From: "Yang, Fei" <fei.yang@intel.com> To: "Hajda, Andrzej" <andrzej.hajda@intel.com>, "intel-gfx@lists.freedesktop.org" <intel-gfx@lists.freedesktop.org> Cc: "dri-devel@lists.freedesktop.org" <dri-devel@lists.freedesktop.org> Subject: RE: [Intel-gfx] [PATCH 4/8] drm/i915/mtl: workaround coherency issue for Media Date: Wed, 19 Apr 2023 18:49:17 +0000 [thread overview] Message-ID: <SN6PR11MB257482462F00C5B347E6CC339A629@SN6PR11MB2574.namprd11.prod.outlook.com> (raw) In-Reply-To: <SN6PR11MB2574462BC46616D1954D40309A629@SN6PR11MB2574.namprd11.prod.outlook.com> >>> diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c >>> b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c >>> index 1803a633ed64..98e682b7df07 100644 >>> --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c >>> +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c >>> @@ -415,12 +415,6 @@ static int ct_write(struct intel_guc_ct *ct, >>> } >>> GEM_BUG_ON(tail > size); >>> >>> - /* >>> - * make sure H2G buffer update and LRC tail update (if this triggering a >>> - * submission) are visible before updating the descriptor tail >>> - */ >>> - intel_guc_write_barrier(ct_to_guc(ct)); >>> - >>> /* update local copies */ >>> ctb->tail = tail; >>> GEM_BUG_ON(atomic_read(&ctb->space) < len + GUC_CTB_HDR_LEN); @@ >>> -429,6 +423,12 @@ static int ct_write(struct intel_guc_ct *ct, >>> /* now update descriptor */ >>> WRITE_ONCE(desc->tail, tail); >>> >>> + /* >>> + * make sure H2G buffer update and LRC tail update (if this triggering a >>> + * submission) are visible before updating the descriptor tail >>> + */ >>> + intel_guc_write_barrier(ct_to_guc(ct)); >> >> The comment above needs update, Never mind, I decided to revert this change because it's not necessary. There is a MMIO write following the ct_write() call which would flush the write combining buffer anyway, so the barrier is redundant here. > >Will update the comment. > >> if this is correct change. The question is why it is correct? If yes, >> it implies that old barrier is incorrect, maybe it should be then separate fix? > > There is WRITE_ONCE(desc->tail, tail) right after the H2G buffer update which is also > seen by the GuC firmware, the barrier is needed for both, thus moved it down a few > lines to cover them all. > >> I am not an expert, but previous location of the barrier seems sane to >> me - assure GuC will see proper buffer, before updating buffer's tail. > > That is correct, but the barrier is needed for both H2G buffer and descriptor, as they are all shared with GuC firmware. > > -Fei > >> And according to commit message this new barrier should flush WC >> buffer, so for me it seems to be different thing. >> Am I missing something? >> >> >> Regards >> Andrzej
WARNING: multiple messages have this Message-ID (diff)
