[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH] e1000e: Don't zero out buffer address in rx des

From: Jason Wang
Subject: Re: [Qemu-devel] [PATCH] e1000e: Don't zero out buffer address in rx descriptor
Date: Fri, 21 Oct 2016 09:58:48 +0800
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.3.0

On 2016年10月18日 22:10, Dmitry Fleytman wrote:
On 17 Oct 2016, at 01:35 AM, Kevin Wolf <address@hidden> wrote:

The e1000e emulation zeroes out any used rx descriptor and then writes a
completely newly constructed value there. By doing this, it doesn't only
update the write-back area of the descriptors (as it's supposed to do),
but it also clears the buffer address, which real hardware doesn't do.

The spec explicitly mentions in chapter 7.1.8 that it is valid for a
driver to reuse a descriptor and only update the status field while
doing so, i.e. reusing the old buffer address:

    If software statically allocates buffers, and uses memory read to
    check for completed descriptors, it simply has to zero the status
    byte in the descriptor to make it ready for reuse by hardware.

This patch fixes the behaviour to leave the buffer address in
descriptors unchanged even after the descriptor has been used.
Hi Kevin,

Reviewed-by: Dmitry Fleytman <address@hidden>

Thanks for catching this!


Applied, thanks.

Signed-off-by: Kevin Wolf <address@hidden>
hw/net/e1000e_core.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/hw/net/e1000e_core.c b/hw/net/e1000e_core.c
index 9fa4116..a5ca97d 100644
--- a/hw/net/e1000e_core.c
+++ b/hw/net/e1000e_core.c
@@ -1280,11 +1280,10 @@ e1000e_write_lgcy_rx_descr(E1000ECore *core, uint8_t 

     struct e1000_rx_desc *d = (struct e1000_rx_desc *) desc;

-    memset(d, 0, sizeof(*d));

     d->length = cpu_to_le16(length);
+    d->csum = 0;

     e1000e_build_rx_metadata(core, pkt, pkt != NULL,
@@ -1293,6 +1292,7 @@ e1000e_write_lgcy_rx_descr(E1000ECore *core, uint8_t 
     d->errors = (uint8_t) (le32_to_cpu(status_flags) >> 24);
     d->status = (uint8_t) le32_to_cpu(status_flags);
+    d->special = 0;

static inline void
@@ -1303,7 +1303,7 @@ e1000e_write_ext_rx_descr(E1000ECore *core, uint8_t *desc,
     union e1000_rx_desc_extended *d = (union e1000_rx_desc_extended *) desc;

-    memset(d, 0, sizeof(*d));
+    memset(&d->wb, 0, sizeof(d->wb));

     d->wb.upper.length = cpu_to_le16(length);

@@ -1327,7 +1327,7 @@ e1000e_write_ps_rx_descr(E1000ECore *core, uint8_t *desc,
     union e1000_rx_desc_packet_split *d =
         (union e1000_rx_desc_packet_split *) desc;

-    memset(d, 0, sizeof(*d));
+    memset(&d->wb, 0, sizeof(d->wb));

     d->wb.middle.length0 = cpu_to_le16((*written)[0]);


reply via email to

[Prev in Thread] Current Thread [Next in Thread]