| Message ID | 226fc2b1-ecb9-19d9-d7f0-2c6bc0fc8edf@omp.ru |
|---|---|
| State | New |
| Headers | show |
| Series | ata: libata-core: fix sloppy parameter type in ata_exec_internal[_sg]() | expand |
On 6/14/22 03:47, Sergey Shtylyov wrote: > Make the 'timeout' parameter to ata_exec_internal_sg() *unsigned int* as > msecs_to_jiffies() that it calls takes just *unsigned int* for the time in > milliseconds. Then follow the suit with ata_exec_internal(), its only > caller; also fix up ata_dev_set_feature(), the only ata_exec_internal()'s > caller that explicitly passes *unsigned long* variable for timeout... > > Found by Linux Verification Center (linuxtesting.org) with the SVACE static > analysis tool. Since you are changing this function signature, can you also make it static since it is only used in libata-core.c ? The declaration in drivers/ata/libata.h is useless. > > Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru> > > --- > This patch is against the 'for-next' branch of Damien's 'libata.git' repo. > > drivers/ata/libata-core.c | 6 +++--- > drivers/ata/libata.h | 4 ++-- > 2 files changed, 5 insertions(+), 5 deletions(-) > > Index: libata/drivers/ata/libata-core.c > =================================================================== > --- libata.orig/drivers/ata/libata-core.c > +++ libata/drivers/ata/libata-core.c > @@ -1472,7 +1472,7 @@ static void ata_qc_complete_internal(str > unsigned ata_exec_internal_sg(struct ata_device *dev, > struct ata_taskfile *tf, const u8 *cdb, > int dma_dir, struct scatterlist *sgl, > - unsigned int n_elem, unsigned long timeout) > + unsigned int n_elem, unsigned int timeout) > { > struct ata_link *link = dev->link; > struct ata_port *ap = link->ap; > @@ -1647,7 +1647,7 @@ unsigned ata_exec_internal_sg(struct ata > unsigned ata_exec_internal(struct ata_device *dev, > struct ata_taskfile *tf, const u8 *cdb, > int dma_dir, void *buf, unsigned int buflen, > - unsigned long timeout) > + unsigned int timeout) > { > struct scatterlist *psg = NULL, sg; > unsigned int n_elem = 0; > @@ -4339,7 +4339,7 @@ unsigned int ata_dev_set_feature(struct > { > struct ata_taskfile tf; > unsigned int err_mask; > - unsigned long timeout = 0; > + unsigned int timeout = 0; > > /* set up set-features taskfile */ > ata_dev_dbg(dev, "set features - SATA features\n"); > Index: libata/drivers/ata/libata.h > =================================================================== > --- libata.orig/drivers/ata/libata.h > +++ libata/drivers/ata/libata.h > @@ -52,11 +52,11 @@ extern u64 ata_tf_read_block(const struc > extern unsigned ata_exec_internal(struct ata_device *dev, > struct ata_taskfile *tf, const u8 *cdb, > int dma_dir, void *buf, unsigned int buflen, > - unsigned long timeout); > + unsigned int timeout); > extern unsigned ata_exec_internal_sg(struct ata_device *dev, > struct ata_taskfile *tf, const u8 *cdb, > int dma_dir, struct scatterlist *sg, > - unsigned int n_elem, unsigned long timeout); > + unsigned int n_elem, unsigned int timeout); > extern int ata_wait_ready(struct ata_link *link, unsigned long deadline, > int (*check_ready)(struct ata_link *link)); > extern int ata_dev_read_id(struct ata_device *dev, unsigned int *p_class,
On 6/14/22 8:24 AM, Damien Le Moal wrote: [...] >> Make the 'timeout' parameter to ata_exec_internal_sg() *unsigned int* as >> msecs_to_jiffies() that it calls takes just *unsigned int* for the time in >> milliseconds. Then follow the suit with ata_exec_internal(), its only >> caller; also fix up ata_dev_set_feature(), the only ata_exec_internal()'s >> caller that explicitly passes *unsigned long* variable for timeout... >> >> Found by Linux Verification Center (linuxtesting.org) with the SVACE static >> analysis tool. > > Since you are changing this function signature, can you also make it > static since it is only used in libata-core.c ? The declaration in > drivers/ata/libata.h is useless. Hopefully you don't mean I should do it in the same patch? :-) [...] MBR, Sergey
On 6/15/22 03:43, Sergey Shtylyov wrote: > On 6/14/22 8:24 AM, Damien Le Moal wrote: > [...] >>> Make the 'timeout' parameter to ata_exec_internal_sg() *unsigned int* as >>> msecs_to_jiffies() that it calls takes just *unsigned int* for the time in >>> milliseconds. Then follow the suit with ata_exec_internal(), its only >>> caller; also fix up ata_dev_set_feature(), the only ata_exec_internal()'s >>> caller that explicitly passes *unsigned long* variable for timeout... >>> >>> Found by Linux Verification Center (linuxtesting.org) with the SVACE static >>> analysis tool. >> >> Since you are changing this function signature, can you also make it >> static since it is only used in libata-core.c ? The declaration in >> drivers/ata/libata.h is useless. > > Hopefully you don't mean I should do it in the same patch? :-) The patch you sent is changing the function signature. So yes, doing it in the same patch is fine with me. I do not see the need for 2 patches for that. Instead of "fix sloppy parameter type in ata_exec_internal_sg", rename the patch "fix ata_exec_internal_sg signature" and then all changes naturally belong to the same patch. > > [...] > > MBR, Sergey
