m68k build failures in -next: undefined reference to `arch_dma_prep_coherent'
I see the following build error in -next:
kernel/dma/direct.o: In function `dma_direct_alloc_pages':
direct.c:(.text+0x4d8): undefined reference to `arch_dma_prep_coherent'
Example: m68k:allnoconfig.
Bisect log is ambiguous and points to the merge of m68k/for-next into -next. Yet, I think the problem is with commit 69878ef47562 ("m68k:
Implement arch_dma_prep_coherent()") which is supposed to introduce the function. The problem is likely that arch_dma_prep_coherent() is only declared if CONFIG_MMU is enabled, but it is called from code outside CONFIG_MMU.
Guenter
---
# bad: [d58b5ab90ee7528126fd5833df7fc5bda8331ce8] Add linux-next specific files for 20190708 # good: [6fbc7275c7a9ba97877050335f290341a1fd8dbf] Linux 5.2-rc7
git bisect start 'HEAD' 'v5.2-rc7'
# bad: [ba30fb6d5d6464bd7d3759408ea7f178d8c9fe87] Merge remote-tracking branch 'crypto/master' git bisect bad ba30fb6d5d6464bd7d3759408ea7f178d8c9fe87
# bad: [4226c2bec4757a3acbf86bb836375d3966ab72ce] Merge remote-tracking branch 'i2c/i2c/for-next' git bisect bad 4226c2bec4757a3acbf86bb836375d3966ab72ce
# good: [e41aad4a290783ec7d3730542cbed0e99b2dcb4a] Merge remote-tracking branch 'tegra/for-next' git bisect good e41aad4a290783ec7d3730542cbed0e99b2dcb4a
# bad: [e495984468b15cb8fa276b63fff3986111a513a5] Merge remote-tracking branch 'ceph/master' git bisect bad e495984468b15cb8fa276b63fff3986111a513a5
# bad: [8e8fefda572360f00854547f3458a9c2cf932ff5] Merge remote-tracking branch 'powerpc/next' git bisect bad 8e8fefda572360f00854547f3458a9c2cf932ff5
# bad: [01fd0e565283d69adf0ff1da95cab5bb4cb58acb] Merge remote-tracking branch 'm68k/for-next' git bisect bad 01fd0e565283d69adf0ff1da95cab5bb4cb58acb
# good: [7dcbe273f9c61b83e398a19152f6529d1676751d] Merge branch 'clk-allwinner' into clk-next git bisect good 7dcbe273f9c61b83e398a19152f6529d1676751d
# good: [59f375866fa68888fedc7bcd2dad381ceb6ba4dc] Merge branch 'clk-sprd' into clk-next git bisect good 59f375866fa68888fedc7bcd2dad381ceb6ba4dc
# good: [49193327486022abc85b24d058aefa666367ffff] Merge branch 'clk-lochnagar' into clk-next git bisect good 49193327486022abc85b24d058aefa666367ffff
# good: [7b4e6f9a4b8daa37a86cd7bca46bf70b522dbb62] csky: Add new asid lib code from arm git bisect good 7b4e6f9a4b8daa37a86cd7bca46bf70b522dbb62
# good: [c6d41a2a548b8b684a006ab77d42c73815105011] Merge remote-tracking branch 'csky/linux-next' git bisect good c6d41a2a548b8b684a006ab77d42c73815105011
# good: [69878ef47562f32e02d0b7975c990e1c0339320d] m68k: Implement arch_dma_prep_coherent() git bisect good 69878ef47562f32e02d0b7975c990e1c0339320d
# good: [ccd1eda4ac7a73806ccc6afa899bb58e00c7772f] Merge remote-tracking branch 'h8300/h8300-next' git bisect good ccd1eda4ac7a73806ccc6afa899bb58e00c7772f
# first bad commit: [01fd0e565283d69adf0ff1da95cab5bb4cb58acb] Merge remote-tracking branch 'm68k/for-next'
Hi Günter,
Quoted 5 lines · Guenter Roeck
I see the following build error in -next:
kernel/dma/direct.o: In function `dma_direct_alloc_pages':
direct.c:(.text+0x4d8): undefined reference to `arch_dma_prep_coherent'
Quoted 8 lines
Example: m68k:allnoconfig.
Bisect log is ambiguous and points to the merge of m68k/for-next into -next. Yet, I think the problem is with commit
69878ef47562("m68k:
Implementarch_dma_prep_coherent()") which is supposed to introduce the function. The problem is likely thatarch_dma_prep_coherent()is only declared ifCONFIG_MMUis enabled, but it is called from code outsideCONFIG_MMU.
Thanks, one more thing to fix in m68k-allnoconfig (did it really build before?)...
Given you say "example", does it fail in real configs, too?
Thanks again!
Gr{oetje,eeting}s,
Geert
Signature · 5 lines
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected] In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds
Quoted 17 lines · Geert Uytterhoeven
Hi Günter,
Quoted 5 lines · Guenter Roeck
I see the following build error in -next:
kernel/dma/direct.o: In function `dma_direct_alloc_pages':
direct.c:(.text+0x4d8): undefined reference to `arch_dma_prep_coherent'Quoted 8 lines
Example: m68k:allnoconfig.
Bisect log is ambiguous and points to the merge of m68k/for-next into -next. Yet, I think the problem is with commit
69878ef47562("m68k:
Implementarch_dma_prep_coherent()") which is supposed to introduce the function. The problem is likely thatarch_dma_prep_coherent()is only declared ifCONFIG_MMUis enabled, but it is called from code outsideCONFIG_MMU.
Quoted 5 lines
Thanks, one more thing to fix in m68k-allnoconfig (did it really build before?)...
Given you say "example", does it fail in real configs, too?
Yes, it does. All nommu builds fail. allnoconfig and tinyconfig just happen to be among those.
Building m68k:allnoconfig ... failed
Building m68k:tinyconfig ... failed
Building m68k:m5272c3_defconfig ... failed
Building m68k:m5307c3_defconfig ... failed
Building m68k:m5249evb_defconfig ... failed
Building m68k:m5407c3_defconfig ... failed
Building m68k:m5475evb_defconfig ... failed
Error is always the same.
The error started with next-20190702. Prior to that, builds were fine, including m68k:allnoconfig and m68k:tinyconfig.
Guenter
Hi Günter,
CC Greg
Quoted 32 lines · Guenter Roeck
Quoted 15 lines · Geert Uytterhoeven
Quoted 5 lines · Guenter Roeck
I see the following build error in -next:
kernel/dma/direct.o: In function `dma_direct_alloc_pages':
direct.c:(.text+0x4d8): undefined reference to `arch_dma_prep_coherent'Quoted 8 lines
Example: m68k:allnoconfig.
Bisect log is ambiguous and points to the merge of m68k/for-next into -next. Yet, I think the problem is with commit
69878ef47562("m68k:
Implementarch_dma_prep_coherent()") which is supposed to introduce the function. The problem is likely thatarch_dma_prep_coherent()is only declared ifCONFIG_MMUis enabled, but it is called from code outsideCONFIG_MMU.Quoted 4 lines
Thanks, one more thing to fix in m68k-allnoconfig (did it really build before?)...
Given you say "example", does it fail in real configs, too?
Yes, it does. All nommu builds fail. allnoconfig and tinyconfig just happen to be among those.
Building m68k:allnoconfig ... failed
Building m68k:tinyconfig ... failed
Building m68k:m5272c3_defconfig ... failed
Building m68k:m5307c3_defconfig ... failed
Building m68k:m5249evb_defconfig ... failed
Building m68k:m5407c3_defconfig ... failed
Building m68k:m5475evb_defconfig ... failed
Quoted 1 line
Error is always the same.
Thanks!
Quoted 2 lines
The error started with next-20190702. Prior to that, builds were fine, including m68k:allnoconfig and m68k:tinyconfig.
Yeah, it started when I queued up the DMA rework.
I didn't double-check when Greg said it was OK for him, as it wouldn't affect Coldfire or mmu. Sorry for that.
And that has just been pulled by Linus... Oops...
Gr{oetje,eeting}s,
Geert
Signature · 5 lines
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected] In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds
Quoted 4 lines · Geert Uytterhoeven
Yeah, it started when I queued up the DMA rework.
I didn't double-check when Greg said it was OK for him, as it wouldn't affect Coldfire or mmu. Sorry for that.
And that has just been pulled by Linus... Oops...
The fix should have been in your inbox for a while..
Hi Christoph,
Quoted 5 lines · Christoph Hellwig
Quoted 4 lines · Geert Uytterhoeven
Yeah, it started when I queued up the DMA rework.
