While studying riscv's cmpxchg.h file, I got really interested in
understanding how RISCV asm implemented the different versions of
{cmp,}xchg.
When I understood the pattern, it made sense for me to remove the
duplications and create macros to make it easier to understand what exactly
changes between the versions: Instruction sufixes & barriers.
I split those changes in 3 levels for each cmpxchg and xchg, resulting a
total of 6 patches. I did this so it becomes easier to review and remove
the last levels if desired, but I have no issue squashing them if it's
better.
Please provide comments.
Thanks!
Leo
Leonardo Bras (6):
riscv/cmpxchg: Deduplicate cmpxchg() asm functions
riscv/cmpxchg: Deduplicate cmpxchg() macros
riscv/cmpxchg: Deduplicate arch_cmpxchg() macros
riscv/cmpxchg: Deduplicate xchg() asm functions
riscv/cmpxchg: Deduplicate xchg() macros
riscv/cmpxchg: Deduplicate arch_xchg() macros
arch/riscv/include/asm/cmpxchg.h | 316 +++++++------------------------
1 file changed, 64 insertions(+), 252 deletions(-)
--
2.40.0
Every arch_cmpxchg define (_relaxed, _acquire, _release, vanilla) contain
it's own define for creating tmp variables and calling the correct internal
macro for the desired version.
Those defines are mostly the same code, so there is no need to keep the 4
copies.
Create a helper define to avoid code duplication.
(This did not cause any change in generated asm)
Signed-off-by: Leonardo Bras <[email protected]>
---
arch/riscv/include/asm/cmpxchg.h | 31 ++++++++++---------------------
1 file changed, 10 insertions(+), 21 deletions(-)
diff --git a/arch/riscv/include/asm/cmpxchg.h b/arch/riscv/include/asm/cmpxchg.h
index c7a13eec4dbcc..e49a2edc6f36c 100644
--- a/arch/riscv/include/asm/cmpxchg.h
+++ b/arch/riscv/include/asm/cmpxchg.h
@@ -202,46 +202,35 @@
#define __cmpxchg_relaxed(ptr, old, new, size) \
___cmpxchg(ptr, old, new, size, "", "", "")
-#define arch_cmpxchg_relaxed(ptr, o, n) \
+#define _arch_cmpxchg(order, ptr, o, n) \
({ \
__typeof__(*(ptr)) _o_ = (o); \
__typeof__(*(ptr)) _n_ = (n); \
- (__typeof__(*(ptr))) __cmpxchg_relaxed((ptr), \
- _o_, _n_, sizeof(*(ptr))); \
+ (__typeof__(*(ptr))) __cmpxchg ## order((ptr), \
+ _o_, _n_, \
+ sizeof(*(ptr)));\
})
+#define arch_cmpxchg_relaxed(ptr, o, n) \
+ _arch_cmpxchg(_relaxed, ptr, o, n)
+
#define __cmpxchg_acquire(ptr, old, new, size) \
___cmpxchg(ptr, old, new, size, "", "", RISCV_ACQUIRE_BARRIER)
#define arch_cmpxchg_acquire(ptr, o, n) \
-({ \
- __typeof__(*(ptr)) _o_ = (o); \
- __typeof__(*(ptr)) _n_ = (n); \
- (__typeof__(*(ptr))) __cmpxchg_acquire((ptr), \
- _o_, _n_, sizeof(*(ptr))); \
-})
+ _arch_cmpxchg(_acquire, ptr, o, n)
#define __cmpxchg_release(ptr, old, new, size) \
___cmpxchg(ptr, old, new, size, "", RISCV_RELEASE_BARRIER, "")
#define arch_cmpxchg_release(ptr, o, n) \
-({ \
- __typeof__(*(ptr)) _o_ = (o); \
- __typeof__(*(ptr)) _n_ = (n); \
- (__typeof__(*(ptr))) __cmpxchg_release((ptr), \
- _o_, _n_, sizeof(*(ptr))); \
-})
+ _arch_cmpxchg(_release, ptr, o, n)
#define __cmpxchg(ptr, old, new, size) \
___cmpxchg(ptr, old, new, size, ".rl", "", " fence rw, rw\n")
#define arch_cmpxchg(ptr, o, n) \
-({ \
- __typeof__(*(ptr)) _o_ = (o); \
- __typeof__(*(ptr)) _n_ = (n); \
- (__typeof__(*(ptr))) __cmpxchg((ptr), \
- _o_, _n_, sizeof(*(ptr))); \
-})
+ _arch_cmpxchg(, ptr, o, n)
#define arch_cmpxchg_local(ptr, o, n) \
(__cmpxchg_relaxed((ptr), (o), (n), sizeof(*(ptr))))
--
2.40.0
Every xchg define (_relaxed, _acquire, _release, vanilla) contain it's own
define for creating tmp variables and selecting the correct asm code for
given variable size.
All those defines are mostly the same code (other than specific barriers),
so there is no need to keep the 4 copies.
Unify those under a more general define, that can reproduce the previous 4
versions.
(This did not cause any change in generated asm)
Signed-off-by: Leonardo Bras <[email protected]>
---
arch/riscv/include/asm/cmpxchg.h | 62 ++++++--------------------------
1 file changed, 10 insertions(+), 52 deletions(-)
diff --git a/arch/riscv/include/asm/cmpxchg.h b/arch/riscv/include/asm/cmpxchg.h
index 13dc608229ef0..23da4d8e6f0c8 100644
--- a/arch/riscv/include/asm/cmpxchg.h
+++ b/arch/riscv/include/asm/cmpxchg.h
@@ -11,7 +11,7 @@
#include <asm/barrier.h>
#include <asm/fence.h>
-#define ___xchg(sfx, prepend, append) \
+#define ____xchg(sfx, prepend, append) \
({ \
__asm__ __volatile__ ( \
prepend \
@@ -22,17 +22,17 @@
: "memory"); \
})
-#define __xchg_relaxed(ptr, new, size) \
+#define ___xchg(ptr, new, size, sfx, prepend, append) \
({ \
__typeof__(ptr) __ptr = (ptr); \
__typeof__(new) __new = (new); \
__typeof__(*(ptr)) __ret; \
switch (size) { \
case 4: \
- ___xchg(".w", "", ""); \
+ ____xchg(".w" sfx, prepend, append); \
break; \
case 8: \
- ___xchg(".d", "", ""); \
+ ____xchg(".d" sfx, prepend, append); \
break; \
default: \
BUILD_BUG(); \
@@ -40,6 +40,9 @@
__ret; \
})
+#define __xchg_relaxed(ptr, new, size) \
+ ___xchg(ptr, new, size, "", "", "")
+
#define arch_xchg_relaxed(ptr, x) \
({ \
__typeof__(*(ptr)) _x_ = (x); \
@@ -48,22 +51,7 @@
})
#define __xchg_acquire(ptr, new, size) \
-({ \
- __typeof__(ptr) __ptr = (ptr); \
- __typeof__(new) __new = (new); \
- __typeof__(*(ptr)) __ret; \
- switch (size) { \
- case 4: \
- ___xchg(".w", "", RISCV_ACQUIRE_BARRIER); \
- break; \
- case 8: \
- ___xchg(".d", "", RISCV_ACQUIRE_BARRIER); \
- break; \
- default: \
- BUILD_BUG(); \
- } \
- __ret; \
-})
+ ___xchg(ptr, new, size, "", "", RISCV_ACQUIRE_BARRIER)
#define arch_xchg_acquire(ptr, x) \
({ \
@@ -73,22 +61,7 @@
})
#define __xchg_release(ptr, new, size) \
-({ \
- __typeof__(ptr) __ptr = (ptr); \
- __typeof__(new) __new = (new); \
- __typeof__(*(ptr)) __ret; \
- switch (size) { \
- case 4: \
- ___xchg(".w", RISCV_RELEASE_BARRIER, ""); \
- break; \
- case 8: \
- ___xchg(".d", RISCV_RELEASE_BARRIER, ""); \
- break; \
- default: \
- BUILD_BUG(); \
- } \
- __ret; \
-})
+ ___xchg(ptr, new, size, "", RISCV_RELEASE_BARRIER, "")
#define arch_xchg_release(ptr, x) \
({ \
@@ -98,22 +71,7 @@
})
#define __xchg(ptr, new, size) \
-({ \
- __typeof__(ptr) __ptr = (ptr); \
- __typeof__(new) __new = (new); \
- __typeof__(*(ptr)) __ret; \
- switch (size) { \
- case 4: \
- ___xchg("w.aqrl", "", ""); \
- break; \
- case 8: \
- ___xchg("d.aqrl", "", ""); \
- break; \
- default: \
- BUILD_BUG(); \
- } \
- __ret; \
-})
+ ___xchg(ptr, new, size, ".aqrl", "", "")
#define arch_xchg(ptr, x) \
({ \
--
2.40.0
Every arch_xchg define (_relaxed, _acquire, _release, vanilla) contain it's
own define for creating tmp variables and calling the correct internal
macro for the desired version.
Those defines are mostly the same code, so there is no need to keep the 4
copies.
Create a helper define to avoid code duplication.
(This did not cause any change in generated asm)
Signed-off-by: Leonardo Bras <[email protected]>
---
arch/riscv/include/asm/cmpxchg.h | 26 +++++++++-----------------
1 file changed, 9 insertions(+), 17 deletions(-)
diff --git a/arch/riscv/include/asm/cmpxchg.h b/arch/riscv/include/asm/cmpxchg.h
index 23da4d8e6f0c8..d13da2286c82a 100644
--- a/arch/riscv/include/asm/cmpxchg.h
+++ b/arch/riscv/include/asm/cmpxchg.h
@@ -43,41 +43,33 @@
#define __xchg_relaxed(ptr, new, size) \
___xchg(ptr, new, size, "", "", "")
-#define arch_xchg_relaxed(ptr, x) \
+#define _arch_xchg(order, ptr, x) \
({ \
__typeof__(*(ptr)) _x_ = (x); \
- (__typeof__(*(ptr))) __xchg_relaxed((ptr), \
- _x_, sizeof(*(ptr))); \
+ (__typeof__(*(ptr))) __xchg ## order((ptr), \
+ _x_, sizeof(*(ptr))); \
})
+#define arch_xchg_relaxed(ptr, x) \
+ _arch_xchg(_relaxed, ptr, x)
+
#define __xchg_acquire(ptr, new, size) \
___xchg(ptr, new, size, "", "", RISCV_ACQUIRE_BARRIER)
#define arch_xchg_acquire(ptr, x) \
-({ \
- __typeof__(*(ptr)) _x_ = (x); \
- (__typeof__(*(ptr))) __xchg_acquire((ptr), \
- _x_, sizeof(*(ptr))); \
-})
+ _arch_xchg(_acquire, ptr, x)
#define __xchg_release(ptr, new, size) \
___xchg(ptr, new, size, "", RISCV_RELEASE_BARRIER, "")
#define arch_xchg_release(ptr, x) \
-({ \
- __typeof__(*(ptr)) _x_ = (x); \
- (__typeof__(*(ptr))) __xchg_release((ptr), \
- _x_, sizeof(*(ptr))); \
-})
+ _arch_xchg(_release, ptr, x)
#define __xchg(ptr, new, size) \
___xchg(ptr, new, size, ".aqrl", "", "")
#define arch_xchg(ptr, x) \
-({ \
- __typeof__(*(ptr)) _x_ = (x); \
- (__typeof__(*(ptr))) __xchg((ptr), _x_, sizeof(*(ptr))); \
-})
+ _arch_xchg(, ptr, x)
#define xchg32(ptr, x) \
({ \
--
2.40.0
In this header every xchg define (_relaxed, _acquire, _release, vanilla)
contain it's own asm file, both for 4-byte variables an 8-byte variables,
on a total of 8 versions of mostly the same asm.
