Linaro-mm-sig

linaro-mm-sig@lists.linaro.org

15 participants
3030 discussions

by Jiapeng Chong

Clean up the following includecheck warning: ./drivers/gpu/drm/drm_gem_shmem_helper.c: linux/module.h is included more than once. Reported-by: Abaci Robot <abaci(a)linux.alibaba.com> Signed-off-by: Jiapeng Chong <jiapeng.chong(a)linux.alibaba.com> --- drivers/gpu/drm/drm_gem_shmem_helper.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/drm_gem_shmem_helper.c b/drivers/gpu/drm/drm_gem_shmem_helper.c index 62192411..7915047 100644 --- a/drivers/gpu/drm/drm_gem_shmem_helper.c +++ b/drivers/gpu/drm/drm_gem_shmem_helper.c @@ -10,7 +10,6 @@ #include <linux/shmem_fs.h> #include <linux/slab.h> #include <linux/vmalloc.h> -#include <linux/module.h> #ifdef CONFIG_X86 #include <asm/set_memory.h> -- 1.8.3.1

3 years, 9 months

Re: [Linaro-mm-sig] [Intel-gfx] [RFC PATCH 1/2] dma-fence: Avoid establishing a locking order between fence classes

by Daniel Vetter

Once more an entire week behind on mails, but this looked interesting enough. On Fri, Dec 03, 2021 at 03:18:01PM +0100, Thomas Hellström wrote: > On Fri, 2021-12-03 at 14:08 +0100, Christian König wrote: > > Am 01.12.21 um 13:16 schrieb Thomas Hellström (Intel): > > > > > > On 12/1/21 12:25, Christian König wrote: > > > > And why do you use dma_fence_chain to generate a timeline for > > > > TTM? > > > > That should come naturally because all the moves must be ordered. > > > > > > Oh, in this case because we're looking at adding stuff at the end > > > of > > > migration (like coalescing object shared fences and / or async > > > unbind > > > fences), which may not complete in order. > > > > Well that's ok as well. My question is why does this single dma_fence > > then shows up in the dma_fence_chain representing the whole > > migration? > > What we'd like to happen during eviction is that we > > 1) await any exclusive- or moving fences, then schedule the migration > blit. The blit manages its own GPU ptes. Results in a single fence. > 2) Schedule unbind of any gpu vmas, resulting possibly in multiple > fences. This sounds like over-optimizing for nothing. We only really care about pipeling moves on dgpu, and on dgpu we only care about modern userspace (because even gl moves in that direction). And modern means that usually even write access is only setting a read fence, because in vk/compute we only set write fences for object which need implicit sync, and _only_ when actually needed. So ignoring read fences for movings "because it's only reads" is actually busted. I think for buffer moves we should document and enforce (in review) the rule that you have to wait for all fences, otherwise boom. Same really like before freeing backing storage. Otherwise there's just too many gaps and surprises. And yes with Christian's rework of dma_resv this will change, and we'll allow multiple write fences (because that's what amdgpu encoded into their uapi). Still means that you cannot move a buffer without waiting for read fences (or kernel fences or anything really). The other thing is this entire spinlock recursion topic for dma_fence, and I'm deeply unhappy about the truckload of tricks i915 plays and hence in favour of avoiding recursion in this area as much as possible. If we really can't avoid it then irq_work to get a new clean context gets the job done. Making this messy and work is imo a feature, lock nesting of same level locks is just not a good&robust engineering idea. /me back to being completely burried I do hope I can find some more time to review a few more of Christian's patches this week though :-/ Cheers, Daniel > 3) Most but not all of the remaining resv shared fences will have been > finished in 2) We can't easily tell which so we have a couple of shared > fences left. > 4) Add all fences resulting from 1) 2) and 3) into the per-memory-type > dma-fence-chain. > 5) hand the resulting dma-fence-chain representing the end of migration > over to ttm's resource manager. > > Now this means we have a dma-fence-chain disguised as a dma-fence out > in the wild, and it could in theory reappear as a 3) fence for another > migration unless a very careful audit is done, or as an input to the > dma-fence-array used for that single dependency. > > > > > That somehow doesn't seem to make sense because each individual step > > of > > the migration needs to wait for those dependencies as well even when > > it > > runs in parallel. > > > > > But that's not really the point, the point was that an (at least to > > > me) seemingly harmless usage pattern, be it real or fictious, ends > > > up > > > giving you severe internal- or cross-driver headaches. > > > > Yeah, we probably should document that better. But in general I don't > > see much reason to allow mixing containers. The dma_fence_array and > > dma_fence_chain objects have some distinct use cases and and using > > them > > to build up larger dependency structures sounds really questionable. > > Yes, I tend to agree to some extent here. Perhaps add warnings when > adding a chain or array as an input to array and when accidently > joining chains, and provide helpers for flattening if needed. > > /Thomas > > > > > > Christian. > > > > > > > > /Thomas > > > > > > > > > > > > > > Regards, > > > > Christian. > > > > > > > > > > > > -- Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch

3 years, 9 months

Re: [Linaro-mm-sig] [RFC PATCH v4 0/2] RDMA/rxe: Add dma-buf support

by Jason Gunthorpe

On Fri, Dec 03, 2021 at 12:51:44PM +0900, Shunsuke Mie wrote: > Hi maintainers, > > Could you please review this patch series? Why is it RFC? I'm confused why this is useful? This can't do copy from MMIO memory, so it shouldn't be compatible with things like Gaudi - does something prevent this? Jason

3 years, 9 months

Re: [Linaro-mm-sig] [RFC PATCH 1/2] dma-fence: Avoid establishing a locking order between fence classes

by Christian König

Am 03.12.21 um 15:50 schrieb Thomas Hellström: > > On 12/3/21 15:26, Christian König wrote: >> [Adding Daniel here as well] >> >> Am 03.12.21 um 15:18 schrieb Thomas Hellström: >>> [SNIP] >>>> Well that's ok as well. My question is why does this single dma_fence >>>> then shows up in the dma_fence_chain representing the whole >>>> migration? >>> What we'd like to happen during eviction is that we >>> >>> 1) await any exclusive- or moving fences, then schedule the migration >>> blit. The blit manages its own GPU ptes. Results in a single fence. >>> 2) Schedule unbind of any gpu vmas, resulting possibly in multiple >>> fences. >>> 3) Most but not all of the remaining resv shared fences will have been >>> finished in 2) We can't easily tell which so we have a couple of shared >>> fences left. >> >> Stop, wait a second here. We are going a bit in circles. >> >> Before you migrate a buffer, you *MUST* wait for all shared fences to >> complete. This is documented mandatory DMA-buf behavior. >> >> Daniel and I have discussed that quite extensively in the last few >> month. >> >> So how does it come that you do the blit before all shared fences are >> completed? > > Well we don't currently but wanted to... (I haven't consulted Daniel > in the matter, tbh). > > I was under the impression that all writes would add an exclusive > fence to the dma_resv. Yes that's correct. I'm working on to have more than one write fence, but that is currently under review. > If that's not the case or this is otherwise against the mandatory > DMA-buf bevhavior, we can certainly keep that part as is and that > would eliminate 3). Ah, now that somewhat starts to make sense. So your blit only waits for the writes to finish before starting the blit. Yes that's legal as long as you don't change the original content with the blit. But don't you then need to wait for both reads and writes before you unmap the VMAs? Anyway the good news is your problem totally goes away with the DMA-resv rework I've already send out. Basically it is now possible to have more than one fence in the DMA-resv object for migrations and all existing fences are kept around until they are finished. Regards, Christian. > > /Thomas >