I didn't double-check when Greg said it was OK for him, as it wouldn't affect Coldfire or mmu. Sorry for that.
And that has just been pulled by Linus... Oops...
Note that the build failure is more subtle: both m5307c3_defconfig and m5475evb_defconfig build fine in m68k/for-linus, but fail in next-20190708. So it fails when combined with other changes, going in through a different tree (the DMA tree?).
Quoted 1 line
The fix should have been in your inbox for a while..
Thanks, your patch fixes that.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Quoted 4 lines · Geert Uytterhoeven
Note that the build failure is more subtle: both
m5307c3_defconfigandm5475evb_defconfigbuild fine in m68k/for-linus, but fail in next-20190708. So it fails when combined with other changes, going in through a different tree (the DMA tree?).
Yes, the dma tree adds the stub for the symbol and adds code relying on that.
Quoted 8 lines · Christoph Hellwig
Quoted 4 lines · Geert Uytterhoeven
Note that the build failure is more subtle: both
m5307c3_defconfigandm5475evb_defconfigbuild fine in m68k/for-linus, but fail in next-20190708. So it fails when combined with other changes, going in through a different tree (the DMA tree?).Yes, the dma tree adds the stub for the symbol and adds code relying on that.
If you give me an ack I can add the fix to the dma-mapping pull request.
Hi Christoph,
Quoted 11 lines · Christoph Hellwig
Quoted 8 lines · Christoph Hellwig
Quoted 4 lines · Geert Uytterhoeven
Note that the build failure is more subtle: both
m5307c3_defconfigandm5475evb_defconfigbuild fine in m68k/for-linus, but fail in next-20190708. So it fails when combined with other changes, going in through a different tree (the DMA tree?).Yes, the dma tree adds the stub for the symbol and adds code relying on that.
If you give me an ack I can add the fix to the dma-mapping pull request.
Thanks, done.
Gr{oetje,eeting}s,
Geert
Signature · 5 lines
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected] In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds
Quoted 41 lines · Geert Uytterhoeven
Hi Günter,
CC Greg
Quoted 32 lines · Guenter Roeck
Quoted 15 lines · Geert Uytterhoeven
Quoted 5 lines · Guenter Roeck
I see the following build error in -next:
kernel/dma/direct.o: In function `dma_direct_alloc_pages':
direct.c:(.text+0x4d8): undefined reference to `arch_dma_prep_coherent'Quoted 8 lines
Example: m68k:allnoconfig.
Bisect log is ambiguous and points to the merge of m68k/for-next into -next. Yet, I think the problem is with commit
69878ef47562("m68k:
Implementarch_dma_prep_coherent()") which is supposed to introduce the function. The problem is likely thatarch_dma_prep_coherent()is only declared ifCONFIG_MMUis enabled, but it is called from code outsideCONFIG_MMU.Quoted 4 lines
Thanks, one more thing to fix in m68k-allnoconfig (did it really build before?)...
Given you say "example", does it fail in real configs, too?
Yes, it does. All nommu builds fail. allnoconfig and tinyconfig just happen to be among those.
Building m68k:allnoconfig ... failed
Building m68k:tinyconfig ... failed
Building m68k:m5272c3_defconfig ... failed
Building m68k:m5307c3_defconfig ... failed
Building m68k:m5249evb_defconfig ... failed
Building m68k:m5407c3_defconfig ... failed
Building m68k:m5475evb_defconfig ... failedQuoted 1 line
Error is always the same.
Thanks!
Quoted 8 lines
Quoted 2 lines
The error started with next-20190702. Prior to that, builds were fine, including m68k:allnoconfig and m68k:tinyconfig.
Yeah, it started when I queued up the DMA rework.
I didn't double-check when Greg said it was OK for him, as it wouldn't affect Coldfire or mmu. Sorry for that.
And that has just been pulled by Linus... Oops...
Oh well, I would have hoped for another rc, not because of the state of mainline but for the state of -next. The next release may be fun, in a negative sense. From the build of next-20190705:
Build results:
total: 158 pass: 147 fail: 11 Qemu test results: total: 364 pass: 41 fail: 323
People have been making lots of last-minute (ie in the week before the commit window opens, and just before a holiday weekend) untested changes. Unfortunately that happens a lot lately.
To be fair, most of the qemu failures are due to messed up btrfs dependencies (or at least I hope so - hard to say with that many failures), but still ...
Guenter