iommu: Enhance IOMMU default DMA mode build options
Subsystems (4)
v13 cover iommu: Enhance IOMMU default DMA mode build options
This is a reboot of Zhen Lei's series from a couple of years ago, which never made it across the line.
I still think that it has some value, so taking up the mantle.
Motivation:
Allow lazy mode be default mode for DMA domains for all ARCHs, and not only those who hardcode it (to be lazy). For ARM64, currently we must use a kernel command line parameter to use lazy mode, which is less than ideal.
I have now included the print for strict/lazy mode, which I originally sent in:
https://lore.kernel.org/linux-iommu/[email protected]/
There was some concern there about drivers and their custom prints conflicting with the print in that patch, but I think that it should be ok.
Differences to v12:
- Rebase to next-20210611 and include patch "iommu: Update "iommu.strict" documentation" as a baseline
- Add Robin's RB tags (thanks!)
- Please let me know if not ok with kernel-parameters.txt update in 3/6
- Add a patch to mark x86 strict cmdline params as deprecated
- Improve wording in Kconfig change and tweak iommu_dma_strict declaration
Differences to v11:
- Rebase to next-20210610
- Drop strict mode globals in Intel and AMD drivers
- Include patch to print strict vs lazy mode
- Include patch to remove argument from iommu_set_dma_strict()
Differences to v10:
- Rebase to v5.13-rc4
- Correct comment and typo in Kconfig (Randy)
- Make Kconfig choice depend on relevant architectures
- Update kernel-parameters.txt for CONFIG_IOMMU_DEFAULT_{LAZY,STRICT}
John Garry (3):
iommu: Deprecate Intel and AMD cmdline methods to enable strict mode
iommu: Print strict or lazy mode at init time
iommu: Remove mode argument from iommu_set_dma_strict()
Zhen Lei (3):
iommu: Enhance IOMMU default DMA mode build options iommu/vt-d: Add support for IOMMU default DMA mode build options iommu/amd: Add support for IOMMU default DMA mode build options
.../admin-guide/kernel-parameters.txt | 8 ++-- drivers/iommu/Kconfig | 41 +++++++++++++++++++ drivers/iommu/amd/amd_iommu_types.h | 6 --- drivers/iommu/amd/init.c | 7 ++-- drivers/iommu/amd/iommu.c | 6 --- drivers/iommu/intel/iommu.c | 16 ++++---- drivers/iommu/iommu.c | 12 ++++-- include/linux/iommu.h | 2 +- 8 files changed, 66 insertions(+), 32 deletions(-)
Signature · 1 line
2.26.2
PATCH v13 1/6 iommu: Deprecate Intel and AMD cmdline methods to enable strict mode
Now that the x86 drivers support iommu.strict, deprecate the custom methods.
--- Documentation/admin-guide/kernel-parameters.txt | 5 +++-- drivers/iommu/amd/init.c | 4 +++- drivers/iommu/intel/iommu.c | 1 + 3 files changed, 7 insertions(+), 3 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +3 -2
@@ -290,7 +290,8 @@ amd_iommu= [HW,X86-64] Pass parameters to the AMD IOMMU driver in the system. Possible values are: - fullflush - enable flushing of IO/TLB entries when + fullflush [Deprecated, use iommu.strict instead] + - enable flushing of IO/TLB entries when they are unmapped. Otherwise they are flushed before they will be reused, which is a lot of faster
@@ -1947,7 +1948,7 @@ bypassed by not enabling DMAR with this option. In this case, gfx device will use physical address for DMA. - strict [Default Off] + strict [Default Off] [Deprecated, use iommu.strict instead] With this option on every unmap_single operation will result in a hardware IOTLB flush operation as opposed to batching them for performance.
drivers/iommu/amd/init.c +3 -1
@@ -3098,8 +3098,10 @@ static int __init parse_amd_iommu_intr(char *str)static int __init parse_amd_iommu_options (char *str) {for (; *str; ++str) { -if (strncmp (str,"fullflush" ,9 ) ==0 ) +if (strncmp (str,"fullflush" ,9 ) ==0 ) { +pr_warn ("amd_iommu=fullflush deprecated; use iommu.strict instead\n" ); amd_iommu_unmap_flush = true; + }if (strncmp (str,"force_enable" ,12 ) ==0 ) amd_iommu_force_enable = true;if (strncmp (str,"off" ,3 ) ==0 )
drivers/iommu/intel/iommu.c +1 -0
@@ -454,6 +454,7 @@ static int __init intel_iommu_setup(char *str)pr_warn ("intel_iommu=forcedac deprecated; use iommu.forcedac instead\n" ); iommu_dma_forcedac = true; }else if (!strncmp (str,"strict" ,6 )) { +pr_warn ("intel_iommu=strict deprecated; use iommu.strict instead\n" );pr_info ("Disable batched IOTLB flush\n" ); intel_iommu_strict =1 ; }else if (!strncmp (str,"sp_off" ,6 )) {
Signature · 1 line
2.26.2
Re: [PATCH v13 1/6] iommu: Deprecate Intel and AMD cmdline methods to enable strict mode
Quoted 25 lines · John Garry
Now that the x86 drivers support iommu.strict, deprecate the custom methods.
Signed-off-by John Garry <[email protected]>--- Documentation/admin-guide/kernel-parameters.txt | 5 +++-- drivers/iommu/amd/init.c | 4 +++- drivers/iommu/intel/iommu.c | 1 + 3 files changed, 7 insertions(+), 3 deletions(-)Documentation/admin-guide/kernel-parameters.txt +2
-1@@ -290,7 +290,8 @@ amd_iommu= [HW,X86-64] Pass parameters to the AMD IOMMU driver in the system. Possible values are: - fullflush - enable flushing of IO/TLB entries when + fullflush [Deprecated, use iommu.strict instead] + - enable flushing of IO/TLB entries when they are unmapped. Otherwise they are flushed before they will be reused, which is a lot of faster
Documentation/admin-guide/kernel-parameters.txt line 1954
Quoted 8 lines
Documentation/admin-guide/kernel-parameters.txt +1
-1@@ -1947,7 +1948,7 @@ bypassed by not enabling DMAR with this option. In this case, gfx device will use physical address for DMA. - strict [Default Off] + strict [Default Off] [Deprecated, use iommu.strict instead] With this option on every unmap_single operation will result in a hardware IOTLB flush operation as opposed to batching them for performance.
