quoted 1 line
Quoted 1 line
Quoted 1 line
Review
Quoted 1 line
(quoted diff) +0
-0b/docs/misc/xen-
Quoted 8 lines
Quoted 8 lines
Quoted 5 lines
docs/misc/xen-command-line.markdown detection of systems known to misbehave
Quoted 5 lines
command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known to misbehaveupon
Quoted 2 lines
Quoted 2 lines
accesses to that port.
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies
verbose).+>
inclusiveThis is a dom0 (or hwdom) specific setting so perhaps dom0-inclusive?
Actually the dom0 iommu options are starting to get unwieldy as they
are
Quoted 8 lines
Quoted 8 lines
conflated with the general host iommu options so I think it may be worthwhile splitting things out into a separate 'dom0-iommu=' top level parameter at this stage. (My reasons are slightly selfish as I intend to add another dom0 iommu option to give it just reserved regions, to avoid unnecessary set-up if we know it will be using PV-IOMMU).
Mapping just the reserved regions is what I actually do for PVH with iommu=inclusive (patch 4/4), so maybe it would make sense to speak
about
Quoted 8 lines
Quoted 6 lines
the
naming here in order to use the same naming for PV and PVH.TBH I don't really like the dom0- prefix, the command line iommu options either apply to all domains or Dom0 only, having domu-inclusive for example makes no sense IMO.
No, I think there are some options that you may want to apply to dom0
only, but these are more like the
dom0_memordom0_max_vpusoptions. Particularly, the inclusive option is probably something that is only desirable for dom0. Clearly dom0-passthrough and dom0-strict are already defined to relate to dom0 only, and options such as 'reserved' should only be specific on the command line in relation to dom0 IMO. For other domains such an option should be specified via xl.cfg.
Quoted 4 lines
Yes, we already have a bunch of those, so then I think dom0-inclusive and dom0-reserved would be appropriate?
dom0-inclusive-mapping or dom0-reserved-mapping seems too long.
Yes, those names are ok, but I still think it better in the long run if we have something like:
dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
where relaxed is the default and 'none' (I think) is equivalent to the current iommu=dom0-passthrough.
Paul
Quoted 2 lines
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 133 lines
Quoted 95 lines
Quoted 1 line · On 31.07.18 at 10:37, <[email protected]> wrote:
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:34
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 6 lines · Paul Durrant
Quoted 6 lines
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:16
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei LiuQuoted 2 lines
Quoted 2 lines
George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>;
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 7 lines
Quoted 7 lines
<[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 2 lines · Paul Durrant
Quoted 2 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]]On
Quoted 6 lines
Quoted 6 lines
Behalf
Quoted 5 lines
Quoted 5 lines
Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; GeorgeDunlap
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
<[email protected]>; Andrew Cooper
<[email protected]>; Ian JacksonQuoted 9 lines
Quoted 9 lines
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 6 lines
Quoted 6 lines
<[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional changeon
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Intel hardware, while AMD hardware will gain the same functionality
of
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
mapping almost everything below the 4GB boundary.
Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------ xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)Quoted 1 line
Quoted 1 line
Quoted 1 line
Review
Quoted 1 line
(quoted diff) +0
-0b/docs/misc/xen-
Quoted 8 lines
Quoted 8 lines
Quoted 5 lines
docs/misc/xen-command-line.markdown detection of systems known to misbehave
Quoted 5 lines
command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known to misbehaveupon
Quoted 2 lines
Quoted 2 lines
accesses to that port.
Quoted 11 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies
verbose).+>
inclusiveThis is a dom0 (or hwdom) specific setting so perhaps dom0-inclusive?
Actually the dom0 iommu options are starting to get unwieldy as they
are
Quoted 3 lines
Quoted 3 lines
conflated with the general host iommu options so I think it may be worthwhile splitting things out into a separate 'dom0-iommu=' top level parameter at this stage. (My reasons are slightly selfish as I intend to
add
Quoted 21 lines
Quoted 5 lines
Quoted 5 lines
another dom0 iommu option to give it just reserved regions, to avoid unnecessary set-up if we know it will be using PV-IOMMU).
Mapping just the reserved regions is what I actually do for PVH with iommu=inclusive (patch 4/4), so maybe it would make sense to speak
about
Quoted 8 lines
Quoted 6 lines
the
naming here in order to use the same naming for PV and PVH.TBH I don't really like the dom0- prefix, the command line iommu options either apply to all domains or Dom0 only, having domu-inclusive for example makes no sense IMO.
No, I think there are some options that you may want to apply to dom0
only, but these are more like the
dom0_memordom0_max_vpusoptions. Particularly, the inclusive option is probably something that is only desirable for dom0. Clearly dom0-passthrough and dom0-strict are already defined to relate to dom0 only, and options such as 'reserved' should only be specific on the command line in relation to dom0 IMO. For other domains such an option should be specified via xl.cfg.Quoted 4 lines
Yes, we already have a bunch of those, so then I think dom0-inclusive and dom0-reserved would be appropriate?
dom0-inclusive-mapping or dom0-reserved-mapping seems too long.
Quoted 7 lines
Yes, those names are ok, but I still think it better in the long run if we have something like:
dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
where relaxed is the default and 'none' (I think) is equivalent to the current iommu=dom0-passthrough.
Or, along the lines of the other reply just sent, e.g.
dom0=pvh,iommu:inclusive;reserved,shadow
But perhaps the difference between , and ; gets too confusing then.
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 138 lines · Jan Beulich
Quoted 138 lines
Quoted 95 lines
Quoted 1 line · On 31.07.18 at 10:37, <[email protected]> wrote:
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:34
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 6 lines · Paul Durrant
Quoted 6 lines
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:16
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei LiuQuoted 2 lines
Quoted 2 lines
George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>;
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 7 lines
Quoted 7 lines
<[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 2 lines · Paul Durrant
Quoted 2 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]]On
Quoted 6 lines
Quoted 6 lines
Behalf
Quoted 5 lines
Quoted 5 lines
Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; GeorgeDunlap
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
<[email protected]>; Andrew Cooper
<[email protected]>; Ian JacksonQuoted 9 lines
Quoted 9 lines
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 6 lines
Quoted 6 lines
<[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional changeon
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Intel hardware, while AMD hardware will gain the same functionality
of
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
mapping almost everything below the 4GB boundary.
Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------ xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)Quoted 1 line
Quoted 1 line
Quoted 1 line
Review
Quoted 1 line
(quoted diff) +0
-0b/docs/misc/xen-
Quoted 8 lines
Quoted 8 lines
Quoted 5 lines
docs/misc/xen-command-line.markdown detection of systems known to misbehave
Quoted 5 lines
command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known to misbehaveupon
Quoted 2 lines
Quoted 2 lines
accesses to that port.
Quoted 11 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies
verbose).+>
inclusiveThis is a dom0 (or hwdom) specific setting so perhaps dom0-inclusive?
Actually the dom0 iommu options are starting to get unwieldy as they
are
Quoted 3 lines
Quoted 3 lines
conflated with the general host iommu options so I think it may be worthwhile splitting things out into a separate 'dom0-iommu=' top level parameter at this stage. (My reasons are slightly selfish as I intend to
add
Quoted 21 lines
Quoted 5 lines
Quoted 5 lines
another dom0 iommu option to give it just reserved regions, to avoid unnecessary set-up if we know it will be using PV-IOMMU).
Mapping just the reserved regions is what I actually do for PVH with iommu=inclusive (patch 4/4), so maybe it would make sense to speak
about
Quoted 8 lines
Quoted 6 lines
the
naming here in order to use the same naming for PV and PVH.TBH I don't really like the dom0- prefix, the command line iommu options either apply to all domains or Dom0 only, having domu-inclusive for example makes no sense IMO.
No, I think there are some options that you may want to apply to dom0
only, but these are more like the
dom0_memordom0_max_vpusoptions. Particularly, the inclusive option is probably something that is only desirable for dom0. Clearly dom0-passthrough and dom0-strict are already defined to relate to dom0 only, and options such as 'reserved' should only be specific on the command line in relation to dom0 IMO. For other domains such an option should be specified via xl.cfg.Quoted 4 lines
Yes, we already have a bunch of those, so then I think dom0-inclusive and dom0-reserved would be appropriate?
dom0-inclusive-mapping or dom0-reserved-mapping seems too long.
Yes, those names are ok, but I still think it better in the long run if we have something like:
dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
Quoted 8 lines
Quoted 2 lines
where relaxed is the default and 'none' (I think) is equivalent to the current iommu=dom0-passthrough.
Or, along the lines of the other reply just sent, e.g.
dom0=pvh,iommu:inclusive;reserved,shadow
But perhaps the difference between , and ; gets too confusing then.
So I think we have the following options:
1. dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
2. dom0=[pvh,][shadow,][iommu=[inclusive;][reserved;][strict;][none;][relaxed]]
3. dom0=[pvh,][shadow,][iommu-inclusive,][iommu-reserved,][iommu-strict,][iommu-none,][iommu-relaxed]
I don't have a strong preference between 1 and 3, but I would prefer to avoid 2 because I think suboptions inside of options it's too complex IMO.
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 147 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 11:05, <[email protected]> wrote:
Quoted 116 lines · Jan Beulich
Quoted 116 lines
Quoted 95 lines
Quoted 1 line · On 31.07.18 at 10:37, <[email protected]> wrote:
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:34
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 6 lines · Paul Durrant
Quoted 6 lines
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:16
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei LiuQuoted 2 lines
Quoted 2 lines
George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>;
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 7 lines
Quoted 7 lines
<[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 2 lines · Paul Durrant
Quoted 2 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]]On
Quoted 6 lines
Quoted 6 lines
Behalf
Quoted 5 lines
Quoted 5 lines
Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; GeorgeDunlap
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
<[email protected]>; Andrew Cooper
<[email protected]>; Ian JacksonQuoted 9 lines
Quoted 9 lines
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 6 lines
Quoted 6 lines
<[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional changeon
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Intel hardware, while AMD hardware will gain the same functionality
of
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
mapping almost everything below the 4GB boundary.
Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------ xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)Quoted 1 line
Quoted 1 line
Quoted 1 line
Review
Quoted 1 line
(quoted diff) +0
-0b/docs/misc/xen-
Quoted 8 lines
Quoted 8 lines
Quoted 5 lines
docs/misc/xen-command-line.markdown detection of systems known to misbehave
Quoted 5 lines
command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known to misbehaveupon
Quoted 2 lines
Quoted 2 lines
accesses to that port.
Quoted 11 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies
verbose).+>
inclusiveThis is a dom0 (or hwdom) specific setting so perhaps dom0-inclusive?
Actually the dom0 iommu options are starting to get unwieldy as they
are
Quoted 3 lines
Quoted 3 lines
conflated with the general host iommu options so I think it may be worthwhile splitting things out into a separate 'dom0-iommu=' top level parameter at this stage. (My reasons are slightly selfish as I intend to
add
Quoted 9 lines
Quoted 5 lines
Quoted 5 lines
another dom0 iommu option to give it just reserved regions, to avoid unnecessary set-up if we know it will be using PV-IOMMU).
