From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from relay.sgi.com (relay1.corp.sgi.com [137.38.102.111]) by oss.sgi.com (Postfix) with ESMTP id 640057F47 for ; Sat, 13 Sep 2014 12:07:59 -0500 (CDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by relay1.corp.sgi.com (Postfix) with ESMTP id 3AAFB8F804B for ; Sat, 13 Sep 2014 10:07:56 -0700 (PDT) Received: from mail-ie0-f170.google.com (mail-ie0-f170.google.com [209.85.223.170]) by cuda.sgi.com with ESMTP id uGCxBpAQc81BRUst (version=TLSv1 cipher=RC4-SHA bits=128 verify=NO) for ; Sat, 13 Sep 2014 10:07:54 -0700 (PDT) Received: by mail-ie0-f170.google.com with SMTP id tp5so2628331ieb.29 for ; Sat, 13 Sep 2014 10:07:53 -0700 (PDT) MIME-Version: 1.0 Date: Sat, 13 Sep 2014 22:37:53 +0530 Message-ID: Subject: [RFD] xfs_fsr: Doubts related to xfs_fsr code From: Somdeep Dey List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============6503212041451765132==" Errors-To: xfs-bounces@oss.sgi.com Sender: xfs-bounces@oss.sgi.com To: xfs@oss.sgi.com --===============6503212041451765132== Content-Type: multipart/alternative; boundary=90e6ba6e8fb83e236d0502f57163 --90e6ba6e8fb83e236d0502f57163 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Hi, While studying the code and attempting to understand it, we have come up against certain doubts that have us in a slight fix. We've included the concerned sections of the code below, along with our specific problem in each of these sections. A little pointer in the right direction would be of great help. (Source file : fsr_xfs_fsr.c) ****** Doubt number 1 ****** (line 331 onwards) int main() if (optind < argc) // If the command line input contains the XFS //filesystem name / file on which xfs_fsr needs to be run on { for (; optind < argc; optind++) { argname =3D argv[optind]; // save target which can be file or filesystem if (lstat64(argname, &sb) < 0) { /* This system call returns a stat64 structure, and thus sets * all fields in it. * On success, zero is returned. On error, -1 is returned, * and errno is set appropriately. */ fprintf(stderr, _("%s: could not stat: %s: %s\n"), progname, argname, strerror(errno)); continue; } // POSIX macros are defined to check the file type using the st_mode field if (S_ISLNK(sb.st_mode)) // Check if path(argname) is a //symbolic link, if so link will be stat-ed and not file {// Hence we run stat64() and save the obtained stat structure struct stat64 sb2; if (stat64(argname, &sb2) =3D=3D 0 && (S_ISBLK(sb2.st_mode) || S_ISCHR(sb2.st_mode))) sb =3D sb2; // check if stat is a success and //if argname(path) is block device ? OR is character device? } ___________________________________________________________________________= _____ We understand that lstat64() and stat64() are used to see if target (file/filesystem) can be stated and if yes then the structure is saved. But we couldn=E2=80=99t exactly understand its use and why both functions a= re used separately. Usually the error could not stat: filename : is followed by Permission denied. Is this related to the root permissions i.e. accessibility ? ****** Doubt number 2 ****** (line 184 onwards) static char * find_mountpoint(char *mtab, char *argname, struct stat64 *sb) while ((t =3D getmntent(mtabp))) { if (S_ISDIR(sb->st_mode)) { /* mount point */ if (stat64(t->mnt_dir, &ms) < 0) continue; if (sb->st_ino !=3D ms.st_ino) continue; if (sb->st_dev !=3D ms.st_dev) continue; if (strcmp(t->mnt_type, MNTTYPE_XFS) !=3D 0) continue; } else { /* device */ struct stat64 sb2; if (stat64(t->mnt_fsname, &ms) < 0) continue; if (sb->st_rdev !=3D ms.st_rdev) continue; if (strcmp(t->mnt_type, MNTTYPE_XFS) !=3D 0) continue; /* * Make sure the mount point given by mtab is accessible * before using it. */ if (stat64(t->mnt_dir, &sb2) < 0) continue; } ___________________________________________________________________________= _____ We just wanted to confirm if the basic working of the function is 1) To obtain a mount table pointer to the mount table (/etc/mtab/ or /proc/mounts). 2) For each entry in the mount table check if it is a directory or device and after performing various comparisons (checks) =E2=80=93 (could you please elabor= ate on the checks performed), this function returns a pointer to the entry. ****** Doubt number 3 ****** (line 677 onwards) static int fsrfs(char *mntdir, xfs_ino_t startino, int targetrange) For the following __s32 buflenout; fshandlep =3D jdm_getfshandle( mntdir ); if ( ! fshandlep ) { fsrprintf(_("unable to get handle: %s: %s\n"), mntdir, strerror( errno )); return -1; } ___________________________________________________________________________= _____ We are a bit confused about what is exactly the file handle being returned by jdm_getfshandle(). Also what exactly is buflenout. Is it a structure field ? ___________________________________________________________________________= _____ while ((ret =3D xfs_bulkstat(fsfd,&lastino, GRABSZ, &buf[0], &buflenout) = =3D=3D 0)) { xfs_bstat_t *p; xfs_bstat_t *endp; if (buflenout =3D=3D 0) goto out0; /* Each loop through, defrag targetrange percent of the files */ count =3D (buflenout * targetrange) / 100; qsort((char *)buf, buflenout, sizeof(struct xfs_bstat), cmp); ___________________________________________________________________________= _____ In the above code snippet we understand that the while loop will run for N(10) passes and defragment top 10% of the defragmented files. However we would appreciate if you could further explain the functions: (ret =3D xfs_bulkstat(fsfd,&lastino, GRABSZ, &buf[0], &buflenout) =3D=3D 0) qsort((char *)buf, buflenout, sizeof(struct xfs_bstat), cmp); We know that the sort function will be used to sort extents based on size and then offset, but a bit more information on how it is exactly working will be really appreciated, as we believe that this sort() has some other purpose. ___________________________________________________________________________= _____ For this mail we have listed only a limited number of doubts that we think are pressing. Based on further explanations that we might receive from you, we will send out another mail for the doubts that still linger. Regards, A-DRS. --90e6ba6e8fb83e236d0502f57163 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Hi,<= /div>