FWIW I'd be inclined to replace both whole descriptions with just something like "Deprecated, equivalent to iommu.strict=1".
drivers/iommu/amd/init.c static int __init parse_amd_iommu_intr(char *str) · line 3102
Quoted 11 lines
drivers/iommu/amd/init.c +2
-1@@ -3098,8 +3098,10 @@ static int __init parse_amd_iommu_intr(char *str)static int __init parse_amd_iommu_options (char *str) {for (; *str; ++str) { -if (strncmp (str,"fullflush" ,9 ) ==0 ) +if (strncmp (str,"fullflush" ,9 ) ==0 ) { +pr_warn ("amd_iommu=fullflush deprecated; use iommu.strict instead\n" );
Nit: maybe we should spell out "...use <option>=1 instead" in all of these messages just in case anyone takes them literally? (I'm not sure the options parse correctly with no argument)
Either way,
Thanks,
Robin.
Quoted 18 lines
drivers/iommu/amd/init.c +1
-0amd_iommu_unmap_flush = true; + }if (strncmp (str,"force_enable" ,12 ) ==0 ) amd_iommu_force_enable = true;if (strncmp (str,"off" ,3 ) ==0 )drivers/iommu/intel/iommu.c +1
-0@@ -454,6 +454,7 @@ static int __init intel_iommu_setup(char *str)pr_warn ("intel_iommu=forcedac deprecated; use iommu.forcedac instead\n" ); iommu_dma_forcedac = true; }else if (!strncmp (str,"strict" ,6 )) { +pr_warn ("intel_iommu=strict deprecated; use iommu.strict instead\n" );pr_info ("Disable batched IOTLB flush\n" ); intel_iommu_strict =1 ; }else if (!strncmp (str,"sp_off" ,6 )) {
Re: [PATCH v13 1/6] iommu: Deprecate Intel and AMD cmdline methods to enable strict mode
Quoted 25 lines · Robin Murphy
Quoted 6 lines
DMA.
- strict [Default Off]
+ strict [Default Off] [Deprecated, use iommu.strict instead] With this option on everyunmap_singleoperation will result in a hardware IOTLB flush operation as opposed to batching them for performance.FWIW I'd be inclined to replace both whole descriptions with just something like "Deprecated, equivalent to iommu.strict=1".
drivers/iommu/amd/init.c static int __init parse_amd_iommu_intr(char *str) · line 3103
Quoted 12 lines
drivers/iommu/amd/init.c +2
-1@@ -3098,8 +3098,10 @@ static int __init parse_amd_iommu_intr(char *str) static int __init parse_amd_iommu_options(char *str) { for (; *str; ++str) { - if (strncmp(str, "fullflush", 9) == 0) + if (strncmp(str, "fullflush", 9) == 0) { + pr_warn("amd_iommu=fullflush deprecated; use iommu.strict instead\n");Nit: maybe we should spell out "...use <option>=1 instead" in all of these messages just in case anyone takes them literally?
Quoted 1 line
(I'm not sure the options parse correctly with no argument)
I don't think they do.
But I'll take both suggestions on board.
Quoted 4 lines
Either way,
Acked-by Robin Murphy <[email protected]>
Cheers!
PATCH v13 2/6 iommu: Print strict or lazy mode at init time
In reply to John GarryAs well as the default domain type, it's useful to know whether strict or lazy for DMA domains, so add this info in a separate print.
The (stict/lazy) mode may be also set via iommu.strict earlyparm, but this will be processed prior to iommu_subsys_init(), so that print will be accurate for drivers which don't set the mode via custom means.
For the drivers which set the mode via custom means - AMD and Intel drivers
- they maintain prints to inform a change in policy or that custom cmdline methods to change policy are deprecated.
--- drivers/iommu/iommu.c | 5 +++++ 1 file changed, 5 insertions(+)
drivers/iommu/iommu.c +5 -0
@@ -138,6 +138,11 @@ static int __init iommu_subsys_init(void) (iommu_cmd_line &IOMMU_CMD_LINE_DMA_API ) ? "(set via kernel command line)" : ""); +pr_info ("DMA domain TLB invalidation policy: %s mode %s\n" , + iommu_dma_strict ?"strict" :"lazy" , + (iommu_cmd_line &IOMMU_CMD_LINE_STRICT ) ? +"(set via kernel command line)" :"" ); +return 0 ; }subsys_initcall (iommu_subsys_init);
Signature · 1 line
2.26.2
PATCH v13 3/6 iommu: Enhance IOMMU default DMA mode build options
In reply to John GarryFrom: Zhen Lei <[email protected]>
First, add build options IOMMU_DEFAULT_{LAZY|STRICT}, so that we have the opportunity to set {lazy|strict} mode as default at build time. Then put the two config options in an choice, as they are mutually exclusive.
--- .../admin-guide/kernel-parameters.txt | 3 +- drivers/iommu/Kconfig | 40 +++++++++++++++++++ drivers/iommu/iommu.c | 2 +- 3 files changed, 43 insertions(+), 2 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +2 -1
@@ -2052,9 +2052,10 @@ throughput at the cost of reduced device isolation. Will fall back to strict mode if not supported by the relevant IOMMU driver. - 1 - Strict mode (default). + 1 - Strict mode. DMA unmap operations invalidate IOMMU hardware TLBs synchronously. + unset - Use value of CONFIG_IOMMU_DEFAULT_{LAZY,STRICT}. Note: on x86, the default behaviour depends on the equivalent driver-specific parameters, but a strict mode explicitly specified by either method takes
drivers/iommu/Kconfig +40 -0
@@ -90,6 +90,46 @@ config IOMMU_DEFAULT_PASSTHROUGH If unsure, say N here. +choice + prompt "IOMMU default DMA IOTLB invalidation mode" + depends on IOMMU_DMA + + default IOMMU_DEFAULT_STRICT + help + This option allows an IOMMU DMA IOTLB invalidation mode to be + chosen at build time, to override the default mode of each ARCH, + removing the need to pass in kernel parameters through command line. + It is still possible to provide common boot params to override this + config. + + If unsure, keep the default. + +config IOMMU_DEFAULT_STRICT + bool "strict" + help + For every IOMMU DMA unmap operation, the flush operation of IOTLB and + the free operation of IOVA are guaranteed to be done in the unmap + function. + +config IOMMU_DEFAULT_LAZY + bool "lazy" + help + Support lazy mode, where for every IOMMU DMA unmap operation, the + flush operation of IOTLB and the free operation of IOVA are deferred. + They are only guaranteed to be done before the related IOVA will be + reused. + + The isolation provided in this mode is not as secure as STRICT mode, + such that a vulnerable time window may be created between the DMA + unmap and the mappings cached in the IOMMU IOTLB or device TLB + finally being invalidated, where the device could still access the + memory which has already been unmapped by the device driver. + However this mode may provide better performance in high throughput + scenarios, and is still considerably more secure than passthrough + mode or no IOMMU. + +endchoice + config OF_IOMMU def_bool y depends on OF && IOMMU_API
drivers/iommu/iommu.c +1 -1
@@ -29,7 +29,7 @@ static struct kset *iommu_group_kset;static DEFINE_IDA (iommu_group_ida);static unsigned int iommu_def_domain_type__read_mostly ; -static bool iommu_dma_strict__read_mostly = true; +static bool iommu_dma_strict__read_mostly =IS_ENABLED (CONFIG_IOMMU_DEFAULT_STRICT );static u32 iommu_cmd_line__read_mostly ;struct iommu_group {
Signature · 1 line
2.26.2
PATCH v13 4/6 iommu/vt-d: Add support for IOMMU default DMA mode build options
In reply to John GarryFrom: Zhen Lei <[email protected]>
Make IOMMU_DEFAULT_LAZY default for when INTEL_IOMMU config is set, as is current behaviour.
Also delete global flag intel_iommu_strict:
- In intel_iommu_setup(), call iommu_set_dma_strict(true) directly. Also remove the print, as iommu_subsys_init() prints the mode and we have already marked this param as deprecated.
- For cap_caching_mode() check in intel_iommu_setup(), call iommu_set_dma_strict(true) directly, and reword the accompanying print and add the missing '\n'.
- For Ironlake GPU, again call iommu_set_dma_strict(true) directly and keep the accompanying print.
--- drivers/iommu/Kconfig | 1 + drivers/iommu/intel/iommu.c | 15 ++++++--------- 2 files changed, 7 insertions(+), 9 deletions(-)
drivers/iommu/Kconfig +1 -0
@@ -94,6 +94,7 @@ choice prompt "IOMMU default DMA IOTLB invalidation mode" depends on IOMMU_DMA + default IOMMU_DEFAULT_LAZY if INTEL_IOMMU default IOMMU_DEFAULT_STRICT help This option allows an IOMMU DMA IOTLB invalidation mode to be
drivers/iommu/intel/iommu.c +6 -9
@@ -361,7 +361,6 @@ int intel_iommu_enabled = 0;EXPORT_SYMBOL_GPL (intel_iommu_enabled);static int dmar_map_gfx =1 ; -static int intel_iommu_strict;static int intel_iommu_superpage =1 ;static int iommu_identity_mapping;static int iommu_skip_te_disable;
@@ -455,8 +454,7 @@ static int __init intel_iommu_setup(char *str) iommu_dma_forcedac = true; }else if (!strncmp (str,"strict" ,6 )) {pr_warn ("intel_iommu=strict deprecated; use iommu.strict instead\n" ); -pr_info ("Disable batched IOTLB flush\n" ); - intel_iommu_strict =1 ; +iommu_set_dma_strict (true); }else if (!strncmp (str,"sp_off" ,6 )) {pr_info ("Disable supported super page\n" ); intel_iommu_superpage =0 ;
@@ -4382,9 +4380,9 @@ int __init intel_iommu_init(void)* is likely to be much lower than the overhead of synchronizing * the virtual and physical IOMMU page-tables. */ -if (!intel_iommu_strict &&cap_caching_mode (iommu->cap)) { -pr_warn ("IOMMU batching is disabled due to virtualization" ); - intel_iommu_strict =1 ; +if (cap_caching_mode (iommu->cap)) { +pr_warn ("IOMMU batching disallowed due to virtualization\n" ); +iommu_set_dma_strict (true); } iommu_device_sysfs_add(&iommu->iommu, NULL, intel_iommu_groups,
@@ -4393,7 +4391,6 @@ int __init intel_iommu_init(void) }up_read (&dmar_global_lock); -iommu_set_dma_strict (intel_iommu_strict);bus_set_iommu (&pci_bus_type, &intel_iommu_ops);if (si_domain && !hw_pass_through)register_memory_notifier (&intel_iommu_memory_nb);
@@ -5702,8 +5699,8 @@ static void quirk_calpella_no_shadow_gtt(struct pci_dev *dev) }else if (dmar_map_gfx) {/* we have to ensure the gfx device is idle before we flush */ pci_info (dev,"Disabling batched IOTLB flush on Ironlake\n" ); - intel_iommu_strict =1 ; - } +iommu_set_dma_strict (true); + } }DECLARE_PCI_FIXUP_HEADER (PCI_VENDOR_ID_INTEL ,0x0040 , quirk_calpella_no_shadow_gtt);DECLARE_PCI_FIXUP_HEADER (PCI_VENDOR_ID_INTEL ,0x0044 , quirk_calpella_no_shadow_gtt);
Signature · 1 line
2.26.2
Lu Baolu
maintainerRe: [PATCH v13 4/6] iommu/vt-d: Add support for IOMMU default DMA mode build options
drivers/iommu/intel/iommu.c int __init intel_iommu_init(void) · line 4385
Quoted 10 lines · John Garry
drivers/iommu/intel/iommu.c +3
-3@@ -4382,9 +4380,9 @@ int __init intel_iommu_init(void)* is likely to be much lower than the overhead of synchronizing * the virtual and physical IOMMU page-tables. */ -if (!intel_iommu_strict &&cap_caching_mode (iommu->cap)) { -pr_warn ("IOMMU batching is disabled due to virtualization" ); - intel_iommu_strict =1 ; +if (cap_caching_mode (iommu->cap)) { +pr_warn ("IOMMU batching disallowed due to virtualization\n" ); +iommu_set_dma_strict (true);
With this change, VM guest will always show this warning. How about removing this message? Users could get the same information through the kernel message added by "[PATCH v13 2/6] iommu: Print strict or lazy mode at init time".
