Skip to contentSkip to conversation

RISC-V: fixup in-flight collision with ARCH_WANT_OPTIMIZE_VMEMMAP rename

PATCH Applied Conor Dooley 4 participants 8 messages Apr 29, 2023 – May 2, 2023 RISC-V

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 RISC-V: fixup in-flight collision with ARCH_WANT_OPTIMIZE_VMEMMAP rename

From: Conor Dooley <[email protected]>

Lukas warned that ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP had been renamed in the mm tree & that RISC-V would need a fixup as part of the merge. The warning was missed however, and RISC-V is selecting the orphaned Kconfig option.

Fixes 89d77f71f493 ("Merge tag 'riscv-for-linus-6.4-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux")

---
I guess you picking this up for your second PR makes the most sense Palmer?

CC Palmer Dabbelt <[email protected]>
CC Linus Torvalds <[email protected]>
CC Andrew Morton <[email protected]>
CC Aneesh Kumar K.V <[email protected]>
CC Lukas Bulwhan <[email protected]>
---
 arch/riscv/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/riscv/Kconfig +1 -1
index 5c88ac4b52be..e1bdb3fb16cc 100644
@@ -46,9 +46,9 @@ config RISCV
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT if MMU
 	select ARCH_WANT_FRAME_POINTERS
 	select ARCH_WANT_GENERAL_HUGETLB if !RISCV_ISA_SVNAPOT
-	select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
 	select ARCH_WANT_HUGE_PMD_SHARE if 64BIT
 	select ARCH_WANT_LD_ORPHAN_WARN if !XIP_KERNEL
+	select ARCH_WANT_OPTIMIZE_VMEMMAP
 	select ARCH_WANTS_THP_SWAP if HAVE_ARCH_TRANSPARENT_HUGEPAGE
 	select BINFMT_FLAT_NO_DATA_START_OFFSET if !MMU
 	select BUILDTIME_TABLE_SORT if MMU
Signature · 7 lines
2.39.2


_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv
#1 lore raw
Quoted 9 lines · Conor Dooley

From: Conor Dooley <[email protected]>

Lukas warned that ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP had been renamed in the mm tree & that RISC-V would need a fixup as part of the merge. The warning was missed however, and RISC-V is selecting the orphaned Kconfig option.

Fixes 89d77f71f493 ("Merge tag 'riscv-for-linus-6.4-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux")

That should have been Lukas Bulwahn <[email protected]>. This is what I get for hand-typing an email address that I could have copied from lore I suppose.

Quoted 35 lines

---
I guess you picking this up for your second PR makes the most sense Palmer?

CC Palmer Dabbelt <[email protected]>
CC Linus Torvalds <[email protected]>
CC Andrew Morton <[email protected]>
CC Aneesh Kumar K.V <[email protected]>
CC Lukas Bulwhan <[email protected]>
---
 arch/riscv/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/riscv/Kconfig +1 -1
index 5c88ac4b52be..e1bdb3fb16cc 100644
@@ -46,9 +46,9 @@ config RISCV
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT if MMU
 	select ARCH_WANT_FRAME_POINTERS
 	select ARCH_WANT_GENERAL_HUGETLB if !RISCV_ISA_SVNAPOT
-	select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
 	select ARCH_WANT_HUGE_PMD_SHARE if 64BIT
 	select ARCH_WANT_LD_ORPHAN_WARN if !XIP_KERNEL
+	select ARCH_WANT_OPTIMIZE_VMEMMAP
 	select ARCH_WANTS_THP_SWAP if HAVE_ARCH_TRANSPARENT_HUGEPAGE
 	select BINFMT_FLAT_NO_DATA_START_OFFSET if !MMU
 	select BUILDTIME_TABLE_SORT if MMU
Signature · 1 line
2.39.2
maintainer Reviewed-by
#2 lore raw
Quoted 8 lines · Conor Dooley
Quoted 7 lines · Conor Dooley

From: Conor Dooley <[email protected]>

Lukas warned that ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP had been renamed in the mm tree & that RISC-V would need a fixup as part of the merge. The warning was missed however, and RISC-V is selecting the orphaned Kconfig option.

Quoted 12 lines
Quoted 2 lines
Fixes 89d77f71f493 ("Merge tag 'riscv-for-linus-6.4-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux")

That should have been Lukas Bulwahn <[email protected]>. This is what I get for hand-typing an email address that I could have copied from lore I suppose.

Quoted 5 lines

---
I guess you picking this up for your second PR makes the most sense Palmer?

Yep, except it crashes b4

$ b4 shazam -lts https://lore.kernel.org/all/20230429-trilogy-jolly-12bf5c53d62d@spud/
Grabbing thread from lore.kernel.org/all/20230429-trilogy-jolly-12bf5c53d62d%40spud/t.mbox.gz
Analyzing 2 messages in the thread
Checking attestation on all messages, may take a moment...
Traceback (most recent call last):
 File "/home/palmer/.local/src/b4/b4/command.py", line 383, in <module>
   cmd()
 File "/home/palmer/.local/src/b4/b4/command.py", line 366, in cmd
   cmdargs.func(cmdargs)
 File "/home/palmer/.local/src/b4/b4/command.py", line 100, in cmd_shazam
   b4.mbox.main(cmdargs)
 File "/home/palmer/.local/src/b4/b4/mbox.py", line 713, in main
   make_am(msgs, cmdargs, msgid)
 File "/home/palmer/.local/src/b4/b4/mbox.py", line 101, in make_am
   am_msgs = lser.get_am_ready(noaddtrailers=cmdargs.noaddtrailers,
 File "/home/palmer/.local/src/b4/b4/__init__.py", line 552, in get_am_ready
   checkmark, trailers, attcrit = lmsg.get_attestation_trailers(attpolicy, maxdays)
 File "/home/palmer/.local/src/b4/b4/__init__.py", line 1319, in get_attestation_trailers
   for attestor in self.attestors:
 File "/home/palmer/.local/src/b4/b4/__init__.py", line 1158, in attestors
   self._load_patatt_attestors()
 File "/home/palmer/.local/src/b4/b4/__init__.py", line 1261, in _load_patatt_attestors
   patatt_config = patatt.get_config_from_git(r'patatt\..*', multivals=['keyringsrc'])
AttributeError: module 'patatt' has no attribute 'get_config_from_git'

So I probably need to just go update my setup, I've got a bunch of wacky stuff. I'll try and remember to do so, but I'm going to scrub through other bits first...

Quoted 30 lines
Quoted 30 lines
CC Palmer Dabbelt <[email protected]>
CC Linus Torvalds <[email protected]>
CC Andrew Morton <[email protected]>
CC Aneesh Kumar K.V <[email protected]>
CC Lukas Bulwhan <[email protected]>
---
 arch/riscv/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/riscv/Kconfig +1 -1
index 5c88ac4b52be..e1bdb3fb16cc 100644
@@ -46,9 +46,9 @@ config RISCV
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT if MMU
 	select ARCH_WANT_FRAME_POINTERS
 	select ARCH_WANT_GENERAL_HUGETLB if !RISCV_ISA_SVNAPOT
-	select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
 	select ARCH_WANT_HUGE_PMD_SHARE if 64BIT
 	select ARCH_WANT_LD_ORPHAN_WARN if !XIP_KERNEL
+	select ARCH_WANT_OPTIMIZE_VMEMMAP
 	select ARCH_WANTS_THP_SWAP if HAVE_ARCH_TRANSPARENT_HUGEPAGE
 	select BINFMT_FLAT_NO_DATA_START_OFFSET if !MMU
 	select BUILDTIME_TABLE_SORT if MMU
Signature · 1 line
2.39.2
#3 lore raw
Quoted 30 lines · Palmer Dabbelt
Quoted 2 lines
Quoted 2 lines

I guess you picking this up for your second PR makes the most sense Palmer?