While = studying the code and attempting to understand it, we have come up against = certain doubts that have us in a slight fix. We've included the concern= ed
=C2=A0sections of the code below, along = with our specific problem in each of these=C2=A0
=C2=A0sections. A little pointer in the right direction would be of g= reat help.

(Source file : fsr_xfs_fsr.c)

****** Doubt number 1 ******
(line 331 onwards)
int mai= n()=C2=A0

if (optind < argc) // If the command line input contains the XFS=C2=A0
//filesystem name / file on which xfs_fsr needs to be run on
{
for (; optind < argc; optind++)=C2=A0
{
a= rgname =3D argv[optind]; = // save target which can be file or filesystem
if (lstat64(argname, &sb) < 0)
{ /* This system call returns a stat64 struc= ture, and thus sets =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 * all f= ields in it.
* On success, zero is returned. =C2=A0On error, -1 = is returned,=C2=A0
=C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 * and =C2=A0errno =C2=A0is set appropriatel= y.
*/
fprintf(stderr,
<= span class=3D"" style=3D"white-space:pre"> _("%s: could not s= tat: %s: %s\n"),
progname, argname, strerror(errno));
<= div class=3D"gmail_extra"> continue;
}=
// POSIX macros are defined to check the file type using the st_m= ode field
if (S_ISLNK(sb.st_mode)) // Check if path(argname) is a=C2=A0
//symboli= c link, if so link will be stat-ed and not file
{// Hence =C2=A0we= run stat64() and save the obtained stat structure
struct stat64 = sb2;
if (stat64(argname, &sb2) =3D=3D 0 && (S_ISBLK(s= b2.st_mode) || = =C2=A0=C2=A0
=C2=A0 =C2=A0 S_ISCHR(sb2.st_mode)))
sb= =3D sb2; // check if sta= t is a success and=C2=A0
//if argname(path) is block device ? O= R is character device? <= /div>
= }
________________________________= ________________________________________________

