| Message ID | 20190920083624.4601-1-chris@chris-wilson.co.uk (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | drm/i915: Prevent bonded requests from overtaking each other on preemption | expand |
On 20/09/2019 09:36, Chris Wilson wrote: > Force bonded requests to run on distinct engines so that they cannot be > shuffled onto the same engine where timeslicing will reverse the order. > A bonded request will often wait on a semaphore signaled by its master, > creating an implicit dependency -- if we ignore that implicit dependency > and allow the bonded request to run on the same engine and before its > master, we will cause a GPU hang. > > We can prevent this inversion by restricting which engines we allow > ourselves to jump to upon preemption, i.e. baking in the arrangement > established at first execution. (We should also consider capturing the > implicit dependency using i915_sched_add_dependency(), but first we need > to think about the constraints that requires on the execution/retirement > ordering.) > > Fixes: 8ee36e048c98 ("drm/i915/execlists: Minimalistic timeslicing") > References: ee1136908e9b ("drm/i915/execlists: Virtual engine bonding") > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> > Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> > --- > drivers/gpu/drm/i915/gt/intel_lrc.c | 19 +++++++++++-------- > 1 file changed, 11 insertions(+), 8 deletions(-) > > diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c > index a99166a2d2eb..7920649e4d87 100644 > --- a/drivers/gpu/drm/i915/gt/intel_lrc.c > +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c > @@ -3755,18 +3755,21 @@ static void > virtual_bond_execute(struct i915_request *rq, struct dma_fence *signal) > { > struct virtual_engine *ve = to_virtual_engine(rq->engine); > + intel_engine_mask_t allowed, exec; > struct ve_bond *bond; > > bond = virtual_find_bond(ve, to_request(signal)->engine); > - if (bond) { > - intel_engine_mask_t old, new, cmp; > + if (!bond) > + return; > > - cmp = READ_ONCE(rq->execution_mask); > - do { > - old = cmp; > - new = cmp & bond->sibling_mask; > - } while ((cmp = cmpxchg(&rq->execution_mask, old, new)) != old); > - } > + /* Restrict the bonded request to run on only the slaved engines */ > + allowed = bond->sibling_mask & ~to_request(signal)->engine->mask; Hmm.. isn't it a miss on the uapi level that we allow master to be mentioned in the list of bonds? That's the only scenario where this line does something I think. So should we just forbid this setup on the uapi level? > + exec = READ_ONCE(rq->execution_mask); > + while (!try_cmpxchg(&rq->execution_mask, &exec, exec & allowed)) > + ; > + > + /* Prevent the master from being re-run on the slaved engines */ > + to_request(signal)->execution_mask &= ~allowed; This sounds unfortunate for future scheduling. There shouldn't be a fundamental reason why next execution for the master couldn't be on an engine which can also be a slave. So if we have: master .veng=vcs0,vcs1 slave .veng=vcs0,vcs1 .bond(master=vcs0, mask=vcs1) .bond(master=vcs1, mask=vcs0) This should be allowed setup but with this change it would fix the master to only be one of the options. Is the real problem that after preemption for timeslicing and subsequent re-submit we miss some hooks to re-evaluate the bonded relationship? I guess looking would be hard to do any peeking from one submission tasklet to another (different engines) to check if one of the pair is already executing again and so to pick the other end correctly? I think in practical terms for media this work since they are not setting it up like my sketch shows. So it could be just fine in practice for current users. Regards, Tvrtko > } > > struct intel_context * >
Quoting Tvrtko Ursulin (2019-09-20 13:24:47) > > On 20/09/2019 09:36, Chris Wilson wrote: > > Force bonded requests to run on distinct engines so that they cannot be > > shuffled onto the same engine where timeslicing will reverse the order. > > A bonded request will often wait on a semaphore signaled by its master, > > creating an implicit dependency -- if we ignore that implicit dependency > > and allow the bonded request to run on the same engine and before its > > master, we will cause a GPU hang. > > > > We can prevent this inversion by restricting which engines we allow > > ourselves to jump to upon preemption, i.e. baking in the arrangement > > established at first execution. (We should also consider capturing the > > implicit dependency using i915_sched_add_dependency(), but first we need > > to think about the constraints that requires on the execution/retirement > > ordering.) > > > > Fixes: 8ee36e048c98 ("drm/i915/execlists: Minimalistic timeslicing") > > References: ee1136908e9b ("drm/i915/execlists: Virtual engine bonding") > > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > > Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> > > Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> > > --- > > drivers/gpu/drm/i915/gt/intel_lrc.c | 19 +++++++++++-------- > > 1 file changed, 11 insertions(+), 8 deletions(-) > > > > diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c > > index a99166a2d2eb..7920649e4d87 100644 > > --- a/drivers/gpu/drm/i915/gt/intel_lrc.c > > +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c > > @@ -3755,18 +3755,21 @@ static void > > virtual_bond_execute(struct i915_request *rq, struct dma_fence *signal) > > { > > struct virtual_engine *ve = to_virtual_engine(rq->engine); > > + intel_engine_mask_t allowed, exec; > > struct ve_bond *bond; > > > > bond = virtual_find_bond(ve, to_request(signal)->engine); > > - if (bond) { > > - intel_engine_mask_t old, new, cmp; > > + if (!bond) > > + return; > > > > - cmp = READ_ONCE(rq->execution_mask); > > - do { > > - old = cmp; > > - new = cmp & bond->sibling_mask; > > - } while ((cmp = cmpxchg(&rq->execution_mask, old, new)) != old); > > - } > > + /* Restrict the bonded request to run on only the slaved engines */ > > + allowed = bond->sibling_mask & ~to_request(signal)->engine->mask; > > Hmm.. isn't it a miss on the uapi level that we allow master to be > mentioned in the list of bonds? That's the only scenario where this line > does something I think. So should we just forbid this setup on the uapi > level? That's just a lot of digging! > > + exec = READ_ONCE(rq->execution_mask); > > + while (!try_cmpxchg(&rq->execution_mask, &exec, exec & allowed)) > > + ; > > + > > + /* Prevent the master from being re-run on the slaved engines */ > > + to_request(signal)->execution_mask &= ~allowed; > > This sounds unfortunate for future scheduling. There shouldn't be a > fundamental reason why next execution for the master couldn't be on an > engine which can also be a slave. So if we have: Note though that we do not reset the execution_mask at any point :) That's actually harder to do than it sounds, as after the bonded execution, they are no longer linked. :| > master > .veng=vcs0,vcs1 > slave > .veng=vcs0,vcs1 > .bond(master=vcs0, mask=vcs1) > .bond(master=vcs1, mask=vcs0) > > This should be allowed setup but with this change it would fix the > master to only be one of the options. It would fix it to the first one it selected and executed on. It can still pick either vcs0 or vcs1 and the slave would then be on vcs1 or vcs0 respectively. > Is the real problem that after preemption for timeslicing and subsequent > re-submit we miss some hooks to re-evaluate the bonded relationship? That doesn't exist, yes. But it's more than that, as we don't have the notion of global preemption -- we don't evaluate between engines whether or not there are cross dependencies. > I guess looking would be hard to do any peeking from one submission > tasklet to another (different engines) to check if one of the pair is > already executing again and so to pick the other end correctly? Hard indeed. I would throw a flag onto the request that says if you preempt me, stop the world (intel_engine_mask_t perhaps). Even that requires some tricks we don't yet have. But we can't touch the other engines within the tasklet unless we can come up with a lockless strategy (hence the strategy of punting to a supreme thread with oversight of all engines, gah.) > I think in practical terms for media this work since they are not > setting it up like my sketch shows. So it could be just fine in practice > for current users. I think your example works better than you think -- we just end up concreted into our first choice and can't jump around hogs in the system. (For example, to prove the above, we can launch two such tasks, with a spinner as both masters and the system should get stuck on both spinners.) However, I'm currently dreading writing the test case for this. -Chris