This is usually bad, as it means any change may be done in up to 8
different places.
Unify those versions by creating a new define with enough parameters to
generate any version of the previous 8.
(This did not cause any change in generated asm)
Signed-off-by: Leonardo Bras <[email protected]>
---
arch/riscv/include/asm/cmpxchg.h | 63 ++++++++++----------------------
1 file changed, 19 insertions(+), 44 deletions(-)
diff --git a/arch/riscv/include/asm/cmpxchg.h b/arch/riscv/include/asm/cmpxchg.h
index e49a2edc6f36c..13dc608229ef0 100644
--- a/arch/riscv/include/asm/cmpxchg.h
+++ b/arch/riscv/include/asm/cmpxchg.h
@@ -11,6 +11,17 @@
#include <asm/barrier.h>
#include <asm/fence.h>
+#define ___xchg(sfx, prepend, append) \
+({ \
+ __asm__ __volatile__ ( \
+ prepend \
+ " amoswap" sfx " %0, %2, %1\n" \
+ append \
+ : "=r" (__ret), "+A" (*__ptr) \
+ : "r" (__new) \
+ : "memory"); \
+})
+
#define __xchg_relaxed(ptr, new, size) \
({ \
__typeof__(ptr) __ptr = (ptr); \
@@ -18,18 +29,10 @@
__typeof__(*(ptr)) __ret; \
switch (size) { \
case 4: \
- __asm__ __volatile__ ( \
- " amoswap.w %0, %2, %1\n" \
- : "=r" (__ret), "+A" (*__ptr) \
- : "r" (__new) \
- : "memory"); \
+ ___xchg(".w", "", ""); \
break; \
case 8: \
- __asm__ __volatile__ ( \
- " amoswap.d %0, %2, %1\n" \
- : "=r" (__ret), "+A" (*__ptr) \
- : "r" (__new) \
- : "memory"); \
+ ___xchg(".d", "", ""); \
break; \
default: \
BUILD_BUG(); \
@@ -51,20 +54,10 @@
__typeof__(*(ptr)) __ret; \
switch (size) { \
case 4: \
- __asm__ __volatile__ ( \
- " amoswap.w %0, %2, %1\n" \
- RISCV_ACQUIRE_BARRIER \
- : "=r" (__ret), "+A" (*__ptr) \
- : "r" (__new) \
- : "memory"); \
+ ___xchg(".w", "", RISCV_ACQUIRE_BARRIER); \
break; \
case 8: \
- __asm__ __volatile__ ( \
- " amoswap.d %0, %2, %1\n" \
- RISCV_ACQUIRE_BARRIER \
- : "=r" (__ret), "+A" (*__ptr) \
- : "r" (__new) \
- : "memory"); \
+ ___xchg(".d", "", RISCV_ACQUIRE_BARRIER); \
break; \
default: \
BUILD_BUG(); \
@@ -86,20 +79,10 @@
__typeof__(*(ptr)) __ret; \
switch (size) { \
case 4: \
- __asm__ __volatile__ ( \
- RISCV_RELEASE_BARRIER \
- " amoswap.w %0, %2, %1\n" \
- : "=r" (__ret), "+A" (*__ptr) \
- : "r" (__new) \
- : "memory"); \
+ ___xchg(".w", RISCV_RELEASE_BARRIER, ""); \
break; \
case 8: \
- __asm__ __volatile__ ( \
- RISCV_RELEASE_BARRIER \
- " amoswap.d %0, %2, %1\n" \
- : "=r" (__ret), "+A" (*__ptr) \
- : "r" (__new) \
- : "memory"); \
+ ___xchg(".d", RISCV_RELEASE_BARRIER, ""); \
break; \
default: \
BUILD_BUG(); \
@@ -121,18 +104,10 @@
__typeof__(*(ptr)) __ret; \
switch (size) { \
case 4: \
- __asm__ __volatile__ ( \
- " amoswap.w.aqrl %0, %2, %1\n" \
- : "=r" (__ret), "+A" (*__ptr) \
- : "r" (__new) \
- : "memory"); \
+ ___xchg("w.aqrl", "", ""); \
break; \
case 8: \
- __asm__ __volatile__ ( \
- " amoswap.d.aqrl %0, %2, %1\n" \
- : "=r" (__ret), "+A" (*__ptr) \
- : "r" (__new) \
- : "memory"); \
+ ___xchg("d.aqrl", "", ""); \
break; \
default: \
BUILD_BUG(); \
--
2.40.0
On Sat, Mar 18, 2023 at 05:00:54AM -0300, Leonardo Bras wrote:
> While studying riscv's cmpxchg.h file, I got really interested in
> understanding how RISCV asm implemented the different versions of
> {cmp,}xchg.