Hello! Replying from my Gmail address, the corporate IMAP server is still AWOL... :-( Not sure what happened to the subject: "[_sg]" somehow got moved before "fix"... On 6/15/22 2:46 AM, Damien Le Moal wrote: [...] >>>> Make the 'timeout' parameter to ata_exec_internal_sg() *unsigned int* as >>>> msecs_to_jiffies() that it calls takes just *unsigned int* for the time in >>>> milliseconds. Then follow the suit with ata_exec_internal(), its only >>>> caller; also fix up ata_dev_set_feature(), the only ata_exec_internal()'s >>>> caller that explicitly passes *unsigned long* variable for timeout... >>>> >>>> Found by Linux Verification Center (linuxtesting.org) with the SVACE static >>>> analysis tool. >>> >>> Since you are changing this function signature, can you also make it >>> static since it is only used in libata-core.c ? The declaration in >>> drivers/ata/libata.h is useless. >> >> Hopefully you don't mean I should do it in the same patch? :-) > > The patch you sent is changing the function signature. Actually, it changes 2 function signatures... > So yes, doing it in the same patch is fine with me. But not with me. Adding a "drove by" signature change doesn't go together well with the "found by SVACE" note, to start with... Moreover, I've created 2 patches already, with the 1st patch making ata_exec_internal_sg() *static*... > I do not see the need for 2 patches for that. This patch as is already does enough to violate the "do one thing per patch" rule, to add even more violations. :-) > Instead of "fix sloppy parameter type in ata_exec_internal_sg", > rename the patch "fix ata_exec_internal_sg signature" and then all changes > naturally belong to the same patch. Thanks, no -- I'd like to keep the emphasis on "sloppy parameter type". MBR, Sergey
Index: libata/drivers/ata/libata-core.c =================================================================== --- libata.orig/drivers/ata/libata-core.c +++ libata/drivers/ata/libata-core.c @@ -1472,7 +1472,7 @@ static void ata_qc_complete_internal(str unsigned ata_exec_internal_sg(struct ata_device *dev, struct ata_taskfile *tf, const u8 *cdb, int dma_dir, struct scatterlist *sgl, - unsigned int n_elem, unsigned long timeout) + unsigned int n_elem, unsigned int timeout) { struct ata_link *link = dev->link; struct ata_port *ap = link->ap; @@ -1647,7 +1647,7 @@ unsigned ata_exec_internal_sg(struct ata unsigned ata_exec_internal(struct ata_device *dev, struct ata_taskfile *tf, const u8 *cdb, int dma_dir, void *buf, unsigned int buflen, - unsigned long timeout) + unsigned int timeout) { struct scatterlist *psg = NULL, sg; unsigned int n_elem = 0; @@ -4339,7 +4339,7 @@ unsigned int ata_dev_set_feature(struct { struct ata_taskfile tf; unsigned int err_mask; - unsigned long timeout = 0; + unsigned int timeout = 0; /* set up set-features taskfile */ ata_dev_dbg(dev, "set features - SATA features\n"); Index: libata/drivers/ata/libata.h =================================================================== --- libata.orig/drivers/ata/libata.h +++ libata/drivers/ata/libata.h @@ -52,11 +52,11 @@ extern u64 ata_tf_read_block(const struc extern unsigned ata_exec_internal(struct ata_device *dev, struct ata_taskfile *tf, const u8 *cdb, int dma_dir, void *buf, unsigned int buflen, - unsigned long timeout); + unsigned int timeout); extern unsigned ata_exec_internal_sg(struct ata_device *dev, struct ata_taskfile *tf, const u8 *cdb, int dma_dir, struct scatterlist *sg, - unsigned int n_elem, unsigned long timeout); + unsigned int n_elem, unsigned int timeout); extern int ata_wait_ready(struct ata_link *link, unsigned long deadline, int (*check_ready)(struct ata_link *link)); extern int ata_dev_read_id(struct ata_device *dev, unsigned int *p_class,
Make the 'timeout' parameter to ata_exec_internal_sg() *unsigned int* as msecs_to_jiffies() that it calls takes just *unsigned int* for the time in milliseconds. Then follow the suit with ata_exec_internal(), its only caller; also fix up ata_dev_set_feature(), the only ata_exec_internal()'s caller that explicitly passes *unsigned long* variable for timeout... Found by Linux Verification Center (linuxtesting.org) with the SVACE static analysis tool. Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru> --- This patch is against the 'for-next' branch of Damien's 'libata.git' repo. drivers/ata/libata-core.c | 6 +++--- drivers/ata/libata.h | 4 ++-- 2 files changed, 5 insertions(+), 5 deletions(-)