3 years, 9 months

Re: [Linaro-mm-sig] [RFC PATCH 1/2] dma-fence: Avoid establishing a locking order between fence classes

by Christian König

[Adding Daniel here as well] Am 03.12.21 um 15:18 schrieb Thomas Hellström: > [SNIP] >> Well that's ok as well. My question is why does this single dma_fence >> then shows up in the dma_fence_chain representing the whole >> migration? > What we'd like to happen during eviction is that we > > 1) await any exclusive- or moving fences, then schedule the migration > blit. The blit manages its own GPU ptes. Results in a single fence. > 2) Schedule unbind of any gpu vmas, resulting possibly in multiple > fences. > 3) Most but not all of the remaining resv shared fences will have been > finished in 2) We can't easily tell which so we have a couple of shared > fences left. Stop, wait a second here. We are going a bit in circles. Before you migrate a buffer, you *MUST* wait for all shared fences to complete. This is documented mandatory DMA-buf behavior. Daniel and I have discussed that quite extensively in the last few month. So how does it come that you do the blit before all shared fences are completed? > 4) Add all fences resulting from 1) 2) and 3) into the per-memory-type > dma-fence-chain. > 5) hand the resulting dma-fence-chain representing the end of migration > over to ttm's resource manager. > > Now this means we have a dma-fence-chain disguised as a dma-fence out > in the wild, and it could in theory reappear as a 3) fence for another > migration unless a very careful audit is done, or as an input to the > dma-fence-array used for that single dependency. > >> That somehow doesn't seem to make sense because each individual step >> of >> the migration needs to wait for those dependencies as well even when >> it >> runs in parallel. >> >>> But that's not really the point, the point was that an (at least to >>> me) seemingly harmless usage pattern, be it real or fictious, ends >>> up >>> giving you severe internal- or cross-driver headaches. >> Yeah, we probably should document that better. But in general I don't >> see much reason to allow mixing containers. The dma_fence_array and >> dma_fence_chain objects have some distinct use cases and and using >> them >> to build up larger dependency structures sounds really questionable. > Yes, I tend to agree to some extent here. Perhaps add warnings when > adding a chain or array as an input to array and when accidently > joining chains, and provide helpers for flattening if needed. Yeah, that's probably a really good idea. Going to put it on my todo list. Thanks, Christian. > > /Thomas > > >> Christian. >> >>> /Thomas >>> >>> >>>> Regards, >>>> Christian. >>>> >>>> >

3 years, 9 months

Re: [Linaro-mm-sig] [RFC PATCH 1/2] dma-fence: Avoid establishing a locking order between fence classes