>
> When I understood the pattern, it made sense for me to remove the
> duplications and create macros to make it easier to understand what exactly
> changes between the versions: Instruction sufixes & barriers.
>
> I split those changes in 3 levels for each cmpxchg and xchg, resulting a
> total of 6 patches. I did this so it becomes easier to review and remove
> the last levels if desired, but I have no issue squashing them if it's
> better.
>
> Please provide comments.
>
> Thanks!
> Leo
>
> Leonardo Bras (6):
> riscv/cmpxchg: Deduplicate cmpxchg() asm functions
> riscv/cmpxchg: Deduplicate cmpxchg() macros
> riscv/cmpxchg: Deduplicate arch_cmpxchg() macros
> riscv/cmpxchg: Deduplicate xchg() asm functions
FWIW, this patch seems to break the build pretty badly:
https://patchwork.kernel.org/project/linux-riscv/patch/[email protected]/
Patches 1 & 5 also add quite a lot of sparse issues (like 1000), but I
think that may be more of an artifact of the testing process as opposed
to something caused by this patchset.
Cheers,
Conor.
> riscv/cmpxchg: Deduplicate xchg() macros
> riscv/cmpxchg: Deduplicate arch_xchg() macros
Hello Conor, thanks for the feedback!
On Sun, 2023-03-19 at 20:35 +0000, Conor Dooley wrote:
> On Sat, Mar 18, 2023 at 05:00:54AM -0300, Leonardo Bras wrote:
> > While studying riscv's cmpxchg.h file, I got really interested in
> > understanding how RISCV asm implemented the different versions of
> > {cmp,}xchg.
> >
> > When I understood the pattern, it made sense for me to remove the
> > duplications and create macros to make it easier to understand what exactly
> > changes between the versions: Instruction sufixes & barriers.
> >
> > I split those changes in 3 levels for each cmpxchg and xchg, resulting a
> > total of 6 patches. I did this so it becomes easier to review and remove
> > the last levels if desired, but I have no issue squashing them if it's
> > better.
> >
> > Please provide comments.
> >
> > Thanks!
> > Leo
> >
> > Leonardo Bras (6):
> > riscv/cmpxchg: Deduplicate cmpxchg() asm functions
> > riscv/cmpxchg: Deduplicate cmpxchg() macros
> > riscv/cmpxchg: Deduplicate arch_cmpxchg() macros
>
> > riscv/cmpxchg: Deduplicate xchg() asm functions
>
> FWIW, this patch seems to break the build pretty badly:
> https://patchwork.kernel.org/project/linux-riscv/patch/[email protected]/
Thanks for pointing out!
It was an intermediary error:
Sufix for amoswap on acquire version was "d.aqrl" instead of the
correct".d.aqrl", and that caused the fail.
I did not notice anything because the next commit made it more general, and thus
removed this line of code. I will send a v2-RFC shortly.
I see that patch 4/6 has 5 fails, but on each one of them I can see:
"I: build output in /ci/workspace/[...]", or
""I: build output in /tmp/[...]".
I could not find any reference to where this is saved, though.
Could you point where can I find the error output, just for the sake of further
debugging?
>
> Patches 1 & 5 also add quite a lot of sparse issues (like 1000), but I
> think that may be more of an artifact of the testing process as opposed
> to something caused by this patchset.
For those I can see the build output diffs. Both added error lines to
conchuod/build_rv64_gcc_allmodconfig.
I tried to mimic this with [make allmodconfig + gcc build with
CROSS_COMPILE=riscv64-linux-gnu- ] but I could not get any error in any patch.
For patch 1/6 it removes as much error lines (-) as it adds (+), and the error
messages are mostly the same, apart for an line number.
For patch 5/6 it actually adds many more lines, but tracking (some of) the
errors gave me no idea why.
>
> Cheers,
> Conor.
Thanks a lot!
Leo
On Tue, Mar 21, 2023 at 03:30:35AM -0300, Leonardo Br?s wrote:
> Hello Conor, thanks for the feedback!
>
>
> On Sun, 2023-03-19 at 20:35 +0000, Conor Dooley wrote:
> > On Sat, Mar 18, 2023 at 05:00:54AM -0300, Leonardo Bras wrote:
> > > While studying riscv's cmpxchg.h file, I got really interested in
> > > understanding how RISCV asm implemented the different versions of
> > > {cmp,}xchg.
> > >
> > > When I understood the pattern, it made sense for me to remove the
> > > duplications and create macros to make it easier to understand what exactly
> > > changes between the versions: Instruction sufixes & barriers.
> > >
> > > I split those changes in 3 levels for each cmpxchg and xchg, resulting a
> > > total of 6 patches. I did this so it becomes easier to review and remove
> > > the last levels if desired, but I have no issue squashing them if it's
> > > better.
> > >
> > > Please provide comments.
