Message ID | 5CDBD157020000780022F1EA@prv1-mh.provo.novell.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | page-alloc: accompany BUG() with printk() | expand |
On Wed, May 15, 2019 at 02:44:07AM -0600, Jan Beulich wrote: > Log information likely relevant for understanding why the BUG()s were > triggering. > > Requested-by: Andrew Cooper <andrew.cooper3@citrix.com> > Signed-off-by: Jan Beulich <jbeulich@suse.com> > Reviewed-by: Wei Liu <wei.liu2@citrix.com>
On Wed, May 15, 2019 at 02:44:07AM -0600, Jan Beulich wrote:
> Log information likely relevant for understanding why the BUG()s were
Why not use panic instead of printk + BUG?
Thanks, Roger.
On 15/05/2019 10:44, Roger Pau Monné wrote: > On Wed, May 15, 2019 at 02:44:07AM -0600, Jan Beulich wrote: >> Log information likely relevant for understanding why the BUG()s were > Why not use panic instead of printk + BUG? Because the backtrace is useful a lot of the time. ~Andrew
On 15/05/2019 09:44, Jan Beulich wrote: > Log information likely relevant for understanding why the BUG()s were > triggering. > > Requested-by: Andrew Cooper <andrew.cooper3@citrix.com> > Signed-off-by: Jan Beulich <jbeulich@suse.com> Reviewed-by: Andrew Cooper <andrew.cooper3@citirx.com>
On Wed, May 15, 2019 at 10:59:20AM +0100, Andrew Cooper wrote: > On 15/05/2019 10:44, Roger Pau Monné wrote: > > On Wed, May 15, 2019 at 02:44:07AM -0600, Jan Beulich wrote: > >> Log information likely relevant for understanding why the BUG()s were > > Why not use panic instead of printk + BUG? > > Because the backtrace is useful a lot of the time. Oh, panic has no backtrace, that's not very useful. Maybe we could add a panic_trace version to avoid having to do a printk + BUG. In any case: Reviewed-by: Roger Pau Monné <roger.pau@citrix.com> Thanks, Roger.
On 15/05/2019 11:09, Roger Pau Monné wrote: > On Wed, May 15, 2019 at 10:59:20AM +0100, Andrew Cooper wrote: >> On 15/05/2019 10:44, Roger Pau Monné wrote: >>> On Wed, May 15, 2019 at 02:44:07AM -0600, Jan Beulich wrote: >>>> Log information likely relevant for understanding why the BUG()s were >>> Why not use panic instead of printk + BUG? >> Because the backtrace is useful a lot of the time. > Oh, panic has no backtrace, that's not very useful. Maybe we could add > a panic_trace version to avoid having to do a printk + BUG. In any > case: TBH, It would be better for BUG() to take a printk-like , but arranging this to work is decidedly non-trivial, because it would have to be arranged from #UD context with some way of passing va_args() behind the scene. ~Andrew
--- a/xen/common/page_alloc.c +++ b/xen/common/page_alloc.c @@ -984,7 +984,15 @@ static struct page_info *alloc_heap_page for ( i = 0; i < (1 << order); i++ ) { /* Reference count must continuously be zero for free pages. */ - BUG_ON((pg[i].count_info & ~PGC_need_scrub) != PGC_state_free); + if ( (pg[i].count_info & ~PGC_need_scrub) != PGC_state_free ) + { + printk(XENLOG_ERR + "pg[%u] MFN %"PRI_mfn" c=%#lx o=%u v=%#lx t=%#x\n", + i, mfn_x(page_to_mfn(pg + i)), + pg[i].count_info, pg[i].v.free.order, + pg[i].u.free.val, pg[i].tlbflush_timestamp); + BUG(); + } /* PGC_need_scrub can only be set if first_dirty is valid */ ASSERT(first_dirty != INVALID_DIRTY_IDX || !(pg[i].count_info & PGC_need_scrub)); @@ -1393,6 +1401,11 @@ static void free_heap_pages( break; default: + printk(XENLOG_ERR + "pg[%u] MFN %"PRI_mfn" c=%#lx o=%u v=%#lx t=%#x\n", + i, mfn_x(page_to_mfn(pg + i)), + pg[i].count_info, pg[i].v.free.order, + pg[i].u.free.val, pg[i].tlbflush_timestamp); BUG(); } @@ -2336,7 +2349,15 @@ void free_domheap_pages(struct page_info for ( i = 0; i < (1 << order); i++ ) { - BUG_ON((pg[i].u.inuse.type_info & PGT_count_mask) != 0); + if ( pg[i].u.inuse.type_info & PGT_count_mask ) + { + printk(XENLOG_ERR + "pg[%u] MFN %"PRI_mfn" c=%#lx o=%u v=%#lx t=%#x\n", + i, mfn_x(page_to_mfn(pg + i)), + pg[i].count_info, pg[i].v.free.order, + pg[i].u.free.val, pg[i].tlbflush_timestamp); + BUG(); + } arch_free_heap_page(d, &pg[i]); }
Log information likely relevant for understanding why the BUG()s were triggering. Requested-by: Andrew Cooper <andrew.cooper3@citrix.com> Signed-off-by: Jan Beulich <jbeulich@suse.com>