From cd078bf95df29d827daca0274a9a13821c11eedb Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 11 May 2018 13:11:45 +0100 Subject: [PATCH] drm/i915/execlists: Use rmb() to order CSB reads MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit We assume that the CSB is written using the normal ringbuffer coherency protocols, as outlined in kernel/events/ring_buffer.c: * (HW) (DRIVER) * * if (LOAD ->data_tail) { LOAD ->data_head * (A) smp_rmb() (C) * STORE $data LOAD $data * smp_wmb() (B) smp_mb() (D) * STORE ->data_head STORE ->data_tail * } So we assume that the HW fulfils its ordering requirements (B), and so we should use a complimentary rmb (C) to ensure that our read of its WRITE pointer is completed before we start accessing the data. The final mb (D) is implied by the uncached mmio we perform to inform the HW of our READ pointer. References: https://bugs.freedesktop.org/show_bug.cgi?id=105064 References: https://bugs.freedesktop.org/show_bug.cgi?id=105888 References: https://bugs.freedesktop.org/show_bug.cgi?id=106185 Fixes: 767a983ab255 ("drm/i915/execlists: Read the context-status HEAD from the HWSP") References: 61bf9719fa17 ("drm/i915/cnl: Use mmio access to context status buffer") Suggested-by: Mika Kuoppala Signed-off-by: Chris Wilson Cc: Mika Kuoppala Cc: Joonas Lahtinen Cc: Tvrtko Ursulin Cc: Michał Winiarski Cc: Rafael Antognolli Cc: Michel Thierry Cc: Timo Aaltonen Tested-by: Timo Aaltonen Acked-by: Michel Thierry Acked-by: Mika Kuoppala Link: https://patchwork.freedesktop.org/patch/msgid/20180511121147.31915-1-chris@chris-wilson.co.uk (cherry picked from commit 77dfedb5be03779f9a5d83e323a1b36e32090105) Signed-off-by: Joonas Lahtinen --- drivers/gpu/drm/i915/intel_lrc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index e3a5f673ff67..8704f7f8d072 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -884,6 +884,7 @@ static void execlists_submission_tasklet(unsigned long data) head = execlists->csb_head; tail = READ_ONCE(buf[write_idx]); + rmb(); /* Hopefully paired with a wmb() in HW */ } GEM_TRACE("%s cs-irq head=%d [%d%s], tail=%d [%d%s]\n", engine->name, -- 2.39.5