Messages in this thread | | | Subject | Re: [PATCH 1/2] GenWQE: Fix endian issues detected by sparse | From | Frank Haverkamp <> | Date | Tue, 07 Jan 2014 13:30:16 +0100 |
| |
Hi Dan,
Am Dienstag, den 07.01.2014, 09:41 +0300 schrieb Dan Carpenter: > Could you also fix this Smatch warning? > drivers/misc/genwqe/card_dev.c:658 do_flash_update() warn: maybe return -EFAULT instead of the bytes remaining?
I thought i fixed this already in my posting: [PATCH] GenWQE: Rework return code for flash-update ioctl from 22.12.2013 14:16:36:
Here the spot:
@@ -565,14 +552,13 @@ static int do_flash_update(struct genwqe
rc = copy_from_user(xbuf, buf, tocopy); if (rc) { - dev_err(&pci_dev->dev, - "err: could not copy all data rc=%d\n", rc); + rc = -EFAULT; goto free_buffer; } crc = genwqe_crc32(xbuf, tocopy, 0xffffffff);
- dev_info(&pci_dev->dev, - "[%s] DMA: 0x%llx CRC: %08x SZ: %ld %d\n", + dev_dbg(&pci_dev->dev, + "[%s] DMA: 0x%llx CRC: %08x SZ: %ld %d\n", __func__, dma_addr, crc, tocopy, blocks_to_flash);
> > Also we shouldn't be doing dev_err() on copy_to/from_user() problems. > The user can trigger those and flood dmesg. It is a DoS (annoying). > > regards, > dan carpenter >
In this patch I also removed most of the dev_err() messages. Fixing the possibility to overflow the logs with error messages from user-side. Greg did not pick up this patch yet, as far as I can see.
Regards
Frank
| |