Mapping just the reserved regions is what I actually do for PVH with iommu=inclusive (patch 4/4), so maybe it would make sense to speak
about
Quoted 3 lines
Quoted 3 lines
the
naming here in order to use the same naming for PV and PVH.Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 5 lines
Quoted 3 lines
TBH I don't really like the dom0- prefix, the command line iommu options either apply to all domains or Dom0 only, having domu-inclusive for example makes no sense IMO.
No, I think there are some options that you may want to apply to dom0
only, but these are more like the
dom0_memordom0_max_vpusoptions. Particularly, the inclusive option is probably something that is onlydesirable
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
for dom0. Clearly dom0-passthrough and dom0-strict are already defined to relate to dom0 only, and options such as 'reserved' should only be specific
on
Quoted 18 lines
Quoted 8 lines
Quoted 7 lines
the command line in relation to dom0 IMO. For other domains such an option should be specified via xl.cfg.
Yes, we already have a bunch of those, so then I think dom0-inclusive and dom0-reserved would be appropriate?
dom0-inclusive-mapping or dom0-reserved-mapping seems too long.
Quoted 7 lines
Yes, those names are ok, but I still think it better in the long run if we have something like:
dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
where relaxed is the default and 'none' (I think) is equivalent to the current iommu=dom0-passthrough.
Or, along the lines of the other reply just sent, e.g.
Quoted 7 lines
Quoted 3 lines
dom0=pvh,iommu:inclusive;reserved,shadow
But perhaps the difference between , and ; gets too confusing then.
So I think we have the following options:
1. dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
Nit: dom0-iommu= (no underscores in new options)
Quoted 6 lines
2. dom0=[pvh,][shadow,][iommu=[inclusive;][reserved;][strict;][none;][relaxed]]
3. dom0=[pvh,][shadow,][iommu-inclusive,][iommu-reserved,][iommu-strict,][iommu-none,][iommu-relaxed]I don't have a strong preference between 1 and 3, but I would prefer to avoid 2 because I think suboptions inside of options it's too complex IMO.
While generally I prefer to limit the number of top level options, in this case I think I'd prefer 1 after all. Or wait - does any pair of the (sub)options actually make sense to be specified? Isn't it rather a choice of five than an enumeration of up to 5? In which case I'd still prefer 2 (as then there's no need for a second separator beside comma), the more that we have at least one example with such sub-options (cpufreq).
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 160 lines · Jan Beulich
Quoted 147 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 11:05, <[email protected]> wrote:
Quoted 116 lines · Jan Beulich
Quoted 116 lines
Quoted 95 lines
Quoted 1 line · On 31.07.18 at 10:37, <[email protected]> wrote:
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:34
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 6 lines · Paul Durrant
Quoted 6 lines
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:16
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei LiuQuoted 2 lines
Quoted 2 lines
George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>;
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 7 lines
Quoted 7 lines
<[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 2 lines · Paul Durrant
Quoted 2 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]]On
Quoted 6 lines
Quoted 6 lines
Behalf
Quoted 5 lines
Quoted 5 lines
Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; GeorgeDunlap
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
<[email protected]>; Andrew Cooper
<[email protected]>; Ian JacksonQuoted 9 lines
Quoted 9 lines
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 6 lines
Quoted 6 lines
<[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional changeon
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Intel hardware, while AMD hardware will gain the same functionality
of
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
mapping almost everything below the 4GB boundary.
Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------ xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)Quoted 1 line
Quoted 1 line
Quoted 1 line
Review
Quoted 1 line
(quoted diff) +0
-0b/docs/misc/xen-
Quoted 8 lines
Quoted 8 lines
Quoted 5 lines
docs/misc/xen-command-line.markdown detection of systems known to misbehave
Quoted 5 lines
command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known to misbehaveupon
Quoted 2 lines
Quoted 2 lines
accesses to that port.
Quoted 11 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies
verbose).+>
inclusiveThis is a dom0 (or hwdom) specific setting so perhaps dom0-inclusive?
Actually the dom0 iommu options are starting to get unwieldy as they
are
Quoted 3 lines
Quoted 3 lines
conflated with the general host iommu options so I think it may be worthwhile splitting things out into a separate 'dom0-iommu=' top level parameter at this stage. (My reasons are slightly selfish as I intend to
add
Quoted 9 lines
Quoted 5 lines
Quoted 5 lines
another dom0 iommu option to give it just reserved regions, to avoid unnecessary set-up if we know it will be using PV-IOMMU).
Mapping just the reserved regions is what I actually do for PVH with iommu=inclusive (patch 4/4), so maybe it would make sense to speak
about
Quoted 3 lines
Quoted 3 lines
the
naming here in order to use the same naming for PV and PVH.Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 5 lines
Quoted 3 lines
TBH I don't really like the dom0- prefix, the command line iommu options either apply to all domains or Dom0 only, having domu-inclusive for example makes no sense IMO.
No, I think there are some options that you may want to apply to dom0
only, but these are more like the
dom0_memordom0_max_vpusoptions. Particularly, the inclusive option is probably something that is onlydesirable
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
for dom0. Clearly dom0-passthrough and dom0-strict are already defined to relate to dom0 only, and options such as 'reserved' should only be specific
on
Quoted 18 lines
Quoted 8 lines
Quoted 7 lines
the command line in relation to dom0 IMO. For other domains such an option should be specified via xl.cfg.
Yes, we already have a bunch of those, so then I think dom0-inclusive and dom0-reserved would be appropriate?
dom0-inclusive-mapping or dom0-reserved-mapping seems too long.
Quoted 7 lines
Yes, those names are ok, but I still think it better in the long run if we have something like:
dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
where relaxed is the default and 'none' (I think) is equivalent to the current iommu=dom0-passthrough.
Or, along the lines of the other reply just sent, e.g.
Quoted 7 lines
Quoted 3 lines
dom0=pvh,iommu:inclusive;reserved,shadow
But perhaps the difference between , and ; gets too confusing then.
So I think we have the following options:
1. dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
Nit: dom0-iommu= (no underscores in new options)
Quoted 3 lines
2. dom0=[pvh,][shadow,][iommu=[inclusive;][reserved;][strict;][none;][relaxed]]
3. dom0=[pvh,][shadow,][iommu-inclusive,][iommu-reserved,][iommu-strict,][iommu-none,][iommu-relaxed]
Quoted 7 lines
Quoted 3 lines
I don't have a strong preference between 1 and 3, but I would prefer to avoid 2 because I think suboptions inside of options it's too complex IMO.
While generally I prefer to limit the number of top level options, in this case I think I'd prefer 1 after all. Or wait - does any pair of the (sub)options actually make sense to be specified?
Yes, for example you can use strict and inclusive at the same time, I think it's something like:
dom0=[pvh,][shadow,][iommu=[inclusive|reserved;][strict|none|relaxed]]
Quoted 5 lines
Isn't it rather a
choice of five than an enumeration of up to 5? In which case I'd still prefer 2 (as then there's no need for a second separator beside comma), the more that we have at least one example with such sub-options (cpufreq).
OK, I can do the nested iommu option inside of dom0 if that's the preference.
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 212 lines
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 10:35
To: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Paul Durrant <[email protected]>; Wei Liu <[email protected]>; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; xen-devel <[email protected]>; Tim (Xen.org) <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 8 lines · Jan Beulich
Quoted 8 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 11:05, <[email protected]> wrote:
Quoted 6 lines · Jan Beulich
Quoted 6 lines
Quoted 6 lines
Quoted 1 line · On 31.07.18 at 10:37, <[email protected]> wrote:
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:34
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin TianQuoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Stefano Stabellini <[email protected]>; Wei Liu
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson
<[email protected]>; Tim
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
<[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 5 lines · Paul Durrant
Quoted 5 lines
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:16
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin TianQuoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 1 line
Quoted 1 line
Stefano Stabellini <[email protected]>; Wei Liu
Quoted 2 lines
Quoted 2 lines
George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>;
Jan
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Beulich
Quoted 7 lines
Quoted 7 lines
<[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 2 lines · Paul Durrant
Quoted 2 lines
-----Original Message-----
From: Xen-devel [mailto:xen-devel-Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
On
Quoted 6 lines
Quoted 6 lines
Behalf
Quoted 5 lines
Quoted 5 lines
Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>;George
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Dunlap
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
<[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson
Quoted 2 lines
Quoted 2 lines
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall
<[email protected]>; Jan
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Beulich
Quoted 1 line
Quoted 1 line
<[email protected]>; Roger Pau Monne
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Subject: [Xen-devel] [PATCH 2/4] iommu: generalize
iommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that
supersedes
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
iommu_inclusive_mapping. This should be a non-functionalchange
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
on
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Intel hardware, while AMD hardware will gain the same
functionality
Quoted 16 lines
Quoted 16 lines
Quoted 16 lines
Quoted 16 lines
Quoted 16 lines
of
Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
mapping almost everything below the 4GB boundary.
Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Cc Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Review
Quoted 1 line
(quoted diff) +0
-0b/docs/misc/xen-
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
docs/misc/xen-command-line.markdown detection of systems known to
Quoted 5 lines
command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known tomisbehave
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
upon
Quoted 7 lines
Quoted 5 lines
accesses to that port.
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies
verbose).+>
inclusiveThis is a dom0 (or hwdom) specific setting so perhaps dom0-
inclusive?
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Actually the dom0 iommu options are starting to get unwieldy
as they
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
are
Quoted 1 line
Quoted 1 line
conflated with the general host iommu options so I think it may
be
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
worthwhile splitting things out into a separate 'dom0-iommu='
top level
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
parameter at this stage. (My reasons are slightly selfish as I
intend to
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
add
Quoted 1 line
Quoted 1 line
Quoted 1 line
another dom0 iommu option to give it just reserved regions, to
avoid
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
unnecessary set-up if we know it will be using PV-IOMMU).
Mapping just the reserved regions is what I actually do for PVH
with
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
iommu=inclusive (patch 4/4), so maybe it would make sense to
speak
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
about
Quoted 8 lines
Quoted 6 lines
the
naming here in order to use the same naming for PV and PVH.TBH I don't really like the dom0- prefix, the command line iommu options either apply to all domains or Dom0 only, having domu-inclusive for example makes no sense IMO.
No, I think there are some options that you may want to apply to
dom0
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
only, but these are more like the
dom0_memordom0_max_vpusoptions.
Quoted 3 lines
Quoted 3 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Particularly, the inclusive option is probably something that is only
desirable
Quoted 1 line
Quoted 1 line
Quoted 1 line
for dom0. Clearly dom0-passthrough and dom0-strict are already
defined to
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
relate to dom0 only, and options such as 'reserved' should only be
specific
Quoted 2 lines
Quoted 2 lines
on
Quoted 1 line
Quoted 1 line
Quoted 1 line
the command line in relation to dom0 IMO. For other domains such
an option
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
should be specified via xl.cfg.
