Mapping IO memory read-write
I'm writing a device driver for a memory-mapped device on x86-64. I'm mapping the device in the kernel using ioremap_cache(). My file_operations.mmap function is as follows:
static int dev_mmap(struct file *filep, struct vm_area_struct *vma) {
vma->vm_page_prot.pgprot|=_PAGE_BIT_RW;
return vm_iomap_memory(vma,START,LEN);
}The user process calls mmap(..,PROT_READ|PROT_WRITE|PROT_EXEC,..). The mapping in /proc/[pid]/maps shows the write bit. However, when looking at the actual page table entry does not have the RW bit set. For example:
virtual address = 0000_0001_8000_0000
cr3 = 0000_0000_700a_e000
phys:0000_0000_700a_e000 = 0000_0000_7e05_5067 (P|RW|US|A|D) phys:0000_0000_7e05_5030 = 0000_0001_66e2_9067 (P|RW|US|A|D) phys:0000_0001_66e2_9000 = 0000_0001_69db_3067 (P|RW|US|A|D) phys:0000_0001_69db_3000 = 0000_0000_8020_0225 (P|US|A|SOFTW1)
Am I doing something wrong?
Jethro
In this case, have you tried reading and writing to the memory segment being mmap'd from userland?
Here's an example mmap'ing device driver if you need to see that:
https://github.com/claudioscordino/mmap_alloc
Quoted 24 lines · Jethro Beekman
I'm writing a device driver for a memory-mapped device on x86-64. I'm mapping
the device in the kernel usingioremap_cache(). My file_operations.mmap function
is as follows:static int dev_mmap(struct file *filep, struct vm_area_struct *vma) { vma->vm_page_prot.pgprot|=_PAGE_BIT_RW; return vm_iomap_memory(vma,START,LEN); }The user process calls
mmap(..,PROT_READ|PROT_WRITE|PROT_EXEC,..). The mapping
in /proc/[pid]/maps shows the write bit. However, when looking at the actual
page table entry does not have the RW bit set. For example:virtual address = 0000_0001_8000_0000
cr3 = 0000_0000_700a_e000phys:0000_0000_700a_e000 = 0000_0000_7e05_5067 (P|RW|US|A|D) phys:0000_0000_7e05_5030 = 0000_0001_66e2_9067 (P|RW|US|A|D) phys:0000_0001_66e2_9000 = 0000_0001_69db_3067 (P|RW|US|A|D) phys:0000_0001_69db_3000 = 0000_0000_8020_0225 (P|US|A|SOFTW1)
Quoted 9 lines
Am I doing something wrong?
Jethro
_______________________________________________
Kernelnewbies mailing list
Kernelnewbies at kernelnewbies.org
http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.kernelnewbies.org/pipermail/kernelnewbies/attachments/20160113/5cec8b47/attachment.html
Quoted 5 lines
In this case, have you tried reading and writing to the memory segment being mmap'd from userland?
Here's an example mmap'ing device driver if you need to see that:
https://github.com/claudioscordino/mmap_alloc
That would not have worked, but your example code did give me an idea. I was using MAP_PRIVATE while I should've been using MAP_SHARED. It works now. Thanks!
Jethro
Quoted 39 lines · Kenneth Adam Miller
In this case, have you tried reading and writing to the memory segment being mmap'd from userland?
Here's an example mmap'ing device driver if you need to see that:
https://github.com/claudioscordino/mmap_allocOn Tue, Jan 12, 2016 at 6:10 PM, Jethro Beekman <[email protected] <mailto:[email protected]>> wrote:
I'm writing a device driver for a memory-mapped device on x86-64. I'm mapping the device in the kernel using ioremap_cache(). My file_operations.mmap function is as follows: static int dev_mmap(struct file *filep, struct vm_area_struct *vma) { vma->vm_page_prot.pgprot|=_PAGE_BIT_RW; return vm_iomap_memory(vma,START,LEN); } The user process calls mmap(..,PROT_READ|PROT_WRITE|PROT_EXEC,..). The mapping in /proc/[pid]/maps shows the write bit. However, when looking at the actual page table entry does not have the RW bit set. For example: virtual address = 0000_0001_8000_0000 cr3 = 0000_0000_700a_e000 phys:0000_0000_700a_e000 = 0000_0000_7e05_5067 (P|RW|US|A|D) phys:0000_0000_7e05_5030 = 0000_0001_66e2_9067 (P|RW|US|A|D) phys:0000_0001_66e2_9000 = 0000_0001_69db_3067 (P|RW|US|A|D) phys:0000_0001_69db_3000 = 0000_0000_8020_0225 (P|US|A|SOFTW1) Am I doing something wrong? Jethro _______________________________________________ Kernelnewbies mailing list Kernelnewbies at kernelnewbies.org <mailto:[email protected]> http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies