| Message ID | 1556369542-13247-23-git-send-email-info@metux.net |
|---|---|
| State | Not Applicable |
| Delegated to: | David Miller |
| Headers | show |
| Series | [01/41] drivers: tty: serial: dz: use dev_err() instead of printk() | expand |
Le 27/04/2019 à 14:52, Enrico Weigelt, metux IT consult a écrit : > Fix checkpatch warnings by using pr_err(): > > WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ... > #109: FILE: drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c:109: > + printk(KERN_ERR > > WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ... > #128: FILE: drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c:128: > + printk(KERN_ERR > > WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ... > + printk(KERN_ERR > > WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ... > + printk(KERN_ERR > > Signed-off-by: Enrico Weigelt <info@metux.net> Reviewed-by: Christophe Leroy <christophe.leroy@c-s.fr> But is that really worth doing those changes ? If we want to do something useful, wouldn't it make more sense to introduce the use of dev_err() in order to identify the faulting device in the message ? Christophe > --- > drivers/tty/serial/cpm_uart/cpm_uart_cpm1.c | 6 ++---- > drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c | 6 ++---- > 2 files changed, 4 insertions(+), 8 deletions(-) > > diff --git a/drivers/tty/serial/cpm_uart/cpm_uart_cpm1.c b/drivers/tty/serial/cpm_uart/cpm_uart_cpm1.c > index 56fc527..aed61e9 100644 > --- a/drivers/tty/serial/cpm_uart/cpm_uart_cpm1.c > +++ b/drivers/tty/serial/cpm_uart/cpm_uart_cpm1.c > @@ -71,8 +71,7 @@ int cpm_uart_allocbuf(struct uart_cpm_port *pinfo, unsigned int is_con) > dpmemsz = sizeof(cbd_t) * (pinfo->rx_nrfifos + pinfo->tx_nrfifos); > dp_offset = cpm_dpalloc(dpmemsz, 8); > if (IS_ERR_VALUE(dp_offset)) { > - printk(KERN_ERR > - "cpm_uart_cpm1.c: could not allocate buffer descriptors\n"); > + pr_err("cpm_uart_cpm1.c: could not allocate buffer descriptors\n"); > return -ENOMEM; > } > dp_mem = cpm_dpram_addr(dp_offset); > @@ -90,8 +89,7 @@ int cpm_uart_allocbuf(struct uart_cpm_port *pinfo, unsigned int is_con) > > if (mem_addr == NULL) { > cpm_dpfree(dp_offset); > - printk(KERN_ERR > - "cpm_uart_cpm1.c: could not allocate coherent memory\n"); > + pr_err("cpm_uart_cpm1.c: could not allocate coherent memory\n"); > return -ENOMEM; > } > > diff --git a/drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c b/drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c > index 40cfcf4..a0fccda 100644 > --- a/drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c > +++ b/drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c > @@ -106,8 +106,7 @@ int cpm_uart_allocbuf(struct uart_cpm_port *pinfo, unsigned int is_con) > dpmemsz = sizeof(cbd_t) * (pinfo->rx_nrfifos + pinfo->tx_nrfifos); > dp_offset = cpm_dpalloc(dpmemsz, 8); > if (IS_ERR_VALUE(dp_offset)) { > - printk(KERN_ERR > - "cpm_uart_cpm.c: could not allocate buffer descriptors\n"); > + pr_err("cpm_uart_cpm.c: could not allocate buffer descriptors\n"); > return -ENOMEM; > } > > @@ -125,8 +124,7 @@ int cpm_uart_allocbuf(struct uart_cpm_port *pinfo, unsigned int is_con) > > if (mem_addr == NULL) { > cpm_dpfree(dp_offset); > - printk(KERN_ERR > - "cpm_uart_cpm.c: could not allocate coherent memory\n"); > + pr_err("cpm_uart_cpm.c: could not allocate coherent memory\n"); > return -ENOMEM; > } > >
On 29.04.19 17:59, Christophe Leroy wrote: > If we want to do something useful, wouldn't it make more sense to > introduce the use of dev_err() in order to identify the faulting device > in the message ? Well, I could get the struct device* pointer via pinfo.port->dev, but I wasn't entirely sure that it's always defined before these functions could be called. Shall I change it to dev_*() ? --mtx
On Mon, Apr 29, 2019 at 05:59:04PM +0200, Christophe Leroy wrote: > Le 27/04/2019 à 14:52, Enrico Weigelt, metux IT consult a écrit : > > Fix checkpatch warnings by using pr_err(): > > > > WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ... > > #109: FILE: drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c:109: > > + printk(KERN_ERR > > > > WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ... > > #128: FILE: drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c:128: > > + printk(KERN_ERR > > > > WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ... > > + printk(KERN_ERR > > > > WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ... > > + printk(KERN_ERR > > > > Signed-off-by: Enrico Weigelt <info@metux.net> > > Reviewed-by: Christophe Leroy <christophe.leroy@c-s.fr> > > But is that really worth doing those changes ? > > If we want to do something useful, wouldn't it make more sense to introduce > the use of dev_err() in order to identify the faulting device in the message > ? +1 for switching to dev_*().