Yes, we already have a bunch of those, so then I think dom0-
inclusive
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
and dom0-reserved would be appropriate?
dom0-inclusive-mapping or dom0-reserved-mapping seems too
long.
Quoted 13 lines
Quoted 13 lines
Quoted 13 lines
Quoted 8 lines
Yes, those names are ok, but I still think it better in the long run if we have something like:
dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
where relaxed is the default and 'none' (I think) is equivalent to the current iommu=dom0-passthrough.
Or, along the lines of the other reply just sent, e.g.
dom0=pvh,iommu:inclusive;reserved,shadow
Quoted 9 lines
Quoted 5 lines
Quoted 1 line
But perhaps the difference between , and ; gets too confusing then.
So I think we have the following options:
1. dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
Nit: dom0-iommu= (no underscores in new options)
Quoted 1 line
2.
dom0=[pvh,][shadow,][iommu=[inclusive;][reserved;][strict;][none;][relaxe d]]
Quoted 1 line
Quoted 1 line
3. dom0=[pvh,][shadow,][iommu-inclusive,][iommu-reserved,][iommu-
strict,][iommu-none,][iommu-relaxed]
Quoted 5 lines
Quoted 4 lines
I don't have a strong preference between 1 and 3, but I would prefer to avoid 2 because I think suboptions inside of options it's too complex IMO.
Quoted 8 lines
Quoted 3 lines
While generally I prefer to limit the number of top level options, in this case I think I'd prefer 1 after all. Or wait - does any pair of the (sub)options actually make sense to be specified?
Yes, for example you can use strict and inclusive at the same time, I think it's something like:
dom0=[pvh,][shadow,][iommu=[inclusive|reserved;][strict|none|relaxed]]
I think it may be:
dom0=[pvh,][shadow,][iommu=[inclusive|reserved|none;][strict|relaxed]]
where 'none' in the first part means the second part is moot. (Not sure how to express that cleanly).
Paul
Quoted 11 lines
Quoted 6 lines
Isn't it rather a
choice of five than an enumeration of up to 5? In which case I'd still prefer 2 (as then there's no need for a second separator beside comma), the more that we have at least one example with such sub-options (cpufreq).OK, I can do the nested iommu option inside of dom0 if that's the preference.
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
In reply to Roger Pau MonnéQuoted 178 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 11:34, <[email protected]> wrote:
Quoted 123 lines · Jan Beulich
Quoted 123 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 11:05, <[email protected]> wrote:
Quoted 121 lines · Jan Beulich
Quoted 121 lines
Quoted 95 lines
Quoted 1 line · On 31.07.18 at 10:37, <[email protected]> wrote:
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:34
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 6 lines · Paul Durrant
Quoted 6 lines
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:16
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei LiuQuoted 2 lines
Quoted 2 lines
George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>;
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 7 lines
Quoted 7 lines
<[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 2 lines · Paul Durrant
Quoted 2 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]]On
Quoted 6 lines
Quoted 6 lines
Behalf
Quoted 5 lines
Quoted 5 lines
Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; GeorgeDunlap
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
<[email protected]>; Andrew Cooper
<[email protected]>; Ian JacksonQuoted 9 lines
Quoted 9 lines
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 6 lines
Quoted 6 lines
<[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional changeon
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Intel hardware, while AMD hardware will gain the same functionality
of
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
Quoted 28 lines
mapping almost everything below the 4GB boundary.
Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------ xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)Quoted 1 line
Quoted 1 line
Quoted 1 line
Review
Quoted 1 line
(quoted diff) +0
-0b/docs/misc/xen-
Quoted 8 lines
Quoted 8 lines
Quoted 5 lines
docs/misc/xen-command-line.markdown detection of systems known to misbehave
Quoted 5 lines
command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known to misbehaveupon
Quoted 2 lines
Quoted 2 lines
accesses to that port.
Quoted 11 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies
verbose).+>
inclusiveThis is a dom0 (or hwdom) specific setting so perhaps dom0-inclusive?
Actually the dom0 iommu options are starting to get unwieldy as they
are
Quoted 3 lines
Quoted 3 lines
conflated with the general host iommu options so I think it may be worthwhile splitting things out into a separate 'dom0-iommu=' top level parameter at this stage. (My reasons are slightly selfish as I intend to
add
Quoted 13 lines
Quoted 5 lines
Quoted 5 lines
another dom0 iommu option to give it just reserved regions, to avoid unnecessary set-up if we know it will be using PV-IOMMU).
Mapping just the reserved regions is what I actually do for PVH with iommu=inclusive (patch 4/4), so maybe it would make sense to speak
about
Quoted 7 lines
Quoted 6 lines
the
naming here in order to use the same naming for PV and PVH.TBH I don't really like the dom0- prefix, the command line iommu options either apply to all domains or Dom0 only, having domu-inclusive for example makes no sense IMO.
Quoted 6 lines
Quoted 6 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 1 line
No, I think there are some options that you may want to apply to dom0
only, but these are more like the
dom0_memordom0_max_vpusoptions. Particularly, the inclusive option is probably something that is onlydesirable
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
for dom0. Clearly dom0-passthrough and dom0-strict are already defined to relate to dom0 only, and options such as 'reserved' should only be
specific
Quoted 33 lines
Quoted 20 lines
on
Quoted 19 lines
Quoted 8 lines
Quoted 7 lines
the command line in relation to dom0 IMO. For other domains such an option should be specified via xl.cfg.
Yes, we already have a bunch of those, so then I think dom0-inclusive and dom0-reserved would be appropriate?
dom0-inclusive-mapping or dom0-reserved-mapping seems too long.
Quoted 8 lines
Yes, those names are ok, but I still think it better in the long run if we
have something like:
dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
where relaxed is the default and 'none' (I think) is equivalent to the current iommu=dom0-passthrough.
Or, along the lines of the other reply just sent, e.g.
Quoted 7 lines
Quoted 3 lines
dom0=pvh,iommu:inclusive;reserved,shadow
But perhaps the difference between , and ; gets too confusing then.
So I think we have the following options:
1. dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
Nit: dom0-iommu= (no underscores in new options)
Quoted 3 lines
2. dom0=[pvh,][shadow,][iommu=[inclusive;][reserved;][strict;][none;][relaxed]]
3. dom0=[pvh,][shadow,][iommu-inclusive,][iommu-reserved,][iommu-strict,][iommu-none,][iommu-relaxed]Quoted 7 lines
Quoted 3 lines
I don't have a strong preference between 1 and 3, but I would prefer to avoid 2 because I think suboptions inside of options it's too complex IMO.
While generally I prefer to limit the number of top level options, in this case I think I'd prefer 1 after all. Or wait - does any pair of the (sub)options actually make sense to be specified?
Yes, for example you can use strict and inclusive at the same time, I think it's something like:
dom0=[pvh,][shadow,][iommu=[inclusive|reserved;][strict|none|relaxed]]
Quoted 8 lines
Quoted 5 lines
Isn't it rather a
choice of five than an enumeration of up to 5? In which case I'd still prefer 2 (as then there's no need for a second separator beside comma), the more that we have at least one example with such sub-options (cpufreq).OK, I can do the nested iommu option inside of dom0 if that's the preference.
Well, no, then let's go with 1 (with the dash) I would say.
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 232 lines
-----Original Message-----
From: Jan Beulich [mailto:[email protected]]
Sent: 31 July 2018 10:42
To: Roger Pau Monne <[email protected]>
Cc: Julien Grall <[email protected]>; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Paul Durrant <[email protected]>; Wei Liu <[email protected]>; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; xen-devel <[email protected]>; Tim (Xen.org) <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 10 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 11:34, <[email protected]> wrote:
Quoted 8 lines · Jan Beulich
Quoted 8 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 11:05, <[email protected]> wrote:
Quoted 6 lines · Jan Beulich
Quoted 6 lines
Quoted 6 lines
Quoted 1 line · On 31.07.18 at 10:37, <[email protected]> wrote:
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:34
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin TianQuoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Stefano Stabellini <[email protected]>; Wei Liu
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson
<[email protected]>; Tim
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
<[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 5 lines · Paul Durrant
Quoted 5 lines
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:16
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin TianQuoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 1 line
Quoted 1 line
Stefano Stabellini <[email protected]>; Wei Liu
Quoted 2 lines
Quoted 2 lines
George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>;
Jan
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Beulich
Quoted 7 lines
Quoted 7 lines
<[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 2 lines · Paul Durrant
Quoted 2 lines
-----Original Message-----
From: Xen-devel [mailto:xen-devel-Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
On
Quoted 6 lines
Quoted 6 lines
Behalf
Quoted 5 lines
Quoted 5 lines
Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>;George
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Dunlap
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
<[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson
Quoted 2 lines
Quoted 2 lines
Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall
<[email protected]>; Jan
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Beulich
Quoted 1 line
Quoted 1 line
<[email protected]>; Roger Pau Monne
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Subject: [Xen-devel] [PATCH 2/4] iommu: generalize
iommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that
supersedes
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
iommu_inclusive_mapping. This should be a non-functionalchange
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
on
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Intel hardware, while AMD hardware will gain the same
functionality
Quoted 16 lines
Quoted 16 lines
Quoted 16 lines
Quoted 16 lines
Quoted 16 lines
Quoted 16 lines
of
Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
mapping almost everything below the 4GB boundary.
Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Cc Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Review
Quoted 1 line
(quoted diff) +0
-0b/docs/misc/xen-
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
docs/misc/xen-command-line.markdown detection of systems known to
Quoted 5 lines
command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known tomisbehave
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
upon
Quoted 7 lines
Quoted 5 lines
accesses to that port.
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies `verbose`).
+>
inclusiveThis is a dom0 (or hwdom) specific setting so perhaps dom0-
inclusive?
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Actually the dom0 iommu options are starting to get unwieldy
as they
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
are
Quoted 1 line
Quoted 1 line
conflated with the general host iommu options so I think it may
be
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
worthwhile splitting things out into a separate 'dom0-iommu='
top level
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
parameter at this stage. (My reasons are slightly selfish as I
intend to
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
add
Quoted 1 line
Quoted 1 line
Quoted 1 line
another dom0 iommu option to give it just reserved regions, to
avoid
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
unnecessary set-up if we know it will be using PV-IOMMU).
Mapping just the reserved regions is what I actually do for PVH
with
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
iommu=inclusive (patch 4/4), so maybe it would make sense to
speak
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
about
Quoted 4 lines
Quoted 4 lines
the
naming here in order to use the same naming for PV and PVH.TBH I don't really like the dom0- prefix, the command line
iommu
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 2 lines
options either apply to all domains or Dom0 only, having domu-inclusive for example makes no sense IMO.
No, I think there are some options that you may want to apply to
dom0
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
only, but these are more like the
dom0_memordom0_max_vpusoptions.
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Particularly, the inclusive option is probably something that is only
desirable
Quoted 1 line
Quoted 1 line
Quoted 1 line
for dom0. Clearly dom0-passthrough and dom0-strict are already
defined to
Quoted 4 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
relate to dom0 only, and options such as 'reserved' should only be
specific
Quoted 2 lines
Quoted 2 lines
on
Quoted 1 line
Quoted 1 line
Quoted 1 line
the command line in relation to dom0 IMO. For other domains such
an option
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
should be specified via xl.cfg.
Yes, we already have a bunch of those, so then I think dom0-
inclusive
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
and dom0-reserved would be appropriate?
dom0-inclusive-mapping or dom0-reserved-mapping seems too
long.
Quoted 16 lines
Quoted 16 lines
Quoted 16 lines
Quoted 15 lines
Quoted 9 lines
Yes, those names are ok, but I still think it better in the long run if we
have something like:
dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
where relaxed is the default and 'none' (I think) is equivalent to the current iommu=dom0-passthrough.
Or, along the lines of the other reply just sent, e.g.
dom0=pvh,iommu:inclusive;reserved,shadow
But perhaps the difference between , and ; gets too confusing then.
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
So I think we have the following options:
1. dom0_iommu=[inclusive,][reserved,][strict,][none,][relaxed]
Nit: dom0-iommu= (no underscores in new options)
Quoted 1 line
2.
dom0=[pvh,][shadow,][iommu=[inclusive;][reserved;][strict;][none;][relaxe d]]
Quoted 1 line
Quoted 1 line
Quoted 1 line
3. dom0=[pvh,][shadow,][iommu-inclusive,][iommu-reserved,][iommu-
strict,][iommu-none,][iommu-relaxed]
Quoted 5 lines
Quoted 5 lines
Quoted 4 lines
I don't have a strong preference between 1 and 3, but I would prefer to avoid 2 because I think suboptions inside of options it's too complex IMO.
Quoted 8 lines
Quoted 3 lines
While generally I prefer to limit the number of top level options, in this case I think I'd prefer 1 after all. Or wait - does any pair of the (sub)options actually make sense to be specified?
Yes, for example you can use strict and inclusive at the same time, I think it's something like:
dom0=[pvh,][shadow,][iommu=[inclusive|reserved;][strict|none|relaxed]]
Quoted 7 lines
Quoted 6 lines
Isn't it rather a
choice of five than an enumeration of up to 5? In which case I'd still prefer 2 (as then there's no need for a second separator beside comma), the more that we have at least one example with such sub-options (cpufreq).
Quoted 5 lines
Quoted 2 lines
OK, I can do the nested iommu option inside of dom0 if that's the preference.
Well, no, then let's go with 1 (with the dash) I would say.
+1
Paul
Quoted 1 line
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
In reply to Paul DurrantQuoted 94 lines
Quoted 85 lines
Quoted 1 line · On 31.07.18 at 10:27, <[email protected]> wrote:
-----Original Message-----
From: Roger Pau Monne
Sent: 31 July 2018 09:16
To: Paul Durrant <[email protected]>
Cc: [email protected]; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 2 lines · Paul Durrant
Quoted 2 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]] OnBehalf
Quoted 7 lines
Quoted 7 lines
Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>;Tim
Quoted 1 line
Quoted 1 line
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan
Beulich
Quoted 35 lines
Quoted 35 lines
<[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingIntroduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.Note that is a noop for ARM hardware.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Kevin Tian <[email protected]>--- docs/misc/xen-command-line.markdown | 14 ++++++ xen/drivers/passthrough/arm/iommu.c | 4 ++ xen/drivers/passthrough/iommu.c | 6 +++ xen/drivers/passthrough/vtd/extern.h | 2 - xen/drivers/passthrough/vtd/iommu.c | 6 --- xen/drivers/passthrough/vtd/x86/vtd.c | 66 +------------------------ xen/drivers/passthrough/x86/iommu.c | 70 +++++++++++++++++++++++++++ xen/include/xen/iommu.h | 2 + 8 files changed, 97 insertions(+), 73 deletions(-)Quoted 6 lines
docs/misc/xen-command-line.markdown detection of systems known to misbehave
Quoted 6 lines
docs/misc/xen- renamed from docs/misc/xen-command-line.markdown +0
-0command-line.markdown
index65b4754418..91a8bfc9a6100644docs/misc/xen-command-line.markdown +0
-0@@ -1198,6 +1198,17 @@ detection of systems known to misbehaveupon
Quoted 9 lines
Quoted 5 lines
accesses to that port.
Quoted 1 line
Quoted 1 line
Enable IOMMU debugging code (implies
verbose).+>
inclusiveThis is a dom0 (or hwdom) specific setting so perhaps dom0-inclusive?
Actually the dom0 iommu options are starting to get unwieldy as they are
conflated with the general host iommu options so I think it may be worthwhile splitting things out into a separate 'dom0-iommu=' top level parameter at this stage. (My reasons are slightly selfish as I intend to add another dom0 iommu option to give it just reserved regions, to avoid unnecessary set-up if we know it will be using PV-IOMMU).
Quoted 8 lines
Mapping just the reserved regions is what I actually do for PVH with iommu=inclusive (patch 4/4), so maybe it would make sense to speak about the
naming here in order to use the same naming for PV and PVH.TBH I don't really like the dom0- prefix, the command line iommu options either apply to all domains or Dom0 only, having domu-inclusive for example makes no sense IMO.
Quoted 7 lines
No, I think there are some options that you may want to apply to dom0 only, but these are more like the
dom0_memordom0_max_vpusoptions. Particularly, the inclusive option is probably something that is only desirable for dom0. Clearly dom0-passthrough and dom0-strict are already defined to relate to dom0 only, and options such as 'reserved' should only be specific on the command line in relation to dom0 IMO. For other domains such an option should be specified via xl.cfg.
So perhaps "dom0=iommu-inclusive" etc?
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
In reply to Roger Pau MonneQuoted 5 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
Introduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.
So first of all - what's the motivation behind this change? So far we had no need for hacks line the VT-d side one on AMD. I don't think this should be widened without there being indication of a problem with non-niche AMD systems.
xen/drivers/passthrough/x86/iommu.c line 23
Quoted 7 lines
xen/drivers/passthrough/x86/iommu.c +1
-0@@ -20,6 +20,8 @@#include <xen/softirq.h> #include <xsm/xsm.h> +#include <asm/setup.h>
Why? And if it's needed here now, can it be dropped from the VT-d file where you removed the code?
Quoted 50 lines
xen/drivers/passthrough/x86/iommu.c +46
-0@@ -132,6 +134,74 @@ void arch_iommu_domain_destroy(struct domain *d) { } +void __hwdom_init arch_iommu_hwdom_init(struct domain *d) +{ +unsigned long i, j, tmp, top, max_pfn; + +if ( iommu_passthrough || !is_pv_domain (d) ) +return ; + +BUG_ON (!is_hardware_domain (d)); + + max_pfn = (GB (4 ) >>PAGE_SHIFT ) -1 ; + top =max (max_pdx,pfn_to_pdx (max_pfn) +1 ); + +for ( i =0 ; i < top; i++ ) + { +unsigned long pfn =pdx_to_pfn (i); +bool map; +int rc =0 ; + +/* +* Set up 1:1 mapping for dom0. Default to include only +* conventional RAM areas and let RMRRs include needed reserved +* regions. When set, the inclusive mapping additionally maps in +* every pfn up to 4GB except those that fall in unusable ranges. +*/ +if ( pfn > max_pfn && !mfn_valid (_mfn (pfn)) ) +continue ; + +if ( iommu_inclusive && pfn <= max_pfn ) + map = !page_is_ram_type (pfn,RAM_TYPE_UNUSABLE ); +else + map =page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ); + +if ( !map ) +continue ; + +/* Exclude Xen bits */ +if (xen_in_range (pfn) ) +continue ; + +/* +* If dom0-strict mode is enabled then exclude conventional RAM +* and let the common code map dom0's pages. +*/ +if ( iommu_dom0_strict && +page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ) ) +continue ;
xen/drivers/passthrough/x86/iommu.c void arch_iommu_domain_destroy(struct domain *d) · line 192
Quoted 10 lines
xen/drivers/passthrough/x86/iommu.c +10
-0+ + tmp =1 << (PAGE_SHIFT -PAGE_SHIFT_4K ); +for ( j =0 ; j < tmp; j++ ) + { +int ret =iommu_map_page (d, pfn * tmp + j, pfn * tmp + j, + IOMMUF_readable|IOMMUF_writable); + +if ( !rc ) + rc = ret; + }
To VT-d specific code was this way to also cope with ia64. I don't see the need for this to be a loop when the code is now x86- specific.
xen/drivers/passthrough/x86/iommu.c void arch_iommu_domain_destroy(struct domain *d) · line 198
Quoted 5 lines
xen/drivers/passthrough/x86/iommu.c +5
-0+if ( rc ) +printk (XENLOG_WARNING "d%d: IOMMU mapping failed: %d\n" , + d->domain_id, rc); + +if (!(i & (0xfffff >> (PAGE_SHIFT -PAGE_SHIFT_4K ))))
Same here for the unnecessary shifting.
xen/drivers/passthrough/x86/iommu.c void arch_iommu_domain_destroy(struct domain *d) · line 203
Quoted 5 lines
xen/drivers/passthrough/x86/iommu.c +5
-0+process_pending_softirqs (); + } + + +}
Double blank line above here when there should be none at all.
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 10 lines · Jan Beulich
Quoted 5 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
Introduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.So first of all - what's the motivation behind this change? So far we had no need for hacks line the VT-d side one on AMD. I don't think this should be widened without there being indication of a problem with non-niche AMD systems.
OK, I can leave the default on for Intel and off for everything else, but I will introduce the generic dom0-iommu= option anyway.
Quoted 10 lines
xen/drivers/passthrough/x86/iommu.c line 23
Quoted 7 lines
xen/drivers/passthrough/x86/iommu.c +1
-0@@ -20,6 +20,8 @@#include <xen/softirq.h> #include <xsm/xsm.h> +#include <asm/setup.h> Why? And if it's needed here now, can it be dropped from the VT-d file where you removed the code?
setup.h is needed for xen_in_range. And yes, I can drop it from the VTd code then.