On 20/09/2019 13:42, Chris Wilson wrote: > Quoting Tvrtko Ursulin (2019-09-20 13:24:47) >> >> On 20/09/2019 09:36, Chris Wilson wrote: >>> Force bonded requests to run on distinct engines so that they cannot be >>> shuffled onto the same engine where timeslicing will reverse the order. >>> A bonded request will often wait on a semaphore signaled by its master, >>> creating an implicit dependency -- if we ignore that implicit dependency >>> and allow the bonded request to run on the same engine and before its >>> master, we will cause a GPU hang. >>> >>> We can prevent this inversion by restricting which engines we allow >>> ourselves to jump to upon preemption, i.e. baking in the arrangement >>> established at first execution. (We should also consider capturing the >>> implicit dependency using i915_sched_add_dependency(), but first we need >>> to think about the constraints that requires on the execution/retirement >>> ordering.) >>> >>> Fixes: 8ee36e048c98 ("drm/i915/execlists: Minimalistic timeslicing") >>> References: ee1136908e9b ("drm/i915/execlists: Virtual engine bonding") >>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> >>> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> >>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> >>> --- >>> drivers/gpu/drm/i915/gt/intel_lrc.c | 19 +++++++++++-------- >>> 1 file changed, 11 insertions(+), 8 deletions(-) >>> >>> diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c >>> index a99166a2d2eb..7920649e4d87 100644 >>> --- a/drivers/gpu/drm/i915/gt/intel_lrc.c >>> +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c >>> @@ -3755,18 +3755,21 @@ static void >>> virtual_bond_execute(struct i915_request *rq, struct dma_fence *signal) >>> { >>> struct virtual_engine *ve = to_virtual_engine(rq->engine); >>> + intel_engine_mask_t allowed, exec; >>> struct ve_bond *bond; >>> >>> bond = virtual_find_bond(ve, to_request(signal)->engine); >>> - if (bond) { >>> - intel_engine_mask_t old, new, cmp; >>> + if (!bond) >>> + return; >>> >>> - cmp = READ_ONCE(rq->execution_mask); >>> - do { >>> - old = cmp; >>> - new = cmp & bond->sibling_mask; >>> - } while ((cmp = cmpxchg(&rq->execution_mask, old, new)) != old); >>> - } >>> + /* Restrict the bonded request to run on only the slaved engines */ >>> + allowed = bond->sibling_mask & ~to_request(signal)->engine->mask; >> >> Hmm.. isn't it a miss on the uapi level that we allow master to be >> mentioned in the list of bonds? That's the only scenario where this line >> does something I think. So should we just forbid this setup on the uapi >> level? > > That's just a lot of digging! It's simple, in set_engines__bond in the bonds loop we just do "if (master == bond) oh_no_you_wont". Am I missing something? >>> + exec = READ_ONCE(rq->execution_mask); >>> + while (!try_cmpxchg(&rq->execution_mask, &exec, exec & allowed)) >>> + ; >>> + >>> + /* Prevent the master from being re-run on the slaved engines */ >>> + to_request(signal)->execution_mask &= ~allowed; >> >> This sounds unfortunate for future scheduling. There shouldn't be a >> fundamental reason why next execution for the master couldn't be on an >> engine which can also be a slave. So if we have: > > Note though that we do not reset the execution_mask at any point :) > That's actually harder to do than it sounds, as after the bonded > execution, they are no longer linked. :| > >> master >> .veng=vcs0,vcs1 >> slave >> .veng=vcs0,vcs1 >> .bond(master=vcs0, mask=vcs1) >> .bond(master=vcs1, mask=vcs0) >> >> This should be allowed setup but with this change it would fix the >> master to only be one of the options. > > It would fix it to the first one it selected and executed on. It can > still pick either vcs0 or vcs1 and the slave would then be on vcs1 or > vcs0 respectively. > >> Is the real problem that after preemption for timeslicing and subsequent >> re-submit we miss