Yep, except it crashes b4

$ b4 shazam -lts https://lore.kernel.org/all/20230429-trilogy-jolly-12bf5c53d62d@spud
Grabbing thread from lore.kernel.org/all/20230429-trilogy-jolly-12bf5c53d62d%40spud/t.mbox.gz
Analyzing 2 messages in the thread
Checking attestation on all messages, may take a moment...
Traceback (most recent call last):
File "/home/palmer/.local/src/b4/b4/command.py", line 383, in <module>
cmd()
File "/home/palmer/.local/src/b4/b4/command.py", line 366, in cmd
cmdargs.func(cmdargs)
File "/home/palmer/.local/src/b4/b4/command.py", line 100, in cmd_shazam
b4.mbox.main(cmdargs)
File "/home/palmer/.local/src/b4/b4/mbox.py", line 713, in main
make_am(msgs, cmdargs, msgid)
File "/home/palmer/.local/src/b4/b4/mbox.py", line 101, in make_am
am_msgs = lser.get_am_ready(noaddtrailers=cmdargs.noaddtrailers,
File "/home/palmer/.local/src/b4/b4/__init__.py", line 552, in get_am_ready
checkmark, trailers, attcrit = lmsg.get_attestation_trailers(attpolicy, maxdays)
File "/home/palmer/.local/src/b4/b4/__init__.py", line 1319, in get_attestation_trailers
for attestor in self.attestors:
File "/home/palmer/.local/src/b4/b4/__init__.py", line 1158, in attestors
self._load_patatt_attestors()
File "/home/palmer/.local/src/b4/b4/__init__.py", line 1261, in _load_patatt_attestors
patatt_config = patatt.get_config_from_git(r'patatt\..*', multivals=['keyringsrc'])
AttributeError: module 'patatt' has no attribute 'get_config_from_git'
Quoted 3 lines

So I probably need to just go update my setup, I've got a bunch of wacky stuff. I'll try and remember to do so, but I'm going to scrub through other bits first...

Indeed, you have a newer version of b4 without also updating the patatt dependency. If you're using b4 from git, you need to "git submodule update".

Regards,
-K

maintainer
#4 lore raw
Quoted 31 lines · On Sat, 29 Apr 2023 14:48:18 PDT (-0700), [email protected] wrote:
Quoted 31 lines
Quoted 2 lines · Palmer Dabbelt
Quoted 2 lines

I guess you picking this up for your second PR makes the most sense Palmer?

Yep, except it crashes b4

$ b4 shazam -lts https://lore.kernel.org/all/20230429-trilogy-jolly-12bf5c53d62d@spud
Grabbing thread from lore.kernel.org/all/20230429-trilogy-jolly-12bf5c53d62d%40spud/t.mbox.gz
Analyzing 2 messages in the thread
Checking attestation on all messages, may take a moment...
Traceback (most recent call last):
File "/home/palmer/.local/src/b4/b4/command.py", line 383, in <module>
cmd()
File "/home/palmer/.local/src/b4/b4/command.py", line 366, in cmd
cmdargs.func(cmdargs)
File "/home/palmer/.local/src/b4/b4/command.py", line 100, in cmd_shazam
b4.mbox.main(cmdargs)
File "/home/palmer/.local/src/b4/b4/mbox.py", line 713, in main
make_am(msgs, cmdargs, msgid)
File "/home/palmer/.local/src/b4/b4/mbox.py", line 101, in make_am
am_msgs = lser.get_am_ready(noaddtrailers=cmdargs.noaddtrailers,
File "/home/palmer/.local/src/b4/b4/__init__.py", line 552, in get_am_ready
checkmark, trailers, attcrit = lmsg.get_attestation_trailers(attpolicy, maxdays)
File "/home/palmer/.local/src/b4/b4/__init__.py", line 1319, in get_attestation_trailers
for attestor in self.attestors:
File "/home/palmer/.local/src/b4/b4/__init__.py", line 1158, in attestors
self._load_patatt_attestors()
File "/home/palmer/.local/src/b4/b4/__init__.py", line 1261, in _load_patatt_attestors
patatt_config = patatt.get_config_from_git(r'patatt\..*', multivals=['keyringsrc'])
AttributeError: module 'patatt' has no attribute 'get_config_from_git'
Quoted 5 lines
Quoted 3 lines

So I probably need to just go update my setup, I've got a bunch of wacky stuff. I'll try and remember to do so, but I'm going to scrub through other bits first...

Indeed, you have a newer version of b4 without also updating the patatt dependency. If you're using b4 from git, you need to "git submodule update".

Thanks, it's fixed. This one's in the queue for testing, but it looks generally fine. Might be a bit slower than usual, as I'm also running the glibc tests.

maintainer
#5 lore raw
In reply to Conor Dooley
Quoted 8 lines · Conor Dooley
Quoted 7 lines · Conor Dooley

From: Conor Dooley <[email protected]>

Lukas warned that ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP had been renamed in the mm tree & that RISC-V would need a fixup as part of the merge. The warning was missed however, and RISC-V is selecting the orphaned Kconfig option.