Quoted 56 lines
Quoted 56 lines
xen/drivers/passthrough/x86/iommu.c +52
-0@@ -132,6 +134,74 @@ void arch_iommu_domain_destroy(struct domain *d) { } +void __hwdom_init arch_iommu_hwdom_init(struct domain *d) +{ +unsigned long i, j, tmp, top, max_pfn; + +if ( iommu_passthrough || !is_pv_domain (d) ) +return ; + +BUG_ON (!is_hardware_domain (d)); + + max_pfn = (GB (4 ) >>PAGE_SHIFT ) -1 ; + top =max (max_pdx,pfn_to_pdx (max_pfn) +1 ); + +for ( i =0 ; i < top; i++ ) + { +unsigned long pfn =pdx_to_pfn (i); +bool map; +int rc =0 ; + +/* +* Set up 1:1 mapping for dom0. Default to include only +* conventional RAM areas and let RMRRs include needed reserved +* regions. When set, the inclusive mapping additionally maps in +* every pfn up to 4GB except those that fall in unusable ranges. +*/ +if ( pfn > max_pfn && !mfn_valid (_mfn (pfn)) ) +continue ; + +if ( iommu_inclusive && pfn <= max_pfn ) + map = !page_is_ram_type (pfn,RAM_TYPE_UNUSABLE ); +else + map =page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ); + +if ( !map ) +continue ; + +/* Exclude Xen bits */ +if (xen_in_range (pfn) ) +continue ; + +/* +* If dom0-strict mode is enabled then exclude conventional RAM +* and let the common code map dom0's pages. +*/ +if ( iommu_dom0_strict && +page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ) ) +continue ; + + tmp =1 << (PAGE_SHIFT -PAGE_SHIFT_4K ); +for ( j =0 ; j < tmp; j++ ) + { +int ret =iommu_map_page (d, pfn * tmp + j, pfn * tmp + j, + IOMMUF_readable|IOMMUF_writable);
Quoted 8 lines
xen/drivers/passthrough/x86/iommu.c void arch_iommu_domain_destroy(struct domain *d) · line 192
Quoted 4 lines
xen/drivers/passthrough/x86/iommu.c +4
-0+ +if ( !rc ) + rc = ret; + }To VT-d specific code was this way to also cope with ia64. I don't see the need for this to be a loop when the code is now x86- specific.
Oh, I wondered about this and TBH I couldn't figure out why it's this way, now I understand.
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 8 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 17:33, <[email protected]> wrote:
Quoted 6 lines · Jan Beulich
Quoted 5 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
Introduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.
Quoted 7 lines
Quoted 4 lines
So first of all - what's the motivation behind this change? So far we had no need for hacks line the VT-d side one on AMD. I don't think this should be widened without there being indication of a problem with non-niche AMD systems.
OK, I can leave the default on for Intel and off for everything else, but I will introduce the generic dom0-iommu= option anyway.
Hmm, I've always been wishing we'd change to a default of off for VT-d as well - imo we shouldn't by default assume broken firmware. Kevin?
As to AMD - you still don't really say why this would be needed there. I'm not in favor of workarounds when there's nothing to work around. IOW - if the logic isn't needed on AMD, do we need this code movement in the first place?
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 13 lines · Jan Beulich
Quoted 7 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 17:33, <[email protected]> wrote:
Quoted 5 lines · Jan Beulich
Quoted 5 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
Introduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.Quoted 6 lines
Quoted 4 lines
So first of all - what's the motivation behind this change? So far we had no need for hacks line the VT-d side one on AMD. I don't think this should be widened without there being indication of a problem with non-niche AMD systems.
OK, I can leave the default on for Intel and off for everything else, but I will introduce the generic dom0-iommu= option anyway.
Quoted 8 lines
Hmm, I've always been wishing we'd change to a default of off for VT-d as well - imo we shouldn't by default assume broken firmware. Kevin?
As to AMD - you still don't really say why this would be needed there. I'm not in favor of workarounds when there's nothing to work around. IOW - if the logic isn't needed on AMD, do we need this code movement in the first place?
This is sufficiently broken on AMD as well. Take a look at all the IVRS/IVHT reports we've had on xen-devel.
Furthermore, on by default is the only reasonable option. Dom0 is specifically given cpu mappings of the reserved regions, therefore it should have matching IOMMU mappings.
~Andrew
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 19 lines
Quoted 1 line
Quoted 1 line · On 01.08.18 at 10:32, <[email protected]> wrote:
Quoted 17 lines · Jan Beulich
Quoted 7 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 17:33, <[email protected]> wrote:
Quoted 5 lines · Jan Beulich
Quoted 5 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
Introduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.Quoted 6 lines
Quoted 4 lines
So first of all - what's the motivation behind this change? So far we had no need for hacks line the VT-d side one on AMD. I don't think this should be widened without there being indication of a problem with non-niche AMD systems.
OK, I can leave the default on for Intel and off for everything else, but I will introduce the generic dom0-iommu= option anyway.
Hmm, I've always been wishing we'd change to a default of off for VT-d as well - imo we shouldn't by default assume broken firmware. Kevin?
Quoted 7 lines
Quoted 4 lines
As to AMD - you still don't really say why this would be needed there. I'm not in favor of workarounds when there's nothing to work around. IOW - if the logic isn't needed on AMD, do we need this code movement in the first place?
This is sufficiently broken on AMD as well. Take a look at all the IVRS/IVHT reports we've had on xen-devel.
We've had IVRS issues, yes, and we now have command line options to deal with them. That's entirely unrelated to memory ranges though.
Quoted 3 lines
Furthermore, on by default is the only reasonable option. Dom0 is specifically given cpu mappings of the reserved regions, therefore it should have matching IOMMU mappings.
I don't think I can uniformly agree with this: There are reserved regions which specifically would better not have IOMMU mappings:
LAPIC, IO-APIC, MMCFG just to name things I can immediately think of.
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 7 lines · Jan Beulich
Quoted 3 lines
Furthermore, on by default is the only reasonable option. Dom0 is specifically given cpu mappings of the reserved regions, therefore it should have matching IOMMU mappings.
I don't think I can uniformly agree with this: There are reserved regions which specifically would better not have IOMMU mappings:
LAPIC, IO-APIC, MMCFG just to name things I can immediately think of.
Right, but that's not what I said.
For any CPU mapping, we should have a matching IOMMU mapping. This is how hardware behaves in a native system, and is how all operating systems are written to behave. Breaking this expectation will break software in unexpected ways.
For the items you list there, we don't give dom0 CPU mappings (because restricting them is indeed a sensible thing), and the "matching" mapping in the IOMMU is also not present.
~Andrew
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 10 lines
Quoted 1 line
Quoted 1 line · On 01.08.18 at 11:20, <[email protected]> wrote:
Quoted 7 lines · Jan Beulich
Quoted 3 lines
Furthermore, on by default is the only reasonable option. Dom0 is specifically given cpu mappings of the reserved regions, therefore it should have matching IOMMU mappings.
I don't think I can uniformly agree with this: There are reserved regions which specifically would better not have IOMMU mappings:
LAPIC, IO-APIC, MMCFG just to name things I can immediately think of.
Quoted 6 lines
Right, but that's not what I said.
For any CPU mapping, we should have a matching IOMMU mapping. This is how hardware behaves in a native system, and is how all operating systems are written to behave. Breaking this expectation will break software in unexpected ways.
Well, LAPIC is a clear exception to this rule, as from devices' point of view the 0xFEExxxxx range has a completely different meaning.
Quoted 3 lines
For the items you list there, we don't give dom0 CPU mappings (because restricting them is indeed a sensible thing), and the "matching" mapping in the IOMMU is also not present.
Not exactly: We still emulate those ranges for CPU side accesses, while we don't emulate IOMMU / device side ones. So there's a difference to native hardware anyway.
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 9 lines · Jan Beulich
Quoted 9 lines
Quoted 1 line
Quoted 1 line · On 01.08.18 at 11:20, <[email protected]> wrote:
Quoted 7 lines · Jan Beulich
Quoted 3 lines
Furthermore, on by default is the only reasonable option. Dom0 is specifically given cpu mappings of the reserved regions, therefore it should have matching IOMMU mappings.
I don't think I can uniformly agree with this: There are reserved regions which specifically would better not have IOMMU mappings:
LAPIC, IO-APIC, MMCFG just to name things I can immediately think of.
Quoted 8 lines
Quoted 6 lines
Right, but that's not what I said.
For any CPU mapping, we should have a matching IOMMU mapping. This is how hardware behaves in a native system, and is how all operating systems are written to behave. Breaking this expectation will break software in unexpected ways.
Well, LAPIC is a clear exception to this rule, as from devices' point of view the 0xFEExxxxx range has a completely different meaning.
And furthermore is magic anyway when it comes to DMA's hitting that region.
Quoted 7 lines
Quoted 4 lines
For the items you list there, we don't give dom0 CPU mappings (because restricting them is indeed a sensible thing), and the "matching" mapping in the IOMMU is also not present.
Not exactly: We still emulate those ranges for CPU side accesses, while we don't emulate IOMMU / device side ones. So there's a difference to native hardware anyway.
We can't emulate the IOMMU side at all, nor would we want to in the context of disallowing dom0 access to those bits of real hardware.
~Andrew
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
In reply to Jan BeulichQuoted 27 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]] On Behalf Of Jan Beulich
Sent: 01 August 2018 09:21
To: Roger Pau Monne <[email protected]>
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; xen-devel <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 13 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 17:33, <[email protected]> wrote:
Quoted 10 lines · Jan Beulich
Quoted 5 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
Introduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.So first of all - what's the motivation behind this change? So far we had no need for hacks line the VT-d side one on AMD. I don't think this should be widened without there being indication of a problem with non-niche AMD systems.
Quoted 6 lines
Quoted 2 lines
OK, I can leave the default on for Intel and off for everything else, but I will introduce the generic dom0-iommu= option anyway.
Hmm, I've always been wishing we'd change to a default of off for VT-d as well - imo we shouldn't by default assume broken firmware. Kevin?
I suspect that a seriously large number of Xen users will find their systems fail to boot if the default is changed. I'm testing on a Dell R730 with (to my knowledge) up-to-date firmware. Not exactly a rare system and turning off inclusive mapping will cause it to wedge during boot.
Paul
Quoted 6 lines
As to AMD - you still don't really say why this would be needed there. I'm not in favor of workarounds when there's nothing to work around. IOW - if the logic isn't needed on AMD, do we need this code movement in the first place?
Quoted 8 lines
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
Quoted 20 lines
Quoted 20 lines
Quoted 1 line · On 01.08.18 at 10:33, <[email protected]> wrote:
From: Xen-devel [mailto:[email protected]] On Behalf Of Jan Beulich
Sent: 01 August 2018 09:21Quoted 15 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 17:33, <[email protected]> wrote:
Quoted 10 lines · Jan Beulich
Quoted 5 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
Introduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.So first of all - what's the motivation behind this change? So far we had no need for hacks line the VT-d side one on AMD. I don't think this should be widened without there being indication of a problem with non-niche AMD systems.
OK, I can leave the default on for Intel and off for everything else, but I will introduce the generic dom0-iommu= option anyway.
Quoted 8 lines
Quoted 3 lines
Hmm, I've always been wishing we'd change to a default of off for VT-d as well - imo we shouldn't by default assume broken firmware. Kevin?
I suspect that a seriously large number of Xen users will find their systems fail to boot if the default is changed. I'm testing on a Dell R730 with (to my knowledge) up-to-date firmware. Not exactly a rare system and turning off inclusive mapping will cause it to wedge during boot.
Hmm, that's rather sad indeed.
