| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <CAACQVJrO_PN8LBY0ovwkdxGsyvW_gGN7C3MxnuW+jjdS_75Hhw@mail.gmail.com>
Date: Fri, 18 Oct 2019 12:04:35 +0530
From: Vasundhara Volam <vasundhara-v.volam@...adcom.com>
To: Jakub Kicinski <jakub.kicinski@...ronome.com>
Cc: Sheetal Tigadoli <sheetal.tigadoli@...adcom.com>,
Rafał Miłecki <zajec5@...il.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Michal Simek <michal.simek@...inx.com>,
Rajan Vaja <rajan.vaja@...inx.com>,
Scott Branden <scott.branden@...adcom.com>,
Ray Jui <ray.jui@...adcom.com>,
Vikram Prakash <vikram.prakash@...adcom.com>,
Jens Wiklander <jens.wiklander@...aro.org>,
Michael Chan <michael.chan@...adcom.com>,
"David S. Miller" <davem@...emloft.net>,
Vikas Gupta <vikas.gupta@...adcom.com>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
tee-dev@...ts.linaro.org, bcm-kernel-feedback-list@...adcom.com,
Netdev <netdev@...r.kernel.org>
Subject: Re: [PATCH V2 3/3] bnxt_en: Add support to collect crash dump via ethtool
On Fri, Oct 18, 2019 at 12:52 AM Jakub Kicinski
<jakub.kicinski@...ronome.com> wrote:
>
> On Thu, 17 Oct 2019 17:31:22 +0530, Sheetal Tigadoli wrote:
> > From: Vasundhara Volam <vasundhara-v.volam@...adcom.com>
> >
> > Driver supports 2 types of core dumps.
> >
> > 1. Live dump - Firmware dump when system is up and running.
> > 2. Crash dump - Dump which is collected during firmware crash
> > that can be retrieved after recovery.
> > Crash dump is currently supported only on specific 58800 chips
> > which can be retrieved using OP-TEE API only, as firmware cannot
> > access this region directly.
> >
> > User needs to set the dump flag using following command before
> > initiating the dump collection:
> >
> > $ ethtool -W|--set-dump eth0 N
> >
> > Where N is "0" for live dump and "1" for crash dump
> >
> > Command to collect the dump after setting the flag:
> >
> > $ ethtool -w eth0 data Filename
> >
> > Cc: Michael Chan <michael.chan@...adcom.com>
> > Signed-off-by: Vasundhara Volam <vasundhara-v.volam@...adcom.com>
> > Signed-off-by: Sheetal Tigadoli <sheetal.tigadoli@...adcom.com>
> > ---
> > drivers/net/ethernet/broadcom/bnxt/bnxt.h | 3 ++
> > drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 36 +++++++++++++++++++++--
> > drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h | 2 ++
> > 3 files changed, 39 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
> > index 0943715..3e7d1fb 100644
> > --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
> > +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
> > @@ -1807,6 +1807,9 @@ struct bnxt {
> >
> > u8 num_leds;
> > struct bnxt_led_info leds[BNXT_MAX_LED];
> > + u16 dump_flag;
> > +#define BNXT_DUMP_LIVE 0
> > +#define BNXT_DUMP_CRASH 1
> >
> > struct bpf_prog *xdp_prog;
> >
> > diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
> > index 51c1404..1596221 100644
> > --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
> > +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
> > @@ -3311,6 +3311,23 @@ static int bnxt_get_coredump(struct bnxt *bp, void *buf, u32 *dump_len)
> > return rc;
> > }
> >
> > +static int bnxt_set_dump(struct net_device *dev, struct ethtool_dump *dump)
> > +{
> > + struct bnxt *bp = netdev_priv(dev);
> > +
> > +#ifndef CONFIG_TEE_BNXT_FW
> > + return -EOPNOTSUPP;
> > +#endif
>
> if (!IS_ENABLED(...))
> return x;
>
> reads better IMHO
Okay.
>
> But also you seem to be breaking live dump for systems with
> CONFIG_TEE_BNXT_FW=n
Yes, we are supporting set_dump only if crash dump is supported.
>
> > + if (dump->flag > BNXT_DUMP_CRASH) {
> > + netdev_err(dev, "Supports only Live(0) and Crash(1) dumps.\n");
>
> more of an _info than _err, if at all
I made this err, as we are returning error on invalid flag value. I
can modify the log to
something like "Invalid dump flag. Supports only Live(0) and Crash(1)
dumps.\n" to make
it more like error log.
>
> > + return -EINVAL;
> > + }
> > +
> > + bp->dump_flag = dump->flag;
> > + return 0;
> > +}
> > +
> > static int bnxt_get_dump_flag(struct net_device *dev, struct ethtool_dump *dump)
> > {
> > struct bnxt *bp = netdev_priv(dev);
> > @@ -3323,7 +3340,12 @@ static int bnxt_get_dump_flag(struct net_device *dev, struct ethtool_dump *dump)
> > bp->ver_resp.hwrm_fw_bld_8b << 8 |
> > bp->ver_resp.hwrm_fw_rsvd_8b;
> >
> > - return bnxt_get_coredump(bp, NULL, &dump->len);
> > + dump->flag = bp->dump_flag;
> > + if (bp->dump_flag == BNXT_DUMP_CRASH)
> > + dump->len = BNXT_CRASH_DUMP_LEN;
> > + else
> > + bnxt_get_coredump(bp, NULL, &dump->len);
> > + return 0;
> > }
> >
> > static int bnxt_get_dump_data(struct net_device *dev, struct ethtool_dump *dump,
> > @@ -3336,7 +3358,16 @@ static int bnxt_get_dump_data(struct net_device *dev, struct ethtool_dump *dump,
> >
> > memset(buf, 0, dump->len);
> >
> > - return bnxt_get_coredump(bp, buf, &dump->len);
> > + dump->flag = bp->dump_flag;
> > + if (dump->flag == BNXT_DUMP_CRASH) {
> > +#ifdef CONFIG_TEE_BNXT_FW
> > + return tee_bnxt_copy_coredump(buf, 0, dump->len);
> > +#endif
> > + } else {
> > + return bnxt_get_coredump(bp, buf, &dump->len);
> > + }
> > +
> > + return 0;
> > }
> >
> > void bnxt_ethtool_init(struct bnxt *bp)
> > @@ -3446,6 +3477,7 @@ void bnxt_ethtool_free(struct bnxt *bp)
> > .set_phys_id = bnxt_set_phys_id,
> > .self_test = bnxt_self_test,
> > .reset = bnxt_reset,
> > + .set_dump = bnxt_set_dump,
> > .get_dump_flag = bnxt_get_dump_flag,
> > .get_dump_data = bnxt_get_dump_data,
> > };
> > diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
> > index b5b65b3..01de7e7 100644
> > --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
> > +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
> > @@ -59,6 +59,8 @@ struct hwrm_dbg_cmn_output {
> > #define HWRM_DBG_CMN_FLAGS_MORE 1
> > };
> >
> > +#define BNXT_CRASH_DUMP_LEN (8 << 20)
> > +
> > #define BNXT_LED_DFLT_ENA \
> > (PORT_LED_CFG_REQ_ENABLES_LED0_ID | \
> > PORT_LED_CFG_REQ_ENABLES_LED0_STATE | \
>
Thanks,
Vasundhara
Powered by blists - more mailing lists