by Christian König

Am 01.12.21 um 13:16 schrieb Thomas Hellström (Intel): > > On 12/1/21 12:25, Christian König wrote: >> Am 01.12.21 um 12:04 schrieb Thomas Hellström (Intel): >>> >>> On 12/1/21 11:32, Christian König wrote: >>>> Am 01.12.21 um 11:15 schrieb Thomas Hellström (Intel): >>>>> [SNIP] >>>>>> >>>>>> What we could do is to avoid all this by not calling the callback >>>>>> with the lock held in the first place. >>>>> >>>>> If that's possible that might be a good idea, pls also see below. >>>> >>>> The problem with that is >>>> dma_fence_signal_locked()/dma_fence_signal_timestamp_locked(). If >>>> we could avoid using that or at least allow it to drop the lock >>>> then we could call the callback without holding it. >>>> >>>> Somebody would need to audit the drivers and see if holding the >>>> lock is really necessary anywhere. >>>> >>>>>> >>>>>>>> >>>>>>>>>> >>>>>>>>>> /Thomas >>>>>>>>> >>>>>>>>> Oh, and a follow up question: >>>>>>>>> >>>>>>>>> If there was a way to break the recursion on final put() >>>>>>>>> (using the same basic approach as patch 2 in this series uses >>>>>>>>> to break recursion in enable_signaling()), so that none of >>>>>>>>> these containers did require any special treatment, would it >>>>>>>>> be worth pursuing? I guess it might be possible by having the >>>>>>>>> callbacks drop the references rather than the loop in the >>>>>>>>> final put. + a couple of changes in code iterating over the >>>>>>>>> fence pointers. >>>>>>>> >>>>>>>> That won't really help, you just move the recursion from the >>>>>>>> final put into the callback. >>>>>>> >>>>>>> How do we recurse from the callback? The introduced fence_put() >>>>>>> of individual fence pointers >>>>>>> doesn't recurse anymore (at most 1 level), and any callback >>>>>>> recursion is broken by the irq_work? >>>>>> >>>>>> Yeah, but then you would need to take another lock to avoid >>>>>> racing with dma_fence_array_signaled(). >>>>>> >>>>>>> >>>>>>> I figure the big amount of work would be to adjust code that >>>>>>> iterates over the individual fence pointers to recognize that >>>>>>> they are rcu protected. >>>>>> >>>>>> Could be that we could solve this with RCU, but that sounds like >>>>>> a lot of churn for no gain at all. >>>>>> >>>>>> In other words even with the problems solved I think it would be >>>>>> a really bad idea to allow chaining of dma_fence_array objects. >>>>> >>>>> Yes, that was really the question, Is it worth pursuing this? I'm >>>>> not really suggesting we should allow this as an intentional >>>>> feature. I'm worried, however, that if we allow these containers >>>>> to start floating around cross-driver (or even internally) >>>>> disguised as ordinary dma_fences, they would require a lot of >>>>> driver special casing, or else completely unexpeced WARN_ON()s and >>>>> lockdep splats would start to turn up, scaring people off from >>>>> using them. And that would be a breeding ground for hairy >>>>> driver-private constructs. >>>> >>>> Well the question is why we would want to do it? >>>> >>>> If it's to avoid inter driver lock dependencies by avoiding to call >>>> the callback with the spinlock held, then yes please. We had tons >>>> of problems with that, resulting in irq_work and work_item >>>> delegation all over the place. >>> >>> Yes, that sounds like something desirable, but in these containers, >>> what's causing the lock dependencies is the enable_signaling() >>> callback that is typically called locked. >>> >>> >>>> >>>> If it's to allow nesting of dma_fence_array instances, then it's >>>> most likely a really bad idea even if we fix all the locking order >>>> problems. >>> >>> Well I think