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
In reply to Paul DurrantQuoted 41 lines
From: Paul Durrant [mailto:[email protected]]
Sent: Wednesday, August 1, 2018 4:34 PMQuoted 2 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]] OnBehalf
Quoted 7 lines
Of Jan Beulich
Sent: 01 August 2018 09:21
To: Roger Pau Monne <[email protected]>
Cc: Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>;Tim
Quoted 10 lines
(Xen.org) <[email protected]>; Julien Grall <[email protected]>; xen-devel <[email protected]>
Subject: Re: [Xen-devel] [PATCH 2/4] iommu: generalizeiommu_inclusive_mappingQuoted 5 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 17:33, <[email protected]> wrote:
Quoted 3 lines · Jan Beulich
Quoted 3 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
Introduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional changeon
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Intel hardware, while AMD hardware will gain the same functionality
of
Quoted 7 lines
Quoted 7 lines
Quoted 6 lines
Quoted 1 line
mapping almost everything below the 4GB boundary.
So first of all - what's the motivation behind this change? So far we had no need for hacks line the VT-d side one on AMD. I don't think this should be widened without there being indication of a problem with non-niche AMD systems.
Quoted 6 lines
Quoted 2 lines
OK, I can leave the default on for Intel and off for everything else, but I will introduce the generic dom0-iommu= option anyway.
Hmm, I've always been wishing we'd change to a default of off for VT-d as well - imo we shouldn't by default assume broken firmware. Kevin?
Quoted 6 lines
I suspect that a seriously large number of Xen users will find their systems fail to boot if the default is changed. I'm testing on a Dell R730 with (to my knowledge) up-to-date firmware. Not exactly a rare system and turning off inclusive mapping will cause it to wedge during boot.
Paul
Though I agree with Jan in concept, the reality means we have to keep default on as before. :-)
Thanks
Kevin
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 2/4] iommu: generalize iommu_inclusive_mapping
In reply to Jan BeulichQuoted 13 lines · Jan Beulich
Quoted 13 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 17:33, <[email protected]> wrote:
Quoted 10 lines · Jan Beulich
Quoted 5 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
Introduce a new iommu=inclusive generic option that supersedes
iommu_inclusive_mapping. This should be a non-functional change on Intel hardware, while AMD hardware will gain the same functionality of mapping almost everything below the 4GB boundary.So first of all - what's the motivation behind this change? So far we had no need for hacks line the VT-d side one on AMD. I don't think this should be widened without there being indication of a problem with non-niche AMD systems.
Quoted 6 lines
Quoted 2 lines
OK, I can leave the default on for Intel and off for everything else, but I will introduce the generic dom0-iommu= option anyway.
Hmm, I've always been wishing we'd change to a default of off for VT-d as well - imo we shouldn't by default assume broken firmware. Kevin?
I'm all for it, but even in newish systems (as recently reported by people trying PVH Dom0) RMRR entries are missing and will make some devices unusable.
Quoted 4 lines
As to AMD - you still don't really say why this would be needed there. I'm not in favor of workarounds when there's nothing to work around. IOW - if the logic isn't needed on AMD, do we need this code movement in the first place?
There's nothing VTd specific in this code at all, I could see it living in a VTd specific file if it had quirks specific to VTd, but I think the code is generic enough that it warrants being shared, so that people running on AMD hw can also enable it if required.
As requested I've left this off for AMD and on by default on Intel, so the default functionality is not changed at all.
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
PATCH 3/4 x86/iommu: reorder conditions used in the inclusive iommu mappings
In reply to Roger Pau MonneIn order to place all the map conditions in a single if ... else conditional.
No functional change.
---
Cc: Jan Beulich <[email protected]>
--- xen/drivers/passthrough/x86/iommu.c | 23 ++++++++++------------- 1 file changed, 10 insertions(+), 13 deletions(-)
xen/drivers/passthrough/x86/iommu.c +10 -13
@@ -158,19 +158,9 @@ void __hwdom_init arch_iommu_hwdom_init(struct domain *d)* regions. When set, the inclusive mapping additionally maps in * every pfn up to 4GB except those that fall in unusable ranges. */ -if ( pfn > max_pfn && !mfn_valid (_mfn (pfn)) ) -continue ; - -if ( iommu_inclusive && pfn <= max_pfn ) - map = !page_is_ram_type (pfn,RAM_TYPE_UNUSABLE ); -else - map =page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ); - -if ( !map ) -continue ; - -/* Exclude Xen bits */ -if (xen_in_range (pfn) ) +if ( (pfn > max_pfn && !mfn_valid (_mfn (pfn))) || +/* Exclude Xen bits */ +xen_in_range (pfn) )continue ;/*
@@ -179,6 +169,13 @@ void __hwdom_init arch_iommu_hwdom_init(struct domain *d)*/ if ( iommu_dom0_strict &&page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ) ) + map = false; +else if ( iommu_inclusive && pfn <= max_pfn ) + map = !page_is_ram_type (pfn,RAM_TYPE_UNUSABLE ); +else + map =page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ); + +if ( !map )continue ; tmp =1 << (PAGE_SHIFT -PAGE_SHIFT_4K );
Signature · 7 lines
2.18.0 _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 3/4] x86/iommu: reorder conditions used in the inclusive iommu mappings
Quoted 52 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]] On Behalf Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Jan Beulich <[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 3/4] x86/iommu: reorder conditions used in the inclusive iommu mappingsIn order to place all the map conditions in a single if ... else conditional.
No functional change.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Jan Beulich <[email protected]>--- xen/drivers/passthrough/x86/iommu.c | 23 ++++++++++------------- 1 file changed, 10 insertions(+), 13 deletions(-)xen/drivers/passthrough/x86/iommu.c +3
-13@@ -158,19 +158,9 @@ void __hwdom_init arch_iommu_hwdom_init(structdomain *d) * regions. When set, the inclusive mapping additionally maps in * every pfn up to 4GB except those that fall in unusable ranges. */ -if ( pfn > max_pfn && !mfn_valid (_mfn (pfn)) ) -continue ; - -if ( iommu_inclusive && pfn <= max_pfn ) - map = !page_is_ram_type (pfn,RAM_TYPE_UNUSABLE ); -else - map =page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ); - -if ( !map ) -continue ; - -/* Exclude Xen bits */ -if (xen_in_range (pfn) ) +if ( (pfn > max_pfn && !mfn_valid (_mfn (pfn))) || +/* Exclude Xen bits */ +xen_in_range (pfn) )continue ;/*
xen/drivers/passthrough/x86/iommu.c void __hwdom_init arch_iommu_hwdom_init(struct · line 178
Quoted 10 lines
xen/drivers/passthrough/x86/iommu.c +6
-0@@ -179,6 +169,13 @@ void __hwdom_init arch_iommu_hwdom_init(structdomain *d) */ if ( iommu_dom0_strict &&page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ) ) + map = false; +else if ( iommu_inclusive && pfn <= max_pfn ) + map = !page_is_ram_type (pfn,RAM_TYPE_UNUSABLE ); +else + map =page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ); +
Maybe better as...
If ( page_is_ram_type(pfn, RAM_TYPE_CONVENTIONAL) )
map = !iommu_dom0_strict;
else if ( !page_is_ram_type(pfn, RAM_TYPE_UNUSABLE) )
map = (iommu_inclusive && pfn <= max_pfn);
else
map = false;
(I think that logic is correct).
Paul
Quoted 4 lines
xen/drivers/passthrough/x86/iommu.c +1
-0+if ( !map )continue ;tmp = 1 << (PAGE_SHIFT - PAGE_SHIFT_4K);
Quoted 8 lines
--
2.18.0_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 3/4] x86/iommu: reorder conditions used in the inclusive iommu mappings
Quoted 16 lines · Paul Durrant
xen/drivers/passthrough/x86/iommu.c void __hwdom_init arch_iommu_hwdom_init(struct · line 178
Quoted 13 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]] On Behalfxen/drivers/passthrough/x86/iommu.c +6
-0@@ -179,6 +169,13 @@ void __hwdom_init arch_iommu_hwdom_init(structdomain *d) */ if ( iommu_dom0_strict &&page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ) ) + map = false; +else if ( iommu_inclusive && pfn <= max_pfn ) + map = !page_is_ram_type (pfn,RAM_TYPE_UNUSABLE ); +else + map =page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ); +Maybe better as...
Quoted 8 lines
If (
page_is_ram_type(pfn, RAM_TYPE_CONVENTIONAL))
map = !iommu_dom0_strict;
else if ( !page_is_ram_type(pfn,RAM_TYPE_UNUSABLE) )
map = (iommu_inclusive&& pfn <=max_pfn);
else
map = false;(I think that logic is correct).
Yes I think it's correct. Should have looked more closely when moving this code around.
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
PATCH 4/4 x86/iommu: add PVH support to the inclusive options
In reply to Roger Pau MonneSeveral people have reported hardware issues (malfunctioning USB controllers) due to iommu page faults. Those faults are caused by missing RMRR (VTd) or IRVS (AMD-Vi) entries in the ACPI tables. Those can be worked around on VTd hardware by manually adding RMRR entries on the command line, this is however limited to Intel hardware and quite cumbersome to do.
In order to solve those issues add PVH support to the inclusive option that identity maps all regions marked as reserved in the memory map. Note that regions used by devices emulated by Xen (LAPIC, IO-APIC or PCIe MCFG regions) are specifically avoided. Note that this option currently relies on no MSIX MMIO areas residing in a reserved region, or else Xen won't be able to trap those accesses.
---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
--- docs/misc/xen-command-line.markdown | 16 ++++-- xen/drivers/passthrough/x86/iommu.c | 82 +++++++++++++++++++++++------ 2 files changed, 77 insertions(+), 21 deletions(-)
docs/misc/xen-command-line.markdown +11 -5
@@ -1203,11 +1203,17 @@ detection of systems known to misbehave upon accesses to that port. > Default: `true` >> Use this to work around firmware issues providing incorrect RMRR or IVMD ->> entries. Rather than only mapping RAM pages for IOMMU accesses for Dom0, ->> with this option all pages up to 4GB, not marked as unusable in the E820 ->> table, will get a mapping established. Note that this option is only ->> applicable to a PV dom0. Also note that if `dom0-strict` mode is enabled ->> then conventional RAM pages not assigned to dom0 will not be mapped. +>> entries. The behaviour of this option is slightly different between a PV and +>> a PVH Dom0: +>> +>> * For a PV Dom0 all pages up to 4GB not marked as unusable in the memory +>> map will get a mapping established. Note that if `dom0-strict` mode is +>> enabled then conventional RAM pages not assigned to dom0 will not be +>> mapped. +>> +>> * For a PVH Dom0 all memory regions marked as reserved in the memory map +>> that don't overlap with any MMIO region from emulated devices will be +>> identity mapped. ### iommu\_dev\_iotlb\_timeout > `= <integer>`
xen/drivers/passthrough/x86/iommu.c +66 -16
115 lines not on this page · raw message
Signature · 7 lines
2.18.0 _______________________________________________ Xen-devel mailing list [email protected] https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 4/4] x86/iommu: add PVH support to the inclusive options
Quoted 94 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]] On Behalf Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 4/4] x86/iommu: add PVH support to the inclusive optionsSeveral people have reported hardware issues (malfunctioning USB controllers) due to iommu page faults. Those faults are caused by missing RMRR (VTd) or IRVS (AMD-Vi) entries in the ACPI tables. Those can be worked around on VTd hardware by manually adding RMRR entries on the command line, this is however limited to Intel hardware and quite cumbersome to do.
