PATCH: fs/proc/task_mmu.c Kernel2.6.0-test9: to add physical memory info in /proc/PID/status
From: edjard
Date: Fri Nov 21 2003 - 16:11:07 EST
Hi There,
We needed to add detailed info at /proc/PID/status of a given
process PID. Here is how we did it. Below is an example of output
followed by the pattch for /proc/PID/status for 2.6.0-test9.
TIA,
BR,
Edjard
-----------------------------------------------
10LE/INdT - Labaratorio de Linux
JOSE - Jungle Open Source dEvelopers
Manaus, Amazon - Brazil
-----------------------------------------------
Here is an output of a new /proc/PID/stat
Originally the command 'cat /proc/PID/status' does not provide the
information in physical memory of the process with PID. As you can see in
the example below physical memory allocation is detailed.
>
> $ cat /proc/1718/status
> Name: opera
> State: S (sleeping)
> SleepAVG: 101%
> Tgid: 1718
> Pid: 1718
> PPid: 1631
> TracerPid: 0
> Uid: 500 500 500 500
> Gid: 501 501 501 501
> FDSize: 256
> Groups: 501
> VmSize: 35792 kB
> VmLck: kB
> VmRSS: 21156 kB
> VmData: 5920 kB
> PhysicalData: 5556 kB
> VmStk: 68 kB
> PhysicalStk: 40 kB
> VmExe: 5124 kB
> PhysicalExe: 3408 kB
> VmLib: 19244 kB
> PhysicalLib: 10240 kB
> HeapSize: 5184 KB
> PhysicalHeap: 5124 KB
> Threads: 1
> SigPnd: 0000000000000000
> ShdPnd: 0000000000000000
> SigBlk: 0000000100000000
> SigIgn: 8000000010001000
> SigCgt: 0000000080010000
> CapInh: 0000000000000000
> CapPrm: 0000000000000000
> CapEff: 0000000000000000
>
-------------------------------------------------------------------------
--- linux/fs/proc/task_mmu.c 2003-10-25 14:43:00.000000000 -0400
+++ linux/fs/proc/task_mmu.c-10LE 2003-11-21 13:51:42.000000000
-0400 @@
-3,44 +3,87 @@
#include <linux/seq_file.h>
#include <asm/uaccess.h>
+
+void phys_size_10LE(struct mm_struct *mm, unsigned long start_address,
unsigned long end_address, unsigned long *size) {
+ pgd_t *my_pgd;
+ pmd_t *my_pmd;
+ pte_t *my_pte;
+ unsigned long page;
+
+ for (page = start_address; page < end_address; page += PAGE_SIZE) { +
my_pgd = pgd_offset(mm, page);
+ if (pgd_none(*my_pgd) || pgd_bad(*my_pgd)) break;
+ my_pmd = pmd_offset(my_pgd, page);
+ if (pmd_none(*my_pmd) || pmd_bad(*my_pmd)) break;
+ my_pte = pte_offset_map(my_pmd, page);
+ if (pte_present(*my_pte)) {
+ *size += PAGE_SIZE;
+ }
+ }
+}
+
char *task_mem(struct mm_struct *mm, char *buffer)
{
- unsigned long data = 0, stack = 0, exec = 0, lib = 0;
- struct vm_area_struct *vma;
-
- down_read(&mm->mmap_sem);
- for (vma = mm->mmap; vma; vma = vma->vm_next) {
- unsigned long len = (vma->vm_end - vma->vm_start) >> 10;
- if (!vma->vm_file) {
- data += len;
- if (vma->vm_flags & VM_GROWSDOWN)
- stack += len;
- continue;
- }
- if (vma->vm_flags & VM_WRITE)
- continue;
- if (vma->vm_flags & VM_EXEC) {
- exec += len;
- if (vma->vm_flags & VM_EXECUTABLE)
- continue;
- lib += len;
- }
- }
- buffer += sprintf(buffer,
- "VmSize:\t%8lu kB\n"
- "VmLck:\t%8lu kB\n"
- "VmRSS:\t%8lu kB\n"
- "VmData:\t%8lu kB\n"
- "VmStk:\t%8lu kB\n"
- "VmExe:\t%8lu kB\n"
- "VmLib:\t%8lu kB\n",
- mm->total_vm << (PAGE_SHIFT-10),
- mm->locked_vm << (PAGE_SHIFT-10),
- mm->rss << (PAGE_SHIFT-10),
- data - stack, stack,
- exec - lib, lib);
- up_read(&mm->mmap_sem);
- return buffer;
+ unsigned long data = 0, stack = 0, exec = 0, lib = 0;
+ unsigned long phys_data = 0, phys_stack = 0, phys_exec = 0, phys_lib =
0; + unsigned long phys_brk = 0;
+ struct vm_area_struct *vma;
+ down_read(&mm->mmap_sem);
+ for (vma = mm->mmap; vma; vma = vma->vm_next) {
+ unsigned long len = (vma->vm_end - vma->vm_start) >> 10;
+
+ if (!vma->vm_file) {
+ phys_size_10LE(mm, vma->vm_start, vma->vm_end, &phys_data); +
if (vma->vm_flags & VM_GROWSDOWN) {
+ stack += len;
+ phys_size_10LE(mm, vma->vm_start, vma->vm_end, &phys_stack);
+ }
+ else {
+ data += len;
+ }
+ continue;
+ }
+
+ if (vma->vm_flags & VM_WRITE)
+ continue;
+
+ if (vma->vm_flags & VM_EXEC) {
+ exec += len;
+ phys_size_10LE(mm, vma->vm_start, vma->vm_end, &phys_exec); +
if (vma->vm_flags & VM_EXECUTABLE) {
+ continue;
+ }
+ lib += len;
+ phys_size_10LE(mm, vma->vm_start, vma->vm_end, &phys_lib);
+ }
+ }
+ phys_size_10LE(mm, mm->start_brk, mm->brk, &phys_brk);
+ buffer += sprintf(buffer,
+ "VmSize:\t%8lu kB\n"
+ "VmLck:\t%8lu kB\n"
+ "VmRSS:\t%8lu kB\n"
+ "VmData:\t%8lu kB\n"
+ "PhysicalData:\t%8lu kB\n"
+ "VmStk:\t%8lu kB\n"
+ "PhysicalStk:\t%8lu kB\n"
+ "VmExe:\t%8lu kB\n"
+ "PhysicalExe:\t%8lu kB\n"
+ "VmLib:\t%8lu kB\n"
+ "PhysicalLib:\t%8lu kB\n"
+ "VmHeap: \t%8lu KB\n"
+ "PhysicalHeap: \t%8lu KB\n",
+ mm->total_vm << (PAGE_SHIFT-10),
+ mm->locked_vm << (PAGE_SHIFT-10),
+ mm->rss << (PAGE_SHIFT-10),
+ data, (phys_data - phys_stack) >> 10,
+ stack, phys_stack >> 10,
+ exec - lib, (phys_exec - phys_lib) >> 10,
+ lib, phys_lib >> 10,
+ (mm->brk - mm->start_brk) >> 10,
+ phys_brk >> 10
+ );
+ up_read(&mm->mmap_sem);
+ return buffer;
}
unsigned long task_vsize(struct mm_struct *mm)
--------------------- END OF PATCH -----------------------------
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/