some hooks to re-evaluate the bonded relationship? > > That doesn't exist, yes. But it's more than that, as we don't have the > notion of global preemption -- we don't evaluate between engines whether > or not there are cross dependencies. > >> I guess looking would be hard to do any peeking from one submission >> tasklet to another (different engines) to check if one of the pair is >> already executing again and so to pick the other end correctly? > > Hard indeed. I would throw a flag onto the request that says if you > preempt me, stop the world (intel_engine_mask_t perhaps). Even that > requires some tricks we don't yet have. But we can't touch the other > engines within the tasklet unless we can come up with a lockless > strategy (hence the strategy of punting to a supreme thread with > oversight of all engines, gah.) > >> I think in practical terms for media this work since they are not >> setting it up like my sketch shows. So it could be just fine in practice >> for current users. > > I think your example works better than you think -- we just end up > concreted into our first choice and can't jump around hogs in the > system. (For example, to prove the above, we can launch two such tasks, > with a spinner as both masters and the system should get stuck on both > spinners.) > > However, I'm currently dreading writing the test case for this. No, you are right, as usual.. I was wrongly thinking the veng mask for the master context would be changed while it is only a single request. I have no complaints on that. I don't see that there is much point trying to improve on it right now. Even the issue of master in the bond list is passable since now we handle it gracefully. Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Regards, Tvrtko
Quoting Chris Wilson (2019-09-20 09:36:24) > Force bonded requests to run on distinct engines so that they cannot be > shuffled onto the same engine where timeslicing will reverse the order. > A bonded request will often wait on a semaphore signaled by its master, > creating an implicit dependency -- if we ignore that implicit dependency > and allow the bonded request to run on the same engine and before its > master, we will cause a GPU hang. Thinking more, it should not directly cause a GPU hang, as the stuck request should be timesliced away, and each preemption should be enough to keep hangcheck at bay (though we have evidence it may not). So at best it runs at half-speed, at worst a third (if my model is correct). -Chris
Quoting Tvrtko Ursulin (2019-09-20 15:51:35) > > On 20/09/2019 13:42, Chris Wilson wrote: > > Quoting Tvrtko Ursulin (2019-09-20 13:24:47) > >> > >> On 20/09/2019 09:36, Chris Wilson wrote: > >>> Force bonded requests to run on distinct engines so that they cannot be > >>> shuffled onto the same engine where timeslicing will reverse the order. > >>> A bonded request will often wait on a semaphore signaled by its master, > >>> creating an implicit dependency -- if we ignore that implicit dependency > >>> and allow the bonded request to run on the same engine and before its > >>> master, we will cause a GPU hang. > >>> > >>> We can prevent this inversion by restricting which engines we allow > >>> ourselves to jump to upon preemption, i.e. baking in the arrangement > >>> established at first execution. (We should also consider capturing the > >>> implicit dependency using i915_sched_add_dependency(), but first we need > >>> to think about the constraints that requires on the execution/retirement > >>> ordering.) > >>> > >>> Fixes: 8ee36e048c98 ("drm/i915/execlists: Minimalistic timeslicing") > >>> References: ee1136908e9b ("drm/i915/execlists: Virtual engine bonding") > >>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > >>> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> > >>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> > >>> --- > >>> drivers/gpu/drm/i915/gt/intel_lrc.c | 19 +++++++++++-------- > >>> 1 file changed, 11 insertions(+), 8 deletions(-) > >>> > >>> diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c > >>> index a99166a2d2eb..7920649e4d87 100644 > >>> --- a/drivers/gpu/drm/i915/gt/intel_lrc.c > >>> +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c > >>> @@ -3755,18 +3755,21 @@ static void > >>> virtual_bond_execute(struct i915_request *rq, struct dma_fence *signal) > >>> { > >>> struct virtual_engine *ve = to_virtual_engine(rq->engine); > >>> + intel_engine_mask_t allowed, exec; > >>> struct ve_bond *bond; > >>> > >>> bond = virtual_find_bond(ve, to_request(signal)->engine); > >>> - if (bond) { > >>> - intel_engine_mask_t old, new, cmp; > >>> + if (!bond) > >>> + return; > >>> > >>> - cmp = READ_ONCE(rq->execution_mask); > >>> - do { > >>> - old = cmp; > >>> - new = cmp & bond->sibling_mask; > >>> - } while ((cmp = cmpxchg(&rq->execution_mask, old, new)) != old); > >>> - } > >>> + /* Restrict the bonded request to run on only the slaved engines */ > >>> + allowed = bond->sibling_mask & ~to_request(signal)->engine->mask; > >> > >> Hmm.. isn't it a miss on the uapi level that we allow master to be > >> mentioned in the list of bonds? That's the only scenario where this line > >> does something I think. So should we just forbid this setup on the uapi > >> level? > > > > That's just a lot of digging! > > It's simple, in set_engines__bond in the bonds loop we just do "if > (master == bond) oh_no_you_wont". Am I missing something? There doesn't even need to be a bond, which is something I forgot to handle. So I'm looking at something more like static void virtual_bond_execute(struct i915_request *rq, struct dma_fence *signal) { struct virtual_engine *ve = to_virtual_engine(rq->engine); intel_engine_mask_t allowed, exec; struct ve_bond *bond; allowed = ~to_request(signal)->engine->mask; bond = virtual_find_bond(ve, to_request(signal)->engine); if (bond) allowed &= bond->sibling_mask; /* Restrict the bonded request to run on only the available engines */ exec = READ_ONCE(rq->execution_mask); while (!try_cmpxchg(&rq->execution_mask, &exec, exec & allowed)) ; /* Prevent the master from being re-run on the bonded engines */ to_request(signal)->execution_mask &= ~allowed; } (The joy of trying to write a test case.) -Chris
On 20/09/2019 15:57, Chris Wilson wrote: > Quoting Chris Wilson (2019-09-20 09:36:24) >> Force bonded requests to run on distinct engines so that they cannot be >> shuffled onto the same engine where timeslicing will reverse the order. >> A bonded request will often wait on a semaphore signaled by its master, >> creating an implicit dependency -- if we ignore that implicit dependency >> and allow the bonded request to run on the same engine and before its >> master, we will cause a GPU hang. > > Thinking more, it should not directly cause a GPU hang, as the stuck request > should be timesliced away, and each preemption should be enough to keep > hangcheck at bay (though we have evidence it may not). So at best it runs > at half-speed, at worst a third (if my model is correct). But I think it is still correct to do since we don't have the coupling information on re-submit. Hm.. but don't we need to prevent slave from changing engines as well? Regards, Tvrtko
> -----Original Message----- > From: Intel-gfx <intel-gfx-bounces@lists.freedesktop.org> On Behalf Of Tvrtko > Ursulin > Sent: Friday, September 20, 2019 8:12 AM > To: Chris Wilson <chris@chris-wilson.co.uk>; intel-gfx@lists.freedesktop.org > Subject: Re: [Intel-gfx] [PATCH] drm/i915: Prevent bonded requests from > overtaking each other on preemption > > > On 20/09/2019 15:57, Chris Wilson wrote: > > Quoting Chris Wilson (2019-09-20 09:36:24) > >> Force bonded requests to run on distinct engines so that they cannot be > >> shuffled onto the same engine where timeslicing will reverse the order. > >> A bonded request will often wait on a semaphore signaled by its master, > >> creating an implicit dependency -- if we ignore that implicit dependency > >> and allow the bonded request to run on the same engine and before its > >> master, we will cause a GPU hang. > > > > Thinking more, it should not directly cause a GPU hang, as the stuck request > > should be timesliced away, and each preemption should be enough to keep > > hangcheck at bay (though we have evidence it may not). So at best it runs > > at half-speed, at worst a third (if my model is correct). > > But I think it is still correct to do