Best regards,
baolu
Re: [PATCH v13 4/6] iommu/vt-d: Add support for IOMMU default DMA mode build options
Quoted 14 lines · Lu Baolu
drivers/iommu/intel/iommu.c int __init intel_iommu_init(void) · line 4386
Quoted 11 lines · John Garry
drivers/iommu/intel/iommu.c +3
-3@@ -4382,9 +4380,9 @@ int __init intel_iommu_init(void)* is likely to be much lower than the overhead of synchronizing * the virtual and physical IOMMU page-tables. */ - if (!intel_iommu_strict && cap_caching_mode(iommu->cap)) { - pr_warn("IOMMU batching is disabled due to virtualization"); - intel_iommu_strict = 1; + if (cap_caching_mode(iommu->cap)) { + pr_warn("IOMMU batching disallowed due to virtualization\n"); + iommu_set_dma_strict(true);With this change, VM guest will always show this warning.
Would they have got it before also normally?
I mean, default is intel_iommu_strict=0, so if cap_caching_mode(iommu->cap) is true and intel_iommu_strict not set to 1 elsewhere previously, then we would get this print.
Quoted 4 lines
How about
removing this message? Users could get the same information through the kernel message added by "[PATCH v13 2/6] iommu: Print strict or lazy mode at init time".
I think that the print from 2/6 should occur before this print.
Regardless I would think that you would still like to be notified of this change in policy, right?
However I now realize that the print is in a loop per iommu, so we would get it per iommu:
for_each_active_iommu(iommu, drhd) {
/*
* The flush queue implementation does not perform
* page-selective invalidations that are required for efficient
* TLB flushes in virtual environments. The benefit of batching
* is likely to be much lower than the overhead of synchronizing
* the virtual and physical IOMMU page-tables.
*/
if (!intel_iommu_strict && cap_caching_mode(iommu->cap)) {
pr_warn("IOMMU batching is disabled due to virtualization");
intel_iommu_strict = 1;
}
...
}I need to change that. How about this:
bool print_warning = false;
for_each_active_iommu(iommu, drhd) {
/*
* The flush queue implementation does not perform
* page-selective invalidations that are required for efficient
* TLB flushes in virtual environments. The benefit of batching
* is likely to be much lower than the overhead of synchronizing
* the virtual and physical IOMMU page-tables.
*/
if (!print_warning && cap_caching_mode(iommu->cap)) {
pr_warn("IOMMU batching disallowed due to virtualization\n");
iommu_set_dma_strict(true);
print_warning = true;
}
...
}or use pr_warn_once().
Thanks,
John
Robin Murphy
reviewerRe: [PATCH v13 4/6] iommu/vt-d: Add support for IOMMU default DMA mode build options
Quoted 67 lines · John Garry
Quoted 15 lines · Lu Baolu
drivers/iommu/intel/iommu.c int __init intel_iommu_init(void) · line 4387
Quoted 12 lines · John Garry
drivers/iommu/intel/iommu.c +3
-3@@ -4382,9 +4380,9 @@ int __init intel_iommu_init(void)* is likely to be much lower than the overhead of synchronizing * the virtual and physical IOMMU page-tables. */ - if (!intel_iommu_strict && cap_caching_mode(iommu->cap)) { - pr_warn("IOMMU batching is disabled due to virtualization"); - intel_iommu_strict = 1; + if (cap_caching_mode(iommu->cap)) { + pr_warn("IOMMU batching disallowed due to virtualization\n"); + iommu_set_dma_strict(true);With this change, VM guest will always show this warning.
Would they have got it before also normally?
I mean, default is intel_iommu_strict=0, so if
cap_caching_mode(iommu->cap)is true andintel_iommu_strictnot set to 1 elsewhere previously, then we would get this print.Quoted 4 lines
How about
removing this message? Users could get the same information through the kernel message added by "[PATCH v13 2/6] iommu: Print strict or lazy mode at init time".I think that the print from 2/6 should occur before this print.
Regardless I would think that you would still like to be notified of this change in policy, right?
However I now realize that the print is in a loop per iommu, so we would get it per iommu:
for_each_active_iommu(iommu, drhd){
/*
* The flush queue implementation does not perform
* page-selective invalidations that are required for efficient
* TLB flushes in virtual environments. The benefit of batching
* is likely to be much lower than the overhead of synchronizing
* the virtual and physical IOMMU page-tables.
*/
if (!intel_iommu_strict &&cap_caching_mode(iommu->cap)) {pr_warn("IOMMU batching is disabled due to virtualization");intel_iommu_strict= 1;
}
...
}I need to change that. How about this:
bool
print_warning= false;
for_each_active_iommu(iommu, drhd){
/*
* The flush queue implementation does not perform
* page-selective invalidations that are required for efficient
* TLB flushes in virtual environments. The benefit of batching
* is likely to be much lower than the overhead of synchronizing
* the virtual and physical IOMMU page-tables.
*/
if (!print_warning &&cap_caching_mode(iommu->cap)) {pr_warn("IOMMU batching disallowed due to virtualization\n");iommu_set_dma_strict(true);
print_warning= true;
}
...
}
Quoted 1 line
or use
pr_warn_once().
Maybe even downgrade it to pr_info_once(), since AIUI it's not really anything scary?
I suppose you could technically fake up a domain on the stack to get the global setting out of iommu_get_dma_strict(), or perhaps give iommu_set_dma_strict() a cheeky return value to indicate what the previous setting was, in order to suppress the message entirely if strict is already set, but I'm not at all convinced it's worth the bother.
Robin.
Lu Baolu
maintainerRe: [PATCH v13 4/6] iommu/vt-d: Add support for IOMMU default DMA mode build options
In reply to John GarryHi John,
Quoted 17 lines · John Garry
Quoted 15 lines · Lu Baolu
drivers/iommu/intel/iommu.c int __init intel_iommu_init(void) · line 4387
Quoted 12 lines · John Garry
drivers/iommu/intel/iommu.c +3
-3@@ -4382,9 +4380,9 @@ int __init intel_iommu_init(void)* is likely to be much lower than the overhead of synchronizing * the virtual and physical IOMMU page-tables. */ - if (!intel_iommu_strict && cap_caching_mode(iommu->cap)) { - pr_warn("IOMMU batching is disabled due to virtualization"); - intel_iommu_strict = 1; + if (cap_caching_mode(iommu->cap)) { + pr_warn("IOMMU batching disallowed due to virtualization\n"); + iommu_set_dma_strict(true);With this change, VM guest will always show this warning.
Quoted 3 lines
Would they have got it before also normally?
I mean, default is intel_iommu_strict=0, so if
cap_caching_mode(iommu->cap)is true andintel_iommu_strictnot set to 1 elsewhere previously, then we would get this print.
Yes. You are right.
Quoted 47 lines
Quoted 5 lines
How about
removing this message? Users could get the same information through the kernel message added by "[PATCH v13 2/6] iommu: Print strict or lazy mode at init time".I think that the print from 2/6 should occur before this print.
Regardless I would think that you would still like to be notified of this change in policy, right?
However I now realize that the print is in a loop per iommu, so we would get it per iommu:
for_each_active_iommu(iommu, drhd){
/*
* The flush queue implementation does not perform
* page-selective invalidations that are required for efficient
* TLB flushes in virtual environments. The benefit of batching
* is likely to be much lower than the overhead of synchronizing
* the virtual and physical IOMMU page-tables.
*/
if (!intel_iommu_strict &&cap_caching_mode(iommu->cap)) {pr_warn("IOMMU batching is disabled due to virtualization");intel_iommu_strict= 1;
}
...
}I need to change that. How about this:
bool
print_warning= false;
for_each_active_iommu(iommu, drhd){
/*
* The flush queue implementation does not perform
* page-selective invalidations that are required for efficient
* TLB flushes in virtual environments. The benefit of batching
* is likely to be much lower than the overhead of synchronizing
* the virtual and physical IOMMU page-tables.
*/
if (!print_warning &&cap_caching_mode(iommu->cap)) {pr_warn("IOMMU batching disallowed due to virtualization\n");iommu_set_dma_strict(true);
print_warning= true;
}
...
}
Quoted 1 line
or use
pr_warn_once().
From my p.o.v, pr_xxxx_once() is better.
How about using a pr_info_once()? I don't think it's a warning, it's just a policy choice in VM environment.
Best regards,
baolu
Re: [PATCH v13 4/6] iommu/vt-d: Add support for IOMMU default DMA mode build options
On 18/06/2021 02:46, Lu Baolu wrote:
Hi baolu,
Quoted 20 lines
Quoted 20 lines
I need to change that. How about this:
bool
print_warning= false;
for_each_active_iommu(iommu, drhd){
/*
* The flush queue implementation does not perform
* page-selective invalidations that are required for efficient
* TLB flushes in virtual environments. The benefit of batching
* is likely to be much lower than the overhead of synchronizing
* the virtual and physical IOMMU page-tables.
*/
if (!print_warning &&cap_caching_mode(iommu->cap)) {pr_warn("IOMMU batching disallowed due to virtualization\n");iommu_set_dma_strict(true);
print_warning= true;
}
...
}
Quoted 6 lines
Quoted 1 line
or use
pr_warn_once().From my p.o.v,
pr_xxxx_once()is better.How about using a
pr_info_once()? I don't think it's a warning, it's just a policy choice in VM environment.
ok, I can go with that, which Robin mostly agrees with.
Thanks,
John
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu
PATCH v13 5/6 iommu/amd: Add support for IOMMU default DMA mode build options
In reply to John GarryFrom: Zhen Lei <[email protected]>
Make IOMMU_DEFAULT_LAZY default for when AMD_IOMMU config is set, which matches current behaviour.
For "fullflush" param, just call iommu_set_dma_strict(true) directly.
Since we get a strict vs lazy mode print already in iommu_subsys_init(), and maintain a deprecation print when "fullflush" param is passed, drop the prints in amd_iommu_init_dma_ops().
Finally drop global flag amd_iommu_unmap_flush, as it has no longer has any purpose.
--- drivers/iommu/Kconfig | 2 +- drivers/iommu/amd/amd_iommu_types.h | 6 ------ drivers/iommu/amd/init.c | 3 +-- drivers/iommu/amd/iommu.c | 6 ------ 4 files changed, 2 insertions(+), 15 deletions(-)
drivers/iommu/Kconfig +1 -1
@@ -94,7 +94,7 @@ choice prompt "IOMMU default DMA IOTLB invalidation mode" depends on IOMMU_DMA - default IOMMU_DEFAULT_LAZY if INTEL_IOMMU + default IOMMU_DEFAULT_LAZY if (AMD_IOMMU || INTEL_IOMMU) default IOMMU_DEFAULT_STRICT help This option allows an IOMMU DMA IOTLB invalidation mode to be
drivers/iommu/amd/amd_iommu_types.h +0 -6
@@ -779,12 +779,6 @@ extern u16 amd_iommu_last_bdf;/* allocation bitmap for domain ids */ extern unsigned long *amd_iommu_pd_alloc_bitmap; -/* -* If true, the addresses will be flushed on unmap time, not when -* they are reused -*/ -extern bool amd_iommu_unmap_flush; -/* Smallest max PASID supported by any IOMMU in the system */ extern u32 amd_iommu_max_pasid;
drivers/iommu/amd/init.c +1 -2
@@ -161,7 +161,6 @@ u16 amd_iommu_last_bdf; /* largest PCI device id we haveto handle */ LIST_HEAD (amd_iommu_unity_map);/* a list of required unity mappings we find in ACPI */ -bool amd_iommu_unmap_flush;/* if true, flush on every unmap */ LIST_HEAD (amd_iommu_list);/* list of all AMD IOMMUs in the system */
@@ -3100,7 +3099,7 @@ static int __init parse_amd_iommu_options(char *str)for (; *str; ++str) {if (strncmp (str,"fullflush" ,9 ) ==0 ) {pr_warn ("amd_iommu=fullflush deprecated; use iommu.strict instead\n" ); - amd_iommu_unmap_flush = true; +iommu_set_dma_strict (true); }if (strncmp (str,"force_enable" ,12 ) ==0 ) amd_iommu_force_enable = true;
drivers/iommu/amd/iommu.c +0 -6
@@ -1775,12 +1775,6 @@ void amd_iommu_domain_update(struct protection_domain *domain)static void __init amd_iommu_init_dma_ops (void ) { swiotlb = (iommu_default_passthrough () || sme_me_mask) ?1 :0 ; - -if (amd_iommu_unmap_flush) -pr_info ("IO/TLB flush on unmap enabled\n" ); -else -pr_info ("Lazy IO/TLB flushing enabled\n" ); -iommu_set_dma_strict (amd_iommu_unmap_flush); }int __init amd_iommu_init_api (void )
Signature · 6 lines
2.26.2 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
PATCH v13 6/6 iommu: Remove mode argument from iommu_set_dma_strict()
In reply to John GarryWe only ever now set strict mode enabled in iommu_set_dma_strict(), so just remove the argument.
--- drivers/iommu/amd/init.c | 2 +- drivers/iommu/intel/iommu.c | 6 +++--- drivers/iommu/iommu.c | 5 ++--- include/linux/iommu.h | 2 +- 4 files changed, 7 insertions(+), 8 deletions(-)
drivers/iommu/amd/init.c +1 -1
@@ -3099,7 +3099,7 @@ static int __init parse_amd_iommu_options(char *str)for (; *str; ++str) {if (strncmp (str,"fullflush" ,9 ) ==0 ) {pr_warn ("amd_iommu=fullflush deprecated; use iommu.strict instead\n" ); -iommu_set_dma_strict (true); +iommu_set_dma_strict (); }if (strncmp (str,"force_enable" ,12 ) ==0 ) amd_iommu_force_enable = true;
drivers/iommu/intel/iommu.c +3 -3
@@ -454,7 +454,7 @@ static int __init intel_iommu_setup(char *str) iommu_dma_forcedac = true; }else if (!strncmp (str,"strict" ,6 )) {pr_warn ("intel_iommu=strict deprecated; use iommu.strict instead\n" ); -iommu_set_dma_strict (true); +iommu_set_dma_strict (); }else if (!strncmp (str,"sp_off" ,6 )) {pr_info ("Disable supported super page\n" ); intel_iommu_superpage =0 ;
@@ -4382,7 +4382,7 @@ int __init intel_iommu_init(void)*/ if (cap_caching_mode (iommu->cap)) {pr_warn ("IOMMU batching disallowed due to virtualization\n" ); -iommu_set_dma_strict (true); +iommu_set_dma_strict (); } iommu_device_sysfs_add(&iommu->iommu, NULL, intel_iommu_groups,
@@ -5699,7 +5699,7 @@ static void quirk_calpella_no_shadow_gtt(struct pci_dev *dev) }else if (dmar_map_gfx) {/* we have to ensure the gfx device is idle before we flush */ pci_info (dev,"Disabling batched IOTLB flush on Ironlake\n" ); -iommu_set_dma_strict (true); +iommu_set_dma_strict (); } }DECLARE_PCI_FIXUP_HEADER (PCI_VENDOR_ID_INTEL ,0x0040 , quirk_calpella_no_shadow_gtt);
drivers/iommu/iommu.c +2 -3
@@ -349,10 +349,9 @@ static int __init iommu_dma_setup(char *str) }early_param ("iommu.strict" , iommu_dma_setup); -void iommu_set_dma_strict (bool strict) +void iommu_set_dma_strict (void ) { -if (strict || !(iommu_cmd_line &IOMMU_CMD_LINE_STRICT )) - iommu_dma_strict = strict; + iommu_dma_strict = true; }bool iommu_get_dma_strict (struct iommu_domain *domain)
include/linux/iommu.h +1 -1
@@ -476,7 +476,7 @@ int iommu_enable_nesting(struct iommu_domain *domain);int iommu_set_pgtable_quirks (struct iommu_domain *domain,unsigned long quirks); -void iommu_set_dma_strict (bool val); +void iommu_set_dma_strict (void );bool iommu_get_dma_strict (struct iommu_domain *domain); extern int report_iommu_fault(struct iommu_domain *domain, struct device *dev,
Signature · 6 lines
2.26.2 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerRe: [PATCH v13 6/6] iommu: Remove mode argument from iommu_set_dma_strict()
Quoted 61 lines · John Garry
We only ever now set strict mode enabled in
iommu_set_dma_strict(), so just remove the argument.Signed-off-by John Garry <[email protected]>Reviewed-by Robin Murphy <[email protected]>--- drivers/iommu/amd/init.c | 2 +- drivers/iommu/intel/iommu.c | 6 +++--- drivers/iommu/iommu.c | 5 ++--- include/linux/iommu.h | 2 +- 4 files changed, 7 insertions(+), 8 deletions(-)drivers/iommu/amd/init.c +1