diff --git a/drivers/tty/serial/cpm_uart/cpm_uart_cpm1.c b/drivers/tty/serial/cpm_uart/cpm_uart_cpm1.c index 56fc527..aed61e9 100644 --- a/drivers/tty/serial/cpm_uart/cpm_uart_cpm1.c +++ b/drivers/tty/serial/cpm_uart/cpm_uart_cpm1.c @@ -71,8 +71,7 @@ int cpm_uart_allocbuf(struct uart_cpm_port *pinfo, unsigned int is_con) dpmemsz = sizeof(cbd_t) * (pinfo->rx_nrfifos + pinfo->tx_nrfifos); dp_offset = cpm_dpalloc(dpmemsz, 8); if (IS_ERR_VALUE(dp_offset)) { - printk(KERN_ERR - "cpm_uart_cpm1.c: could not allocate buffer descriptors\n"); + pr_err("cpm_uart_cpm1.c: could not allocate buffer descriptors\n"); return -ENOMEM; } dp_mem = cpm_dpram_addr(dp_offset); @@ -90,8 +89,7 @@ int cpm_uart_allocbuf(struct uart_cpm_port *pinfo, unsigned int is_con) if (mem_addr == NULL) { cpm_dpfree(dp_offset); - printk(KERN_ERR - "cpm_uart_cpm1.c: could not allocate coherent memory\n"); + pr_err("cpm_uart_cpm1.c: could not allocate coherent memory\n"); return -ENOMEM; } diff --git a/drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c b/drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c index 40cfcf4..a0fccda 100644 --- a/drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c +++ b/drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c @@ -106,8 +106,7 @@ int cpm_uart_allocbuf(struct uart_cpm_port *pinfo, unsigned int is_con) dpmemsz = sizeof(cbd_t) * (pinfo->rx_nrfifos + pinfo->tx_nrfifos); dp_offset = cpm_dpalloc(dpmemsz, 8); if (IS_ERR_VALUE(dp_offset)) { - printk(KERN_ERR - "cpm_uart_cpm.c: could not allocate buffer descriptors\n"); + pr_err("cpm_uart_cpm.c: could not allocate buffer descriptors\n"); return -ENOMEM; } @@ -125,8 +124,7 @@ int cpm_uart_allocbuf(struct uart_cpm_port *pinfo, unsigned int is_con) if (mem_addr == NULL) { cpm_dpfree(dp_offset); - printk(KERN_ERR - "cpm_uart_cpm.c: could not allocate coherent memory\n"); + pr_err("cpm_uart_cpm.c: could not allocate coherent memory\n"); return -ENOMEM; }
Fix checkpatch warnings by using pr_err(): WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ... #109: FILE: drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c:109: + printk(KERN_ERR WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ... #128: FILE: drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c:128: + printk(KERN_ERR WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ... + printk(KERN_ERR WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ... + printk(KERN_ERR Signed-off-by: Enrico Weigelt <info@metux.net> --- drivers/tty/serial/cpm_uart/cpm_uart_cpm1.c | 6 ++---- drivers/tty/serial/cpm_uart/cpm_uart_cpm2.c | 6 ++---- 2 files changed, 4 insertions(+), 8 deletions(-)