From: Matthew Brost <matthew.brost@intel.com>
To: Matthew Auld <matthew.auld@intel.com>
Cc: intel-xe@lists.freedesktop.org
Subject: Re: [Intel-xe] [PATCH v4 2/7] drm/xe/ct: hold fast_lock when reserving space for g2h
Date: Thu, 6 Jul 2023 03:43:39 +0000 [thread overview]
Message-ID: <ZKY4a0N1ocB0geeg@DUT025-TGLU.fm.intel.com> (raw)
In-Reply-To: <20230705160602.237213-11-matthew.auld@intel.com>
On Wed, Jul 05, 2023 at 05:06:05PM +0100, Matthew Auld wrote:
> Reserving and checking for space on the g2h side relies on the
> fast_lock, and not the CT lock since we need to release space from the
> fast CT path. Make sure we hold it when checking for space and reserving
> it. The main concern is calling __g2h_release_space() as we are reserving
> something and since the info.space and info.g2h_outstanding operations
> are not atomic we can get some nonsense values back.
>
> Signed-off-by: Matthew Auld <matthew.auld@intel.com>
> Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
> Cc: José Roberto de Souza <jose.souza@intel.com>
> ---
> drivers/gpu/drm/xe/xe_guc_ct.c | 22 +++++++++++++++-------
> 1 file changed, 15 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
> index b7aecc480098..f8c1a2ca89f7 100644
> --- a/drivers/gpu/drm/xe/xe_guc_ct.c
> +++ b/drivers/gpu/drm/xe/xe_guc_ct.c
> @@ -346,7 +346,10 @@ static bool h2g_has_room(struct xe_guc_ct *ct, u32 cmd_len)
>
> static bool g2h_has_room(struct xe_guc_ct *ct, u32 g2h_len)
> {
> - lockdep_assert_held(&ct->lock);
> + if (!g2h_len)
> + return true;
> +
> + lockdep_assert_held(&ct->fast_lock);
>
> return ct->ctbs.g2h.info.space > g2h_len;
> }
> @@ -367,15 +370,15 @@ static void h2g_reserve_space(struct xe_guc_ct *ct, u32 cmd_len)
> ct->ctbs.h2g.info.space -= cmd_len;
> }
>
> -static void g2h_reserve_space(struct xe_guc_ct *ct, u32 g2h_len, u32 num_g2h)
> +static void __g2h_reserve_space(struct xe_guc_ct *ct, u32 g2h_len, u32 num_g2h)
> {
> XE_BUG_ON(g2h_len > ct->ctbs.g2h.info.space);
>
> if (g2h_len) {
> - spin_lock_irq(&ct->fast_lock);
> + lockdep_assert_held(&ct->fast_lock);
> +
> ct->ctbs.g2h.info.space -= g2h_len;
> ct->g2h_outstanding += num_g2h;
> - spin_unlock_irq(&ct->fast_lock);
> }
> }
>
> @@ -499,21 +502,26 @@ static int __guc_ct_send_locked(struct xe_guc_ct *ct, const u32 *action,
> }
> }
>
> + if (g2h_len)
> + spin_lock_irq(&ct->fast_lock);
> retry:
> ret = has_room(ct, len + GUC_CTB_HDR_LEN, g2h_len);
> if (unlikely(ret))
> - goto out;
> + goto out_unlock;
>
> ret = h2g_write(ct, action, len, g2h_fence ? g2h_fence->seqno : 0,
> !!g2h_fence);
> if (unlikely(ret)) {
> if (ret == -EAGAIN)
> goto retry;
> - goto out;
> + goto out_unlock;
> }
>
> - g2h_reserve_space(ct, g2h_len, num_g2h);
> + __g2h_reserve_space(ct, g2h_len, num_g2h);
> xe_guc_notify(ct_to_guc(ct));
> +out_unlock:
> + if (g2h_len)
> + spin_unlock_irq(&ct->fast_lock);
> out:
> return ret;
> }
> --
> 2.41.0
>
next prev parent reply other threads:[~2023-07-06 3:44 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-05 16:06 [Intel-xe] [PATCH v4 0/7] Try to handle TLB invalidations from CT fast-path Matthew Auld
2023-07-05 16:06 ` [Intel-xe] [PATCH v4 1/7] drm/xe: hold mem_access.ref for " Matthew Auld
2023-07-06 3:51 ` Matthew Brost
2023-07-06 8:29 ` Matthew Auld
2023-07-06 14:50 ` Matthew Brost
2023-07-05 16:06 ` [Intel-xe] [PATCH v4 2/7] drm/xe/ct: hold fast_lock when reserving space for g2h Matthew Auld
2023-07-06 3:43 ` Matthew Brost [this message]
2023-07-05 16:06 ` [Intel-xe] [PATCH v4 3/7] drm/xe/tlb: increment next seqno after successful CT send Matthew Auld
2023-07-06 3:59 ` Matthew Brost
2023-07-06 9:42 ` Matthew Auld
2023-07-06 15:15 ` Matthew Brost
2023-07-06 15:22 ` Matthew Auld
2023-07-05 16:06 ` [Intel-xe] [PATCH v4 4/7] drm/xe/ct: serialise fast_lock during CT disable Matthew Auld
2023-07-06 4:00 ` Matthew Brost
2023-07-05 16:06 ` [Intel-xe] [PATCH v4 5/7] drm/xe/gt: tweak placement for signalling TLB fences after GT reset Matthew Auld
2023-07-06 4:01 ` Matthew Brost
2023-07-05 16:06 ` [Intel-xe] [PATCH v4 6/7] drm/xe/tlb: also update seqno_recv during reset Matthew Auld
2023-07-06 4:05 ` Matthew Brost
2023-07-06 10:02 ` Matthew Auld
2023-07-05 16:06 ` [Intel-xe] [PATCH v4 7/7] drm/xe: handle TLB invalidations from CT fast-path Matthew Auld
2023-07-06 4:14 ` Matthew Brost
2023-07-05 16:10 ` [Intel-xe] ✓ CI.Patch_applied: success for Try to handle TLB invalidations from CT fast-path (rev2) Patchwork
2023-07-05 16:11 ` [Intel-xe] ✗ CI.checkpatch: warning " Patchwork
2023-07-05 16:12 ` [Intel-xe] ✓ CI.KUnit: success " Patchwork
2023-07-05 16:16 ` [Intel-xe] ✓ CI.Build: " Patchwork
2023-07-05 16:16 ` [Intel-xe] ✓ CI.Hooks: " Patchwork
2023-07-05 16:17 ` [Intel-xe] ✓ CI.checksparse: " Patchwork
2023-07-05 17:02 ` [Intel-xe] ○ CI.BAT: info " Patchwork
2023-07-06 15:23 ` [Intel-xe] [PATCH v4 0/7] Try to handle TLB invalidations from CT fast-path Souza, Jose
2023-07-06 15:48 ` Matthew Auld
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=ZKY4a0N1ocB0geeg@DUT025-TGLU.fm.intel.com \
--to=matthew.brost@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=matthew.auld@intel.com \
/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: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox