diff for duplicates of <527D6B88.3010409@cogentembedded.com> diff --git a/a/1.txt b/N1/1.txt index 4055be4..c919fc4 100644 --- a/a/1.txt +++ b/N1/1.txt @@ -3,8 +3,7 @@ Hello. On 11/08/2013 12:17 PM, Marc Kleine-Budde wrote: >>>> Index: linux-can-next/drivers/net/can/rcar_can.c ->>>> =================================================================== ->>>> --- /dev/null +>>>> =================================>>>> --- /dev/null >>>> +++ linux-can-next/drivers/net/can/rcar_can.c >>>> @@ -0,0 +1,893 @@ [...] @@ -71,7 +70,7 @@ this way... Yes, the mailbox interrupts in MIER1 are enabled only for used mailboxes. ->> + if (unlikely(mier1 == 0xffffffff)) +>> + if (unlikely(mier1 = 0xffffffff)) >> + netif_stop_queue(ndev); > Then you have a race condition in your tx-complete handler @@ -79,7 +78,7 @@ this way... Yes, I'm seeing it now... -> mier1 == 0xffffffff you must wait until _all_ mailboxes are transmitted +> mier1 = 0xffffffff you must wait until _all_ mailboxes are transmitted That 0xffffffff criterion seems wrong for me now, I changed the algorithm and moved the criterion but didn't update it. The correct one seems to be: diff --git a/a/content_digest b/N1/content_digest index 988486e..75c80a2 100644 --- a/a/content_digest +++ b/N1/content_digest @@ -4,7 +4,7 @@ "ref\0527CAC38.6040800@pengutronix.de\0" "From\0Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>\0" "Subject\0Re: [PATCH v3] can: add Renesas R-Car CAN driver\0" - "Date\0Sat, 09 Nov 2013 01:54:00 +0300\0" + "Date\0Fri, 08 Nov 2013 21:54:11 +0000\0" "To\0Marc Kleine-Budde <mkl@pengutronix.de>\0" "Cc\0netdev@vger.kernel.org" wg@grandegger.com @@ -18,8 +18,7 @@ "On 11/08/2013 12:17 PM, Marc Kleine-Budde wrote:\n" "\n" ">>>> Index: linux-can-next/drivers/net/can/rcar_can.c\n" - ">>>> ===================================================================\n" - ">>>> --- /dev/null\n" + ">>>> =================================>>>> --- /dev/null\n" ">>>> +++ linux-can-next/drivers/net/can/rcar_can.c\n" ">>>> @@ -0,0 +1,893 @@\n" "[...]\n" @@ -86,7 +85,7 @@ "\n" " Yes, the mailbox interrupts in MIER1 are enabled only for used mailboxes.\n" "\n" - ">> +\tif (unlikely(mier1 == 0xffffffff))\n" + ">> +\tif (unlikely(mier1 = 0xffffffff))\n" ">> +\t\tnetif_stop_queue(ndev);\n" "\n" "> Then you have a race condition in your tx-complete handler\n" @@ -94,7 +93,7 @@ "\n" " Yes, I'm seeing it now...\n" "\n" - "> mier1 == 0xffffffff you must wait until _all_ mailboxes are transmitted\n" + "> mier1 = 0xffffffff you must wait until _all_ mailboxes are transmitted\n" "\n" " That 0xffffffff criterion seems wrong for me now, I changed the algorithm \n" "and moved the criterion but didn't update it. The correct one seems to be:\n" @@ -190,4 +189,4 @@ "\n" WBR, Sergei -260663788bba038af7a385e1c9df56f7453eac8e489b4dfd0a3bb616729476a8 +fbb6ab93a52c637354d78e1e3b15d28188c1dcacda1713d0cbddce0aa68ebe23
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.