Re: [PATCH v2 02/18] gpio: uapi: define uAPI v2

On Wed, Aug 5, 2020 at 7:19 AM Kent Gibson [off-list ref] wrote:

[snip]

quoted

quoted

+/*
+ * Maximum number of requested lines.
+ *
+ * Must be a multiple of 8 to ensure 32/64-bit alignment of structs.
+ */
+#define GPIOLINES_MAX 64
+
+/* The number of __u64 required for a bitmap for GPIOLINES_MAX lines */
+#define GPIOLINES_BITMAP_SIZE  __KERNEL_DIV_ROUND_UP(GPIOLINES_MAX, 64)
+

In what circumstances can this be different than 1? It's worth
documenting here I suppose.

In terms of the API definition, GPIOLINES_MAX can be anything you want
and the definitions are still valid.  In practice in the mainline kernel
it would always be 1 for ABI compatibility.

Chiselling GPIOLINES_MAX <= 64 into stone could simplify things a bit,
as all the bitmaps reduce to a single __u64.  Would you prefer that?

I'm not sure I follow. We need to chisel some max value in stone. Up
to that point it's been 64. We can make it more and the bitmap API
would handle it alright but if we don't, then this
__KERNEL_DIV_ROUND_UP() is unnecessary. Limiting it to 64 makes things
very simple thanks to fitting into a __u64 though. I've personally
never needed to request even half that so I guess this value's fine?

quoted

quoted

+/*
+ * The maximum number of configuration attributes associated with a line
+ * request.
+ */
+#define GPIOLINE_NUM_ATTRS_MAX 10
+

How did you choose this number? I mean: it's reasonable - just asking
for clarification.

I didn't want to constrain the possible configurations by making it too
small, particularly allowing for future attributes, but wanted to keep the
request size down so it can still comfortably fit on the stack.
The gpioline_request stands at 592 bytes, which is already substantially
larger than the 364 bytes of the v1 request, and each additional config
attribute slot adds another 24 bytes.

So 10 seemed like a happy medium.

Makes sense.

quoted

quoted

+/**
+ * enum gpioline_flag_v2 - &struct gpioline_attribute.flags values
+ */
+enum gpioline_flag_v2 {
+       GPIOLINE_FLAG_V2_USED                   = 1UL << 0, /* line is not available for request */
+       GPIOLINE_FLAG_V2_ACTIVE_LOW             = 1UL << 1,
+       GPIOLINE_FLAG_V2_INPUT                  = 1UL << 2,
+       GPIOLINE_FLAG_V2_OUTPUT                 = 1UL << 3,
+       GPIOLINE_FLAG_V2_EDGE_RISING            = 1UL << 4,
+       GPIOLINE_FLAG_V2_EDGE_FALLING           = 1UL << 5,
+       GPIOLINE_FLAG_V2_OPEN_DRAIN             = 1UL << 6,
+       GPIOLINE_FLAG_V2_OPEN_SOURCE            = 1UL << 7,
+       GPIOLINE_FLAG_V2_BIAS_PULL_UP           = 1UL << 8,
+       GPIOLINE_FLAG_V2_BIAS_PULL_DOWN         = 1UL << 9,
+       GPIOLINE_FLAG_V2_BIAS_DISABLED          = 1UL << 10,
+};
+
+/**
+ * struct gpioline_values - Values of GPIO lines
+ * @bits: a bitmap containing the value of the lines, set to 1 for active
+ * and 0 for inactive.  Note that this is the logical value, which will be
+ * the opposite of the physical value if the line is configured as active
+ * low.
+ */
+struct gpioline_values {
+       __u64 bits[GPIOLINES_BITMAP_SIZE];
+};
+

We can set values only for a subset of requested lines but AFAICT we
can't read values of only a subset of lines. Would it be difficult to
remove this limitation? While reading values always succeeds - even if
the line is in input mode and has edge detected - I think that someone
may want to request the max number of lines without reading all their
values each time. Maybe consider merging this with struct
gpioline_set_values?

That is correct.

I considered that corner case to be unlikely, as a major point of
requesting lines together is to be able to perform collective operations
on them as atomically as possible.  If you only want subsets then
request them as separate subsets.

And yet this version implements heterogeneous config and setting edge
detection and values of subsets of requested lines. :)

Do you have a case in mind where you would have overlapping subsets?

No, not really but then I also don't have a use-case for setting only
a certain subset of lines.

