Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
From: Ian Munsie
The afu fd release path was identified as a significant bottleneck in
the overall performance of cxl. While an optimal AFU design would
minimise the need to close & reopen the AFU fd, it is not always
practical to avoid.
The bottleneck seems to be down to the cal
Hi Mikey,
> +/* wrappers around afu_* file ops which are EXPORTED */
This is fine, though alternatively you could export the original
functions directly from file.c (feel free to rename them to your
versions if you do change it) - I don't really mind either way :)
> +static void cxl_pci_reset_se
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Looks like a good change to make sure the PSL & AFU is not still active
across a reboot :)
Wondering if this should be Cc: Stable - has there been any demonstrable
stability increase after a reboot with this?
Acked-by: Ian Munsie
Excerpts from Michael Neuling's message of 2015-05-19
I think this is missing the From line to indicate I wrote it ;)
Otherwise...
Acked-by: Ian Munsie
;-)
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Hooray, one previously useless level of indirection now has a purpose :)
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
> +void assign_psn_space(struct cxl_context *ctx);
Should we consider renaming this to cxl_assign_psn_space to namespace it
now that it's potentially callable from elsewhere in the kernel?
> +int afu_check_and_enable(struct cxl_afu *afu);
Likewise?
Acked-by: Ian Mu
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Excerpts from Michael Neuling's message of 2015-05-19 21:22:32 +1000:
> +if (ctx->kernel) {
> +sr |= CXL_PSL_SR_An_R | (mfmsr() & MSR_SF) | CXL_PSL_SR_An_HV;
> +pid = 0;
> +} else {
> +sr |= CXL_PSL_SR_An_PR | CXL_PSL_SR_An_R;
CXL_PSL_SR_An_R is common between these
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
Cheers,
-Ian
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Hey Michael,
Excerpts from Michael Ellerman's message of 2014-10-02 16:02:37 +1000:
> > +/* ioctls */
> > +struct cxl_ioctl_start_work {
> > +__u64 wed;
> > +__u64 amr;
> > +__u64 reserved1;
> > +__u32 reserved2;
> > +__s16 num_interrupts; /* -1 = use value from afu descriptor
Excerpts from Michael Neuling's message of 2014-10-08 19:55:02 +1100:
> +static ssize_t afu_read(struct file *file, char __user *buf, size_t count,
> +loff_t *off)
...
> +for (;;) {
> +prepare_to_wait(&ctx->wq, &wait, TASK_INTERRUPTIBLE);
> +if (ctx_event_pending(ctx
correctly.
Signed-off-by: Ian Munsie
---
drivers/misc/cxl/file.c | 20 +++-
1 file changed, 15 insertions(+), 5 deletions(-)
diff --git a/drivers/misc/cxl/file.c b/drivers/misc/cxl/file.c
index 847b7e6..378b099 100644
--- a/drivers/misc/cxl/file.c
+++ b/drivers/misc/cxl/file.c
From: Ian Munsie
If afu_read() returned due to a signal or the AFU file descriptor being
opened non-blocking it would not call finish_wait() before returning,
which could lead to a crash later when something else wakes up the wait
queue.
This patch restructures the wait logic to ensure that the
From: Ian Munsie
In certain circumstances the PSL can send an interrupt for a segment
miss that the kernel has already handled. This can happen if multiple
translations for the same segment are queued in the PSL before the
kernel has restarted the first translation.
The CXL driver did not
Excerpts from Michael Ellerman's message of 2014-10-27 17:41:00 +1100:
> On Mon, 2014-27-10 at 04:24:35 UTC, Ian Munsie wrote:
> > From: Ian Munsie
> >
> > In certain circumstances the PSL can send an interrupt for a segment
>
> Define PSL before using it please.
Excerpts from Aneesh Kumar K.V's message of 2014-10-28 01:38:41 +1100:
> I guess you are missing too many fixes in one patch.
>
> 1) One cleanup
> 2) Fix for masking ea correctly
> 3) And fix for not erroring out when a slb is already in the slb cache.
ok, I'll split it up
> > +/* This finds a f
From: Ian Munsie
This patch simplifies the process of finding a free segment table entry
by disabling the secondary hash. This reduces the number of possible
entries in the segment table for a given address from 16 to 8.
Due to the large segment sizes we use it is extremely unlikely that the
From: Ian Munsie
In certain circumstances the PSL (Power Service Layer, which provides
translation services for CXL hardware) can send an interrupt for a
segment miss that the kernel has already handled. This can happen if
multiple translations for the same segment are queued in the PSL before
In certain circumstances the PSL (Power Service Layer, which provides
translation services for CXL hardware) can send an interrupt for a segment miss
that the kernel has already handled. This can happen if multiple translations
for the same segment are queued in the PSL before the kernel has restar
From: Ian Munsie
This patch makes copro_calculate_slb mask the ESID by the correct mask
for 1T vs 256M segments.
This has no effect by itself as the extra bits were ignored, but it
makes debugging the segment table entries easier and means that we can
directly compare the ESID values for
From: Ian Munsie
This moves the segment table hash calculation from cxl_load_segment into
find_free_sste since that is the only place it is actually used.
Signed-off-by: Ian Munsie
---
drivers/misc/cxl/fault.c | 34 ++
1 file changed, 18 insertions(+), 16
From: Ian Munsie
If an AFU has a hardware bug that causes it to acknowledge a context
terminate or remove while that context has outstanding transactions, it
is possible for the kernel to receive an interrupt for that context
after we have removed it from the context list.
The kernel will not
1066 OPAL ICS 2050 Level cxl-afu0.0
514: 0 OPAL ICS 2048 Level cxl-card0-err
515: 0 OPAL ICS 2049 Level cxl-afu0.0-err
Signed-off-by: Michael Neuling
Signed-off-by: Ian Munsie
---
drivers/misc/cxl/cxl.h | 13 +--
drivers/misc
From: Ian Munsie
We had a known sleep while atomic bug if a CXL device was forcefully
unbound while it was in use. This could occur as a result of EEH, or
manually induced with something like this while the device was in use:
echo :01:00.0 > /sys/bus/pci/drivers/cxl-pci/unbind
The is
From: Ian Munsie
In the event that something goes wrong in the hardware and it is unable
to complete a process element comment we would end up polling forever,
effectively making the associated process unkillable.
This patch adds a timeout to the process element command code path, so
that we
From: Ian Munsie
If a context is being detached and we get a translation fault for it
there is little point getting it's mm and handling the fault, so just
respond with an address error and return earlier.
Signed-off-by: Ian Munsie
---
drivers/misc/cxl/fault.c | 6 ++
1 file chang
From: Ian Munsie
Upon inspection of the implementation specific registers, it was
discovered that the high bit of the implementation specific RXCTL
register was enabled, which enables the DEADB00F debug feature.
The debug feature causes MMIO reads to a disabled AFU to respond with
0xDEADB00F
From: Ian Munsie
When we deactivate the AFU directed mode we free the scheduled process
area, but did not clear the register in the hardware that has a pointer
to it.
This should be fine since we will have already cleared out every context
and we won't do anything that would cause the har
From: Ian Munsie
In this particular error path we have already allocated the AFU
interrupts, but have not yet set the status to STARTED. The detach
context code will only attempt to release the interrupts if the context
is in state STARTED, so in this case the interrupts would remain
allocated
From: Ian Munsie
If we need to force detach a context (e.g. due to EEH or simply force
unbinding the driver) we should prevent the userspace contexts from
being able to access the Problem State Area MMIO region further, which
they may have mapped with mmap().
This patch unmaps any mapped MMIO
This patch will definitely need to go to stable - we've run into issues
a couple of times when something has gone wrong on an AFU and ended up
taking down the whole system as a result of this bug.
Cheers,
-Ian
___
Linuxppc-dev mailing list
Linuxppc-dev@
This one needs to go to stable - I've hit it a couple of times while
testing bad AFUs and it results in an unkillable process.
Cheers,
-Ian
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
This one would be nice to go to stable, but I'm not sure if it's
critical enough to justify it since it only reduces the number of
available interrupts (and therefore, contexts) that can be used by the
card (so, maybe you can only run 507 contexts simultaneously instead of
509)...
Cheers,
-Ian
__
This one would be nice to go to stable, but I'm not sure it really meets
the rules. It could be a problem for userspace error paths checking the
result of MMIO reads, but only if the AFU has actually been unexpectedly
disabled somehow yet the PSL is still responding...
I don't think this is a high
This one should go to stable - this was the first bug uncovered after
fixing the sleep while atomic and force unbinding the driver.
Cheers,
-Ian
Excerpts from Ian Munsie's message of 2014-12-08 19:18:01 +1100:
> From: Ian Munsie
>
> If we need to force detach a context (e.g.
Excerpts from Ian Munsie's message of 2014-12-08 19:18:01 +1100:
> From: Ian Munsie
>
> If we need to force detach a context (e.g. due to EEH or simply force
> unbinding the driver) we should prevent the userspace contexts from
> being able to access the Problem State Area
From: Ian Munsie
An issue was introduced with "cxl: Unmap MMIO regions when detaching a
context" (b123429e6a9e8d03aacf888d23262835f0081448) where closing a
context normally could also unmap the problem state area of other
contexts currently using the AFU.
It was also discovered th
Ryan Grimm
Signed-off-by: Ian Munsie
---
drivers/misc/cxl/pci.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
index 2ccd0a9..f801c28 100644
--- a/drivers/misc/cxl/pci.c
+++ b/drivers/misc/cxl/pci.c
@@ -319,6 +319,7 @@ stati
Acked-By: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
From: Ian Munsie
This patch adds tracepoints throughout the cxl driver, which can provide
insight into:
- Context lifetimes
- Commands sent to the PSL and AFU and their completion status
- Segment and page table misses and their resolution
- PSL and AFU interrupts
- slbia calls from the powerpc
From: Ian Munsie
If a doorbell IPI comes in while a thread is in nap power saving, the
doorbell interrupt won't be replayed by the hardware since it is edge
sensitive. Currently we are not replaying these interrupts in software,
which can cause threads to miss IPIs that come in during
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
From: Ian Munsie
It was discovered that if a process mmaped their problem state area they
were able to access one page more than expected, potentially allowing
them to access the problem state area of an unrelated process.