my use-case where I hit a dead end may illustrate what >>> worries me here: >>> >>> 1) We use a dma-fence-array to coalesce all dependencies for ttm >>> object migration. >>> 2) We use a dma-fence-chain to order the resulting dm_fence into a >>> timeline because the TTM resource manager code requires that. >>> >>> Initially seemingly harmless to me. >>> >>> But after a sequence evict->alloc->clear, the dma-fence-chain feeds >>> into the dma-fence-array for the clearing operation. Code still >>> works fine, and no deep recursion, no warnings. But if I were to add >>> another driver to the system that instead feeds a dma-fence-array >>> into a dma-fence-chain, this would give me a lockdep splat. >>> >>> So then if somebody were to come up with the splendid idea of using >>> a dma-fence-chain to initially coalesce fences, I'd hit the same >>> problem or risk illegaly joining two dma-fence-chains together. >>> >>> To fix this, I would need to look at the incoming fences and iterate >>> over any dma-fence-array or dma-fence-chain that is fed into the >>> dma-fence-array to flatten out the input. In fact all >>> dma-fence-array users would need to do that, and even >>> dma-fence-chain users watching out for not joining chains together >>> or accidently add an array that perhaps came as a disguised >>> dma-fence from antother driver. >>> >>> So the purpose to me would be to allow these containers as input to >>> eachother without a lot of in-driver special-casing, be it by >>> breaking recursion on built-in flattening to avoid >>> >>> a) Hitting issues in the future or with existing interoperating >>> drivers. >>> b) Avoid driver-private containers that also might break the >>> interoperability. (For example the i915 currently driver-private >>> dma_fence_work avoid all these problems, but we're attempting to >>> address issues in common code rather than re-inventing stuff >>> internally). >> >> I don't think that a dma_fence_array or dma_fence_chain is the right >> thing to begin with in those use cases. >> >> When you want to coalesce the dependencies for a job you could either >> use an xarray like Daniel did for the scheduler or some hashtable >> like we use in amdgpu. But I don't see the need for exposing the >> dma_fence interface for those. > > This is because the interface to our migration code takes just a > single dma-fence as dependency. Now this is of course something we > need to look at to mitigate this, but see below. Yeah, that's actually fine. >> >> And why do you use dma_fence_chain to generate a timeline for TTM? >> That should come naturally because all the moves must be ordered. > > Oh, in this case because we're looking at adding stuff at the end of > migration (like coalescing object shared fences and / or async unbind > fences), which may not complete in order. Well that's ok as well. My question is why does this single dma_fence then shows up in the dma_fence_chain representing the whole migration? That somehow doesn't seem to make sense because each individual step of the migration needs to wait for those dependencies as well even when it runs in parallel. > But that's not really the point, the point was that an (at least to > me) seemingly harmless usage pattern, be it real or fictious, ends up > giving you severe internal- or cross-driver headaches. Yeah, we probably should document that better. But in general I don't see much reason to allow mixing containers. The dma_fence_array and dma_fence_chain objects have some distinct use cases and and using them to build up larger dependency structures sounds really questionable. Christian. > > /Thomas > > >> >> Regards, >> Christian. >> >>