Not difficult to remove the limitation - I just didn't see sufficient
benefit.

Using the same structure for setting and getting values is a benefit
IMO. If it's not a difficult task, then I think it's worth adding it.

quoted

quoted

+/**
+ * struct gpioline_set_values - Values to set a group of GPIO lines
+ * @mask: a bitmap identifying the lines to set.
+ * @bits: a bitmap containing the value of the lines, set to 1 for active
+ * and 0 for inactive.  Note that this is the logical value, which will be
+ * the opposite of the physical value if the line is configured as active
+ * low.
+ */
+struct gpioline_set_values {
+       __u64 mask[GPIOLINES_BITMAP_SIZE];
+       __u64 bits[GPIOLINES_BITMAP_SIZE];
+};
+
+/**
+ * enum gpioline_attr_id - &struct gpioline_attribute.id values
+ */
+enum gpioline_attr_id {
+       GPIOLINE_ATTR_ID_FLAGS                  = 1,
+       GPIOLINE_ATTR_ID_OUTPUT_VALUES          = 2,
+       GPIOLINE_ATTR_ID_DEBOUNCE               = 3,
+};
+
+/**
+ * struct gpioline_attribute - a configurable attribute of a line
+ * @id: attribute identifier with value from &enum gpioline_attr_id
+ * @padding: reserved for future use and must be zero filled
+ * @flags: if id is GPIOLINE_ATTR_ID_FLAGS, the flags for the GPIO line,
+ * with values from enum gpioline_flag_v2, such as
+ * GPIOLINE_FLAG_V2_ACTIVE_LOW, GPIOLINE_FLAG_V2_OUTPUT etc, OR:ed
+ * together.  This overrides the default flags contained in the &struct
+ * gpioline_config for the associated line.
+ * @values: if id is GPIOLINE_ATTR_ID_OUTPUT_VALUES, the values to which
+ * the lines will be set
+ * @debounce_period: if id is GPIOLINE_ATTR_ID_DEBOUNCE, the desired
+ * debounce period, in microseconds
+ */
+struct gpioline_attribute {
+       __u32 id;
+       __u32 padding;
+       union {
+               __u64 flags;
+               struct gpioline_values values;
+               __u32 debounce_period;
+       };
+};

I'm afraid that if we don't have enough padding here (at the end),
we'll end up wanting to add a new attribute at some point whose
argument won't fit. Maybe have a specific field in the union that's
even larger than __u64?

I'm satisfied with the 64-bit value restriction.

I don't want to go adding another 8 bytes of pad per attribute on the
off chance that we ever find such an attribute, and that we couldn't
find some other solution like using the __u32 padding, or user the
gpioline_config padding, or split it over two attributes....

Fair enough.

quoted

quoted

+
+/**
+ * struct gpioline_config_attribute - a configuration attribute associated
+ * with one or more of the requested lines.
+ * @mask: a bitmap identifying the lines to which the attribute applies
+ * @attr: the configurable attribute
+ */
+struct gpioline_config_attribute {
+       __u64 mask[GPIOLINES_BITMAP_SIZE];
+       struct gpioline_attribute attr;
+};
+
+/**
+ * struct gpioline_config - Configuration for GPIO lines
+ * @flags: flags for the GPIO lines, with values from enum
+ * gpioline_flag_v2, such as GPIOLINE_FLAG_V2_ACTIVE_LOW,
+ * GPIOLINE_FLAG_V2_OUTPUT etc, OR:ed together.  This is the default for
+ * all requested lines but may be overridden for particular lines using
+ * attrs.

So I'm having a hard time with this. I understand that the thinking
behind it was: use the flags field to set all lines to INPUT by
default and only set certain lines to OUTPUT with attrs. This would
make life easier for user-space but it complicates the kernel code and
I also believe that any such simplification should be handled by
user-space libraries, not be exposed by kernel uAPI. My personal
preference would be to drop the flags field and only handle attributes
(maybe even define a special macro to set all bits in mask -
GPIOLINE_CONFIG_ALL_LINES or something) on a first-in-wins basis. I'm
open to other suggestions though.

I think I've addressed this elsewhere, and still think it is worthwhile
and very low cost.  I thought it was an easy win when I added it, and
still do.

Happy to change the attrs to first-in-wins though - the validation of
the attrs is still my biggest bugbear with this version.

Yes, I read your other reply. Ok, makes sense to have default flags
with an attribute for overrides. This just needs very explicit
documentation.

Bartosz