In order to solve those issues add PVH support to the inclusive option that identity maps all regions marked as reserved in the memory map. Note that regions used by devices emulated by Xen (LAPIC, IO-APIC or PCIe MCFG regions) are specifically avoided. Note that this option currently relies on no MSIX MMIO areas residing in a reserved region, or else Xen won't be able to trap those accesses.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>--- docs/misc/xen-command-line.markdown | 16 ++++-- xen/drivers/passthrough/x86/iommu.c | 82 +++++++++++++++++++++++-- ---- 2 files changed, 77 insertions(+), 21 deletions(-)docs/misc/xen- renamed from docs/misc/xen-command-line.markdown +0
-0command-line.markdown
index91a8bfc9a6..c7c9a38c19100644docs/misc/xen-command-line.markdown +0
-0@@ -1203,11 +1203,17 @@ detection of systems known to misbehave uponaccesses to that port.
Quoted 3 lines
Default:
trueQuoted 1 line
Use this to work around firmware issues providing incorrect RMRR or
IVMD
->> entries. Rather than only mapping RAM pages for IOMMU accesses for Dom0,
->> with this option all pages up to 4GB, not marked as unusable in the E820 ->> table, will get a mapping established. Note that this option is only ->> applicable to a PV dom0. Also note that ifdom0-strictmode is enabled ->> then conventional RAM pages not assigned to dom0 will not be mapped. +>> entries. The behaviour of this option is slightly different between a PV and
+>> a PVH Dom0:
+>>
+>> * For a PV Dom0 all pages up to 4GB not marked as unusable in the memory+>> map will get a mapping established. Note that if `dom0-strict` mode is +>> enabled then conventional RAM pages not assigned to dom0 will not be +>> mapped.+>>
+>> * For a PVH Dom0 all memory regions marked as reserved in the memory map
+>> that don't overlap with any MMIO region from emulated devices will be
+>> identity mapped.### iommu\_dev\_iotlb\_timeout
Quoted 1 line
= <integer>xen/drivers/passthrough/x86/iommu.c +2
-0@@ -20,6 +20,8 @@#include <xen/softirq.h> #include <xsm/xsm.h> +#include <asm/apicdef.h> +#include <asm/io_apic.h> #include <asm/setup.h> void iommu_update_ire_from_apic(
xen/drivers/passthrough/x86/iommu.c void arch_iommu_domain_destroy(struct domain · line 141
Quoted 6 lines
xen/drivers/passthrough/x86/iommu.c +2
-0@@ -134,11 +136,62 @@ void arch_iommu_domain_destroy(struct domain *d) { } +static bool __hwdom_init pv_inclusive_map(unsigned long pfn, +unsigned long max_pfn)
Perhaps pv_hwdom_inclusive_map() (and similarly pvh_hwdom_inclusive_map()) to make it obvious that they are intended only for the hardware domain. (I know the annotation makes this reasonably obvious but other hwdom-specific functions seem to carry this in their name).
Paul
Quoted 107 lines
xen/drivers/passthrough/x86/iommu.c +50
-1+{ +/* +* If dom0-strict mode is enabled then exclude conventional RAM +* and let the common code map dom0's pages. +*/ +if ( iommu_dom0_strict &&page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ) ) +return false; +if ( iommu_inclusive && pfn <= max_pfn ) +return !page_is_ram_type (pfn,RAM_TYPE_UNUSABLE ); + +return page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ); +} + +static bool __hwdom_init pvh_inclusive_map(const struct domain *d, +unsigned long pfn) +{ +unsigned int i; + +/* +* Ignore any address below 1MB, that's already identity mapped by the +* domain builder. +*/ +if ( pfn <PFN_DOWN (MB (1 )) ) +return false; + +/* Only add reserved regions. */ +if ( !page_is_ram_type (pfn,RAM_TYPE_RESERVED ) ) +return false; + +/* Check that it doesn't overlap with the LAPIC */ +if ( pfn ==PFN_DOWN (APIC_DEFAULT_PHYS_BASE ) ) +return false; +/* ... or the IO-APIC */ +for ( i =0 ; i < nr_ioapics; i++ ) +if ( pfn ==PFN_DOWN (domain_vioapic (d, i)->base_address) ) +return false; +/* ... or the PCIe MCFG regions. */ +for ( i =0 ; i < pci_mmcfg_config_num; i++ ) + { +unsigned long addr =PFN_DOWN (pci_mmcfg_config[i].address); + +if ( pfn >= addr + (pci_mmcfg_config[i].start_bus_number <<8 ) && + pfn < addr + (pci_mmcfg_config[i].end_bus_number <<8 ) ) +return false; + } + +return true; +} + void __hwdom_init arch_iommu_hwdom_init(struct domain *d) {unsigned long i, j, tmp, top, max_pfn; -if ( iommu_passthrough || !is_pv_domain (d) ) +if ( iommu_passthrough )return ;BUG_ON(!is_hardware_domain(d));xen/drivers/passthrough/x86/iommu.c +0
-1@@ -149,7 +202,6 @@ void __hwdom_init arch_iommu_hwdom_init(struct domain *d)for ( i =0 ; i < top; i++ ) {unsigned long pfn =pdx_to_pfn (i); -bool map;int rc =0 ;/*xen/drivers/passthrough/x86/iommu.c +12
-14@@ -163,25 +215,23 @@ void __hwdom_init arch_iommu_hwdom_init(struct domain *d) xen_in_range(pfn) )continue ; -/* -* If dom0-strict mode is enabled then exclude conventional RAM -* and let the common code map dom0's pages. -*/ -if ( iommu_dom0_strict && -page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ) ) - map = false; -else if ( iommu_inclusive && pfn <= max_pfn ) - map = !page_is_ram_type (pfn,RAM_TYPE_UNUSABLE ); -else - map =page_is_ram_type (pfn,RAM_TYPE_CONVENTIONAL ); - -if ( !map ) +if (is_pv_domain (d) ? !pv_inclusive_map (pfn, max_pfn) + : !pvh_inclusive_map (d, pfn) )continue ; tmp =1 << (PAGE_SHIFT -PAGE_SHIFT_4K );for ( j =0 ; j < tmp; j++ ) { -int ret =iommu_map_page (d, pfn * tmp + j, pfn * tmp + j, +int ret; + +if (iommu_use_hap_pt (d) ) + { +ASSERT (is_hvm_domain (d)); + ret =set_identity_p2m_entry (d, pfn * tmp + j, p2m_access_rw, +0 ); + } +else + ret =iommu_map_page (d, pfn * tmp + j, pfn * tmp + j, IOMMUF_readable|IOMMUF_writable);if ( !rc )
Quoted 8 lines
--
2.18.0_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 4/4] x86/iommu: add PVH support to the inclusive options
Quoted 98 lines · Paul Durrant
Quoted 98 lines
-----Original Message-----
From: Xen-devel [mailto:[email protected]] On Behalf Of Roger Pau Monne
Sent: 27 July 2018 16:32
To: [email protected]
Cc: Stefano Stabellini <[email protected]>; Wei Liu <[email protected]>; George Dunlap <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Tim (Xen.org) <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>; Roger Pau Monne <[email protected]>
Subject: [Xen-devel] [PATCH 4/4] x86/iommu: add PVH support to the inclusive optionsSeveral people have reported hardware issues (malfunctioning USB controllers) due to iommu page faults. Those faults are caused by missing RMRR (VTd) or IRVS (AMD-Vi) entries in the ACPI tables. Those can be worked around on VTd hardware by manually adding RMRR entries on the command line, this is however limited to Intel hardware and quite cumbersome to do.
In order to solve those issues add PVH support to the inclusive option that identity maps all regions marked as reserved in the memory map. Note that regions used by devices emulated by Xen (LAPIC, IO-APIC or PCIe MCFG regions) are specifically avoided. Note that this option currently relies on no MSIX MMIO areas residing in a reserved region, or else Xen won't be able to trap those accesses.
Signed-off-by Roger Pau Monné <[email protected]>---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>--- docs/misc/xen-command-line.markdown | 16 ++++-- xen/drivers/passthrough/x86/iommu.c | 82 +++++++++++++++++++++++-- ---- 2 files changed, 77 insertions(+), 21 deletions(-)docs/misc/xen- renamed from docs/misc/xen-command-line.markdown +0
-0command-line.markdown
index91a8bfc9a6..c7c9a38c19100644docs/misc/xen-command-line.markdown +0
-0@@ -1203,11 +1203,17 @@ detection of systems known to misbehave uponaccesses to that port.
Quoted 3 lines
Default:
trueQuoted 1 line
Use this to work around firmware issues providing incorrect RMRR or
IVMD
->> entries. Rather than only mapping RAM pages for IOMMU accesses for Dom0,
->> with this option all pages up to 4GB, not marked as unusable in the E820 ->> table, will get a mapping established. Note that this option is only ->> applicable to a PV dom0. Also note that ifdom0-strictmode is enabled ->> then conventional RAM pages not assigned to dom0 will not be mapped. +>> entries. The behaviour of this option is slightly different between a PV and
+>> a PVH Dom0:
+>>
+>> * For a PV Dom0 all pages up to 4GB not marked as unusable in the memory+>> map will get a mapping established. Note that if `dom0-strict` mode is +>> enabled then conventional RAM pages not assigned to dom0 will not be +>> mapped.+>>
+>> * For a PVH Dom0 all memory regions marked as reserved in the memory map
+>> that don't overlap with any MMIO region from emulated devices will be
+>> identity mapped.### iommu\_dev\_iotlb\_timeout
Quoted 1 line
= <integer>xen/drivers/passthrough/x86/iommu.c +2
-0@@ -20,6 +20,8 @@#include <xen/softirq.h> #include <xsm/xsm.h> +#include <asm/apicdef.h> +#include <asm/io_apic.h> #include <asm/setup.h> void iommu_update_ire_from_apic(@@ -134,11 +136,62 @@ void arch_iommu_domain_destroy(struct domain *d) { }
Quoted 4 lines
xen/drivers/passthrough/x86/iommu.c void arch_iommu_domain_destroy(struct domain *d) · line 140
Quoted 2 lines
xen/drivers/passthrough/x86/iommu.c +2
-0+static bool __hwdom_init pv_inclusive_map(unsigned long pfn, +unsigned long max_pfn)Perhaps
pv_hwdom_inclusive_map()(and similarlypvh_hwdom_inclusive_map()) to make it obvious that they are intended only for the hardware domain. (I know the annotation makes this reasonably obvious but other hwdom-specific functions seem to carry this in their name)
Given the naming conversation in patch 2 I think this should be named hwdom_inclusive_map and the new function should be hwdom_reserved_map. I think the pv/pvh prefix can be avoided if you think the reserved mapping is helpful for classic PV also.
