On Tue, 19 May 2015, Daniel Granat wrote:
> Signed-off-by: Daniel Granat <[email protected]> > --- > scripts/coccinelle/misc/of_table.cocci | 26 +++++++++++++++++++++----- > 1 file changed, 21 insertions(+), 5 deletions(-) > > diff --git a/scripts/coccinelle/misc/of_table.cocci > b/scripts/coccinelle/misc/of_table.cocci > index 3c93404..74f7dbb 100644 > --- a/scripts/coccinelle/misc/of_table.cocci > +++ b/scripts/coccinelle/misc/of_table.cocci > @@ -1,6 +1,6 @@ > -/// Make sure of_device_id tables are NULL terminated > +/// Make sure (of_device_id | i2c_device_id | platform_device_id) tables are > NULL terminated > // > -// Keywords: of_table > +// Keywords: of_table i2c_table platform_table > // Confidence: Medium > // Options: --include-headers > > @@ -13,7 +13,7 @@ virtual report > identifier var, arr; > expression E; > @@ > -struct of_device_id arr[] = { > +struct \(of_device_id \| i2c_device_id \| platform_device_id\) arr[] = { > ..., > { > .var = E, This case (the context case) needs a ..., E, ... case like the patch rule has. > @@ -24,7 +24,8 @@ struct of_device_id arr[] = { > identifier var, arr; > expression E; > @@ > -struct of_device_id arr[] = { > +( > +struct \(of_device_id \| i2c_device_id \| platform_device_id\) arr[] = { > ..., > { > .var = E, > @@ -32,19 +33,34 @@ struct of_device_id arr[] = { > + }, > + { } > }; > +| > +struct \(of_device_id \| i2c_device_id \| platform_device_id\) arr[] = { > + ..., > + { ..., E, ... }, > ++ { }, > +}; > +) > > @r depends on org || report@ > position p1; > identifier var, arr; > expression E; > @@ > -struct of_device_id arr[] = { > +( > +struct \(of_device_id \| i2c_device_id \| platform_device_id\) arr[] = { > ..., > { > .var = E, > } > @p1 > }; > +| > +struct \(of_device_id \| i2c_device_id \| platform_device_id\) arr[] = { > + ..., > + { ..., E, ... }, > + @p1 It is not a good idea to put the @p1 on the comma. The comma might not be there, and if it is not there Coccinelle seems to crash. A more graceful behavior would be for the rule simply not to match, since a comma that isn't there doesn't have a position. In any case, if you put the @p1 on the }, things are better. I made the following test case, if you want to make a quick check of the rule. All four options should give four results. static const struct platform_device_id ab85xx_rtc_ids[] = { { "ab8500-rtc", (kernel_ulong_t)&ab8500_rtc_ops, }, { "ab8540-rtc", (kernel_ulong_t)&ab8540_rtc_ops, }, }; static const struct platform_device_id ab85xx_rtc_ids[] = { { "ab8500-rtc", (kernel_ulong_t)&ab8500_rtc_ops, }, { "ab8540-rtc", (kernel_ulong_t)&ab8540_rtc_ops, } }; static const struct platform_device_id ab85xx_rtc_ids[] = { { .a = "ab8500-rtc", .b = (kernel_ulong_t)&ab8500_rtc_ops, }, { .a = "ab8540-rtc", .b = (kernel_ulong_t)&ab8540_rtc_ops, }, }; static const struct platform_device_id ab85xx_rtc_ids[] = { { .a = "ab8500-rtc", .b = (kernel_ulong_t)&ab8500_rtc_ops, }, { .a = "ab8540-rtc", .b = (kernel_ulong_t)&ab8540_rtc_ops, } }; julia > +}; > +) > > @script:python depends on org@ > p1 << r.p1; > -- > 1.9.1 > > -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