-1@@ -3099,7 +3099,7 @@ static int __init parse_amd_iommu_options(char *str)for (; *str; ++str) {if (strncmp (str,"fullflush" ,9 ) ==0 ) {pr_warn ("amd_iommu=fullflush deprecated; use iommu.strict instead\n" ); -iommu_set_dma_strict (true); +iommu_set_dma_strict (); }if (strncmp (str,"force_enable" ,12 ) ==0 ) amd_iommu_force_enable = true;drivers/iommu/intel/iommu.c +3
-3@@ -454,7 +454,7 @@ static int __init intel_iommu_setup(char *str) iommu_dma_forcedac = true; }else if (!strncmp (str,"strict" ,6 )) {pr_warn ("intel_iommu=strict deprecated; use iommu.strict instead\n" ); -iommu_set_dma_strict (true); +iommu_set_dma_strict (); }else if (!strncmp (str,"sp_off" ,6 )) {pr_info ("Disable supported super page\n" ); intel_iommu_superpage =0 ;@@ -4382,7 +4382,7 @@ int __init intel_iommu_init(void)*/ if (cap_caching_mode (iommu->cap)) {pr_warn ("IOMMU batching disallowed due to virtualization\n" ); -iommu_set_dma_strict (true); +iommu_set_dma_strict (); } iommu_device_sysfs_add(&iommu->iommu, NULL, intel_iommu_groups,@@ -5699,7 +5699,7 @@ static void quirk_calpella_no_shadow_gtt(struct pci_dev *dev) }else if (dmar_map_gfx) {/* we have to ensure the gfx device is idle before we flush */ pci_info (dev,"Disabling batched IOTLB flush on Ironlake\n" ); -iommu_set_dma_strict (true); +iommu_set_dma_strict (); } }DECLARE_PCI_FIXUP_HEADER (PCI_VENDOR_ID_INTEL ,0x0040 , quirk_calpella_no_shadow_gtt);drivers/iommu/iommu.c +0
-0
drivers/iommu/iommu.c static int __init iommu_dma_setup(char *str) · line 354
Quoted 9 lines
drivers/iommu/iommu.c +2
-3@@ -349,10 +349,9 @@ static int __init iommu_dma_setup(char *str) }early_param ("iommu.strict" , iommu_dma_setup); -void iommu_set_dma_strict (bool strict) +void iommu_set_dma_strict (void ) { -if (strict || !(iommu_cmd_line &IOMMU_CMD_LINE_STRICT )) - iommu_dma_strict = strict; + iommu_dma_strict = true;
Sorry, I still can't get how iommu.strict kernel option works.
static int __init iommu_dma_setup(char *str)
{
int ret = kstrtobool(str, &iommu_dma_strict);
if (!ret)
iommu_cmd_line |= IOMMU_CMD_LINE_STRICT;
return ret;
}early_param("iommu.strict", iommu_dma_setup);
The bit IOMMU_CMD_LINE_STRICT is only set, but not used anywhere. Hence, I am wondering how could it work? A bug or I missed anything?
Best regards,
baolu
Quoted 17 lines
drivers/iommu/iommu.c +0
-0}bool iommu_get_dma_strict (struct iommu_domain *domain)include/linux/iommu.h +1
-1@@ -476,7 +476,7 @@ int iommu_enable_nesting(struct iommu_domain *domain);int iommu_set_pgtable_quirks (struct iommu_domain *domain,unsigned long quirks); -void iommu_set_dma_strict (bool val); +void iommu_set_dma_strict (void );bool iommu_get_dma_strict (struct iommu_domain *domain); extern int report_iommu_fault(struct iommu_domain *domain, struct device *dev,
Re: [PATCH v13 6/6] iommu: Remove mode argument from iommu_set_dma_strict()
Quoted 17 lines
drivers/iommu/iommu.c static int __init iommu_dma_setup(char *str) · line 353
Quoted 10 lines
drivers/iommu/iommu.c +2
-3@@ -349,10 +349,9 @@ static int __init iommu_dma_setup(char *str) } early_param("iommu.strict", iommu_dma_setup); -void iommu_set_dma_strict (bool strict) +void iommu_set_dma_strict (void ) { - if (strict || !(iommu_cmd_line & IOMMU_CMD_LINE_STRICT)) - iommu_dma_strict = strict; + iommu_dma_strict = true;Sorry, I still can't get how iommu.strict kernel option works.
static int
__initiommu_dma_setup(char *str){
int ret =kstrtobool(str, &iommu_dma_strict);
Quoted 7 lines
if (!ret)
iommu_cmd_line|=IOMMU_CMD_LINE_STRICT; return ret;
}early_param("iommu.strict", iommu_dma_setup);The bit
IOMMU_CMD_LINE_STRICTis only set, but not used anywhere.
It is used in patch 2/6:
+ pr_info("DMA domain TLB invalidation policy: %s mode %s\n",
+ iommu_dma_strict ? "strict" : "lazy",
+ (iommu_cmd_line & IOMMU_CMD_LINE_STRICT) ?
+ "(set via kernel command line)" : "");
Quoted 2 lines
Hence,
I am wondering how could it work? A bug or I missed anything?
It is really just used for informative purpose now.
Thanks,
john
Lu Baolu
maintainerRe: [PATCH v13 6/6] iommu: Remove mode argument from iommu_set_dma_strict()
Quoted 32 lines · John Garry
Quoted 17 lines
drivers/iommu/iommu.c static int __init iommu_dma_setup(char *str) · line 353
Quoted 10 lines
drivers/iommu/iommu.c +2
-3@@ -349,10 +349,9 @@ static int __init iommu_dma_setup(char *str) } early_param("iommu.strict", iommu_dma_setup); -void iommu_set_dma_strict (bool strict) +void iommu_set_dma_strict (void ) { - if (strict || !(iommu_cmd_line & IOMMU_CMD_LINE_STRICT)) - iommu_dma_strict = strict; + iommu_dma_strict = true;Sorry, I still can't get how iommu.strict kernel option works.
static int
__initiommu_dma_setup(char *str){
int ret =kstrtobool(str, &iommu_dma_strict);Quoted 7 lines
if (!ret)
iommu_cmd_line|=IOMMU_CMD_LINE_STRICT; return ret;
}early_param("iommu.strict", iommu_dma_setup);The bit
IOMMU_CMD_LINE_STRICTis only set, but not used anywhere.It is used in patch 2/6:
+ pr_info("DMA domain TLB invalidation policy: %s mode %s\n", +
iommu_dma_strict? "strict" : "lazy",
+ (iommu_cmd_line&IOMMU_CMD_LINE_STRICT) ?
+ "(set via kernel command line)" : "");
Quoted 4 lines
Quoted 2 lines
Hence,
I am wondering how could it work? A bug or I missed anything?It is really just used for informative purpose now.
I am clear now. Thanks!
Best regards,
baolu
Robin Murphy
reviewerRe: [PATCH v13 6/6] iommu: Remove mode argument from iommu_set_dma_strict()
In reply to Lu BaoluQuoted 75 lines · Lu Baolu
Quoted 63 lines · John Garry
We only ever now set strict mode enabled in
iommu_set_dma_strict(), so just remove the argument.Signed-off-by John Garry <[email protected]>Reviewed-by Robin Murphy <[email protected]>---
drivers/iommu/amd/init.c | 2 +-
drivers/iommu/intel/iommu.c| 6 +++---
drivers/iommu/iommu.c | 5 ++---
include/linux/iommu.h | 2 +-4 files changed, 7 insertions(+), 8 deletions(-)drivers/iommu/amd/init.c +1
-1@@ -3099,7 +3099,7 @@ static int __init parse_amd_iommu_options(char *str) for (; *str; ++str) { if (strncmp(str, "fullflush", 9) == 0) { pr_warn("amd_iommu=fullflush deprecated; use iommu.strict instead\n"); - iommu_set_dma_strict(true); + iommu_set_dma_strict(); } if (strncmp(str, "force_enable", 12) == 0)
amd_iommu_force_enable= true;drivers/iommu/intel/iommu.c +3
-3@@ -454,7 +454,7 @@ static int __init intel_iommu_setup(char *str) iommu_dma_forcedac = true; } else if (!strncmp(str, "strict", 6)) { pr_warn("intel_iommu=strict deprecated; use iommu.strict instead\n"); - iommu_set_dma_strict(true); + iommu_set_dma_strict(); } else if (!strncmp(str, "sp_off", 6)) { pr_info("Disable supported super page\n"); intel_iommu_superpage = 0;@@ -4382,7 +4382,7 @@ int __init intel_iommu_init(void)*/ if (cap_caching_mode(iommu->cap)) { pr_warn("IOMMU batching disallowed due to virtualization\n"); - iommu_set_dma_strict(true); + iommu_set_dma_strict(); } iommu_device_sysfs_add(&iommu->iommu, NULL, intel_iommu_groups,@@ -5699,7 +5699,7 @@ static void quirk_calpella_no_shadow_gtt(struct pci_dev *dev) } else if (dmar_map_gfx) { /* we have to ensure the gfx device is idle before we flush */ pci_info(dev, "Disabling batched IOTLB flush on Ironlake\n"); - iommu_set_dma_strict(true); + iommu_set_dma_strict(); } }
DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0040, quirk_calpella_no_shadow_gtt);drivers/iommu/iommu.c +0
-0drivers/iommu/iommu.c static int __init iommu_dma_setup(char *str) · line 353
Quoted 8 lines
drivers/iommu/iommu.c +2
-3@@ -349,10 +349,9 @@ static int __init iommu_dma_setup(char *str) } early_param("iommu.strict", iommu_dma_setup); -void iommu_set_dma_strict (bool strict) +void iommu_set_dma_strict (void ) { - if (strict || !(iommu_cmd_line & IOMMU_CMD_LINE_STRICT)) - iommu_dma_strict = strict; + iommu_dma_strict = true;Sorry, I still can't get how iommu.strict kernel option works.
Quoted 3 lines
static int
__initiommu_dma_setup(char *str){
int ret =kstrtobool(str, &iommu_dma_strict);
Note that this is the bit that does the real work - if the argument parses OK then iommu_dma_strict is reassigned with the appropriate value. The iommu_cmd_line stuff is a bit of additional bookkeeping, basically just so we can see whether default values have been overridden.
Robin.
Quoted 28 lines
if (!ret)
iommu_cmd_line|=IOMMU_CMD_LINE_STRICT; return ret;
}early_param("iommu.strict", iommu_dma_setup);The bit
IOMMU_CMD_LINE_STRICTis only set, but not used anywhere. Hence, I am wondering how could it work? A bug or I missed anything?Best regards,
baoluQuoted 15 lines
drivers/iommu/iommu.c +0
-0} bool iommu_get_dma_strict(struct iommu_domain *domain)include/linux/iommu.h +1
-1@@ -476,7 +476,7 @@ int iommu_enable_nesting(struct iommu_domain *domain); int iommu_set_pgtable_quirks(struct iommu_domain *domain, unsigned long quirks); -void iommu_set_dma_strict (bool val); +void iommu_set_dma_strict (void ); bool iommu_get_dma_strict(struct iommu_domain *domain); extern int report_iommu_fault(struct iommu_domain *domain, struct device *dev,
Lu Baolu
maintainerRe: [PATCH v13 6/6] iommu: Remove mode argument from iommu_set_dma_strict()
Hi Robin,
Quoted 16 lines · Robin Murphy
Quoted 16 lines
drivers/iommu/iommu.c static int __init iommu_dma_setup(char *str) · line 353
Quoted 13 lines
drivers/iommu/iommu.c +2
-3@@ -349,10 +349,9 @@ static int __init iommu_dma_setup(char *str) } early_param("iommu.strict", iommu_dma_setup); -void iommu_set_dma_strict (bool strict) +void iommu_set_dma_strict (void ) { - if (strict || !(iommu_cmd_line & IOMMU_CMD_LINE_STRICT)) - iommu_dma_strict = strict; + iommu_dma_strict = true;Sorry, I still can't get how iommu.strict kernel option works.
Quoted 5 lines
Quoted 3 lines
static int
__initiommu_dma_setup(char *str){
int ret =kstrtobool(str, &iommu_dma_strict);Note that this is the bit that does the real work - if the argument parses OK then
iommu_dma_strictis reassigned with the appropriate value. Theiommu_cmd_linestuff is a bit of additional bookkeeping, basically just so we can see whether default values have been overridden.
Ah, get it. Thanks a lot. I missed this part and naively thought it just converts a string to integer.
Best regards,
baolu