since we don't have the coupling > information on re-submit. Hm.. but don't we need to prevent slave from > changing engines as well? Unless I'm missing something, the proposal here is to set the engines in stone at first submission, and never change them? If so, that does sound overly restrictive, and will prevent any kind of rebalancing as workloads (of varying slave counts) come and go. During the original design it was called out that the workloads should be pre-empted atomically. That allows the entire bonding mask to be re-evaluated at every context switch and so we can then rebalance. Still not easy to achieve I agree :-( > > Regards, > > Tvrtko > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoting Tvrtko Ursulin (2019-09-20 16:12:23) > > On 20/09/2019 15:57, Chris Wilson wrote: > > Quoting Chris Wilson (2019-09-20 09:36:24) > >> Force bonded requests to run on distinct engines so that they cannot be > >> shuffled onto the same engine where timeslicing will reverse the order. > >> A bonded request will often wait on a semaphore signaled by its master, > >> creating an implicit dependency -- if we ignore that implicit dependency > >> and allow the bonded request to run on the same engine and before its > >> master, we will cause a GPU hang. > > > > Thinking more, it should not directly cause a GPU hang, as the stuck request > > should be timesliced away, and each preemption should be enough to keep > > hangcheck at bay (though we have evidence it may not). So at best it runs > > at half-speed, at worst a third (if my model is correct). > > But I think it is still correct to do since we don't have the coupling > information on re-submit. Hm.. but don't we need to prevent slave from > changing engines as well? Yes, it still looks like a sensible enough patch (even if I am biased because I think it is cute), especially in light of how we only run the bond_execute once and do not reconfigure the execution_mask on unsubmit. Still working on the test cases to exercise timeslicing + submit/bonding. -Chris
Quoting Bloomfield, Jon (2019-09-20 16:50:57) > > -----Original Message----- > > From: Intel-gfx <intel-gfx-bounces@lists.freedesktop.org> On Behalf Of Tvrtko > > Ursulin > > Sent: Friday, September 20, 2019 8:12 AM > > To: Chris Wilson <chris@chris-wilson.co.uk>; intel-gfx@lists.freedesktop.org > > Subject: Re: [Intel-gfx] [PATCH] drm/i915: Prevent bonded requests from > > overtaking each other on preemption > > > > > > On 20/09/2019 15:57, Chris Wilson wrote: > > > Quoting Chris Wilson (2019-09-20 09:36:24) > > >> Force bonded requests to run on distinct engines so that they cannot be > > >> shuffled onto the same engine where timeslicing will reverse the order. > > >> A bonded request will often wait on a semaphore signaled by its master, > > >> creating an implicit dependency -- if we ignore that implicit dependency > > >> and allow the bonded request to run on the same engine and before its > > >> master, we will cause a GPU hang. > > > > > > Thinking more, it should not directly cause a GPU hang, as the stuck request > > > should be timesliced away, and each preemption should be enough to keep > > > hangcheck at bay (though we have evidence it may not). So at best it runs > > > at half-speed, at worst a third (if my model is correct). > > > > But I think it is still correct to do since we don't have the coupling > > information on re-submit. Hm.. but don't we need to prevent slave from > > changing engines as well? > > Unless I'm missing something, the proposal here is to set the engines in stone at first submission, and never change them? For submission here, think execution (submission to actual HW). (We have 2 separate phases that all like to be called submit()!) > If so, that does sound overly restrictive, and will prevent any kind of rebalancing as workloads (of varying slave counts) come and go. We are only restricting this request, not the contexts. We still have balancing overall, just not instantaneous balancing if we timeslice out of this request -- we put it back onto the "same" engine and not another. Which is in some ways is less than ideal, although strictly we are only saying don't put it back onto an engine we have earmarked for our bonded request, and so we avoid contending with our parallel request reducing that to serial (and often bad) behaviour. [So at the end of this statement, I'm more happy with the restriction ;] > During the original design it was called out that the workloads should be pre-empted atomically. That allows the entire bonding mask to be re-evaluated at every context switch and so we can then rebalance. Still not easy to achieve I agree :-( The problem with that statement is that atomic implies a global scheduling decision. Blood, sweat and tears. Of course, with your endless scheme, scheduling is all in the purview of the user :) -Chris
Quoting Chris Wilson (2019-09-20 17:03:34) > Quoting Bloomfield, Jon (2019-09-20 16:50:57) > > > -----Original Message----- > > > From: Intel-gfx <intel-gfx-bounces@lists.freedesktop.org> On Behalf Of Tvrtko > > > Ursulin > > > Sent: Friday, September 20, 2019 8:12 AM > > > To: Chris Wilson <chris@chris-wilson.co.uk>; intel-gfx@lists.freedesktop.org > > > Subject: Re: [Intel-gfx] [PATCH] drm/i915: Prevent bonded requests from > > > overtaking each other on preemption > > > > > > > > > On 20/09/2019 15:57, Chris Wilson wrote: > > > > Quoting Chris Wilson (2019-09-20 09:36:24) > > > >> Force bonded requests to run on distinct engines so that they cannot be > > > >> shuffled onto the same engine where timeslicing will reverse the order. > > > >> A bonded request will often wait on a semaphore signaled by its master, > > > >> creating an implicit dependency -- if we ignore that implicit dependency > > > >> and allow the bonded request to run on the same engine and before its > > > >> master, we will cause a GPU hang. > > > > > > > > Thinking more, it should not directly cause a GPU hang, as the stuck request > > > > should be timesliced away, and each preemption should be enough to keep > > > > hangcheck at bay (though we have evidence it may not). So at best it runs > > > > at half-speed, at worst a third (if my model is correct). > > > > > > But I think it is still correct to do since we don't have the coupling > > > information on re-submit. Hm.. but don't we need to prevent slave from > > > changing engines as well? > > > > Unless I'm missing something, the proposal here is to set the engines in stone at first submission, and never change them? > > For submission here, think execution (submission to actual HW). (We have > 2 separate phases that all like to be called submit()!) s/2/3/ -Chris
> -----Original Message----- > From: Chris Wilson <chris@chris-wilson.co.uk> > Sent: Friday, September 20, 2019 9:04 AM > To: Bloomfield, Jon <jon.bloomfield@intel.com>; intel- > gfx@lists.freedesktop.org; Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> > Subject: RE: [Intel-gfx] [PATCH] drm/i915: Prevent bonded requests from > overtaking each other on preemption > > Quoting Bloomfield, Jon (2019-09-20 16:50:57) > > > -----Original Message----- > > > From: Intel-gfx <intel-gfx-bounces@lists.freedesktop.org> On Behalf Of > Tvrtko > > > Ursulin > > > Sent: Friday, September 20, 2019 8:12 AM > > > To: Chris Wilson <chris@chris-wilson.co.uk>; intel-gfx@lists.freedesktop.org > > > Subject: Re: [Intel-gfx] [PATCH] drm/i915: Prevent bonded requests from > > > overtaking each other on preemption > > > > > > > > > On 20/09/2019 15:57, Chris Wilson wrote: > > > > Quoting Chris Wilson (2019-09-20 09:36:24) > > > >> Force bonded requests to run on distinct engines so that they cannot be > > > >> shuffled onto the same engine where timeslicing will reverse the order. > > > >> A bonded request will often wait on a semaphore signaled by its master, > > > >> creating an implicit dependency -- if we ignore that implicit dependency > > > >> and allow the bonded request to run on the same engine and before its > > > >> master, we will cause a GPU hang. > > > > > > > > Thinking more, it should not directly cause a GPU hang, as the stuck > request > > > > should be timesliced away, and each preemption should be enough to > keep > > > > hangcheck at bay (though we have evidence it may not). So at best it runs > > > > at half-speed, at worst a third (if my model is correct). > > > > > > But I think it