Quoted 4 lines
Quoted 2 lines
Fixes 89d77f71f493 ("Merge tag 'riscv-for-linus-6.4-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux")

That should have been Lukas Bulwahn <[email protected]>.

I think I got that right, but the email bounced. I'm not sure if it's just the list issues, though.

Quoted 38 lines

This is what I get for hand-typing an email address that I could have copied from lore I suppose.

Quoted 35 lines

---
I guess you picking this up for your second PR makes the most sense Palmer?

CC Palmer Dabbelt <[email protected]>
CC Linus Torvalds <[email protected]>
CC Andrew Morton <[email protected]>
CC Aneesh Kumar K.V <[email protected]>
CC Lukas Bulwhan <[email protected]>
---
 arch/riscv/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/riscv/Kconfig +1 -1
index 5c88ac4b52be..e1bdb3fb16cc 100644
@@ -46,9 +46,9 @@ config RISCV
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT if MMU
 	select ARCH_WANT_FRAME_POINTERS
 	select ARCH_WANT_GENERAL_HUGETLB if !RISCV_ISA_SVNAPOT
-	select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
 	select ARCH_WANT_HUGE_PMD_SHARE if 64BIT
 	select ARCH_WANT_LD_ORPHAN_WARN if !XIP_KERNEL
+	select ARCH_WANT_OPTIMIZE_VMEMMAP
 	select ARCH_WANTS_THP_SWAP if HAVE_ARCH_TRANSPARENT_HUGEPAGE
 	select BINFMT_FLAT_NO_DATA_START_OFFSET if !MMU
 	select BUILDTIME_TABLE_SORT if MMU
Signature · 1 line
2.39.2
#6 lore raw
Quoted 9 lines · Palmer Dabbelt
Quoted 8 lines · Conor Dooley
Quoted 7 lines · Conor Dooley

From: Conor Dooley <[email protected]>

Lukas warned that ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP had been renamed in the mm tree & that RISC-V would need a fixup as part of the merge. The warning was missed however, and RISC-V is selecting the orphaned Kconfig option.

Quoted 6 lines
Quoted 4 lines
Quoted 2 lines
Fixes 89d77f71f493 ("Merge tag 'riscv-for-linus-6.4-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux")

That should have been Lukas Bulwahn <[email protected]>.

I think I got that right, but the email bounced. I'm not sure if it's just the list issues, though.

B4 probably just replied to the address in my original email, rather than the corrected one?

Quoted 31 lines
Quoted 31 lines

This is what I get for hand-typing an email address that I could have copied from lore I suppose.

Quoted 27 lines

---
I guess you picking this up for your second PR makes the most sense Palmer?

CC Palmer Dabbelt <[email protected]>
CC Linus Torvalds <[email protected]>
CC Andrew Morton <[email protected]>
CC Aneesh Kumar K.V <[email protected]>
CC Lukas Bulwhan <[email protected]>
---
 arch/riscv/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/riscv/Kconfig +0 -1
index 5c88ac4b52be..e1bdb3fb16cc 100644
@@ -46,9 +46,9 @@ config RISCV
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT if MMU
 	select ARCH_WANT_FRAME_POINTERS
 	select ARCH_WANT_GENERAL_HUGETLB if !RISCV_ISA_SVNAPOT
-	select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
 	select ARCH_WANT_HUGE_PMD_SHARE if 64BIT
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
arch/riscv/Kconfig +1 -0
 	select ARCH_WANT_LD_ORPHAN_WARN if !XIP_KERNEL
+	select ARCH_WANT_OPTIMIZE_VMEMMAP
 	select ARCH_WANTS_THP_SWAP if HAVE_ARCH_TRANSPARENT_HUGEPAGE
 	select BINFMT_FLAT_NO_DATA_START_OFFSET if !MMU
 	select BUILDTIME_TABLE_SORT if MMU
Signature · 1 line
2.39.2

_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv

#7 lore raw
In reply to Conor Dooley
Quoted 6 lines · Conor Dooley

Lukas warned that ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP had been renamed in the mm tree & that RISC-V would need a fixup as part of the merge. The warning was missed however, and RISC-V is selecting the orphaned Kconfig option.

Applied, thanks!

[1/1] RISC-V: fixup in-flight collision with ARCH_WANT_OPTIMIZE_VMEMMAP rename

      https://git.kernel.org/palmer/c/26b0812f4cf8

Best regards,

Signature · 1 line
Palmer Dabbelt <[email protected]>