| Message ID | 20191123174946.891350-1-chris@chris-wilson.co.uk (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | drm/i915/gt: Mark the execlists->active as the primary volatile access | expand |
Quoting Chris Wilson (2019-11-23 17:49:46) > Since we want to do a lockless read of the current active request, and > that request is written to by process_csb also without serialisation, we > need to instruct gcc to take care in reading the pointer itself. > > Otherwise, we have observed execlists_active() to report 0x40. > > [ 2400.760381] igt/para-4098 1..s. 2376479300us : process_csb: rcs0 cs-irq head=3, tail=4 > [ 2400.760826] igt/para-4098 1..s. 2376479303us : process_csb: rcs0 csb[4]: status=0x00000001:0x00000000 > [ 2400.761271] igt/para-4098 1..s. 2376479306us : trace_ports: rcs0: promote { b9c59:2622, b9c55:2624 } > [ 2400.761726] igt/para-4097 0d... 2376479311us : __i915_schedule: rcs0: -2147483648->3, inflight:0000000000000040, rq:ffff888208c1e940 > > which is impossible! > Fixes: df403069029d ("drm/i915/execlists: Lift process_csb() out of the irq-off spinlock") -Chris
Hi Chris, On Sat, Nov 23, 2019 at 05:49:46PM +0000, Chris Wilson wrote: > Since we want to do a lockless read of the current active request, and > that request is written to by process_csb also without serialisation, we > need to instruct gcc to take care in reading the pointer itself. > > Otherwise, we have observed execlists_active() to report 0x40. > > [ 2400.760381] igt/para-4098 1..s. 2376479300us : process_csb: rcs0 cs-irq head=3, tail=4 > [ 2400.760826] igt/para-4098 1..s. 2376479303us : process_csb: rcs0 csb[4]: status=0x00000001:0x00000000 > [ 2400.761271] igt/para-4098 1..s. 2376479306us : trace_ports: rcs0: promote { b9c59:2622, b9c55:2624 } > [ 2400.761726] igt/para-4097 0d... 2376479311us : __i915_schedule: rcs0: -2147483648->3, inflight:0000000000000040, rq:ffff888208c1e940 > > which is impossible! > > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Tested-by: Andi Shyti <andi.shyti@intel.com> Reviewed-by: Andi Shyti <andi.shyti@intel.com> Andi
diff --git a/drivers/gpu/drm/i915/gt/intel_engine.h b/drivers/gpu/drm/i915/gt/intel_engine.h index bc3b72bfa9e3..01765a7ec18f 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine.h +++ b/drivers/gpu/drm/i915/gt/intel_engine.h @@ -100,9 +100,7 @@ execlists_num_ports(const struct intel_engine_execlists * const execlists) static inline struct i915_request * execlists_active(const struct intel_engine_execlists *execlists) { - GEM_BUG_ON(execlists->active - execlists->inflight > - execlists_num_ports(execlists)); - return READ_ONCE(*execlists->active); + return *READ_ONCE(execlists->active); } static inline void diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c index 062dd8ac472a..4adc6c8d02eb 100644 --- a/drivers/gpu/drm/i915/gt/intel_lrc.c +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c @@ -2186,6 +2186,10 @@ static void process_csb(struct intel_engine_cs *engine) while (*execlists->active) execlists_schedule_out(*execlists->active++); + /* Point active to the new ELSP; prevent overwriting */ + WRITE_ONCE(execlists->active, execlists->pending); + set_timeslice(engine); + /* switch pending to inflight */ GEM_BUG_ON(!assert_pending_valid(execlists, "promote")); execlists->active = @@ -2194,8 +2198,6 @@ static void process_csb(struct intel_engine_cs *engine) execlists_num_ports(execlists) * sizeof(*execlists->pending)); - set_timeslice(engine); - WRITE_ONCE(execlists->pending[0], NULL); } else { GEM_BUG_ON(!*execlists->active);
Since we want to do a lockless read of the current active request, and that request is written to by process_csb also without serialisation, we need to instruct gcc to take care in reading the pointer itself. Otherwise, we have observed execlists_active() to report 0x40. [ 2400.760381] igt/para-4098 1..s. 2376479300us : process_csb: rcs0 cs-irq head=3, tail=4 [ 2400.760826] igt/para-4098 1..s. 2376479303us : process_csb: rcs0 csb[4]: status=0x00000001:0x00000000 [ 2400.761271] igt/para-4098 1..s. 2376479306us : trace_ports: rcs0: promote { b9c59:2622, b9c55:2624 } [ 2400.761726] igt/para-4097 0d... 2376479311us : __i915_schedule: rcs0: -2147483648->3, inflight:0000000000000040, rq:ffff888208c1e940 which is impossible! Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> --- drivers/gpu/drm/i915/gt/intel_engine.h | 4 +--- drivers/gpu/drm/i915/gt/intel_lrc.c | 6 ++++-- 2 files changed, 5 insertions(+), 5 deletions(-)