3 years, 9 months

[PATCH] drm/amdkfd: Use max() instead of doing it manually

by Jiapeng Chong

Fix following coccicheck warning: ./drivers/gpu/drm/amd/amdkfd/kfd_svm.c:2193:16-17: WARNING opportunity for max(). Reported-by: Abaci Robot <abaci(a)linux.alibaba.com> Signed-off-by: Jiapeng Chong <jiapeng.chong(a)linux.alibaba.com> --- drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index f2db49c..4f7e7b1 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -2190,7 +2190,7 @@ void schedule_deferred_list_work(struct svm_range_list *svms) start = mni->interval_tree.start; last = mni->interval_tree.last; - start = (start > range->start ? start : range->start) >> PAGE_SHIFT; + start = max(start, range->start) >> PAGE_SHIFT; last = (last < (range->end - 1) ? last : range->end - 1) >> PAGE_SHIFT; pr_debug("[0x%lx 0x%lx] range[0x%lx 0x%lx] notifier[0x%lx 0x%lx] %d\n", start, last, range->start >> PAGE_SHIFT, -- 1.8.3.1

3 years, 9 months

Re: [Linaro-mm-sig] [syzbot] WARNING in __dma_map_sg_attrs

by Christoph Hellwig

This means the virtgpu driver uses dma mapping helpers but has not set up a DMA mask (which most likely suggests it is some kind of virtual device). On Wed, Dec 01, 2021 at 10:18:21AM -0800, syzbot wrote: > Hello, > > syzbot found the following issue on: > > HEAD commit: c5c17547b778 Merge tag 'net-5.16-rc3' of git://git.kernel... > git tree: upstream > console output: https://syzkaller.appspot.com/x/log.txt?x=13a73609b00000 > kernel config: https://syzkaller.appspot.com/x/.config?x=bf85c53718a1e697 > dashboard link: https://syzkaller.appspot.com/bug?extid=10e27961f4da37c443b2 > compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2 > > Unfortunately, I don't have any reproducer for this issue yet. > > IMPORTANT: if you fix the issue, please add the following tag to the commit: > Reported-by: syzbot+10e27961f4da37c443b2(a)syzkaller.appspotmail.com > > ------------[ cut here ]------------ > WARNING: CPU: 2 PID: 17169 at kernel/dma/mapping.c:188 __dma_map_sg_attrs+0x181/0x1f0 kernel/dma/mapping.c:188 > Modules linked in: > CPU: 0 PID: 17169 Comm: syz-executor.3 Not tainted 5.16.0-rc2-syzkaller #0 > Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014 > RIP: 0010:__dma_map_sg_attrs+0x181/0x1f0 kernel/dma/mapping.c:188 > Code: 00 00 00 00 00 fc ff df 48 c1 e8 03 80 3c 10 00 75 71 4c 8b 3d 70 6d b1 0d e9 db fe ff ff e8 86 ff 12 00 0f 0b e8 7f ff 12 00 <0f> 0b 45 31 e4 e9 54 ff ff ff e8 70 ff 12 00 49 8d 7f 50 48 b8 00 > RSP: 0018:ffffc90002c0fb20 EFLAGS: 00010216 > RAX: 0000000000013018 RBX: 0000000000000020 RCX: ffffc900037d4000 > RDX: 0000000000040000 RSI: ffffffff8163d361 RDI: ffff8880182ae4d0 > RBP: ffff8880182ae088 R08: 0000000000000002 R09: ffff888017ba054f > R10: ffffffff8163d242 R11: 000000000008808a R12: 0000000000000000 > R13: ffff888024ca5700 R14: 0000000000000001 R15: 0000000000000000 > FS: 00007fa269e34700(0000) GS:ffff88802cb00000(0000) knlGS:0000000000000000 > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 > CR2: 000000000040c120 CR3: 000000006c77c000 CR4: 0000000000150ee0 > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 > Call Trace: > <TASK> > dma_map_sgtable+0x70/0xf0 kernel/dma/mapping.c:264 > drm_gem_map_dma_buf+0x12a/0x1e0 drivers/gpu/drm/drm_prime.c:633 > __map_dma_buf drivers/dma-buf/dma-buf.c:675 [inline] > dma_buf_map_attachment+0x39a/0x5b0 drivers/dma-buf/dma-buf.c:954 > drm_gem_prime_import_dev.part.0+0x85/0x220 drivers/gpu/drm/drm_prime.c:939 > drm_gem_prime_import_dev drivers/gpu/drm/drm_prime.c:982 [inline] > drm_gem_prime_import+0xc8/0x200 drivers/gpu/drm/drm_prime.c:982 > virtgpu_gem_prime_import+0x49/0x150 drivers/gpu/drm/virtio/virtgpu_prime.c:166 > drm_gem_prime_fd_to_handle+0x21d/0x550 drivers/gpu/drm/drm_prime.c:318 > drm_prime_fd_to_handle_ioctl+0x9b/0xd0 drivers/gpu/drm/drm_prime.c:374 > drm_ioctl_kernel+0x27d/0x4e0 drivers/gpu/drm/drm_ioctl.c:782 > drm_ioctl+0x51e/0x9d0 drivers/gpu/drm/drm_ioctl.c:885 > vfs_ioctl fs/ioctl.c:51 [inline] > __do_sys_ioctl fs/ioctl.c:874 [inline] > __se_sys_ioctl fs/ioctl.c:860 [inline] > __x64_sys_ioctl+0x193/0x200 fs/ioctl.c:860 > do_syscall_x64 arch/x86/entry/common.c:50 [inline] > do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 > entry_SYSCALL_64_after_hwframe+0x44/0xae > RIP: 0033:0x7fa26c8beae9 > Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 bc ff ff ff f7 d8 64 89 01 48 > RSP: 002b:00007fa269e34188 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 > RAX: ffffffffffffffda RBX: 00007fa26c9d1f60 RCX: 00007fa26c8beae9 > RDX: 00000000200004c0 RSI: 00000000c00c642e RDI: 0000000000000005 > RBP: 00007fa26c918f6d R08: 0000000000000000 R09: 0000000000000000 > R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 > R13: 00007ffc0019c51f R14: 00007fa269e34300 R15: 0000000000022000 > </TASK> > > > --- > This report is generated by a bot. It may contain errors. > See https://goo.gl/tpsmEJ for more information about syzbot. > syzbot engineers can be reached at syzkaller(a)googlegroups.com. > > syzbot will keep track of this issue. See: > https://goo.gl/tpsmEJ#status for how to communicate with syzbot. ---end quoted text---