From: "Yang, Fei" <fei.yang@intel.com> To: "Hajda, Andrzej" <andrzej.hajda@intel.com>, "intel-gfx@lists.freedesktop.org" <intel-gfx@lists.freedesktop.org> Cc: "dri-devel@lists.freedesktop.org" <dri-devel@lists.freedesktop.org> Subject: Re: [Intel-gfx] [PATCH 4/8] drm/i915/mtl: workaround coherency issue for Media Date: Wed, 19 Apr 2023 18:49:17 +0000 [thread overview] Message-ID: <SN6PR11MB257482462F00C5B347E6CC339A629@SN6PR11MB2574.namprd11.prod.outlook.com> (raw) In-Reply-To: <SN6PR11MB2574462BC46616D1954D40309A629@SN6PR11MB2574.namprd11.prod.outlook.com> >>> diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c >>> b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c >>> index 1803a633ed64..98e682b7df07 100644 >>> --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c >>> +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c >>> @@ -415,12 +415,6 @@ static int ct_write(struct intel_guc_ct *ct, >>> } >>> GEM_BUG_ON(tail > size); >>> >>> - /* >>> - * make sure H2G buffer update and LRC tail update (if this triggering a >>> - * submission) are visible before updating the descriptor tail >>> - */ >>> - intel_guc_write_barrier(ct_to_guc(ct)); >>> - >>> /* update local copies */ >>> ctb->tail = tail; >>> GEM_BUG_ON(atomic_read(&ctb->space) < len + GUC_CTB_HDR_LEN); @@ >>> -429,6 +423,12 @@ static int ct_write(struct intel_guc_ct *ct, >>> /* now update descriptor */ >>> WRITE_ONCE(desc->tail, tail); >>> >>> + /* >>> + * make sure H2G buffer update and LRC tail update (if this triggering a >>> + * submission) are visible before updating the descriptor tail >>> + */ >>> + intel_guc_write_barrier(ct_to_guc(ct)); >> >> The comment above needs update, Never mind, I decided to revert this change because it's not necessary. There is a MMIO write following the ct_write() call which would flush the write combining buffer anyway, so the barrier is redundant here. > >Will update the comment. > >> if this is correct change. The question is why it is correct? If yes, >> it implies that old barrier is incorrect, maybe it should be then separate fix? > > There is WRITE_ONCE(desc->tail, tail) right after the H2G buffer update which is also > seen by the GuC firmware, the barrier is needed for both, thus moved it down a few > lines to cover them all. > >> I am not an expert, but previous location of the barrier seems sane to >> me - assure GuC will see proper buffer, before updating buffer's tail. > > That is correct, but the barrier is needed for both H2G buffer and descriptor, as they are all shared with GuC firmware. > > -Fei > >> And according to commit message this new barrier should flush WC >> buffer, so for me it seems to be different thing. >> Am I missing something? >> >> >> Regards >> Andrzej
next prev parent reply other threads:[~2023-04-19 18:49 UTC|newest] Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top 2023-04-17 6:24 [PATCH 0/8] drm/i915/mtl: Define MOCS and PAT tables for MTL fei.yang 2023-04-17 6:24 ` [Intel-gfx] " fei.yang 2023-04-17 6:24 ` [PATCH 1/8] drm/i915/mtl: Set has_llc=0 fei.yang 2023-04-17 6:24 ` [Intel-gfx] " fei.yang 2023-04-19 10:59 ` Andi Shyti 2023-04-19 12:50 ` Andrzej Hajda 2023-04-19 14:10 ` Das, Nirmoy 2023-04-17 6:24 ` [PATCH 2/8] drm/i915/mtl: Define MOCS and PAT tables for MTL fei.yang 2023-04-17 6:24 ` [Intel-gfx] " fei.yang 2023-04-19 11:01 ` Andi Shyti 2023-04-19 16:00 ` Yang, Fei 2023-04-19 16:00 ` Yang, Fei 2023-04-19 13:59 ` Andrzej Hajda 2023-04-19 16:03 ` Yang, Fei 2023-04-19 