> > >
> > > Thanks!
> > > Leo
> > >
> > > Leonardo Bras (6):
> > > riscv/cmpxchg: Deduplicate cmpxchg() asm functions
> > > riscv/cmpxchg: Deduplicate cmpxchg() macros
> > > riscv/cmpxchg: Deduplicate arch_cmpxchg() macros
> >
> > > riscv/cmpxchg: Deduplicate xchg() asm functions
> >
> > FWIW, this patch seems to break the build pretty badly:
> > https://patchwork.kernel.org/project/linux-riscv/patch/[email protected]/
>
> Thanks for pointing out!
>
> It was an intermediary error:
> Sufix for amoswap on acquire version was "d.aqrl" instead of the
> correct".d.aqrl", and that caused the fail.
>
> I did not notice anything because the next commit made it more general, and thus
> removed this line of code. I will send a v2-RFC shortly.
>
> I see that patch 4/6 has 5 fails, but on each one of them I can see:
> "I: build output in /ci/workspace/[...]", or
> ""I: build output in /tmp/[...]".
I don't push the built artifacts anywhere, just the brief logs -
although the "failed to build" log isn't very helpful other than telling
you the build broke.
That seems like a bug w.r.t. where tuxmake prints its output. I'll try
to fix that.
> I could not find any reference to where this is saved, though.
> Could you point where can I find the error output, just for the sake of further
> debugging?
>
> >
> > Patches 1 & 5 also add quite a lot of sparse issues (like 1000), but I
> > think that may be more of an artifact of the testing process as opposed
> > to something caused by this patchset.
>
> For those I can see the build output diffs. Both added error lines to
> conchuod/build_rv64_gcc_allmodconfig.
> I tried to mimic this with [make allmodconfig + gcc build with
> CROSS_COMPILE=riscv64-linux-gnu- ] but I could not get any error in any patch.
If you can't replicate them, it's probably because they come from
sparse. I only really mentioned it here in case you went looking at the
other patches in the series and were wondering why things were so amiss.
> For patch 1/6 it removes as much error lines (-) as it adds (+), and the error
> messages are mostly the same, apart for an line number.
I don't see a line number difference, but rather an increase in the
number of times the same error lines have appeared in the output.
I don't find the single-line output from sparse to really be very
helpful, I should probably have a bit of a think how to present that
information better.
> For patch 5/6 it actually adds many more lines, but tracking (some of) the
> errors gave me no idea why.
Probably just sparse being unhappy with some way the macros were
changed - but some of it ("Should it be static?" bits) look very much
like the patch causing things to be rebuilt only for the "after the
patch" build, but somehow not for the "before" build.
On Tue, Mar 21, 2023 at 4:49 AM Conor Dooley <[email protected]> wrote:
>
> On Tue, Mar 21, 2023 at 03:30:35AM -0300, Leonardo Brás wrote:
> > Hello Conor, thanks for the feedback!
> >
> >
> > On Sun, 2023-03-19 at 20:35 +0000, Conor Dooley wrote:
> > > On Sat, Mar 18, 2023 at 05:00:54AM -0300, Leonardo Bras wrote:
> > > > While studying riscv's cmpxchg.h file, I got really interested in
> > > > understanding how RISCV asm implemented the different versions of
> > > > {cmp,}xchg.
> > > >
> > > > When I understood the pattern, it made sense for me to remove the
> > > > duplications and create macros to make it easier to understand what exactly
> > > > changes between the versions: Instruction sufixes & barriers.
> > > >
> > > > I split those changes in 3 levels for each cmpxchg and xchg, resulting a
> > > > total of 6 patches. I did this so it becomes easier to review and remove
> > > > the last levels if desired, but I have no issue squashing them if it's
> > > > better.
> > > >
> > > > Please provide comments.
> > > >
> > > > Thanks!
> > > > Leo
> > > >
> > > > Leonardo Bras (6):
> > > > riscv/cmpxchg: Deduplicate cmpxchg() asm functions
> > > > riscv/cmpxchg: Deduplicate cmpxchg() macros
> > > > riscv/cmpxchg: Deduplicate arch_cmpxchg() macros
> > >
> > > > riscv/cmpxchg: Deduplicate xchg() asm functions
> > >
> > > FWIW, this patch seems to break the build pretty badly:
> > > https://patchwork.kernel.org/project/linux-riscv/patch/[email protected]/
> >
> > Thanks for pointing out!
> >
> > It was an intermediary error:
> > Sufix for amoswap on acquire version was "d.aqrl" instead of the
> > correct".d.aqrl", and that caused the fail.
> >
> > I did not notice anything because the next commit made it more general, and thus
> > removed this line of code. I will send a v2-RFC shortly.
> >
> > I see that patch 4/6 has 5 fails, but on each one of them I can see:
> > "I: build output in /ci/workspace/[...]", or
> > ""I: build output in /tmp/[...]".
>
> I don't push the built artifacts anywhere, just the brief logs -
> although the "failed to build" log isn't very helpful other than telling
> you the build broke.
> That seems like a bug w.r.t. where tuxmake prints its output. I'll try
> to fix that.
Thanks for that :)
>
> > I could not find any reference to where this is saved, though.
> > Could you point where can I find the error output, just for the sake of further
> > debugging?
> >
> > >
> > > Patches 1 & 5 also add quite a lot of sparse issues (like 1000), but I
> > > think that may be more of an artifact of the testing process as opposed
> > > to something caused by this patchset.