3 years, 9 months

Re: [Linaro-mm-sig] [RFC PATCH 1/2] dma-fence: Avoid establishing a locking order between fence classes

by Christian König

Am 01.12.21 um 12:04 schrieb Thomas Hellström (Intel): > > On 12/1/21 11:32, Christian König wrote: >> Am 01.12.21 um 11:15 schrieb Thomas Hellström (Intel): >>> [SNIP] >>>> >>>> What we could do is to avoid all this by not calling the callback >>>> with the lock held in the first place. >>> >>> If that's possible that might be a good idea, pls also see below. >> >> The problem with that is >> dma_fence_signal_locked()/dma_fence_signal_timestamp_locked(). If we >> could avoid using that or at least allow it to drop the lock then we >> could call the callback without holding it. >> >> Somebody would need to audit the drivers and see if holding the lock >> is really necessary anywhere. >> >>>> >>>>>> >>>>>>>> >>>>>>>> /Thomas >>>>>>> >>>>>>> Oh, and a follow up question: >>>>>>> >>>>>>> If there was a way to break the recursion on final put() (using >>>>>>> the same basic approach as patch 2 in this series uses to break >>>>>>> recursion in enable_signaling()), so that none of these >>>>>>> containers did require any special treatment, would it be worth >>>>>>> pursuing? I guess it might be possible by having the callbacks >>>>>>> drop the references rather than the loop in the final put. + a >>>>>>> couple of changes in code iterating over the fence pointers. >>>>>> >>>>>> That won't really help, you just move the recursion from the >>>>>> final put into the callback. >>>>> >>>>> How do we recurse from the callback? The introduced fence_put() of >>>>> individual fence pointers >>>>> doesn't recurse anymore (at most 1 level), and any callback >>>>> recursion is broken by the irq_work? >>>> >>>> Yeah, but then you would need to take another lock to avoid racing >>>> with dma_fence_array_signaled(). >>>> >>>>> >>>>> I figure the big amount of work would be to adjust code that >>>>> iterates over the individual fence pointers to recognize that they >>>>> are rcu protected. >>>> >>>> Could be that we could solve this with RCU, but that sounds like a >>>> lot of churn for no gain at all. >>>> >>>> In other words even with the problems solved I think it would be a >>>> really bad idea to allow chaining of dma_fence_array objects. >>> >>> Yes, that was really the question, Is it worth pursuing this? I'm >>> not really suggesting we should allow this as an intentional >>> feature. I'm worried, however, that if we allow these containers to >>> start floating around cross-driver (or even internally) disguised as >>> ordinary dma_fences, they would require a lot of driver special >>> casing, or else completely unexpeced WARN_ON()s and lockdep splats >>> would start to turn up, scaring people off from using them. And that >>> would be a breeding ground for hairy driver-private constructs. >> >> Well the question is why we would want to do it? >> >> If it's to avoid inter driver lock dependencies by avoiding to call >> the callback with the spinlock held, then yes please. We had tons of >> problems with that, resulting in irq_work and work_item delegation >> all over the place. > > Yes, that sounds like something desirable, but in these containers, > what's causing the lock dependencies is the enable_signaling() > callback that is typically called locked. > > >> >> If it's to allow nesting of dma_fence_array instances, then it's most >> likely a really bad idea even if we fix all the locking order problems. > > Well I think my use-case where I hit a dead end may illustrate what > worries me here: > > 1) We use a dma-fence-array to coalesce all dependencies for ttm > object migration. > 2) We use a dma-fence-chain to order the resulting dm_fence into a > timeline because the TTM resource manager code requires that. > > Initially seemingly harmless to me. > > But after a sequence evict->alloc->clear, the dma-fence-chain feeds > into the dma-fence-array for the clearing operation. Code still works > fine, and no deep recursion, no warnings. But if I were to add another > driver to the system that instead feeds a dma-fence-array into a > dma-fence-chain, this would give me a lockdep splat. > > So then if somebody were to come up with the splendid idea of using a > dma-fence-chain to initially coalesce fences, I'd hit the same problem > or risk illegaly joining two dma-fence-chains together. > > To fix this, I would need to look at the incoming fences and iterate > over any dma-fence-array or dma-fence-chain that is fed into the > dma-fence-array to flatten out the input. In fact all dma-fence-array > users would need to do that, and even dma-fence-chain users watching > out for not joining chains together or accidently add an array that > perhaps came as a disguised dma-fence from antother driver. > > So the purpose to me would be to allow these containers as input to > eachother without a lot of in-driver special-casing, be it by breaking > recursion on built-in flattening to avoid > > a) Hitting issues in the future or with existing interoperating drivers. > b) Avoid driver-private containers that also might break the > interoperability. (For example the i915 currently driver-private > dma_fence_work avoid all these problems, but we're attempting to > address issues in common code rather than re-inventing stuff internally). I don't think that a dma_fence_array or dma_fence_chain is the right thing to begin with in those use cases. When you want to coalesce the dependencies for a job you could either use an xarray like Daniel did for the scheduler or some hashtable like we use in amdgpu. But I don't see the need for exposing the dma_fence interface for those. And why do you use dma_fence_chain to generate a timeline for TTM? That should come naturally because all the moves must be ordered. Regards, Christian.

