qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Finn Thain <fthain@telegraphics.com.au>
To: Jason Wang <jasowang@redhat.com>, qemu-devel@nongnu.org
Cc: Aleksandar Rikalo <aleksandar.rikalo@rt-rk.com>,
	Herve Poussineau <hpoussin@reactos.org>,
	Laurent Vivier <laurent@vivier.eu>,
	qemu-stable@nongnu.org
Subject: [PATCH 09/10] dp8393x: Don't stop reception upon RBE interrupt assertion
Date: Sat, 14 Dec 2019 12:25:57 +1100	[thread overview]
Message-ID: <f0acfe7f91ac42560a07661cd0f07143fce338c3.1576286757.git.fthain@telegraphics.com.au> (raw)
In-Reply-To: <cover.1576286757.git.fthain@telegraphics.com.au>

Section 5.4.7 of the datasheet explains that the RBE interrupt is an
"early warning". It indicates that the last RBA is still available to
receive a packet. It doesn't imply actual receive buffer exhaustion.

This is an important distinction because Linux will not check and clear
the RBE interrupt until it receives another packet. But that won't
happen if can_receive returns false. So the SONIC becomes deaf (until
reset).

Fix this with a new flag to indicate actual receive buffer exhaustion.

Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
---
 hw/net/dp8393x.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/hw/net/dp8393x.c b/hw/net/dp8393x.c
index 24f2e19f07..8e66b1f5de 100644
--- a/hw/net/dp8393x.c
+++ b/hw/net/dp8393x.c
@@ -157,6 +157,7 @@ typedef struct dp8393xState {
     /* Hardware */
     uint8_t it_shift;
     bool big_endian;
+    bool last_rba_is_full;
     qemu_irq irq;
 #ifdef DEBUG_SONIC
     int irq_level;
@@ -311,6 +312,12 @@ static void dp8393x_do_read_rra(dp8393xState *s)
 {
     int width, size;
 
+    /* Already on the last RBA? */
+    s->last_rba_is_full = s->regs[SONIC_ISR] & SONIC_ISR_RBE;
+    if (s->last_rba_is_full) {
+        return;
+    }
+
     /* Read memory */
     width = (s->regs[SONIC_DCR] & SONIC_DCR_DW) ? 2 : 1;
     size = sizeof(uint16_t) * 4 * width;
@@ -334,7 +341,7 @@ static void dp8393x_do_read_rra(dp8393xState *s)
         s->regs[SONIC_RRP] = s->regs[SONIC_RSA];
     }
 
-    /* Check resource exhaustion */
+    /* Warn the host if this entry is the last one */
     if (s->regs[SONIC_RRP] == s->regs[SONIC_RWP])
     {
         s->regs[SONIC_ISR] |= SONIC_ISR_RBE;
@@ -703,8 +710,9 @@ static int dp8393x_can_receive(NetClientState *nc)
 
     if (!(s->regs[SONIC_CR] & SONIC_CR_RXEN))
         return 0;
-    if (s->regs[SONIC_ISR] & SONIC_ISR_RBE)
+    if (s->last_rba_is_full) {
         return 0;
+    }
     return 1;
 }
 
-- 
2.23.0



  parent reply	other threads:[~2019-12-14  1:35 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-12-14  1:25 [PATCH 00/10] Fixes for DP8393X SONIC device emulation Finn Thain
2019-12-14  1:25 ` [PATCH 08/10] dp8393x: Implement packet size limit and RBAE interrupt Finn Thain
2019-12-14  1:25 ` Finn Thain [this message]
2019-12-14  1:25 ` [PATCH 06/10] dp8393x: Clear RRRA command register bit only when appropriate Finn Thain
2019-12-14 13:31   ` Philippe Mathieu-Daudé
2019-12-14  1:25 ` [PATCH 02/10] dp8393x: Clean up endianness hacks Finn Thain
2019-12-14  1:25 ` [PATCH 05/10] dp8393x: Update LLFA register Finn Thain
2019-12-14  1:25 ` [PATCH 10/10] dp8393x: Don't clobber packet checksum Finn Thain
2019-12-14 13:21   ` Philippe Mathieu-Daudé
2019-12-14  1:25 ` [PATCH 07/10] dp8393x: Implement TBWC0 and TBWC1 registers to restore buffer state Finn Thain
2019-12-14  1:25 ` [PATCH 01/10] dp8393x: Mask EOL bit from descriptor addresses Finn Thain
2019-12-14 13:35   ` Philippe Mathieu-Daudé
2019-12-14 23:21     ` Finn Thain
2019-12-14  1:25 ` [PATCH 04/10] dp8393x: Don't advance RX descriptor twice Finn Thain
2019-12-14  1:25 ` [PATCH 03/10] dp8393x: Have dp8393x_receive() return the packet size Finn Thain
2019-12-14 13:26   ` Philippe Mathieu-Daudé
2019-12-14  1:43 ` [PATCH 00/10] Fixes for DP8393X SONIC device emulation no-reply
2019-12-14  2:52   ` Finn Thain
2019-12-14 13:38     ` Philippe Mathieu-Daudé
2019-12-14 13:45       ` Eric Blake
2019-12-14 17:17 ` Aleksandar Markovic
2019-12-14 23:16   ` Finn Thain
2019-12-14 23:32     ` Aleksandar Markovic
2019-12-14 23:35       ` Aleksandar Markovic
2019-12-20  4:24         ` Finn Thain
2019-12-23 17:17           ` Philippe Mathieu-Daudé
2019-12-24  0:12             ` NetBSD/arc on MIPS Magnum, was " Finn Thain
2019-12-24  4:33               ` Finn Thain
2019-12-24  6:53                 ` Hervé Poussineau
2020-01-06 22:15                   ` Finn Thain
2019-12-16  0:36     ` Finn Thain
2019-12-20  4:21       ` Finn Thain
2019-12-20 11:38 ` Aleksandar Markovic
2019-12-20 12:03   ` Aleksandar Markovic
2019-12-20 12:54   ` Laurent Vivier
2019-12-20 23:22     ` Finn Thain
2019-12-21 12:03       ` Aleksandar Markovic

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=f0acfe7f91ac42560a07661cd0f07143fce338c3.1576286757.git.fthain@telegraphics.com.au \
    --to=fthain@telegraphics.com.au \
    --cc=aleksandar.rikalo@rt-rk.com \
    --cc=hpoussin@reactos.org \
    --cc=jasowang@redhat.com \
    --cc=laurent@vivier.eu \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-stable@nongnu.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).