> >
> > For those I can see the build output diffs. Both added error lines to
> > conchuod/build_rv64_gcc_allmodconfig.
> > I tried to mimic this with [make allmodconfig + gcc build with
> > CROSS_COMPILE=riscv64-linux-gnu- ] but I could not get any error in any patch.
>
> If you can't replicate them, it's probably because they come from
> sparse. I only really mentioned it here in case you went looking at the
> other patches in the series and were wondering why things were so amiss.
Oh, it makes sense.
>
> > For patch 1/6 it removes as much error lines (-) as it adds (+), and the error
> > messages are mostly the same, apart for an line number.
>
> I don't see a line number difference, but rather an increase in the
> number of times the same error lines have appeared in the output.
> I don't find the single-line output from sparse to really be very
> helpful, I should probably have a bit of a think how to present that
> information better.
Oh, I see.
The number at the beginning relates to the number of times the error happens.
Ok it makes sense to me now :)
>
> > For patch 5/6 it actually adds many more lines, but tracking (some of) the
> > errors gave me no idea why.
>
> Probably just sparse being unhappy with some way the macros were
> changed - but some of it ("Should it be static?" bits) look very much
> like the patch causing things to be rebuilt only for the "after the
> patch" build, but somehow not for the "before" build.
Humm, not sure I could understand that last part:
What I get is that the patch 5/6 is causing things to be rebuilt, and
it was not like that on 1-4/6.
Is that what you said?
If so, and you are doing it as an incremental build, changing the
macros in 5/6 should be triggering rebuilds, but it does not make
sense to not be rebuilt in 1-4/6 as they change the same macros.
Thanks for the feedback!
Best regards,
Leo
On Tue, Mar 21, 2023 at 02:01:59PM -0300, Leonardo Bras Soares Passos wrote:
> On Tue, Mar 21, 2023 at 4:49 AM Conor Dooley <[email protected]> wrote:
> > On Tue, Mar 21, 2023 at 03:30:35AM -0300, Leonardo Brás wrote:
> > > On Sun, 2023-03-19 at 20:35 +0000, Conor Dooley wrote:
> > > > On Sat, Mar 18, 2023 at 05:00:54AM -0300, Leonardo Bras wrote:
> > > > > While studying riscv's cmpxchg.h file, I got really interested in
> > > > > understanding how RISCV asm implemented the different versions of
> > > > > {cmp,}xchg.
> > > > >
> > > > > When I understood the pattern, it made sense for me to remove the
> > > > > duplications and create macros to make it easier to understand what exactly
> > > > > changes between the versions: Instruction sufixes & barriers.
> > > > >
> > > > > I split those changes in 3 levels for each cmpxchg and xchg, resulting a
> > > > > total of 6 patches. I did this so it becomes easier to review and remove
> > > > > the last levels if desired, but I have no issue squashing them if it's
> > > > > better.
> > > > >
> > > > > Please provide comments.
> > > > >
> > > > > Thanks!
> > > > > Leo
> > > > >
> > > > > Leonardo Bras (6):
> > > > > riscv/cmpxchg: Deduplicate cmpxchg() asm functions
> > > > > riscv/cmpxchg: Deduplicate cmpxchg() macros
> > > > > riscv/cmpxchg: Deduplicate arch_cmpxchg() macros
> > > >
> > > > > riscv/cmpxchg: Deduplicate xchg() asm functions
> > > >
> > > > FWIW, this patch seems to break the build pretty badly:
> > > > https://patchwork.kernel.org/project/linux-riscv/patch/[email protected]/
> > >
> > > Thanks for pointing out!
> > >
> > > It was an intermediary error:
> > > Sufix for amoswap on acquire version was "d.aqrl" instead of the
> > > correct".d.aqrl", and that caused the fail.
> > >
> > > I did not notice anything because the next commit made it more general, and thus
> > > removed this line of code. I will send a v2-RFC shortly.
> > >
> > > I see that patch 4/6 has 5 fails, but on each one of them I can see:
> > > "I: build output in /ci/workspace/[...]", or
> > > ""I: build output in /tmp/[...]".
> >
> > I don't push the built artifacts anywhere, just the brief logs -
> > although the "failed to build" log isn't very helpful other than telling
> > you the build broke.
> > That seems like a bug w.r.t. where tuxmake prints its output. I'll try
> > to fix that.
>
> Thanks for that :)
I've pushed what I think is a fix, the wrong log file was being grepped
for errors in the case of a failed build.
> > > I could not find any reference to where this is saved, though.
> > > Could you point where can I find the error output, just for the sake of further
> > > debugging?
> > >
> > > >
> > > > Patches 1 & 5 also add quite a lot of sparse issues (like 1000), but I
> > > > think that may be more of an artifact of the testing process as opposed
> > > > to something caused by this patchset.
> > >
> > > For those I can see the build output diffs. Both added error lines to
> > > conchuod/build_rv64_gcc_allmodconfig.
> > > I tried to mimic this with [make allmodconfig + gcc build with
> > > CROSS_COMPILE=riscv64-linux-gnu- ] but I could not get any error in any patch.
> >
> > If you can't replicate them, it's probably because they come from
> > sparse. I only really mentioned it here in case you went looking at the
> > other patches in the series and were wondering why things were so amiss.
>
> Oh, it makes sense.
