Re: [PATCH v2 1/3] iommu/amd: Do not request ACS when IOMMU is not going to be initialized

[Rong Zhang <i@rong.moe>]

Hi Sairaj,

On Fri, 2026-08-21 at 16:12 +0530, Sairaj Kodilkar wrote:
> On 8/20/2026 11:55 PM, Rong Zhang wrote:
> > The AMD IOMMU Initialization State Machine has the following state
> > transition diagram (only the very first states are showed, and the
> > `IOMMU_' prefix is omitted):
> > 
> >    START_STATE
> >         |
> >         v
> > [0] detect_ivrs() --> NOT_FOUND
> >         | ok
> >         v
> >   IVRS_DETECTED
> >         |
> >         v
> > [1] amd_iommu_disabled? (amd_iommu=off) --> IOMMU_CMDLINE_DISABLED
> >         | no
> >         v
> > [2] early_amd_iommu_init()
> >         |
> >         +-- [3] amd_iommu_detected? (!iommu=off && ...) -+
> >         | yes                                            |
> >         +-- ... --> IOMMU_INIT_ERROR <-------------------+
> >         | ok
> >         v
> > IOMMU_ACPI_FINISHED
> >         |
> >         v
> >        ...
> > 
> > [0] always calls pci_request_acs() as long as there's a valid IVRS table
> > and no Stoney Ridge graphics. This is not optimal as ACS is not required
> > in an [amd_]iommu=off boot.
> > 
> > In a normal boot, ACS is requested due to amd_iommu_detect() requesting
> > IVRS_DETECTED.
> > 
> >     pci_request_acs+0x9/0x18
> >     iommu_go_to_state+0x106/0x1a20
> >     amd_iommu_detect+0x1c/0x50
> >     pci_iommu_alloc+0x26/0x40
> >     mm_core_init+0xa/0x120
> >     start_kernel+0x527/0x7a0
> >     x86_64_start_reservations+0x24/0x30
> >     x86_64_start_kernel+0xd1/0xe0
> >     common_startup_64+0x13e/0x158
> > 
> > This is intended to ensure ACS is requested before the PCI core
> > initialization, or else a !CONFIG_IRQ_REMAP, nointremap or intremap=off
> > boot would be broken.
> > 
> > However, in an amd_iommu=off boot, the state machine still requests ACS
> > at the exact same time, as amd_iommu_detect() has nothing to do with
> > amd_iommu_disabled.
> > 
> > Even worse, in an iommu=off boot, though amd_iommu_detect() bails out
> > early, ACS is still requested due to amd_iommu_prepare() requesting
> > IOMMU_ACPI_FINISHED, which is called by irq_remapping_prepare() thanks
> > to CONFIG_X86_LOCAL_APIC (always set on X86_64) and CONFIG_IRQ_REMAP
> > (enabled by defconfig), unless nointremap or intremap=off is also passed
> > to cmdline.
> > 
> >     pci_request_acs+0x9/0x18
> >     iommu_go_to_state+0x106/0x1a20
> >     amd_iommu_prepare+0x15/0x40
> >     irq_remapping_prepare+0x43/0x60
> >     enable_IR_x2apic+0x22/0x190
> >     x86_64_probe_apic+0xa/0x50
> >     apic_intr_mode_init+0x70/0xd0
> >     x86_late_time_init+0x28/0x40
> >     start_kernel+0x6f9/0x7a0
> >     ...
> > 
> > In both cases, [2] is still gated due to the [1] or [3] check, so that
> > IOMMU can be disabled per cmdline.
> > 
> > Technically, it makes no sense to detect IVRS at all in an
> > [amd_]iommu=off boot or if IOMMU is not supported due to platform
> > settings. This is probably why amd_iommu_detect() bails out before
> > requesting IVRS_DETECTED. Apparently only bailing out there is not
> > sufficient, and the bailing-out paths should really have been parts of
> > the state machine.
> > 
> > Clean up the initialization routines by moving the bailing-out paths and
> > [1] to the right place in the state machine (i.e., before [0]), and
> > always requesting IVRS_DETECTED in amd_iommu_detect() to initialize the
> > state machine early and properly.
> > 
> > Signed-off-by: Rong Zhang <i@rong.moe>
> > ---
> >  drivers/iommu/amd/init.c | 25 +++++++++++--------------
> >  1 file changed, 11 insertions(+), 14 deletions(-)
> > 
> > diff --git a/drivers/iommu/amd/init.c b/drivers/iommu/amd/init.c
> > index 40726dfef273..a720796cca3b 100644
> > --- a/drivers/iommu/amd/init.c
> > +++ b/drivers/iommu/amd/init.c
> > @@ -3470,6 +3470,8 @@ static void amd_iommu_apply_erratum_snp(void)
> >  #endif
> >  }
> >  
> > +static bool amd_iommu_sme_check(void);
> > +
> >  /****************************************************************************
> >   *
> >   * AMD IOMMU Initialization State Machine
> > @@ -3482,7 +3484,13 @@ static int __init state_next(void)
> >  
> >  	switch (init_state) {
> >  	case IOMMU_START_STATE:
> > -		if (!detect_ivrs()) {
> > +		if (no_iommu || amd_iommu_disabled) {
> > +			init_state	= IOMMU_CMDLINE_DISABLED;
> > +			ret		= -EINVAL;
> > +		} else if ((iommu_detected && !gart_iommu_aperture) || !amd_iommu_sme_check()) {
> 
> Looks like you can drop the condition
> (iommu_detected && !gart_iommu_aperture)
> 
> It was added by the commit 6631ee9d00, which set iommu_detected = 1 and
> gart_iommu_aperture = 0 in iommu driver. But it is later removed and no
> longer present in the lastest iommu code.
> 
> Latest code can have following two conditions
> 
> 1. When CONFIG_GART_IOMMU=y
> x86/kernel/aperture_64.c sets both iommu_detected and
> gart_iommu_aperture to 1
> 
> 2. When CONFIG_GART_IOMMU=n
> amd_iommu_detect is called only when both iommu_detected and
> gart_iommu_aperture are zero.

Makes sense. Will clean it up in v3.

Thanks,
Rong

> 
> Thanks
> Sairaj.
> 
> 
> 
> > +			init_state	= IOMMU_INIT_ERROR;
> > +			ret		= -EINVAL;
> > +		} else if (!detect_ivrs()) {
> >  			init_state	= IOMMU_NOT_FOUND;
> >  			ret		= -ENODEV;
> >  		} else {
> > @@ -3490,13 +3498,8 @@ static int __init state_next(void)
> >  		}
> >  		break;
> >  	case IOMMU_IVRS_DETECTED:
> > -		if (amd_iommu_disabled) {
> > -			init_state = IOMMU_CMDLINE_DISABLED;
> > -			ret = -EINVAL;
> > -		} else {
> > -			ret = early_amd_iommu_init();
> > -			init_state = ret ? IOMMU_INIT_ERROR : IOMMU_ACPI_FINISHED;
> > -		}
> > +		ret = early_amd_iommu_init();
> > +		init_state = ret ? IOMMU_INIT_ERROR : IOMMU_ACPI_FINISHED;
> >  		break;
> >  	case IOMMU_ACPI_FINISHED:
> >  		early_enable_iommus();
> > @@ -3695,12 +3698,6 @@ void __init amd_iommu_detect(void)
> >  {
> >  	int ret;
> >  
> > -	if (no_iommu || (iommu_detected && !gart_iommu_aperture))
> > -		goto disable_snp;
> > -
> > -	if (!amd_iommu_sme_check())
> > -		goto disable_snp;
> > -
> >  	ret = iommu_go_to_state(IOMMU_IVRS_DETECTED);
> >  	if (ret)
> >  		goto disable_snp;
> >