On Fri, 28 Oct 2022, Gwan-gyeong Mun <gwan-gyeong.mun@xxxxxxxxx> wrote:
> Resend, because some content was accidentally omitted from the previous
> reply.
> Please ignore the previous email.
>
> Hi all,
>
> I should have written the original commit message more accurately, but
> it seems that it was written inaccurately.
>
> If the FIELD_PREP macro is expanded, the following macros are used.
>
> #define FIELD_PREP(_mask, _val) \
> ({ \
> __BF_FIELD_CHECK(_mask, 0ULL, _val, "FIELD_PREP: "); \
> ((typeof(_mask))(_val) << __bf_shf(_mask)) & (_mask); \
> })
>
>
> #define __BF_FIELD_CHECK(_mask, _reg, _val, _pfx) \
> ({ \
> BUILD_BUG_ON_MSG(!__builtin_constant_p(_mask), \
> _pfx "mask is not constant"); \
> BUILD_BUG_ON_MSG((_mask) == 0, _pfx "mask is zero"); \
> BUILD_BUG_ON_MSG(__builtin_constant_p(_val) ? \
> ~((_mask) >> __bf_shf(_mask)) & (_val) : 0, \
> _pfx "value too large for the field"); \
> BUILD_BUG_ON_MSG(__bf_cast_unsigned(_mask, _mask) > \
> __bf_cast_unsigned(_reg, ~0ull), \
> _pfx "type of reg too small for mask"); \
> __BUILD_BUG_ON_NOT_POWER_OF_2((_mask) + \
> (1ULL << __bf_shf(_mask))); \
> })
>
> Among them, a build error is generated by the lower part of the
> __BF_FIELD_CHECK() macro.
>
> BUILD_BUG_ON_MSG(__bf_cast_unsigned(_mask, _mask) > \
> __bf_cast_unsigned(_reg, ~0ull), \
> _pfx "type of reg too small for mask"); \
>
>
> Here, if you apply an argument to this macro, it will look like the
> following.
>
> __bf_cast_unsigned(field_msk, field_msk) > __bf_cast_unsigned(0ULL, ~0ull)
>
> The result is always false because an unsigned int value of type
> field_msk is not always greater than the maximum value of unsigned long
> long .
> So, a build error occurs due to the following part of the clang compiler
> option.
>
> [-Werror,-Wtautological-constant-out-of-range-compare]
>
> You can simply override this warning in Clang by adding the build option
> below, but this seems like a bad attempt
>
> i915/Makefile
> CFLAGS_i915_hwmon.o += -Wno-tautological-constant-out-of-range-compare
>
> The easiest way to solve this is to use a constant value, not a
> variable, as an argument to FIELD_PREP.
>
> And since the REG_FIELD_PREP() macro suggested by Jani requires a const
> expression as the first argument, it cannot be changed with this macro
> alone in the existing code, it must be changed to input a constant value
> as shown below.
We've added REG_FIELD_PREP() precisely to avoid the problems with the
types and ranges, as we want it to operate on u32. It also uses
__is_constexpr() to avoid dependencies on compiler implementation and
optimizations.
Please use REG_FIELD_PREP() and a constant value. Maybe rethink the
interface if needed.
BR,
Jani.
>
> diff --git a/drivers/gpu/drm/i915/i915_hwmon.c
> b/drivers/gpu/drm/i915/i915_hwmon.c
> index 08c921421a5f..abb3a194c548 100644
> --- a/drivers/gpu/drm/i915/i915_hwmon.c
> +++ b/drivers/gpu/drm/i915/i915_hwmon.c
> @@ -101,7 +101,7 @@ hwm_field_read_and_scale(struct hwm_drvdata *ddat,
> i915_reg_t rgadr,
>
> static void
> hwm_field_scale_and_write(struct hwm_drvdata *ddat, i915_reg_t rgadr,
> - const u32 field_msk, int nshift,
> + int nshift,
> unsigned int scale_factor, long lval)
> {
> u32 nval;
> @@ -111,8 +111,8 @@ hwm_field_scale_and_write(struct hwm_drvdata *ddat,
> i915_reg_t rgadr,
> /* Computation in 64-bits to avoid overflow. Round to nearest. */
> nval = DIV_ROUND_CLOSEST_ULL((u64)lval << nshift, scale_factor);
>
> - bits_to_clear = field_msk;
> - bits_to_set = REG_FIELD_PREP(field_msk, nval);
> + bits_to_clear = PKG_PWR_LIM_1;
> + bits_to_set = REG_FIELD_PREP(PKG_PWR_LIM_1, nval);
>
> hwm_locked_with_pm_intel_uncore_rmw(ddat, rgadr,
> bits_to_clear, bits_to_set);
> @@ -406,7 +406,6 @@ hwm_power_write(struct hwm_drvdata *ddat, u32 attr,
> int chan, long val)
> case hwmon_power_max:
> hwm_field_scale_and_write(ddat,
> hwmon->rg.pkg_rapl_limit,
> - PKG_PWR_LIM_1,
> hwmon->scl_shift_power,
> SF_POWER, val);
> return 0;
>
>
>
> In addition, if there is no build problem regardless of the size of the
> type as the first argument in FIELD_PREP, it is possible through the
> following modification.