is still correct to do since we don't have the coupling > > > information on re-submit. Hm.. but don't we need to prevent slave from > > > changing engines as well? > > > > Unless I'm missing something, the proposal here is to set the engines in stone > at first submission, and never change them? > > For submission here, think execution (submission to actual HW). (We have > 2 separate phases that all like to be called submit()!) > > > If so, that does sound overly restrictive, and will prevent any kind of > rebalancing as workloads (of varying slave counts) come and go. > > We are only restricting this request, not the contexts. We still have > balancing overall, just not instantaneous balancing if we timeslice out > of this request -- we put it back onto the "same" engine and not another. > Which is in some ways is less than ideal, although strictly we are only > saying don't put it back onto an engine we have earmarked for our bonded > request, and so we avoid contending with our parallel request reducing > that to serial (and often bad) behaviour. > > [So at the end of this statement, I'm more happy with the restriction ;] > > > During the original design it was called out that the workloads should be pre- > empted atomically. That allows the entire bonding mask to be re-evaluated at > every context switch and so we can then rebalance. Still not easy to achieve I > agree :-( > > The problem with that statement is that atomic implies a global > scheduling decision. Blood, sweat and tears. Agreed - It isn't fun. Perhaps it doesn't matter anyway. Once GuC is offloading the scheduling it should be able to do a little more wrt rebalancing. Let's make it a GuC headache instead. > > Of course, with your endless scheme, scheduling is all in the purview of > the user :) Hey, don't tarnish me with that brush. I don't like it either. Actually, it's your scheme technically. I just asked for a way to enable HPC workloads, and you enthusiastically offered heartbeats&non-persistence. So shall history be written :-) > -Chris
diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c index a99166a2d2eb..7920649e4d87 100644 --- a/drivers/gpu/drm/i915/gt/intel_lrc.c +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c @@ -3755,18 +3755,21 @@ static void virtual_bond_execute(struct i915_request *rq, struct dma_fence *signal) { struct virtual_engine *ve = to_virtual_engine(rq->engine); + intel_engine_mask_t allowed, exec; struct ve_bond *bond; bond = virtual_find_bond(ve, to_request(signal)->engine); - if (bond) { - intel_engine_mask_t old, new, cmp; + if (!bond) + return; - cmp = READ_ONCE(rq->execution_mask); - do { - old = cmp; - new = cmp & bond->sibling_mask; - } while ((cmp = cmpxchg(&rq->execution_mask, old, new)) != old); - } + /* Restrict the bonded request to run on only the slaved engines */ + allowed = bond->sibling_mask & ~to_request(signal)->engine->mask; + exec = READ_ONCE(rq->execution_mask); + while (!try_cmpxchg(&rq->execution_mask, &exec, exec & allowed)) + ; + + /* Prevent the master from being re-run on the slaved engines */ + to_request(signal)->execution_mask &= ~allowed; } struct intel_context *
Force bonded requests to run on distinct engines so that they cannot be shuffled onto the same engine where timeslicing will reverse the order. A bonded request will often wait on a semaphore signaled by its master, creating an implicit dependency -- if we ignore that implicit dependency and allow the bonded request to run on the same engine and before its master, we will cause a GPU hang. We can prevent this inversion by restricting which engines we allow ourselves to jump to upon preemption, i.e. baking in the arrangement established at first execution. (We should also consider capturing the implicit dependency using i915_sched_add_dependency(), but first we need to think about the constraints that requires on the execution/retirement ordering.) Fixes: 8ee36e048c98 ("drm/i915/execlists: Minimalistic timeslicing") References: ee1136908e9b ("drm/i915/execlists: Virtual engine bonding") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> --- drivers/gpu/drm/i915/gt/intel_lrc.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-)