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 4/4] x86/iommu: add PVH support to the inclusive options
In reply to Roger Pau MonneQuoted 8 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
Several people have reported hardware issues (malfunctioning USB controllers) due to iommu page faults. Those faults are caused by missing RMRR (VTd) or IRVS (AMD-Vi) entries in the ACPI tables. Those can be worked around on VTd hardware by manually adding RMRR entries on the command line, this is however limited to Intel hardware and quite cumbersome to do.
Quoted 6 lines
In order to solve those issues add PVH support to the inclusive option that identity maps all regions marked as reserved in the memory map. Note that regions used by devices emulated by Xen (LAPIC, IO-APIC or PCIe MCFG regions) are specifically avoided. Note that this option currently relies on no MSIX MMIO areas residing in a reserved region, or else Xen won't be able to trap those accesses.
But that would be a firmware bug anyway: These are sub-ranges of PCI device BARs, and those must not overlap reserved ranges in E820.
xen/drivers/passthrough/x86/iommu.c line 24
Quoted 8 lines
xen/drivers/passthrough/x86/iommu.c +2
-0@@ -20,6 +20,8 @@#include <xen/softirq.h> #include <xsm/xsm.h> +#include <asm/apicdef.h> +#include <asm/io_apic.h>
Why? You're looking for the guest view of things, which I don't think you can derive from definitions in these two headers.
Quoted 8 lines
xen/drivers/passthrough/x86/iommu.c +8
-0+static bool __hwdom_init pvh_inclusive_map(const struct domain *d, +unsigned long pfn) +{ +unsigned int i; + +/* +* Ignore any address below 1MB, that's already identity mapped by the +* domain builder.
xen/drivers/passthrough/x86/iommu.c void arch_iommu_domain_destroy(struct domain *d) · line 171
Quoted 10 lines
xen/drivers/passthrough/x86/iommu.c +10
-0+*/ +if ( pfn <PFN_DOWN (MB (1 )) ) +return false; + +/* Only add reserved regions. */ +if ( !page_is_ram_type (pfn,RAM_TYPE_RESERVED ) ) +return false; + +/* Check that it doesn't overlap with the LAPIC */ +if ( pfn ==PFN_DOWN (APIC_DEFAULT_PHYS_BASE ) )
I.e. this should interact with vlapic.c.
xen/drivers/passthrough/x86/iommu.c void arch_iommu_domain_destroy(struct domain *d) · line 176
Quoted 5 lines
xen/drivers/passthrough/x86/iommu.c +5
-0+return false; +/* ... or the IO-APIC */ +for ( i =0 ; i < nr_ioapics; i++ ) +if ( pfn ==PFN_DOWN (domain_vioapic (d, i)->base_address) ) +return false;
Ah, nr_ioapics legitimately comes from that header. But do you really need that? Can't you use d->arch.hvm_domain.nr_vioapics?
xen/drivers/passthrough/x86/iommu.c void arch_iommu_domain_destroy(struct domain *d) · line 185
Quoted 9 lines
xen/drivers/passthrough/x86/iommu.c +9
-0+/* ... or the PCIe MCFG regions. */ +for ( i =0 ; i < pci_mmcfg_config_num; i++ ) + { +unsigned long addr =PFN_DOWN (pci_mmcfg_config[i].address); + +if ( pfn >= addr + (pci_mmcfg_config[i].start_bus_number <<8 ) && + pfn < addr + (pci_mmcfg_config[i].end_bus_number <<8 ) ) +return false; + }
Same here - this would better use domain state.
Also end_bus_number is inclusive iirc, so the above doesn't cover the entire range as it seems.
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 4/4] x86/iommu: add PVH support to the inclusive options
Quoted 15 lines · Jan Beulich
Quoted 8 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
Several people have reported hardware issues (malfunctioning USB controllers) due to iommu page faults. Those faults are caused by missing RMRR (VTd) or IRVS (AMD-Vi) entries in the ACPI tables. Those can be worked around on VTd hardware by manually adding RMRR entries on the command line, this is however limited to Intel hardware and quite cumbersome to do.
Quoted 6 lines
In order to solve those issues add PVH support to the inclusive option that identity maps all regions marked as reserved in the memory map. Note that regions used by devices emulated by Xen (LAPIC, IO-APIC or PCIe MCFG regions) are specifically avoided. Note that this option currently relies on no MSIX MMIO areas residing in a reserved region, or else Xen won't be able to trap those accesses.
Quoted 3 lines
But that would be a firmware bug anyway: These are sub-ranges of PCI device BARs, and those must not overlap reserved ranges in E820.
Considering this is already a workaround for broken firmware I won't be surprised that on certain hardware BARs overlap with reserved ranges.
Quoted 11 lines
xen/drivers/passthrough/x86/iommu.c line 24
Quoted 8 lines
xen/drivers/passthrough/x86/iommu.c +2
-0@@ -20,6 +20,8 @@#include <xen/softirq.h> #include <xsm/xsm.h> +#include <asm/apicdef.h> +#include <asm/io_apic.h> Why? You're looking for the guest view of things, which I don't think you can derive from definitions in these two headers.
I can use the domain state for the lapic and the ioapic...
Quoted 11 lines
xen/drivers/passthrough/x86/iommu.c void arch_iommu_domain_destroy(struct domain *d) · line 185
Quoted 9 lines
xen/drivers/passthrough/x86/iommu.c +9
-0+/* ... or the PCIe MCFG regions. */ +for ( i =0 ; i < pci_mmcfg_config_num; i++ ) + { +unsigned long addr =PFN_DOWN (pci_mmcfg_config[i].address); + +if ( pfn >= addr + (pci_mmcfg_config[i].start_bus_number <<8 ) && + pfn < addr + (pci_mmcfg_config[i].end_bus_number <<8 ) ) +return false; + }Same here - this would better use domain state.
... but pci mcfg regions are currently added after the iommu initialization in the HVM dom0 builder. I can shuffle that call so that iommu initialization is performed after the mcfg regions are added.
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 4/4] x86/iommu: add PVH support to the inclusive options
Quoted 26 lines · Roger Pau Monné
Quoted 7 lines · Jan Beulich
Quoted 7 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
xen/drivers/passthrough/x86/iommu.c +0
-0@@ -20,6 +20,8 @@#include <xen/softirq.h> #include <xsm/xsm.h> Quoted 5 lines
xen/drivers/passthrough/x86/iommu.c line 24
Quoted 2 lines
xen/drivers/passthrough/x86/iommu.c +2
-0+#include <asm/apicdef.h> +#include <asm/io_apic.h> Why? You're looking for the guest view of things, which I don't think you can derive from definitions in these two headers.
I can use the domain state for the lapic and the ioapic...
Quoted 10 lines
Quoted 9 lines
xen/drivers/passthrough/x86/iommu.c +9
-0+/* ... or the PCIe MCFG regions. */ +for ( i =0 ; i < pci_mmcfg_config_num; i++ ) + { +unsigned long addr =PFN_DOWN (pci_mmcfg_config[i].address); + +if ( pfn >= addr + (pci_mmcfg_config[i].start_bus_number <<8 ) && + pfn < addr + (pci_mmcfg_config[i].end_bus_number <<8 ) ) +return false; + }
Quoted 6 lines
Quoted 1 line
Same here - this would better use domain state.
... but pci mcfg regions are currently added after the iommu initialization in the HVM dom0 builder. I can shuffle that call so that iommu initialization is performed after the mcfg regions are added.
So that's not so easy. The current hvm_mmcfg used to store the MMCFG data in the domain struct is private to io.c, I could move it's declaration to a header file, but maybe it's better to just use pci_mmcfg_config directly?
Thanks, Roger.
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 4/4] x86/iommu: add PVH support to the inclusive options
Quoted 35 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 17:27, <[email protected]> wrote:
Quoted 26 lines · Roger Pau Monné
Quoted 7 lines · Jan Beulich
Quoted 7 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
xen/drivers/passthrough/x86/iommu.c +0
-0@@ -20,6 +20,8 @@#include <xen/softirq.h> #include <xsm/xsm.h> Quoted 5 lines
xen/drivers/passthrough/x86/iommu.c line 24
Quoted 2 lines
xen/drivers/passthrough/x86/iommu.c +2
-0+#include <asm/apicdef.h> +#include <asm/io_apic.h> Why? You're looking for the guest view of things, which I don't think you can derive from definitions in these two headers.
I can use the domain state for the lapic and the ioapic...
Quoted 10 lines
Quoted 9 lines
xen/drivers/passthrough/x86/iommu.c +9
-0+/* ... or the PCIe MCFG regions. */ +for ( i =0 ; i < pci_mmcfg_config_num; i++ ) + { +unsigned long addr =PFN_DOWN (pci_mmcfg_config[i].address); + +if ( pfn >= addr + (pci_mmcfg_config[i].start_bus_number <<8 ) && + pfn < addr + (pci_mmcfg_config[i].end_bus_number <<8 ) ) +return false; + }Quoted 6 lines
Quoted 1 line
Same here - this would better use domain state.
... but pci mcfg regions are currently added after the iommu initialization in the HVM dom0 builder. I can shuffle that call so that iommu initialization is performed after the mcfg regions are added.
Quoted 4 lines
So that's not so easy. The current
hvm_mmcfgused to store the MMCFG data in the domain struct is private toio.c, I could move it's declaration to a header file, but maybe it's better to just usepci_mmcfg_configdirectly?
Or add a helper function in that file?
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Re: [PATCH 4/4] x86/iommu: add PVH support to the inclusive options
In reply to Roger Pau MonnéQuoted 13 lines
Quoted 1 line
Quoted 1 line · On 31.07.18 at 17:15, <[email protected]> wrote:
Quoted 11 lines · Jan Beulich
Quoted 10 lines
Quoted 1 line
Quoted 1 line · On 27.07.18 at 17:31, <[email protected]> wrote:
xen/drivers/passthrough/x86/iommu.c +9
-0+/* ... or the PCIe MCFG regions. */ +for ( i =0 ; i < pci_mmcfg_config_num; i++ ) + { +unsigned long addr =PFN_DOWN (pci_mmcfg_config[i].address); + +if ( pfn >= addr + (pci_mmcfg_config[i].start_bus_number <<8 ) && + pfn < addr + (pci_mmcfg_config[i].end_bus_number <<8 ) ) +return false; + }
Quoted 6 lines
Quoted 1 line
Same here - this would better use domain state.
... but pci mcfg regions are currently added after the iommu initialization in the HVM dom0 builder. I can shuffle that call so that iommu initialization is performed after the mcfg regions are added.
Or at least leave a todo comment. After all MCFG regions might further be added while Dom0 boots and parses AML.
Jan
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel