Re: [RFC PATCH v1 0/9] deferred_probe_timeout logic clean up

On 2022-05-26 01:15:39 [-0700], Saravana Kannan wrote: > Yoshihiro/Geert, Hi Saravana, > If you can test this patch series and confirm that the NFS root case > works, I'd really appreciate that. The two patches you sent earlier, plus this series, plus diff --git a/drivers/base/core.c

Re: [PATCH v1] driver core: Extend deferred probe timeout on driver registration

On 2022-05-24 10:46:49 [-0700], Saravana Kannan wrote: > > Removing probe_timeout_waitqueue (as suggested) or setting the timeout > > to 0 avoids the delay. > > In your case, I think it might be working as intended? Curious, what > was the call stack in your case where it was blocked? Why is

Re: [PATCH v1] driver core: Extend deferred probe timeout on driver registration

On 2022-05-23 20:43:06 [-0700], Saravana Kannan wrote: … > Thanks for all the help. I think I know what's going on. I, too got here because my boot recently was extended by 10 seconds and bisected to that commit in question. > If you revert the following commit, then you'll see that your device

Re: [RFC PATCH v9 02/13] x86: always set IF before oopsing from page fault

- stepping on del button while browsing though CCs. On 2019-04-04 09:47:27 [-0600], Tycho Andersen wrote: > > Hmm. do_exit() isn't really meant to be "try your best to leave the > > system somewhat usable without returning" -- it's a function that, > > other than in OOPSes, is called from a

Re: [PATCH] iommu/amd: fix unused symbol iommu_table_lock

On 2018-05-06 11:53:26 [+0200], Danilo Krummrich wrote: > When CONFIG_IRQ_REMAP is not set iommu_table_lock is unused and hece > a warning is generated. > > Fixes: ea6166f4b83e9 ("iommu/amd: Split irq_lookup_table out of the > amd_iommu_devtable_lock") > Signed-off-by: Danilo Krummrich

[PATCH 2/3] iommu/amd: Cleanup locking in __attach/detach_device()

mplementation of those functions. Add lockdep checks where appropriate. Signed-off-by: Anna-Maria Gleixner <anna-ma...@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 70 +-- 1 file chang

[PATCH 3/3] iommu/amd: Do not flush when device is busy

; Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 32 ++-- 1 file changed, 10 insertions(+), 22 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index f66a5d0b7c62..a801678ae1b4 100644 -

[PATCH 1/3] iommu/amd: Prevent possible null pointer dereference and infinite loop

g. Move the check with the explanation of the do_detach() code into the caller detach_device() and return immediately. Throw an error, when hitting the condition in cleanup_domain(). Signed-off-by: Anna-Maria Gleixner <anna-ma...@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bige...@linutr

iommu/amd: avoid irqs_disabled() check

Primary motivation was to get rid of the "WARN_ON(!irqs_disabled());" check. The first patch avoids a possible loop (if cleanup_domain() is invoked `entry->domain == NULL' then it loops for ever). The irqs_disabled() check has been replaced with a lockdep_assert_held() check. Sebastian

Re: [PATCH] iommu/amd: fix unused-variable warning

On 2018-04-20 11:28:36 [+0200], Tobias Regnery wrote: > The iommu_table_lock is only used by code inside an ifdef CONFIG_IRQ_REMAP > block. This leads to the following warning with CONFIG_IRQ_REMAP=n: > > amd_iommu.c:86:24: warning: 'iommu_table_lock' defined but not used > [-Wunused-variable] >

Re: [PATCH] iommu: amd: hide unused iommu_table_lock

ommu_table_lock); > > This moves the definition next to the user, within the #ifdef protected > section of the file. > > Fixes: ea6166f4b83e ("iommu/amd: Split irq_lookup_table out of the > amd_iommu_devtable_lock") > Signed-off-by: Arnd Bergmann <a...@arndb.de> Acked-

[PATCH 08/10] iommu/amd: drop the lock while allocating new irq remap table

hile the lock is dropped since the same device can only be probed once. However I check for both cases, just to be sure. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 65 +-- 1 file changed, 4

[PATCH 05/10] iommu/amd: remove the special case from alloc_irq_table()

irq_domain_mutex so the initialization of iommu->irte_ops->set_allocated() should not race against other user. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 34 -- 1 file changed, 20 insertions(+),

[PATCH 09/10] iommu/amd: make amd_iommu_devtable_lock a spin_lock

change its type to a spin_lock. I *think* that we might even be able to remove the lock because all its current user seem to have their own protection. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 14 +++--- 1 file changed, 7

[PATCH 07/10] iommu/amd: factor out setting the remap table for a devid

. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 23 --- 1 file changed, 12 insertions(+), 11 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index ea120c7b46c9..11ea2d656be8 100644 --- a/d

[PATCH 10/10] iommu/amd: return proper error code in irq_remapping_alloc()

1' instead `-ENOMEM' what was intended. In order to correct this, I assign -ENOMEM to index. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/iommu/amd_iommu.c

[PATCH 06/10] iommu/amd: use `table' instead `irt' as variable name in amd_iommu_update_ga()

The variable of type struct irq_remap_table is always named `table' except in amd_iommu_update_ga() where it is called `irt'. Make it consistent and name it also `table'. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 10 +-

[PATCH 04/10] iommu/amd: split irq_lookup_table out of the amd_iommu_devtable_lock

zed with its own (iommu's) lock. So split out get_irq_table() out of amd_iommu_devtable_lock's lock. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/iommu/amd_iommu.c

[PATCH 03/10] iommu/amd: split domain id out of amd_iommu_devtable_lock

-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 12 +--- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index d4c2b1a11924..fcfdce70707d 100644 --- a/drivers/iommu/amd_iommu.c +++ b/d

[PATCH 02/10] iommu/amd: turn dev_data_list into a lock less list

it. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 28 ++-- drivers/iommu/amd_iommu_types.h | 2 +- 2 files changed, 11 insertions(+), 19 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c

[PATCH 01/10] iommu/amd: take into account that alloc_dev_data() may return NULL

s required. Cc: Baoquan He <b...@redhat.com> Fixes: df3f7a6e8e85 ("iommu/amd: Use is_attach_deferred call-back") Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/i

[PATCH 00/10 v3] iommu/amd: lock splitting & GFP_KERNEL allocation

Hi, this is the rebase on top of iommu/x86/amd of my last series. It takes Scotts comments into consideration from my v2. It contains lock splitting and GFP_KERNEL allocation of remap-table. Mostly cleanup. The patches were boot tested on an AMD EPYC 7601. Sebastian

Re: [PATCH 00/10 v2] iommu/amd: lock splitting & GFP_KERNEL allocation

On 2018-03-17 16:43:39 [-0500], Scott Wood wrote: > If that's worth the lock dropping then fine (though why does only one > of the two allocations use GFP_KERNEL?), but it doesn't need to be a That was a mistake, I planned to keep both as GFP_KERNEL. > raw lock if the non-allocating users are

Re: [PATCH 00/10 v2] iommu/amd: lock splitting & GFP_KERNEL allocation

On 2018-03-17 14:49:54 [-0500], Scott Wood wrote: > On Fri, 2018-03-16 at 21:18 +0100, Sebastian Andrzej Siewior wrote: > > The goal here is to make the memory allocation in get_irq_table() not > > with disabled interrupts and having as little raw_spin_lock as > > possible

[PATCH 08/10] iommu/amd: factor out setting the remap table for a devid

. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 23 --- 1 file changed, 12 insertions(+), 11 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index 6ee8ef22ad51..7dd4c27a941d 100644 --- a/d

[PATCH 06/10] iommu/amd: remove the special case from get_irq_table()

get_irq_table() has a special ioapic argument. If set then it will pre-allocate / reserve the first 32 indexes. The argument is only once true and it would make get_irq_table() a little simpler if we would extract the special bits to the caller. Signed-off-by: Sebastian Andrzej Siewior <b

[PATCH 00/10 v2] iommu/amd: lock splitting & GFP_KERNEL allocation

The goal here is to make the memory allocation in get_irq_table() not with disabled interrupts and having as little raw_spin_lock as possible while having them if the caller is also holding one (like desc->lock during IRQ-affinity changes). I reverted one patch one patch in the iommu while

[PATCH 10/10] iommu/amd: make amd_iommu_devtable_lock a spin_lock

change its type to a spin_lock. I *think* that we might even be able to remove the lock because all its current user seem to have their own protection. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 14 +++--- 1 file changed, 7

[PATCH 09/10] iommu/amd: drop the lock while allocating new irq remap table

ot;devid" entry appears in irq_lookup_table while the lock is dropped since the same device can only be probed once. It is more likely that another device added an `alias' entry. However I check for both cases, just to be sure. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de>

[PATCH 07/10] iommu/amd: use `table' instead `irt' as variable name in amd_iommu_update_ga()

The variable of type struct irq_remap_table is always named `table' except in amd_iommu_update_ga() where it is called `irt'. Make it consistent and name it also `table'. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 10 +-

[PATCH 05/10] Revert "iommu/amd: Avoid locking get_irq_table() from atomic context"

t Wood <sw...@redhat.com> Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 29 +++-- 1 file changed, 7 insertions(+), 22 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index 5191319d9f0

[PATCH 04/10] iommu/amd: split irq_lookup_table out of the amd_iommu_devtable_lock

zed with its own (iommu's) lock. So split out get_irq_table() out of amd_iommu_devtable_lock's lock. The new lock is a raw_spin_lock because modify_irte_ga() is called while desc->lock is held (which is raw). Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd

[PATCH 03/10] iommu/amd: split domain id out of amd_iommu_devtable_lock

-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 12 +--- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index 0e57ce2c258b..692b2e3b9af1 100644 --- a/drivers/iommu/amd_iommu.c +++ b/d

[PATCH 01/10] iommu/amd: take into account that alloc_dev_data() may return NULL

s required. Cc: Baoquan He <b...@redhat.com> Fixes: df3f7a6e8e85 ("iommu/amd: Use is_attach_deferred call-back") Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/i

[PATCH 02/10] iommu/amd: turn dev_data_list into a lock less list

it. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 28 ++-- drivers/iommu/amd_iommu_types.h | 2 +- 2 files changed, 11 insertions(+), 19 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c

Re: [PATCH 05/10] iommu/amd: remove the special case from get_irq_table()

On 2018-03-15 16:19:17 [+0100], Joerg Roedel wrote: > Okay, if the irq-layer does the needed locking, then we don't need > another lock here. There is the modify_irte_ga() path for the > virtualized irq routing into KVM guests, but there should be no KVM > guests running when setup the ioapic

Re: [PATCH 05/10] iommu/amd: remove the special case from get_irq_table()

On 2018-03-15 13:53:42 [+0100], Joerg Roedel wrote: > On Fri, Feb 23, 2018 at 11:27:31PM +0100, Sebastian Andrzej Siewior wrote: > > @@ -4103,10 +4093,26 @@ static int irq_remapping_alloc(struct irq_domain > > *domain, unsigned int virq, > > return ret; > &g

[PATCH 10/10] iommu/amd: make amd_iommu_devtable_lock a spin_lock

change its type to a spin_lock. I *think* that we might even be able to remove the lock because all its current user seem to have their own protection. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 14 +++--- 1 file changed, 7

[PATCH 05/10] iommu/amd: remove the special case from get_irq_table()

get_irq_table() has a special ioapic argument. If set then it will pre-allocate / reserve the first 32 indexes. The argument is only once true and it would make get_irq_table() a little simpler if we would extract the special bits to the caller. Signed-off-by: Sebastian Andrzej Siewior <b

[PATCH 09/10] iommu/amd: declare irq_remap_table's and amd_iommu's lock as a raw_spin_lock

te and flush the iommu. The latter is also required during table allocation. I currently don't see a feasible way of getting this done without turning both locks raw so here it is. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_

[PATCH 01/10] iommu/amd: take into account that alloc_dev_data() may return NULL

s required. Cc: Baoquan He <b...@redhat.com> Fixes: df3f7a6e8e85 ("iommu/amd: Use is_attach_deferred call-back") Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/i

[PATCH 06/10] iommu/amd: use `table' instead `irt' as variable name in amd_iommu_update_ga()

The variable of type struct irq_remap_table is always named `table' except in amd_iommu_update_ga() where it is called `irt'. Make it consistent and name it also `table'. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 10 +-

[PATCH 07/10] iommu/amd: factor out setting the remap table for a devid

. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 23 --- 1 file changed, 12 insertions(+), 11 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index d3a05d794218..b763fcbd790d 100644 --- a/d

[PATCH 08/10] iommu/amd: drop the lock while allocating new irq remap table

ot;devid" entry appears in irq_lookup_table while the lock is dropped since the same device can only be probed once. It is more likely that another device added an `alias' entry. However I check for both cases, just to be sure. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de>

iommu/amd: lock splitting & GFP_KERNEL allocation

Hi, I have no idea why but suddenly my A10 box complained loudly about locking and memory allocations within the iommu code under RT. Looking at the code it has been like this for a longer time so the iommu must have appeared recently (well there was a bios upgrade due to other issues so it might

[PATCH 04/10] iommu/amd: split irq_lookup_table out of the amd_iommu_devtable_lock

zed with its own (iommu's) lock. So split out get_irq_table() out of amd_iommu_devtable_lock's lock. The new lock is a raw_spin_lock because modify_irte_ga() is called while desc->lock is held (which is raw). Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd

[PATCH 03/10] iommu/amd: split domain id out of amd_iommu_devtable_lock

-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 12 +--- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index 53aece41ddf2..958efe311057 100644 --- a/drivers/iommu/amd_iommu.c +++ b/d

[PATCH 02/10] iommu/amd: turn dev_data_list into a lock less list

it. Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 28 ++-- drivers/iommu/amd_iommu_types.h | 2 +- 2 files changed, 11 insertions(+), 19 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c

Re: [PATCH v2] iommu/iova: Use raw_cpu_ptr() instead of get_cpu_ptr() for ->fq

On 2017-09-21 17:21:40 [+0200], Sebastian Andrzej Siewior wrote: > get_cpu_ptr() disabled preemption and returns the ->fq object of the > current CPU. raw_cpu_ptr() does the same except that it not disable > preemption which means the scheduler can move it to another CPU after i

[PATCH v2] iommu/iova: Use raw_cpu_ptr() instead of get_cpu_ptr() for ->fq

ure itself is protected with a spin_lock. This change shouldn't matter however on RT it does because the sleeping lock can't be accessed with disabled preemption. Cc: Joerg Roedel <j...@8bytes.org> Cc: iommu@lists.linux-foundation.org Reported-by: vina...@gmail.com Signed-off-by: Sebastian Andrze

Re: [PATCH] iommu/amd: Use raw_cpu_ptr() instead of get_cpu_ptr() for ->flush_queue

On 2017-09-11 22:22:11 [-0400], Vinod Adhikary wrote: > Dear all, Hi, > Thank you for the great community support and support from Sebastian to > provide me this patch. I wanted to send this email to inform you and > perhaps get some information on how I could keep myself updated on updates > in

[PATCH] iommu/amd: Use raw_cpu_ptr() instead of get_cpu_ptr() for ->flush_queue

Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/amd_iommu.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index 4ad7e5e31943..943efbc08128 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/io

Re: [PATCH 2/3] iommu/iova: don't disable preempt around this_cpu_ptr()

On 2017-06-28 11:22:05 [+0200], Joerg Roedel wrote: > On Tue, Jun 27, 2017 at 06:16:47PM +0200, Sebastian Andrzej Siewior wrote: > > Commit 583248e6620a ("iommu/iova: Disable preemption around use of > > this_cpu_ptr()") disables preemption while accessing a per-CPU var

[PATCH 2/3] iommu/iova: don't disable preempt around this_cpu_ptr()

r CPU down (and none for up) so we are good. Cc: Joerg Roedel <j...@8bytes.org> Cc: iommu@lists.linux-foundation.org Cc: Andrew Morton <a...@linux-foundation.org> Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/iova.c | 9 +++-- 1 file changed,

[PATCH 3/3] iommu/vt-d: don't disable preemption while accessing deferred_flush()

<dw...@infradead.org> Cc: Joerg Roedel <j...@8bytes.org> Cc: iommu@lists.linux-foundation.org Cc: Andrew Morton <a...@linux-foundation.org> Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/intel-iommu.c | 8 ++-- 1 file changed, 2 insertions(+),

[PATCH 4/5] iommu/iova: don't disable preempt around this_cpu_ptr()

different CPU (which means he needs protection against concurrent access). Cc: Joerg Roedel <j...@8bytes.org> Cc: iommu@lists.linux-foundation.org Cc: Chris Wilson <ch...@chris-wilson.co.uk> Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de> --- drivers/iommu/iova.c |