Re: Kernel 2.4.x and 2.4.1-preX - Higher latency then 2.2.xkernels?

[Shawn Starr]

Will this patch work with the low-latency patch? I have a few other patches in this kernel (one
fixing the ps hang issue).

Andrew Morton wrote:

> Shawn Starr wrote:
> >
> > Andrew, the patch HAS made a difference. For example, while untaring glibc-2.2.1.tar.gz the
> > system was not sluggish (mouse movements in X) etc.
> >
> > Seems to be a go for latency improvements on this system.
>
> Shawn,
>
> could you please try this patch in a pristine 2.4.1-pre10? It
> gets reiserfs down to 4 milliseconds worst case.  If the
> system's interactivity is still sluggish with this then
> reiserfs isn't the cause.
>
> Thanks.
>
> --- linux-2.4.1-pre10/include/linux/reiserfs_fs.h       Tue Jan 23 19:28:16 2001
> +++ linux-akpm/include/linux/reiserfs_fs.h      Sun Jan 28 22:37:11 2001
> @@ -1161,7 +1161,8 @@
>  #define fs_generation(s) ((s)->u.reiserfs_sb.s_generation_counter)
>  #define get_generation(s) atomic_read (&fs_generation(s))
>  #define FILESYSTEM_CHANGED_TB(tb)  (get_generation((tb)->tb_sb) != (tb)->fs_gen)
> -#define fs_changed(gen,s) (gen != get_generation (s))
> +#define __fs_changed(gen,s) (gen != get_generation (s))
> +#define fs_changed(gen,s) ({if (current->need_resched) schedule(); __fs_changed(gen,s);})
>
>
>  /***************************************************************************/
> --- linux-2.4.1-pre10/fs/reiserfs/journal.c     Tue Jan 23 19:28:15 2001
> +++ linux-akpm/fs/reiserfs/journal.c    Sun Jan 28 22:31:12 2001
> @@ -2649,6 +2649,8 @@
>        }
>  #endif
>        wait_on_buffer(bh) ;
> +      if (current->need_resched)
> +       schedule();
>      }
>      retry_count++ ;
>    }
> @@ -3085,6 +3087,8 @@
>      /* copy all the real blocks into log area.  dirty log blocks */
>      if (test_bit(BH_JDirty, &cn->bh->b_state)) {
>        struct buffer_head *tmp_bh ;
> +      if (current->need_resched)
> +        schedule();
>        tmp_bh = getblk(p_s_sb->s_dev, reiserfs_get_journal_block(p_s_sb) +
>                      ((cur_write_start + jindex) % JOURNAL_BLOCK_COUNT),
>                                        p_s_sb->s_blocksize) ;
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [e-mail]
> Please read the FAQ at http://www.tux.org/lkml/