This was due to a simple off by one error in the mmap fault handler
From: Ian Munsie
This patch makes the mmap call fail outright if the requested region is
larger than the problem state area assigned to the context so the error
is reported immediately rather than waiting for an attempt to access an
address out of bounds.
Although we never expect users to map
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
We are probably also missing an idr_destroy(&afu->contexts_idr); in
cxl_release_afu() as well if you wanted to send a patch for that.
Cheers,
-Ian
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.or
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Excerpts from Sukadev Bhattiprolu's message of 2015-07-17 11:51:04 +1000:
> Are you seeing this on big-endian or little-endian system?
>
> IIRC, I saw the opposite behavior on an LE system a few months ago.
> i.e. without 1028ccf5, 'perf listf|grep syscall' failed.
>
> Applying 1028ccf5, seemed t
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
From: Ian Munsie
If the system has a PCI device with a memory-controller device node,
kexec-lite would spew hundreds of double free warnings and eventually
segfault. This would result in a "kexec load failed" message from
petitboot.
This was due to kexec_memory_map() searching f
Whoops - subject line was supposed to say "kexec-lite: ..."
-Ian
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Excerpts from Anton Blanchard's message of 2015-07-22 09:18:44 +1000:
> Hi,
>
> > > Nice catch! I wonder if we should be checking for device_type
> > > "memory". Ben?
> >
> > Yes. That's what Linux does.
>
> Ian: I made that change, and slightly modified your commit message.
> Look ok?
Looks go
From: Ian Munsie
userspace programs using cxl currently have to use two strategies for
dealing with MMIO errors simultaneously. They have to check every read
for a return of all Fs in case the adapter has gone away and the kernel
has not yet noticed, and they have to deal with SIGBUS in case the
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
Excerpts from Daniel Axtens's message of 2015-08-13 14:11:20 +1000:
> +/* Only warn if we detached while the link was OK.
Only because mpe is sure to pick this up (I personally don't mind) -
block comments should start with /* on a line by itself.
> +
hough in the long run I wonder if we
should consider making the contexts increase the refcount of the AFU so
that we can be sure that the AFU structure will outlive the contexts?
That would be a more significant rework though, and this patch is needed
either
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
s pass, but if you happen to do another
respin please move the style changes into a separate patch.
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
e in with this
patch makes the diff harder to follow than necessary (though not as hard
as the last one). If you happen to do a v5, please put the coding style
changes in a separate patch, but otherwise it looks fine:
Acked-by: Ian Munsie
___
Linuxppc-dev m
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Once cxlflash has been merged we might drop this, but until then:
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Thanks for splitting this out - much easier to follow now :)
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Thanks :)
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
Acked-by: Ian Munsie
___
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
From: Ian Munsie
This adds an afu_driver_ops structure with event_pending and
deliver_event callbacks. An AFU driver can fill these out and associate
it with a context to enable passing custom AFU specific events to
userspace.
The cxl driver will call event_pending() during poll, select, read
.
Signed-off-by: Michael Neuling
Signed-off-by: Ian Munsie
---
drivers/misc/cxl/api.c | 21 +
drivers/misc/cxl/cxl.h | 3 +++
include/misc/cxl.h | 7 +++
3 files changed, 31 insertions(+)
diff --git a/drivers/misc/cxl/api.c b/drivers/misc/cxl/api.c
index e0f0c78..5f0b22e
Excerpts from Michael Ellerman's message of 2015-08-19 14:49:30 +1000:
> Do we really need the accessors? They don't buy anything I can see over just
> using ctx->priv directly.
The reasoning there is because we don't currently expose the contents of
stuct cxl_context to afu drivers, rather they j
Excerpts from Michael Ellerman's message of 2015-08-19 15:30:46 +1000:
> On Wed, 2015-08-19 at 14:19 +1000, Ian Munsie wrote:
> > From: Ian Munsie
> >
> > This adds an afu_driver_ops structure with event_pending and
> > deliver_event callbacks. An AFU driver
From: Ian Munsie
This patch adds support for enabling and context switching the Target
Address Register in Power8. The TAR is a new special purpose register
that can be used for computed branches with the bctar[l] (branch
conditional to TAR) instruction in the same manner as the count and link
From: Ian Munsie
This patch adds support for enabling and context switching the Target
Address Register in Power8. The TAR is a new special purpose register
that can be used for computed branches with the bctar[l] (branch
conditional to TAR) instruction in the same manner as the count and link
From: Ian Munsie
This patch adds a new line to /proc/interrupts to account for the
doorbell interrupts that each hardware thread has received. The total
interrupt count in /proc/stat will now also include doorbells.
# cat /proc/interrupts
CPU0 CPU1 CPU2 CPU3
16
1 - 100 of 457 matches
Mail list logo