We understand that lstat64() and= stat64() are used to see if target
=C2=A0(= file/filesystem) can be stated and if yes then the structure is saved.=C2= =A0
But we couldn=E2=80=99t exactly underst= and its use and why both functions are used separately.
Usually the error could not stat: filename : is followed by Pe= rmission denied.
=C2=A0Is this related to t= he root permissions i.e. accessibility ?
****** Doubt number 2 ******

(line 184 onwards= )
static char *
find_mountpoint(char *mtab, char *argname, struct stat64 *sb)

while ((t =3D = getmntent(mtabp))) {
if (S_ISDIR(sb->st_mode)) { /* mount point */
if (stat64= (t->mnt_dir, &ms) < 0)
continue;
if (sb->= st_ino !=3D ms.st_ino)
continue;
if (sb->st_dev != =3D ms.st_dev)
continue;
if (strcmp(t->mnt_type, MN= TTYPE_XFS) !=3D 0)
continue;
<= span class=3D"" style=3D"white-space:pre"> } else { /* device */
struct stat64 = sb2;

<= span class=3D"" style=3D"white-space:pre"> if (stat64(t->mnt_fs= name, &ms) < 0)
continue;
if (sb->st_rdev != =3D ms.st_rdev)
continue;
if (strcmp(t->mnt_type, M= NTTYPE_XFS) !=3D 0)
continue;
<= br>
/*
* Make sure the mount point given by mtab is ac= cessible
* before using it.
*/
if (stat64(t= ->mnt_dir, &sb2) < 0)
continue;
}
______________________________________________________= __________________________
We just wanted t= o confirm if the basic working of the function is
1) To obtain a mount table pointer to the mount table (/etc/mtab/ or= /proc/mounts).
2) =C2=A0For each entry in = the mount table check if it is a directory or device and after
=C2=A0performing various comparisons (checks) =E2=80=93= (could you please elaborate on the checks
= =C2=A0 performed), this function returns a pointer to the entry.=C2=A0

****** Dou= bt number 3 ******

(line 677 onwards)
static int<= /div>
fsrfs(char *mntdir, xfs_ino_t startino, int= targetrange)

For the following=C2=A0

<= div class=3D"gmail_extra">__s32 = buflenout;

fshandlep = =3D jdm_getfshandle( mntdir );
if ( ! fshandlep ) {
fsrpr= intf(_("unable to get handle: %s: %s\n"),
=C2=A0 =C2=A0= =C2=A0 =C2=A0 =C2=A0mntdir, strerror( errno ));
return -1;
<= div class=3D"gmail_extra"> }
_______________________________________= _________________________________________
<= br>
We are a bit confused about what is exa= ctly the file handle being returned
=C2=A0b= y jdm_getfshandle().
Also what exactly is b= uflenout. Is it a structure field ?
_______= _________________________________________________________________________

while (= (ret =3D xfs_bulkstat(fsfd,&lastino, GRABSZ, &buf[0], &buflenou= t) =3D=3D 0))
{
xfs_bstat_t *p;
xfs_bstat_t *end= p;

if (buflenout =3D=3D 0)
= goto out0;

/* Each lo= op through, defrag targetrange percent of the files */
count =3D (= buflenout * targetrange) / 100;

<= div class=3D"gmail_extra"> qsort((char *)buf, buflenout, sizeof(struct xfs_bstat), cmp);
_____________________________________________________= ___________________________

In the above code snippet we understand that the whil= e loop will run for N(10)=C2=A0
passes and = defragment top 10% of the defragmented files.
However we would appreciate if you could further explain the functions:<= /div>

(ret = =3D xfs_bulkstat(fsfd,&lastino, GRABSZ, &buf[0], &buflenout) = =3D=3D 0)

qsort((char *)buf, buflenout, sizeof(struct xfs_bstat), cmp);

We know that th= e sort function will be used to sort extents based on size and then=C2=A0
offset, but a bit more information on how it= is exactly working will be really=C2=A0
ap= preciated, as we believe that this sort() has some other purpose.
____________________________________________________= ____________________________



For this mail we have listed only a limited number of= doubts that we think are pressing. Based on further explanations that we m= ight receive from you, we will send out another =C2=A0mail for the doubts t= hat still linger.

Regards,
A-DRS.
--90e6ba6e8fb83e236d0502f57163-- --===============6503212041451765132== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs --===============6503212041451765132==--