16:03 ` Yang, Fei 2023-04-19 14:36 ` Das, Nirmoy 2023-04-19 16:05 ` Yang, Fei 2023-04-19 16:05 ` Yang, Fei 2023-04-17 6:24 ` [PATCH 3/8] drm/i915/mtl: Add PTE encode function fei.yang 2023-04-17 6:24 ` [Intel-gfx] " fei.yang 2023-04-19 11:02 ` Andi Shyti 2023-04-19 12:51 ` Andrzej Hajda 2023-04-19 15:11 ` Das, Nirmoy 2023-04-19 15:11 ` [Intel-gfx] " Das, Nirmoy 2023-04-17 6:24 ` [PATCH 4/8] drm/i915/mtl: workaround coherency issue for Media fei.yang 2023-04-17 6:24 ` [Intel-gfx] " fei.yang 2023-04-19 10:59 ` Andi Shyti 2023-04-19 12:38 ` Andi Shyti 2023-04-19 15:14 ` Das, Nirmoy 2023-04-19 15:14 ` [Intel-gfx] " Das, Nirmoy 2023-04-19 15:40 ` Andrzej Hajda 2023-04-19 16:37 ` Yang, Fei 2023-04-19 16:37 ` Yang, Fei 2023-04-19 18:49 ` Yang, Fei [this message] 2023-04-19 18:49 ` Yang, Fei 2023-04-17 6:25 ` [PATCH 5/8] drm/i915/mtl: end support for set caching ioctl fei.yang 2023-04-17 6:25 ` [Intel-gfx] " fei.yang 2023-04-19 11:08 ` Andi Shyti 2023-04-19 13:05 ` Andrzej Hajda 2023-04-19 16:56 ` Yang, Fei 2023-04-19 16:56 ` Yang, Fei 2023-04-17 6:25 ` [PATCH 6/8] drm/i915: preparation for using PAT index fei.yang 2023-04-17 6:25 ` [Intel-gfx] " fei.yang 2023-04-19 11:17 ` Andi Shyti 2023-04-19 11:17 ` [Intel-gfx] " Andi Shyti 2023-04-17 6:25 ` [PATCH 7/8] drm/i915: use pat_index instead of cache_level fei.yang 2023-04-17 6:25 ` [Intel-gfx] " fei.yang 2023-04-19 12:16 ` Andi Shyti 2023-04-17 6:25 ` [PATCH 8/8] drm/i915: Allow user to set cache at BO creation fei.yang 2023-04-17 6:25 ` [Intel-gfx] " fei.yang 2023-04-19 12:23 ` Andi Shyti 2023-04-19 12:23 ` [Intel-gfx] " Andi Shyti 2023-04-17 6:32 ` [Intel-gfx] [PATCH 0/8] drm/i915/mtl: Define MOCS and PAT tables for MTL Timo Aaltonen 2023-04-17 6:43 ` Yang, Fei 2023-04-17 6:43 ` Yang, Fei 2023-04-24 20:00 ` Jordan Justen 2023-04-24 20:00 ` Jordan Justen 2023-04-17 11:59 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915/mtl: Define MOCS and PAT tables for MTL (rev5) Patchwork 2023-04-17 11:59 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork 2023-04-17 12:15 ` [Intel-gfx] ✗ Fi.CI.BAT: failure " Patchwork 2023-04-19 18:09 [PATCH 0/8] drm/i915/mtl: Define MOCS and PAT tables for MTL fei.yang 2023-04-19 18:09 ` [Intel-gfx] [PATCH 4/8] drm/i915/mtl: workaround coherency issue for Media fei.yang 2023-04-19 21:12 [PATCH 0/8] drm/i915/mtl: Define MOCS and PAT tables for MTL fei.yang 2023-04-19 21:12 ` [Intel-gfx] [PATCH 4/8] drm/i915/mtl: workaround coherency issue for Media fei.yang 2023-04-19 22:07 ` Andi Shyti 2023-04-19 23:00 [PATCH 0/8] drm/i915/mtl: Define MOCS and PAT tables for MTL fei.yang 2023-04-19 23:00 ` [Intel-gfx] [PATCH 4/8] drm/i915/mtl: workaround coherency issue for Media fei.yang 2023-04-20 8:26 ` Andrzej Hajda 2023-04-20 11:36 ` Das, Nirmoy 2023-04-20 11:36 ` Das, Nirmoy 2023-04-20 20:52 ` Matt Roper
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=SN6PR11MB257482462F00C5B347E6CC339A629@SN6PR11MB2574.namprd11.prod.outlook.com \ --to=fei.yang@intel.com \ --cc=andrzej.hajda@intel.com \ --cc=dri-devel@lists.freedesktop.org \ --cc=intel-gfx@lists.freedesktop.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.