>
> >
> > > For patch 1/6 it removes as much error lines (-) as it adds (+), and the error
> > > messages are mostly the same, apart for an line number.
> >
> > I don't see a line number difference, but rather an increase in the
> > number of times the same error lines have appeared in the output.
> > I don't find the single-line output from sparse to really be very
> > helpful, I should probably have a bit of a think how to present that
> > information better.
>
> Oh, I see.
> The number at the beginning relates to the number of times the error happens.
> Ok it makes sense to me now :)
>
> >
> > > For patch 5/6 it actually adds many more lines, but tracking (some of) the
> > > errors gave me no idea why.
> >
> > Probably just sparse being unhappy with some way the macros were
> > changed - but some of it ("Should it be static?" bits) look very much
> > like the patch causing things to be rebuilt only for the "after the
> > patch" build, but somehow not for the "before" build.
>
> Humm, not sure I could understand that last part:
> What I get is that the patch 5/6 is causing things to be rebuilt, and
> it was not like that on 1-4/6.
> Is that what you said?
> If so, and you are doing it as an incremental build, changing the
> macros in 5/6 should be triggering rebuilds, but it does not make
> sense to not be rebuilt in 1-4/6 as they change the same macros.
Right, it is an incremental build.
First it builds the tree with a patch applied, then it checks out HEAD~1
and builds that tree. Then it goes back to the tree with the patch
applied and builds it again. The output from builds 2 & 3 are compared
to see if any errors snuck in.
In theory, that should ensure that, as builds 2 & 3 have had the same
diff to the previous one just in opposite directions, the same files get
rebuilt - but I am a little worried that ccache gets involved sometimes
and leads to the before/after builds not being exactly the same.
They may very well be real issues as your refactor has caused the
casting in the macros to change or w/e. Not my area of expertise by any
stretch of the imagination, but the lkp sparse is out of date & doesn't
run any more so I figure it's better to be running the stuff, even if it
does sometimes result in a splurge of errors like this than forget about
poor aul sparse.
Cheers,
Conor.
On Tue, Mar 21, 2023 at 4:48 PM Conor Dooley <[email protected]> wrote:
>
> On Tue, Mar 21, 2023 at 02:01:59PM -0300, Leonardo Bras Soares Passos wrote:
> > On Tue, Mar 21, 2023 at 4:49 AM Conor Dooley <[email protected]> wrote:
> > > On Tue, Mar 21, 2023 at 03:30:35AM -0300, Leonardo Brás wrote:
> > > > On Sun, 2023-03-19 at 20:35 +0000, Conor Dooley wrote:
> > > > > On Sat, Mar 18, 2023 at 05:00:54AM -0300, Leonardo Bras wrote:
> > > > > > While studying riscv's cmpxchg.h file, I got really interested in
> > > > > > understanding how RISCV asm implemented the different versions of
> > > > > > {cmp,}xchg.
> > > > > >
> > > > > > When I understood the pattern, it made sense for me to remove the
> > > > > > duplications and create macros to make it easier to understand what exactly
> > > > > > changes between the versions: Instruction sufixes & barriers.
> > > > > >
> > > > > > I split those changes in 3 levels for each cmpxchg and xchg, resulting a
> > > > > > total of 6 patches. I did this so it becomes easier to review and remove
> > > > > > the last levels if desired, but I have no issue squashing them if it's
> > > > > > better.
> > > > > >
> > > > > > Please provide comments.
> > > > > >
> > > > > > Thanks!
> > > > > > Leo
> > > > > >
> > > > > > Leonardo Bras (6):
> > > > > > riscv/cmpxchg: Deduplicate cmpxchg() asm functions
> > > > > > riscv/cmpxchg: Deduplicate cmpxchg() macros
> > > > > > riscv/cmpxchg: Deduplicate arch_cmpxchg() macros
> > > > >
> > > > > > riscv/cmpxchg: Deduplicate xchg() asm functions
> > > > >
> > > > > FWIW, this patch seems to break the build pretty badly:
> > > > > https://patchwork.kernel.org/project/linux-riscv/patch/[email protected]/
> > > >
> > > > Thanks for pointing out!
> > > >
> > > > It was an intermediary error:
> > > > Sufix for amoswap on acquire version was "d.aqrl" instead of the
> > > > correct".d.aqrl", and that caused the fail.
> > > >
> > > > I did not notice anything because the next commit made it more general, and thus
> > > > removed this line of code. I will send a v2-RFC shortly.
> > > >
> > > > I see that patch 4/6 has 5 fails, but on each one of them I can see:
> > > > "I: build output in /ci/workspace/[...]", or
> > > > ""I: build output in /tmp/[...]".
> > >
> > > I don't push the built artifacts anywhere, just the brief logs -
> > > although the "failed to build" log isn't very helpful other than telling
> > > you the build broke.
> > > That seems like a bug w.r.t. where tuxmake prints its output. I'll try
> > > to fix that.
> >
> > Thanks for that :)
>
> I've pushed what I think is a fix, the wrong log file was being grepped
> for errors in the case of a failed build.
Thanks!
>
> > > > I could not find any reference to where this is saved, though.
> > > > Could you point where can I find the error output, just for the sake of further
> > > > debugging?
> > > >
> > > > >
> > > > > Patches 1 & 5 also add quite a lot of sparse issues (like 1000), but I
> > > > > think that may be more of an artifact of the testing process as opposed
> > > > > to something caused by this patchset.
> > > >
> > > > For those I can see the build output diffs. Both added error lines to
> > > > conchuod/build_rv64_gcc_allmodconfig.
> > > > I tried to mimic this with [make allmodconfig + gcc build with
> > > > CROSS_COMPILE=riscv64-linux-gnu- ] but I could not get any error in any patch.
> > >
> > > If you can't replicate them, it's probably because they come from
> > > sparse. I only really mentioned it here in case you went looking at the
> > > other patches in the series and were wondering why things were so amiss.
> >
> > Oh, it makes sense.
> >
> > >
> > > > For patch 1/6 it removes as much error lines (-) as it adds (+), and the error
> > > > messages are mostly the same, apart for an line number.
> > >
> > > I don't see a line number difference, but rather an increase in the
> > > number of times the same error lines have appeared in the output.
> > > I don't find the single-line output from sparse to really be very
> > > helpful, I should probably have a bit of a think how to present that
> > > information better.
> >
> > Oh, I see.
> > The number at the beginning relates to the number of times the error happens.
> > Ok it makes sense to me now :)
> >
> > >
> > > > For patch 5/6 it actually adds many more lines, but tracking (some of) the
> > > > errors gave me no idea why.
> > >
> > > Probably just sparse being unhappy with some way the macros were
> > > changed - but some of it ("Should it be static?" bits) look very much
> > > like the patch causing things to be rebuilt only for the "after the
> > > patch" build, but somehow not for the "before" build.
> >
> > Humm, not sure I could understand that last part:
> > What I get is that the patch 5/6 is causing things to be rebuilt, and
> > it was not like that on 1-4/6.
> > Is that what you said?
> > If so, and you are doing it as an incremental build, changing the
> > macros in 5/6 should be triggering rebuilds, but it does not make
> > sense to not be rebuilt in 1-4/6 as they change the same macros.
>
> Right, it is an incremental build.
> First it builds the tree with a patch applied, then it checks out HEAD~1
> and builds that tree. Then it goes back to the tree with the patch
> applied and builds it again. The output from builds 2 & 3 are compared
> to see if any errors snuck in.
> In theory, that should ensure that, as builds 2 & 3 have had the same
> diff to the previous one just in opposite directions, the same files get
> rebuilt - but I am a little worried that ccache gets involved sometimes
> and leads to the before/after builds not being exactly the same.
Makes sense to me.
>
> They may very well be real issues as your refactor has caused the
> casting in the macros to change or w/e. Not my area of expertise by any
> stretch of the imagination, but the lkp sparse is out of date & doesn't
> run any more so I figure it's better to be running the stuff, even if it
> does sometimes result in a splurge of errors like this than forget about
> poor aul sparse.
I agree with you.
Better deal with sporadic spurious errors than not testing at all.
>
> Cheers,
> Conor.
Best regards,
Leo
On Wed, Mar 22, 2023 at 12:42:56AM -0300, Leonardo Bras Soares Passos wrote:
> On Tue, Mar 21, 2023 at 4:48 PM Conor Dooley <[email protected]> wrote:
> > On Tue, Mar 21, 2023 at 02:01:59PM -0300, Leonardo Bras Soares Passos wrote:
> > > On Tue, Mar 21, 2023 at 4:49 AM Conor Dooley <[email protected]> wrote:
> > > > On Tue, Mar 21, 2023 at 03:30:35AM -0300, Leonardo Brás wrote:
> > > > > For patch 5/6 it actually adds many more lines, but tracking (some of) the
> > > > > errors gave me no idea why.
> > > >
> > > > Probably just sparse being unhappy with some way the macros were
> > > > changed - but some of it ("Should it be static?" bits) look very much
> > > > like the patch causing things to be rebuilt only for the "after the
> > > > patch" build, but somehow not for the "before" build.
> > >
> > > Humm, not sure I could understand that last part:
> > > What I get is that the patch 5/6 is causing things to be rebuilt, and
> > > it was not like that on 1-4/6.
> > > Is that what you said?
> > > If so, and you are doing it as an incremental build, changing the
> > > macros in 5/6 should be triggering rebuilds, but it does not make
> > > sense to not be rebuilt in 1-4/6 as they change the same macros.
> >
> > Right, it is an incremental build.
> > First it builds the tree with a patch applied, then it checks out HEAD~1
> > and builds that tree. Then it goes back to the tree with the patch
> > applied and builds it again. The output from builds 2 & 3 are compared
> > to see if any errors snuck in.
> > In theory, that should ensure that, as builds 2 & 3 have had the same
> > diff to the previous one just in opposite directions, the same files get
> > rebuilt - but I am a little worried that ccache gets involved sometimes
> > and leads to the before/after builds not being exactly the same.
>
> Makes sense to me.
Oh, the other thing that I didn't notice until now is that, if, as
in your case, HEAD~1 for the patch currently being tested does not build
but the HEAD build does, the HEAD build will have more sparse coverage
than the HEAD~1 one. Cos of that, sparse is likely to find more issues in
the after build, and you end up with +1000 error count like patch 5 in
the v1.
If you look at your most recent version, that doesn't have build issues,
patch 5 gets the all-clear:
https://patchwork.kernel.org/project/linux-riscv/list/?series=732217
Not sure why I didn't notice that before, I completely forgot that 4/6
had build issues and it should've been immediately obvious to me why 5/6
had a bunch of extra sparse warnings...