> (Since this modification modifies include/linux/bitfield.h , I will send
> it as a separate patch.
> )
>
> However, it seems that we need to have Jani's confirm whether it is okay
> to use FIELD_PREP() instead of REG_FIELD_PREP() which is forced to u32
> return type in i915.
>
> diff --git a/include/linux/bitfield.h b/include/linux/bitfield.h
> index c9be1657f03d..6e96799b6f38 100644
> --- a/include/linux/bitfield.h
> +++ b/include/linux/bitfield.h
> @@ -9,7 +9,7 @@
>
> #include <linux/build_bug.h>
> #include <asm/byteorder.h>
> -
> +#include <linux/overflow.h>
> /*
> * Bitfield access macros
> *
> @@ -69,7 +69,7 @@
> ~((_mask) >> __bf_shf(_mask)) & (_val)
> : 0, \
> _pfx "value too large for the field"); \
> BUILD_BUG_ON_MSG(__bf_cast_unsigned(_mask, _mask) > \
> - __bf_cast_unsigned(_reg, ~0ull), \
> + __bf_cast_unsigned(_reg,
> type_max(__unsigned_scalar_typeof(_reg))), \
> _pfx "type of reg too small for mask"); \
> __BUILD_BUG_ON_NOT_POWER_OF_2((_mask) + \
> (1ULL << __bf_shf(_mask))); \
> @@ -84,7 +84,7 @@
> */
> #define FIELD_MAX(_mask) \
> ({ \
> - __BF_FIELD_CHECK(_mask, 0ULL, 0ULL, "FIELD_MAX: "); \
> + __BF_FIELD_CHECK(_mask,
> type_min(__unsigned_scalar_typeof(_mask)),
> type_min(__unsigned_scalar_typeof(_mask)), "FIELD_MAX: "); \
> (typeof(_mask))((_mask) >> __bf_shf(_mask)); \
> })
>
> @@ -97,7 +97,7 @@
> */
> #define FIELD_FIT(_mask, _val) \
> ({ \
> - __BF_FIELD_CHECK(_mask, 0ULL, 0ULL, "FIELD_FIT: "); \
> + __BF_FIELD_CHECK(_mask,
> type_min(__unsigned_scalar_typeof(_mask)),
> type_min(__unsigned_scalar_typeof(_val)), "FIELD_FIT: "); \
> !((((typeof(_mask))_val) << __bf_shf(_mask)) & ~(_mask)); \
> })
>
> @@ -111,7 +111,7 @@
> */
> #define FIELD_PREP(_mask, _val)
> \
> ({ \
> - __BF_FIELD_CHECK(_mask, 0ULL, _val, "FIELD_PREP: "); \
> + __BF_FIELD_CHECK(_mask,
> type_min(__unsigned_scalar_typeof(_mask)), _val, "FIELD_PREP: "); \
> ((typeof(_mask))(_val) << __bf_shf(_mask)) & (_mask); \
> })
>
> @@ -125,7 +125,7 @@
> */
> #define FIELD_GET(_mask, _reg) \
> ({ \
> - __BF_FIELD_CHECK(_mask, _reg, 0U, "FIELD_GET: "); \
> + __BF_FIELD_CHECK(_mask, _reg,
> type_min(__unsigned_scalar_typeof(_reg)), "FIELD_GET: "); \
> (typeof(_mask))(((_reg) & (_mask)) >> __bf_shf(_mask)); \
> })
>
>
> Br,
>
> G.G.
>
> On 10/27/22 9:32 PM, Dixit, Ashutosh wrote:
>> On Thu, 27 Oct 2022 10:16:47 -0700, Nick Desaulniers wrote:
>>>
>>
>> Hi Nick,
>>
>>> Thanks, I can repro now.
>>>
>>> I haven't detangled the macro soup, but I noticed:
>>>
>>> 1. FIELD_PREP is defined in include/linux/bitfield.h which has the
>>> following comment:
>>> 18 * Mask must be a compilation time constant.
>>
>> I had comments about this here:
>>
>> https://lore.kernel.org/intel-gfx/87ilk7pwrw.wl-ashutosh.dixit@xxxxxxxxx/
>>
>> The relevant part being:
>>
>> ---- {quote} ----
>>>>> ./include/linux/bitfield.h:71:53: note: expanded from macro '__BF_FIELD_CHECK'
>>>>> BUILD_BUG_ON_MSG(__bf_cast_unsigned(_mask, _mask) > \
>>
>> So clang seems to break here at this line in __BF_FIELD_CHECK (note ~0ull
>> also occurs here):
>>
>> BUILD_BUG_ON_MSG(__bf_cast_unsigned(_mask, _mask) > \
>> __bf_cast_unsigned(_reg, ~0ull), \
>> _pfx "type of reg too small for mask"); \
>>
>> So it goes through previous checks including the "mask is not constant"
>> check. As Nick Desaulniers mentions "__builtin_constant_p is evaluated
>> after most optimizations have run" so by that time both compilers (gcc and
>> clang) have figured out that even though _mask is coming in as function
>> argument it is really the constant below:
>>
>> #define PKG_PWR_LIM_1 REG_GENMASK(14, 0)
>>
>> But it is not clear why clang chokes on this "type of reg too small for
>> mask" check (and gcc doesn't) since everything is u32.
>> ---- {end quote} ----
>>
>>>
>>> 2. hwm_field_scale_and_write only has one callsite.
>>>
>>> The following patch works:
>>
>> If we need to fix it at our end yes we can come up with one of these
>> patches. But we were hoping someone from clang/llvm can comment about the
>> "type of reg too small for mask" stuff. If this is something which needs to
>> be fixed in clang/llvm we probably don't want to hide the issue.
>>
>>>
>>> ```
>>> diff --git a/drivers/gpu/drm/i915/i915_hwmon.c
>>> b/drivers/gpu/drm/i915/i915_hwmon.c
>>> index 9e9781493025..6ac29d90b92a 100644
>>> --- a/drivers/gpu/drm/i915/i915_hwmon.c
>>> +++ b/drivers/gpu/drm/i915/i915_hwmon.c
>>> @@ -101,7 +101,7 @@ hwm_field_read_and_scale(struct hwm_drvdata *ddat,
>>> i915_reg_t rgadr,
>>>
>>> static void
>>> hwm_field_scale_and_write(struct hwm_drvdata *ddat, i915_reg_t rgadr,
>>> - u32 field_msk, int nshift,
>>> + int nshift,
>>> unsigned int scale_factor, long lval)
>>> {
>>> u32 nval;
>>> @@ -111,8 +111,8 @@ hwm_field_scale_and_write(struct hwm_drvdata
>>> *ddat, i915_reg_t rgadr,
>>> /* Computation in 64-bits to avoid overflow. Round to nearest. */
>>> nval = DIV_ROUND_CLOSEST_ULL((u64)lval << nshift, scale_factor);
>>>
>>> - bits_to_clear = field_msk;
>>> - bits_to_set = FIELD_PREP(field_msk, nval);
>>> + bits_to_clear = PKG_PWR_LIM_1;
>>> + bits_to_set = FIELD_PREP(PKG_PWR_LIM_1, nval);
>>>
>>> hwm_locked_with_pm_intel_uncore_rmw(ddat, rgadr,
>>> bits_to_clear, bits_to_set);
>>> @@ -406,7 +406,6 @@ hwm_power_write(struct hwm_drvdata *ddat, u32
>>> attr, int chan, long val)
>>> case hwmon_power_max:
>>> hwm_field_scale_and_write(ddat,
>>> hwmon->rg.pkg_rapl_limit,
>>> - PKG_PWR_LIM_1,
>>> hwmon->scl_shift_power,
>>> SF_POWER, val);
>>> return 0;
>>> ```
>>> Though I'm not sure if you're planning to add further callsites of
>>> hwm_field_scale_and_write with different field_masks?
>>
>> I have reasons for keeping it this way, it's there in the link above if you
>> are interested.
>>
>>>
>>> Alternatively, (without the above diff),
>>>
>>> ```
>>> diff --git a/include/linux/bitfield.h b/include/linux/bitfield.h
>>> index c9be1657f03d..6f40f12bcf89 100644
>>> --- a/include/linux/bitfield.h
>>> +++ b/include/linux/bitfield.h
>>> @@ -8,6 +8,7 @@
>>> #define _LINUX_BITFIELD_H
>>>
>>> #include <linux/build_bug.h>
>>> +#include <linux/const.h>
>>> #include <asm/byteorder.h>
>>>
>>> /*
>>> @@ -62,7 +63,7 @@
>>>
>>> #define __BF_FIELD_CHECK(_mask, _reg, _val, _pfx) \
>>> ({ \
>>> - BUILD_BUG_ON_MSG(!__builtin_constant_p(_mask), \
>>> + BUILD_BUG_ON_MSG(!__is_constexpr(_mask), \
>>> _pfx "mask is not constant"); \
>>> BUILD_BUG_ON_MSG((_mask) == 0, _pfx "mask is zero"); \
>>> BUILD_BUG_ON_MSG(__builtin_constant_p(_val) ? \
>>> ```
>>> will produce:
>>> error: call to __compiletime_assert_407 declared with 'error'
>>> attribute: FIELD_PREP: mask is not constant
>>>
>>> I haven't tested if that change is also feasible (on top of fixing
>>> this specific instance), but I think it might help avoid more of these
>>> subtleties wrt. __builtin_constant_p that depende heavily on compiler,
>>> compiler version, optimization level.
>>
>> Not disagreeing, can do something here if needed.
>>
>> Thanks.
>> --
>> Ashutosh
--
Jani Nikula, Intel Open Source Graphics Center