3 years, 9 months

Re: [Linaro-mm-sig] [RFC PATCH 1/2] dma-fence: Avoid establishing a locking order between fence classes

by Christian König

Am 01.12.21 um 11:15 schrieb Thomas Hellström (Intel): > [SNIP] >> >> What we could do is to avoid all this by not calling the callback >> with the lock held in the first place. > > If that's possible that might be a good idea, pls also see below. The problem with that is dma_fence_signal_locked()/dma_fence_signal_timestamp_locked(). If we could avoid using that or at least allow it to drop the lock then we could call the callback without holding it. Somebody would need to audit the drivers and see if holding the lock is really necessary anywhere. >> >>>> >>>>>> >>>>>> /Thomas >>>>> >>>>> Oh, and a follow up question: >>>>> >>>>> If there was a way to break the recursion on final put() (using >>>>> the same basic approach as patch 2 in this series uses to break >>>>> recursion in enable_signaling()), so that none of these containers >>>>> did require any special treatment, would it be worth pursuing? I >>>>> guess it might be possible by having the callbacks drop the >>>>> references rather than the loop in the final put. + a couple of >>>>> changes in code iterating over the fence pointers. >>>> >>>> That won't really help, you just move the recursion from the final >>>> put into the callback. >>> >>> How do we recurse from the callback? The introduced fence_put() of >>> individual fence pointers >>> doesn't recurse anymore (at most 1 level), and any callback >>> recursion is broken by the irq_work? >> >> Yeah, but then you would need to take another lock to avoid racing >> with dma_fence_array_signaled(). >> >>> >>> I figure the big amount of work would be to adjust code that >>> iterates over the individual fence pointers to recognize that they >>> are rcu protected. >> >> Could be that we could solve this with RCU, but that sounds like a >> lot of churn for no gain at all. >> >> In other words even with the problems solved I think it would be a >> really bad idea to allow chaining of dma_fence_array objects. > > Yes, that was really the question, Is it worth pursuing this? I'm not > really suggesting we should allow this as an intentional feature. I'm > worried, however, that if we allow these containers to start floating > around cross-driver (or even internally) disguised as ordinary > dma_fences, they would require a lot of driver special casing, or else > completely unexpeced WARN_ON()s and lockdep splats would start to turn > up, scaring people off from using them. And that would be a breeding > ground for hairy driver-private constructs. Well the question is why we would want to do it? If it's to avoid inter driver lock dependencies by avoiding to call the callback with the spinlock held, then yes please. We had tons of problems with that, resulting in irq_work and work_item delegation all over the place. If it's to allow nesting of dma_fence_array instances, then it's most likely a really bad idea even if we fix all the locking order problems. Christian. > > /Thomas > > >> >> Christian. >> >>> >>> >>> Thanks, >>> >>> /Thomas >>> >>>

3 years, 9 months

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

Linaro-mm-sig