From: "Yuval Mintz" <yuvalmin@broadcom.com>
To: davem@davemloft.net, netdev@vger.kernel.org
Cc: ariele@broadcom.com, eilong@broadcom.com,
"Yuval Mintz" <yuvalmin@broadcom.com>
Subject: [PATCH net 5/5] bnx2x: Fix VF stats sync
Date: Wed, 28 Aug 2013 01:13:04 +0300 [thread overview]
Message-ID: <1377641584-2939-6-git-send-email-yuvalmin@broadcom.com> (raw)
In-Reply-To: <1377641584-2939-1-git-send-email-yuvalmin@broadcom.com>
From: Ariel Elior <ariele@broadcom.com>
Since the PF gathers statistics for the VF, when the VF is about to unload
we must synchronize the release of its statistics buffer with the PF, so that
no DMA operation will be made to that address after the buffer release.
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
---
drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c | 26 ++++++++++++++++++++++-
drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c | 11 ++++++++++
drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h | 3 +++
3 files changed, 39 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
index b7efe27..e8706e1 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
@@ -2786,6 +2786,18 @@ int bnx2x_vf_init(struct bnx2x *bp, struct bnx2x_virtf *vf, dma_addr_t *sb_map)
return 0;
}
+struct set_vf_state_cookie {
+ struct bnx2x_virtf *vf;
+ u8 state;
+};
+
+void bnx2x_set_vf_state(void *cookie)
+{
+ struct set_vf_state_cookie *p = (struct set_vf_state_cookie *)cookie;
+
+ p->vf->state = p->state;
+}
+
/* VFOP close (teardown the queues, delete mcasts and close HW) */
static void bnx2x_vfop_close(struct bnx2x *bp, struct bnx2x_virtf *vf)
{
@@ -2836,7 +2848,19 @@ static void bnx2x_vfop_close(struct bnx2x *bp, struct bnx2x_virtf *vf)
op_err:
BNX2X_ERR("VF[%d] CLOSE error: rc %d\n", vf->abs_vfid, vfop->rc);
op_done:
- vf->state = VF_ACQUIRED;
+
+ /* need to make sure there are no outstanding stats ramrods which may
+ * cause the device to access the VF's stats buffer which it will free
+ * as soon as we return from the close flow.
+ */
+ {
+ struct set_vf_state_cookie cookie;
+
+ cookie.vf = vf;
+ cookie.state = VF_ACQUIRED;
+ bnx2x_stats_safe_exec(bp, bnx2x_set_vf_state, &cookie);
+ }
+
DP(BNX2X_MSG_IOV, "set state to acquired\n");
bnx2x_vfop_end(bp, vf, vfop);
}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c
index fed9b15..86436c7 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c
@@ -1993,3 +1993,14 @@ void bnx2x_afex_collect_stats(struct bnx2x *bp, void *void_afex_stats,
estats->mac_discard);
}
}
+
+void bnx2x_stats_safe_exec(struct bnx2x *bp,
+ void (func_to_exec)(void *cookie),
+ void *cookie){
+ if (down_timeout(&bp->stats_sema, HZ/10))
+ BNX2X_ERR("Unable to acquire stats lock\n");
+ bnx2x_stats_comp(bp);
+ func_to_exec(cookie);
+ __bnx2x_stats_start(bp);
+ up(&bp->stats_sema);
+}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h
index 853824d..f358450 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h
@@ -539,6 +539,9 @@ struct bnx2x;
void bnx2x_memset_stats(struct bnx2x *bp);
void bnx2x_stats_init(struct bnx2x *bp);
void bnx2x_stats_handle(struct bnx2x *bp, enum bnx2x_stats_event event);
+void bnx2x_stats_safe_exec(struct bnx2x *bp,
+ void (func_to_exec)(void *cookie),
+ void *cookie);
/**
* bnx2x_save_statistics - save statistics when unloading.
--
1.8.1.227.g44fe835
next prev parent reply other threads:[~2013-08-27 22:13 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-08-27 22:12 [PATCH net 0/5] bnx2x: Bug fixes patch series Yuval Mintz
2013-08-27 22:13 ` [PATCH net 1/5] bnx2x: vf mark stats started Yuval Mintz
2013-08-27 22:13 ` [PATCH net 2/5] bnx2x: Fix move FP memory deallocations Yuval Mintz
2013-08-27 22:13 ` [PATCH net 3/5] bnx2x: Fix functionality of configuring vlan list Yuval Mintz
2013-08-27 22:13 ` [PATCH net 4/5] bnx2x: Fix VF memory leak unload Yuval Mintz
2013-08-27 22:13 ` Yuval Mintz [this message]
2013-08-28 2:04 ` [PATCH net 0/5] bnx2x: Bug fixes patch series David Miller
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=1377641584-2939-6-git-send-email-yuvalmin@broadcom.com \
--to=yuvalmin@broadcom.com \
--cc=ariele@broadcom.com \
--cc=davem@davemloft.net \
--cc=eilong@broadcom.com \
--cc=netdev@vger.kernel.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: 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;
as well as URLs for NNTP newsgroup(s).